JPS60102739A - Baking and exposure device for semiconductor - Google Patents

Abstract

PURPOSE:To improve the precision of subsequent positioning using one part of alignment marks by positioning a mask and a wafer by using all of a plurality of alignment marks on the first shot when baking and exposing a semiconductor. CONSTITUTION:The upper section of a wafer 40 is divided into a plurality of areas P1-P32, and exposed at every area P1-P32. Four areas T are collected to form an area P in principle. A plurality of alignment marks A, B are put on the areas P. The areas P are not formed from four areas T in the peripheral section of the wafer 40 as an exception. All of the alignment marks A, B cannot be put. Accordingly, the wafer is positioned from the areas P (such as an area 2) where there are all of alignment marks on the first exposure, and exposed after the second time in the areas P (such as an area 1) where there are few alignment marks.

Description

DETAILED DESCRIPTION OF THE INVENTION [Technical Field] The present invention relates to a semiconductor printing and exposure apparatus for transferring an integrated circuit pattern onto a photoresist layer of a wafer.

In particular, it relates to automatic alignment of a photomask or reticle (hereinafter both will be referred to as a mask) and a wafer in a stepper that performs stepwise exposure of a wafer.

[Prior art]

Semiconductor exposure equipment has made remarkable progress with the development of semiconductor integration technology, and its mechanisms include equipment that performs close-contact or extremely close-up printing, equipment that performs 1x printing using a lens or mirror system,
Alternatively, there is a wide variety of devices such as devices (so-called steppers) that print a reduced projection image on a wafer in steps.

As one of the mask and wafer alignment methods in a stepper, each shot is
There is a TTL die-by-die alignment method that aligns the disk and wafer. This alignment method has the advantage that highly accurate positioning can be achieved because alignment is performed for each shot.

Incidentally, in order to save time, it is usual to bake a plurality of chips at once per shot. However, a part of the chip is out of place at the periphery of the wafer, so when the mask-side alignment mark is located in that part, there is no corresponding wafer-side mark, so alignment becomes impossible and It was possible to sacrifice other chips included in the. Regarding this, an exposure apparatus has been proposed that aims to increase the number of chips obtained by providing means for selecting all or part of a plurality of alignment marks per shot. )
.

There are also cases where alignment marks are missing inside the wafer, making alignment impossible; however, an exposure apparatus has been proposed that solves this problem (Japanese Patent Application No. 1,40896/1982).

However, there is a positional deviation between the wafer and the mask, such as an error in the rotational direction, which cannot be obtained from measurement information of a portion of the plurality of alignment marks (one of the alignment marks or each half of the marks).

[Purpose of the invention]

The present invention has been proposed in view of the above, and an object of the present invention is to provide a semiconductor exposure apparatus that can improve alignment accuracy when aligning using some alignment marks.

〔Example〕

Hereinafter, the present invention will be described in detail with reference to the drawings.

FIG. 1(A) shows the main structure of a stepper equipped with an alignment mark detection system.

In the figure, 1 is a mask provided with an integrated circuit pattern and alignment marks, and is held on a mask stage 24. 6 is a reduction projection lens, 4 is a wafer provided with a photosensitive layer, and provided with alignment marks. 5
is a wafer stage, which holds the wafer 4. The mask stage 24 and the wafer stage 5 are arranged in the plane (X direction, Y direction) and in the rotational direction (θ
direction).

For example, first, the wafer stage 5 is moved in the XY directions to perform a step-and-repeat motion, and the mask stay i;'24 is moved in the XYθ directions to initially set the mask 1 in the apparatus. Next, move the mask stage 24 to
The mask 1 and the wafer 4 are aligned by moving in the Y direction and moving the wafer stage 5 in the θ direction. However, alignment may be performed by moving the mask stage for both XY 3-θ, or conversely, alignment may be performed by moving only the wafer stage. Incidentally, the step movement of the wafer stage 5 is executed according to a shot sequence stored in a control device (not shown).

Also, 20 and 20' are the alignment marks of mask 1, and 2
1 and 21' are alignment marks on the wafer 4.

