TW200540579A - Pre-measurement processing method, exposure system and substrate processing equipment - Google Patents

Abstract

The invention is provided to highly efficiently manufacture high-performance and high-quality micro devices at a high throughput. Prior to carrying in a wafer W to exposure equipment (200) which is to expose the wafer W, a mark formed on the wafer W is measured by inline measuring equipment (400), and the exposure equipment (200) is informed of the measurement results and/or results obtained by calculating the measurement results. The exposure equipment (200) optimizes measuring conditions based on the informed results, and then perform processes such as alignment.

Description

200540579 IX. Description of the invention: [Technical field to which the invention belongs] Ben Maoming, "Related to the high-precision and high-level lithography steps used in the manufacture of semiconductor elements, liquid crystal display elements, photographic elements, thin-film magnetic heads, etc. Pre-measurement processing method, exposure system, and substrate processing device for forming circuit patterns based on production capacity. [Prior technology] Various 7L semiconductor devices, liquid crystal display elements, photographic elements (CCD: Charge C + upled DeVlce), thin-film magnetic heads, and other components. It is manufactured by exposing multiple layers of patterns on the substrate using an exposure device. Therefore, when exposing the patterns of the second layer and later on the substrate, it is necessary to expose the patterned irradiation areas and photomasks on the substrate. The pattern image alignment 'must be equal to the alignment (alignment) of the plate and the reticle. Therefore, the pattern-exposed substrate on the stage coordinate 2: the κ layer is attached to each irradiation area (day Japanese film image 庶 ·} · 々, 丄. 〃 5 shapes, respectively forming registration marks (When the alignment bar will be formed with a pair # exposure F set and the board is brought into the exposure device, borrow The position of the mark (the value of the seatpost: to measure the # 记 之 &#; τ on the platform coordinate system) by the reader and the reader. Eight times, according to the measured mark position and your nCi The upper Α 亜 and the reticle are shown on the picture, and are aligned (alignment (positioning) of an irradiation area port on the substrate). Align the i ~ ^ amount of the alignment mark " 来 Ithough It is known that there are various irradiation areas on the substrate. From the perspective of improving the productivity / alignment of the grain-to-grain (D / D) alignment, the current point of view, for example, ... —44429 6 As disclosed in 200540579, Japanese Patent Application Publication No. 62_84516, etc., the regularity of the EGA (Enhanced Global Alignment) has been accurately specified by statistical methods on the substrate. Go mainstream.

The so-called EGA 'means that the pointer knows and shoots a plurality of samples (for example, about 7 to 15) selected in advance to measure the position of the alignment mark' so that the measured values and the design of the alignment mark In order to minimize the position error, statistical calculations are performed using a least square method or the like, and after calculating the position coordinates (irradiation arrangement) of all the irradiation areas on the substrate, the substrate stage is performed in accordance with the calculated irradiation arrangement. By & m, the main linear errors (remaining rotation errors of the substrate, the orthogonality errors of the stage coordinate system (or '' ', radiation arrangement), and the linear expansion and contraction of the substrate (s⑶ "called ), Substrate (center position) offset (parallel movement), etc. In addition, non-linear deformation of the substrate due to processing such as grinding or thermal expansion, and error in the stage coordinate between exposure devices (between stage coordinate systems) = Poor), the substrate's adsorption state #, and a non-linear irradiation arrangement: Poor. The technique used to remove such non-linear errors (random errors) is ^ GCM (Grid Compensation Matching: Compensation Matching). Mingshi y … Sign juice, day, day, week, day, week, week, day, week, week, week, week, week, week, week, week, week, week, week, week, week, week, day, week, etc.) Based on the results of EGA, take # EGA measurement again, take out the non-linear components. The extracted non-linear component is equal to your value; keep it, and in the subsequent exposure procedures, use this transformation to correct the value, cool value; enter the correction of the irradiation position (for example, Ling Zhao Japanese Patent Laid-Open No. 2001-3452 No. A) It is different from the exposure program. 7 200540579 Before each exposure condition and processing, the reference wafer is used to measure the non-linear component (the offset of each irradiation). This offset is stored as a conversion correction file in advance. In the exposure program, a correction correction block corresponding to the exposure conditions is used to correct each irradiation position (for example, refer to Japanese Patent Application Laid-Open No. 2002-3531 21). The applicant of this case is based on a predetermined evaluation function. To evaluate the difference (non-linear error component) between the position of the irradiation arrangement after removing the linear error component by the above-mentioned EGA method (the non-linear error component), and determine the function to represent the nonlinear component according to the evaluation result. According to this function, To correct the irradiation arrangement, it is in the application (Japanese Patent Application No. 2003-49421). Moreover, in order to improve the superposition accuracy of the circuit pattern, it is known that the projection optical system of the exposure device used in the previous step exposure is measured in advance. The deformation caused by it is registered in the database as deformation data in advance. From this, the deformation data and the exposure history of the substrate, The same image distortion as the image distortion according to the previous step is generated in the same way as the exposure device used in the next step exposure. The imaging characteristics of the projection optical system of the next step exposure device are adjusted in batch units. Matching (SDM: Super Distortion Matching) (for example, refer to Japanese Patent Application Laid-Open No. 2000-451, and Japanese Patent Application Laid-Open No. 2001-338860, etc.). In addition, the following technologies have also been proposed for focusing adjustment technology On the surface of the substrate on which the element is formed, there is a step difference due to the circuit pattern and the like formed in the previous step, so a surface shape measuring device attached to the exposure device for measuring the surface shape of the substrate is used to measure the surface shape of the substrate in the exposure process to find the most Correct the focusing position based on this position (eg, according to 8 200540579 JP 2002-4321 7 f Lu Ba Ru, ~ A newspaper) ° Also, adjust the focusing position of the projection optical system that becomes the exposure device In terms of the reference optimal focus position determination technology, there are also multiple numbers along the direction of the optical axis of the projection light. Position, a test pattern exposed wells salmon, Jie onto the test substrate, the check, the developing 'after the most finely resolved pattern of poly ~ as the best focus by the focusing position. As mentioned above, for substrates that are moved in and out of light, before the exposure process is performed, the mark position is measured. ^ Various materials related to the surface shape and other substrates.

σίΙ ‘According to this theory’, calculate the appropriate value — use this correction value to apply the positioning of the substrate to perform warming ^ * First, it will form a high-precision circuit pattern on the substrate. (Patent Document 1) Japanese Patent Laid-Open No. 6-4444429 (Patent Document 2) Japanese Patent Laid-Open No. 62-8451 6 (Patent Document 3) Japanese Patent Laid-Open No. 2001-345243 (Patent Document 4) Japanese Patent Laid-Open 2002-353121 (Patent Document 5) Japanese Patent Laid-Open Publication No. 2000-36451 (Patent Literature 6) Japanese Patent Laid-Open Publication No. 200 1-338860 (Patent Literature 7) Japanese Patent Laid-Open Publication No. 2000-43221 [ SUMMARY OF THE INVENTION However, according to the above-mentioned conventional technology, measurement of various information of a mark position or a surface-shaped substrate is performed on a substrate carried into an exposure device before the exposure process is performed. Therefore, for example, the mark is deformed or ^ cannot be measured with high accuracy. In some cases, there is no guarantee; other problems, and the interruption of the exposure process due to misalignment or the need to quantify the amount of other monthly data, leading to a reduction in production capacity (re-measurement per unit time). 9 200540579 questions. In particular, the above-mentioned EGA, GCM, SDM, etc., in order to perform complex arithmetic processing, it takes a certain amount of time before the solution (correction coefficient) is calculated. During this period, the substrate exposure process must be waited, so the correction value is calculated. It must be performed per batch unit or processing unit, and the optimal correction cannot be performed for each substrate or each irradiation. ^ 'In addition, when an abnormality occurs in the previous step and the precision required to form the pattern of the substrate cannot be formed, it is useless to perform the next exposure step', so it is necessary to prevent this useless operation efficiently. > The present invention is made in view of the problems of such a conventional technology, and an object thereof is to be able to manufacture high-performance, high-quality micro-devices and the like with high productivity and high efficiency. According to the first aspect of the present invention, a pre-measurement processing method is provided, including: a pre-measurement step (S21), measuring a mark formed on the substrate before carrying the substrate into an exposure device (for exposing the substrate); and notification Step (S22), the relevant waveform material of the mark measured in the previous measurement step is notified to the exposure device, the analysis and disposition set separately from the exposure device, and to manage at least one of these devices, the location is more than that. The waiting device is at least one device among the upper management devices. Here, the so-called "waveform material" refers to a measurement signal (original waveform data) output from a detection sensor such as a CCD provided in a measurement device used when measuring a mark, or a certain signal is applied to the measurement signal. (Predetermined) Signals that are processed (for example, electrical filtering, etc.) and have substantially the same content as the measurement signal (in terms of measurement results, information that essentially becomes the same result). That is, in this specification, the so-called "waveform data" includes not only the "original waveform data" output from the detection sensor: 200540579, but also the "original waveform data", which is subjected to the "processing" described above. "Waveform data" concept. X ′ also includes image data in the original waveform data (for example, when the two-dimensional measurement mark is two-dimensional image data). In addition, the above-mentioned predetermined processing includes compression processing, extended interval processing, smoothing processing, and the like. In the present invention, since the substrate mark is moved into the exposure device before the advancement is measured, for example, when the mark is formally or N-shaped by the exposure device, the mark deformation or mark breakage is not ruled out. Mark, f matter, implementation of statistical calculation processing, etc., identify the combination of marks with small errors, and worse, when the formal measurement with the exposure device, you can choose the best standard or the most private measurement conditions. Therefore, due to the misalignment of the exposure device, the re-measurement or processing interruption of the mark formed by Wu is reduced, so that a full-scale measurement can ensure sufficient alignment accuracy. Zhuang = After the mark is measured in the pre-measurement step, before the substrate is brought into the exposure process, it takes a certain amount of time before the exposure process can be performed. Therefore, it can be completed in advance based on the measurement results measured in advance. Various complex = statistical calculation processing, which can omit the measurement and recording of the statistical device and the statistical calculation processing used by the exposure device. Thereby, after the substrate is brought into the exposure apparatus, exposure processing can be performed in advance, and optimum position correction can be performed on each substrate or each irradiation. Also, in order to notify the waveform data, for example, the characteristic difference between the measurement device that uses the previous measurement step for different measurements and the measurement device that the exposure device uses for formal measurement (due to the sensor, imaging optical system, and illumination optical system 11 200540579 Differences in characteristics caused by differences such as differences; Differences in characteristics caused by differences in these environmental changes or long-term changes; and Differences in characteristics caused by differences in signal processing algorithms, etc.) The two are matched in a positive way so that the measurement results of both can be evaluated on the same basis. In the pre-measurement processing method of the first aspect of the present invention, there is further provided an evaluation step (S22), which evaluates the pre-measurement = data measured by the step according to a predetermined evaluation standard; the notification step ㈣ according to the evaluation step: For the evaluation result, you can choose to notify or prohibit the notification of the waveform data. In this month, the notification step can also open / notify the parity result without notification of the waveform data. Of course, all waveform data can also be notified, but generally there is a large amount of data, so from the viewpoint of communication burden, etc., all notifications are not good. Therefore, it is sometimes possible to omit the notification and reduce the communication burden. According to the second aspect of the present invention, the system provides a pre-measurement processing method, including: a pre-measurement step (S21), measuring a mark formed on the substrate before carrying the substrate into an exposure setting (for exposing the substrate); The evaluation step (S22) evaluates the prior measurement step = the mark of the measurement according to the predetermined evaluation standard; and the notification step (S23), which notifies the exposure of the evaluation result obtained by the evaluation step or evaluation-related information. Device, an analysis device provided separately from the gas exposure device, and at least one device located in a higher management device than the devices in order to manage at least one of these devices ^. Since the present invention performs pre-measurement before the substrate mark is carried into the exposure apparatus, the same as the pre-measurement process of the first aspect of the present invention described above, during the formal measurement of the exposure apparatus, fewer alignment errors occur, & Month b 12 200540579, current production capacity and ensuring sufficient material accuracy, and various computational processing are also introduced 仃 'II This can quickly expose the substrate carried into the exposure device 2 processing This raises the production capacity and on each substrate or each The best place for irradiation is not the waveform data like the above. For example, since the measurement result of the unmarked position is notified, the amount of transferred data is also reduced, and the load is small. ° Negatively according to the third aspect of the present invention, a pre-measurement method is provided, which includes a pre-measurement step (S41), and measures a plurality of numbers formed on the substrate before carrying the substrate into an exposure device (for exposing the substrate). And the correction information calculation step (with 2 ~ 349, S36, S37), based on the measurement result measured in the previous measurement step, calculate the correction information (including the linear correction coefficient from which the design position error from the mark is minimized) And nonlinear correction factors). Since the present invention calculates the correction coefficient according to the measurement results measured beforehand, in the exposure device, the calculated correction information can be used to quickly locate and perform the edge processing of the edge substrate picked up, thereby improving the productivity and Optimal position correction is performed on each substrate or each irradiation. According to a fourth aspect of the present invention, a pre-measurement processing method is provided, which includes a pre-measurement step (S61), and a plurality of 'test 1' are formed on the substrate before the substrate is brought into an exposure device (the substrate is exposed by the last exposure). Mark position; image distortion calculation step (S62 ~ S67 in S55A), based on the measurement result measured in the previous measurement step, calculate the image distortion of the projection optical system of another exposure device that has exposed the substrate; and calculate the correction information Step (S 5 5 B, S 5 5 C), calculate the image distortion information based on the image distortion calculated in the image distortion calculation step 13 200540579, and calculate the image distortion information of the projection optical system provided in the exposure device in advance. Image distortion correction information (used to deform the image generated by the other exposure device in the exposure device). In the present invention, since the image distortion and the image distortion correction information generated in the step w are based on the measurement results measured in advance, the calculated image distortion correction information is used in the exposure device of the next step. Change the imaging characteristics of the projection optical system, etc.

