JPH10312958A - Fabrication of semiconductor integrated circuit device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To avoid the effect of misalignment due to coma in a process where normal illumination is mixed with modified illumination by using the normal illumination or the modified illumination, respectively, when a photomask having an inspection mark for normal illumination or modified illumination is exposed to a semiconductor wafer. SOLUTION: Alignment inspection marks corresponding to ring band illumination are arranged at four corners of a mask corresponding to ring band illumination. The inspection mark for ring band illumination serves as an alignment inspection mark for forming an active region 34, a word line 31 and a data line 32, respectively. Alignment inspection marks corresponding to normal illumination are arranged at four corners of a mask corresponding to normal illumination. The inspection mark for normal illumination serves as an alignment inspection mark for forming an opening and a plug electrode 35, respectively. The ring band illumination is used for exposure of the active region 34, the word line 31 and the data line 32 while the normal illumination is used for exposure of the opening and the plug electrode 35.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

The present invention relates to a technique for manufacturing a semiconductor integrated circuit device, and more particularly to a light exposure technique in a lithography process.

[0002]

2. Description of the Related Art Generally, a semiconductor integrated circuit device (hereinafter referred to as I
Called C. Is a process in which a pattern (hereinafter, referred to as a mask pattern) drawn on a photomask (hereinafter, referred to as a mask) called a reticle is transferred to a semiconductor wafer (hereinafter, referred to as a wafer) by lithography and etching. Is a repetition of In this lithography and etching process, the pattern transferred to the wafer in the previous step and the pattern transferred to the subsequent step need to be overlapped with high accuracy. Therefore, in a light exposure step in an IC lithography process, a target pattern (hereinafter, referred to as a target) for superposition is previously drawn on each mask in order to perform pattern superposition with high precision, and the target is drawn on a wafer. It inspects the degree of overlap between the copied target and the target on the mask and corrects misalignment.

[0003] In a reduction projection exposure apparatus (hereinafter, referred to as a stepper) used for manufacturing today's ultra-miniaturized ICs, alignment offset and focus stability are particularly problematic. The stepper uses a lens that uses the difference between the refractive index of air and glass, but the refractive index of the air fluctuates due to various factors such as temperature, pressure, humidity, carbon dioxide concentration, and oxygen concentration. And the exposure characteristics such as the projection magnification change. Further, a rise in the temperature of the lens due to exposure also contributes to these fluctuations. As a result, when the stepper is viewed on a scale of 0.01 μm, the exposure characteristics change with time. Therefore,
When exposing with a stepper, inspections such as offset of focus and alignment, rotation of magnification, and the like are performed using the target printed in the previous step, and these errors are corrected.

[0004] However, during this inspection operation, the production operation by the stepper is stopped, so that the throughput is reduced. In addition, since this operation is performed by actually exposing the wafer manually, there is a concern that an artificial error may occur. Therefore, a technique has been developed in which these errors are automatically measured without exposing the wafer, and the errors are automatically corrected. That is, the alignment inspection mark formed at a predetermined position of the mask (different from a target used in a normal exposure process).
Hereinafter, it is called a mark. ) Is irradiated with an exposure light beam through a projection lens to form an image on a slit formed in the XY table. A photo sensor is provided immediately below the slit on the XY table, and detects the intensity of light transmitted through the slit. The image forming position of the mark is obtained by slightly moving the XY table in each direction, obtaining the position where the detected light amount becomes maximum, and measuring the position of the XY table at that time. Since this automatic correction method uses an exposure optical system,
When a shift of the exposure optical system causes a shift of the image forming position of the pattern, the shift is always corrected by the automatic correction technique.

Examples of the light exposure apparatus are described in “Electronic Materials November 1995 Separate Volume”, published on November 22, 1996 by the Industrial Research Institute, Inc., p.
There is.

