JP2010186918A - Alignment method, exposure method and exposure device, device manufacturing method, and exposure system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To easily make nonlinear corrections on an in-shot error with high precision. <P>SOLUTION: Many alignment marks given to a plurality of shots on a wafer are previously detected and a plurality of (conditions for) models where a nonlinear component of strain of the wafer is described are determined based upon precise detection results thereof. In wafer exposure (lot processing), a smaller number of alignment marks are detected in a step 608 or 606 to obtain the strain of the wafer, and the plurality of predetermined (conditions for) models are selected based upon rough results thereof in a step 618. The selected models are used to position patterns at the plurality of shots in order, and exposure is carried out. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to an alignment method, an exposure method and an exposure apparatus, a device manufacturing method, and an exposure system, and more specifically, for example, a semiconductor element, a liquid crystal display element, an imaging element such as a CCD, a plasma display element, a thin film magnetic head, and the like. Alignment method for aligning a pattern with a sensitive substrate such as a wafer, an exposure method and an exposure apparatus for exposing the sensitive substrate using the alignment method, and device manufacturing using the exposure method The present invention relates to a method and an exposure system including the exposure apparatus.

  In the manufacturing process of electronic devices such as semiconductor elements and liquid crystal display elements, a step-and-repeat type projection exposure apparatus (stepper) or a step-and-scan type projection exposure apparatus (scanner) is mainly used. ing.

  In this type of exposure apparatus, a plurality of patterns formed on a reticle are arranged on a substrate such as a wafer or a glass plate (hereinafter collectively referred to as a wafer) coated with a sensitive agent (photoresist) via a projection optical system. Transferred to each of the shot areas. In particular, in the case of a semiconductor element, since dozens of layers of patterns are formed on a wafer in a superimposed manner, high overlay accuracy (positioning accuracy) of patterns between the respective layers is required.

  Therefore, in the exposure process, in order to align the wafer, in recent years, a predetermined statistical calculation is performed using the detection result of the position information of some of the plurality of alignment marks, and the arrangement of all shot areas or the shot areas is determined. An enhanced global alignment (EGA) system that obtains wafer distortion (in-shot error) within a shot region with high accuracy in addition to the arrangement is widely employed (see, for example, Patent Document 1, Patent Document 2, etc.). .

  However, in wafer alignment (EGA) performed during the exposure process (lot processing), only linear components are considered for the in-shot error. On the other hand, the nonlinear component has been set as a fixed parameter for each lot of wafers (one lot is 25 sheets or 50 sheets). For this reason, a change in the temperature of the wafer while it is transported to the exposure apparatus, a difference in the adsorption state of each wafer by the stage (wafer holder), a change in the temperature of the immersion liquid (in the case of an immersion type exposure apparatus), or the heat of the reticle Due to expansion or the like, the nonlinear component may change during the lot processing, which may deteriorate the overlay accuracy. On the other hand, in order to correct the nonlinear component of the in-shot error for each wafer, more alignment marks are given in each shot area on the wafer, and the position information of the more alignment marks is obtained during the lot processing, for example, Measuring by bi-die (D / D) alignment causes a reduction in throughput.

  In view of such circumstances, the inventor has previously proposed an invention relating to an alignment method or the like that can reduce the influence on throughput and perform nonlinear correction of an optimal shot arrangement for each wafer (see Patent Document 3).

  However, the overlay accuracy required for the exposure apparatus is becoming increasingly strict, and a nonlinear error for each shot region (in the shot region) of the wafer causes an unacceptable level of overlay error. For this reason, it is now becoming difficult to sufficiently cope with this even with the invention described in Patent Document 3.

US Pat. No. 4,780,617 US Pat. No. 6,876,946 US Patent Application Publication No. 2006/0040191

  The present invention has been made under the circumstances described above. From a first viewpoint, the present invention is an alignment method for aligning a pattern with a plurality of partitioned regions arranged on an object, the plurality of partitioned regions. A model for detecting a plurality of marks attached to at least a part of the first partition region, obtaining a nonlinear component of the distortion of the object based on the detection result, and describing the distortion of the object based on the nonlinear component Determining a plurality of conditions for: detecting a number of marks less than the plurality of marks attached to at least a part of the plurality of divided areas, and, based on the detection result, out of the plurality of conditions And aligning the pattern with the plurality of partitioned regions using the model corresponding to the one condition.

  According to this, a plurality of marks (many marks) attached to at least a part of the first divided areas of the plurality of divided areas arranged on the object are detected in advance, and the detection result (the precise detection thereof) is detected. Based on the result, a nonlinear component of the distortion of the object is obtained, and a plurality of models (conditions for describing the nonlinear component of the distortion of the object) are determined based on the nonlinear component. When aligning, a smaller number of marks than the plurality of marks attached to at least a part of the second partitioned areas of the plurality of partitioned areas are detected and determined in advance based on the detection result. One of a plurality of models (conditions for) is selected, and the pattern corresponding to the one condition (selected model) is used to align a pattern with a plurality of partitioned regions. Without hardly lowering Tsu bets, it is possible to align with high accuracy the pattern in each divided area.

  From a second viewpoint, the present invention uses the alignment method of the present invention to align a pattern with a plurality of partitioned areas arranged on an object, and sequentially transfers the pattern to the plurality of partitioned areas. Exposure method.

  According to this, since the alignment method of the present invention is used, it is possible to sequentially transfer the pattern to a plurality of partitioned regions arranged on the object without reducing the throughput.

  According to a third aspect of the present invention, there is provided a device manufacturing method comprising: a step of forming a pattern on an object using the exposure method of the present invention; and a step of processing the object on which the pattern is formed. is there.

  According to a fourth aspect of the present invention, there is provided an exposure apparatus that sequentially transfers a pattern to a plurality of partitioned areas arranged on an object, the optical system projecting the pattern onto the object; and the plurality of partitioned areas A mark detection system for detecting a plurality of marks attached to the plurality of marks; detecting a part of the plurality of marks attached to at least a part of the plurality of partition regions using the mark detection system; One condition for a model describing a nonlinear component of the distortion of the object using a plurality of conditions determined in advance by detecting the plurality of marks attached to at least a part of the plurality of partitioned regions. An exposure apparatus comprising: a control system selected from among the plurality of divided regions using the model corresponding to the one condition.

  According to this, a part of a plurality of marks attached to at least a part of a plurality of partitioned areas is detected by the control system using a mark detection system, and a nonlinear component of object distortion is detected using the detection result. Is selected from a plurality of conditions determined in advance by detecting the plurality of marks attached to at least a part of the plurality of partition regions, and the selected condition is The pattern is aligned with the plurality of partitioned regions using the corresponding model. Therefore, even in a situation where a nonlinear component of distortion as described above occurs, the pattern can be aligned with high accuracy without reducing the throughput.

  According to a fifth aspect of the present invention, there is provided an exposure apparatus according to the present invention; a measuring apparatus for detecting a plurality of marks attached to a plurality of partitioned regions arranged on an object; For a model that detects a plurality of marks attached to at least a part of a plurality of partitioned areas, obtains a nonlinear component of the distortion of the object based on the detection result, and describes the distortion of the object based on the nonlinear component An exposure system comprising: a host computer that determines a plurality of conditions;

  According to this, many marks are detected in advance by the host computer using a measuring device, and a nonlinear component of the distortion of the object is obtained based on the detection result, and further, the distortion of the object is calculated based on the nonlinear component. A plurality of models (conditions for) describing nonlinear components are determined. Therefore, the exposure apparatus constituting the exposure system of the present invention can detect a small number of marks and select one from a plurality of models (conditions) determined in advance precisely based on rough results. Furthermore, the pattern can be aligned with high accuracy using the selected model.

