JP2003086483A - Aligning method, aligning device and aligner - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an aligning method, an aligning device and an aligner capable of aligning with high accuracy, even if the arrangement of marks includes non-linear errors. SOLUTION: The method includes a step 203 of successively measuring positions of marks each formed on a plurality of substrates and deciding whether the measurement is completed, a step 205 of determining conversion parameters (B, Θ, S) for minimizing an overall error of all the measurement positions, on the basis of residues about respective measurement positions, when describing the relations between the measurement positions and positions in design, a step 206 of correcting the positions of objects to be aligned in design, by using the determined conversion parameters and a previously stored correction table and determining the positions of the marks, and steps 207 and 208 of moving the plurality of substrates to successively align them, by using the determined positions of the marks as reference positions.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an apparatus provided with precise alignment means, such as a reduction projection type exposure apparatus for projecting and exposing an electronic circuit pattern onto a semiconductor substrate, to accurately position a plurality of objects relative to each other. The present invention relates to a positioning method and a positioning device for matching. Furthermore, the present invention relates to an exposure apparatus including the alignment device.

[0002]

2. Description of the Related Art The degree of integration of semiconductors represented by dynamic RAMs (DRAMs) has increased remarkably in recent years, and the pattern size formed on semiconductor elements has been on the order of submicrons as the degree of integration increases. From such a background, in an exposure apparatus for manufacturing a device such as a semiconductor, technological development for improving the alignment accuracy between an original plate such as a mask and a substrate such as a wafer has been actively conducted. As a semiconductor exposure apparatus, a reduction projection type stepper or a scanner which is a scan exposure apparatus is widely used. FIG. 12 is a schematic view of an exposure apparatus according to a conventional example and a diagram showing alignment marks, and FIG. 12A is a schematic view showing an example of a reduction projection type semiconductor exposure apparatus according to the conventional example. It is a thing.

In FIG. 1, an exposure light flux emitted from an exposure illumination system (not shown) moves an electronic circuit pattern formed on a reticle R onto an XY stage 11 which can be two-dimensionally moved through a projection optical system 1. The wafer W placed on the wafer is projected and exposed. In the figure, S is an optical system for alignment,
In the figure, the position in the x direction is detected.
A similar alignment optical system (not shown) is mounted, and the position in the y direction is detected by this. Prior to the exposure, the relative alignment between the reticle R and the wafer W is performed by the following procedure.

A wafer W is transferred by a wafer transfer device (not shown).
When is mounted on the XY stage 11, the CPU 9 causes the stage driving device 10 to position the alignment mark M1x formed on the first measurement shot SS1 within the visual field range of the alignment optical system S. A command is sent to drive the XY stage 11. Here, the luminous flux emitted from the positioning illumination device 2 that irradiates the non-exposure light is passed through the beam splitter 3, the reticle R, and the projection optical system 1 to cause the positioning mark M1x (hereinafter referred to as a wafer mark). Is illuminating.

FIG. 12B is a view showing the wafer mark M1x, in which rectangular patterns of the same shape are formed at a constant pitch λ _p.
It has been arranged in multiple. The light flux reflected from the wafer mark M1x reaches the beam splitter 3 again through the projection optical system 1 and the reticle R, is reflected there, and is reflected there through the imaging optical system 4 onto the image pickup surface of the image pickup device 5 on the wafer surface. M1
Form an image WM of x. In the imaging device 5, the mark M
The 1x image is photoelectrically converted and converted into a two-dimensional digital signal train by the A / D converter 6. Reference numeral 7 indicates an integrating device. As shown in FIG. 12 (b), the accumulator 7 has an A / D
A processing window Wp is set for the wafer mark image WM 'converted into a digital signal by the converter 6, and a moving average process is performed in the Y direction shown in FIG. Is converted into a one-dimensional digital signal sequence S (x).

Numeral 8 is a position detecting device, and for the one-dimensional digital signal sequence S (x) output from the integrating device 7,
Pattern matching is performed using a template pattern stored in advance, the address position of S (x) having the highest degree of matching with the template pattern is detected, and this address position is output to the central processing unit (CPU) 9. To do. Since this output signal is a mark position with the image pickup surface of the image pickup device 5 as a reference, the CPU 9 determines the wafer mark from the relative position between the image pickup device 5 and the reticle R, which is obtained in advance by a method not shown. The position a _x1 of M1x with respect to the reticle R is calculated. Thus, the amount of positional deviation in the x direction of the first measurement shot has been measured. Next, the CPU 9 drives the XY stage 11 so that the y-direction measurement mark M1y of the first measurement shot falls within the visual field range of the y-direction alignment optical system. here,
The positional deviation amount a _y1 in the y direction is measured by the same procedure as the measurement in the x direction. With the above, the measurement in the first measurement shot SS1 is completed.

Next, the CPU 9 moves to the second measurement shot SS2 and measures the amount of positional deviation in the x and y directions in the same procedure as the first. Similarly, the present apparatus performs measurement for a predetermined number n of measurement shots (n = 8 in FIG. 12B), and measures the positional deviation measurement values a _xi , a _yi (i = 1, 2, ..., N) are stored.

The CPU 9 aligns the wafer W relative to the reticle R in the following manner based on the amount of positional deviation in each measurement shot thus obtained.