The exposure process can improve the productivity and implement the best Z-image distortion correction on each substrate or each irradiation. According to a fifth aspect of the present invention, a pre-measurement processing method is provided. The method includes the following steps: measuring the uranium of the substrate into an exposure device (for exposing the substrate), and measuring a phase-shifting focus mark formed on the substrate And a focus correction information calculation step, based on the measurement result measured in the previous measurement step, to obtain a focus error when the exposure is performed by another exposure device that has exposed the substrate, to calculate the exposure of the substrate with the exposure device Hour = Focus correction information used. Since the present invention measures the phase-shifted focus zenith formed on the substrate in advance, and calculates the focus correction information based on the measurement results, in the exposure device of the next + step, the calculated focus correction information is used to perform the maximum = 'Focus adjustment' can quickly carry the exposed substrate into the exposure process, so the external production capacity and the best focus correction on each substrate or each irradiation. According to a sixth aspect of the present invention, a pre-measurement processing method is provided, which includes a pre-measurement step (S74), and measures the surface shape of the substrate before carrying the substrate into an exposure device (for exposing the substrate); and repairing the substrate; The calculation step (S76) due to 14 200540579 indicates that the focus correction information used when the substrate is exposed by the exposure device differs from the measurement result measured in the previous measurement step. Since the present invention measures the surface shape of the substrate in advance, and calculates the focus correction information based on the measurement results, in the exposure device in the next step, the calculated focus correction information is used to perform the optimal focus adjustment, which can quickly carry in the Since the substrate is subjected to exposure processing, it is possible to increase the productivity and implement the best focus correction on each substrate or each irradiation. According to a seventh aspect of the present invention, a pre-measurement processing method is provided, including: a pre-measurement step of measuring a plurality of mark positions formed on the substrate before carrying the substrate into an exposure device (for exposing the substrate); temperature A measurement step for measuring the measurement device used in the measurement in the prior measurement step < δHari measurement device to transfer the substrate to the transport device of the exposure device, and at least one of the exposure device Temperature change in the device; a prediction step to predict a change in the position of the mark measured in the previous measurement step based on the temperature change measured in the temperature measurement step; and a correction information calculation step based on a prediction result predicted in the prediction step Calculate the correction information (including the linear correction coefficient and non-linear correction coefficient from which each design position error of the mark is minimized). The present invention is the same as the prior measurement processing method of the third aspect of the present invention. The position of the mark on the substrate is measured in advance. However, if a temperature change occurs during the substrate transportation, the previously measured information is made by the expansion and contraction of the substrate. The actual position of the mark changes < The change of the mark position accompanying the temperature change is theoretically based on the thermal expansion coefficient of the substrate, etc. 15 200540579 or using the test substrate # to measure the relationship between the temperature change and the mark change in advance, or to measure the temperature during the exposure process The relationship between the change and the change in the marker position can be obtained from the materials. Since the present invention predicts a change in the position of a mark accompanied by a temperature change, and calculates correction information based on the position information corrected based on the change, it is possible to implement more accurate position correction. According to an eighth aspect of the present invention, a pre-measurement processing method is provided, including: a pre-measurement step (S21), and measuring a mark position and a mark shape on the substrate before carrying the substrate into an exposure device (for exposing the substrate). , Pattern line width, pattern defect, focus error, surface roughness, at least one of temperature, humidity, and air pressure in another exposure device that has exposed the substrate; and a judgment step (S25, S26, S29), based on the prior measurement The test results in the steps are used to determine whether the substrate should be moved into the exposure device. If the current step is abnormal and the pattern formed on the substrate cannot be formed with the required accuracy, the next exposure step becomes useless. Since the present invention is to measure the marks or patterns on the substrate in advance before carrying the substrate into the exposure device, or to measure the environmental information such as the temperature in the exposure device in the steps before the measurement in advance, when the abnormality actually occurs or the possibility of the abnormality is high It can stop the substrate from being carried into the exposure device, so it can prevent useless processing, and can increase the actual operation rate of the exposure device. According to a ninth aspect of the present invention, there is provided a pre-measurement processing method including the pre-measurement step 'Uranium in which the substrate is brought into an exposure device (for exposing the substrate)', and information on the substrate; and Optimizing step 'According to the operating conditions of the exposure device, optimize the measurement conditions of the previous step 16 200540579. Here, the operation status of the exposure device includes: the case where the operation reference of the exposure device deviates from the established reference, etc., in order to match the implementation status of such implemented corrections, and re-measurement due to measurement errors such as information about the substrate. Status, or exposure processing interruption and stop status of the exposure device. In addition, the measurement conditions include measurement items such as a measurement of a mark position or a measurement of a surface shape of a substrate, a measurement number such as a measured number of marks, a data amount per measurement, and the like, and the measurement condition is preferably such that the exposure is not reduced. The range of production capacity is maximized and optimized. For example, in the exposure device, when correction or retry occurs, it takes only time to delay the exposure process. In other words, even if this part of the increase is measured in advance, it will not adversely affect the throughput of the exposure process. In the pre-measurement step, the more the measurement items, the number of measurements, the amount of data, etc., the more detailed analysis and calculation of the more correct correction value can be performed. Since the present invention optimizes the measurement conditions according to the operating conditions of the exposure device, it does not reduce the throughput of the exposure process, but can analyze and calculate the correct correction value in more detail, and can improve the exposure accuracy. According to the viewpoint 10 of the present invention, “a method for measuring and processing in advance is provided” includes: a procedure for measuring in advance, and measuring the information about the substrate in advance before carrying the substrate into an exposure device (for exposing the substrate); Step 'According to the periodicity of the measurement results measured by the previous measurement step, the measurement conditions of the previous measurement step are optimized. Here, the periodicity includes the batch input cycle, the substrate processing cycle in the batch, and the time of year and month. In addition, the measurement conditions include valid measurement items for analyzing the cause of the abnormality, the number of measurements, and the amount of data for each measurement. 17 200540579 For example, if the previous step of the batch is free of obstacles or abnormalities, it is mostly invested in a cycle. When the period becomes longer 疋 Ai K < 〖Moon shape, in the previous step, the batch speculates that the obstacle or abnormality occurs, and ten pairs of towels are born again. Since the present invention is based on the nature, the measurement conditions of the pre-measurement step are ^ special = because the pre-measurement is performed with effective measurement conditions, so the cause of the obstacle or abnormality can be corrected = special. Cui Quer's second U viewpoint 'is to provide a pre-measurement process to expose the substrate). Prior to the pre-measurement ... ... (using Du Ertu's information about the substrate 丨 and the most soil steps' according to the pre-measurement steps The number of errors is to optimize the measurement conditions of the previous measurement step = of: 'measurement conditions' includes the number of valid measurement items for analyzing the cause of the abnormality, the amount of data of the parent-measurement, etc. In the second step, when errors often occur Therefore, the present invention is based on the number of errors, the prior measurement factor = Article: optimization, more specifically analysis of the original obstacle or abnormal measurement by the governor, so it can be more accurately specified The cause of the obstacle or anomaly .: The view of the 12th brother of the present invention is to provide a pre-measurement method to warm the board. Before the board is brought into the exposure device (used = substrate), the f information about the substrate is measured in advance; Step: According to the measurement result measured in the previous measurement step, optimize the conditions for collecting relevant data when the soil is exposed by the exposure device. Here, the conditions for data collection 'Including whether to collect data, 18 200540579 types of collected data, and the amount of data, etc., the second and second inventions are based on the results measured in advance.' The second exposure device is "under ..." Na: If the results measured beforehand are good, Then, in the light device, consider not to collect the same data as if it were 1 :: or the result of the previous measurement is not ^, then measure and collect data, or: other types of related Data measurement can be used to improve the efficiency of data collection. V. The 13th point of this Benming provides a pre-measurement processing method 'equipped with: a pre-measurement step, before the substrate is brought into an exposure device (using = exposed substrate) ) Prior to measuring the information about the substrate in advance; and the optimization step, according to the collection conditions of the data collected during the exposure of the substrate by the exposure device 'to optimize the data collection conditions of the previous measurement step: The present invention optimizes the data collection conditions in advance: the data collection conditions according to the data collection conditions of the exposure device, so for example: the material is collected by ex-ante measurement to expose The collected data can be repeatedly collected for the same data. There may be cases where the efficiency is not good. In this case, it is possible to improve the efficiency of data collection by avoiding repeated collection. In addition, in the first to the first丨 In the method of pre-measurement of 3 viewpoints, the 4 steps of the measurement are performed by the measuring device installed in the coating and developing device (connected to the exposure flat in the line), or independently of the exposure device. According to the fourth aspect of the present invention, an exposure system is provided which includes an exposure device (200, 13) for exposing a substrate; a pre-measurement device (400), and the substrate is carried in Before the exposure device, it was used to measure the 19 200540579 mark formed on the substrate, and a notification device (400, 450, and even; ^ m ^, Han inverse winding), used to measure the mark measured by the previous measurement device. The waveform is collected, and the exposure device is further provided with: an evaluation device (450, 600, 13) to evaluate the marks measured in the previous measurement step; analysis of setting and independent setting of the exposure device i 6〇〇, and to manage at least one of such devices is located in the upper management apparatus of ⑼ more such devices 00 ...) of at least one device. In this case, it is preferred that the notification device

The evaluation result of the evaluation device can choose to notify or prohibit notification of the waveform data. Can be related to the aforementioned first aspect of the present invention! The m-point prior measurement method achieves the same effect. The notification device can choose to notify or prohibit notification of the waveform data according to the evaluation result of the evaluation device. According to a fifteenth aspect of the present invention, there is provided an exposure system including: an exposure device (200, 13) for exposing a substrate; a pre-measurement step (400) for measuring and forming the side substrate before the side substrate is carried into the exposure device. Evaluation of the substrate's evaluation step 4 5 0 'Evaluate the marks measured by the prior measurement device in accordance with a predetermined evaluation standard; and the notification device (400, 4500 and the connected rural line)' The obtained evaluation result or information related to the evaluation notifies the exposure device, the analysis and placement 600 independently provided by the exposure device, and a management device that is located higher than the I to manage at least one of the devices At least one of (500 or 700) wearing can achieve the same effect as the pre-measurement method of the second aspect of the present invention. According to a sixteenth aspect of the present invention, there is provided an exposure system including: 20 200540579 A preliminary measurement device 400 ′ is used for measuring the substrate before the substrate is brought into the exposure device (200, i 3) (for exposing the substrate). At least one of a mark position, a mark shape, a pattern line width, a pattern defect, a focus error, a surface shape, and a temperature, humidity, and air pressure in another exposure device that has exposed the substrate; and a judging device (450, 600, 1 3), based on the measurement results measured by the previous measurement device, determine whether the substrate should continue to be carried into the exposure device. The same effect as that of the pre-measurement processing method according to the eighth aspect of the present invention can be achieved. According to the seventeenth aspect of the present invention, a substrate processing device (3 00) is provided. The substrate is subjected to an exposure process such as an exposure process or an exposure process within 200 of an exposure device that exposes a pattern to a substrate, and then applies the same to the substrate. It has a predetermined processing; it has a pre-measurement device 400 for measuring the mark position, mark shape, pattern line width, pattern defect, and focus on the substrate before the substrate is brought into the exposure device (the substrate is exposed through the pattern of the photomask). At least one of error, surface shape, temperature, humidity, and air pressure in another exposure device that has exposed the substrate; and a judging device 45, which judges whether the substrate should continue based on the measurement result measured by the prior measurement device A loading process into the exposure device is performed. Thereby, the same effect as that of the pre-measurement processing method according to the third aspect of the present invention can be achieved. As another example, in the #light system of the fourteenth to sixteenth aspects of the present invention, the event measuring device is provided in a coating and developing device connected to the exposure device line. According to the present invention ', it is possible to produce high-performance, South-quality micro-components and the like with high productivity and high efficiency. 21 200540579 [Embodiment] Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. (Exposure system) First, the overall configuration of the exposure system of this embodiment will be described with reference to FIG. This exposure system 100 is for processing substrates such as semiconductor wafers or glass plates. It is placed in a substrate processing plant that manufactures micro-devices and other equipment. As shown in Figure 1, it is equipped with an exposure device (with a light source such as a laser light source) 200, a coating and developing device (adjacent to the "picture!" In the figure !, shown as "transfer vehicle") 300, and an in-line measuring device disposed within the coating and developing device 200 400. In the first figure, for the convenience of illustration, the coating and developing device 300 (including the exposure device 2000 and the in-line measuring device 4) is used as an integrated substrate processing device, and only one is shown. , But in fact, a plurality of substrate processing apparatuses are provided. The substrate processing device is used to perform a k cloth step (used to apply photoresist such as photoresist) on the substrate, and an exposure step (projection exposure of the pattern image of the photomask or reticle on the substrate coated with ruble photosensitizer) , And developing steps (developing the substrate after the exposure step). In addition, the exposure device system 100 is provided with an exposure step management controller 500 that centrally manages the exposure steps performed by each exposure device 2000, that is, is located at a higher position than the exposure device 200 for managing the exposure device. Management device; analysis system 600 for various arithmetic processing or analysis processing; the main production management system 700 in the factory is located at the exposure step management controller 500 (exposure device 200) or analysis system 600 (in-line measuring device 4) 〇〇) or the off-line measuring machine 800 described later is the upper level to manage these; and 22 200540579 off-line measuring machine 800. Among the devices constituting the exposure system 100, at least each of the substrate processing devices (200, 300) and the offline measuring machine 800 is installed in a clean room where temperature and humidity are controlled. In addition, each device is connected through a network such as a local area network (LM: Local Area Netwrk) or a dedicated line (wired or wireless) installed in a substrate processing factory, so that data can be appropriately transmitted between these devices. communication. In each substrate processing apparatus, the exposure apparatus 2000 and the coating and developing apparatus 300 are connected in-line with each other. The connection within this line refers to the connection through a transporting device (a robotic arm or a slide for automatic board transfer). The in-line measuring device 400 will be described in detail later, and it is provided as one of a plurality of processing units arranged in the coating and developing device 300, and before the substrate is brought into the exposure device 2 ', it is a device that measures various information about the substrate in advance. The offline measuring machine 800 is a measuring device that is set independently from other devices. For this exposure system, a single or multiple ones are set. (Exposure Apparatus) The configuration of the exposure wearer 200 provided in each substrate processing apparatus will be described with reference to Fig. 2 '. Of course, the exposure device 200 is preferably an exposure device of stepping and scanning method 1 (scanning exposure method). Here, for example, the exposure device for the + advance-mode (one exposure method) will be described. In the following description, the χγζ orthogonal coordinate system shown in Fig. 2 is set, and the positional relationship between the members 2 will be described with reference to the χγζ orthogonal coordinate system. XYZ orthogonal coordinate system Qiao Ding Fan 1 sets the X-axis and ζ-axis to the paper surface in parallel, and the 系 -axis system to the paper surface is vertical and universal setting. The XYZ coordinate system in the figure 2 23 200540579 is actually set on a plane parallel to the horizontal plane, and the Z axis system is set in the vertical direction. In FIG. 2, when the illumination optical system i outputs a control signal (for instructing the exposure light to be emitted) from an exposure control device 13 described later, the exposure light EL having a substantially uniform illumination is emitted to illuminate the reticle. 2. The optical axis of the exposure light EL is set parallel to the z-axis direction. In the case of exposure light, for example, ’line (wavelength 436_ten lines (wavelength%-),

W excimer laser (wavelength 248nm), ArF $ molecular laser (wavelength 193nm), F2 laser (wavelength 157nm). ~ The reticle 2 has a fine pattern (for transfer to a photoresist-coated wafer (substrate Η)) and is held on the reticle holder 3. The reticle holder 3 is mounted on the base The method of movement and slight rotation in the plane on 4 is supported. The exposure control device for the entire operation of the control device is to control the movement of the reticle stage 3 and set the reticle through the driving device 5 'on the base 4. Position of 2. When the exposure light EL is emitted from the illumination optical system], the pattern image of the reticle 2 is projected through the projection optical system 6 'to each illuminated area that becomes a component part on the wafer w. Projected light Reading Lu Chu's instructions on Fu Chu, Yuan B has a plurality of pre-dried elements. As for the glass material of the optical element, it is transmitted on a daily basis. The wavelength of the exposure light EL is selected from the quartz system.

