TW493205B - Method and apparatus for wafer metrology - Google Patents

Abstract

A surface metrology device, comprising a metrology unit receiving information from a measurement region of a surface; and a first imaging camera with a first field-of-view containing the measurement region.

Description

493205 A7 B7 V. Description of the invention (1) Background of the invention a. Field of the invention The present invention relates to optical measurement methods, and more particularly to a thin-film reflection measurement and profile measurement on a semiconductor crystal circle line. Relevant technical description In the integrated circuit (IC) industry, in the wafer manufacturing technology, small key size process equipment is targeted. Technical factors such as the ratio of light wavelength to key dimensions in manufacturing equipment, and economic factors such as wafer output, manufacturing cost (C〇〇) and overall equipment effectiveness (EE) are quite important. IC manufacturing requires Hundreds of manufacturing steps. In some manufacturing steps, substances are formed successively on top of a substrate. Generally, a thin film layer is deposited on a wafer in a previous process and is subsequently flattened to a high accuracy. If so, the leveling step is usually performed by chemical mechanical polishing (CMP). After the CMP process step, the thickness of the remaining film is within its allowable error range. Optical methods are convenient for measuring the thickness of this film because light is nondestructive and invasive. The light characteristics of the measurement surface or the light interaction of the measurement film are affected to obtain the characteristics of the film on the wafer. As critical dimensions on wafers shrink, optical measurement techniques must be advanced to achieve the required accuracy. Economic factors are technologies that affect the development of semiconductor equipment. These paper sizes are applicable to China National Standard (CNS) A4 specifications (210 X 297 mm) (Please read the precautions on the back before filling out this page. Printed by the Bureau ’s Consumer Cooperatives 4-493205 A7 B7 Printed by the Consumer Property Cooperatives of the Intellectual Property Bureau of the Ministry of Economy .Because all processes are performed in a strictly controlled environment. The size of the mechanical equipment is also an important factor. In addition, no matter how complicated the process or measurement method is, easy operation is also very important. Factors and economic factors can be expressed in graphs such as c00 and 0EE. In the 1990s, 1C manufacturers attached importance to economic factors such as OEE and C00, as well as improvement in yield. A basic demand system It is to accurately measure its geometry and the amount of contamination caused during the wafer manufacturing process. Based on this demand, the `` measurement main frame / device '' is used, which is part of the IC manufacturing line Integration. Regardless of "offline", the result is to make the process control device better. At present, in the advanced process of IC, the yield can be as high as 80 ~ 90%, while the past yield is 50 ~ 60% or lower. However, the cost of manufacturing equipment will increase 3 to 5 times due to this excellent yield. There are two major problems with this offline measurement control method: (1) You must wait for this A test measurement of the measurement main frame system to confirm the results of each process, and (2) it is difficult for the process engineer to achieve and maintain the optimal process parameters. 'In the future, measurement in the integrated (online) measurement method The equipment will be set in the process equipment, and it must face the need to reduce costs. Integrated measurement methods include integrating the measurement system into the process tool, so that the measurement time can be substantially reduced 'and shortened between Process --- ^ ---, ------ • Installation (Please read the precautions on the back before filling this page

ϋ n ϋ · ϋ n 1 ϋ ^ · ϋ ^ 1 I Stupid · This paper size applies to China National Standard (CNS) A4 (210 x 297 mm-5- 493205) Printed by the Consumer Cooperative of Intellectual Property Bureau of the Ministry of Economic Affairs A7 B7 V. Description of the invention (3) Feedback time between control and measurement system. By measuring the critical dimensions of each wafer after processing, the process tool will obtain the latest wafer information without pausing manufacturing. If so, it can be achieved Good wafer-to-wafer control. The measurement method of the whole assembly is also significant to reduce the cost of test wafers, improve manufacturing quality, and maintain on-time operations to reduce operating costs and reduce the number of scratched wafers. From the above, Many processes are manufactured by microelectronics, which are conducive to integrated measurement, including CP, plasma etching, chemical vapor deposition, and stone engraving. Known technologies for chemical devices used for thin film measurement are facing industry requirements. At present, it is mostly limited to the ability to accurately measure the thickness of the film. Generally speaking, the online devices of the conventional technology are limited to measuring the thickness of about 80 nm. However, the industry still measures the thickness of the film The requirements of about tens of nanometers, and even the conventional technology device, are limited to the ability to quickly measure all the characteristics of the wafer surface, because the semiconductor industry needs improved devices and methods for applying integrated thin film measurement to It provides the advantages lacking in the conventional technology. Schematic illustration Fig. 1 is an overall view of the system hardware according to the embodiment of the present invention. Fig. 2 A and Fig. 2 B are illustrations of the novelty of the embodiment of the present invention. Figure 4 shows an embodiment of the wafer alignment device. Figure 5 shows the calibration of the wafer alignment device. Figure 6 shows the present invention improves the accuracy of the wafer alignment device. Figure 7 shows the use of a large-scale view. Field (LF0V) camera and a small field of view (SF0V) camera to avoid exploring the wafer in the process. The paper size of the paper applies the Chinese National Standard (CNS) A4 specification (210 X 297 mm)- 6-1 .-! ---------- • Installation ---- Ί * — · — Order --------- (Please read the precautions on the back before filling this page) Printed by A7 of the Intellectual Property Bureau of the Ministry of Economic Affairs and Consumer Cooperatives ________ B7 V. Description of the invention (4) Specific area% Figure 8 illustrates the use of LF〇V The advantage is that it is easy to determine the size during training. ® 9 represents a mirror-like autofocus system. Figure 10 illustrates an asymmetric mirror-like autofocus system. Figure 11 illustrates the sensitivity of tilt in an asymmetric mirror-like autofocus method. Figure 12 A specific embodiment of the autofocus system is shown. Figure 13 illustrates the improved sensitivity of surface displacement measurement using a symmetrical mirror autofocus system. Figure 14 shows the surface tilt sensitivity of the symmetrical mirror autofocus system. Figure 1 5 FIG. 16 shows another embodiment of the autofocus system. FIG. 16 illustrates the surface tilt sensitivity of the embodiment shown in FIG. 15. Figure 17 illustrates an embodiment of a wafer support. Figure 18 illustrates an embodiment of a wafer support. Figure 19 illustrates an embodiment of a wafer support. Figure 20 illustrates an embodiment of a wafer support. Figure 21 illustrates details of an embodiment of a wafer supporter. Figure 22 illustrates details of an embodiment of a wafer supporter. Figure 23 illustrates an embodiment of wafer immersion hardware before wafer immersion. Instrument 24 illustrates an embodiment of wafer immersion hardware at the beginning of wafer immersion. Figure 25 illustrates an embodiment of a wafer immersion hardware after the wafer immersion stage. This paper size applies to China National Standard (CNS) A4 specification (210 X 297 mm) —J --------- Packing ----- ^ ---- Order -------- -(Please read the precautions on the back before filling this page) 493205 A7 _B7_ V. Description of the invention (5) Figure 2 & Description of wafer rotation in rotating wafer chuck Figure 2 7 illustrates one of the integrated measuring devices Examples. — ^ ------ Equipment ----- ^ ---- Order --------- (Please read the notes on the back before filling out this page) Employee Cooperatives of Intellectual Property Bureau, Ministry of Economic Affairs Printed main component comparison table 1001 10 1 10 3 1 4 10 5 110 111 1 2 0 1 3 0 13 5 13 7 13 8 13 9 1 4 0, 1 4 1 14 5 14 6 1 5 Ό 1 609 1 90 Reflectance Gauge Vacuum Chuck Light Source Optical Fiber First Beam Splitter Second Beam Splitter Wafer Measurement Area Window Collimator Optical Coil Spectrometer Spectral Fiber Aperture Lens Large Field-of-View Camera Small Field-of-View Camera Autofocus Objective Lens Combination First Optical Path Plate This paper is sized for China National Standard (CNS) A4 (210 X 297 mm) -8-493205

V. Description of the invention (6) Board 7 ^ Detailed description Fig. 1 shows the overall view of the system hardware according to the embodiment of the present invention. Figure 1 shows: reflection measurement meter 100, vacuum chuck 1 0, vacuum chuck symmetry axis 10, light source fiber 1 〇3, first beam splitter 10, second beam splitter 1 〇 5. Semiconductor wafer 1 1 0, measurement area 1 1 1, window 12 0, collimator 13 0, optical coil 135, first imaging optical combination 1 3 7, second imaging optical combination 丨 3 8, third Imaging optics 1 39, 0 (including calibration filter) 140 and 1 4 1, spectral fiber 1 4 5, aperture lens 1 4 6, large field of view camera 1 50, small field of view camera 1 6 0. Autofocus objective lens combination 190, first optical path plate 195, and second optical path plate 197. The semiconductor wafer 110 is connected to the vacuum chuck 101, and its center of gravity is fixed relative to the laboratory system. However, the vacuum chuck is rotated around the vacuum chuck symmetry axis 102. The reflection measuring device combination 100 includes a window 120 and first and second optical path plates 195 and 197. The first optical path plate 195 is freely movable along the y-axis, and is driven by a driver in the embodiment. The second optical path plate 197 is connected to the first optical path plate, and the second optical path plate is moved along the X axis relative to the first optical path plate. The objective lens assembly 19 '0 is abutted against the second optical path plate and is movable along the z-axis. Therefore, the embodiment shown in FIG. 1 has four-dimensional spatial degrees of freedom: movement along the (X, y, Z) axis and rotation of the vacuum chuck about the axis of symmetry of the vacuum chuck. In the embodiment shown in FIG. 1, in addition to the components on the second optical circuit board, the paper size applies the Chinese National Standard (CNS) A4 specification (210 x 297 mm) -9--^ --- r-- ---- Equipment ----- ^ ---- Order --------- (Please read the notes on the back before filling out this page) Printed by the Employees' Cooperative of the Intellectual Property Bureau of the Ministry of Economic Affairs 493205 Economy Printed by A7 B7 of the Consumer Cooperatives of the Ministry of Intellectual Property Bureau V. Invention Description (7) 'All optical components are connected and relatively fixed to the first optical circuit board 195. The objective lens set 190 is connected to the second optical path plate 197. This is because the objective lens system can move freely along the X axis. In addition, the objective lens assembly can be moved along the z-axis to focus on the semiconductor wafer 1 10. The first and second optical boards can move along the X and Y axes to access the entire wafer surface. The rotation of the wafer connected to the vacuum chuck can be combined with the movement of the first and second optical path plates along the X and Y axes, so that the measurement device can quickly approach the surface of the semiconductor wafer or reduce the obstacles therebetween. It is possible to complete a wafer with a diameter of 200 mm, and directly measure wafers with a diameter of 300 mm or larger. The reflection measurement meter combination 100 measures the selected area on the semiconductor wafer 110 to locate a specific area of the semiconductor. The surface of the semiconductor wafer is taken by a large field of view camera and a small field of view camera. image. The large field-of-view camera has a lens of about 20 m x 27 mm, while the small field-of-view camera has a lens of about 1 m x 1 · 3 m. The reflectometer combination 100 has a wideband (ultraviolet, visible, infrared) reflection measurement system. In an embodiment, the light source (not shown) may be a gas light lamp connected with an optical fiber, and is connected to the system with the optical fiber 103 as the light source. The photocoil 1 3 5 transmits the collimated light from the lens combination 130 to the beam splitter 104. The light source directly emits a monitoring beam from the optical fiber of the light source through the beam splitter, and the monitoring beam does not interact with the measurement area 1 1/1. The light beam split by the beam splitter and shines on the surface of the wafer is the measurement area. The measurement beam is reflected from the surface of the wafer, and its spectrum changes according to the film on the surface of the wafer. As the measuring beam is reflected by several mirrors 1 3 5 and returned to the beam, the paper size applies the Chinese National Standard (CNS) A4 specification (210 x 297 mm) -10-—r-J _------- ---- ^ ---- ^ --------- (Please read the notes on the back before filling out this page) 493205 Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs A7 B7 V. Description of the invention ( 8) Isolator. The ^ th imaging optical combination 1 3 7 focuses the measuring beam on the aperture mirror 1 4 6. The light shining on the aperture mirror passes through the spectral fiber 14 5 and is converted into the spectrum 14 0. This spectrum represents the most primitive information about the thin film on the wafer. The monitoring beam is similar to the above, but passes through a different aperture mirror 1 4 6 and a spectral fiber 1 4 5 to form a spectrum 1 4 1. The monitoring spectrum of this measurement represents the spectral system and optical components, which can correct the characteristics of the instrument in the measurement spectrum of the thin film. Part of the measurement beam is reflected by the aperture mirror 1 4 6 and refocused on a small field of view (S F 0 V) camera. The image formed is the indication pattern on the semiconductor wafer 1 4 5. The aperture mirror itself is also imaged at SFOV, forming a black spot on the wafer pattern. The black dot indicates the precise position of the thickness measurement on the wafer pattern. As described above, the relative spectrum of reflected light and incident light of the semiconductor wafer 110 can be measured. The thickness of the film on the measurement area 110 can be determined by the reflected "measurement beam 〃" and the incident "monitor beam 以" by the optical principle in the conventional technology. The advantage of the present invention over the conventional technology is that the mirror array is used to scan in a space. If the light beam is reflected from the mirror array and collimated fairly well, the scanning will not damage the overall performance of the system. However, due to scattering, the beam cannot be completely collimated and aligned, because the beam should be as small as possible when scanning the target. 'In the present invention, the main optical scanning system moves in one dimension on the first optical path plate, while the optical scanning system in other two-dimensional space is relative to the coordinate system fixed in the laboratory, but only in one-dimensional spatial direction (X) Is relative to the first optical path plate. Because the loop scan length does not exceed the diameter of the wafer. In this paper size, the Chinese National Standard (CNS) A4 specification (210 X 297 mm) is applicable. -11---f ---------- install ----- ^ ---- order-- ------- (Please read the precautions on the back before filling out this page) 493205 A7 B7 Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs 5. Description of the invention (9) The conventional device is fixed in the device The target scan is two-dimensional, and the scan length is twice the wafer diameter. Yet another advantage of the present invention is that the optical path remains fixed regardless of the scanning position. Because if the target is the focus, the amount of light beam scattering that has an optical reflection from the wafer surface will not change. In the conventional device, the scanning of the wafer surface changes all the entire optical path lengths and the amount of scattering of the collimated beam. The embodiment in FIG. 1 requires special attention. The semiconductor wafer 1 i 0 is placed on the reflection measurement meter combination 100. In another embodiment, the semiconductor wafer can be placed in a pool under an optical system, which can be seen as the bottom. It is necessary to be different from the wafer processing in FIG. 1, and the device has one side facing up. In this embodiment, the optical system (including the main window) may be lowered toward the semiconductor wafer, or the semiconductor wafer may be raised toward the optical system. Compared with Fig. 1, the system of this embodiment can rotate 180 degrees around a horizontal axis. It is also possible to make a general rotation of the relative structure of Fig. 1, such as a 90 degree rotation. The impact generated by these rotations lies in the technology of wafer processing. In another specific embodiment of the present invention, there may be no need for water in the measurement path, that is, the equipment is dry. In such an embodiment, the orientation of the device relative to the laboratory is arbitrary. For example, the embodiment of Fig. 1 may have its side facing down or upside down, while some optics are best redesigned, but not necessarily necessary. As known, the reflective optics used in the present invention have at least three advantages. Fresnal reflection occurs in refractive elements (such as lenses) (please read the precautions on the back before filling this page). --- The size of Huaben paper is in accordance with Chinese National Standard (CNS) A4 (210 X 297 mm) -12- Printed by the Intellectual Property Bureau Employee Consumer Cooperative of the Ministry of Economic Affairs 493205 A7 __ B7 V. Description of the invention (10) Surface, and Will be a source of systematic noise in the system. For example, even if no wafer appears, it will be affected by the reflection of the objective lens before it reaches the photodetector, because this kind of light does not represent the cost of the wafer and is noise. On the contrary, the reflective element is not affected by the reflection of F r e s n a 1. The refractive element can simultaneously limit the bandwidth of light in both directions through the refractive element. A preferred embodiment is to reduce Fresnal reflection by using a refractive element with a retroreflective coating (ARC). Generally speaking, the A R C series resonance structure 'can operate well within a limited wavelength range. Outside this range, its transmission property is reduced, thereby limiting the bandwidth of the system. At the same time, the refractive index of most substances is a complex function related to wavelength. The imaginary part (K) of the refractive index indicates the attenuation of light through the substance at a specific wavelength. Therefore, many lens materials have a large K 可以, which can limit the system's wavelength bandwidth to a specific range. The second advantage of the reflective element is that it prevents light from being attenuated when passing through the lens material. The third problem is that the refractive element must be color corrected. The real part (N) of the refractive index is also a function of wavelength. N 値 affects the focal length of the lens. Because the lens has chromatic aberration, that is, different colors focus at different distances, usually the different components in the system are corrected with materials with spectral N 値 characteristics. Because the reflective element is not focused by refraction, it is not affected by color aberrations caused by changes in the refractive index N. The reflective element has considerable restrictions on the aperture and geometry, making the refractive element more suitable for use in specific embodiments. In these embodiments, the optical elements used for immersing the semiconductor wafer in water are color corrected. The design of these optical elements also considers water as an optical element. In the embodiment shown in FIG. 1, the optical measurement system is based on a relative experiment. The paper size is in accordance with China National Standard (CNS) A4 (210 X 297 mm). ---- l · —Order --------- ^ 9, (Please read the notes on the back before filling this page) 493205 A7 B7 V. Description of the invention (11) The room is a fixed window 1 2 0, and another embodiment other than FIG. 1 is a new window embodiment. Figure 2A illustrates a conventional device having a single large relatively laboratory fixed window. Figure 2A includes wafer 200, wafer surface 201, receiving wall 201, objective lens assembly 07, beam splitter 235, optical loop 237, and window 202. It is important to note that the conventional device utilizes a single large window 2 02. In order to measure accurately, the window 202 must have comparable optical quality. And due to the size of the window, it will cause considerable overhead. FIG. 2B is the new method of the present invention, and the illustration includes: wafer 200, column 2 01, small scanning window 2 0 2, light detection element 20 3, beam splitter 235, mirror 237, and optical fiber 2 〇4, optical combination 205, lighting device 206, and objective lens combination 207. In Fig. 2B, part of the optical system is fixed to the water column of the objective lens assembly 