[0006]

As the IC becomes ultrafine, a so-called modified illumination method for changing the illumination shape of a stepper has been applied in order to increase the resolution. However, when this modified illumination method is applied to a stepper, although the alignment by the mark is appropriate,
The present inventor has clarified that there is a problem that an incorrect circuit alignment occurs in an actual circuit pattern (hereinafter, referred to as an actual pattern).
This phenomenon becomes conspicuous when ordinary circular illumination and deformed illumination are mixed.

Therefore, the present inventor analyzed this phenomenon by simulation, and when a modified illumination method was applied to a stepper, the influence of the coma aberration of the lens depending on the shape of the illumination caused by the conventional aberration. We investigated the phenomenon that displacement occurs due to fluctuations in illumination.

The present invention has been made based on this finding, and an object of the present invention is to provide a semiconductor integrated circuit device capable of avoiding the influence of misalignment due to coma even in a process in which ordinary illumination and modified illumination are mixed. It is to provide a manufacturing method.

The above and other objects and novel features of the present invention will become apparent from the description of the present specification and the accompanying drawings.

[0010]

The outline of a typical invention among the inventions disclosed in the present application is as follows.

That is, a photomask manufacturing step for manufacturing a plurality of photomasks on which a plurality of mutually related practical mask patterns and alignment inspection marks are respectively formed, and a method for manufacturing each of the practical masks on each photomask on the same semiconductor wafer. An exposure step in which the pattern and the alignment inspection mark are sequentially exposed so as to overlap, an actualizing step in which the respective practical mask patterns and the alignment inspection mark respectively exposed in the semiconductor wafer are exposed, and An inspection step of inspecting a positional deviation between an exposed alignment inspection mark pattern and an alignment inspection mark pattern which is later exposed, and a method for manufacturing a semiconductor integrated circuit device, comprising: , The alignment mask for normal illumination as each photomask. A photomask having a mask and a photomask having a deformed illumination alignment inspection mark are manufactured, and in the exposing step, when the photomask having the normal illumination alignment inspection mark is exposed on the semiconductor wafer, Normal illumination is used, and the modified illumination is used when a photomask having the alignment inspection mark for modified illumination is exposed on the semiconductor wafer.

According to the above-described means, the positional deviation caused by the difference between the normal illumination and the modified illumination can be offset by the alignment inspection mark for normal illumination and the alignment inspection mark for modified illumination. For this reason, for example, at the time of the alignment inspection of the normal illumination alignment inspection mark pattern in which the normal illumination alignment inspection mark is revealed, the normal illumination alignment inspection mark pattern is exposed by the modified illumination to the target to be aligned. With respect to the amount of misalignment between the deformed illumination alignment inspection mark pattern that has become apparent, the amount of misalignment associated with the difference between the normal illumination and the modified illumination can be excluded. Therefore, by measuring the amount of misalignment between the normal illumination alignment inspection mark pattern and the annular illumination alignment inspection mark pattern, the actual mask pattern formed on the photomask having the normal illumination alignment inspection mark becomes visible. The true misregistration between the practical pattern and the actual pattern revealed by the practical mask pattern formed on the photomask having the deformed illumination alignment inspection mark is measured. In other words, by correcting the misalignment amount between the normal illumination alignment check mark pattern and the modified illumination alignment check mark pattern, the true misalignment amount between the actual patterns to be revealed is corrected. As a result, the yield of the method for manufacturing a semiconductor integrated circuit device can be improved. After doing so, it is desirable that correction of the position shift due to the influence of the lens aberration of the practical pattern be separately performed in advance. That is, it is effective to perform position correction according to the shape of each practical pattern on the mask or by adjusting the lens.

[0013]

DESCRIPTION OF THE PREFERRED EMBODIMENTS FIG. 1 is a process chart showing main steps of an IC manufacturing method according to an embodiment of the present invention. FIG. 2 is a schematic view showing a stepper used in the IC manufacturing method according to one embodiment of the present invention. FIG. 3 and subsequent figures are explanatory diagrams for explaining the operation.