The CPU 9 is designed for each measurement shot.
Mark position d_i = [D_xi, A_yi]^TWafer mark measurement
The actual mark position a obtained by_i = [A_xi，
a_yi]^TWhen trying to superimpose by correction conversion to
Correction residual e_i = [E_xi, E_yi] ^T Corrected position g including
_i = [G_xi, G_yi]^T = [A_xi+ E_xi, A_xi+ E_xi]^T 
And d_i It is assumed that the relationship is expressed as Equation 1.

[0010]

[Equation 1]

Further, the conversion parameters A and S are calculated so that the sum of squares of the residual error e _i of the correction shown in Expression 2 is minimized.

[0012]

[Equation 2]

Next, the CPU 9 drives the XY stage based on a predetermined conversion parameter defined by A and S,
All the shots formed on the wafer are exposed by performing step & repeat so as to minimize the error between the measured mark position and the designed mark position. Here, A and S are shown like Formula 3.

[0014]

[Equation 3]

In Equation 3, a _x and a _y respectively represent the growth of the wafer in the x and y directions, and θ _x and θ _y respectively represent the x-axis and y-axis rotation components of the shot array. Further, S represents the parallel displacement of the entire wafer.

[0016]

According to the method in the above-mentioned conventional example, all the exposure shots (ES1 to ES3).
There is an advantage that the throughput of the apparatus is improved because the position shift is not measured in 2) and the position is adjusted using a limited number of sample shots.

However, when a non-linear distortion of the exposure shot array occurs, such a linear correction causes a large non-linear error, which causes a problem that the correction accuracy is deteriorated. Further, even if a technique such as alignment for each exposure shot or linear correction limited to only a partial area, so-called zone alignment is used to improve the correction accuracy, it is difficult to achieve both throughput and accuracy. Furthermore, when exposing the layout that generated the marks, if there is a step direction difference offset or a scan direction difference offset due to the stage drive mechanism of the exposure apparatus, etc., a partial linear zone alignment or exposure shot is performed. The alignment method in which the weight of the measured value in the vicinity is added to the correction formula has a drawback that it is difficult to improve the alignment accuracy.

An object of the present invention is to perform alignment of the entire substrate by forming a plurality of alignment objects on a substrate to be processed and selecting a plurality of measurement objects from the plurality of alignment objects. PROBLEM TO BE SOLVED: To provide a aligning method and aligning device capable of aligning with high accuracy even when an array of aligning objects includes a non-linear error, and to provide an aligner including the aligning device. Is to provide.

[0019]

In order to solve the above-mentioned problems, in the alignment method of the present invention, a plurality of alignment objects are formed in advance on each of a plurality of substrates in accordance with a predetermined arrangement, and each of the positions is adjusted. A positioning method for sequentially positioning the substrates at predetermined reference positions by positioning the alignment objects, wherein the positions of the alignment objects formed on the substrate are sequentially measured. When the first step and the relationship between the positions measured by the first step and their designed positions are described by a predetermined conversion parameter, an overall error of all the measured positions based on the residual error for each measured position. A second step of determining the conversion parameter that minimizes the above, the conversion parameter determined in the second step, and a correction table stored in advance. And a third step of correcting the designed position of each alignment object to determine the position of each alignment object, and the position of each alignment object determined by the third step as the reference. As a position, a fourth step of moving the other substrates so as to sequentially align the substrates is characterized.

In the present invention, in the third step, some of the plurality of substrates are previously aligned, and the displacement amount measurement result of the alignment object is described by a predetermined conversion parameter. Determine the conversion parameter so that the overall error of the difference is minimized, cluster the residuals for each measurement value determined from the conversion parameter, statistically processing the residuals for each class, It is preferable that the correction amount of each of the alignment objects is determined. Further, in the third step, the positions of the respective alignment objects are sequentially measured in advance on some of the plurality of substrates, and the relationship between the measured values and their designed positions is converted into a predetermined value. The conversion parameter is determined so that the overall error of the residual when described by parameters is minimized, and the design position of each alignment object is converted according to the determined conversion parameter to obtain the conversion parameter. The correction amount of each alignment object may be determined by statistically processing the residual between the position of each alignment object and the measured position. Further, the statistical processing can use an average value or a median value of residuals, or an average value or a median of residuals in the vicinity of the alignment object to be obtained.

In the alignment method, the correction table may be a correction amount corresponding to each alignment target or a correction amount corresponding to the coordinates of the substrate. Further, the correction table is preferably the same as the correction table referred to when the overlay object is generated, a duplicated table, or a table rewritten from the outside.

In order to solve the above-mentioned problems, in the alignment apparatus of the present invention, a plurality of alignment objects are previously formed on each of a plurality of substrates according to a predetermined arrangement, and the alignment objects are aligned. A positioning device for positioning the substrate to a predetermined reference position by means of the above, and first position measuring means for sequentially measuring the position of each alignment object formed on one substrate, Total measurement based on the residual for each measurement position when the relationship between the position of each of the alignment objects measured by the first position measurement means and their design position is described by a predetermined conversion parameter First parameter determining means for determining the conversion parameter so that the overall position error is minimized; and the variation determined by the first parameter determining means. Position determining means for determining the position of each alignment object by correcting the design position of each alignment object using a parameter and a correction table stored in advance in the storage means, and the position determination And a substrate moving unit that moves the positions of the respective alignment objects determined by the unit to sequentially align the other substrates with each other as the reference position.