207. The bottom of the column 201 is a small window 202. The column side 209 is upright with a small gap between the wafer and the water flow from the supply pipe 206 into the column. The combination of surface tension and viscosity keeps the water in it. The height of the gap determines the flow of water, which keeps the water flow in the pillar between the wafer 200 and the small window 202. An additional ejection hole can be used to remove air bubbles, it is noted that the water column forms an optical element, and a specific embodiment may include an extended water tank. 1 靑 Referring to FIG. 2B, the watertight scanning optical combination 2 0 5 has a lighting device 2 6 'which receives light from the optical fiber 2 0 4. The illuminating device transmits a light beam (which may pass through a collimator) 'through a beam splitter 2 3 5 to an objective lens combination 2 07. Objective lens combination 2 0 7 This beam is condensed. The paper size is in accordance with Chinese National Standard (CNS) A4 (210 X 297 mm) -14-(Please read the precautions on the back before filling this page) ·· Binding --- -Printed by the Consumer Cooperative of Intellectual Property Bureau of the Ministry of Economic Affairs, printed by the Employee Cooperative of the Intellectual Property Bureau of the Ministry of Economic Affairs, printed by 493205 A7 B7 V. Description of the invention (12) Focus on the wafer, collect the reflected light and transmit it to the mirror 2 3 7 Beam (collimated or uncollimated). The mirror shifts the light reflected to the wafer 200 and enters the light detection element 203. The detection element has an aperture mirror spectrometer and a vision system (not shown) to make the The precise position of the optical module 2 05 can be positioned at a predetermined position on the wafer. A mechanical transfer platform (not shown) scans the component with its water column and optical elements. The present invention has two advantages over those using a single large window and water bath. First, the target is always observed from the same part of the same window, because it has little effect on the quality of the measurement. (These effects can be resolved by calibration). Secondly, because it is smaller than the window of the conventional technology, it is easier to obtain high-quality surface luminosity. By properly designing the geometry of the water column determined by the window 202, the column side 200, and the supply line 206, the water flow can flush any bubbles out of the pillars on both sides of the wafer. Compared with the conventional technology, the relatively small area of the present invention is easier to achieve without the need to lower the wafer to the height of the water bath with a special wafer processor as in the conventional technology. To make the measurement more accurate, the present invention uses a reference reflector to correct the slowly changing characteristics of the measurement system. Figure 3 shows an embodiment with a reference reflector. Figure 3 includes wafer 300, window 3 0, reference reflector 309, reference volume wall 310, reference volume 3 1 1, main water volume 3 0 1, objective lens combination 3 7 and loop optics 3 3 5 ° In Figure 3, the reference volume wall 3 1 0 refers to the reference volume 3 1 1 and the paper size to the Chinese National Standard (CNS) A4 (210 X 297 mm) -15-" " " " " --r -------------- Γ III ^ · I ------- (Please read the notes on the back before filling this page) 493205 Ministry of Economic Affairs Printed by the Consumer Property Cooperative of Intellectual Property Bureau A7 B7 V. Description of the invention (13) The main water capacity is separated by 3 0 1 and the reference volume 3 1 1 can be filled with air, water or suitable substances. The present invention confirms that the characteristics of the reference reflector 309 are stable for a long time. When the volume 3 1 1 series is not filled with water, the distance between the window 3 0 2 and the reference reflector can be adjusted. When the objective lens combination 3 7 7 is located below the window 3 2 at the same distance, To focus the reflector while the wafer is in focus. In a preferred embodiment, the volume is filled with an inert solid, and the height from the window 302 to the reflective surface can be adjusted appropriately. The reference reflector 3 0 9 can be chopped, fused sand dioxide, or other inert substances. It may include a deposition substance deposited on the substrate to stabilize mechanical and optical characteristics. In a preferred embodiment, the reference reflector includes a fused silica substrate and a chromium film formed on the top surface. The reference reflector of another embodiment is formed on a lower surface with silicon and a reflective oxide layer. Referring to Fig. 3, the reference reflector 309, the reference volume wall 3 10, and the window 3 02 may be combined in different ways. In a preferred embodiment, the reflector and the window are sealed to the window. In another embodiment, the reflector, the reference volume wall and the window are fixed to the body with a polymer glue, such as resin or super glue. In other embodiments, the volume 309 is not sealed in the main volume 301, and the elements can be connected together or held in one place by mechanical elements such as stoppers or springs. The reference volume seal is used to reserve the reflective characteristics of the reflector, that is, to avoid being dirty or eroded by a material water bath, such as a CMP polishing slurry. The sealing method for the volume 3 1 1 is to prevent damage or damage caused by different thermal expansion coefficients during operation or transportation. The paper size is applicable to China National Standard (CNS) A4 (210 X 297 mm) -16 -r _--------- install ----- ^ ---- order --------- ^ 9. (Please read the notes on the back before filling this page) Ministry of Economy Printed by the Intellectual Property Bureau's Consumer Cooperatives 493205 A7 _ B7__ V. Explanation of Invention (14) Omission. θ In the preferred embodiment, the reference reflector 3 0 9 is placed directly close to the objective lens combination 3 7. The objective lens combination can be scanned in one dimension and moved to the reference reflector. However, in the embodiment where the wafer scans the target, the reference reflector will do the same. When the wafer of the embodiment is on the objective lens combination, as shown in FIG. 3, another embodiment can make the objective lens combination higher than the wafer, or an arbitrary inclined position. According to the above description of the present invention, the reference spectrum generated by the reference reflector 309 is collected in intervals. With the collected spectrum, the reference spectrum can be used to calculate the film thickness on the surface of the wafer 300. It is best to collect the reference spectrum before wafer measurement. There are several ways to substitute the reference spectrum generated by the reflector into the data reduction algorithm. In the embodiment, each spectrum from the wafer is The spectrum of the reflector is orthogonal. Calibrating the measuring device can correct the spectrum of the wafer to measure the corresponding collected spectrum, the adjustment can adjust the above algorithm, and get the correct answer of the calibrated wafer. The reference spectrum can be used in the same way in the calibration algorithm when measuring, because the changes after the final system calibration and the current measurement will not affect the results of the algorithm. As mentioned above, the embodiments of the present invention (see Figs. 1 and 3) may include a reference reflector and a dual spectrum analyzer. The initial value of the measurement is the spectrum S, which is the output of the sample under the system's relative reflection test. In addition to the characteristics of the sample, S is determined by the wide band of illumination (ultraviolet, visible, infrared). The optical system, optical detector, digitizer, and other components are included in the measurement system. The characteristics of these measurement systems make the sample paper size applicable to the Chinese National Standard (CNS) A4 specification (210 X 297 mm) -17- --τ ---'------ install ----- „- --- Order —-------- (Please read the notes on the back before filling this page) 493205 A7 Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economy Because of the effect. The present invention uses an approximately orthogonal incident reflection measuring instrument to accurately measure the refractive index of a thin film to obtain its thickness. In a preferred embodiment of the present invention, the cost function is constructed by the instrument. The obtained spectrum can be used to determine the thickness without calculating the reflection characteristics of the surface. As mentioned above, the device of the present invention includes a calibration reflector // a reference reflector and a two-beam spectrometer. The maximum number is the spectrum S ′, which represents the system output corresponding to the reflection of the sample under test. In addition to the characteristics of the sample obtained from the test, the spectrum s also depends on the amount of the bulb, optical system, detector, and digitizer. The characteristics of the measurement system. The information obtained from the test sample is accurate, so an accurate measurement system must consider these factors at the same time. Some measurement system characteristics change significantly with time, while others remain fixed. In the embodiment of the present invention An arc lamp is used as the light source, the arc flashes very quickly, the light source fiber 103 is bent (as shown in Figure 1), and the length of the light path is changed rapidly due to the scanning. The aging of the electric lamp will cause a slow According to the present invention, a dual spectrometer can simultaneously collect two kinds of spectra, one is a sample's reflection spectrum and the other is a monitoring spectrum. The monitoring spectrum During the test, it will have no effect on the sample, as shown in Figure 1. In Figure 1, the sample of the "g measurement type" is a semiconductor wafer 1 1 0. This sample can be used as a reference reflector or a correction reflector at the same time, as described above. This should be removed by an accurate film thickness measurement system (please read the precautions on the back before filling this page)-binding-binding ----- Φ This paper size applies to Chinese national standards CNS) A4 specification (210 x 297 mm) -18-493205 Employees ’cooperatives printed leftovers A7 B7 in the Intellectual Property Bureau of the Ministry of Economic Affairs 5. Description of the invention (16) Η 1 stomach knowledge 4 The light path system and the amount to determine the monitoring spectrum The light of the spectrum measurement is similar to each other except that the measurement spectrum is transmitted from the measurement area 1 1 1 ° The preferred embodiment of the two types of beams shown in Figure 1, where the light sources of the two beams can be the same In Figure 1, the 'optical splitter' separates the reflected beam from the monitoring beam, and