The stepper shown in FIG. 2 is used in the method of manufacturing an IC according to this embodiment. First, the outline of the stepper 10 will be described. The light emitted from the light source 11 is applied to the mask 70 supported on the mask table 17 via the fly-eye lens 12, the illumination shape adjustment aperture 13, the first condenser lens 14, the mirror 15, and the second condenser lens 16. The mask pattern drawn on the mask 70 is transmitted through the projection lens 18 to the sample stage 19.
Is projected onto the wafer 28 held by vacuum suction. The sample table 19 is moved by the Z-table drive 21 to the projection lens 18.
Is mounted on a Z table 20 which is moved in the optical axis direction (hereinafter, referred to as Z direction). Z table 20 is X
It is mounted on an XY table 22 that is moved in the XY directions by a Y table driving device 23. The Z table driving device 21 and the XY table driving device 23 are configured to be controlled by the controller 24. The controller 24 is a mirror 2 fixed to the Z table 20
The position of the sample table 19 is monitored by measuring the position of the sample table 19 with the laser length measuring device 26. In addition, the controller 24 operates the mask table 1
Table driving device 27 driving light source 7 and light source 1
1 is controlled.

The wafer 28 held on the sample stage 19
Is configured to be detected by a focus position detection device (not shown) including a detection light emitting unit and a light receiving unit, and the focus is automatically adjusted by this detection.

In the present embodiment, FIG.
Exposure is performed using the three types of illumination shape adjustment apertures shown in FIGS. The aperture A shown in FIG. 3A is for illumination (hereinafter, referred to as normal illumination) in which the light transmitting portion Aa and the light shielding portion Ab are arranged concentrically. The normal illumination A is effective for resolving the contact hole, and the ratio (coherent factor) of the radius r of the outer periphery of the light transmitting portion Aa when the outer radius R of the light shielding portion Ab is set to “1”.
Preferably, σ is 0.3 or 0.4. The aperture B shown in FIG. 3B is for illumination in which the light transmitting portions Ba are arranged in an annular shape (hereinafter, referred to as annular illumination). The annular illumination B is effective for exposing a wiring pattern, and it is preferable that the outer diameter σ of the light transmitting portion Ba is 0.7 and the inner diameter σ is 0.4. The aperture C illustrated in FIG. 3C is for illumination in which the light transmitting portion Ca is divided into four (hereinafter, referred to as four-division illumination). The quadrant illumination C is effective for resolving the wiring pattern, and the outer diameter σ
It is preferable that the circular light-transmitting portion Ca of which is 0.7 is shielded by a light-shielding portion Cb of a cross band having a width of σ0.3. The annular illumination B and the quadrant illumination C are examples of the modified illumination with respect to the normal illumination A.

FIG. 4 showing the relationship between the three types of illumination and the displacement will now be described. FIG. 4A shows an alignment inspection mark 1 formed on a mask. The alignment inspection mark 1 has a light-transmitting portion (hereinafter, referred to as a space) 3 formed at a central portion of a light shielding portion 2. It consists of a pattern consisting of FIG. 4B shows a sample 4 on which the alignment inspection mark 1 of FIG. 4A has been transferred and processed, and the sample 4 has a concave portion 5 formed by a space 3. FIG. 4C shows a waveform diagram of the position detection with respect to the concave portion 5 in the sample 4 of FIG. 4B. The center position of the concave portion 5 is obtained from the width 6 of the concave portion 5, and the amount of displacement is measured. Will be. FIG. 4D shows the relationship between the width of the space 3 and the displacement of the concave portion 5, in which the horizontal axis represents the value of the space width and the vertical axis represents the displacement. In FIG. 4D, curve 7 is a case where exposure is performed using normal illumination A, curve 8 is a case where exposure is performed using annular illumination B, and curve 9 is a case where exposure is performed using quadrant illumination C. Is shown.