# 1 n Towering Stone 4 Learn materials first. The wafer W is loaded on the Z carrier a 8 through the wafer holder 7. Within H ^-I, Z Battle Gate 8. The components of the projection optical system 6 'can be adjusted toward the z-axis, in order to adjust the imaging characteristics H (magnification or distortion, etc.) of the projection optical system 6 described later.

v, universal U moves, and can rotate slightly around X. The adjustment of the imaging characteristics of the projection optical system 6 can be performed by changing the air pressure between the optical elements. The z stage 8 is a stage for finely adjusting the position in the z-axis direction of the wafer w, and is mounted on the XY stage 9. The χγ stage 9 is a stage for moving the wafer W in the χγ plane. In addition, although the figure is saved in%, a stage for slightly rotating the wafer W in the XY plane and a stage for changing the angle with respect to the ζ axis and adjusting the tilt of the wafer w with respect to the χγ plane are also provided.

At the upper end of the wafer holder 7, an L-shaped moving mirror 10 is mounted, and a laser interferometer ^ is disposed at a position facing the mirror of the moving mirror 10. Although it is simplified in the second illustration, the “moving 豸 1G” is composed of a flat mirror (having a mirror surface perpendicular to the X axis) and a flat mirror (having a mirror surface perpendicular to the γ axis). Another field radio interferometer 11 is a laser interferometer for 2㈣χ axis (radiating a laser beam along the X-axis' to the moving mirror H) and a laser interferometer for the ¥ -axis (producer Y-axis, for the moving mirror) H) irradiated with laser beam), the X coordinate and γ coordinate of Crystal® Retainer 7 are measured with a laser interferometer for = interferometer and Z axis. The quarterly m f ′, γ coordinates, and rotation angle information are supplied to the stage drive. The second and second messages are used as position information, and are output from the cutting table drive system 2 / control device 13. Exposure control clothes set 13 series side monitoring willow "" two are not shown in the second ®, also in the reticle holder two ... and the moving mirror and laser stem set on the wafer holding g7, cattle ... The reticle holder ... Yz position and other information are supplied to the exposure 25 200540579 Light control device 1 3. On the side of the projection optical system 6, an off-axis photographic alignment sensor 14 is set. This alignment sensor 14 series FIA (Field image

Alignment) alignment device. The alignment sensor 14 is a sensor for measuring an alignment mark formed on the crystal i w t. In the alignment sensor 14, the halogen k 1 5 passes through the optical fiber 16 and is incident on the illumination light for illuminating the wafer w. Here, as for the light source of the illumination light, _ lamp i 5 is a halogen lamp, and the wavelength range of the emitted light is 500 to 800 nm, so that the photoresist coated on the wafer w is not sensitive to light. The wavelength region and the wide wavelength band range can reduce the influence of the surface reflectance of the wafer w on the wavelength characteristics. After the illuminating light emitted from the alignment sensor 14 is reflected by 稜鏡 17, ::: is emitted on the wafer w. The alignment sensor 14 transmits the reflected light from the circle W through 稜鏡 17, converts the detection result into an electrical signal, and outputs it to the alignment &f; processing system # 5 of j8. The alignment signal processing system 18 obtains a position in the χγ plane of the alignment mark according to the detection result from the alignment sensor 14, and this position is used as wafer position information, and is output to the exposure control device 13. The exposure control device 13 'controls the operation of the entire exposure device based on the position information output from the stage driving system 12 and the wafer position information output from the alignment signal processing system 18. Specifically, the exposure control device 13 performs various calculations described below based on the position data output from the alignment signal processing system and 18 households, and various data supplied from the in-line measuring device described later as necessary, A driving control signal is output to the driving system 12. The driving system 12 drives the χγ 26 200540579 stage 9 or the Z stage 8 stepwise based on the drive control signal. At this time, the exposure control is on the basis of the wafer W; ^ + one hundred, in a way of forming, the drive control unit 14 is driven to detect a 12-round drive-out drive signal, and the XY stage 9 is driven. Right. Move the system 12 to the detection result of the sensor 14 and turn the signal processing system 18 on. Alas. If it is output to the alignment cry 14 ... According to the detection result, for example, the deviation of the measurement center of the alignment sensor 叩 14 and the optical axis AX of the reticle system 6 is measured :: the center of the image (projection optics The heart pot is moved away (baseline amount). Moreover, the pair M ^ + pair measured by the device 14 is measured, and the X temple of the wafer W is finally controlled based on the value obtained by adding the above baseline amount; ® # ^ Kl, t4 ^ m wide and Y-blocks, and then align each irradiation area with the exposure position. (Coating and developing device) Second, the coating and developing device 300 and the substrate conveying device provided for each substrate processing device ' It will be described with reference to Fig. 3. The coating and developing device 30 is connected in-line with the room surrounding the exposure device 2000. The coating and developing device 3GG towel is disposed across the central portion of the container. .W round conveying line 301. At one end of this conveying line 301, a crystal load σ 3 0 2 (for storing unexposed or processed by the substrate processing device of the previous step = most of the said W ®), With wafer stage 303 (for storing the substrate processing unit, exposure steps and display Most of the wafers in the step W), stand separately on the transfer line 30j, and have a shutter port (not shown) with a shutter on the side of the room of the 6-man exposure device 2000. Also set along the coating The conveyor line 300 of the display device 300 is provided with a main cloth 31 0 on one side and a developing portion 3 2 0 along the other side. The coating portion 31 〇 is a photoresist coating portion 311 (for coating the photoresist On wafer w), pre-baking 27.200540579 baking device 312 (composed of a hot plate for baking the photoresistance on the wafer w in advance), and cooling device 313 (for cooling the preheated wafer w). The developing unit 320 is a post-baking device 321 (for photoresistance on the wafer w after baking exposure processing, so-called PEB (p0st_Exposure test)), and a cooling device 322 (for cooling ρΕβ 的 晶 w) and a developing device 323 (for developing the photoresist on the wafer W). In addition, in this embodiment, before the wafer W is transferred to the exposure device 200, it is installed in the line. Useful 重重 ^ 、 日 彳 旦 # θ ^ In-line measuring device 400 for measuring the relevant information of the Japanese yen and w if nothing happens. Although not shown, it can also be set in-line for measurement Set (for measuring the shape of the photoresist pattern (photoresist pattern) formed on the wafer w developed by the developer 1 set 323). This device is for the shape of the photoresist pattern formed on the wafer W ( ”Column such as: pattern line width, pattern overlap error, etc.) However, in this embodiment, from the viewpoint of reducing the cost of the device, this pattern shape error is also measured by the in-line measuring device 400.

For the units constituting the coating section 310 (the photoresist coating section 311, the pre-baking device 312, and the cooling device 313), and the early parts of the developing section (post-baking device 321, cooling device 322, and development) Device 323),: line :: the composition and configuration of the measuring device, the representation in Figure 3 is expedient, 'on the horizon, a number of other processing units or buffers are also set, and each unit is a spatial configuration, also Robots or lifts in each single W. In addition, two jins and five are set up for carrying wafers and one elevator temple. In addition, the order of processing is not continuously the same. == The path is processed through each unit. It is optimized based on the viewpoints of the processing content of the processing unit or the speeding up of the overall processing time. 28 200540579 . The exposure control device 13 as the main control system provided by the exposure device 200, the coating section 310 and the developing section 320, the in-line measuring device 400, and the analysis system 600 are connected by wire or wireless, and each processing is received and transmitted. Signal to start or process completion. In addition, the original k-waveform data measured by the in-line measuring device 400 (from the photo element 4 2 2 1 described below _ under-requested or signal-processed data has the same original signal wave as the original == materials with the same content or can be restored to the original waveform data), 'The measurement results processed by the established algorithm or the evaluation results after evaluation of the measurement results are directly transmitted to the exposure control device 13 or through the analysis system 6〇〇 Σ (notification) is transmitted to the exposure control device 13. The exposure control device 13 records the transmitted information in a memory device such as a hard disk (attached to the exposure control device 13). In the exposure device 200, it is arranged approximately along an extension line provided on the central axis of the conveyance 3G1 of the coating and developing device _! The guide member 201 is a guide member 202 orthogonal to the upper part of the end of the guide member 201. ^ Add 2 guides: Please configure slide 203 (up and down along the first guide member ^ Slide ⑽ is configured with 1 胄 204 (holding the wafer w in a manner that can rotate and move up and down privately). T ^ η ,, 〇nc / J, and the 2nd guide member 202 is assigned to the 2nd # 205 (the 2nd guide member 202 is extended from the 202nd member system to the Yugong Gong Mo Le) 2nd guide 9η, ^ Every day, the wafer loading position of 0 and σ 9 is placed near the machine V that is orthogonal to the second guide member 2ΰ2 of the second arm, and is located near the intersection of the guide member 201 and the second guide member 202. In order to perform wafer f pre-alignment, a delivery pin is provided (can rotate 29 200540579 rotation and up and down movement), and a notch (notch portion) on the outer periphery of the wafer W is provided around the delivery pin 206. And two positions of the wafer edge portion, or a position detection device (not shown, for detecting the orientation plane and wafer edge portion formed on the outer periphery of the wafer w). The first guide member 201 and the second The guide member 202, the slider 203, the first arm 204, the second arm 205, and the delivery pin 206 constitute a wafer loading system (substrate transfer device). A temperature sensor (for measuring Expose Temperature in the interior of device 200), humidity sensor (for measuring humidity), and environmental sensor DT1 (atmospheric pressure sensor for measuring atmospheric pressure, etc.), temperature sensor (for measuring substrate processing) Temperature outside the device (ie, clean room), humidity sensor (for measuring humidity), and environmental sensor DT2 (atmospheric pressure sensor for measuring atmospheric pressure, etc.), environmental sensor DT3 (for To measure the temperature or humidity or pressure near the conveying line 301), and the environmental sensor DT4 (used to measure the temperature or humidity or pressure in the line measuring device 400 in the 1 line, etc.), such sensing is DT1 The detection signal to DT4 is supplied to the exposure control device 13 'and recorded in a memory device such as a hard disk attached to the exposure control device 13 for a certain period of time. (In-line measuring device) Next, the in-line measuring device 400 will be described. The in-line measuring device 400 is provided with a pre-measurement sensor, and the pre-measurement sensor is provided with at least one type corresponding to the type of the tribute of the relevant substrate, that is, a measurement item, for example, an example of an alignment mark formed on a wafer or others Markers, lines for measuring patterns, shapes, sensors for defects, sensing states for measuring the surface shape (flatness) of wafers, focus sensors, etc. For sensors, in accordance with measurement items, crystal 30 200540579 circle The state, resolution, and stomach # and the other are elastic and corresponding. It is better to set the type of compound, and select and use it as appropriate. Also, the offline measuring machine 8ηη can also be used from the outside. ~ Li 钬 800, once crying 4ηη, it is omitted, so this pit is omitted. However, in-line measurement of the state 4 0 0 14 offline measuring machine "Li 铽 800 can of course also adopt the principle of the measurement and measurement) or measurement items are different By. (Including, example for in-line measurement using pre-measurement sensors (used to align with the printed film on the wafer), Η, °° 1), refer to the figure Explain. "、, brother 4

As shown in Fig. 4, the in-line measuring device 400 includes a pre-measurement sensor 410 and a pre-measurement control device 45. In addition, although it is not shown in the figure, it also has a stage 1 (for adjusting the position of the wafer to be measured in the wt χγχ axis direction and the Z axis to adjust the tilt), and a laser interference system (for measuring the wafer W Position and status). The stage device is composed of a χγ stage, a Z stage, and a wafer holder. These are for the exposure device 2000, which is the same as the previously described Η stage 9, the Z stage 8, and the wafer holder 7. Of the composition. The laser interferometer system has the same configuration as the moving mirror 10 and laser interferometer u of the exposure device 200. The pre-measurement sensor 4 0 of the in-line measuring device 400 is a sensor for measuring the position of the alignment mark formed on the wafer W, and can be used with a photographic alignment sensor 14 (equipped with an exposure device 200). The same. Here, for example, the sensor used in the FIA (Field Image Alignment) method will be described. The sensor used in the LSA (Laser Step A1 ignment) method or the LIA (Laser Interferometric Alignment) method will be described. 31 200540579 In addition, the sensor of the LSA method can be used to measure the alignment state of the alignment mark when the field radiation is irradiated before the base radiation is formed; Its "μ, Α alignment sensor is a diffractive grid-shaped alignment mark formed on the surface of the substrate π ^ ^ ^, + irradiating wavelengths in two directions is still ^ ^ 2 of 7 Diffraction light, detection of alignment fatality based on the phase of the interference light — Ding Ban ^ " Alignment sensor of poor position in line. In-line measurement 4 0 0 system disk exposure flower dream doctor 9 π π & The situation of H exposure especially device 200 is the same, in these 3