In the figure, one rough alignment mark on the wafer 4 is drawn for convenience, but in reality, it is assumed that the same number of sets as the number of exposed shots are arranged.

Note that the alignment mark on mask 1''2 and the alignment mark on the wafer are aligned so that when a system other than the same-magnification projection system is used, the dimensions of both alignment marks do not change even if the projection or back projection is performed. Here, the dimensions of the wafer alignment mark are divided by the dimensions of the mask alignment mark to obtain the reduction magnification.

28 is a rotor 4-sided mirror (polygon mirror) that rotates around a rotation axis 29. The laser beam emitted from the laser light source 22 is passed through the rotating polygon mirror 28 by the lens 26.
It converges to a point 61 on the upper surface.

Lenses 32, 33, and 34 are lenses for relaying light beams, and 65 is a triangular prism.Since the apex of the triangular prism 65 coincides with the optical axis, one scan of the laser beam is divided into left and right halves, the first half and the second half. It will be done. Reference numeral 39 denotes a prism block for converting laser beam scanning in the direction in the drawing to scanning in the direction perpendicular to the drawing, and is configured in a shape as shown in FIG. 1 (Bl), for example.

42 is a half mirror forming a photoelectric detection system (43, 44, 45, 46), 43 is a mirror, 544 is a lens, 4
5 is a spatial filter, 46 is a condenser lens,
47 is a photodetector. 48°49.50.51 is a total reflection mirror, 52 is a prism, and 56 is an f-θ objective lens.

Further, 57 is a condenser lens, and 58 is a photodetector for detecting a synchronization signal. A part of the laser light incident on the semi-transparent mirror 42 is condensed by the condenser lens 57 and enters the photodetector 58. The photodetector 58 thus serves to detect the start and end points of the laser beam scan.

As you can see from the figure, the signal detection system consists of completely symmetrical left and right systems, and when the operator side is placed toward the front of the paper, the system shown with a dash will detect the signal on the right, and the system without a dash will detect the signal on the left. We call them systems, and the same parts are given the same number with a dash.

The relay lenses 62, 33, 34 form the origin of the deflection from the rotating polygon mirror 28 at the pupil 56 in the aperture position 55 of the objective lens 56. Therefore, the sheet-like or spot-like laser light is scanned over the mask and wafer by the rotation of the rotating polygon mirror 28.

In addition, in the objective lens system, an objective lens 56, an aperture 55
, the mirror 51 and the prism 52 are movable in the XY directions by a moving means (not shown), and the measurement positions of the mask 1 and the wafer 4 can be changed arbitrarily. For example,
When the mirror 51 moves in the direction indicated by the arrow A in the figure, the objective lens 56 and the diaphragm 55 simultaneously move in the entrance direction, and the prism 5 moves to keep the optical path length constant.
2 also moves in the A direction by an amount that is 1/2 of the amount of movement of the mirror 51.

On the other hand, when moving in the Y direction, the entire optical system for position detection moves in the Y direction (direction perpendicular to the plane of the paper).

Since the objective lens system can be moved in this way,
It is possible to cope with mark detection when the position of the alignment mark differs depending on the mask or when a mask setting mark is provided separately from the alignment mark.

100 is wafer stage 5 and mask stage 24
The details of its operation will be described later with reference to FIG. 6. C is a control circuit that controls the operation;
Further, M is a storage circuit in which a predetermined shot arrangement and step order are stored, and the amount of correction for positional deviation will also be stored later.

A signal selection circuit S selects both or one of the detection signals from the photodetectors 47 and 47' according to the shot arrangement and step order determined by the memory circuit M. 0 is an arithmetic circuit that calculates x, which is necessary for alignment, from two columns of detection signals.
The movement limits in the y and θ directions and the pitch error are calculated. Based on these, the mask stage 24 is moved to the alignment position via the first drive circuit, and the pitch error is stored in the memory circuit M. Furthermore, when one of the detection signals is selected, the pitch error stored in the memory circuit M is read out and used to correct the calculation result of the alignment data obtained with one of the detection signals. D2 is a second drive circuit, which drives the wafer stage 5 in steps according to the shot arrangement and step order determined by the memory circuit M. If the detection signal from the photodetectors 47, 47' is not complete because it corresponds to only one side or a partial mark, the signal selection circuit S determines this and applies it to an appropriate arithmetic expression, which will be described later.