The conditions for exposure of the stepper 10 when the relationship shown in FIG. 4D is obtained are as follows. Projection lens 1
8 has an NA of 0.6, an exposure wavelength λ of 365 nm, and a coma at a resolution position of space 3 of the projection lens 18 of 0.
4λ. In normal illumination A, σ was set to 0.3. In the annular illumination B, the outer diameter σ of the light transmitting portion Ba was set to 0.7 and the inner diameter σ was set to 0.4. In the four-division illumination C, the circular translucent portion Ca having an outer diameter σ of 0.7 is replaced by a cross-shaped light shielding portion Cb having a width of σ 0.3.
And was shielded from light. Note that the NA of the projection lens, the exposure wavelength, and the like are not limited to these values.

According to FIG. 4D, it is understood that the amounts of positional deviation in the curve 7 by the normal illumination A, the curve 8 by the annular illumination B, and the curve 9 by the four-division illumination C are different. In particular, the difference is remarkable between the curve 7 by the normal illumination, the curve 8 by the annular illumination, and the curve 9 by the quadrant illumination, and the difference between the curve 8 by the annular illumination and the curve 9 by the quadrant illumination. Is small. This is considered because normal illumination is normal incidence illumination, while annular illumination and quadrant illumination both belong to oblique incidence illumination. According to FIG. 4D, the space is 0.4 μm.
It is understood that, in the case of m, the positional deviation amount difference is small in all of the curves 7, 8 and 9.

Now, if the displacement of the space 3 having a width of 0.8 μm is observed using the normal illumination and the quadrant illumination, according to the curves 7 and 9 in FIG. Is 0.07 μm, the displacement due to the four-division illumination is 0.17 μm, and the difference between the displacements is 0.1 μm. Here, except for the difference between the normal illumination and the four-division illumination, the exposure conditions of the two are the same. Therefore, the difference in the amount of displacement is caused by the difference between the normal illumination and the four-division illumination. According to the analysis by the inventor's simulation, it has been found that the coma aberration of the projection lens fluctuates under the influence of the normal illumination and the four-division illumination, thereby causing a difference in the displacement amount.

In other words, by changing the normal illumination to the four-division illumination, the position of the space 3 is shifted, and it is detected as if the alignment inspection mark 1 was shifted. For example, when the alignment inspection mark 1 with the space 3 of 0.8 μm is used, it is determined that the reduction magnification or the like has changed by 0.1 μm even though there is no reduction magnification or expansion / contraction of the wafer. If the above-described correction is performed using this value, a position error of 0.1 μm or more occurs at a position where there is no coma aberration or a position where the coma aberration is opposite. The present invention uses an alignment inspection mark that is not easily affected by coma aberration, and selects an alignment inspection mark in the alignment reference layer and the alignment layer so that the influence of coma aberration is uniform. The device is devised so as to avoid the detection of the mark being displaced beforehand and to ensure proper alignment.

Hereinafter, a method of manufacturing an IC according to an embodiment of the present invention will be described mainly on a lithography step. In order to make the explanation easy to understand, a specific description will be given by taking a DRAM memory cell as an example.