the monitoring beam directly passes through the beam splitter to the spectrometer 141. The reflected beam passes through the objective lens to the sample, returns to the objective lens and beam splitting The reflector reflects a beam parallel to the monitor beam to the spectrometer 140 °. Among them, there are many conventional optical elements between the beam splitter and the mirror in FIG. 1, which are not shown in the embodiment. The path of the two beams is preferably as similar as possible. When the sample is under test, that is, the wafer has been polished, the spectrum S r is a reflection spectrum and the monitoring spectrum is a spectrum sm. The monitoring spectrum is used to correct rapid changes in the system, such as flicker of the illumination light source. The invention can use the reference spectrum S r to correct the slowly changing characteristics of the system. The spectrum S r represents a very stable reflection of the sample in the system. At the same time, the corresponding monitoring spectrum S m is also collected, and whenever a new wafer is placed in the instrument, the system can collect S r. At the same time, the invention can use the correction spectrum S c to correct the characteristics of fixed or slow change in the system. S c is the reflection spectrum obtained from a known sample. In addition to its corresponding monitoring spectrum S c m, it includes the relevant corrected reference spectrum S c r and its corresponding monitoring spectrum S c r m. The latter two spectra are collected before or after S c, and are collected when the system is in operation for a considerable period of time 'about 3 months. This paper size applies to China National Standard (CNS) A4 specification (210 X 297 mm) -19---p ---, ------ pack ----- ^ ---- order --- ------ S ----- (Please read the notes on the back before filling this page) 493205 Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs A7 B7 V. Invention Description (17) Familiar with this technology In other words, the above-mentioned spectra are described in a rough form, and the spectra will be destroyed by different components. Detector The first current and the readable current will destroy the measured spectrum. The incident reflections in the system will also destroy these spectra, especially the reflection spectrum. Because ‘all spectra will collect these unwanted contributions. The monitoring spectrum can be corrected by subtracting the dark spectrum, the dark spectrum, which collects the part of the monitoring light source where the illumination light source is blocked. The reflection spectrum can be corrected by subtracting the ', dark spectrum of the unreflected area in the spectrometer. Data reduction methods that determine film thickness can have a cost reduction function attached. A better cost function includes the above eight spectra and measurement parameters, and other information about the sample and measurement system. In general applications, there may be only a single parameter, the thickness of the film on the sample. In other embodiments, it may be two parameters, the thickness of two thin film layers, or the coefficient of one thin film layer and another layer affecting the refractive index. More than two parameters can be used at the same time. The cost function in the preferred embodiment is used to balance the characteristics of the measurement system, so as to reduce the dependence on the characteristics of the samples in the test. Information about the measurement system includes noise in the spectrum to weight the optimization conditions of the system. The information of the sample includes the thickness and optical characteristics of different substances on the sample. For example, conventional techniques, various types and methods of parameter reduction are all possible better cost functions as follows: Weighting number, R p is based on the paper size of parameter p. Applicable to China National Standard (CNS) A4 (210 X 297 mm) -20-

--Order --------- (Please read the notes on the back before filling out this page) Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs 493205 A7 ____B7_ V. Description of the invention (18) Theoretical reflection of calculations 値, R r is another theoretical reflection 値 calculated based on the corrected reflector. All spectra, reflection chirps, and weighted chirps are a function of wavelength. The numerator includes two terms C7 1 = S S rmS c r S cmRc and crl = SmSrSCSci'mRp, all of which are affected by the characteristics of the system. Because the characteristics of the system do not affect its minimum value. For example: S and S m have the same light source flickering speed. For example, the number of perforations (NA) of the s r m and S r 1 systems have the same effect on S and S r 1. Many versions of the cost function also exist in the teachings of the present invention, and until the impact of the measurement system is offset each. Compared with the above, a simplified version is also co-existed in the present invention, such as removing the monitoring spectrum, or missing four kinds of reflector-related spectra. In addition to the above-mentioned flat method, minimize the method as much as possible, such as the mi n i — ma X or L 1 method. It should be noted that R P and R r in the above are theoretical 値 and not measured reflection 値. Therefore, the method of the present invention does not use the spectrum obtained by an absolute measurement or a relative measurement method as in some conventional techniques. Quickly aligning the wafer to the optical system is another focus of the present invention. FIG. 4 shows an embodiment of the wafer aligner of the present invention. Figure 4 includes: wafer 4 0 3, rotating chuck 4 0 2, motor 4 1 2, water 4 0 4, window 4 0 5, water height 4 1 0, motor housing 4 0 0, rotating seal 4 〇1, Light source 4 ◦ 7, Light 4 1 3, Alignment window 408, Detector '406, and Sink wall 411. In FIG. 4, the rotation chuck 4 2 holds the wafer 403, and the motor 4 1 2 drives the rotation chuck to rotate about an axis (not shown). Water 4 0 4 is filled above the main window 4 05 to a water height of 4 1 0 and exceeds the water tank. The paper size applies the Chinese National Standard (CNS) A4 specification (210 X 297 mm) -21-11 .-- ------- • Equipment ---- I .--- --Order ------- 丨 Stupid L (Please read the precautions on the back before filling out this page) 493205 Employees of Intellectual Property Bureau, Ministry of Economic Affairs Printed by the Consumer Cooperative A7 B7 V. Description of the invention (19) The wall 4 1 1% rotating seal 4 0 1 is to isolate the motor housing 4 0 from water. The light source 4 7 is also housed in a dry casing. The light source generates light 413 through the alignment window 408 and enters the water from the dry housing. The detector is located in the drying volume below the window 405. Some light 4 1 3 hits the wafer 403 and is blocked. The rest of the light passes through the main window 405 and enters the drying volume below it. It is irradiated onto the detector. Rotating the chuck 4 02 rotates the wafer 403. When the wafer rotates, the edge of the wafer is directly moved radially above the detector (from left to right in Figure 4). The radial motion is caused by the wafer being deviated from the center of the chuck or the wafer itself is not very round. In addition to the margin of machining, the flat parts and notches on the chuck can also cause wafers to be out of round. The radial movement of the edge of the wafer 4 0 3 above the detector 4 6 changes the shadow of the light 4 1 3 on the detector. The detector can be a single long detector, which is an imaging diode. Body, or array detector, also known as c CD. In the foregoing case, the amount of light falling on the detector indicates the edge position of the wafer. When the edge of the wafer 403 moves to the position of FIG. 4, the light falling on the detector decreases. In general, the output of the detector, I, is a function of the wafer edge position X e, I = f (X e) (1), which is not necessarily linear but must be unitary because of its trans-X e = f-1 (I) (2) This paper size is applicable to China National Standard (CNS) A4 (210 X 297 mm) -22-Order --------- 3b (Please read the Note: Please fill in this page again.) · 493205 A7 B7 V. Description of the invention (20 can be used to determine the position of the edge. In another embodiment, the detector may include an array type detector, each array The elements in it have different positions X a. In this case, the density of light falling on different detection elements can be represented by a waveform: I (Xa) two g (Xe) (3) It can be processed as an algorithm, h is as follows X e = h (I (X a)) (4) The functions g and f will be further complicated by wave optics (Figure 4 illustrates particle optics), and the correction will determine f and g. Figure 5 illustrates wafer pairs Correction of the positive device. Figure 5 includes: spiral wafer 50, chuck 503, spiral edges 504 and 505, detector 506, and light source 507. Printed by the Consumer Cooperative of the Intellectual Property Bureau of the Ministry of Economic Affairs —H ------------ (Please read the note on the back? Matters before filling out this page) The thickness of the spiral wafer 500 can be compared with the silicon wafer, which is double-layered and clean. Mechanically opaque metals, such as stainless steel, have a mechanical coefficient that ensures that the center is aligned with the center of the chuck 50. When the chuck is turned, the helical edges 5 0 4 and 5 0 5 blocks the amount of light from the light source 5 0 7 reaching the detector 5 0 6. When the spiral rotates, the system records the output of the detector, which is a function of the angle. The radius of the spiral edge 5 0 5 The system is discontinuous. When the spiral system is located on the detector, the size of the paper is applicable to the Chinese National Standard (CNS) A4 specification (210 X 297 mm) -23-493205 A7 B7 V. Description of the invention (21). ; ^ Radius is a function of angular displacement and starts at the discontinuity of 5 0 5. Because the function g (Xe) or f (Xe) can be recorded, because f-1 or h can be used as true The wafer can be calculated. The above measurement can be calculated from the position of the flat part or the missing corner on the wafer within the 360 ° rotation range including f 1 or h from I. And the position of the wafer center relative to the chuck. Figure 6 is another embodiment of the improved wafer alignment accuracy of the present invention. Figure 6 includes a beam splitter 600, a lens 600, and a reference detector 602. , Light source 6 0 7, window 6 0 8, light 6 13, collimator lens 6 10 and wafer 6 06. Generally, the intensity of the light source 6 7 can be changed as a function of time and temperature. In order to correct and compensate for this part, part of the light can be twisted by the beam splitter 600, focused by the lens 601, and detected by the reference detector 602. The output of the reference detector can be used to control the intensity of the light source or to correct the data converted by the light source change. Fig. 6 shows another lighting method, in which the light source 607 generates divergent