1 is a diagram showing a schematic configuration of a lithography system according to an embodiment. FIG. 2 is a view showing a schematic configuration of one exposure apparatus constituting the lithography system of FIG. 1. FIG. 3A to FIG. 3J are diagrams for explaining a non-linear component of a typical in-shot error. It is a figure for demonstrating an example of arrangement | positioning of the sample shot of the multipoint EGA in a shot. FIG. 5A is a diagram showing an example of the arrangement of alignment marks provided in each shot area on the wafer measured in advance measurement, and FIG. 5B is a multi-point measurement in a shot during lot processing. It is a figure which shows an example of arrangement | positioning of the alignment mark provided in each shot area | region measured by (2). The series of processes including alignment and exposure, etc. of one lot of wafers in the exposure apparatus 110 _k, a flowchart showing the simplified processing algorithm of the main controller (internal CPU). 7 is a flowchart showing an example of step 508 (subroutine for high-order correction optimization measurement processing) in FIG. 6.

  Hereinafter, an embodiment of the present invention will be described with reference to FIGS.

FIG. 1 schematically shows a configuration of a lithography system 100 according to an embodiment suitable for applying the alignment method and the exposure method according to the present invention. The lithography system 100 includes N exposure apparatuses 110 _{1 to} 110 _N , an overlay measurement apparatus (hereinafter abbreviated as “measurement apparatus”) 130, a storage device 140, a terminal server 150, and a host computer (hereinafter “host”). 160) and the like.

Exposure apparatuses 110 _{1 to} 110 _N , measurement apparatus 130, and terminal server 150 are connected to a local area network (LAN) 170. A host 160 is connected to the terminal server 150. A storage device 140 is connected to the host 160 via a communication path 180 such as a scuzzy (SCSI). That is, on the hardware configuration, communication paths among the exposure apparatuses 110 _{1 to} 110 _N , the measurement apparatus 130, the terminal server 150, and the host 160 (and the storage device 140) are secured.

The exposure apparatuses 110 _{1 to} 110 _N are at least one scanning exposure apparatus, for example, a step-and-scan type projection exposure apparatus (that is, a scanner), and has an ability to adjust distortion of a projected image. projection exposure apparatus (hereinafter, unless otherwise necessary to distinguish, simply exposure apparatus called a) 110 k _(k is any of 1 to N) are included.

FIG. 2 shows a schematic configuration of the exposure apparatus 110 _k . The exposure apparatus 110 _k includes an illumination system IOP, a reticle stage RST that holds the reticle R, a projection optical system PL, a wafer stage WST on which the wafer W is placed, a control system for these, and the like.

  The illumination system IOP includes a light source and an illumination optical system connected to the light source via a light transmission optical system. The illumination optical system includes, for example, an illumination uniformizing optical system including an optical integrator, a beam splitter, a reticle blind, and the like (all not shown) as disclosed in, for example, US Patent Application Publication No. 2003/0025890. . Illumination system IOP uses illumination light IL to illuminate a slit-like illumination area defined by a reticle blind on reticle R on which a circuit pattern or the like is drawn (a long and narrow rectangular shape extending in the X-axis direction (the direction perpendicular to the plane of FIG. 2)). Illuminate with almost uniform illumination. As the illumination light IL, for example, ArF excimer laser light (wavelength 193 nm) (or KrF excimer laser light (wavelength 248 nm)) is used.

  Reticle stage RST is arranged below illumination system IOP in FIG. 2 (on the −Z side). On reticle stage RST, reticle R on which a pattern is formed is fixed, for example, by vacuum suction. Here, reticle stage RST can be finely driven in a horizontal plane (XY plane) by a reticle stage drive system 22 including a linear motor and the like, and has a predetermined stroke range in a scanning direction (here, Y-axis direction). It can be driven at the scanning speed specified in. Position information (including rotation information about the Z axis) of the reticle stage RST in the XY plane is, for example, about 0.25 nm by a reticle laser interferometer (hereinafter referred to as “reticle interferometer”) 16 via a movable mirror 15. Is always detected with a resolution of. Position information of reticle stage RST from reticle interferometer 16 is sent to main controller 50, which drives reticle stage RST via reticle stage drive system 22 based on the position information of reticle stage RST. (Control.

  Projection optical system PL is arranged below reticle stage RST in FIG. 2 (on the −Z side), and the direction of optical axis AX is the Z-axis direction. As the projection optical system PL, for example, only a lens element is used as an optical element, that is, includes a plurality of lens elements 27, 29, 30, 31,... Arranged at predetermined intervals along the optical axis AX direction. A constructed, for example, double-sided telecentric refractive system is used. The lens elements 27, 29, 30, 31... Are held inside the lens barrel 32. The projection magnification of the projection optical system PL is, for example, 1/5 (or 1/4). For this reason, when the reticle R is illuminated by the illumination light IL from the illumination system IOP, the illumination light IL that has passed through the reticle R passes through the projection optical system PL and is within the illumination area (illumination area) of the illumination light IL. A reduced image (partially inverted image) of the reticle R pattern is formed in an area (exposure area) conjugate to the illumination area on the wafer W coated with a resist (sensitizer) on the surface (pattern latent in the resist). An image is formed).

The exposure apparatus 110 _k is provided with an imaging characteristic correction apparatus for correcting imaging characteristics of the projection optical system PL, for example, various aberrations. This imaging characteristic correction device corrects changes in imaging characteristics of the projection optical system PL itself due to changes in atmospheric pressure, absorption of illumination light, and the like, and shot regions (for example, the previous layer) of a preceding specific layer on the wafer W ( Hereinafter, the projection image of the pattern of the reticle R is distorted according to the distortion of the pattern transferred to the shot). The imaging characteristics of the projection optical system PL include spherical aberration (aberration at the imaging position), coma aberration (aberration of magnification), astigmatism, curvature of field, distortion (distortion), etc. Has a function of correcting these various aberrations, but in the following description, the image formation characteristic correction device mainly relates to distortion of the projected image (transfer image) (in principle, including aberration of magnification). Only correction will be performed.

  In FIG. 2, the lens element 27 that is the closest to the reticle R constituting the projection optical system PL is fixed to the support member 28, and the lens elements 29, 30, 31,... Following the lens element 27 are lens mirrors of the projection optical system PL. It is fixed to the cylinder 32. The support member 28 is projected optically via a plurality of (three in this case) drive elements that can be expanded and contracted, for example, piezo elements 11a, 11b, and 11c (however, the drive element 11c on the back side of the paper is not shown in FIG. 2). It is connected to the lens barrel 32 of the system PL. The drive voltage applied to the drive elements 11a, 11b, and 11c is independently controlled by the imaging characteristic control unit 12, whereby the lens element 27 is arbitrarily inclined and optical axis with respect to the plane orthogonal to the optical axis AX. It is configured to be movable in the AX direction. The driving amount of the lens element 27 by each driving element is strictly measured by a position sensor (not shown), and the position is maintained at a target position by servo control. Note that the optical axis AX of the projection optical system PL indicates the common optical axis of a fixed lens element such as the lens element 29.

  In the above description, for the sake of convenience of explanation, it is assumed that only the lens element 27 is movable. However, actually, in the projection optical system PL, a plurality of lens elements or lens groups include the above lens. Like the element 27, it is configured to be movable. In addition, in the imaging characteristic correction apparatus, in addition to the configuration for correcting distortion of the projection image by driving a part of the optical elements constituting the projection optical system PL, A configuration in which the distortion of the projected image is corrected by controlling the gas pressure (adjusting the refractive index) may be employed.

  Wafer stage WST is arranged below projection optical system PL in FIG. 2 (on the −Z side). Wafer holder 9 is mounted on wafer stage WST. A wafer W is held on the wafer holder 9 by vacuum suction or the like.

