JPS5828748A - Positioning device for transcribing device - Google Patents

Abstract

PURPOSE:To reduce greatly the alignment error, by performing the positioning previously to plural transcription regions on a wafer before the transcription, storing the information on the matching position and then controlling a shifting device during the transcription and based on the above-mentioned information. CONSTITUTION:For an exposure system of the step-and-repreat system, a mask 25 is positioned at the lower part of an alignment optical system 40 before the transcription. Then the shift of rotation between the mask 25 and the xy direction of the stages 7 and 11 of (x) and (y) axes is compensated by means of the system 40. A wafer 15 is set on a wafer stage 14, and the mask 25 is positioned to each of plural transcription regions on the wafer 15 with use of the marks provided to the mask 25 and the wafer 15. Then the information on the matching position is stored. In the transcription mode, the mask 25 is positioned at the lower part of a light source 46. Then the shift of the stages 11 and 7 is controlled based on the above-mentioned information to perform the transcription successively.

Description

[Detailed description of the invention] Ru.

Traditionally, it has been used as a transfer device. After relatively aligning the mask and wafer at the alignment position, that is, the base of the alignment optical system, the two are clamped together to prevent their relative positions from shifting, and then moved to the exposure position, that is, the base of the exposure optical system. and. There is a method in which the alignment position and the exposure position are matched, and after the alignment is completed, the alignment optical system is retracted and exposure is performed.

However, in the former case, even if a positional shift occurs between the wafer and mask when they are moved to the exposure position, this cannot be easily corrected, and in the latter case, the alignment optical system must be retracted. Because of that. There is a drawback that the mechanical strength and accuracy of the alignment optical system are unstable.

These drawbacks become noticeable in so-called step-and-repeat exposure systems. Here, in the step-and-repeat method, the exposure area is set smaller than the wafer size j2. This method exposes the entire wafer by moving the wafer or mask and exposing each small area. In this step-and-repeat method, the wafer or mask must move back and forth multiple times between the alignment position and the exposure position. Furthermore, even in the latter case, it is necessary to retreat the alignment optical system every time each area is exposed, which causes the problem that the number of processing per unit time, so-called throughput, cannot be increased.

In recent years, in order to form patterns with narrower line widths on wafers, in addition to exposure using visible light, several transfer techniques using X-rays with extremely short wavelengths have been used. For example, when transferring a pattern using the proximity method in which the mask and wafer are placed slightly apart. The problem is that the X-ray irradiation time, the so-called transfer time, including the sensitivity of the resist applied to the wafer, is much longer than the exposure time of visible light, etc.The source of the radiation and the alignment optical system are naturally separated. is placed in the correct position. Therefore, as stated earlier. The relative position of the mask and wafer is adjusted at the alignment position, and the wafer and mask are moved to the exposure position, that is, the X-ray irradiation position, while being clamped at that relative position, and transfer is performed. In this state, the wafer and mask are held together for a long time (for example, 1
During this period, the relative positions of the wafer and mask may shift due to mechanical fluctuations such as the wafer mounting stage clamp. If one end shifts during transfer, an accurately positioned pattern can no longer be transferred onto the wafer. As a result, there has been a drawback that the use of X-rays to transfer patterns with narrower line widths is not fully utilized.

The present invention was developed against the background of these circumstances. It is a step-and-repeat type transcription device.

There is no need to retreat the alignment device each time each small area is exposed. It is an object of the present invention to provide an alignment device for a transfer device, which eliminates the need to reciprocate a wafer and a mask many times between an alignment position and a transfer position, and improves throughput during alignment and exposure. .

to this end. In the present invention. Prior to transfer, the mask and wafer are moved relative to each other. An alignment detection device that detects the alignment position of both at each step position and generates information regarding the relative position, and a storage means that stores the information generated thereby. The present invention includes means for controlling a moving device so as to relatively move the mask and the wafer step by step based on the above information during transfer.

Embodiments of the present invention will be described below with reference to the drawings.