light. In the embodiment, the lens 6 1 3 is collimated to become the light 6 1 3. In another embodiment, a collimated light source may be used, and in other embodiments, a light diffusing element connected to the light source may be used to harmonize the spatial mode of the light source. FIG. 7 illustrates the use of a large field of view (L FOV) camera and a small field of view (S FOV) camera to avoid fumbling to locate a specific area of a wafer. FIG. 7 includes crystal 700, LFOV702, SFOV703, LFOV image 704, and SFOV image 701. LF〇V 7 0 2 series is usually larger than crystal 7 0, and the paper size in the wafer applies the Chinese National Standard (CNS) A4 specification (210 X 297 mm) -24-(Please read the precautions on the back before (Fill in this page) ▼ Binding Printed by the Employees' Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs 493205 A7 B7 V. Description of Invention (22) (Please read the precautions on the back before filling in this page> Heart position is greater than a certain level. Because it can be moved to the crystal of the non-oriented wafer, where the LFOV image 704 can be found. Once the LFOV image is found, the system can know the orientation of the wafer and its center Because it can locate the SF0V7 0 3 on the SFOV image 7 0 1. This step is decisive, it takes less time than only using SF0V, or it is actually correct Positive wafer. Figure 8 shows that the size of the crystal can be easily determined during training by using the LFOV camera. Figure 8 includes the crystal 800, the inner crystal street diameter 8 4 0, and the inner crystal lane diameter 8 5 0 , Crystal features 80 5 a-c, large viewing area 802, small viewing area images 803, 804, and 8 1 and measurement station 806. Printed by the Consumer Cooperative of the Intellectual Property Bureau of the Ministry of Economic Affairs for training, the operator can view one side of the wafer and even position the wafer so that the inner crystal street diameter is 8 4 0 and lane diameter 8 5 0 appear vertical and horizontal, as shown in Figure 8. However, if the wafer orientation is not necessary, it can be positioned in other directions in other embodiments. The rough step estimates the crystal size can be used in three The crystal characteristics selected by the operator on the side are 80 5 a-c, and the system can identify images on different crystals on the wafer and locate LF〇V with LF〇V and / or SF〇V cameras. And / or SFOV images 804 and / or 801 to obtain the exact size of the crystal. In this way, the operator does not need to know the size of the crystal in advance. The advantage of the LFOV camera is that it is easy to train the operator to connect to the measurement station And the position of the image on the wafer in SFOV. Ideally, this large viewing area includes all the crystals, as shown in Figure 8. Using the large viewing area 802 ', the operator can select the crystal at SFOV10. The standard of this paper is applicable to Chinese national standards (CNS) A4 specifications (210 X 297 mm) -25-493205 A7 B7 printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economy V. Invention description (23) domain. It is similar to using a map to cruise a specific city. Once the SF 〇 After V is properly positioned, the operator can accurately select the SF 0 V image 8 0 1 and the measurement station 8 0 6, which is similar to accurately finding the correct city with an intersection on a map. In an embodiment, on a crystal There are various measurement stations. In this case, different measurement stations may have different measurement layers. This thickness algorithm, which is the cost function for minimizing the aforementioned parameters, generally requires prior information about the calculation formulas for each stack after measurement. There are multiple measurement stations in different layer stacks on the crystal, or there are general calculation formulas to match different layer stacks. Another important point of the present invention is an autofocus system for a target optic. Figure 9 shows a spectral autofocus system. 9 includes: a sensor combination 99 3. an illumination light source 99 1. a detector 992, a sensor 990, a surface 902, 907, and 920, a shaft 996, an illumination beam 900, and a reflected beam 904. The components in the sensor combination 9 9 3, the illumination light source 9 9 1, the detector 9 9 2 and the sensor 9 9 0 are relatively fixed. The surface 9 0 2 has a correct distance Z between the nominal surface and the combination 9 9 3, which is orthogonal to the axis of the sensor 9 9 6. The surface 9 0 7 is a distance t from the nominal surface displacement. The Γ system is measured from the axis 996 of the sensor 990, and the surface 920 maintains an oblique angle 4 with the nominal surface. The components of the autofocus system are an illumination light source 9'9 1 and a detector 9 9 2. The illumination light source 9 9 1 generates an oblique illumination light beam 900 and the detector 992 detects a reflected light beam 9 0 4 reflected from the surface 902, 907, or 9 2 0. When the surface system is not inclined, the angle between the light beam and the sensor 9 9 6 is 0, and the entire combination 9 9 3 series. The paper size applies to the Chinese National Standard (CNS) A4 (210 X 297 mm) -26-

tr --------- Φ (Please read the notes on the back before filling this page) A7

493205 V. Description of the invention (24 moves up and down relative to the surface, the goal of the autofocus system is set at a distance Γ = 0, so that the distance between the sensor 9 9 0 and the surface 9 0 2 is ′ Z ′ and the oblique angle 4 Nothing. The system completes the setting by adjusting the height of the combination based on the output of the detector. Figure 1 0 illustrates an asymmetric spectral autofocus system. Figure 1 〇1 includes an oblique illumination beam 1 0 1 1 Nominal reflection surface 1 〇〇2, reflection point 10 03, reflected light beam 〇04, detector plane 1005, nominal detection point 1 0 6, curved surface 1 0 0 7, displacement reflection point 1 0 0 8, displacement reflected light 1 〇09 and the displacement curve point 101. An illumination light source (laser, imaging diode, white light, discontinuity, coherence) generates an oblique illumination beam 1 0 1 1 passing through a nominal reflecting surface 10 02, the beam The impact hits the surface of the reflection point 1003, where a reflected beam 1 0 4 is generated. In a special embodiment, it can be noted that the illumination light source is above the nominal reflection surface. In other embodiments, the illumination light source may be unfocused. The reflected beam hits the detector flat 1 0 0 5 is at the nominal detection point 1 0 0 6. If the reflecting surface is displaced to another position, such as 1 07, a displacement reflection point 1 008, a displacement reflection light 1 0 0 9 and a displacement will be caused. Detection point 1 〇1 〇. As the surface moves downwards, the detection point 1 0 1 0 is moved from the nominal detection point 1 0 6 to the detection position 'f 1. The detector can be a two-cell Detector, a position sensing detector (PSD), or a CCD array, or other spatial detection optical detector. The autofocus system can adjust the distance of the sensor combination on the surface so that f 1 is Zero. — ^ — J ------ Equipment ----- ^ ---- Order --------- (Please read the notes on the back before filling this page) Intellectual Property of the Ministry of Economic Affairs The paper size printed by the Bureau ’s Consumer Cooperatives applies the Chinese National Standard (CNS) A4 specification (210 X 297 public love) “27 _ Printed by the Employees’ Cooperatives of the Intellectual Property Bureau of the Ministry of Economy 493205 A7 _____ B7 V. Description of the invention (25) 1 i illustrates the tilt sensitivity of the asymmetric spatial autofocus method. Figure 1 1 includes: oblique illumination beam 1 100, nominal reflecting surface 1 1 2 0, reflection 1 1〇3, reflected beam 1 104, detection plane 1 105, nominal detection point 1 106, curved plane 1 1 07, displacement reflected light 1 1 2 2 and displacement detection point 1 1 2 3 When the surface 1 1 20 generates an oblique reflected beam 1 1 22, its position 1 1 2 3 on the detector plane is caused by the displacement f 2 from the nominal detection point 1 1 0 6 due to the inclination. In the embodiment, the distance from the hole to the sample is substantially corrected, however, the autofocus can bring the hole to a different height to compensate for the displacement amount f 2. The new height may have errors due to tilt in the optical system. FIG. 12 shows an embodiment of the auto-focusing system of the present invention, and FIG. 12 includes a combination 1 2 9 3, a lamp 1 2 9 1, a spherical mirror 1 2 50, a detector plane 1 2 05, and a lens 1 29 5 , Beam splitter 1 294, reflected beam 1 204, beam 1 200, measuring surface 1 20 2 and detector 1 2 9 2. The embodiment shown in Fig. 12 is sensitive to surface displacement, but not sensitive to surface tilt. In the embodiment, the detector 1 2 9 2 and the light bulb 1 2 9 1 are located on the same side of the combination 1 2 9 3. The spherical mirror 1 250 is in the conventional position on the detector plane in the conventional device, and the lens 12'9 5 selectively focuses the light beam within a point of a wafer or an imaging diode. The beam splitter 1 2 9 4 allows the detector to be positioned at the same optical position as the bulb. The mirror 1 2 5 0 series is selected so that its focal length is half of the distance of the nominal reflected beam 1 2 0 4 'because' the imaging surface is located on this paper. The Chinese National Standard (CNS) A4 specification (210 X 297 mm) is applicable. -28--^ --------- install ----- ^ ---- order --------- (Please read the precautions on the back before filling this page) 493205 Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs A7 B7 V. Invention Description (26) on the mirror. The system in Figure i 1 is not bilaterally symmetric, and the light only passes from left to right. The system shown in Figure 12 is left-right symmetrical, and the light can pass from right to left or from left to right. Figure 13 illustrates the increased sensitivity to the amount of surface displacement obtained using symmetrical space autofocus. In FIG. 13, the oblique beam 1 300, the entrance hole 1301, the illumination source 1391, the detector 1392, the focusing plane 1 3 14, the nominal reflection point 1 303, the nominal reflected beam 1 304, the mirror 1320, and the nominal detection point 1305, displacement surface 1307, displacement reflection point 1 3 1 0, target point 1 3 1 5, return reflected beam 1 3 1 3, mirror reflection point 1 3 1 2, imaging point 1 3 1 6, displacement second reflection point 1 3 1 7. Re-reflected beam 1 3 1 8 and detection point 13 19 ° In Figure 1 3, the illumination light source 1 3 9 1 and the detector 1 3 9 2 can allow light to pass through. The oblique light beam 1 3 0 0 passes this detector, but it is practically impossible. However, as a skilled person, this geometrical effect can be achieved with the optical splitter shown in Fig. 13 or by moving the access hole 1 301 and the detector in opposite directions. The mirror has a focal