In this type of sensing 11, there are more than 2 sensors in f, preferably more than 3 sensors in 2 ways, and they can be used separately according to the characteristics and conditions of μ. It is also possible to provide a sensor for measuring the asymmetry of a mark to be measured disclosed in Japanese Patent Application Laid-Open No. 3, as disclosed in W publication. In FIG. 4, the sensor 41o is measured in advance, and the illumination light IU0 is guided through an optical source 4il from an external illumination light source such as a toothed lamp. The illumination light iu〇 passes through the condenser lens 412 and irradiates the field-of-view dividing diaphragm 413. The 413 towel is divided in the field of view. Although not shown in the figure, a marking illumination aperture (consisting of a wide rectangular opening) and a focus detection slit (using an illumination aperture across the marking are formed in the center). A pair of narrow rectangular openings is configured). Illumination light IL10 is divided by a field-of-view dividing aperture 413 into a first light beam (for illuminating the alignment mark area on the substrate w) for marking illumination and a second light beam (before alignment) for detecting the focus position. The illumination light IL20 divided by the field of view is transmitted through the lens system 414, reflected by the half mirror 415 and the mirror 416, and then transmitted through the objective lens 417, reflected by the beam 418, and irradiates the marked area (including the wafer w). The alignment mark am) is close to its attached 32 • 200540579. When the illumination light IL 2 0 is irradiated, the reflected light on the surface of the substrate w is reflected by 稜鏡 418, passes through the objective lens 417, is reflected by the reflection mirror 416, and then passes through the half mirror 415. Then, after passing through the lens system 419, the reflected light reaching 420 'is split into two directions. The first branched light transmitted through the beam splitter 420 forms an image of the alignment mark AM on the index plate 421. In addition, the light from this image and the index mark on the index plate 421 is incident on the photographic element 4 2 2 (consisting of a one-dimensional CCD). The image of the mark AM and the index mark is formed. surface. On the other hand, the second branch light system reflected by the beam splitter 420 enters the light shielding plate 423. The light-shielding plate 423 blocks light that enters a predetermined rectangular area, and transmits light that enters areas other than the rectangular area. Therefore, the light shielding plate 423 blocks the branched light of the first light beam and transmits the branched light corresponding to the second light beam. The branched light passing through the light-shielding plate 423 is incident on the line sensor 425 (consisting of a one-dimensional CCD) through the pupil splitting mirror 424 when the telecentricity collapses, so that the image of the focus detection slit is imaged on Light receiving surface of In-Line Sensor® 425. Here, in order to break the telecentricity between the substrate W and the imaging element 422, if the substrate W is displaced in a direction parallel to the optical axis of the illumination light and the reflected light, the image is aligned on the light receiving surface of the imaging element 422 The position of the image marked AM is not changed without focusing on the light receiving surface of the photographing element 422. In contrast, the reflected light incident on the line sensor 4 2 5 is as described above, and its telecentricity collapses. Therefore, if the substrate W is displaced in a direction parallel to the optical axis of the illumination light and the reflected light, it is imaged on the line sensor. The focus detection slit 33.200540579 on the light receiving surface of 425 shifts the image position in a direction crossing the optical axis of the branch light. With this property, as long as the offset amount is measured with respect to the image reference position on the line sensor 425, the optical axis position (focus position) of the illumination light and reflected light of the substrate W can be detected. For details of this technology, refer to, for example, Japanese Patent Application Laid-Open No. 7-321030. In addition, the pre-measurement step using the in-line measuring device 400 is performed after the wafer W is carried into the coating and developing device 300, preferably after the photoresist is applied, and before the alignment process in which the light φ is set to 2000. . The installation location of the in-line measuring device 400 is not limited to that in this embodiment. For example, it may be installed in a room of an exposure device in addition to being placed in a coating and developing device 300. But'dang in-line measuring device 4. 〇When installed in the coating and developing device 300, it has the advantage that the size and shape of the exposed photoresist pattern can be measured immediately (wafer processing) Secondly, the processing for the wafer w shown in Figure 5 also includes each package. : The action 'is briefly explained. First, from the factory production management system and system 700 shown in the figure, the processing start command is output to the exposure control device 13 through the network and the exposure step management controller 500. The exposure control device 根据 outputs various control signals to the exposure device 200, the coating section 310, the developing section 320, and the in-line measuring device 400 according to this start command. When the control signal is output ', the wafer is taken out from the wafer stage 302.] The wafer is picked up by the transfer line and transferred to the photoresist coating unit 311. The photoresist is applied in order along the transfer line 30 b. Device 312 and cooling device χ M3 (sio) 'Into the stage device of the in-line measuring device 400, and carry out the alignment 34 200540579 Although the photoresist can be used as the opposite mark beforehand measurement processing (SI 1) . However, here, the sequence is performed after the processing (S10) and the pre-measurement processing (S11). The in-line measurement state 400 is measured beforehand (S11), and is performed to form the alignment mark position on the wafer W, then 1. The measurement result of the previous measurement processing (S11) (for example, the location information of the marked seat A ^ 軚) is related to the original signal waveform (the work of the sensor measuring sensor 41 〇 _ M, ,, &gt; Α The output of the camera module 422 itself) through the §fl line directly or through the agency ’s fresh film system 600 notifies the exposure control device 1V and the exposure control device 13 according to the notified information to the exposure device 2 0 0 Measure the alignment of the wafer w, 槲 槲 4a. For the Drought Age, the mark (measurement), the count is displayed, and the lighting conditions (such as: Zhaoming wavelength, Zhaoming intensity, dark field illumination) Or bright vision, ... ^ 〇T ..., month, or whether to optimize the processing through the phase difference plate 2). In addition, in order to reduce the processing load of the light control device 13, a part or all of this optimization processing may be implemented in the analysis system, and then the analysis result is transmitted to the exposure control device 13. f 、 〇 禾 得 k 芏 This process (S1 2) or parallel to this process, Cheng Xi said, "1 w # 事 刖 测 ϊprocess (SI 1) on the day of completion 0 W is delivered to the exposure device ^ 9 0 ^ vl ^ # 1 # 204. Then, when α pieces 203 followers first! The guide member is located near the magic pin 206. Then, the brother 1 # 204 turns around, and then the wafer w bu ο 因 Yinyou brother 1 arm 204 is delivered to the delivery pin A, here, to Wafer w = β starting wire &amp; * <outline datum center position 9f) r,,; L ^, ... Manjri® W is delivered to the 2nd arm / 0 2nd guide member 202 is transported to sm lL η ^ The loading position of% Sunday®, which is moved into the wafer holder 35 on the Yen stage 8 and 9 here. 200540579 After performing alignment processing including mark measurement under optimized measurement conditions, The pattern of the reticle is exposed and transferred to each irradiation area on the wafer w (S13).

After the exposure process is completed, the wafer W is transported along the second guide member and the first guide member to the transfer line 3001 of the coating and developing device 3⑽, and then sequentially along the transfer line 3101, afterwards. The baking device 32m and the cooling device 322 are transferred to the developing device 323. Next, a mask pattern corresponding to the unevenness of the element pattern of the reticle is formed in each irradiated area of the wafer W developed by a developing device such as (S14). After performing this development on the wafer W, set the line width and overlap error of the pattern to be formed as needed. Set in-line measurement = 400 $ for other measuring devices. [Check with this measuring device, It is stored in the wafer stage 3 () 3. This lithography step is completed. For example, a batch of wafers in the wafer stage 303 is transferred to other processing clothes for the remaining time (S15) and photoresist peeling (g 1 6). In the above description, although the in-line measuring device 400 provided in the coating and developing device 300 is used to perform the pre-measurement on the wafer W, the off-line measuring device 800 can also be used. The above-mentioned wafer process processing is performed separately by each substrate processing device, and each substrate processing device is managed and controlled by the exposure step at $ 50. Comprehensive control management. That is, the 'exposure step management controller 500 stores various information (used to control the processing of each batch or each wafer processed by the exposure system 100), and various related information such as various numbers or exposure history data. A memory device attached to it. Then, based on this information, in order to properly process each batch, control and manage each exposure device f 200. In addition, the exposure step 36 • 200540579 step management controller 500 is a + w # ^ ^ output of various conditions used by the registration processing station M ^,? W B7J of each exposure device 200 (number and configuration of sample exposure, Knowing that there are more points in the shot, and less, 50,000, or 50,000 points or 1 point method, the signal waveform of the signal is used to calculate the algorithm m # Fly 1. The processing of the SDM i (TM time and money conditions (considered later) SDM or GCM alignment correction I / Masaru special)), this is registered in each exposure device = light step management controller 50. It also stores various information such as exposure data and recording data. Based on these data, Appropriately control and manage each exposure device 200. Another 'angle + analysis system 600' will remove you from the following @ 系 播 先 虞 置 200, coating and developing device 300, * light breaking 2000, , Shield θ 800 Rongjiluzhuang *. First source, in-line measuring device 400, offline measuring machine: clothing, collection of various data through the network, and analysis. (Pipeline processing) Added in-line pre-measurement steps (using the above) In-line measuring device 400) 'This can not avoid the delay caused by the brother in the μ circle processing, but the following tube is applied Line processing, the Department of Forestry 〃 μ said this shame to prevent delay. Refer to Figure 6 to explain the tube processing. 由 绫 绫 言 言 言 言, w 么 ϊ w 骤 ϊ Wafer processing is a photoresistive processing step A (formation of light and formation #,,), pre-measurement step B (using in-line measurement cry 400), exposure step C (alignment and exposure), development step]) (heat treatment and development after exposure, _ J Known), A 7 &lt; 6 steps of the pattern size measurement step E (in the case of measuring the photoresist pattern). With these 6 steps, a number of wafers W junctions η u η (three in the figure e) are processed in parallel for pipeline processing. Specifically, the exposure step C of the wafer (the previous measurement step B of the preceding wafer W) is performed on the same day, and the warehouse / μ ^, thereby suppressing the impact on the overall capacity 37.200540579 is extremely small. In addition, after the development step D is performed, a photoresist size measurement step is performed. The h-shaped measurement step B and the photoresist size measurement step are performed in a time sequence that is not heavy with each other. The tube is measured linearly, so there is no need to set a photoresist size measurement device, and there is no adverse effect on productivity (alignment optimization). Figure 7 shows the alignment optimization program flow using in-line pre-measurement

Process map. First of all, the in-line measuring device 400 obtains the data to be measured in the exposure device (alignment sensor 14) through the communication of the exposure device 200 or the analysis system 600 or the main production management system 700 in the factory. Alignment mark = meter position information and mark detection parameters (correlation number of processing algorithm of signal waveform), such as limit level etc. (S2G). Next, the in-line measuring device and drive: a stage device. Align the crystal H w with the mark of the object, and sequentially position it near the detection position of the prior measurement sensor 41 〇, and implement the alignment of the position of the YI Measure (S21). τ Secondly, 'in-line measurement n 400, according to the original signal waveform data of the marker output from the photographic element 422 or the signal processed data, according to the predetermined evaluation criteria' to evaluate the marker as measured by the exposure device 200 , Mark the suitability, and calculate the score indicating the evaluation level. In the form of heart application, "Although the pre-measurement control device is installed for 45 years; and the record" Huai "when all the pre-measurement results are transmitted to the analysis system So or the exposure is first installed i 200 (exposure control device) 3), Calculations and scores can be performed on the receiving side. The description of this score will be described later. When the rating 38 200540579 is better than the threshold determined in advance, the rating and the marking table do not use the mark measured by the exposure ling 200 as the appropriate information (the transmission exposure I is set to 2GG, when the rating is more predetermined If the threshold value is bad, the knife and the mark will transmit the appropriate information (NG) indicating the mark measured by the exposure device to the exposure device 200 (S22). Moreover, when the judgment is good, it is better to deal with the situation. The information of the score A NG-from the beginning, pass ,,,, etc. In addition, in principle, it is better to transmit the signal waveform data of all the markers measured in the line measurement line to the exposure device 200 'but The right for all measurement marks to transmit the signal waveform data #, it will consume communication time, which will lead to a reduction in money, and on the pure side of the data ;,: a storage medium must be prepared beforehand with a large memory. Therefore, this implementation is evil, only For the mark that is judged to be inappropriate or the mark that cannot be measured (measurement error mark), the measured mark signal waveform data is transmitted. Also, in the present form +, it is determined whether to send an f-signal. It is carried out by the pre-measurement control device 450. From this information and the in-line measuring device 400, which will be described later, the information of the exposure device 2GG can also be notified through the analysis system_Notification of the exposure device £ 2GG. The person who has notified the exposure I will set it to 200 for explanation. In addition, in the case where the information is transmitted to the exposure device 200 through the analysis system_, the analysis system 600 can also be used to perform-part or all of the processing performed by the exposure device 200- The result is transmitted to the exposure device 200. The information of the analysis system 600 can also be transmitted to the exposure device 200 through the main production management system 700 and the exposure step management controller 500 in the factory. '39 200540579 Inside the exposure device (align the sensor 感), record the results of the measurement on the wafer (mark position information or mark signal waveform data, etc.) in the internal memory of the exposure device and transfer it to the external analysis system 6〇〇 In the system where the S memory is stored and recorded, the evaluation result in the exposure device can also be aligned with the measurement result of the sensor 14, and only regarding the judgment of the measurement is not suitable Marks that cannot be measured (measurement error marks), record the measurement results at that time. Then when receiving the information transmission in step S22, determine whether the mark detection error (NG) is set in advance before receiving the 2GG towel for receiving such information. Above noon (S23) 'In the case where the mark detection error is more than the allowable number and the original signal waveform data of the mark is transmitted, according to the data, if it is not transmitted, it is obtained from the in-line measuring instrument 400. The whole or a part of the original signal waveform is used to perform the optimization of the marker detection parameters ^. S24). X, the optimization processing of the marker detection parameters can also be performed by the in-line measuring device 400 beforehand measurement control device 45 In s23, when the mark detection error does not reach the set allowable number, the process of transferring the wafer w to the exposure apparatus 200 is continued, and the exposure process is continued (s28). 2. After the optimization processing of the mark detection parameters is performed, it is again judged whether the mark detection error is greater than the set allowable number (S25). When the mark detection error does not reach the set allowable number, the wafer W is directed toward the exposure device 2. The conveyance process of 〇 continues the exposure process (S28). After performing the mark detection parameter, after the optimization process is performed, when an N-shape of mark detection inconsistency that exceeds the setting allowance occurs, according to the information registered in advance, it is set in advance in the exploration area designated by the event month_J. The position of the coordinates on the design of other marks. 200540579 Priority order to determine whether to explore other marks