  Wafer stage WST is driven with a predetermined stroke in the X-axis direction and Y-axis direction by wafer stage drive system 24 including a linear motor and the like, and also rotates in the Z-axis rotation direction (θz direction) and in the X-axis rotation direction. It is finely driven in the (θx direction) and rotation direction around the Y axis (θy direction). Therefore, the wafer holder 9 can be tilted in any direction with respect to the best imaging plane of the projection optical system PL by the wafer stage drive system 24, and can be finely moved in the optical axis AX direction (Z-axis direction). It is configured to be rotatable around a Z axis parallel to the axis AX. In place of wafer stage WST, a stage that can move (including rotation in the θz direction) in the XY plane, and a wafer holder 9 held on the stage, can be finely moved in the Z-axis direction, θx direction, and θy direction. A table may be used.

  Position information (including yawing (rotation in the θz direction) information) and tilt information (pitching (rotation in the θx direction) information and rolling (rotation in the θy direction) information) of the wafer stage WST in the XY plane are the wafer laser. An interferometer (hereinafter abbreviated as a wafer interferometer) 18 is constantly detected through a movable mirror 17 with a resolution of, for example, about 0.25 nm. The position information (or speed information) of wafer stage WST is sent to main controller 50, and main controller 50 determines the position of wafer stage WST in the XY plane (in the θz direction) based on the position information (or speed information). (Including rotation) is controlled via the wafer stage drive system 24.

  Further, a reference mark plate FM having a surface set substantially at the same height as the surface of wafer W is fixed to the upper surface of wafer stage WST. On the surface of the fiducial mark plate FM, a first fiducial mark for reticle alignment and a second fiducial mark for baseline measurement of the alignment system 8 described later are formed in a predetermined positional relationship.

  On the side surface of the projection optical system PL, there is an off-axis alignment system 8 for detecting an alignment mark (wafer mark) attached to each shot on the wafer W and a second reference mark on the reference mark plate FM. Is provided. As the alignment system 8, for example, an FIA (Field Image Alignment) system of an image processing system is used. In addition to the FIA system, a target mark is irradiated with coherent detection light, and scattered light or diffracted light generated from the target mark is detected, or two diffracted lights (for example, of the same order) generated from the target mark. Of course, it is possible to use an alignment sensor that detects the interference by interfering with each other alone or in combination. The detection result of the alignment system 8 is sent to the main controller 50 via an alignment signal processing system (not shown).

  Further, in the vicinity of the lower end portion of the projection optical system PL, a multipoint focal position for detecting position information (surface position information) regarding the Z-axis direction of the surface of the wafer W at the plurality of detection points in the exposure area and in the vicinity thereof. A detection system (13, 14) is provided. As the multipoint focal position detection system, for example, an oblique incidence type detection system disclosed in US Pat. No. 5,448,332 is used. The multipoint focal position detection system includes an irradiation optical system 13 that emits an imaged light beam obliquely with respect to the optical axis AX toward the best image formation plane of the projection optical system PL, and a reflected light beam from the surface of the wafer W as a slit. And a light receiving optical system 14 for receiving light through The wafer surface position information detected by the multipoint focal position detection system (13, 14) is supplied to the main controller 50. Based on the supplied wafer surface position information, main controller 50 drives wafer stage WST (wafer holder 9) in the Z-axis direction and the tilt direction via wafer stage drive system 24 to focus / Perform leveling control.

In addition, the exposure apparatus 110 _k is provided with a pair of reticle alignment systems (not shown) as disclosed in, for example, US Pat. No. 5,646,413 above the reticle stage RST. . The reticle alignment system is composed of a TTR (Through The Reticle) alignment system using light having the same wavelength as the illumination light IL. The detection signal of the reticle alignment system is supplied to main controller 50 via an alignment signal processing system (not shown).

The main control device 50 is composed of, for example, a microcomputer (or a workstation), and comprehensively controls each component of the exposure apparatus 110 _k . The main controller 50 also controls a coater / developer (not shown) (hereinafter referred to as “C / D”) provided in the exposure apparatus 110 _k . As shown in FIG. 1, the main controller 50 is connected to the LAN 170 and can communicate with main controllers of other exposure apparatuses 110 _{1 to} 110 _N (except 110 _k ).

Next, the operation of the exposure process in the exposure apparatus 110 _k will be briefly described.

  Prior to exposure, main controller 50 causes wafer W to be loaded onto wafer holder 9 using a wafer transfer system (not shown), reticle alignment and baseline measurement of alignment system 8, and wafer alignment (for example, multiple points in a shot) EGA) and other preparatory work is performed. The above reticle alignment, baseline measurement, and the like are disclosed in detail in the aforementioned US Pat. No. 5,646,413. Subsequent multi-point EGA within a shot is disclosed in detail in US Pat. No. 6,876,946. Here, the in-shot multipoint EGA is a statistic using the least square method disclosed in, for example, the above-mentioned US Pat. No. 6,876,946 using position detection data of a plurality of wafer alignment marks in a shot. This means an alignment method (details will be described later) for obtaining a deformation amount including the arrangement coordinates of all shots on the wafer W and the magnification (scaling), rotation, and orthogonality of each shot using a geometric method.

  Main controller 50 sequentially transfers the pattern of reticle R to all shots on wafer W by scanning exposure based on the results of reticle alignment and baseline measurement described above and the results of wafer alignment.

  In the scanning exposure for each shot on wafer W, main controller 50 monitors the position information from reticle interferometer 16 and wafer interferometer 18 while scanning reticle stage RST and wafer stage WST at their respective scanning start positions ( Move to the acceleration start position). Then, main controller 50 relatively drives both stages RST and WST in the Y-axis direction, but in opposite directions. Here, when both stages RST and WST reach their respective target speeds, the pattern area of reticle R starts to be illuminated by illumination light IL, and scanning exposure is started.

  Main controller 50 determines that the speed Vr of reticle stage RST in the Y-axis direction and the movement speed Vw of wafer stage WST in the Y-axis direction at a speed ratio corresponding to the projection magnification of projection optical system PL, particularly during the above-described scanning exposure. The reticle stage RST and wafer stage WST are synchronously controlled so as to be maintained. At this time, main controller 50 adjusts the synchronous drive of both stages RST and WST or drives lens element 27 via the imaging characteristic correction device to project the pattern of reticle R onto the wafer W. Correct distortion. Here, the adjustment of the synchronous drive of both stages RST and WST means that reticle R (reticle stage RST) and wafer W (wafer stage WST) at the time of scanning exposure via reticle stage drive system 22 and wafer stage drive system 24. And adjusting the speed ratio in the scanning direction, and slightly shifting the scanning direction of both stages RST and WST. According to the former, the magnification of the projected image in the scanning direction can be corrected, and according to the latter, the projected image can be distorted.

  Then, different areas of the pattern area of the reticle R are sequentially illuminated with the illumination light IL, and the illumination of the entire pattern area is completed, thereby completing the scanning exposure of the first shot on the wafer W. Thereby, the circuit pattern of the reticle R is reduced and transferred to the first shot via the projection optical system PL.

  When the scanning exposure for the first shot is completed, main controller 50 moves (steps) wafer stage WST to the scanning start position (acceleration start position) for the next second shot. Then, similarly to the above, the scanning exposure for the second shot is performed. Thereafter, the same operation is performed for the third and subsequent shots. In this manner, the stepping operation between shots and the scanning exposure operation for the shots are repeated, and the pattern of the reticle R is transferred to all shots on the wafer W by the step-and-can method.

Among other exposure apparatuses, a projection exposure apparatus which is a scanning exposure apparatus is configured in the same manner as the projection exposure apparatus 110 _k . The projection exposure apparatus comprising a static exposure apparatus is basically configured similarly to the projection exposure apparatus 110 _k in FIG. However, in these static exposure apparatuses, the respective reticle stages are configured so that they can only be finely driven in the X, Y, and θz directions in the XY plane, and the illumination light within the circular field of the projection optical system PL. The irradiation area of IL (corresponding to the above-described illumination area and exposure area) is larger than that of the scanning exposure apparatus of FIG. 2, and is different in that, for example, it has the same size as the scanning exposure range of the scanning exposure apparatus. This is because these projection exposure apparatuses perform exposure with both the wafer stage and the reticle stage stationary.