plane 1 3 1 4 and is imaged on itself. The center of the focal plane is located at the nominal reflection point 1 3 0 3 and is orthogonal to the nominal reflected beam 1 3 0 4. For this nominal surface, the mirror 1320 reflects the reflected light 'beam 1 3 0 4 along its path. This back-reflected beam is reflected again from the nominal reflection point 1 3 0 3 of the wafer, and returns to the detector along the same path as the oblique beam 1 3 0 0. The nominal inflection point 1 3 0 5 is located along the path of the oblique beam 1 300. For the displacement surface 1 307, the paper size is applicable to the Chinese National Standard (CNS) A4 specification (210 X 297 mm) -29- i > --------- installation ----- ^- --Order ----- (Please read the note on the back? Matters before filling out this page) Boxing 493205 Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs A7 B7 V. Description of the invention (27) Oblique beam anti_self displacement Reflective point 1 3 1 0. The reflected beam passes through the target point on the focusing plane 1 3 1 4] _ 3 1 5. The returned reflected light beam 1 3 1 3 passes from the mirror reflection point 1 3 1 2 through the imaging point 1 3 1 6 to a displacement of the second reflection point 1 3 1 7. The re-reflected light beam 1 3 1 8 hits the detector at the detection point 1 3 1 9, which is at a distance from the nominal detection point f 3. The detection distance f 3 is greater than four times the detection distance f 1 of a similar autofocus system. When (increased, the angle between light rays such as 1 3 1 3 and 1 3 0 4 will increase and the sensitivity will increase. Because the sensitivity of the symmetrical system in Figure 13 is greater than or equal to four times the equivalent asymmetric system. Figure 14 indicates the insensitivity of the symmetrical spatial autofocus system. Figure 14 includes: surface 1 4 0 2, inclined surface 1 4 2 0, oblique light 1 4 0 0, reflected light 1 4 0 4, reflected light 1 4 2 1. Mirror 142 2 and detector plane 1 426. The surface 1 4 2 0 in Figure 14 is located at the correct distance from the sensor combination, but is inclined at an angle 4. The reflected light reflected from the mirror 1 4 2 2 1 4 2 1 is connected to I. It is inclined at the nominal reflected light, but is imaged on the mirror itself, because the reflected light hits the detector at the nominal point 1 4 0 3 of the detection plane. Yes, with this symmetrical system, there will be no error shift due to tilt. The advantage of the SSA method is that it can be structured in a direct way to generate a 'linear error signal, and the autofocus position is zero traverse. Differential signal from any line imaging detector (CCD, PSD or bi-cell) can be used to provide The error signal is returned to the correct focal length. The operating range of the SSA is determined by the f / # of the optical element, the size of the imaging detector, and the paper size applicable to China National Standard (CNS) A4 (210 X 297 mm) -30 -| > — " · ------ · 装 —l · —Order --------- (Please read the notes on the back before filling this page)

493205 V. Description of the invention (28) Determined by the inclination of the surface. FIG. 15 shows another embodiment of the autofocus system. FIG. 5 includes a lens 1500 and a plane mirror 1501. The embodiment of FIG. 15 includes a lens 15 0 0 and a plane mirror 15 0 1, which are replaced by the spherical mirror (see FIG. 14) in other embodiments. The focal length of the mirror is equal to the distance from the lens center to the nominal reflection point. The lens-to-mirror lens is the focal length of the lens. If the sensitivity is reduced, the coefficient is approximately 2 compared to the spherical mirror system. Figure 16 illustrates the insensitivity of the embodiment of Figure 15 to surface tilt. A lens 1600 and a plane mirror 1601 are shown in FIG. 16. In the above description and drawings, the light beam is represented by a line. In some embodiments, the light can be focused, such as a lens focused on the water surface. When the light beam is focused on reflection, the symmetrical system roughly refocuses the light beam on the re-reflection point on the surface. The present invention is also related to wafer holding. In many implementations, the wafer holder receives a wafer from a robotic arm, and the device is configured to make the side of the wafer downward. In a preferred embodiment of the present invention, the wafer holder holds the wafer in a wafer ring supported by a wafer supporter, as shown in FIG. 17. The wafer ring in the embodiment has at least three positions relative to the chuck. In the first position shown in FIG. 18, the wafer ring is far away from the chuck 'and the robot arm can move the wafer to the wafer ring. As shown in the second position 'shown in FIG. 19, the wafer ring pulls the wafer up from the chuck', thereby creating a vacuum to hold the chuck on the wafer. In the third position shown in Figure 20, the wafer ring system is slightly separated from the chuck, thereby: 1) it cannot contact the wafer. The paper size is applicable to the Chinese National Standard (CNS) A4 specification (210 X 297 mm). ) _ 31-(Please read the precautions on the back before filling out this page ----- l · — — — Order II ------, Printed by the Intellectual Property Bureau of the Ministry of Economic Affairs and Consumer Cooperatives 493205 Intellectual Property of the Ministry of Economic Affairs A7 B7 printed by the Bureau ’s Consumer Cooperative. V. Invention Description (29) 2) It is close to the chuck, which can catch the wafer when the vacuum holding state is broken. In this position, the wafer can be rotated on the chuck while maintaining safety. The wafer holder shown in Figs. 17-20 has three different advantages compared to the conventional one. In the conventional technique, a suction cup on a rod is used to pick up a wafer and lower it to a measurement position, which is docked at various points along its edge, or slightly bent by the downward force of the rod. Compared with the present invention, a person skilled in the art cannot achieve the following _ 1) Orienting the wafer 2) When the vacuum is removed, the wafer can be dropped to a significant distance 3) Cannot hold the chuck to hold the crystal in the chuck There are several ways to circle. The wafer-based instruments are oriented with the device side down. It is only necessary to make contact in a unique area around the wafer, such as a 3 mm ring-shaped peripheral portion of the wafer periphery. There is still significant uncertainty about where the robotic arm will set the wafer's off-hand position. At the same time, in order not to obscure the line of sight from below the wafer, the instrument can see the surface of the wafer as much as possible. In the embodiment of the present invention, FIGS. 21 and 22 illustrate two methods that meet the above-mentioned restrictions. In FIG. 21, the wafer supporter is abutted against the wafer ring to hold the wafer, and the top surface of the wafer supporter is inclined, so that the wafer supporter only contacts the crystal regardless of the position of the wafer. Rounded edges. As shown in the embodiment, at least three locations surrounding the wafer are used for support. Each crystal circle support is significantly narrower (vertical to the direction of the drawing) to minimize blurring. In Figure 21, the center of the wafer is coaxial with the wafer ring. Because of the uncertainty of the off-hand position, the wafer can be left or right. The crystal paper size is in accordance with Chinese National Standard (CNS) A4 (210 X 297 mm) -32----- 1 _-- I ------- l · III ^ --- I --- -I (Please read the precautions on the back before filling this page) 493205 Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs A7 B7 V. Description of the invention (30) The length of the brace waste is determined by the position of the hand Certainty. In another embodiment shown in FIG. 22, the wafer support has two inclined portions, and the outer inclined portions bring the wafer closer to the center without having to consider the off-hand position of the wafer. In another embodiment, the edges of the wafer are grasped rather than held by a holder or a vacuum chuck. The gripper can selectively rotate the entire wafer to align one or more axes. In a preferred embodiment, the wafer held by the chuck is lowered into a water bath for optical measurement. If the wafer is dropped directly into the water, the generated air bubbles will be attached to the surface, which will adversely affect the image recognition and thickness measurement. In order to reduce the generation of air bubbles, the wafer is lowered into the water in a way that is not parallel to the water surface during operation. In the conventional technology, it is avoided by using a rod and a convex transport. In the present invention, a connecting rod is added in FIGS. 2 to 25, which has two supporting pivots, which are respectively located on both sides of the chuck assembly, and are used to support the weight of the chuck assembly and drive it to move up and down. . The chuck is located at the bottom of the chuck assembly, and the movement of the chuck housing is further restricted by the combination of a stationary tilting anchor, a connecting rod and a tilting arm. These three elements are approximately coplanar and perpendicular, and bisect the connection by the two support pivots. The link is pivotally connected between the tilt anchor and the tilt arm. In the embodiment shown in FIG. 23, the support pivot is located at the highest position. 'When the support pivot is lowered to the position shown in FIG. 24, the chuck housing is pulled down to the left end of the link, and the connection The lever is turned to a further horizontal position. When this is the case, it pushes the tilt arm to the right, and turns the chuck housing in a clockwise direction in turn in the above steps. In this way, the wafer (the size of the paper is applicable to the Chinese National Standard (CNS) A4 specification (210 X 297 mm) -33-—u--Λ ------ packed ----- ^- --- Order --------- ^ 9. (Please read the notes on the back before filling out this page> 493205 Α7 V. Description of the invention (31) * # 'It is at the bottom of the chuck combination ) Is fixed at a non-horizontal angle (please read the precautions on the back before filling in this page > degree when it is immersed in the water surface. When the chuck combination is more lowered, the water level will rise. In this implementation In the example, when the supporting pivot is further lowered, the link rotates above the horizontal plane of gravity S, and the inclined arm is pulled to the left, so that the chuck housing rotates counterclockwise. Finally, the supporting pivot is turned to the bottom The chuck housing is approximately horizontal by the action of the support pivot and connecting rod. The rotation of the chuck housing and its precise position depend on the difficulty of stopping the movement when the support pivot travels to the bottom. It can be adjusted to The wafer is moved parallel to the X and Y of the optical system under the support plane. The support plane is provided with a main window, which enables the optical system See the wafer. In the embodiment, a sensor is used to control the height of the water surface. The low-height sensor is used to confirm that there is sufficient water in the system during operation, and the high-height sensor is used to prevent Water Overflow. Printed by the Consumer Cooperative of the Intellectual Property Bureau of the Ministry of Economic Affairs. In other embodiments, wafers can be placed flat in water, which can make bubbles extremely trapped. There are many devices available to remove these Bubbles, including jets of water flowing around the wafer (and the window), compressed waterjets (such as wipers, high-speed jets of water), and ultrasound. It is possible to identify the image of the bubble in imaging or spectral measurement analysis, and Controls the removal of air bubbles from the substrate. The wafer is not cleaned after chemical mechanical polishing, and there is still polishing slurry on it. Over time, the polishing slurry will make the optical components between the window and the wafer dirty. Airborne Particles will also cause the window to become dirty. Several technologies, used alone or together, can be used to clean the main window. The preferred methods include: This paper size applies the Chinese National Standard (CNS) A4 specification (210 X 297 mm) -34-493205 A7 B7 printed by the Consumer Cooperative of the Intellectual Property Bureau of the Ministry of Economic Affairs 5. Description of the invention (32) 1) Diversion water 2) Wet the window completely with detergent (such as unit alcohol) 3) Cover the window completely with water-absorbent cloth. 