Further, in the exposure apparatus of the present invention, the aligning device can be provided in the means for aligning the substrate.

According to the present invention, a correction table holding a correction value for correcting a non-linear error for each alignment object is constructed in advance, and the design position is converted by a linear or preset non-linear conversion parameter, and the position is converted. Improving the alignment accuracy by calculating the position of the alignment target and by correcting the position calculated for each alignment target in the correction table so that any non-linear error can be corrected. Can be achieved.

[0025]

BEST MODE FOR CARRYING OUT THE INVENTION In a preferred embodiment of the present invention, a plurality of alignment objects (patterns formed at each shot position) formed in advance in a predetermined arrangement on a plurality of substrates are predetermined. Is a positioning method for sequentially positioning to the reference position of, and the first step of sequentially measuring the position of each alignment object formed on one substrate, the measurement position and their design In order to minimize the overall error (sum of squares of residual error ei) of all measured positions based on the residual error (ei) for each measured position when describing the relationship with the position by a predetermined conversion parameter. The second step of determining the conversion parameters (B, Θ, S), the position of each alignment object is determined according to the determined conversion parameters, and the correction test stored in advance for each shot is performed. A third step of correcting the position of the object by the correction amount table (O _i), to move the other substrate so the position of each alignment object the determined sequentially positioned in the reference position And a fourth step.

Further, in a preferred embodiment of the present invention, there is provided an alignment device for sequentially aligning a plurality of alignment objects formed in advance in a predetermined array on each of a plurality of substrates at a predetermined reference position, First position measuring means (CPU 9, alignment optical system S, stage driving device 10, etc.) for sequentially measuring the position of each alignment object formed on one substrate, and the measurement positions and their positions. A first conversion parameter is determined so that the total error of all measurement positions based on the residual for each measurement position when the relationship with the design position is described by a predetermined conversion parameter is minimized. Parameter determination means (CP
U9), position determining means (CPU9) for determining the corrected position by referring to the determined conversion parameter and the correction table stored in advance in the storage means (12), and each of the determined alignment targets. Substrate moving means (CPU 9, stage driving device 10, etc.) for moving the other substrate so that the position of 1 is sequentially positioned at the reference position.

Here, the description in parentheses indicates the corresponding components in the embodiments described later. Further, in the alignment method and / or the alignment apparatus of the present embodiment, the position of the alignment object can be measured by measuring the amount of deviation from the designed position. In addition, as the correction table of the alignment target,
It is possible to use a correction deviation amount that is a set for each object or a list of data that is represented by a set of a substrate coordinate and a deviation amount. Further, the correction table preliminarily aligns all or a part of the plurality of substrates in advance, and the alignment error of a so-called preceding wafer exposed by using the result is directly changed to a non-linear component in which a linear error is eliminated. Separated, and further divided into certain clusters as needed,
The result of performing the statistical processing for each cluster can be used.

[0028]

EXAMPLES Examples of the present invention will be specifically described below. (First Embodiment) FIG. 1 is a schematic view showing an exposure apparatus having an alignment apparatus according to the first embodiment of the present invention.
In the figure, the functions of the alignment optical system S, the A / D conversion device 6, the integration device 7, and the position detection device 8 are the same as in the conventional example, so a detailed description thereof will be omitted here, but in the present embodiment. A storage device 12 is added to the configuration of the conventional example. The alignment method and alignment device according to this embodiment will be described below.

FIG. 2 is a flow chart showing an example of the exposure sequence in the exposure apparatus of FIG. When the wafer W1 is placed on the XY stage 11 by the wafer transfer device (not shown) (step 201), the CPU 9 causes the alignment mark M1x formed on the first sample shot SS1 to be aligned with the alignment optical system S. Command is sent to the stage drive device 10 so that the XY stage 11 is driven so that the XY stage 11 is positioned within the visual field range (step 20
2). The light flux emitted from the positioning illumination means 2 that emits non-exposure light is reflected by the beam splitter 3, reticle R, and
And the alignment mark M1 via the projection optical system 1.
illuminating x. The alignment mark M1x is shown in FIG.
It is the lattice mark shown in (b).

The A / D converter 6, the accumulator 7, and the position detector 8 are arranged in the same manner as described in the conventional example.
A relative positional deviation amount between the reticle R and the mark M1x is obtained. Next, the CPU 9 drives the XY stage 11 so that the alignment mark M1y in the y direction falls within the visual field range of the alignment optical system S (step 202), and M
The relative positional deviation amount between the reticle R and the mark M1y is obtained by the same method as for 1x.

The CPU 9 moves the XY stage 11 so that the x-direction measurement mark M2x of the second sample shot SS2 falls within the visual field range of the alignment optical system S, and thereafter SS2, SS3, SS4, ... , SS8 and the exposure shots formed on the wafer W1 with respect to preset sample shots, the deviation amounts in the x direction and the y direction are measured. An exposure shot and a sample shot located on the wafer W1 at this time are shown in FIG.
It shows in (a).