  Returning to FIG. 1, the measurement apparatus 130 measures the overlay state of the patterns formed on the wafer as described above. The measurement apparatus 130 detects the alignment mark formed on the wafer together with the pattern, similar to the alignment measurement (EGA) described above. Thereby, the relative position (position shift) between marks formed in different layers is obtained as an overlay error. The result is supplied to the host 160. The measuring device 130 is used when analyzing the overlay state of patterns formed on the original process layer prior to exposure on the current process layer.

The main control device and measurement device 130 provided in each of the exposure devices 110 _{1 to} 110 _N are connected to the LAN 170 and communicate with the host 160 via the LAN 170 and the terminal server 150. Here, the terminal server 150 is configured as a gateway processor for absorbing differences in communication protocols between the LAN 170 and the host 160. The function of the terminal server 150 enables communication between the host 160 and the exposure apparatuses 110 _{1 to} 110 _N and the measurement apparatus 130 connected to the LAN 170.

The host 160 is a large computer that centrally manages the lithography system 100. The host 160 periodically exposes exposure histories (for example, lot names of wafers processed by the exposure apparatuses, process program names, and processing times) sent from the exposure apparatuses 110 _{1 to} 110 _N and exposures. Information relating to the distortion of the projected images of the devices 110 _{1 to} 110 _N is stored in the storage device 140. The host 160 schedules the operation of all exposure apparatuses in the exposure process (including all device processing processes) of all wafers (lots) based on the exposure history.

In addition, the host 160 uses the result of wafer alignment (EGA) performed prior to exposure in the exposure apparatuses 110 _{1 to} 110 _N , the result of overlay error measured by the measurement apparatus 130, and the like to perform overlay. Correction information for eliminating the alignment error is created, and the created correction information is stored in the storage device 140. When the host 160 instructs the exposure apparatus 110 _k to expose the wafer, the host 160 reads out necessary correction information from the storage device 140 and transmits it to the exposure apparatus 110 _k . The creation of the correction information will be described later.

  The host 160 is provided with an input / output device 161 such as a display, which is a man-machine interface, a keyboard, and a mouse. The operator can also input the correction parameters described above via the input / output device 161.

  The LAN 170 can be either a bus type LAN or a ring type LAN. In this embodiment, the carrier type medium access / contention detection (CSMA / CD) type bus type LAN of the IEEE 802 standard is adopted. Yes.

In the lithography system 100 of the present embodiment, particularly when a semiconductor device is manufactured, dozens of layers of patterns are overlaid on the wafer, so that high overlay accuracy (positioning accuracy) of the patterns between the layers is high. Required. Therefore, the exposure apparatus 110 _k adjusts the synchronous drive of both stages RST and WST according to the correction information transmitted from the host 160 together with the exposure instruction, and also uses the imaging characteristic correction apparatus to correct the distortion of the pattern projection image. Correct and execute the exposure process.

Here, as described above, the correction information transmitted from the host 160 includes wafer alignment (multi-shot EGA in the shot) performed by the exposure apparatuses 110 _{1 to} 110 _N performed prior to exposure and measurement of overlay error by the measurement apparatus 130 as described above. Etc., based on the result of the above. Here, in the in-shot multipoint EGA, only a correctable linear component is considered with respect to wafer distortion (in-shot error) in each shot to which a pattern is transferred. On the other hand, the nonlinear component has been conventionally considered for each lot (one lot is 25 or 50 wafers).

  However, a stage grid error between different exposure apparatuses (units), a difference in wafer adsorption mechanism for each exposure apparatus, a difference in stage adsorption state for each wafer, and a change in wafer temperature while being transferred from the C / D to the exposure apparatus However, due to thermal expansion of the reticle or the like, the nonlinear component of wafer distortion (in-shot error) changes during lot processing, which may deteriorate the overlay accuracy.

3A to 3J show typical in-shot error components. Here, dx and dy are in-shot errors in the X-axis direction and the Y-axis direction, respectively. 3A shows a so-called offset component (dx = K ₁ , dy = K ₂ ), and FIG. 3B shows a magnification component (dx = K ₃ x, dy = K ₄ y). C) shows rhombus formation (dx = K ₅ y, dy = K ₆ x), and FIG. 3D shows eccentric magnification components (dx = K ₇ x ² , dy = K ₈ y ² ). 3 (E) shows the base formation (dx = K ₉ xy, dy = K ₁₀ xy), and FIG. 3 (F) shows the fan-shaped components (dx = K ₁₁ y ² , dy = K ₁₂ x ² ). 3 (G) shows a C-shaped magnification component (dx = K ₁₃ y ³ , dy = K ₁₄ x ³ ), and FIG. 3 (H) shows an accordion component (dx = K ₁₅ x ² y, dy = K). ₁₆ xy ² ), the C-shaped distortion component (dx = K ₁₇ xy ² , dy = K ₁₈ x ² y) in FIG. 3 (I), and the river flow component (dx = K) in FIG. 3 (J). ₁₉ y , _{Dy = K} 20 ^{x 3)} are shown, respectively. The components shown in FIGS. 3D to 3J are the above-described nonlinear components.

  Therefore, in the exposure based on the mark measurement result in the in-shot multipoint EGA, it is also necessary to consider the nonlinear component of the in-shot error. Here, in order to accurately consider the nonlinear component, it is necessary to increase the number of measurement points in the shot, that is, to add more alignment marks and perform the measurement. However, measuring lots of alignment marks for each wafer (and also for each shot) during lot processing in order to correct the nonlinear component of in-shot error will drastically reduce throughput. It cannot be adopted as it is.

  Therefore, in the present embodiment, the non-linear component of the in-shot error is corrected for each wafer and for each shot without reducing the throughput by the method described below, thereby ensuring high overlay accuracy. It is said.

In the lithography system 100 of this embodiment, prior measurement is performed prior to wafer processing by the exposure apparatus 110 _k . In the preliminary measurement, at least one set of correction information (nonlinear error correction information) for correcting the nonlinear component of the in-shot error is created.

  In the pre-measurement, first, in accordance with an instruction from the host 160 (or operator), an alignment mark (wafer mark for wafer alignment (multi-point EGA in shot)) with respect to the wafer to be exposed (wafer in lot) is measured by the measuring device 130. ) Is measured. The details of the shot multi-point EGA are disclosed in US Pat. No. 6,876,946.

  FIG. 4 shows a plurality of shots arranged on the wafer W. Measurement of alignment marks (wafer marks) for in-shot multipoint EGA is performed for all or part of these shots. When the alignment mark (wafer mark) is measured for only a part of the shots, at least the shots shown in FIG. 4 (rough measurement target shots described later) Sa1, Sa2, Sb1, Sb2, Sc1, Sc2 , Sd1 and Sd2. A shot in which alignment marks (wafer marks) for multi-point EGA in a shot are measured is called a sample shot.

In the pre-measurement, a plurality of alignment marks M _ij (i, j = 1 to 8) given in each shot are detected in the arrangement shown in FIG. 5A for the sample shot. In the present embodiment, the shot is square. A total of 64 alignment marks M _ij (i, j = 1 to 8) are provided in the shot, 8 in the X-axis direction and 8 in the Y-axis direction. The interval between the X-axis direction and the Y-axis direction is ΔD. The detection points (alignment marks) are desirably arranged evenly in the shot, but are not necessarily limited to this. The measurement device 130 detects the positions (X position, Y position) ξ _ij and ζ _ij of the alignment mark M _ij (i, j = 1 to 8) on the wafer coordinate system (or stage coordinate system). The detection result is transferred to the host 160.

The host 160 uses the detection results ξ _ij and ζ _ij of the alignment mark M _ij (i, j = 1 to 8), and uses nonlinear error correction information (also simply referred to as correction information) for correcting the nonlinear component of the in-shot error. ).