In FIG. 1, a column 2 is mounted on a base 1 so as to be movable in the X-axis direction (horizontal direction in the figure) via a bearing 3.
The column 2 is driven by a motor 4 fixed on the base 1 via a feed screw 4a. On the column 2, the y
A shaft stage 7 is mounted so as to be movable in the y-axis direction (direction perpendicular to the plane of the drawing).

It is driven by a y-axis motor 8 fixed to the column 2. Furthermore, an X-axis stage 11 is mounted on the y-axis stage Z so as to be movable in the X-axis direction via rolling elements 12.

It is driven by an X-axis motor 13 fixed on the y-axis stage 7. A wafer stage 14 for mounting a wafer 15 is fixed on the X-axis stage 11, and this wafer stage 14 corrects the taper of the wafer 15 itself, and also supports the wafer 15 and a mask (described later) 2.
5 and a leveling mechanism to create a gap between
The column 2, which has a wafer orientation mechanism for correcting errors in the rotational direction of the wafer 150 with respect to the axis stage 11, has an upper extension portion 2a,
This part is provided with a mask holder 21 for attaching the mask 25 and a mask orientation mechanism 22 for correcting rotational errors when attaching the mask 25. T-x.

It is rotatable within an Xy-axis plane that includes each movement direction of the y-axis stages 11 and 7.

At the ends of the X-axis stage 11 and the column extension portion 2a, there are plane reflection mirrors 26 for position detection in the X-axis direction.
and 27 are attached. Although not shown, a plane reflection mirror for position detection in the y-axis direction is also attached to each end.

Further, a length measuring laser light source 31 and a laser interferometer 32 are fixed to the end of the column 2. between the laser light source 31 and the X-axis stage 11, and between the interferometer 32 and the column 2.
There is a half mirror 63 and a reflective mirror 3 between them, respectively.
4 is disposed, and another reflective mirror 36 is disposed above the mirror filters 3 and 34 at a position opposite to the mirror 27. The laser light source 31 and the interferometer 32 are mirror 2
6 and the mirror 27 is connected to a length measuring circuit 35 for determining the relative distance between the mirror 27 and the mirror 27. A portion of the laser beam from the laser light source 31 passes through the half mirror 33, is reflected by the reflecting mirror 26, is reflected downward and to the left by the mirrors 33 and 34, and reaches the laser interferometer 32. On the other hand, the remainder of the laser beam reflected by the half mirror 33 is reflected to the right by the mirror 36 and to the left by the mirror 27, and is again reflected downward by the mirror 36 and left by the mirror 34, and is reflected to the tiginometer 32. reach. Then, the relative distance between the X-axis stage 11 and the column, that is, the wafer 15 and the mask 25 is determined by comparing both incident lights in the length measuring circuit '55.

The determined distance value is input to the control device 50. What is the relative distance between the wafer 15 and the mask 25?

The same process is performed in both the X-axis direction and the y-axis direction. Therefore, the output signal of the length measuring circuit is.

This value corresponds to the relative distance between the mirror 26 and the mirror 27 in the X-axis direction (for example, the distance from the mirror 26 to the mirror 27), and this means that the relative distance between the wafer 15 and the mask 25 in the X-axis direction means. Furthermore, since the relative distance in the y-axis direction is determined by a similar length measuring circuit, the relative two-dimensional position of the wafer 15 and the mask 25 (for example, the coordinate value of the wafer 15 with respect to the mask 25) is finally measured.

On the other hand, an alignment optical system 40 for observing alignment marks provided on each of the mask 25 and the wafer 15 is fixed to a base 41 that is integrated with the base 1 above the mask holder 21. The observation light is input to a mark detection section 42 provided on the side and converted into an electrical signal representing the position of each mark. Based on the input of the electrical signal, the one-position displacement processing section 43 detects the direction and amount of displacement between each mark on the wafer 15 and the mask 25, and generates a signal corresponding to the displacement between the two marks. Therefore, this signal represents the misalignment between the mask 25 and the wafer 15.