When the measurement of the sample shot is completed in step 203, the CPU 9 obtains the designed position d _i = [d _xi , d _yi ] ^T of each exposure shot from the actual measurement obtained by the wafer mark measurement (step 204). Shot position a
_{When i} = [a _xi , a _yi ] ^T is attempted to be superimposed by correction conversion, a correction position g _i = [g _xi , g _{yi including} a correction residual e _i = [e _xi , e _yi ] ^T ] ^T = [a _xi , + e _xi ,
It is assumed that the relationship between a _xi + e _xi ] ^T and d _i is represented by Expression 4.

[0033]

[Equation 4]

The CPU 9 calculates the conversion parameters B, Θ and S such that the sum of squares of the correction residuals e _i becomes the minimum under the condition that V shown in the equation 5 becomes the minimum. (Step 205).

[0035]

[Equation 5]

B, Θ, and S in the equation 4 are shown as the equation 6.

[0037]

[Equation 6]

In Equation 6, β_x, Β_y Each wafer
Represents the elongation in the x and y directions of _x , Θ_y Each show
Represents the rotation components of the x-axis and the y-axis of the dot array. Also,
S represents the parallel displacement of the entire wafer. these
The conversion parameters of are the patterns formed on the wafer W1.
Error is the cause of deviation from the ideal position
It represents a rate component, a rotation component, and a parallel shift component.

Next, the CPU 9 calculates the conversion parameter β
_The array position according to _x , β _y , θ _x , θ _y , s _x , s _y is calculated for each exposure shot. On the other hand, the storage device 12 stores a non-linear error stored in advance for each exposure shot. This pre-stored error is O _i = [o _xi ,
o _yi ] ^T , the i-th exposure shot (ES
The position of i) is shown as Equation 7.

[0040]

[Equation 7]

Next, the CPU 9 step-and-repeat-drives the XY stage 11 in accordance with the lattice obtained by converting the designed shot arrangement lattice according to the obtained conversion parameters, and sequentially exposes each shot of the wafer W1 (step 206). When the shot exposure is completed, the wafer W1 is discharged by the wafer transfer device (step 207) and stored in a wafer storage carrier (not shown). In step 208, it is determined whether or not it is the final wafer. In the case of No in step 208, that is, when the next wafer W2 is placed on the XY stage 11 similarly to the wafer W1, the CPU 9 still determines the wafer W.
Similar to 1, the alignment mark is measured for each sample shot and the conversion parameter is calculated.

FIG. 3B shows that the storage device 12 of FIG.
It is a figure which shows the stored correction value for every exposure shot. one
Generally, it is considered that the nonlinear error is constant throughout the lot.
To be Therefore, the same calculation is performed for the next wafer W2.
The conversion parameters issued and stored as shown in FIG.
Correction value O stored in advance in the apparatus 12 for each exposure shot _i
The exposure position is calculated by using the above equation (7).

Next, the CPU 9 step-and-repeat-drives the XY stage 11 in accordance with the grid obtained by converting the shot array grid on the design by the obtained conversion parameter, and sequentially exposes each shot of the wafer W2 (step 206), and all the shots are formed. When the shot exposure is completed, the wafer W2 is discharged by the wafer transfer device (step 207) and is stored in a wafer storage carrier (not shown). The above sequence is repeated until the lot is completed (Yes in step 208).

Here, a method of creating the correction table will be described. In the present embodiment, one or several wafers out of a plurality of wafers to be exposed in advance are extracted, and actually aligned and exposed. This is measured by a shift amount measuring device (not shown). Similar to alignment, this measured value has a linear component represented by conversion parameters B, Θ, S and a non-linear component after linear correction. Next, the CPU 9 separates the linear component and the nonlinear component in the same sequence as the wafer alignment, and stores the nonlinear component as it is in the correction table of the storage device 12.

The correction table may be deleted at the end of the job, or may be transferred to an online host computer (not shown) and transferred from the online host computer when a similar lot is exposed to correct the storage device 12. You may rewrite the table. Also, create a histogram of the nonlinear component of the exposure result, reject non-linear data that exceeds a predetermined outlier limit from the average value or median value, and use the average value or median value of the non-linear data in the vicinity of the shot. Good. This has the effect of eliminating non-linear components due to measurement errors and measurement variations. Although these judgment criteria are abnormal limit values specified by the operator, the CPU 9 can also rewrite the correction amount of a specific shot into an input value specified by the operator from a console device (not shown) and store it in the storage device 12. ing.

In this embodiment, instead of feeding back the exposure result of the preceding wafer, the exposure apparatus preliminarily performs alignment for all shots or a plurality of shots for several wafers, and the nonlinear residual after linear correction is performed. May be statistically processed and stored in the correction table.
FIG. 4 is a flowchart showing the exposure sequence in this case. In step 401, the preceding wafer is selected. The statistical processing may be average value, median value, or average value, median value processing using a predetermined specified abnormal value limit. Moreover, you may implement fitting by a higher-order function. By performing only the alignment, it is possible to omit the exposure / development / deviation measurement and improve the throughput. At this time, the preceding wafer may be aligned without resist coating. By doing so, it is possible to eliminate the influence of the unevenness of resist coating on the measurement accuracy, and it is possible to create a correction table with higher accuracy.