  Prior to creation, a model (correction model) for creating correction information, for example, an EGA calculation model and a shot high-order correction model (including conditions for the order and coefficient pair) are specified. As the EGA calculation model, a 6-parameter linear model given by the following equation (1), a 10-parameter linear model given by the following equation (2), an in-shot averaging model, etc. are adopted.

ここで、（Ｘ，Ｙ）はアライメントマークの実際の位置（検出位置）、（ｘ，ｙ）はアライメントマークの設計上の位置、（ｘｃ，ｙｃ）はショット中心の設計上の位置、（Ｏｘ，Ｏｙ）はウエハオフセット、Θはウエハローテーション、（Γｘ，Γｙ）はウエハスケーリング、Ωはウエハ直交度、（γｘ，γｙ）はショットスケーリング、θはショットローテーション、ωはショット内直交度である。これらのパラメータを、アライメント補正パラメータと総称する。なお、ショット内に付与されるアライメントマークの数（検出点の数）に応じて、指定できるモデルが限定される。ショット高次補正モデルについては後述する。

Here, (X, Y) is the actual position (detection position) of the alignment mark, (x, y) is the design position of the alignment mark, (xc, yc) is the design position of the shot center, and (Ox , Oy) is wafer offset, Θ is wafer rotation, (Γx, Γy) is wafer scaling, Ω is wafer orthogonality, (γx, γy) is shot scaling, θ is shot rotation, and ω is in-shot orthogonality. These parameters are collectively referred to as alignment correction parameters. Note that the models that can be specified are limited according to the number of alignment marks (number of detection points) provided in the shot. The shot higher-order correction model will be described later.

The host 160 applies the specified calculation model and removes the wafer component (shot arrangement error) from the detection results ξ _ij and ζ _ij (i, j = 1 to 8). First, the host 160 uses the detection results ξ _ij , ζ _ij to minimize the alignment residual error ε = Σ _{ij = 1 to 8} ((ξ _ij −X _ij ) ² + (ζ _ij −Y _ij ) ² ). Thus, the least square method is applied to determine the alignment correction parameter in the model. Here, X _ij and Y _ij are design positions x _ij and y _ij (i, j = 1 to 8) of the alignment make M _ij (i, j = 1 to 8) transformed by applying the calculation model. ). That is, for example, X and Y on the left side obtained by substituting the design positions x _ij and y _ij into x and y on the right side of Formula (1) or Formula (2). Note that the design positions x _ij and y _ij are obtained in advance.

The host 160 uses the determined alignment correction parameter to remove the wafer component (shot arrangement error) from the detection results ξ _ij , ζ _ij (i, j = 1 to 8), and to obtain a residual error (that is, an in-shot error). dξ _ij = ξ _ij −X ′ _ij , dζ _ij = ζ _ij −Y ′ _ij (i, j = 1 to 8) is obtained. Here, X ′ _ij and Y ′ _ij are design positions of alignment marks M _ij (i, j = 1 to 8) converted by applying a calculation model, for example, Expression (1) or Expression (2). x _ij , y _ij (i, j = 1 to 8). However, the alignment correction parameter obtained above is substituted into the calculation model. The in-shot errors dξ _ij and dζ _ij obtained in this way include only nonlinear components.

  If a higher-order component is included in the wafer component (shot arrangement error), it is similarly removed. Details of handling higher order components are disclosed, for example, in US 2006/0040191.

The host 160 obtains in-shot errors dξ _ij and dζ _ij for alignment marks M _ij in all sample shots. Then, for each mark M _ij , the in-shot errors dξ _ij and dζ _ij are averaged. Here, jump data (data whose value is extremely different from others) is excluded. Hereinafter, the averaged in-shot errors are referred to as in-shot errors dξ _ij and dζ _ij unless there is a possibility of confusion.

The host 160 applies the previously specified shot high-order correction model (including the conditions for the order and coefficient pair), and calculates the in-shot errors dξ _ij and dζ _ij (i, j = 1 to 8), an in-shot high-order correction coefficient (also simply referred to as a coefficient) is obtained. Here, the shot high-order correction model is given as, for example, a two-variable polynomial of the positions X and Y of the following alignment marks (detection points).

dx = K ₁ + K ₅ Y + K ₁₁ Y ² + K ₁₉ Y ³ + K ₂₉ Y ⁴ + K ₄₁ Y ⁵ + K ₅₅ Y ⁶ (3)
dy = K ₂ + K ₄ Y + K ₆ X + K ₈ Y ² + K ₁₀ XY + K ₁₄ Y ³ + K ₁₆ XY ²
+ K ₂₂ Y ⁴ + K ₂₄ XY ³ + K ₃₂ Y ⁵ + K ₃₄ XY ⁴ + K ₄₄ Y ⁶ + K ₄₆ XY ⁵ (4)
Here, the combination of the order of the polynomial and the included coefficient is designated as the condition of the correction model. For example, in the case of the above example, the order is 6, and the coefficient sets K ₁ , K ₂ , K ₄ , K ₅ , K ₆ , K ₈ , K ₁₀ , K ₁₁ , K ₁₄ , K ₁₆ , K ₁₉ , K ₂₂ , Designated as K ₂₄ , K ₂₉ , K ₃₂ , K ₃₄ , K ₄₁ , K ₄₄ , K ₄₆ , K ₅₅ .

When specifying conditions for a shot high-order correction model, particularly, a set of coefficients, it is preferable to remove terms with strong correlation (fix the corresponding coefficients to zero) in order to stably calculate the high-order coefficients. For example, as in equation (4), the X ² Y term and the X ³ term are excluded from the ^third-order term. It should be noted that the terms that cannot be corrected due to the limitation of the function of the exposure apparatus 110 _k are excluded in advance.

The host 160 uses the in-shot errors dξ _ij and dζ _ij to minimize the square error E = Σ _{ij = 1 to 8} W _ij ((dξ _ij −dx _ij ) ² + (dζ _ij −dy _ij ) ² ). As such, the in-shot high-order correction coefficient is determined by applying the method of least squares. Here, W _ij is a weight for the detection point ij. Dx _ij and dy _ij are the positions of the alignment marks (detection points) M _ij (i, j = 1 to 8) in the shot, the positions X _ij and Y _ij of the shot higher-order correction model, for example, Expression (3) and Expression It is obtained by substituting into (4).

The host 160 follows the above-described procedure from the detection results ξ _ij and ζ _ij (i, j = 1 to 8) from the specified correction model (EGA calculation model and shot high-order correction model (for the order and coefficient pair). In-shot high-order correction coefficient is determined every time. The determined in-shot high-order correction coefficient is recorded (registered) in the storage device 140 in association with the designated correction model.

  A set of coefficients corresponding to one correction model is referred to as nonlinear error correction information (or simply correction information). The number of correction models is not limited to one, and a plurality of correction models are designated, and a plurality of correction information is created correspondingly. Of the correction information created for each of the plurality of correction models, the correction information that gives the smallest square error E is further recorded (registered) in the storage device 140 as the best correction information. Thereby, the preliminary measurement is completed.

  In addition, a certain tendency is often observed in the in-shot error according to the lot processing condition (process condition). Therefore, pre-measurement is preferably performed for each lot.

In wafer processing (lot processing) by the exposure apparatus 110 _k, alignment of the wafer with respect to the reticle pattern at the time of exposure is performed using the correction information created (registered) in the above prior measurement.

Next, a series of processes including alignment and exposure of one lot of wafers in the exposure apparatus 110 _k is shown in a flowchart of FIG. 6 in which the processing algorithm of the main controller 50 (inner CPU) is simplified, and This will be described with reference to other drawings as appropriate.

  First, in step 502, the main controller 50 waits for the EGA measurement sample shot (EGA measurement shot) and the EGA calculation model to be specified. In the present embodiment, eight shots Sa1, Sa2, Sb1, Sb2, Sc1, Sc2, Sd1, and Sd2 shown in FIG. 4 are selected as EGA measurement shots. As the EGA calculation model, one of the above-described 6-parameter linear model, 10-parameter linear model, in-shot averaging model, and the like is designated. It is assumed that the model specified here is always a zero-order (no correction) or first-order (linear) correction model for the in-shot error.