FIG. 2(A) is a diagram showing alignment marks on the mask surface. Mark 20 (FIG. 1) on mask 1 consists of orthogonal pars 75 and 76, and similarly mark 20' consists of bars 75' and 76'.

Also, when the marks 21 and 21' on the sea urchin are back-projected onto the mask 1 through the projection lens, the mark 21 consists of parallel pars 71, 72, 73, and 74 in an orthogonal relationship, and the mark 21' consists of 71' , 72', 73', and 74'. The details of this alignment mark are JP-A-53-9
0872.

Here, the alignment mark is divided into two by the axis of symmetry and is called as follows.

Bar 71.72.75 marks the first part bar 73.7
4.76 second part mark par 72', 76'
, 76' to the third partial mark par 71', 72'
, 75' to the fourth partial mark laser beam is shown in Figure 2 (A
)60.60', the left detection system scans the hair alignment mark in the Y direction from the front of the device (bottom of the page), and the right detection system scans the hair alignment mark in the minus Y direction from the back of the device (top of the page). Figure 2 (B) is an enlarged view of the left detection system mark of the prisoner in Figure 2.

When the laser beam 60 scans the mark in FIG. 2(B),
The photodetectors 47 and 47' are equipped with the following components as shown in FIG. 2(C).
Signals 871 to 876 are obtained. 871.872.

S73... are signals corresponding to marks par 71, 72, 73..., respectively.

Therefore, the detection signal S71. S75,872. S73゜S
76.574(7) Intervals Wl, W2, W3,
By measuring W4 by W1, the amount of misalignment between the mask and the wafer can be detected.

X of the first partial mark and second partial mark of the left signal system
The amount of deviation △XL, △YL in the direction and Y direction is △XL=(W
l-W2-W3+W4)/4 Set (1) ΔYL=(-
WI+W2+W3-W4)/4... can be expressed by formula (2).

Also, the amount of deviation of the third partial mark and fourth partial mark of the right signal system, ΔXR, ΔYR, is ΔXR=(-W]'+W2'+W3'-W4')7M
-...Formula (3) △yR=(-w1'+w2'-w3'+
w4-')/4-・Equation (4) is obtained.

Therefore, the amount of deviation in the rotational direction △θ is △θ = (△YR - △YL) / (XR - XL) ... Formula (5) However, XL and XR are the centers of gravity of the left mark and right mark from the center of the mask. is the distance to the position and therefore (
XR-XL) is the distance between the left and right marks.

Note that the deviation in the X direction of the right detection system h) △XR and the X of the left detection system
The difference ΔXR−ΔXL from the directional deviation amount ΔXL should originally be zero since the left and right mark intervals on the mask and the left and right mark intervals on the wafer should be equal. However, in reality, it may not become zero, and this difference in the amount of deviation in the X direction (ΔXR−ΔXL)/2 is called a pitch error.

The main reasons why pitch errors occur are (a) expansion or contraction of the wafer or mask (1)) warping of the wafer (C) mask accuracy (d) printing magnification error and distortion in the previous process ( e) There is a lens effect of the resist layer, etc.

This amount of pitch error is included in the amount of deviation of the right detection system and the left detection system, and as a result, it is determined whether or not the amount of deviation is within the allowable range. This is not a problem in implementation.

On the other hand, auto-alignment using partial marks to which the present invention is applied is performed using the following partial mark combinations in both auto-alignment using one mark of a peripheral shot and auto-alignment due to missing alignment marks. In other words, ■the first part mark and the second part mark.

■Third part mark and fourth part mark.

■First part mark and fourth part mark.

■There are four combinations of the second part mark and the third part mark.

The positional deviation amount △X, ΔY for each combination is calculated as follows: In the case of ■...Equation (1) and equation (2) above.

In the case of ■...Equations (3) and (4) above.