In the present embodiment, the CPU 9 may perform only alignment for two cases, that is, the case where the resist is not applied and the case where the resist is applied to the same preceding wafer. The respective conversion parameters and non-linear components obtained from these two examples are stored in the storage device 12 as a data file. Next, the CPU 9 sequentially aligns the wafers according to the flowchart of FIG. 4 (step 402). At this time, regarding the conversion parameters obtained for each wafer, the CPU 9
The offset obtained by subtracting the correction parameter of the resist-coated wafer from the correction parameter of the resist-uncoated wafer stored in 2 is added, and this is set as a new correction parameter,
Further, the non-linear component of the resist-uncoated wafer is overwritten on the correction table, and the exposure position is calculated according to Equation 7.

Next, a method of obtaining the correction table when the directional offset peculiar to the apparatus is generated in the exposure result of the preceding wafer or the alignment result of the preceding wafer will be described.
The device-specific direction offset includes a step direction offset in which an offset occurs depending on the direction in which the stage is stepped, a scan direction difference offset in which an offset occurs depending on the scanning direction in the scanner, and the like. When such a systematic non-linear component is generated, the distribution of the non-linear component does not follow the normal distribution but has a pole for each generation factor. Therefore, in order to correct the measurement variation of the measurement system of the exposure result of the preceding wafer or the alignment result of the preceding wafer, simple statistical processing (step 40
In the case of 6), the distribution of these directional offsets has a pole, and the directional offsets of neighboring shots are not always in the same system, so the correction accuracy is rather deteriorated.

Therefore, in this embodiment, in the case of Yes in step 405, the correction residuals are clustered and classified into a number of classes according to the factors of occurrence, that is, about 2 to 4 (step 407). This classification algorithm uses a discriminant analysis method that maximizes the interclass variance. By subjecting each classified class to the statistical processing (step 408) and determining the correction amount of each alignment mark, there is no influence of the direction difference offset and the measurement variation of the measurement system is reduced. Higher precision alignment is possible by reducing the number.

As an example of the case where the scanning direction offset occurs, FIG. 5 which is an example where a simple neighborhood averaging process is performed, and FIG. 6 which is an example where neighborhood averaging is performed by clustering are shown. Here, FIG. 5A is a diagram showing a correction value for each exposure shot stored in advance in the storage device 12 of FIG. 1, and FIG. 5B is a diagram showing a histogram of the correction value. Further, FIG. 6A is a diagram showing the correction value for each exposure shot stored in advance in the storage device 12 of FIG.
(B) is a diagram showing a histogram of correction values. By clustering, statistical processing can be performed for each scan direction, and optimal averaging can be performed for each scan direction.

In FIG. 4, the correction table described above is further registered (step 409), the wafer is regenerated (step 410), and the alignment and exposure using the correction table are performed (step 411).

This method also has the effect of automatically extracting the directional offset from the exposure result or the alignment result of the preceding wafer. In that case, the statistical result (average value, median value, etc.) for each class may be adopted as the correction value for the shots belonging to each class in the correction table.

Further, in this embodiment, a preset non-linear conversion parameter (for example, a quadratic term) may be added in addition to the linear conversion parameter. In this case, the error component from the preset non-linear component can be corrected.

(Second Embodiment) Next, a second embodiment of the present invention will be described. In the first embodiment, the correction amount in the correction table is stored as a pair with the exposure shot, but in the present embodiment, the correction amount is stored as a pair of the position coordinate to be corrected on the wafer and the correction amount. ) Is stored in. In the first embodiment, when creating the correction table, it was necessary to associate the deviation amount data output from the deviation amount measuring device and the shot data with the shot held by the exposure apparatus. It is also possible to convert the data output from the deviation amount measuring device into the on-wafer coordinates of the measurement point and store it in the correction table as it is. This method has a merit that it is not necessary to associate shot information of the deviation amount verification device with shot information of the exposure device by storing a set of coordinates and a correction amount. Further, when searching the vicinity of a specific shot, it becomes possible to compare the shot and the shot coordinate with information of only the correction table.

(Example of Semiconductor Production System) Next, an example of a production system of semiconductor devices (semiconductor chips such as IC and LSI, liquid crystal panels, CCDs, thin film magnetic heads, and micromachines) will be described. This is for maintenance services such as troubleshooting and regular maintenance of manufacturing equipment installed in semiconductor manufacturing plants, or software provision.
This is done using a computer network outside the manufacturing plant.

FIG. 7 shows the entire system cut out from a certain angle. In the figure, 101 is a vendor (device supplier) that provides semiconductor device manufacturing equipment.
It is a business office of. As an example of the manufacturing apparatus, a semiconductor manufacturing apparatus for various processes used in a semiconductor manufacturing factory, for example, pre-process equipment (exposure apparatus, resist processing apparatus, lithography apparatus such as etching apparatus, heat treatment apparatus, film forming apparatus, planarization apparatus) Equipment) and post-process equipment (assembling equipment, inspection equipment, etc.). Within the business office 101, a host management system 1 that provides a maintenance database for manufacturing equipment
08, a plurality of operation terminal computers 110, and a local area network (LAN) 109 that connects these to construct an intranet. Host management system 1
08 is provided with a gateway for connecting the LAN 109 to the Internet 105, which is an external network of the office, and a security function for restricting access from the outside.