FIG. 5A shows a memory cell 30 of a DRAM.
FIG. 5B is a sectional view of a main part. FIG.
In (a), many word lines 31 are wired in the vertical direction (hereinafter, referred to as Y direction) on the paper surface, and many data lines 32 are arranged in the horizontal direction on the paper surface (hereinafter, referred to as X direction). Wired. Capacitor 33 is formed above word line 31 and data line 32. On the active region 34 formed in the gap between the adjacent word lines 31, 31, a plug electrode 35 is provided in an elongated shape so as to be in contact with the active region 34 and extend to a region other than the active region 34. In addition, the data line 32 is wired so as to partially overlap the plug electrode 35. An opening 36 is formed over the active region 34.
, And the lower electrode 33 a of the capacitor 33 is connected through a conductor layer filled in the opening 36. The MISFET constituting the memory cell includes a gate insulating film 37, a gate electrode 38, a source 39 which is an n-type high-concentration impurity region, and a drain 40. The plug electrode 35 penetrates the silicon oxide film 41 on the source 39, and the data line 32 is laid on the silicon oxide film 41. Silicon oxide film 4 on data line 32
The lower electrode 33 a of the capacitor 33 is formed on the second 2, and the lower electrode 33 a is connected to the drain 40. The capacitor 33 includes a lower electrode 33a and a dielectric film 33b.
And the upper electrode 33c. An interlayer insulating film 43 made of a silicon oxide film and a wiring 44 made of aluminum are laid on the uppermost layer.

In the lithography step in the method of manufacturing the memory cell of the DRAM having the above configuration, the stepper 10 is used under the following exposure conditions. The wavelength of the exposure light beam is 365 nm, and the NA of the projection lens 18 is 0.60.
It is. For exposure of the active region 34, the word line 31, and the data line 32, the annular illumination B of FIG. 3B is used. In the exposure of the opening 36 and the plug electrode 35, σ in FIG.
Three normal lights A are used. As the mask, for example, a translucent phase shift mask described in JP-A-4-136854 or JP-A-6-175347 is used. The alignment of each layer is managed as follows. Word line 31 is aligned with active region 34. The opening 36, the plug electrode 35, and the data line 32 are aligned with the word line 31.

In the method of manufacturing an IC according to the present embodiment, two types of marks which are hardly affected by coma aberration are manufactured in advance in a mask manufacturing process as shown in FIGS. The illustration of the shifter portion in the translucent phase shift mask is omitted. By the way,
The mask manufacturing step 51 is a step of writing a pattern described by the circuit design data on a mask blank using an electron beam writing apparatus (not shown), and manufacturing a mask through lithography and etching.

The mask shown in FIG. 6 is a mask 60 having an alignment inspection mark corresponding to the annular illumination B. That is, a light-shielding film 62 made of a material such as chrome is applied to one main surface of the mask blank 61 of the mask 60, and a pattern is drawn by opening slits in the light-shielding film 62 by lithography and etching. I have. A desired mask pattern 63 is arranged at the center of the mask 60, and an alignment inspection mark (hereinafter, referred to as an annular illumination alignment inspection mark) 64 corresponding to orbicular illumination B is provided at four corners of the mask 60. Are arranged respectively.

In the present embodiment, the annular illumination alignment inspection mark 64 is an alignment inspection mark for forming the active region 34, the word line 31, and the data line 32, respectively. In other words, a mask for forming the active region 34 having the alignment inspection mark 64 for annular illumination, a mask for forming the word line 31 having the alignment inspection mark 64 for annular illumination, and the alignment inspection mark for annular illumination A mask for forming the data line 32 having 64 is manufactured. Incidentally, a mask for forming the word line 31 (hereinafter, referred to as a word line mask) is shown in FIG. 6 for convenience.

The alignment inspection mark 64 for annular illumination comprises a light shielding portion 65 and three spaces 66 formed in the light shielding portion 65 in a square frame shape. , 66 constitute two lines 67. That is, the alignment inspection mark 64 for annular illumination is constituted by a pattern including three spaces and two lines. The width (W) of the space 66 and the width (w) of the line 67 are 0.
The pitch is set at 5 μm, that is, the pitch (P) is also set at 1.0 μm.

The mask shown in FIG. 7 is a mask 70 having an alignment inspection mark for normal illumination A. That is, a light-shielding film 72 made of a material such as chromium is adhered to one main surface of the mask blank 71 of the mask 70, and a pattern is drawn by opening a slit in the light-shielding film 72 by lithography and etching. I have. A desired mask pattern 73 is arranged at the center of the mask 70, and alignment inspection marks (hereinafter, referred to as alignment inspection marks for normal illumination) 74 corresponding to the normal illumination A are provided at the four corners of the mask 70, respectively. Are located.