  When the EGA measurement shot and the EGA calculation model are designated by the operator, the process proceeds to the next step 504, and the main control device 50 detects the rough measurement target shot for the high-order correction in the shot, and the detection point in the target shot. Wait for the number and placement of shots and the shot higher-order correction model (including conditions for order and coefficient pairs) to be specified. Here, as an example, the same shots Sa1, Sa2, Sb1, Sb2, Sc1, Sc2, Sd1, and Sd2 as the EGA measurement shot are selected as the rough measurement target shots. From the viewpoint of improving the throughput, it is preferable to select the rough measurement target shot so that the EGA measurement shot is included. As the shot high-order correction model, for example, the above-described equations (3) and (4) are specified. Here, the condition for the combination of the order and the coefficient is limited by the number of detection points. It should be noted that a plurality of shot high-order correction models (including conditions for the order and coefficient pair) can be specified.

In the present embodiment, in the above-described preliminary measurement, for example, as shown in FIG. 5A, 64 alignment marks in a shot are detected, whereas in-shot multipoint measurement ( In the in-shot rough measurement), as few alignment marks as possible are detected. For example, as shown in FIG. 5B, a total of 16 alignment marks m _ij (i, j = 1 to 4) are detected, four in the X-axis direction and four in the Y-axis direction. Here, the interval Δd of the alignment marks in the X-axis direction and the Y-axis direction is sufficiently larger than the interval ΔD of the alignment marks M _ij (i, j = 1 to 8) used in the previous preliminary measurement. (Δd >> ΔD). It is desirable that the detection points (alignment marks) are evenly arranged in the shot. Further, a part of the alignment mark M _ij (i, j = 1 to 8) is selected as the sample mark m _ij of the multi-point EGA in the shot during the lot processing. However, this is not always necessary.

  As described above, if the error in a shot changes during lot processing, the operator can change the error within the shot for each wafer, for each shot, or for each shot, according to the change in the tendency of the error for each wafer or shot. A rough measurement target shot for high-order correction optimization and the number and arrangement of detection points in the rough measurement target shot are designated. It is also possible for the main controller 50 to automatically specify these.

  When the rough measurement target shot, the number and arrangement of detection points in the target shot, and the shot higher-order correction model (including conditions for the order and coefficient pair) are specified by the operator, main controller 50 Then, the process proceeds to the next step 506, and it is determined whether or not the in-shot high-order correction optimization function is valid (on) or not (off). If the determination in step 506 is affirmative, that is, if the in-shot high-order correction optimization function is on, the process proceeds to a high-order correction optimization measurement process subroutine in step 508.

  In the subroutine of step 508, as shown in FIG. 7, main controller 50 first determines in step 602 whether or not the shot to be processed is a rough measurement target shot. If the determination in step 602 is negative, the main controller 50 determines in step 610 whether the shot to be processed is an EGA sample shot. If this determination is affirmative, main controller 50 measures the sample mark attached to the sample shot in step 612 and then proceeds to step 614. Here, in step 612, a sample mark corresponding to the EGA calculation model previously specified in step 502 is measured. Here, the measurement of the mark means that the main controller 50 detects the position of the mark relative to the index center using the alignment system 8, and based on the detection result and the measurement value of the wafer interferometer 18 at the time of detection. Means to obtain the position coordinates of the mark in a predetermined reference coordinate system (for example, a stage coordinate system).

  On the other hand, if the determination in step 610 is negative, that is, if the shot to be processed is not an EGA sample shot, the main controller 50 immediately proceeds to step 614.

  On the other hand, if the determination in step 602 is affirmed, the main controller 50 determines in step 604 whether or not the shot to be processed is an EGA sample shot. If this determination is affirmative, the main controller 50 determines that all rough measurement target marks and sample marks (here, for example, FIG. 5 ()) attached to the rough measurement target shot and the EGA sample shot. After measuring the 16 rough measurement target marks (at least a part of which also serves as EGA sample marks) shown in B), the process proceeds to step 614. When there is a sample mark in addition to the rough measurement target mark, in step 604, main controller 50 performs measurement of the rough measurement target mark and the sample mark.

  On the other hand, if the determination in step 604 is negative, main controller 50 executes measurement of all rough measurement target marks attached to the rough measurement target shot in step 608, and then proceeds to step 614. To do.

  In step 614, main controller 50 determines whether or not mark measurement has been completed for all measurement target shots. If this determination is negative, control returns to step 602.

  When the mark measurement for all shots to be measured is completed, main controller 614 ends the mark measurement process and proceeds to step 616. In step 616, main controller 50 performs a calculation based on the EGA calculation model previously specified in step 502 using the measurement results of the position information of all the sample marks, and performs the calculation on all the wafers to be processed. An array of shots (wafer parameters) or an array of all shots (wafer parameters) and shot parameters are obtained.

Next, in step 618, main controller 50 obtains in-shot errors dξ _ij and dζ _ij from the measurement results of all rough measurement target marks for each rough measurement target shot, and obtains them. Based on the tendency of the in-shot error (non-linear error), correction information that gives the minimum square error E is selected from a plurality of non-linear error correction information registered in advance measurement, and is obtained from the host 160. The calculation of the square error E is as described above. Here, when the required overlay accuracy differs depending on the position in the shot, the square error E may be obtained by giving a different weight W _ij to each detection point ij.

  When the process of step 618 ends, the subroutine process ends, and the process returns to step 514 of the main routine of FIG.

  On the other hand, if the determination in step 506 is negative, that is, if the in-shot high-order correction optimization function is off, main controller 50 performs the processing of the normal EGA measurement subroutine of step 510 and then step 514. Here, in the subroutine of step 510, measurement of the EGA sample mark and EGA calculation based on the measurement result are performed and the above-described alignment correction parameter is obtained.

  In step 514, main controller 50 applies the in-shot high-order correction coefficient to the EGA result to expose the shot to be processed. Here, if the in-shot high-order correction optimization function in step 506 is off, in this step 514, the main controller 50 adds the linear correction obtained in step 510 to the EGA calculation model specified in step 502. By applying the information (the alignment correction parameter described above), the best correction information (in-shot high-order correction information obtained from the host) is added to the shot high-order correction model (order and correction term to be used) specified in step 504. Apply information registered in advance measurement). Based on these EGA calculation model and shot high-order correction model, the position coordinates on the stage coordinate system of the shot to be processed on the wafer are calculated. Here, as described above, by applying the EGA calculation model, the linear components of the shot arrangement error and the intra-shot error are corrected, and by applying the shot higher-order correction model, the nonlinear component of the intra-shot error is corrected. Based on this calculation result and the results of reticle alignment and baseline measurement, the lens element 27 is driven via the imaging characteristic correction device, and the synchronous driving of the reticle stage RST and the wafer stage WST is slightly corrected. The above-described scanning exposure is executed. Then, the pattern of the reticle R is sequentially transferred to all shots on the wafer W. The details of driving the lens element for correcting the distortion of the projected image are described in, for example, US Pat. No. 5,117,255, and the details of the stage driving for correcting the distortion of the projected image are, for example, For example, it is disclosed in US Pat. No. 6,235,438.

  On the other hand, even when the in-shot high-order correction optimization function is on, exposure of the processing target shot is performed in the same procedure as described above. However, when calculating the position coordinates on the stage coordinate system of the shot to be processed on the wafer in step 514, the main controller 50 converts the EGA calculation model specified in step 502 into the linearity obtained in step 616. The correction information (the alignment correction parameter described above) is applied, and the correction information selected in step 618 is applied to the shot higher-order correction model specified in step 504.

  Then, when the exposure for all shots on the wafer is completed, the process proceeds to step 516. In step 516, main controller 50 determines whether or not the exposure of all wafers in the lot has been completed. If this determination is negative, control returns to step 506 and repeats the processing from step 506 onward. On the other hand, when the exposure of all the wafers in the lot is completed, the determination in step 516 is affirmed, and the series of processes of this routine is ended.