Even if I have all tears I 丨 Fenfenπ Μ ~ Subject candidates

The mark of π and δ means that the wafer number is not more than 1, and the wafer is not transferred to the exposure device 2000 to detect the error, and the wafer W is excluded (excluded from the processing step) ( S29). Also, in s29, when the number of wafers W to be excluded exceeds the number set in advance ^, wafers w of the full batch including the wafer W are excluded. The process of excluding the wafer W is not limited to the case described in the above embodiment. According to all the preliminary measurement results described below (the amount of wafer deformation predicted based on the mark position information Λ, the focus error, the pattern line width, the pattern defect, and the temperature difference in the device, etc.), the pattern exposure processing of the wafer is judged In the case of poor performance (unable to obtain good components), the wafer removal process is performed in the same manner as described above. On the other hand, it is necessary to correct the difference between the sensors of the in-line measuring device 400 and the exposure device 2000 (the difference in characteristics between the pre-measurement sensor 41 and the alignment sensor 14), including the signal processing algorithm Difference). Check the original signal waveform data of the marker transmitted from the in-line measurement 400 and the original signal waveform data of the same mark using the exposure device 200 (alignment sensor 14). The results are recorded in accordance with the 41.200540579 evaluation of the measurement results of the exposure device 200 (alignment sensor 14) of the same mark, and the score correction value is optimized. x, usually, the alignment process of the exposure device _, because at least the original k-shaped waveform data is marked for the mark record where a detection error has occurred, so it can also be used to blunt the original waveform data, detection parameters, and detection errors. The information is transmitted to the iridium, the search to the analysis system 600 or the in-line measurement state 400, which is compared with the original signal waveform data of the cake in the in-line measuring device 400, and the score correction value is optimized to be consistent. I make the evaluation of the same-mark detection ... and 'The above-mentioned sensor's characteristic difference correction processing is explained for the in-line test 2400 and the exposure device 200, but it can also be used for the offline measurement machine_ The same applies to the difference in characteristics between the sensors of the exposure device 200. Based on the above test results score, t'erming. After each pattern is obtained, the mark signal diagram is calculated. $ 料 料 丹 / α &lt; Eight, the four-inch quantifications of the pattern width error of the private note in $, etc., in Gyula Wai θ $ in each special &amp; , Riding optimization weighting, the sum of the :::: value is defined as the detection result score, and the value ratio between the pre-set: 2 whether there is a mark. Here, in order to correctly determine the "marked original information suitability and inappropriateness", it is preferable to optimize each weighting of a plurality of feature amounts in each exposure process or batch. While. Check the edge portion of the original signal waveform data, and find out the marking characteristics of the Okazu Onkyo "See regularity (eg, uniformity) or the regularity of the pattern interval (eg .hk ,, #J, Sent uniformity) is used as a feature quantity. Here, the so-called "edge J,: 2" refers to the boundary between the patterned portion and the non-patterned portion, as the boundary between the line portion and the gap portion of the line and gap mark. Therefore, the search alignment γ mark (3 marks) shown in the figure of Chu Q ^ Λ and Brother (Α) is shown as an example. The force Π is used to explain. On average, we smoothed the waveform after the noise: offset, and calculated the flatness shown in Figure 8 = intensity t cloth. Secondly, calculated the boundary of the signal intensity shown in Figure 8: ^ ^ Measured 20 Peaks M ~ P20 (Edge candidates for the line pattern and gap pattern edges SML2, SML3, so that the remaining edges shown in Figure 8 (D) are supplemented by El to El0. (Condition 1) The value must be considered as an edge allowance Within the value range. Therefore, go to: Candidates track P5, P6, pi〇, pil generated by noise NZ2, NZ3 ^ Condition 2) If there is The waveform of the edge of the line pattern is in the Y direction: in the case of a shape, a negative peak must appear after the positive peak. The peaks pi and p2 ^ caused by noise _ are caused by the complement. ^ The peak (condition 3) is in the γ direction. In the case of tracking waveforms', the distance from the positive peak to the next negative air sy: direction is considered in the range of the Υ direction of the line pattern, but for the line pattern smu, sml2 of the target mark ==, and in the direction range, within the line 2, , Since the edge candidate removes the peaks Pl3, ρΐ4, ρΐ7, and ρΐ8 caused by noise-like, = ^ 2. Starting from the edge candidate El with the smallest coordinate value of 椤 ', person', and according to the size of γ; For the information of the 6 edge candidates, calculate the feature amount of the pattern shown. (Character 1) Calculate the "line pattern width is a predetermined value (== D for summer, and pay according to the following formula)

DLW △ W1 = (YE2-YE1) — 43 • 200540579

Δ W2 = (ΥΕ4- ΥΕ3) ~ DLW

△ W3 = (YE6-YE5) -DLW Find the line pattern width error Jing (1 {== 1 ~ 3), and use the standard deviation of the difference as the feature AU and the edge candidate E1: E ^ error coordinate value as YE1 ~ YE6). Γ (Feature 2) of E6 calculates the feature quantity A2 of the "line pattern interval DLD2", according to the Ding value (-DLD1, Δ D1 = (YE3- YE2)-DLD1 ^ Δ D2 = (YE5- YE4)- DLD2 Find the standard deviation of line pattern interval error and interval error ^ as the feature amount A2. -Extract the line pattern (feature 3). Calculate the standard for "edge shapes are calculated by calculating the peak value of edge candidate 边缘 J:;: 3 The difference between the line pattern width and the line pattern interval is good. Good, the edge shape uniformity is also smaller, and the smaller the deviation is, the higher the "higher." The lower the lower, the more the correlation algorithm is used. # 令 田 in the private waveform detection situation, the higher the score :: shape, this correlation value can also be equal to the previous r? Mark and the optimization of the mark detection parameters, (E.g. illumination wavelength BΛ…, offset, alignment EGA meter, mold gas / bright intensity 'presence or absence of phase difference illumination, etc.), conditions of E optimization object. In this case, find each treatment ^ .. 余^ Component, make this error; i component becomes the minimum processing condition. Take two from 44 200540579 (irradiation alignment distortion correction (gcm )) First of all, it means the calculation pattern of illumination array deformation used by EGA. (1) The calculation pattern of illumination array deformation of EGA (up to 1st order) is shown below. △ X = Cx —10Wx + Cx —0lWy + Cx—sxSx + Cx-sySy + Cx-00 (Equation 1) △ Y = Cy-10Wx + Cy-OlWy + Cy-sxSx + Cy-sySy + Cy-00 (Equation 2) The meaning of each variable is as follows.

Wx, Wy ·· Measurement point position with wafer center as origin

Sx, Sy: Measurement point position with the irradiation center as the origin

Cx-10: Wafer Calibration X

Cx—01 · Wafer rotation

Cx-sx: irradiation calibration X

Cx-sy · Irradiation rotation

Cx-00: Offset X

Cy-10: wafer rotation

Cy-01: wafer calibration

Cy-sx: irradiation rotation

Cy-sy: irradiation calibration

Cy_00: Bias γ and 'If the above variables are used to represent 1Λ, τ, then the crystal father's degree is one (Cx—0l + Cy—1 〇), and the illumination is orthogonal — one ^ and then CCx ~ sy + Cy-sx).

Then, according to which one of the above parameters is used to calculate the 令 m Lingshi calculation mode + / number, which is also called EGA H formula (,, first count processing mode) as the formula), 10 夂 number M n -number branch formula ( It is usually called EGA mode 1 u, several branches type (multiple * branches in irradiation), average mode in irradiation. 45.200540579 The 6-mesh pull-out method refers to the above-mentioned wafer rotation, offset center "... Said the nuclear formula of the circle mark U, Y), 入, Y). The so-called 10 parameters: In the parameter mode, the lighting calibration (χ, γ) and the imaging mode are added. The so-called average mode of illumination refers to Calculate the measured value of 4 marks in the irradiation, calculate a representative value of the irradiation, and make the number I use the same parameters as the above 6-parameter mode (the position of the EGA calculation mode. Ding Ge ... Shooting (2) the platform coordinate 2 The calculation pattern of the irradiation array deformation up to the first stage is as follows: 0 △ Cx — 20Wx2 + cx—UWxWy + Cx_02Wy2 + Cx—10Wx + Cx—〇iWy + Cx—〇 ++ Cx—sxSx + Cx—SySy (Equation 3) △ Cy —20Wx2 + Cy—llWxWy + Cy-02Wy2

+ Cy-10Wx + Cy-OlWy + Cy ^ 〇〇 + Cy__sxSx + Cy.sySy (Equation 4) (3) The calculation pattern of the irradiation array deformation up to the 3rd order of the stage coordinates is as follows: Δ Cx_30Wx3 + Cx_21Wx2Wy + Cx ^ .l2WxWy2 + Cx_03Wy3 + Cx a 2 0Wx2 + Cx_llWxWy + Cx —0 2Wy2 + Cx a 10Wx + Cx a OlWy + Cx — oo + Cx — sxSx + Cx a sySy (Equation 5) 46 200540579

Δ Υ- Cy_30Wx3 + Cy ^ 2 1 Wx2Wy + Cy_l 2WxWy2 + Cy_03Wy3 + Cy-20Wx2 + Cy-llWxWy + Cy_02Wy2 + Cy —l0Wx + Cy —OlWy + Cy—OO + Cy —sxSx + Cy_sySy (Equation 6) and, In the case of one-point measurement, the irradiation correction coefficients Cx-sx, Cx-Sy, Cy-sx, and Cy-Sy of (Formula n to (Formula 6) are removed (that is, regarded as "0"). Figure 9 shows GCM (Grid Compensation f0r Matching) is used to correct the linearity of the irradiation arrangement caused by the difference between the stage and the grid number of the stage and the deformation of the process. First, determine the designation in advance. The GCM line's pre-measurement switch (a switch that can be set by the user) is ON (0N) or OFF (〇FF) (S3i). When the GCM line measures the on-off relationship in advance, it is determined in advance. (Prepared) high-order correction coefficient (S32), perform ega measurement_ / calculation (S36) of exposure device 2000, and apply EGA measurement / calculation result of S36 to apply high-order correction coefficient determined by S32 to perform exposure processing (S38) In S31, when the GCM line is measured in advance, the relationship is opened. Shape, to determine whether it is the target wafer measured in advance in the GCM line (S33). When it is not the target wafer in the gcm line measured in advance, then for the preceding wafer, it is decided to use the high-order correction coefficient used for exposure ( S34), the EGA measurement / unevenness of the exposure device 2000 is performed (S36), and at the GA measurement / calculation result of S36, the high-order correction coefficient determined by S34 is used to perform the exposure processing (s38). In S33, In the case of GCM measurement wafers, it is the measurement radiation specified in advance to the in-line measuring device 47, 200540579 4 0 0, and the B ++ 7 lamp is used to measure the $. According to the measurement results, it is regarded as a sub-path. 1 0% + t Mu 1 υ The optimization process flow of the South-order correction coefficient not in the picture, calculates the optimized ancient-stage correction coefficient of the same order (S35). The optimization process of this high-order correction coefficient shall be It will be described later. It is considered that the "Besch exposure device 2 0 0" ρ "a," 丨 丨 θ A &gt;, and the £ GA measurement / calculation of πυ (S36), the EGA measurement / calculation result at S36, shore ς π applies the higher-order correction factor determined to perform exposure processing (S38). In-line measuring device 400 And the exposure well 奘 罟 9ηΛ „μ * Especially 200 rooms, the wafer fe obtained from the measurement of the non-linear components caused by the device (wafer deformation (wafer ^ ~, w is shown daily). : Non-linear component of the shape) The difference is obvious ^ It is necessary to use a reference wafer and calculate the appropriate correction value in advance. At this time, using the oblique ^ T will use the EGA measurement results measured for the reference wafer or overlapping measurements. In addition, according to all the "Hai tendency" of the irradiation array deformation calculated by the in-line measuring device-400 in-line measurement step, a plurality of high-order correction coefficients (registered in each order corresponding to the 200 side of the exposure clothing in advance) (Usually 3 steps, but 4 steps can also be used to select the best order and the higher order correction coefficients corresponding to the correction coefficients. Exposure 200 'is the result of measuring the irradiation to perform the normal EGA calculation. Deformation linear correction (correction of linear components) is performed in conjunction with non-linear correction (correction of non-linear component block difference) of wafer deformation using the ^ gate P white ^ jE coefficient to perform irradiation array deformation correction to perform exposure processing. 70 High-order correction coefficient The correction coefficients of order 0 and 1. Here, when calculated based on the EGA measurement / calculation results, the order 0 and &quot; are both repeated, so the correction coefficient must be subtracted to pass t. The order 0 and order 48 calculated by the EGA are calculated according to the conditions of the high-order EGA and the normal EGA for the components of the right blood fiber ^… irradiated by the irradiation itself. Private ^, ten pairs of 咼 order terms Positive coefficients, still use high-order EGA, U: / 4. Separate high-order (more than 2 orders) and low-order (G-order and 1 P white dagger) ° Again, when the high-order correction coefficient is calculated based on the overlapping measurement results, since the remaining unresolvable margins are obtained, ^ 7. ~ remaining difference, the sign of the correction coefficient is reversed to use 0. Secondly, referring to FIG. 10, the explanation The high-order correction coefficient used for in-line measurement is used first. First, the in-line measuring device 400 is used to measure the alignment mark on the wafer w beforehand (S41). Secondly, the EGA calculation mode and optimization optimized by high-order EGA are specified. Order and correction coefficient (S42, S43). Then, calculate the high-order EGA correction coefficient (S44), and repeat for the specified number of wafers: Draw out this correction coefficient (S44, S45). ▲ Use the high-order EGA for the best As for the calculation mode of the converted EGA, there are 6 / digit-counter-type 10-meter-counter-counter-type and irradiation-averaging mode. For the case of 1-point measurement in irradiation, the 6-parameter mode is specified. The situation of multi-point measurement in irradiation is Designated to use 1G #number mode, average within irradiation 6 'parameter mode at any point within the irradiation. For the designation of the order of optimization with high-order EGA, if the order is 3, use the irradiation arrangement shown in (Equation 5) and (Equation 6) Deformation calculation mode, f is 2nd order, then the irradiation arrangement deformation calculation mode shown in (Equation 3) and (Equation 4) is used. Because the correction coefficients of 0th to 3rd order of (Equation 5) and (Equation 6) are used. The meaning of each component is shown in Figure 丨 丨 and Figure 2 and therefore refer to it. 49 • 200540579 The so-called designation of optimization with high-order EGA and # correction coefficient means that in order to stabilize the correction result of the first-order, the correlation High correction factor is removed (drink. For example, in the case of a 3rd order term, the correction coefficients of MxWy ^ are removed in the Wx3, Wx2Wy, WxWy2, and Wy3u ^ I corrections, thereby sometimes obtaining a stable high Θ L positive, ". Results. The higher the order of ρ, the higher the order. The more effective the removal and designation of the relevant higher-order correction coefficient is. Once in S45 in Figure 10, 'If the difference in the correction coefficient for the specified number of wafers is completed, the skipped wafer data is discharged (S46). This skipped wafer The discharge of data is the process of removing the residual squared sum of the high-order correction of each wafer from the wafer data that exceeds the interval. It can also replace the residual sum of squares and divide the high-order correction = position dispersion by the measurement result position. The value obtained by the dispersion (called the coefficient of determination, # 0 ~ !. The closer to 0, the larger the residual becomes. The dispersion of the measurement result position is the sum of the higher-order correction position dispersion and the residual dispersion) as the threshold. The person judges whether the calculation of the high-order correction coefficient is completed (S4 7) for the combination of all the conditions of the optimized order and the correction coefficient with the directed EGA (S4 7). If it is not completed, it returns to S43 to repeat the process. , Advance to S48 'to determine whether to use The first-order EGA is optimized. The calculation of the formula is completed. If it is not finished, the process will be repeated. If it is completed, the process will proceed to S49e. After being eliminated), the averaged higher-order correction coefficient is used in the combination of the most two, and the highest-order correction coefficient with the smallest sum of residuals after the high-order correction is selected to be used (S49). Also, in this embodiment, the system 50 200540579 will be described for the high-order EGA of the third order, but the same applies to the high-order EGA of the fourth order or higher. The pre-measurement result using the in-line measuring device 400 or the pre-measurement control device 450 is used to calculate and use the EGA. Or the correction value of the irradiation arrangement of GCM, and the result is notified to the exposure device 2000. The wafer w is in the in-line measuring device 400, and is removed from the in-line measuring device 400, and transported before being brought into the exposure device 200. If there are various environmental changes (temperature changes) in the path and the exposure device 2000, the wafer w will change to thermal expansion or contraction according to the self-thermal expansion rate according to the temperature change. Calculation result contains the corresponding thermal expansion or contraction of the error.