In the present embodiment, the alignment mark 74 for normal illumination is an alignment mark for forming the opening 36 and the plug electrode 35, respectively. That is,
A mask for forming the opening 36 having the alignment mark 74 for normal illumination, and the alignment mark 7 for normal illumination
4 and a mask for forming the plug electrode 35 having 4 are manufactured. By the way, in FIG.
(Hereinafter, referred to as an opening mask) 70 for forming the plug electrode 35 is schematically illustrated.
Illustration of a mask for formation is omitted.

The normal illumination alignment inspection mark 74 is constituted by a pattern comprising a light shielding portion 75 and a single space (hereinafter referred to as an isolated space) 76 formed in the light shielding portion 75 in a square frame shape. Isolated space 7
The width (W) of No. 6 is set to 0.4 μm.

In the exposure step 52, for example, when the word line 31 is exposed on the wafer 28, the word line mask 60 shown in FIG. 6 is mounted on the mask table 17 of the stepper 10. In this case, the illumination shape adjustment aperture 13 is adjusted to execute the annular illumination B. The exposure light having a wavelength of 365 nm emitted from the light source 11 is subjected to annular illumination B by the illumination shape adjustment aperture 13.
Is adjusted to irradiate the word line mask 60. Both the annular illumination alignment inspection mark 64 drawn at the four corners of the word line mask 60 and the word line pattern 63 corresponding to the word line drawn at the center have an NA of 0.60.
Is reduced and projected onto a wafer 28 held in vacuum on a sample stage 19 through the projection lens 18 of FIG.

Thereafter, X is controlled by the controller 24.
When the Y table driving device 23 is operated, the XY table 22
Is appropriately moved in the X and Y directions, so that
Exposure is sequentially performed on the entire surface of the wafer 28 by AND repeat.

The wafer 28 on which the transfer of the word line mask 60 has been completed is lowered from the sample stage 19 and sent to the developing step 53 and the etching step 54 in sequence. Development step 53
And the etching step 54, the wafer 2
8, a word line 31 is formed, and an annular illumination alignment inspection mark pattern 68 (see FIG. 9) is formed by the annular illumination alignment inspection mark 64. Then, in the alignment inspection step 55, the size and position of the annular illumination alignment inspection mark pattern (hereinafter, referred to as the annular illumination mark pattern) 68 are determined by a pattern inspection wafer inspection using an optical microscope or a scanning electron microscope. It is measured by an apparatus (hereinafter, referred to as a patterned wafer inspection apparatus, not shown).

FIG. 8A shows the relationship between the alignment inspection mark for annular illumination and the positional deviation for the three types of illumination.
(B) and (c) will be described. FIG. 8A shows three spaces 66 and two lines 67 of the annular illumination alignment inspection mark 64. FIG. 8B
Is an annular illumination mark pattern 68 lithographically formed by the annular illumination alignment inspection mark 64 of FIG.
Is shown. FIG. 8C shows a waveform chart of a result of dimension measurement for the annular illumination mark pattern 68 of FIG. 8B. In this case, the center of the width S of the central annular illumination mark pattern 68 is determined, the amount of displacement is measured (hereinafter, referred to as measurement method S), and the annular illumination marks on both sides are used. The center of the maximum width L of the mark patterns 68, 68 is obtained, and the amount of displacement is measured (hereinafter, referred to as a measuring method L). FIG. 8D shows the relationship between the value of the width of the space 66 and the line 67 and the amount of displacement of the annular illumination mark pattern 68. The width of the space 66 and the line 67 is plotted on the horizontal axis. The vertical axis indicates the displacement amount. In FIG. 8D, a curve 81S indicates a normal illumination A.
The curve 82S is a displacement curve due to the measurement method S when exposed using the annular illumination B, and the curve 83S is a quadrant illumination C when exposed using the annular illumination B. The curve 81L is a displacement curve due to the measuring method S when the exposure is performed after exposure, and a curve 81L is a displacement curve according to the measuring method L when the exposure is performed using the normal illumination A.
L is a measurement method L when exposed using the annular illumination B
83L is a displacement curve due to the measurement method L when exposed using the quadrant illumination C,
Are respectively shown.