On the other hand, 102 to 104 are manufacturing factories of semiconductor manufacturing manufacturers who are users of manufacturing equipment. The manufacturing factories 102 to 104 may be factories belonging to different manufacturers or may be factories belonging to the same manufacturer (for example, a factory for a pre-process and a factory for a post-process). In each of the factories 102 to 104, a plurality of manufacturing apparatuses 106 and a local area network (LAN) 111 that connects them and constructs an intranet, respectively.
And a host management system 107 as a monitoring device for monitoring the operating status of each manufacturing device 106. Host management system 1 provided in each factory 102-104
07 includes a gateway for connecting the LAN 111 in each factory to the Internet 105, which is an external network of the factory. As a result, it becomes possible to access the host management system 108 on the vendor 101 side from the LAN 111 of each factory via the Internet 105, and only the limited user is permitted to access by the security function of the host management system 108. Specifically, each manufacturing apparatus 1 is connected via the Internet 105.
In addition to notifying the vendor of the status information indicating the operating status of 06 (for example, a symptom of a manufacturing apparatus in which a trouble has occurred) from the factory side, response information corresponding to the notification (for example,
It is possible to receive maintenance information such as information instructing how to deal with troubles, software and data for dealing with troubles, latest software, and help information from the vendor side. A communication protocol (TCP / IP) generally used on the Internet is used for data communication between the factories 102 to 104 and the vendor 101 and data communication on the LAN 111 in each factory. Instead of using the Internet as an external network outside the factory, it is also possible to use a leased line network (ISDN or the like) which is highly secure without being accessed by a third party. Further, the host management system is not limited to the one provided by the vendor, and a user may construct a database and place it on an external network so that the user can access the database from a plurality of factories.

FIG. 8 is a conceptual diagram showing the entire system of this embodiment cut out from an angle different from that shown in FIG. In the above example, a plurality of user factories each provided with a manufacturing apparatus and a management system of the vendor of the manufacturing apparatus are connected by an external network, and production management of each factory or at least one unit of the factory is performed via the external network. The information of the manufacturing apparatus was data-communicated. On the other hand, in this example, a factory equipped with manufacturing apparatuses of a plurality of vendors and a management system of each vendor of the plurality of manufacturing apparatuses are connected by an external network outside the factory, and maintenance information of each manufacturing apparatus is displayed. It is for data communication. In the figure, reference numeral 201 denotes a manufacturing plant of a manufacturing device user (semiconductor device manufacturing maker), and a manufacturing device for performing various processes is installed on the manufacturing line of the factory.
3. The film forming processing device 204 is installed. Although only one manufacturing factory 201 is shown in FIG. 7, a plurality of factories are actually networked in the same manner. The respective devices in the factory are connected by a LAN 206 to form an intranet, and the host management system 205 manages the operation of the manufacturing line. On the other hand, host management systems 211, 221, for performing remote maintenance of the supplied equipments, are provided at the respective business establishments of the vendors (equipment supply manufacturers) such as the exposure equipment manufacturer 210, the resist processing equipment manufacturer 220, and the film formation equipment manufacturer 230. 231, which, as mentioned above, comprises a maintenance database and a gateway for external networks. The host management system 205 that manages each device in the user's manufacturing plant and the management systems 211, 221, and 231 of the vendor of each device are connected by the external network 200 such as the Internet or a dedicated line network. In this system, if a trouble occurs in any of the series of manufacturing equipment on the manufacturing line, the operation of the manufacturing line is suspended, but the vendor of the equipment in trouble receives remote maintenance via the Internet 200. This enables quick response and minimizes production line downtime.

Each manufacturing apparatus installed in the semiconductor manufacturing factory is provided with a display, a network interface, and a computer for executing network access software and apparatus operation software stored in a storage device. The storage device is a built-in memory, a hard disk, or a network file server. The network access software includes a dedicated or general-purpose web browser, and provides a user interface with a screen, an example of which is shown in FIG. 9, on the display. The operator who manages the manufacturing apparatus at each factory refers to the screen, and the manufacturing apparatus model (401), serial number (402), trouble subject (403), date of occurrence (404), urgency (4)
05), symptom (406), coping method (407), progress (4)
08) etc. is input to the input items on the screen. The input information is transmitted to the maintenance database via the Internet, and the appropriate maintenance information as a result is returned from the maintenance database and presented on the display. The user interface provided by the web browser is a hyperlink function (410-412) as shown in the figure.
The operator can access more detailed information on each item, pull out the latest version of software used in the manufacturing equipment from the software library provided by the vendor, and use the operation guide (help information for the factory operator's reference). ) Can be withdrawn.
Here, the maintenance information provided by the maintenance database also includes the information on the features of the present invention described above, and the software library also provides the latest software for implementing the features of the present invention.