Here, an example of the marks on the wafer 15 and mask 25 handled by the mark detection section 42 and the positional deviation processing section 43 will be explained based on FIG. 2. As shown in FIG. 2, marks M are placed on the mask 25 at predetermined intervals.
, M' are provided, and a mark M is provided on the wafer 15.

Provide a rectangular mark W that is inserted into M' (
Note that the diagonally shaded portion of the center mark in FIG. 1 is a reflective portion that reflects light, for example. ) and as mentioned above, the alignment optical system 4
Observe marks M, M' and mark W at the same time using 0.

The mark detection section 42, for example, scans a laser spot in the direction of arrow S in FIG. 2, and photoelectrically detects reflected light from each mark. At this time, the mark detection unit 42 sends a high-level signal at each mark, for example, to the interval portions a and b.
At this point, a low level signal is generated. Based on the human input of this signal, the positional deviation processing section 43 detects the sizes of the gaps a and b.

A signal corresponding to the direction and amount of deviation between mark MM' and mark W is generated. In addition, as the mark detection section 42, the ITV
It is also possible to use an image input device such as the one described above to determine the displacement of the mark based on the image scanning signal.

Note that each mark on the wafer 15 and the mask 25 is provided as shown in FIG. 4, for example.

Each stage 11.7 is provided in a plurality of transfer areas on the wafer 15.
1 so that the position can be aligned with the moving direction X+V of
Two marks Wx, Wy for one transfer area P
be established. Of course, two marks k1 also correspond to the mask 25 having the area Q where the circuit pattern is drawn.
M'x.

MM'Y is provided. In the case of this device, mask 25
The wafer 15 moves within the xy plane, and marks MM'x, Wx, and MM are formed in each transfer area.
' Transfer is performed with Y and mark wy aligned, that is, overlapping each other.

By aligning the mark MM'' and the mark W in this way, the pattern of the mask 25 is formed in the region P, for example, by aligning the mark MM'' with the mark W.
It matches the pattern formed by the second transfer.

Returning to FIG. 1 again, a light source 46 that generates energy rays such as visible light and X-rays for transfer is provided on the base 41 at a position apart from the alignment optical system 40 along the moving direction of the column 2. It is being The light source 46 is a drive unit 47
The irradiation of light beams, X-rays, etc. to the wafer 15 is controlled by. The drive unit 47 includes, for example, a shutter control circuit that blocks light from an aperture through which a light beam or the like passes. The positional deviation processing section 43 and length measurement circuit 35 are connected to a control device 50,
This control device 50 includes the motors 4, 8 and 13, and the drive section 47.
It is connected to the. This control device 50 is based on instructions from a computer 51 that performs predetermined control and transfers and stores data.

03) It controls the driving and state detection of each part.

FIG. 3 is a block diagram for explaining the control device 5o in detail. Regarding the control device 5°, only the configuration for driving the X-axis and y-axis stages 11 and 7 is shown as a block diagram. A signal Dt corresponding to the amount of deviation in the X-axis direction between the wafer 15 and the mask 25 generated by the positional deviation processing unit 43, and a signal Dt corresponding to the relative distance in the X-axis direction between the mask 25 and the wafer 15 measured by the length measuring circuit 35. signal D2. A signal D5. corresponding to the amount of deviation between the mask 25 and the wafer 15 in the y-axis direction. A signal D4 corresponding to the relative distance in the y-axis direction is input to the arithmetic processing circuit 1oo.

This arithmetic processing circuit 1oo exchanges information regarding positioning with the computer 51, and uses signals such as D3 to ensure that the amount of deviation in the X- and Y-axis directions between the mask 25 and the wafer 15 is below a predetermined value. The signal D2. Mask 25 based on D3 man power
and the relative distance values of the wafer 15 in the X-axis and y-axis directions are stored in the X memory 103. It controls the operation of storing and reading data into and from the Y memory 104. It goes without saying that the X memory 103 and the Y memory 104 may be internal memories of the computer 51.