  In step 618, when the tendency of the error in the shot differs depending on the area on the wafer, for example, when the tendency differs for each of the areas Sa, Sb, Sc, and Sd shown in FIG. 4, optimum correction information is selected for each area. Good. In this case, in step 504, as the rough measurement target shots, the shots Sa1 and Sa2 for the region Sa, the shots Sb1 and Sb2 for the region Sb, the shots Sc1 and Sc2 for the region Sc, and the shot Sd1 for the region Sd. , Sd2 is specified. Furthermore, when the tendency of error in a shot differs for each shot, the region is defined by including only one shot in each region.

As described above in detail, according to the lithography system 100 of the present embodiment, the host 160 uses the measuring apparatus 130 in advance to detect a number of alignment marks attached to a plurality of shots on the wafer. Based on the detection result, a plurality of models (conditions) for accurately describing a nonlinear component of wafer distortion are determined. Then, the wafer processing by the exposure apparatus 110 _k (lot processing) times, the main controller 50 to the exposure apparatus 110 _k comprises detects the alignment mark of the small number, based on the detection result, the plurality of shots on the wafer Obtain an array and select one of a plurality of predetermined models (conditions for the selected model), and align the reticle pattern on the plurality of shots using the obtained array and the selected model. . As a result, it is possible to align and transfer the pattern to each shot with high accuracy without substantially reducing the throughput.

In the above embodiment, the pre-measurement performed by the host 160, the main controller 50 of exposure apparatus 110 _k comprises at lot process wafer alignment or the like is performed, the present invention by the host 160 and the main control unit 50 The case where the alignment method is executed has been described. However, the present invention is not limited to this, and the alignment method of the present invention can be applied to any system including an apparatus that requires highly accurate alignment for each of a plurality of shot areas on a wafer. The processing executed by the host 160 may be executed by the main controller 50.

  Further, in the above-described embodiment, the alignment method of the present invention is applied to the global alignment type wafer alignment. The present invention is not limited to this, and it is also possible to apply to wafer alignment of a die-by-die (D / D) alignment method.

  In the above-described embodiment, the case where the present invention is applied to a dry type exposure apparatus that exposes the wafer W without using liquid (water) has been described. No. 99/49504, European Patent Application Publication No. 1,420,298, International Publication No. 2004/055803, US Pat. No. 6,952,253, etc. The present invention is also applied to an immersion exposure apparatus that forms an immersion space including an optical path of illumination light between the wafer and the wafer, and exposes the wafer with illumination light through the liquid in the projection optical system and immersion space. Can do. In the immersion exposure apparatus, the temperature change of the immersion liquid also causes the fluctuation of the non-linear component of the wafer distortion (in-shot error) during the lot processing. Therefore, by applying the alignment method and the exposure method of the present invention, Further, it becomes possible to correct a non-linear component of wafer distortion (in-shot error) caused by the temperature change of the immersion liquid.

  In the above embodiment, the case where the present invention is applied to a scanning exposure apparatus such as a step-and-scan method has been described. However, the present invention is not limited to this, and the present invention is applied to a stationary exposure apparatus such as a stepper. May be. The present invention can also be applied to a step-and-stitch type reduction projection exposure apparatus, a proximity type exposure apparatus, or a mirror projection aligner that combines shots. Further, as disclosed in, for example, US Pat. No. 6,590,634, US Pat. No. 5,969,441, US Pat. No. 6,208,407, etc. The present invention can also be applied to a multi-stage type exposure apparatus provided with a stage. In addition, as disclosed in, for example, US Patent Application Publication No. 2007/0127006, etc., an exposure including a measurement stage including a measurement member (for example, a reference mark and / or a sensor) separately from the wafer stage. The present invention can also be applied to an apparatus.

  Further, the projection optical system in the exposure apparatus of the above embodiment may be not only a reduction system but also any of the same magnification and enlargement systems, and the projection optical system PL may be any of a reflection system and a catadioptric system as well as a refraction system. The projected image may be either an inverted image or an erect image. In addition, the illumination area and the exposure area described above are rectangular in shape, but the shape is not limited to this, and may be, for example, an arc, a trapezoid, or a parallelogram.

The light source of the exposure apparatus of the above embodiment is not limited to the ArF excimer laser, but is a KrF excimer laser (output wavelength 248 nm), F ₂ laser (output wavelength 157 nm), Ar ₂ laser (output wavelength 126 nm), Kr ₂ laser ( It is also possible to use a pulse laser light source with an output wavelength of 146 nm, an ultrahigh pressure mercury lamp that emits a bright line such as g-line (wavelength 436 nm), i-line (wavelength 365 nm), and the like. A harmonic generator of a YAG laser or the like can also be used. In addition, as disclosed in, for example, U.S. Pat. No. 7,023,610, a single wavelength laser beam in an infrared region or a visible region oscillated from a DFB semiconductor laser or a fiber laser as vacuum ultraviolet light, For example, a harmonic which is amplified by a fiber amplifier doped with erbium (or both erbium and ytterbium) and wavelength-converted into ultraviolet light using a nonlinear optical crystal may be used.

  In the above embodiment, it is needless to say that the illumination light IL of the exposure apparatus is not limited to light having a wavelength of 100 nm or more, and light having a wavelength of less than 100 nm may be used. For example, in recent years, in order to expose a pattern of 70 nm or less, EUV (Extreme Ultraviolet) light in a soft X-ray region (for example, a wavelength region of 5 to 15 nm) is generated using an SOR or a plasma laser as a light source, and its exposure wavelength Development of an EUV exposure apparatus using an all-reflection reduction optical system designed under (for example, 13.5 nm) and a reflective mask is underway. In this apparatus, a configuration in which scanning exposure is performed by synchronously scanning a mask and a wafer using arc illumination is conceivable, and therefore the present invention can be suitably applied to such an apparatus. In addition, the present invention can be applied to an exposure apparatus using a charged particle beam such as an electron beam or an ion beam.

  In the above-described embodiment, a light transmission mask (reticle) in which a predetermined light-shielding pattern (or phase pattern / dimming pattern) is formed on a light-transmitting substrate is used. Instead of this reticle, For example, as disclosed in US Pat. No. 6,778,257, an electronic mask (variable shaping mask, which forms a transmission pattern, a reflection pattern, or a light emission pattern based on electronic data of a pattern to be exposed, as disclosed in US Pat. No. 6,778,257. For example, a non-light emitting image display element (spatial light modulator) including a DMD (Digital Micro-mirror Device) may be used.

  Further, for example, the present invention can be applied to an exposure apparatus (lithography system) that forms line and space patterns on a wafer by forming interference fringes on the wafer.

  Further, as disclosed in, for example, US Pat. No. 6,611,316, two reticle patterns are synthesized on a wafer via a projection optical system, and one scan exposure is performed on one wafer. The present invention can also be applied to an exposure apparatus that performs double exposure of shots almost simultaneously.

  Note that the object on which the pattern is to be formed in the above embodiment (the object to be exposed to the energy beam) is not limited to the wafer, but other objects such as a glass plate, a ceramic substrate, a film member, or a mask blank. But it ’s okay.

  The use of the exposure apparatus is not limited to the exposure apparatus for semiconductor manufacturing. For example, an exposure apparatus for liquid crystal that transfers a liquid crystal display element pattern onto a square glass plate, an organic EL, a thin film magnetic head, an image sensor ( CCDs, etc.), micromachines, DNA chips and the like can also be widely applied to exposure apparatuses. Further, in order to manufacture reticles or masks used in not only microdevices such as semiconductor elements but also light exposure apparatuses, EUV exposure apparatuses, X-ray exposure apparatuses, electron beam exposure apparatuses, etc., glass substrates or silicon wafers, etc. The present invention can also be applied to an exposure apparatus that transfers a circuit pattern.