The conditions for exposure of the stepper 10 when the relationship shown in FIG. 8D is obtained are as follows. Projection lens 1
8 has an NA of 0.6, an exposure wavelength λ of 365 nm, and a coma aberration of the projection lens 18 at the resolving position of the annular illumination alignment inspection mark 64 of 0.4λ. The normal illumination A is performed with σ of 0.3, the annular illumination B is performed with σ of the outer diameter of the light transmitting portion Ba of 0.7, and the σ of the inner diameter of 0.4, and the quadrant illumination C is performed with the outer diameter of 0.4. Is performed by shielding the circular light-transmitting portion Ca having a σ of 0.7 with a light-shielding portion Cb of a cruciform band having a width of σ of 0.3.

According to FIG. 8D, a curve 81S according to the measuring method S under normal illumination, a curve 82S according to the measuring method S under annular illumination, and a curve 83S according to the measuring method S under four-division illumination.
It can be understood that the amount of misregistration is different. In particular, the difference is remarkable between the curve 81S according to the measurement method S under normal illumination, the curve 82S according to the measurement method S under annular illumination, and the curve 83S according to the measurement method S under four-division illumination. The difference between the curve 82S according to the measurement method S under illumination and the curve 83S according to the measurement method S under four-division illumination is small. This is considered because normal illumination is normal incidence illumination, while annular illumination and quadrant illumination both belong to oblique incidence illumination.

Similarly, the amount of displacement between the curve 81L according to the measurement method L under normal illumination, the curve 82L according to the measurement method L under annular illumination, and the curve 83L according to the measurement method L under four-division illumination may differ. Understood. In particular, there is a remarkable difference between the curve 81L according to the measurement method L under normal illumination, the curve 82L according to the measurement method L under annular illumination, and the curve 83L according to the measurement method L under four-division illumination. The difference between the curve 82L according to the measurement method L under illumination and the curve 83L according to the measurement method L under four-division illumination is small.

In turn, the space 66 and the line 6 of the alignment inspection mark 64 for annular illumination of the word line mask 60
7, the width (W) is set to 0.5 μm. Therefore, according to FIG. 8D, the amount of displacement under the annular illumination B is about 0 in the case of the curve 82S based on the measurement method S. .
12 μm, about 0.2 in the case of curve 82L according to measurement method L.
That is, 14 μm.

The wafer 28 having undergone the alignment inspection step 55 is sent again to the exposure step 52. In the exposure step 52, for example, when the opening 36 is exposed on the wafer 28, the opening mask 70 shown in FIG. 7 is mounted on the mask table 17 of the stepper 10. In this case, the illumination shape adjustment aperture 13 is adjusted to execute the normal illumination A. When the wavelength emitted from the light source 11 is 3
The exposure light beam of 65 nm is adjusted to the normal illumination A by the illumination shape adjustment aperture 13 and is applied to the opening mask 70. Both the normal illumination alignment inspection mark 74 drawn at the four corners of the opening mask 70 and the opening pattern 73 corresponding to the opening drawn at the center are sampled through the projection lens 18 having an NA of 0.60. The wafer 28 is reduced and projected on the wafer 28 held in vacuum on the table 19 and exposed.

Thereafter, X is controlled by the controller 24.
When the Y table driving device 23 is operated, the XY table 22
Are appropriately moved in the X and Y directions, and exposure is sequentially performed on the entire surface of the wafer 28 in a step-and-repeat manner.