Therefore, in this embodiment, as shown in FIG. 3, a plurality of sensors, such as a measurement temperature, are arranged in various places in the substrate processing apparatus (exposure apparatus 200, coating and developing apparatus 300). Each sensor The detection temperature of the sensor is supplied to the exposure control device 13, and the exposure control device 13 predicts the expansion and contraction of the wafer w based on the detected temperature 'from these sensors. Based on this expansion and contraction, the temperature can be reduced even if a temperature change occurs. Errors caused by changes. This prediction is preferably based on the temperature change and the thermal expansion coefficient of the crystal mesh W, or the in-line measuring device '400 and the exposure device 200 are used to measure the same in the exposure material or the same test substrate. Marking, beforehand :: The relationship between the temperature changes of the various sensors' Dn ~ _ 'can be based on the changes in the degree of h to perform q' learning in the exposure program can make more accurate predictions. In the sensor DT1 ~ Shunzhong, the wafer is used by the sensor, and the in-line measuring device 400 is used for wafers. After the exposure process is performed by the exposure device 200, at least the path through which the β wafer passes is used. The measured value of the sensor (in the device) and the measured value of 51.200540579 are better. Although it is better to predict the crystal, the extension of m can also be lightened. However, in these sensors, it can also be used only Any number of senses

ητο ^ 1 For example, DT1 and j) T4, or DTI to ⑽3, or a combination of dT3 and D Ding 4). The output of Crying _ Huan Cheng was reduced to make the above prediction. (Deformation Correction (SDM)) Normally ’SDM (Super Distnr 10 1 · a u Fang Ru Cince Temple rtl〇n Matching) is based on registration

Π: Deformation and batch size of the projection optical system of each exposure device. Perform two pairs of batches to obtain the deformation of the device exposed in the past. Therefore, = the deformation of the two devices ㈣ 'in each exposure area (the position of the blind area and the deviation). The function to perform the best deformation matching in batches. … In the correction of distortion, we also obtained the 2GG exposed lenses, etc., the number of wheat coffin cases or mounting platform parameter files, and the marking error manufacturing market. To control the imaging characteristic adjustment device (ΜΑΠ) (adjust the position and inclination of the lens optical elements in the projection optical system installed to control the imaging characteristics of the exposure device = shadow optics), change the deformation shape, Match optimization. In addition, in the case of the exposure type of the scanning type, the imaging characteristics can also be adjusted by changing the number of telescopes. Taking the present invention, the in-line / offline pre-deformation measurement is performed to compare the deformation correction in batch units between the exposure devices in the previous step and the next v step, and also the deformation correction in the specified number of wafers and the specified number of exposures. Figure 13 shows the application procedure of deformation correction without in-line measurement. First, determine whether the pre-measurement switch (by the user to change the setting switch) in the SDM line specified in advance is ON (ON) or OFF (S51), and the moon shape is turned off. The deformation correction coefficient (S52) specified (prepared) by the server (here, as part of the exposure of the first figure 52 200540579 light step management controller 500) (S52), perform the EGA measurement of the exposure device 200 (S56), The EGA measurement results of S56 apply the deformation correction determined by S52 to identify the deformation of the projection optical system of the other machine (exposure device that exposes the pattern of the previous layer on the wafer) and the self-numbering machine (to be repeatedly exposed on the previous layer) For the difference in the distortion of the projection optical system of the exposure device used in the development step, the optimized distortion correction coefficient is performed when repeat exposure is performed with a self-numbering machine. In S51, when the pre-measurement switch in the SDM line is turned on, then it is judged whether it is the pre-measurement target wafer in the SDM line (S53). When it is not the pre-measurement wafer in the SDM line, it is decided before After the deformation correction coefficient used for the exposure of the wafer (previous batch) (S54), the EGA measurement of the exposure device 200 (S56) is performed. Based on the EGA measurement result of S56, the deformation correction coefficient determined by S54 is used for exposure processing. (S57>. In addition, the distortion correction coefficient determined by the above S54 is also used to identify the distortion of the projection optical system of the other machine (exposure device that transfers the pattern of the previous layer on the wafer) and the self-numbering machine (to be repeated) The difference in the deformation of the projection optical system of the exposure device used in the development step of the previous layer is optimized when repeat exposure is performed with a self-numbering machine (whether the timing of the optimization is a front wafer or The correction coefficient of deformation in the previous batch). In S53, when the wafer is the subject of SDM measurement in the line, for the measurement lighting specified in advance, the in-line measurement device 400 performs the in-line pre-measurement. The optimization processing flow shown in Fig. 14 (described later) calculates the optimized South-order correction coefficient (information about the image distortion of the projection optical system of another exposure device (other model)) (S55A). 53 • 200540579 Secondly, it is stored in the internal memory of the exposure device 2000 in advance, the memory belonging to the management controller 500 (the above-mentioned SDM server), and the memory belonging to the main system 7 0 0 Memory 'to manage and read out the projection optics of the exposure device 2000 used by the exposure device used in the current step ▲ shape information (information about the image deformation of the projection optical system used in the development step) (S55B). & Gt Stomach Secondly, according to the high-order correction coefficient (deformation information about other machines) calculated by S55A and the deformation information (compared with two _ information) of the self-number machine read out by S55B, the maximum exposure is calculated when the self-number machine is repeatedly exposed. Best deformation correction coefficient (in order to make the pattern deformation on the wafer by the exposure of the number machine consistent with the deformation of the pattern (front layer pattern) formed on the wafer by another machine, and Optimization change Shape correction coefficient, image distortion correction information) (S55C). 'Next, use the exposure device (self-numbering machine) 200, apply the optimization (obtained in step S55 above) deformation correction coefficient, and set and adjust the projection optical system imaging characteristics.丨 The momentum (parameters) of the mechanism (the mechanism that drives the lens in the projection optical system and controls the heat pressure between the lenses), or if it is a scanning exposure device, set the stage parameters such as the stage scanning speed in pattern transfer The setting is corrected, and the exposure processing is performed on the basis of the setting parameter (S57). Regarding the difference of the non-linear component caused by the device between the in-line measuring device 400 and the exposure device 200, a reference crystal must be used in advance Circle to calculate the combined correction value. Use the EGA measurement results measured for the reference wafer or any of the results of repetition, i cover,, and α results. Also, it is preferable to register a plurality of deformation correction coefficient coefficients on the device side in advance based on all of the tendency of the deformed shape measured in advance in the line. &amp; Select the correction corresponding to the order of the good and good-the person, Fenrudi 14 (illustrated #qing correction value) to optimize the line in-line pre-measurement repair In the first-line measurer side, implement the in-line pre-measurement Measure (3 U deformation correction to optimize the step (S62), calculate the correction coefficient (S63). ^ Coefficient ^ specifies the optimization order, if it is the 3rd order, use the calculation formula (Equation 5) and ( If the calculation mode 2 shown in Equation 6) is 2nd order, the calculation mode 2 shown in calculation formulas (Equation 3) and (Expression 4) is used. 'But 疋, the deformation correction case Γ Γ 〈〈 Formula〉) ~ (Equation 6) Irradiation correction coefficients Cx_SX, Cx_Sy, Cy_sx, Cy_sy (removed. The designation of the optimized correction coefficient refers to the removal of the relevant high correction coefficient for the correction (drink, for example, the case of a third-order term) :pass

Among the coefficients of the heart call, the correction coefficient of C and C is removed ', thereby obtaining a stable result of higher-order correction. The higher the order of the higher order, the more effective the removal of the associated high correction coefficient is specified. Secondly, 'determine whether the calculation of the specified wafer and the specified number of exposures-Cheng (S64)' When the situation cannot be completed, repeatedly calculate the correction coefficient, and the situation of the winter * Cheng 'after the skip data is discharged (S56), for optimization For all combinations of the correction coefficients of the second order, it is judged whether the calculation is completed (S66). In the case of ⑽: 'When it is successful, return to S52, repeat the processing, and complete the situation, which is between the wafer and the irradiation room where the measurement is completed (skip the data to exclude), at each corresponding order (2 orders, 3 orders) Order, 4th order, 5th order, ~). Regarding the averaged surface correction coefficients, in the combination of optimization conditions, select the high-order correction coefficient with the smallest residual sum of squares after high-order correction 55.200540579 to As a coefficient used for distortion correction (S67). In addition, the exclusion of the skipped data in S65 can also replace the sum of squared residuals, and remove the data of the squared sum of residuals after the higher-order correction of each irradiation exceeds the threshold. It is also possible to use the dispersion of the position of the measurement result to divide the value of the dispersion of the higher-order correction position (called the determination coefficient 'use a value of 0 ~ !. The closer to 0, the greater the residual error. The dispersion of the measurement result position plus the higher-order correction position Of the dispersion and residuals.) As the threshold. In this embodiment, the deformation correction before the third order is described, but the same applies to the correction of the fourth order or more. (I focal length correction) Figure 15 shows the application of focus correction using in-line pre-measurement ^ • First, determine whether it is 1ST exposure (about the exposure of the first layer) and 1ST exposure. Focus and perform exposure () In the case where it is not 1ST exposure, determine whether to update the segment difference data (when there is no previous data, develop new segment difference data) (s72), and for the situation where the order segment is poorly updated, take the line After the measuring device has completely performed the alignment (S73), the measurement of the component step difference of the irradiated part is performed (S74, S75). Next, the step difference correction amount (data) is calculated and transmitted to the exposure device. When the step difference correction amount is calculated, each measurement is read out The step difference of irradiation: the material of the material-the number of people is converted into the internal coordinate system of irradiation, and the same is performed: two deaths: homogenization. At this time, by the least square approximation, the spline (Fourier) or Fourier level | nu The position shift of the detection point at the tenth and the tenth point of the temple is the same as the position of the difference between the 56 and the 200540579. The measurement center irradiation is based on the irradiation center position as the reference in the 、 X and γ directions. The grid-shaped data are arranged side by side at a specified interval. At this time, if necessary, an interpolation function is also used. For the position data selected in the grid-shaped data, an appropriate # offset and weight are set, and an approximate surface is calculated by measuring the irradiation unit. This approximate surface can be either a plane or a curved surface. Also, the step data of each measurement exposure is converted into data from the difference portion of the approximate surface (offset data). However, the distance from the approximate surface to the first threshold is specified by a parameter. The segment difference data is excluded from the calculation of the approximate surface. In addition, the data specified as the parameter separated from the approximate surface by the second threshold or more (outlier data) is detected, and the outlier data is designated as the parameter with more than a number The measured irradiation is regarded as unsuccessful irradiation, and only the remaining step differences are averaged to calculate the element step difference correction amount. When the average here is 蚪, interpolation is performed as necessary. Also, the abnormal value data detected at this time The system is transmitted to the main production management system 700 in the factory. The main production management system 700 in the factory transmits the abnormal value data to the offline measuring machine 800 (by external crystal (Including a circle defect inspection device or an inspection station). The correction amount is obtained based on the above. The exposure I is set to 2 0, which is the correction amount based on the step difference data measured beforehand. After adjusting the afocal focus (S77), the exposure process is performed. (S79). (Phase-shifted focus monitoring) On the processed wafer, a phase-shifted focus monitoring mark is formed in advance, and before processing with exposure set 200 (the month when the processed wafer is brought into the exposure device) The detector 400 is used to align and measure the phase-shifted focus monitoring mark formed on the processing wafer w 57 .200540579, so that each mark can be measured and, based on the result of this measurement (pre-measurement), at the exposure place =: The best correction value for focus offset and level offset. If the target of the focus monitoring is-180 for the film pattern. For phase shifters other than that, the focus error ΔZ &gt; is designed according to the change of focus to use the image asymmetry. 俾 Enable to convert to overlap error. 仏 Set i chrome lines to the phase shifter. Phase between. However, the phase shift amount of the phase shift section is not 180. But 90, will