In the case of ■... △X = (Wl - W2 + W3'-W4') /
4...Formula (6) ΔY = (-Wl +W2-W3
'+W4') / 4 ・= Formula (force, in case of ■... △X = (-W1'4-W2'-W3 +W4.)
/4 ... Formula (8) △Y = (-W1'-1-W
2'+W3-W4)/4...Equation (9).

In either case, the difference (△XR - △XT,)/2 between the amount of deviation in the X direction of the right detection system and the amount of deviation in the X direction of the left detection system cannot be detected, so if a pitch error occurs If measurement is performed using a partial mark on a wafer that has a wafer that is

Similarly, since the amount of deviation ΔYR in the Y direction of the right detection system and the amount ΔYL of deviation in the Y direction of the left detection system cannot be detected, it is not possible to calculate the amount of deviation Δθ in the rotational direction using equation (5). Therefore, in positioning using partial marks on the first shot of the wafer, it is not possible to correct (1) pitch error and (2) positional deviation in the co-rotation direction, resulting in a decrease in positioning accuracy. In this case, if the alignment using the partial alignment marks is corrected based on the amount of positional deviation caused by the left and right alignment marks previously made, a decrease in accuracy can be prevented. The present invention focuses on this point, and uses left and right alignment marks for the initial alignment of the wafer, thereby preventing a drop in alignment accuracy due to partial alignment marks on the first shot of the wafer. The present invention can be summarized as follows: (1) When the first shot is positioned at one mark according to the shot arrangement predetermined by the operator, the wafer is stepped so that the shot where both marks can be measured is located first. , auto-alignment is performed on that shot, and after the exposure of that shot or all shots is completed, the process returns to the shot of the one-sided mark.

■In addition, if auto-alignment using both marks is not possible in the first two-mark measurement shot (for example, if there is a defect in the mark), measurement using some alignment marks will not be performed, and the next shot where both marks can be measured is Proceed to the next step and perform alignment.

Next, the operation of the semiconductor exposure apparatus according to the embodiment of the present invention will be explained in detail with reference to the flowchart of FIG.

First, the wafer is transported under the projection lens system. Next, when the step-and-repeat operation is started in step 301, it is determined in branch 302 whether or not there is already an exposure shot.

Initially all shots are unexposed, so step 3
Proceed to 03. In step 303, it is determined whether the first shot has both left and right marks arranged based on the shot position information.

Now, FIG. 4, which shows a planar view of the wafer, will be explained. The area indicated by P in the figure is the area of one shot,
It is assumed that one shot consists of four chips indicated by area T. Further, it is assumed that the right mark is at position A and the left mark is at position B at auto-alignment mark position 1, for example, the second shot (the number written at the bottom left of the shot is the shot number). As shown in the figure, since the right mark position A' of the first shot is not on the wafer, one-sided mark alignment is performed using only the left mark.

Therefore, in the case of the shot arrangement shown in FIG. 4, since the first shot is one-sided mark alignment, no auto-alignment operation is performed, and the process immediately branches from step 303 to step 315 (FIG. 3).

Step 315 is a check to see if exposure of all shots on the wafer has been completed, but in this case all shots are still unexposed, so the process proceeds to step 316. Step 31
At step 6, the wafer stage performs a step operation to advance to the next shot, and returns to step 302 again. In other words, if the first shot is one-sided mark alignment, the process immediately steps to the second shot. In the example of FIG. 4, since both marks can be aligned in the second shot, the sequence is from step 302 to step 303.
Then, in step 304, auto-alignment using both marks is executed. If the auto alignment is performed correctly, the process proceeds from step 305 to step 306.
A mask image is printed onto the wafer by a predetermined exposure process.

After the exposure is completed, the process advances to step 307, and it is determined whether there is a shot that has passed exposure (unexposed) before that shot. In this example, since the first shot is a passed shot, the process branches to step 308, and the wafer stage is moved to the first shot.
Return to shot position. In this case, there is an exposure shot before the first shot, and the alignment marks are not both marks, so step 302 → step 309 → step 3
Proceed to 12. In step 312, information obtained by aligning both marks of the second shot.