Next, a semiconductor device manufacturing process using the above-described production system will be described. FIG. 10 shows a flow of the whole manufacturing process of the semiconductor device.
In step 1 (circuit design), a semiconductor device circuit is designed. In step 2 (mask manufacturing), a mask having the designed circuit pattern is manufactured. On the other hand, in step 3 (wafer manufacturing), a wafer is manufactured using a material such as silicon. Step 4 (wafer process) is called a pre-process, and an actual circuit is formed on the wafer by the lithography technique using the mask and the wafer prepared above.
The next step 5 (assembly) is called a post-process, which is a process of forming a semiconductor chip using the wafer manufactured in step 4, and includes an assembly process (dicing, bonding), a packaging process (chip encapsulation), and the like. Including steps. In step 6 (inspection), the semiconductor device manufactured in step 5 undergoes inspections such as an operation confirmation test and a durability test. A semiconductor device is completed through these processes and shipped (step 7). The front-end process and the back-end process are performed in separate dedicated factories, and maintenance is performed for each of these factories by the remote maintenance system described above. Information for production management and device maintenance is also data-communicated between the front-end factory and the back-end factory via the Internet or the leased line network.

FIG. 11 shows a detailed flow of the wafer process. In step 11 (oxidation), the surface of the wafer is oxidized. In step 12 (CVD), an insulating film is formed on the wafer surface. In step 13 (electrode formation), electrodes are formed on the wafer by vapor deposition. In step 14 (ion implantation), ions are implanted in the wafer. Step 1
In 5 (resist processing), a photosensitive agent is applied to the wafer. In step 16 (exposure), the circuit pattern of the mask is printed and exposed on the wafer by the exposure apparatus described above. In step 17 (development), the exposed wafer is developed. In step 18 (etching), parts other than the developed resist image are removed. In step 19 (resist stripping), the resist that is no longer needed after etching is removed. By repeating these steps, multiple circuit patterns are formed on the wafer. Since the manufacturing equipment used in each process is maintained by the remote maintenance system described above, troubles can be prevented in advance, and even if troubles occur, quick recovery is possible, and Productivity can be improved.

[0062]

As described above, according to the alignment method of the present invention, from the positions of a plurality of sample shots,
The amount of deviation of each shot from a predetermined conversion formula when calculating the overall positional deviation of the substrate to be aligned is held as a correction table in advance, and is reflected in the alignment position in sequence during alignment, thereby providing a nonlinear It is possible to prevent the deterioration of accuracy due to the distortion of and to perform highly accurate alignment.

Further, according to the alignment apparatus of the present invention, similarly to the above, it is possible to obtain the alignment apparatus capable of performing the alignment with high accuracy even when the alignment of each alignment object includes a non-linear error. .

Further, according to the exposure apparatus of the present invention, the deviation amount of each shot from the predetermined conversion formula when calculating the positional deviation of the entire substrate to be exposed is held in advance as a correction table, and at the time of exposure, By reflecting it on the alignment position sequentially, it is possible to prevent the accuracy from decreasing due to non-linear distortion.
An exposure apparatus that can perform highly accurate exposure can be obtained.

[Brief description of drawings]

FIG. 1 is a schematic view showing an exposure apparatus having an alignment device according to a first embodiment of the present invention.

2 is a flowchart showing an example of an exposure sequence in the exposure apparatus of FIG.

3A is a diagram showing an example of an exposure shot and a sample shot positioned on a wafer, and FIG. 3B is a diagram showing a correction value for each exposure shot stored in advance in a storage device 12 of FIG. .

FIG. 4 shows a case in which only a few wafers are aligned in advance for all shots or a plurality of shots according to an embodiment of the present invention, and nonlinear residuals after linear correction are statistically processed and stored in a correction table. 3 is a flowchart showing the exposure sequence of FIG.

5 is an example in which a simple neighborhood averaging process according to an embodiment of the present invention is performed, and FIG. 5A is a diagram showing a correction value for each exposure shot stored in advance in the storage device 12 of FIG. (B)
FIG. 6 is a diagram showing a histogram of correction values.

FIG. 6 is an example of clustering and neighborhood averaging according to an embodiment of the present invention, in which (a) is the storage device 12 of FIG.
A diagram showing a correction value for each exposure shot stored in advance in
(B) is a diagram showing a histogram of correction values.

FIG. 7 is a conceptual view of a semiconductor device production system according to an embodiment of the present invention viewed from an angle.

FIG. 8 is a conceptual view of the semiconductor device production system according to the embodiment of the present invention seen from another angle.

FIG. 9 illustrates a specific example of a user interface according to an exemplary embodiment of the present invention.

FIG. 10 is a diagram illustrating a flow of a device manufacturing process according to an embodiment of the present invention.

FIG. 11 is a diagram illustrating a wafer process according to an embodiment of the present invention.

FIG. 12 is a schematic view of an exposure apparatus according to a conventional example and a view showing alignment marks, and (a) is a schematic view showing an example of a reduction projection type semiconductor exposure apparatus according to the conventional example;
(B) is a figure which shows the mark for alignment.

[Explanation of symbols]

1: Projection optical system, 2: Positioning illumination device, 3: Beam splitter, 4: Imaging optical system, 5: Imaging device, 6: A
/ D converter, 7: integrating device, 8: position detecting device, 9:
CPU, 10: stage driving device, 11: XY stage, 12: storage device, S: alignment optical system, R: reticle, W: wafer.