^ Phase focus monitor pattern enters the “i irradiation” to perform in-line pre-measurement, and then calculate the focus offset and leveling offset, and notify the exposure device _ to perform the best focus correction. (Efficiency of device maintenance) The in-line measuring device 400 measures information about the pattern line width or shape and other pattern defects formed on the wafer, evaluates the quality of the pattern, and counts it according to the level. The signal waveform information is notified to the exposure device 2000. The exposure device 200 is based on the evaluation result notified from the in-line measuring device 400, and identifies the pattern defective part and the approaching bad part. Based on the original signal waveform data of the part, various tracking data, and overlapping measurement data and EGA are obtained. (Alignment) The calculation result is selected as the target Z position for analysis. Secondly, various tracking data including defective and approaching defective parts, and overlapping measurement data and EGA (alignment) calculation results are obtained from the exposure device, and the correlation with pattern defects is analyzed. Here, overlapping measurement data can also be obtained from a measurement device other than the exposure device. As for the analysis content, the focus tracking data, exposure tracking data, and synchronization accuracy tracking data are individually analyzed to predict the pattern size control performance. Based on the overlap measurement data and EGA (standard 58-200540579) calculation results, the overlap control performance is predicted. If it is determined to be related to the defect, if necessary, ‘correct the operating parameters of the exposure device i 2GQ, or perform maintenance on the device. Hereinafter, each analysis method will be described. (1) Acquire the focus tracking data in the exposure process on the exposure device 200 side 'based on the focus and vertical data' analysis pattern size control. The z-tracking error of focus and longitudinal, piteh tracking error, and ㈣ tracking error are reflected to the uniformity of irradiation measured beforehand, thereby calculating the ⑴zmean (B) and (B) Z standard deviation ^ (msd). At each image height (considering the curvature of the image plane: the effect of the main optical aberration), the line width values of the z-average and the standard deviation of each z (using SEM, 0CD Φ Sheng every day, day &amp; Values, or calculated values using a space-like simulator, are maintained as tables. And 'in each exposure condition: the line width value table file. In terms of exposure conditions, t has an exposure wavelength λ,

Numerical aperture of projection lens NA, day call aa M,…, moon σ, lighting conditions (usual lighting, variable _, ,, moon) mask pattern type (binary, halftone, Levenson aeve_n), etc., mask line Width, target line width, and pattern spacing. Refer to the above line width from the average hook degree measured in each irradiation and the focus and vertical data in the exposure process: Yu :, ## The line width value of this condition. In this way, in fact, the line width values of the predicted and interpolated predictions are not measured. If an abnormal line width is detected, immediately after exposure, the scan speed is reduced or the segment correction is changed, the control method of the polymerization is changed, or Preventive measures such as device maintenance. (2) Analyze the pattern size control and overlap control according to the synchronization accuracy tracking. Synchronization accuracy refers to the tracking offset of the wafer stage in the exposure slit area during scanning. ), Using moving average (平均值 η) 乂 moving standard deviation value (msd) for evaluation. Moving average 59,200540579 (Xmean / Ymean) will be affected by the displacement in the scan and affect the superposition accuracy. Moving the standard deviation (Xmsd / Ymad) reduces the contrast of the image plane and affects the accuracy of the pattern size. Determine if the equivalent value is an allowable value ^ If it exceeds 2 allowable values, immediately after the exposure, take measures to reduce the scanning speed or update the step correction, the synchronization accuracy control method, the focus control method change or install = maintenance and other defective products countermeasures . t &quot;

(3) Analyze the pattern size control according to the exposure amount tracking data.-In the tracking data, record the exposure amount results every certain time interval. The exposure amount is evaluated during the scanning by the average exposure amount of the slit area at each position. Determine whether this value is within the allowable value. If it exceeds the allowable value, after η exposure, take countermeasures such as reducing the selection speed or changing the focus control method, and device maintenance. s Measure the results to analyze the weight (4) Based on the overlap measurement data and EGA (alignment) measurement control / analysis of the data obtained by using the overlap measurement control or the system incorporated in the exposure device. Determine whether the overlapping measurement result of J ^ inch + W 疋 Bad 邛 knife is within 60 minutes. Furthermore, it is determined whether the overlap offset is de-aligned (aligned). The gate (rho knife) is not within the allowable value. The EGA (alignment) calculation results are compared between wafers and batches to confirm whether there is a large change. (Optimization of measurement conditions) Most: The measurement conditions measured in advance according to the operating state of the exposure device. For example, in the exposure device 200, when only this type of household + + taxi * mile type occurs, only this place The time required will narrow the exposure. Even if it ’s increased by 60 200540579 for a long time used in the pre-measurement part, it will not adversely affect the throughput of exposure processing. On the other hand, in the pre-measurement step, the more measurement items, the number of measurements, and the amount of data, the more detailed analysis or calculation of the correct correction value can be performed. Therefore, it is preferable to optimize the measurement conditions of the previous measurement step in accordance with the operation conditions of the exposure device 200 (such as the interruption state of the exposure processing). The optimization of this case is performed in a range that does not reduce the throughput of the exposure process, so as to maximize the number of measurement items, measurement points, and measurement data. Therefore, it will not adversely affect the productivity, can analyze or calculate the correct correction value in more detail, and can improve the exposure accuracy. (2) Optimization of measurement conditions using periodic pre-measurement-The exposure system described in the above embodiment basically uses an in-line measuring device before all the processing wafers carried in the exposure apparatus 200 are carried into the exposure apparatus. , 400 to measure beforehand. Therefore, all the processed wafers are measured in advance, and according to the measurement results, an abnormal state (for example, the measurement candidate mark cannot be measured, etc.) can also store the occurrence of such an abnormality (the time or frequency of the abnormality, and its abnormal content). ). If you analyze the price, the data of the abnormality situation collected in this way is used to estimate the tendency of the abnormality to occur. Here's what kind of error (abnormality) is used by Idang Su & 4: Condition: to guess at what timing is prone to occur only to what extent (% ^ If the error occurs, it is better to Seed gas) "丨: £ gastric volume) 'to measure what kind of data (data and, based on this speculation, make a difference ... for the purpose). Optimize the measurement conditions when things go wrong. 61 .200540579 For example 'If you look at how often a certain cycle occurs, then ... Go to the mother cycle, and analyze what kind of content is different (pre-storage library, periodic measurement should be measured beforehand, the type of data or the description cycle, Yong Qu ♦ Laoyangli). Optimize the entry cycle (the cycle between offsets, θ + position of the first device exposure bowl, 'month), the wafer cycle within the batch (n wafers apart), Or a continuous cycle (hour or year and month). (3) The use of errors _ Mo Ximu a degree of the measurement conditions of the optimization of the other steps of the optimization steps, often occur in the Lord; 5 ~ the birth of the wrong 丨 moon shape, you must specify the wrong two = : Therefore, the present invention optimizes the pre-measurement steps according to the number of errors. If the cause of the obstacle or abnormality is more specifically analyzed, the effective measurement conditions are used to perform the measurement. It is possible to more accurately identify the cause of the obstacle or abnormality. In the measurement bar of the exposure device, the measurement conditions on the exposure device side are performed in advance. The result of the previous measurement is very good. In the exposure device, consider that you do not need to collect the same samples as the previous measurement person. &lt; Shell material is useless. In order to reduce the amount of waste, it is better to collect the data according to the V and mouth fruit measured in advance (including the data collection related to the exposure device 2 and the exposure device of the substrate). optimize. Not only is there a data tree, but also the collection (measurement) of 5 Xuanbeike itself is also implemented on the exposure device side, also: according to the results of prior measurement) the amount of data collected (data amount, measurement device) is increased or decreased (if the previous measurement result is good, Then reduce the composition of the same lean material on the exposure device side).测量 Measurement conditions for pre-measurement according to the measurement conditions of the exposure device 62. Optimization of 200540579 For example, if the exposure material is also collected on the pre-measurement side, the same data is collected repeatedly, sometimes there is no ㈣n! . At the time of exposure ', optimize the data collection conditions based on the collected data "exposure first measurement step, for example: avoid:,: to improve the efficiency of data collection beforehand. Recollection, (component manufacturing method) : 次 ：： Describes the manufacturing method using the above exposure system in the lithography step. Figure 16 shows, for example, semiconductor wafers such as IC or CD, liquid crystal panels, CCDs, diaphragm magnetic heads, micromachines, etc. The flow chart of the electronic component manufacturing process. As shown in FIG. 16, in the process of electronic components, first, second; f sub-component circuit design and other functional and performance design of components, pattern design to achieve this function (steps S81). Next, a photomask with a circuit pattern formed is fabricated (step S82). In addition, a wafer (silicon substrate) is manufactured using materials such as Shi Xi (step S83). Next, the step S82 is used. The fabricated photomask and the wafer manufactured in step S83 adopt lithography technology to form actual circuits and the like on the wafer (step "S84). Specifically, a, first, an insulating film, an electrode wiring film, or a semiconductor film is formed on the surface of the wafer (step S841), and secondly, a photoresist (photoresist) is applied using a photoresist coating device (coater) on the entire surface of the film ( Step S842). Next, mount the photoresist-coated substrate on a wafer holder of an exposure device, and mount the photomask manufactured in step S83 on a reticle stage to reduce the pattern formed on the photomask. 63.200540579 is transferred onto the wafer (step S843). At this time, in the exposure method Yia Xia, using the above-mentioned alignment method of the present invention, the wafers are sequentially aligned and the pattern of the photomask is sequentially transferred. After the exposure is completed, keep the wafer from the wafer and unload the θ circle at 4 degrees. Unload the RiRi® and use the developing device to perform the development (step S844). Bury it here, say m m; form a photomask on the wafer surface. Patterned photoresist image. Next, on the wafer that has undergone the development process, a rice-engraving device is used to apply a last-name engraving process (step S845) to remove the photoresist remaining on the wafer surface, for example, using a plasma ashing device to remove it (step s 8⑹. In this way, the pattern is waited in each illuminated area of the wafer. Then, the light m tongue layer or the electrode wiring is changed, and this process is sequentially repeated in order to form an actual circuit on the wafer. If a circuit or the like is formed on the wafer, the assembly of ::: pieces is performed (step S85). Specifically, the wafer is cut to be divided into: I circles, and each circle is mounted on a lead frame or package, The connection electrodes are connected, and then the resin is packaged and mounted. Then, manufacturing: component operation confirmation test 1 long-term test and other inspections ^ finished component shipments, etc. Tian You ^ the implementation of the above is to facilitate Those who understand the present invention and remember it are not limited to the present invention, but they are respected this month. They are contained in it. Therefore, the elements disclosed in the above description include all design changes or equivalents that belong to the technical scope of the present invention. &Amp; nt again, on In the embodiment described above, as for the exposure device, the exposure method of the π method is taken as an example, but an exposure device of the step scan type can also be applied. Moreover, it is not only used for exposure devices made of semiconductor elements or liquid crystal displays, Used in ', exposure devices used in the production of indeterminate displays, thin-film magnetic heads, and 64-200540579 photographic elements, etc., and photomasks, where circuit patterns are transferred to glass substrates and cut into wafers. The present invention is also applicable. That is, the present invention can be applied regardless of the exposure side of the exposure device = Shang Tu. Each embodiment of the formula or the use mode is not limited to the exposure of the step scanning method ... 'The same can be applied to the step repeat method or the proximity method (X-ray exposure device, etc.). All kinds of exposure equipment can also completely

In addition, the exposure illumination light (energy beam) used by the exposure device is limited to ultraviolet light ', and x-rays (including so-called light), charged particle beams such as electricity or sub-beams, etc. may be used. It can also be an exposure device used in the manufacture of doves, masks, or reticle. Furthermore, in the above-mentioned embodiment, a light-transmitting photomask in which a predetermined light-shielding pattern (or a phase pattern or a dimming pattern) is formed on a light-transmitting substrate is used, or a predetermined reflection pattern light is used on a light-reflective substrate. Reflective photomasks can also be replaced with electronic photomasks based on the pattern of the pattern to be exposed. Such an electronic photomask 'is disclosed in, for example, U.S. Patent No. 6,778,257. Here, the above-mentioned electronic photomask, which is referred to as U.S. Patent No. 6,778,257, is a concept including both a non-emissive image display element and a self-emissive image display element. Here, 70 non-luminous images are also referred to as Spatial Light Modulators, which are the 65,405,405 elements of the amplitude, phase, or polarization of spatially modulated light. They are classified as transmissive spatial light modulators. And reflective spatial light modulators. The transmissive spatial light modulator includes a transmissive liquid crystal display (1 ^ 0: Liquid Crystal Display), an electro-optic display (ECD), and the like. The reflective spatial light modulator includes a DMD (Digital Mirror Device or Digital Micro-mirror Device), a mirror array, a reflective liquid crystal display element, an electrophoretic display (EPD: ElectroPhoretic Display), and electronic paper (or Electronic ink), and Grating Light Valve. The self-luminous image display device includes a CRT (Cathode Ray Tube), an inorganic EL (Electro Luminescence) display, a field emission display (FED, Field Emission Display), a plasma display (PDP: Plasraa Display Panel), Or a solid-state light source chip with a plurality of light emitting points, a solid-state light source wafer array in which the wafers are arranged in a plurality of arrays, or a solid-state light source array in which a plurality of light-emitting points are mounted on a substrate Display, 〇LED (〇rganic

Light Emitting Diode) display, LD (Laser Diode) display, etc.). In addition, if a fluorescent substance provided in each pixel of a well-known plasma display (pDp) I is removed, it becomes a self-luminous image display device that emits light in the ultraviolet region. Furthermore, the above-mentioned embodiment is described in terms of a system, but the device, inspection device, test device, and device of the present invention. For the application of the present invention to the exposure system, it can also be applied to all the handling equipment, measurement and alignment of other objects. [Schematic description] 66 200540579 The figure is a block diagram showing the overall configuration of the exposure system of the embodiment of the present invention. Fig. 2 is a diagram showing a schematic configuration of an exposure system provided with an exposure system according to an embodiment of the present invention. Fig. 3 is a schematic configuration diagram showing a coating and developing device and the like connected in-line to an exposure device according to an embodiment of the present invention. Fig. 4 is a diagram showing an example of a pre-measurement sensor used in an in-line measuring device and an off-line measuring device according to an embodiment of the present invention. Fig. 5 is a flowchart showing a process flow of the embodiment of the present invention. Fig. 6 is a pipeline processing diagram for explaining the embodiment of the present invention. Fig. 7 is a flowchart showing the alignment optimization procedure of the pre-measurement in the line using the embodiment of the present invention. Fig. 8 is a diagram showing the results of mark photography and processing results according to the embodiment of the present invention. Fig. 9 is a flow chart showing an application procedure of irradiation alignment correction (GCM) beforehand measurement in a line using the embodiment of the present invention. FIG. 10 is a flowchart showing an optimization procedure of the A-order correction coefficient (gcm correction value) measured beforehand in the line using the embodiment of the present invention. Fig. 11 is a diagram showing the contents of the lower-order components of the components of the correction coefficients from 0 to 3 in the embodiment of the present invention. Fig. 12 is a diagram showing the contents of the higher order components among the components of the correction coefficient from order 0 to order 3 in the embodiment of the present invention. Fig. 13 is a flow chart showing an application procedure for measuring the shape change (SDM) 67,200540579 beforehand within the line of the embodiment of the present invention. Fig. 14 is a flowchart showing an optimization procedure of the deformation correction coefficient (SDM correction value) measured beforehand in the line of the embodiment of the present invention. Fig. 15 is a flowchart showing a procedure for correcting the focus step difference measured in-line beforehand. Fig. 16 is a flowchart showing the manufacturing process of electronic components. [Description of main component symbols] W wafer 100 exposure system 200 exposure device 300 coating and developing device 4 0 0 in-line measuring device 41 0 prior measurement sensor 450 prior measurement control device 500 exposure step management controller 600 analysis system 700 in factory Main production management system 8 0 0 Offline measuring machine 68