4) Pull the cloth from the center of the wafer and wipe the cloth perimeter across the wafer. When the main window and the chuck are at the highest position, a small gap is created between them. Operations can be difficult to achieve. An arm system is designed to place the fabric and draw it away from the center, and an ultrasonic cleaner can help remove particles from the surface of the main window. Once the particles are removed, a continuous stream of water is sufficient to keep the window clean. Scrubbing is also necessary to remove particles. A rubber cleaner (such as a car wiper) can remove particles more effectively than the previous method of removing particles with cloth. Placing a tight restriction on the size of the instruments in the CMP (or other) tool and the moving parts of the optical system will make it difficult to scan the entire wafer. In addition, a barrier between water and the optical system may be formed, such as the wafer holder abutting against the wafer ring. In these cases, another advantage of the present invention resides in the fact that the optical element can absolutely "see" any point on the wafer, as shown in the embodiment shown in FIG. In the embodiment shown in FIG. 26, it is viewed from above the wafer. The wafer is on the support, which is held by the wafer ring. As described above, the vacuum clip 'head supports the wafer, and the holder does not touch the wafer at this time. The scanning area of the optical element does not cover the entire area of the wafer. The target is blurred at its starting position because it falls outside the scanning area and above the wafer support. Rotate the wafer to make the target to the paper size. Applicable to China National Standard (CNS) A4 specification (210 X 297 mm) -35---1 --------- pack ----- ^ ---- Order --------- (Please read the notes on the back before filling out this page) 493205 Printed by the Consumers' Cooperative of the Intellectual Property Bureau of the Ministry of Economic Affairs A7 B7 V. Description of invention (33) Turn target Position so that the optics can see the target. This is very important, especially for scanning thickness along the wafer diameter in order to measure the uniformity of radial thickness in the CMP process. The present invention also utilizes a vacuum chuck to level the wafer. It is known that the wafer ring is held on the circumference of the wafer, and the center of the wafer is slightly depressed by gravity or a rod. In the foregoing example, the curvature of the wafer was caused by a different, highly repeatable unevenness. The present invention uses a vacuum chuck to flatten the wafer system for easy focusing, and can reduce the effects caused by different optical path lengths (including through the wafer) and tilt. The requirement of wafer flatness is due to two reasons. In order to avoid the tilt of the wafer surface with respect to the optical beam, such as thickness measurement, the reflection coefficient of the wafer surface is a function of angle. Measurement error. The second reason is related to the focal point. When the optical path passes through water, it is changed due to the unevenness of the crystal circle, which will reduce the focusing ability of the optical system due to the difference. The third reason is related to autofocus. No auto-focusing system will measure the beam with the same effective incident angle, and there will be errors in the focus and the focus position of the measuring beam between the systems. This is because the change in the optical distance in the air compensates for the change in the optical distance in the water and the refraction of the light beam to focus the objective lens. A lens autofocus system is another example of the autofocus system described above. ’This device can transmit and receive the auto-focused light passing through a light splitter, which is a rotating mirror in the system. The beam can be emitted along this collimated measurement or monitoring beam, and finally reaches the wafer for reflection, and returns along a collimated beam (original beam or another beam). And the method used in the CD This paper size applies to the Chinese National Standard (CNS) A4 specification (210 X 297 mm) -36- --r --------- install ----- ^- --- Order --------- φ (Please read the notes on the back before filling out this page) 493205 Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs A7 B7 V. Description of Invention (34) This use. A focused astigmatism is irradiated in one quarter of a grid, and even four grids are set to N, S, E, and W as direction points. If the light beam is scattered in one direction, it will be scattered in the N and S directions and concentrated in the E and W directions. If the light beam diverges in the other direction, it will be scattered in the E and W directions and concentrated in the N and S directions. The scattering and condensing in the quadrant can be detected electronically and used to control the focus of the measurement system. In addition, the wafer alignment device has three different advantages: 1) During the training of the operator, one side of the wafer can always be seen standing up. If the direction with dents is toward the bottom of the screen, System training is easier. 2) The random positioning of the wafer makes image recognition more difficult. The better the initial alignment, the easier the image recognition. 3) For a pair of positive wafers, the hole can have a square cross-section (perpendicular to the measurement or monitoring beam), and the box used for measurement is also square, which can make more light without increasing the size of the smallest box. And can be used for measurement. The present invention relates to another kind of software joystick, and the problem to be solved is how to make the operator of the system software use a motor to control the camera to move to a characteristic fixed point. In a preferred embodiment of the present invention, the user will not need to contact the motor directly to move the camera. The camera's viewing area displayed in the software is less than 0.01% covered in the camera's full motion area. The user needs to move the camera to see the other viewing areas. This problem can be solved by using a device which actually has a joystick and is connected to one or more electric converters. The user can move the joystick position, and the displacement is converted into the relative speed of the motor to move the camera. And the paper size of the camera is applicable to the Chinese National Standard (CNS) A4 specification (210 X 297 mm) 37 --k --- two ------ install ----- ^ ---- order --------- (Please read the notes on the back before filling this page) 493205 Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs A7 B7 V. Invention Description (35) The view area is constantly updated to Allows users to see the area where the camera is located relatively. This type of device is called a joystick. It is a manual device that can easily enable a user to control the position of the camera with the joystick device. More specifically, it is a multi-dimensional joystick. The text below illustrates a standard control software called a slider. The software is controlled by a graphical user interface that is similar to a sliding potentiometer. The user can control the relevant data in the software by the mouse to move the position of the slider on the screen. This control usually has a pair of arrows. When the user selects an arrow, the number of the slider is in the direction of the arrow. Change a certain amount. As in the embodiment of the present invention, it may not have such rocker hardware. Among other software systems, the joystick separate hardware converter is usually replaced by control software. For example, the two-dimensional joystick may be replaced by two sliders, and it may also be replaced by a two-dimensional joystick with four-directional arrows pointing at different 90 degrees. In these systems, each time the user selects one of the four arrows, the camera moves a certain amount of displacement in the direction indicated. In this system, the user can control the movement selected by the platform's moving slider. The amount, or the amount of movement of the text input type. The four-arrow button can be replaced with or added to the four-arrow key on the keyboard. This software solution makes it difficult for the user to move along the axis of the motor, and it is difficult to change it in accordance with the speed of the movement. According to the present invention, the soft system is used to maintain non-axis movement and to easily control the speed of the rocker hardware. . The software control device of the present invention may be referred to as a software joystick hereinafter. The software joystick is designed to control the movement of two paper sizes in the form of coordinates. It is applicable to China National Standard (CNS) A4 (210 X 297 mm) -38------- ---- ^ ---- Order --------- (Please read the notes on the back before filling out this page) Printed by the Employees' Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs 493205 A7 ____B7__ V. Description of the Invention 36) Dimensional direction. It can be indicated by a circle on the screen. Selecting and fixing the mouse button in the control area of the software joystick is similar to holding the handle of a hardware joystick. The cursor position is in the control area. It can be used to decide how to control the movement of the camera. Release the mouse button in the controllable area of the software joystick, which is like releasing the handle of a hardware joystick. In an embodiment, the control area of the software joystick It is a square that surrounds the circle. When the mouse button is released and the cursor is in the circle, the control of the software joystick is restored to the center position. When the mouse button is released in the control area and outside the circle , The control of the software joystick will remain the last selected number. To understand its operation It is conceivable that there is a horizontal line and a vertical line through the center of the circle. The two lines form a Cartesian coordinate system and are divided into four quadrants. The software joystick has two controls 一, one for horizontal control (X axis), one It is vertical control (Y-axis). Each number is a single function of the distance from the center of the circle. This graphical user interface control software allows the user to control the monitoring position of the camera in the movement of the multi-axis coordinate system. At the same time, it can easily modify the speed in two axes or a single axis. Embodiments of the present invention include a system for determining film thickness and profile, and are used to determine non-planar features of depressions, paraboloids, or other wafer surfaces. The details of the embodiment are shown in Appendix A. In the preferred embodiment, the profiler can be connected to the optical circuit board in the reflection meter system 100 shown in Fig. 1. The preferred embodiment also requires the use of a curved bearing to Applies to slow and repeated micrometer movements or sub-micrometers _ (Please read the precautions on the back before filling this page)