Claims (16)

[Claims] 1. A plurality of alignment objects are formed in advance on each of a plurality of substrates in accordance with a predetermined arrangement, and the substrates are sequentially aligned to predetermined reference positions by aligning the alignment objects. A first step of sequentially measuring the positions of the respective alignment objects formed on the substrate, the positions measured by the first step, and their designed positions. And a second conversion step for determining the conversion parameter that minimizes the overall error of all the measurement positions based on the residual error when each relationship is described by a predetermined conversion parameter, and is determined by the second step. Each of the alignment objects is corrected by correcting the design position of each of the alignment objects by using the converted conversion parameter and the correction table stored in advance. A third step of determining a position, and a fourth step of moving the position of each alignment object determined by the third step as the reference position to sequentially align the other substrates. A characteristic alignment method. 2. The third step is an overall residual error when some of the plurality of substrates are aligned in advance and the displacement amount measurement result of the alignment object is described by a predetermined conversion parameter. The conversion parameter is determined so that the error is minimized, the residuals for each measurement value determined from the conversion parameter are clustered, the residuals are statistically processed for each class, and the respective positions are determined. The alignment method according to claim 1, wherein a correction amount of the alignment object is determined. 3. The third step sequentially measures the positions of the respective alignment targets on some of the plurality of substrates in advance, and the relationship between the measured values and their designed positions is calculated. The conversion parameter is determined so that the overall error of the residual when described by the predetermined conversion parameter is minimized, and the design position of each alignment object is converted according to the determined conversion parameter. The correction amount of each alignment object is determined by statistically processing the residual difference between the obtained position of each alignment object and the measured position.
Positioning method described in. 4. The statistical processing is a mean or median of residuals, or a mean or median of residuals in the vicinity of an alignment object to be obtained. Alternatively, the alignment method according to any one of 3 above. 5. The alignment method according to claim 1, wherein the correction table is a correction amount corresponding to each of the alignment objects. 6. The alignment method according to claim 1, wherein the correction table is a correction amount corresponding to coordinates of the substrate. 7. The correction table is one of the same as the correction table referred to when the overlay object is generated, a duplicated table, and a table rewritten from the outside. The alignment method according to claim 5 or 6, wherein. 8. A position at which a plurality of alignment objects are formed in advance on each of a plurality of substrates according to a predetermined arrangement, and the substrates are aligned at a predetermined reference position by aligning each of the alignment objects. A positioning device, comprising: first position measuring means for sequentially measuring the position of each of the positioning objects formed on the substrate; and each of the positioning objects measured by the first position measuring means. When the relationship between the positions of objects and their design positions is described by the predetermined conversion parameters, the conversion parameters are set so that the total error of all the measurement positions based on the residual for each measurement position is minimized. First parameter determining means for determining, the conversion parameter determined by the first parameter determining means, and a correction table stored in advance in the storage means Position determining means for determining the position of each of the alignment objects by correcting the design position of each of the alignment objects using, and the alignment object of each of the alignment objects determined by the position determining means. And a substrate moving unit that moves the other substrates in order to sequentially align the substrates with the position as the reference position. 9. An exposure apparatus for exposing a substrate with a pattern of an original plate, wherein the means for aligning the substrate is provided with the alignment apparatus according to claim 8. 10. The exposure apparatus according to claim 9,
An exposure apparatus that further includes a display, a network interface, and a computer that executes network software, and that enables maintenance apparatus information to be communicated via a computer network. 11. The network software comprises:
A user interface for accessing a maintenance database provided by a vendor or a user of the exposure apparatus, which is connected to an external network of a factory in which the exposure apparatus is installed, is provided on the display, and from the database via the external network. The exposure apparatus according to claim 10, which makes it possible to obtain information. 12. A step of installing a manufacturing apparatus group for various processes including the exposure apparatus according to claim 9 in a semiconductor manufacturing factory, and a plurality of processes using the manufacturing apparatus group. And a step of manufacturing a semiconductor device. 13. Data communication of information relating to at least one of the manufacturing apparatus group between the step of connecting the manufacturing apparatus group by a local area network, and between the local area network and an external network outside the semiconductor manufacturing factory. The method for manufacturing a semiconductor device according to claim 12, further comprising: 14. A database provided by a vendor or a user of the exposure apparatus is accessed through the external network to obtain maintenance information of the manufacturing apparatus by data communication, or a semiconductor manufacturing factory different from the semiconductor manufacturing factory. 14. The semiconductor device manufacturing method according to claim 13, wherein production control is performed by performing data communication with the device via the external network. 15. A group of manufacturing apparatuses for various processes including the exposure apparatus according to claim 9, a local area network for connecting the group of manufacturing apparatuses, and a local area network connected to the outside of the factory. A semiconductor manufacturing factory, having a gateway that enables access to the external network, and enabling data communication of information relating to at least one of the manufacturing apparatus groups. 16. The method according to claim 9, which is installed in a semiconductor manufacturing factory.
12. The exposure apparatus maintenance method according to any one of claims 1 to 11, wherein a vendor or a user of the exposure apparatus provides a maintenance database connected to an external network of the semiconductor manufacturing factory, and the semiconductor manufacturing factory. A step of permitting access to the maintenance database from the inside via the external network; and a step of transmitting the maintenance information accumulated in the maintenance database to the semiconductor manufacturing factory side via the external network. Maintenance method for exposure equipment.