That is, the positional information of the first shot by the one-sided mark is corrected based on the pitch error and the amount of rotational deviation, and then the first shot is aligned.

If the first shot is both mark alignment,
If auto-alignment using both marks is executed at the shot, there is no pass shot, so the process proceeds from branch 307 to steps 315 and 316, and the stage steps to the next shot position. On the other hand, if in step 304 all (
If auto-alignment of the (both) marks is impossible due to missing alignment marks, etc., auto-alignment using partial marks is not performed for that shot, and the process proceeds from step 305 to steps 315→316, and similarly moves to the next shot.

In other words, to summarize the following -1-: ■ If the first shot is a single mark, and the alignment of both marks on the second shot is OK, then shot 2 → shot 1 → shot 3 → shot 4... ■ The first shot If the shot is one mark, and the second shot's both mark alignment is No, shot 3 → shot 1 → shot 2 → shot 4... ■If the first shot is both marks, shot 1 → shot 2 → Shot 3 → Shot 4...
・Exposure is performed in the following order.

In either case, the shot that is first exposed is the shot that has been aligned by both marks.

After even one exposure is completed, all steps are step 3.
The process branches from step 02 to step 309, and after determining whether it is a double mark alignment mark or a single mark alignment mark based on predetermined shot position information, the process proceeds to step 310 to align both marks.

In step 312, one-sided mark alignment is performed.

If alignment using both marks is not performed correctly in step 310 due to missing alignment marks, etc., the process branches from step 311 to step 312, and in step 312 auto-alignment using some marks is executed.

If automatic alignment using a partial mark or a single mark is not possible, the process proceeds to step 314, and the fact that alignment is not possible is displayed on the display panel of the apparatus using characters or the like.

In this case, the unaligned shot may not be exposed and the process may proceed to the next shot, or the operator may perform alignment manually.

If all mark auto-alignment and partial mark auto-alignment are OK, step 310 is performed, respectively.
→311. Step 312 → 313 to step 306
Then, exposure is performed.

When exposure of all shots is completed in this way, the process advances to step 317, and the step-and-repeat operation is completed.

〔Effect of the invention〕

As explained above, according to the present invention, the number of chips acquired for one wafer is increased, and since the initial alignment on the wafer is performed using all of the multiple alignment marks (both marks), This has the effect that shots can be aligned with high precision using other alignment marks.

[Brief explanation of drawings]

FIG. 1(A) is a schematic configuration diagram of a semiconductor printing exposure apparatus according to an embodiment of the present invention, and FIG. 1(B) is a schematic diagram of a prism block used in a semiconductor printing exposure apparatus according to an embodiment of the present invention. 2(A) is a plan view of the mask alignment mark and wafer alignment mark on the mask surface, FIG. 2(B) is an enlarged view of the alignment mark of the left detection system in FIG. 2(A), FIG. 2(C) is an output signal diagram when the alignment mark shown in FIG. 2(B) is optically scanned, and FIG. 3 is a flowchart for explaining the operation of the semiconductor printing exposure apparatus according to the embodiment of the present invention. Figure, 4th
The figure is a plan view of a wafer showing a shot arrangement. 1 ・・・・・・・・・Mask 3 ・・・・・・・・・Reduction projection lens 4 ・・・
...Wafer 20.20'...Mask alignment mark 21.21'...Wafer alignment mark 39, 39'...Prism blocks 71-74...Left-hand wafer Alignment mark 7
5.76...Left mask alignment mark 71'
~74'...Right-hand wafer alignment mark 75
', 76'... Right-hand mask alignment mark 10
0 ・・・・・・・・・Control device for wafer stage and mask stage Figure 4

Claims (1)

[Claims] Step-type semiconductor printing exposure in which the relative position of the mask and the wafer is detected for each shot using a plurality of alignment marks on the mask and the wafer, the mask and the wafer are aligned, and then exposed. The apparatus further comprises means for selecting all or a part of the plurality of alignment marks, and the selection means selects alignment using all of the plurality of alignment marks for the first shot of the wafer. Semiconductor printing exposure equipment.