_ · N 1_ | I ϋ ϋ ϋ--° JI ϋ ϋ n ϋ n i I

493205 A7 B7 V. Description of the Invention (37) Movement. These embodiments are widely applicable to the dielectric or metal structures of a wafer. In another different embodiment, the profile meter can be a sound type or an optical type. The embodiment of the optical profile meter can be used in the above-mentioned autofocus system to determine the profile of the wafer surface. The autofocus system is inherently sensitive to the contour of the wafer surface, because the focus of the light reflected from the wafer surface will be different due to the unevenness of the characteristics of the wafer surface. Another embodiment of the present invention includes a device for high-contrast imaging of a wafer surface. A specific embodiment uses a differential interference contrast (D I C) technology. Polarization technology can be implemented to infer quantitative information about wafer surface measurement using conventional techniques. In a specific embodiment, an integrated interferometer and image spectrometer can be used to determine the contour and material content of the wafer surface at the same time. Other embodiments may include an ellipsometer. The preferred embodiment includes a motion control system, an image recognition system, and software that determines the meaningful amount of measurement data. As described above, the optical system of the present invention can also be used to infer the erosion condition of the wafer surface. It is also suitable for use between wafer proximity metal and dielectric structures. When occurrences or other substances must be included in the data analysis, different embodiments of the present invention can be analyzed using observational calibration or theoretical models or a combination thereof. In the present invention, it is noted that the measurement can be performed at different rates on the wafer surface as described below. The time measurement of data collection, processing, and movement from the measurement area to another measurement area can be adjusted and depends on the required data. Another feature of the present invention is that the precise optical system is positioned on the surface of the wafer. The paper size applies the Chinese National Standard (CNS) A4 specification (210 X 297 mm) -40-(Please read the precautions on the back before filling this page ) ----- ^ ---- Order ------%, printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs 493205 Α7 Β7 V. Invention Description (38) (Please read the notes on the back before filling On this page), the light path board (see Figure 1) is positioned by a DC motor or a guide stud. In the embodiment, the components of the motor are directly mounted on a guide stud (refer to Appendix A). According to the present invention, the connecting element is omitted to achieve a finer, stronger, and alignable driving device. The embodiments of the present invention are integrated into a wafer production line. As described above, different embodiments can make wafers reach different positions under test. A specific embodiment uses a lifter and a feeder element (as detailed in Annex A) to place the wafer at another location and introduce it into the apparatus of the present invention. Fig. 27 is a preferred embodiment integrated into a wafer production line. For the purpose of illustration and not limitation of the present invention, the processing station in the embodiment of Fig. 27 is a polisher. A polishing machine 1 and an integrated surface measurement station are shown in the IS M S 10 series. The polishing machine 1 includes a polishing unit 4, a load-bearing area 18, and a transport system 22. A wafer placed on the carrier 18 is also displayed at the same time. As shown in Figure 27, the measurement station is separate but connected to the processing station. Wafers 16 printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs were taken to and removed from the polishing machine by the carrier 18 through the carrying area 18. The carrier may be a card type or F 0 UP as seen by conventional techniques. The transfer system 22 is a device or a group of devices for transferring wafers to the polisher 1. A specific embodiment includes a robotic arm, such as Equipe W D M — 105. The transfer system can move the wafer to any of the carriers 18, the polishing unit 14, or the ISMS 10. The foregoing embodiments of the present invention are described herein to limit the present invention. Many modifications or equivalent configurations are obvious. This paper size applies to China National Standard (CNS) A4 (210 X 297 mm) -41-

Claims (1)

493205 Attachment 1 (A): No. 89105382 Patent Application Chinese Amendment to Patent Scope Amendments Members are requested to indicate ^^ year / ^ month ¾ referenced ^ 44 ** 4 变 erifContents # 隹 r: ^ [L -. Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs / & Year / ^ Month Pollution Amendment / Correction Factory: '1 《: __ December 1990. 6. Scope of Patent Application 1. A surface measuring device, including: ( (Please read the precautions on the back before filling out this page) A measurement unit that receives information from a measurement area on a surface; and a first camera with a first field of view that includes the measurement area. 2. The surface measurement device as described in item 1 of the scope of patent application, further comprising at least one controllable translation platform connected to the measurement unit to change the position of the measurement area on the surface. 3. The surface measurement device according to item 2 of the scope of the patent application, wherein the at least one translation platform is a direct-drive translation platform. 4. The surface measurement device according to item 1 of the patent application scope, further comprising a rotatable chuck connected to the surface. 5. The surface measuring device according to item 1 of the patent application scope, wherein the surface and the measuring unit have a relative free movement of four degrees. 6. The surface measuring device according to item 1 of the scope of patent application, further comprising a second video camera having a second field of view. 7. The surface measurement device according to item 6 of the scope of patent application, wherein the second field of view is smaller than the first field of view. 8. The surface measuring device according to item 7 of the scope of patent application, wherein the second field of view includes the measuring area. 9. The surface measuring device according to item 1 of the scope of patent application, wherein the measuring unit is a film thickness measuring unit. 10. The surface measurement device according to item 1 of the scope of patent application, wherein the measurement unit is a surface profile measurement unit. This paper size applies to Chinese National Standard (CNS) A4 specifications (210 father 297 mm 1 ~ 493 205 printed by the Intellectual Property Bureau Employee Consumer Cooperatives of the Ministry of Economic Affairs A8 Β8 C8 D8 VI. Patent application scope 1 1 The surface measuring device described in the above item, wherein δHai retest unit is a bridge circle instrument. 1 2. The surface measuring device according to item 1 of the scope of patent application, wherein the surface is wet with liquid. 1 3. A semiconductor processing apparatus comprising: a wafer processing station; and a measurement station separated from but connected to the wafer processing station, wherein the measurement station includes a measurement area where an ultraviolet light source is irradiated on the surface. 1 4 · The semiconductor buried device described in item 13 of the scope of patent application, wherein the measurement area is wetted by a liquid. 1 5 · The semiconductor processing device described in item 13 of the scope of patent application 'wherein The measurement unit includes an optical element including a curved reflective surface. A semiconductor processing device includes: a wafer processing station; and a measurement station separated but connected to the wafer processing station. The measurement station includes a measurement area where an ultraviolet light source is irradiated on the surface, and at least one spectrometer is optically connected to the measurement area on the surface. 1 7 · The semiconductor described in item 16 of the scope of patent application Processing device 'wherein the measurement area is wet with liquid. 18 · —A semiconductor processing device including: a wafer processing station; and a measurement station separated but connected to the wafer processing station, where the measurement The station includes a wafer chuck for rotating the wafer relative to the measurement station. The paper size applies the Chinese National Standard (CNS) A4 specification (210 × 297 mm). (Please read the precautions on the back before filling (This page) 2- 493205 A8 B8 C8 D8 VI. Patent application scope 19. The semiconductor processing device described in item 18 of the patent application scope, wherein the measurement area is wet with liquid. 2 0. A surface reflectometer, including a light source, an object-side optical instrument that can be translated relative to a wafer surface, and at least one photodetector. 2 1. Surface reflection as described in item 20 of the scope of patent application The measurement area on the surface of the wafer is wet with liquid. (Please read the notes on the back before filling this page) Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs -3- This paper size applies to Chinese national standards (CNS) A4 specification (210X297 mm)