Claims (1)

.200540579 10. Scope of patent application: 1. A pre-measurement processing method, which is characterized by the following steps: the pre-measurement step 'measures a mark formed on the substrate before the substrate is brought into an exposure device (for exposing the substrate); and A notification step of notifying the exposure device, the analysis device separately provided with the exposure device, and the waveform data related to the mark measured in the previous measurement step, and in order to manage at least one of the devices, the device is higher than the devices At least one of the management devices. 2. If the prior measurement and processing method for item i of the patent application scope is further-equipped with an evaluation step, the mark measured by the previous measurement step is evaluated according to a predetermined evaluation standard; the notification step is in accordance with the evaluation step. Evaluation result, you can choose to notify or prohibit notification of the waveform data. 3. For example, picture 1 of item 2 of the scope of patent application, Tian Zhan has only completed the measurement and processing method, and the notification step is to notify the evaluation result when waveform data communication is not performed. 4. If the prior measurement and processing method of any of items 1 to 3, ^ ^ d of the scope of patent application, it is further provided with a mark selection, a step, according to the waveform data and the evaluation office notified by the notification step ^ At least one of the results selects the best 胄 A _ not 5 自 from the plurality of marks formed on the substrate, although it is used as a measurement mark when the exposure device advances the side substrate position. 5, such as the scope of patent application! Pre-test I processing method for any of the ancient, earth, earth, earth, earth, earth, earth, earth, earth, earth, earth, earth, earth, earth, earth, earth, earth, earth, earth, earth, etc., its iT, further equipped with the measurement of Niu Jiu, Ji Zuo Jin A and A, and steps, according to the waveform information notified by the private step and At least one of the 1 shell to the fruit is selected. 69.200540579 The best selection when measuring the mark is to take soil removal conditions for the positioning of the substrate. Settlement before any one of the three items is listed in the Marking Order. 6. If the scope of the patent application is applied, the method “its 1if7, formed in Xi I” Cheng Fang, and the marking of the 5 Hai substrate includes at least the following One type: points; a pre-alignment mark for pre-positioning of the substrate or a precise alignment mark for precise positioning of the substrate outline feature of the substrate and a search-alignment mark for searching for the precise alignment mark of the substrate. 7 The pre-measurement processing method as described in item 5 of the application of the kiss patent, which makes the measurement conditions include: in order to perform the substrate in the exposure device: the number of marks used for positioning, the mark configuration, the focus offset, and the measurement Lighting conditions, and statistical processing modes. 8. If the prior measurement and processing method of item 2 or 3 of the scope of patent application, wherein the evaluation step is based on the predetermined evaluation standard, a scored evaluation result is generated. 9. If the prior measurement processing method of any one of claims 1 to 3 of the scope of patent application, it is further equipped with a formal measurement step, after the substrate is moved into the exposure device, the mark formed on the substrate is measured; according to the notice At least one of the waveform data and the evaluation result notified by the step, and the measurement result of the formal measurement step to match the measurement device used in the previous measurement step measurement with the measurement aggregation used in the formal measurement step measurement Marked evaluation criteria. 10. A pre-measurement processing method, comprising the following steps: 70.200540579: a pre-measurement step of 'measuring a mark formed on the substrate before carrying the substrate into an exposure device (for exposing the substrate); an evaluation step, Evaluate the mark measured in the previous measurement step according to a predetermined evaluation standard; and notify ㈣, notify the exposure device of the evaluation result or evaluation information obtained in the evaluation step to the exposure device, and an analysis table provided independently of the exposure device i and at least one of the management devices that is higher than the devices in order to manage at least one of the devices. 11. A pre-measurement processing method, characterized by having the following steps: the pre-measurement step, forming a square in two measurements of carrying the substrate into an exposure device (for exposing the substrate); a plurality of mark positions on the substrate; and The correction tribute calculation step calculates correction information (including the linear correction coefficient and the non-linear correction coefficient from which the design position error of the mark is minimized) based on the measurement results measured in the previous measurement step. 1 2. A pre-measurement processing method, comprising the following steps: The pre-measurement step measures a plurality of mark positions formed on the substrate before the substrate is brought into an exposure device (for exposing the substrate); image distortion calculation Step 'According to the measured yttrium fruit measured in the previous measurement step, the image distortion information of the projection optical system of another exposure device that has exposed the substrate is revealed; and the correction information calculation step is calculated according to the image deformation calculation step. 71.200540579 The image distortion information of the projection optical system of the other exposure device and the image distortion information of the projection optical system provided by the exposure device are obtained in advance to calculate the image distortion correction information (used to make the The distortion of the image produced by another exposure device is generated in this exposure device). 1 3. A pre-measurement processing method, comprising the following steps: a pre-measurement step of measuring a phase shift focus mark formed on the substrate before the substrate is brought into an exposure device (for exposing the substrate); and focus correction Information calculation step, based on the measurement result measured in the previous measurement step, to determine the focus error when the exposure is performed by another exposure device that has exposed the substrate, to calculate the focus used when the substrate is exposed by the exposure device Fix the information. 14. A pre-measurement processing method, comprising the following steps: a pre-measurement step of measuring a surface shape of the substrate before the substrate is brought into an exposure device (for exposing the substrate); and an iW positive investment step, According to the measurement result measured in the previous measurement step, the focus correction information is different (used when the substrate is exposed by the exposure device). 15. A pre-measurement processing method, comprising the following steps: the pre-measurement step, in which the substrate is brought into an exposure device (for exposing the substrate) such as' measure a plurality of mark positions formed on the substrate; temperature measurement A step for measuring the temperature change in at least one of the measurement device used in the prior measurement step measurement 72 200540579, transferring the substrate from the measurement device to the transfer device of the exposure device, and at least one of the exposure device ; A prediction step that predicts a change in the marker position measured by the previous measurement step based on the temperature change measured by the temperature measurement step; and a correction information calculation step that calculates correction information based on the prediction result predicted by the prediction step ( Including linear correction coefficients and non-linear correction coefficients from which the design position error of the mark becomes dare to be small). 16. A pre-measurement processing method, comprising the following steps: The pre-measurement step, before measuring the substrate into an exposure device (for exposing the substrate), 'measure the mark position, mark shape, pattern line width, At least one of pattern defect, focus error, surface shape, temperature, humidity, and air pressure in another exposure device that has exposed the substrate; and a judging step of judging whether the substrate is based on a measurement result measured in the previous measurement step The carrying-in processing into the exposure device should be continued. 0 7. A pre-measurement processing method is characterized by having the following steps: The measuring step is performed before the substrate is brought into the exposure device (for exposing the substrate). Measuring information about the substrate; and an optimization step of optimizing the measurement conditions of the previous measurement step according to the operating condition of the exposure device. 1 8. A pre-measurement processing method, which is characterized by the following steps: 73.200540579: The pre-measurement step, 'pre-measure information about the substrate before the substrate is brought into an exposure device (for exposing the substrate); and the best The optimization step optimizes the measurement conditions of the previous measurement step according to the periodicity paid by the measurement result measured by the previous measurement step. G: the processing method is characterized by having the following steps: the Shili measurement step, before the substrate is brought into the exposure device (for exposing the substrate), the information about the substrate is measured beforehand; and the S step, According to the number of errors obtained from the measurement results measured by the previous measurement step, the measurement conditions of the previous measurement step are optimized. 20. A pre-measurement processing method, characterized by the following steps: Eight pre-measurement steps, measuring the information about the substrate beforehand when the substrate is brought into the exposure device (for exposing the substrate); and an optimization step, according to The measurement result measured in this preliminary measurement step optimizes the collection conditions of the relevant data when the substrate is exposed by the exposure device. 0 steps · 21 kinds of preliminary measurement processing methods, which are characterized by the following steps: When the substrate is brought into an exposure device (for exposing the substrate / something to measure the information about the substrate; and the steps, according to the collected conditions when the substrate is exposed by the exposure device; 'collection conditions, Optimize the data collection conditions of the previous measurement step 74 200540579. 2 2 Ν κ From the previous measurement processing method of any of items 1 to 3 in the patent scope of Sigma, # 士 ”T 'The previous measurement step is It is performed by a measuring device installed in a coating and developing device (connected to the exposure device in a line). 23. If any of the scope of patent application 1 to 3 The method of the prior measurement section of the item 'where' the prior measurement step is performed by a measurement device provided separately from the exposure device. 24. An exposure system characterized by: an exposure device for exposing the substrate; prior measurement A device for measuring a mark formed on the side substrate before the substrate is brought into the exposure device; and a notification device 'for notifying the exposure device and the exposure of the waveform data related to the mark measured in the previous measurement step An analysis device independently provided by the device, and at least one device that is located in a higher management device than the devices in order to manage at least one of the devices. 25. If the exposure system of the scope of application for patent No. 24 is further provided, The evaluation device evaluates the data measured in the previous measurement step according to a predetermined evaluation standard; the notification device can choose to notify or prohibit notification of the waveform data according to the evaluation result of the evaluation device. 26. An exposure system characterized by having : Exposure device for exposing the substrate; prior measurement A device for measuring the mark formed on the substrate before the substrate is brought into the exposure device; 75.200540579 evaluates the mark and evaluates the mark measured by the prior measurement device according to a predetermined evaluation standard; and the notification device is used Notifying the exposure device, the analysis device independently provided with the exposure device, and the evaluation information obtained by the evaluation device, and the higher-level Xuan Xun device in order to manage at least one of these devices At least one of the management devices. 2 7. An exposure system, comprising: a pre-measurement device for measuring a mark position and a mark on the substrate before the substrate is brought into the exposure device (for exposing the substrate). At least one of shape, pattern line width, pattern defect, focus error, surface shape, temperature, humidity, and air pressure in another exposure device that has exposed the substrate; and a judging device based on the measurement results measured by the prior measurement device To determine whether the substrate should continue to be carried into the exposure device . 28. If at least one of the pre-exposure measuring device and pre-examination device is used in the exposure system of any one of the scope of application for patents Nos. 24 to 27, it is set in a coating and developing device (connected to the exposure device in-line). 29. For the exposure system according to item 27 of the patent application scope, wherein at least one of the pre-measurement device and the pre-judgment device is connected offline to the exposure device or is disposed in the exposure device. 30. A substrate processing device is a method for applying a predetermined treatment to a substrate before or after an exposure process in an exposure device that transfers a pattern onto a substrate, and is characterized by: Before being brought into the exposure device (exposing the substrate through the mask of the photomask), it is used to measure the position of the mark on the substrate, the mark 76.200540579, the pattern finding, pattern defects, focus errors, surface shape, and the substrate has been exposed. At least one of temperature, humidity, and air pressure in another exposure device; and a judging device that judges whether the substrate should continue to be carried into the exposure device based on the measurement result measured by the prior measurement device. 31. A pre-measurement system, comprising: a pre-measurement device for forming a measurement on the substrate before the substrate is moved into a predetermined device (equipped with a positioning device for positioning the substrate moved into the substrate). And a notification device for notifying the predetermined device of the related waveform data of the target measured by the prior measurement device, an analysis device provided separately from the predetermined device, and at least for managing such devices -Kind of position: At least one of the higher-level management devices than these devices. 32. A pre-measurement system, characterized by having a pre-measurement device for measuring the substrate before it is moved into a predetermined device (equipped with an alignment device 'for the alignment of the substrate moved into its interior). A mark formed on the substrate; i an evaluation device 'to evaluate the mark measured by the prior measurement device in accordance with the predetermined evaluation standard; and a notification device to notify the evaluation result or evaluation-related information obtained by the evaluation device of the predetermined A device, an analysis device provided separately from the predetermined device, and at least one device in a management device higher than the devices in order to manage at least one of the devices. 33. A pre-measurement system, which is characterized by: 77 • 200540579 Before the event is measured, Linong i, before the substrate is moved into a predetermined device (with the alignment device used to enter the alignment of the substrate inside it) i :: Base: at least one of the position of the mark, the shape of the mark, the pattern line width, the pattern defect, ", the difference surface shape, the substrate has been treated-at least one of temperature, humidity and air dust in a given device And, the judging device i 'judges whether the substrate should continue to be carried into the jumbo device according to the measurement result measured by the pre-macro measurement device. 34. A pre-measurement system, comprising: an acquisition device , Before the substrate is moved into a predetermined device (with the alignment of the substrate that is moved into the interior), it is used to obtain the relevant information of the mark on the substrate that is scheduled to be measured in the existing 4 units; and the pre-measurement Before placing the substrate in the predetermined device, take the γ-Shi-Yi Shi according to the acquisition device. ^ Set the acquired tfL 'to measure the target formed on the substrate. 35. Pre-measurement system such as the scope of application for patent No. 34 The relevant information of the mark includes: the design position information of the mark and the relevant parameters of the processing algorithm when the signal waveform of the mark is processed. At 36, a processing system is characterized in that: the base is based on a predetermined processing device and is provided with a positioning device to carry out the alignment of its inner plate, and after the positioning by the positioning device, the substrate Processing; a measuring device that measures the marks formed on the substrate before the substrate is moved into the processing device; and] the information providing device performs pre-measurement by the prior measuring device 78 .200540579 , The information related to the mark scheduled to be measured on the processing device side is provided to the prior measurement device; the prior measurement device measures the mark formed on the substrate according to the information provided by the information providing device. Eleven, schema: as the next page 79