Next, the operation of this embodiment will be explained with reference to FIGS. 1 and 3.

First, the column 2 is moved by the drive motor 4, and the mask 25 is positioned below the alignment optical system 40. Next, using the alignment optical system 40, the rotational deviation of the mask 25 and the XiY axis stage 11 with respect to the xy force direction is corrected. This correction is performed by the orientation mechanism 22 by observing dedicated marks provided at two locations on the mask 25, for example. This prevents the mask pattern from being printed diagonally onto the wafer 15 (15
) will be stopped.

Next, the wafer 15 is set on the wafer stage 14, and the gap between the wafer 15 and the mask 25 is adjusted. The setting of this gap is 1.Although not shown in the diagram,
This is done using a proximity sensor, etc.

Next, as shown in FIG. 2, the marks MM' provided at at least two locations on the mask 25 and the marks W provided at two corresponding locations on the wafer 15 are made to almost match at those two locations. The X-axis stage 11 and the Y-axis stage 7 are moved by motors 13 and 8, respectively. Then, the rotational deviation of the wafer 15 with respect to the mask 25 is detected by the alignment optical system 40° mark detection section 42 and the positional deviation signal processing section 43, and the wafer 15 is rotated by the orientation mechanism on the wafer stage 14 to correct this deviation. .

Next, move the X-axis and y-axis stage 11.7 to place the wafer 1.
The positioning of the squares 25 is performed for each of the plurality of transfer areas on the image forming apparatus 5. For this purpose, the computer 51 calculates the relative position of each region of the wafer 15 with respect to the mask 25 in a two-dimensional manner corresponding to each region by transmitting the signals D2 and D4 from the length measuring circuit 35 through the calculation circuit 100. It is stored in advance as a specific position. In addition, please use the calculator 5 in advance.
The plurality of two-dimensional positions stored in 1 will be designated positions from now on. First, for one area on the wafer 15, the computer 51 transmits information on the corresponding specified position to the arithmetic processing circuit 100.
Output to. At this time, the arithmetic processing circuit 100.

Based on the signal D2 and the signal D4 of the length measuring circuit 35, the drive circuits 101 and 102 are controlled so that the relative position of the mask 25 and the wafer 15 becomes equal to the designated position. As a result, the pattern of the mask 25 almost matches the pattern formed in one transfer area of the wafer 15.
If there is a slight deviation when the wafer 15 is placed, perfect alignment may not be achieved at this specified position.

(17) At the specified position, the mark MM' on the mask 25 and the mark W on the wafer 15 are simultaneously observed by the alignment optical system 40, and the mark detection section 42 and the positional deviation processing section 43
detects the amount of positional deviation of the wafer 15 with respect to the mask 25, and uses signals D+ and D3 corresponding to this amount of deviation to drive the drive circuits IQ1. Control IQ3, move the stage 11.7, and apply the servo so that the amount of positional deviation in the X-axis direction and the y-axis direction becomes zero. Then, when the amount of positional deviation between the wafer 15 and the mask 25 becomes approximately equal to the preset characteristic value, for example, the distances a and l) shown in FIG. 2, the stage 11. 7, information on the stage position at that time, that is, signal D
2. D4 passes through the arithmetic processing circuit 100, and signal D2 passes through the
As memory 10311 knee Xt, signal D4 is stored in Y memory 104 as Yl. Then calculator 51
outputs information regarding the next specified position to the arithmetic processing circuit 100, and the above-mentioned operation is repeated (18).

The above operation is repeated for all of the specified position information of the stage 11.7 that has been stored in advance in the computer 51, and for each specified position, the X-axis is accurately determined by the marks MM' and The relative position of the stage 11 and the y-axis stage 7 is determined by the X memory 103 and the Y memory 10.
4, sequentially (X2, Y2), (Xs,
Y3) -m-. The position information stored in each of the memories 103 and 104 will be referred to as a matching position corresponding to each area on the wafer 15.