  An electronic device such as a semiconductor element includes a step of designing a function / performance of the device, a step of manufacturing a reticle based on the design step, a step of manufacturing a wafer from a silicon material, and the exposure apparatus (pattern forming apparatus) of the above-described embodiment. ) A lithography step for transferring a mask (reticle) pattern onto a wafer, a development step for developing the exposed wafer, an etching step for removing exposed members other than the portion where the resist remains by etching, and etching is completed. It is manufactured through a resist removal step for removing unnecessary resist, a device assembly step (including a dicing process, a bonding process, and a package process), an inspection step, and the like. In this case, in the lithography step, the exposure method described above is executed using the exposure apparatus of the above embodiment, and a device pattern is formed on the wafer. Therefore, a highly integrated device can be manufactured with high productivity.

  The alignment method of the present invention is suitable for aligning a pattern with each shot provided on an object. The exposure method and the exposure apparatus of the present invention are suitable for accurately overlapping and forming a pattern on an object. The device manufacturing method and the lithography system of the present invention are suitable for manufacturing an electronic device such as a semiconductor element or a liquid crystal display element.

DESCRIPTION OF SYMBOLS 8 ... Alignment system, 12 ... Imaging characteristic control part, 16 ... Reticle interferometer, 18 ... Wafer interferometer, 22 ... Reticle stage drive system, 24 ... Wafer stage drive system, 50 ... Main controller, 100 ... Lithography system, 110 _{1 to} 110 _N (110 _k ): exposure apparatus (scanning exposure apparatus), 130: overlay measurement apparatus, 160: host computer, IOP: illumination system, PL: projection optical system, R: reticle, RST: reticle stage , W ... wafer, WST ... wafer stage.

Claims (26)

An alignment method for aligning a pattern with a plurality of partitioned regions arranged on an object,
Detecting a plurality of marks attached to at least a part of the first partition region of the plurality of partition regions, obtaining a nonlinear component of distortion of the object based on the detection result, and determining the non-linear component of the object based on the nonlinear component Determining a plurality of conditions for the model describing the distortion;
Detecting a smaller number of marks than the plurality of marks attached to at least a portion of the plurality of partitioned areas, and selecting one of the plurality of conditions based on the detection result; Aligning the pattern with the plurality of partitioned regions using the model corresponding to one condition;
An alignment method including:   In the alignment step, the nonlinear component is obtained based on the detection result of the small number of marks, and the one condition for the model that most accurately describes the nonlinear component is selected from the plurality of conditions. The alignment method according to claim 1.   The alignment method according to claim 1 or 2, wherein, in the alignment step, the one condition is selected for each of the object, the plurality of partitioned regions, and one partitioned region.   In the aligning step, a plurality of marks attached to at least a part of the third partition region of the plurality of partition regions are detected, a linear component of the distortion of the object is obtained based on the detection result, and the linear 4. The pattern is aligned by correcting a linear component of the distortion of the object based on a component and correcting the nonlinear component using the model corresponding to the one condition. The alignment method according to one item.   The alignment method according to claim 4, wherein the third partition region is included in the second partition region.   The first and second partition regions are based on at least one of the detection results of the plurality of marks obtained in the determining step and the detection results of the small number of marks obtained in the positioning step. The alignment method according to claim 1, wherein the alignment method is selected from a plurality of partitioned regions.   The alignment method according to claim 1, wherein the second partition region is included in the first partition region. In the determining step, one condition is further specified from the plurality of determined conditions,
The alignment method according to claim 1, wherein in the alignment step, the pattern is aligned using the model corresponding to the specified condition.   The alignment method according to claim 8, wherein the specified condition is one of the plurality of conditions for the model that most accurately describes a detection result of the plurality of marks.   In the determining step, a plurality of marks attached to at least a part of the plurality of partitioned regions are detected, a linear component of the distortion of the object is obtained based on the detection result, and the nonlinear component is used using the linear component The alignment method as described in any one of Claims 1-9 which calculates | requires a component. The plurality of partitioned areas are classified into a plurality of sets,
The said determining step detects a plurality of marks attached to at least a part of a partition area included in the set for each of the plurality of sets and determines a plurality of conditions for the model. The alignment method as described in any one of these.   In the positioning step, for each of the plurality of sets, a number of marks smaller than the plurality of marks provided in at least some of the partition regions included in the set is detected, and the set is set based on the detection result. The alignment method according to claim 11, wherein one of the corresponding plurality of conditions is selected. The model is given as a polynomial,
The alignment method according to claim 1, wherein the condition includes at least one of a type of the polynomial, an order of the polynomial, and a combination of terms included in the polynomial.   Exposure using the alignment method according to any one of claims 1 to 13 to align a pattern with a plurality of partitioned areas arranged on an object and sequentially transfer the pattern to the plurality of partitioned areas. Method. The pattern is projected onto the object via an optical system;
The exposure method according to claim 14, wherein the pattern is aligned by adjusting the optical system. The object is held by a movable body that holds the object and is movable,
16. The exposure method according to claim 14, wherein the pattern is aligned by driving and controlling the moving body that holds the object. The object has a sensitive layer;
The exposure method according to claim 14, wherein the pattern is transferred by irradiating the sensitive layer with an energy beam. Forming a pattern on an object using the exposure method according to any one of claims 14 to 17;
Treating the object on which the pattern has been formed. An exposure apparatus that sequentially transfers a pattern to a plurality of partitioned areas arranged on an object,
An optical system for projecting a pattern onto the object;
A mark detection system for detecting a plurality of marks attached to the plurality of partition regions;
For a model that uses the mark detection system to detect a part of the plurality of marks attached to at least a part of the plurality of partitioned areas and describes a nonlinear component of the distortion of the object based on the detection result One condition is selected from a plurality of conditions determined in advance by detecting the plurality of marks attached to at least a part of the plurality of partitioned areas, and the model corresponding to the one condition is selected. A control system for aligning the pattern with the plurality of partitioned areas;
An exposure apparatus comprising: A plurality of conditions are predetermined for the model,
The control system obtains the nonlinear component based on a detection result of the part of marks, and selects the one condition for the model that most accurately describes the nonlinear component from the plurality of conditions. 19. The exposure apparatus according to 19.   21. The exposure apparatus according to claim 19, wherein the control system selects the one condition for each of an object, a plurality of partitioned areas, and one partitioned area.   The control system detects a plurality of marks attached to at least a part of the plurality of partitioned areas, obtains a linear component of distortion of the object based on the detection result, and determines the object's distortion based on the linear component. The exposure apparatus according to any one of claims 19 to 21, wherein the pattern is aligned by correcting a linear component of distortion and correcting the nonlinear component using the model corresponding to the one condition. . A lens driving device for driving at least a part of the plurality of lens elements constituting the optical system;
The exposure apparatus according to any one of claims 19 to 22, wherein the control system controls the lens driving device to align the pattern. A first moving body that is movable while holding the object;
A second moving body that is movable while holding the mask on which the pattern is formed;
A moving body drive device that measures positional information of the first and second moving bodies and synchronously drives the first and second moving bodies according to the measurement results;
The exposure apparatus according to any one of claims 19 to 23, wherein the control system controls the moving body driving device to align the pattern.   Claims: The pattern is transferred to the aligned partial areas by irradiating the sensitive layer provided on the object with an energy beam through the mask on which the pattern is formed and the optical system. Item 25. The exposure apparatus according to any one of Items 19 to 24. At least one exposure apparatus according to any one of claims 19 to 25;
A measuring device for detecting a plurality of marks attached to a plurality of partitioned areas arranged on the object;
Using the measuring device, a plurality of marks attached to at least a part of the plurality of partitioned regions are detected, a nonlinear component of the distortion of the object is obtained based on the detection result, and the nonlinear component is used based on the nonlinear component. A host computer that determines multiple conditions for the model describing the distortion of the object;
An exposure system comprising:

Images