As described above, the X-axis stage 1 with respect to the mask 25
1 and all alignment positions of the y-axis stage 7 are stored in each memory 1
03.1 (Once stored in Ill, the column 2 is moved by the motor 4 and the mask 25 is positioned below the light source 46.

Next, the computer 51 outputs a transfer start command to the arithmetic processing section 100. Based on this Directive.

(19) Among the matching position information stored in each memory 103.10A, any one (2), for example (χ+, Y+), is taken into the arithmetic processing unit 100. Then, the arithmetic processing unit 100
is the signal D2゜D4 and matching position information (Xl, Y+)
1. For example, move the X-axis and y-axis stage 11.7 so that the difference becomes zero. Drive circuit IQ1, 102
control. That is, using the information (Xl, Yt) as a reference value, the relative positions of the wafer 15 and the mask 25 are controlled so as to always match this reference value. Therefore,
Even if mechanical fluctuations occur for some reason in each stage 11.7 or its feeding mechanism, the relative positions of wafer 15 and mask 25 are always maintained at aligned positions.

When the pattern of the mask 25 matches one area on the wafer 15 in this way, the computer 51 issues a command to the drive section 47 of the light source 46 via the control device 50 to start irradiation with the transfer illumination light or energy beam. And predetermined time: (2
0) 1 on the wafer 15 by irradiating onto the mask 25
The pattern is transferred to two areas.

Incidentally, during the transfer, the drive circuits 101 and 102 are always working,
Continue to maintain alignment. after that.

Irradiation from the light source 46 is stopped, and information on the next matching position (x2°Y2) is taken into the arithmetic processing circuit 100 from the X memory 103° and the Y memory 104.

Then, as described above, each stage 11, 7 is moved to similarly transfer the pattern of the mask 25 to the next area on the wafer 15. In this way, the wafer 15 is
The pattern of the mask 25 is aligned with the pattern in each area above and transferred sequentially.

In the alignment apparatus according to the embodiment of the present invention described above, when each alignment position is determined and stored, each stage 11.7 is moved in the vicinity of the specified position, for example, as shown in FIG. 4 in the X direction. The mark detection section 42 is configured to photoelectrically detect that the mark M M' x and the mark Wx match and generate a pulse signal (21). and.

When the arithmetic processing circuit 100 receives this pulse signal, the signal D2 from the length measuring circuit 35 may be stored in the X memory 103. Of course, the same applies to the X direction.

FIG. 5 is an explanatory diagram showing the configuration of another transfer device using the alignment device according to the present invention. The difference from the configuration shown in FIG. 1 is that the mask 25 is fixed in position with respect to the base 1, and the alignment optical system 40. The light source 46 is made movable relative to the base 41. For this purpose, the alignment optical system 40 and the light source 46 are fixed to the mount 61 so as not to spatially interfere with each other. The mount 61 is the base 41
It is moved in the left and right direction in FIG. 5 by a feed screw 62 which is suspended by bearings 63 and 64 fixed to.

This movement is controlled by a motor 65 which functions similarly to the motor 4 shown in FIG.

(22) It is attached to an arm 60 that is integrated with the base 1.

The X-axis and y-axis stages 11.7 move two-dimensionally relative to the base 1. still.

The rest of the configuration is the same as that in FIG. 1, so the explanation will be omitted.

Also, the mark detection section 42. Drive unit 47 of light source 46. Also regarding control of motors 8, 13, etc.

The control system shown in FIGS. 1 and 3 is used as is. The alignment operation in this case is the same as described above, but
The difference is that during transfer to the wafer 15, the control device 50 controls the motor 65 to be driven instead of the motor 4 shown in FIG.

By configuring the transfer device as in this example, it becomes unnecessary to mount the X-axis and y-axis stages 11.7, which require mechanical stability, on the column 2.1 Mechanical strength and stability are improved. A sufficiently high stage can be used. As a result, there will be an advantage that alignment accuracy will be improved.

(23) According to the embodiment described above, it is not necessary to retreat the alignment optical system every time the wafer 15 is exposed, and the mechanical stability of the alignment optical system can be improved. The positional deviation between the mask 25 and the wafer 15 when they are moved to a certain position can be made smaller than the resolution of the laser interferometer during exposure. Since the alignment optical system 40 and the light source 46 are separated, there is no interference between them, and a highly accurate optical system can be formed by providing each with sufficient mechanical margin.

Furthermore, the throughput during alignment and exposure can be increased.

In addition, as another configuration of the transfer device, as shown in FIG. 5, an X-axis and y-axis stage 11.7 is mounted on the base 1, and only the alignment optical system 40 is made movable in the horizontal direction in the figure. , the light source 46 is arranged above the mask 25 so as to be movable up and down. When using the alignment optical system 40, the light source 46 is retracted upward (24), and during transfer, the alignment optical system 40 is retracted horizontally and the light source 46 is moved downward to a predetermined position. You can also do it. Furthermore, it goes without saying that the present invention can also be used in other exposure apparatuses 1, such as step-and-repeat projection type exposure apparatuses.

As described above, according to the present invention, before transfer, a mask is aligned in advance with respect to a plurality of transfer areas on a wafer, and information on the alignment positions is stored, and during transfer, the stored information is Since the relative positions of the mask and wafer are controlled to be maintained at aligned positions based on the above, positioning errors due to minute movements of the stage that occur during transfer, so-called alignment errors, are extremely low. Therefore, in the manufacture of semiconductor devices such as ICs, patterns with narrower line widths can be managed with good yield.

Furthermore, when transferring by moving the wafer and mask in steps relative to each other, there is no need to observe and check each stop position, that is, the alignment position using the 7-line optical system (25). Therefore, throughput can be improved in the manufacturing process of semiconductor devices.

[Brief explanation of drawings]

FIG. 1 is an explanatory diagram showing one embodiment of the present invention. Figure 2 is a cross-sectional view showing wafer marks and mask marks;
FIG. 3 is a block diagram of the control device (50), FIG. 4 is a perspective view showing wafer marks and mask marks, and FIG. 5 is an explanatory diagram corresponding to FIG. 1 showing another embodiment of the present invention. . [Explanation of symbols of main parts] 7.11 --- One moving device 31, 32.35.40.42.43 -- One matching detection means 103.1 (1-- One storage means 1QQ, IQl, 102- - one control means dO, 42;
, 43-m-positional deviation detection means D+, D5--1 detection output MM', W---- mark (26)

Claims (1)

[Claims] 1. A moving device that moves the mask and the wafer relatively (-) is provided, and the mask and the wafer are moved relatively by a predetermined amount of steps, and the pattern of the mask is placed on the wafer at each step position (=). In a device that repeatedly transfers a pattern overlapping the pattern of an alignment detection device that generates information regarding the position; a storage device that sequentially stores the information; and a movement device that moves the mask and the wafer in steps relative to each other based on the information in the storage device during transfer. 2. The mask and the wafer each have marks for alignment, and the alignment detection device detects two marks between the mark on the mask and the mark on the wafer. Claim 1, characterized in that the moving device includes positional deviation detection means for detecting dimensional positional deviation, and prior to transfer, the control means brings both marks into a predetermined positional relationship according to the detection output thereof. The alignment device according to item 1. 6. The moving device has a moving stage that places a wafer and moves two-dimensionally with respect to the mask, and the alignment detection device detects the relative position of the moving stage with respect to the mask. and a two-dimensional length measuring means for generating information about the two-dimensional length measuring means. During transfer, the control means sets the information of the two-dimensional length measuring means and the information corresponding to each step position stored in the storage means into a predetermined relationship. 3. The positioning apparatus according to claim 2, wherein the moving stage is moved in steps so as to maintain the position.