The wafer 28 on which the transfer of the opening mask 70 has been completed is lowered from the sample stage 19 and sent to the developing step 53 and the etching step 54 in sequence. Through the development step 53 and the etching step 54, the wafer 28
Is formed with an opening 36. Here, the developing step 53
In FIG. 9, as shown in FIG. 9, a normal illumination alignment inspection mark pattern (hereinafter, referred to as a normal illumination mark pattern) 7 by the ordinary illumination alignment inspection mark 74.
8 are formed. Then, in the alignment inspection step 55, the size and position of the normal illumination mark pattern 78 are measured by the patterned wafer inspection apparatus.

Here, the alignment mark 74 for normal illumination of the mask 70 for an opening is an isolated space 76 and the width (W) is set to 0.4 μm.
According to curve 7, the amount of displacement under normal illumination is
That is, about 0.12 μm. On the other hand, as described above with reference to FIG. 8, the positional shift amount of the word line mask 60 with respect to the space 66 (0.5 mm) of the annular inspection mark 64 for annular illumination under annular illumination is represented by the curve 82S obtained by the measurement method S. Is about 0.12 μm. Therefore, the amount of misalignment of the alignment inspection mark 74 for normal illumination in the isolated space 76 having a width of 0.4 μm is substantially equal to the amount of misalignment of the alignment inspection mark 64 for annular illumination by the measurement method S. That is, even if a positional shift occurs due to the difference between the normal illumination A and the annular illumination B, the positional deviation amount of the alignment inspection mark 74 for normal illumination in the isolated space 76 having a width of 0.4 μm due to the variation and the annular zone It is substantially equal to the amount of misalignment of the illumination alignment inspection mark 64 by the measurement method S. That is, the amount of positional deviation caused by the difference between the normal illumination A and the annular illumination B is determined by the 0.4 μm width of the isolated space 76 of the normal illumination alignment inspection mark 74 and the width S of the annular illumination alignment inspection mark 64. Will be offset by the method of the above.

Therefore, as shown in FIG. 9, in the alignment inspection of the normal illumination mark pattern 78, the positional deviation amount between the normal illumination mark pattern 78 and the annular illumination mark pattern 68 is determined by the ordinary illumination mark pattern. The amount of displacement due to the difference between A and the annular illumination B can be taken out of consideration. That is, the center 6 of the annular illumination mark pattern 68
By measuring the position shift amount G of the center 79 of the normal illumination mark pattern 78, the true position shift amount between the word line pattern 63 and the opening pattern 73 is measured. Therefore, since the displacement can be accurately corrected, the yield of the IC manufacturing method can be improved. For example, in the case of the example shown in FIG.
Since the position is displaced rightward in the direction by the amount G, correction to return leftward in the X direction is executed.

As described above, according to the present embodiment, the positional deviation caused by the difference between the normal illumination A and the annular illumination B is set to the position of the normal illumination alignment inspection mark 74 by 0.4 mm.
The offset between the isolated space 76 having a width of μm and the case using the measurement method S for the annular illumination alignment inspection mark 64 allows the normal illumination mark pattern 78 and the ring to be aligned in the alignment inspection of the ordinary illumination mark pattern 78. Regarding the displacement G with respect to the band illumination mark pattern 68,
Since the amount of misalignment due to the difference between the normal illumination A and the annular illumination B can be ignored, the amount of misalignment G between the normal illumination mark pattern 78 and the annular illumination mark pattern 68 can be considered.
Is measured, the true positional deviation between the word line pattern 63 and the opening pattern 73 can be measured. In other words, by executing a correction for improving the positional deviation amount G between the normal illumination mark pattern 78 and the annular illumination mark pattern 68, the true positional deviation between the word line pattern 63 and the opening pattern 73 is corrected. As a result, the yield of the IC manufacturing method can be improved.

The above inspection principle and correction principle are relatively applicable not only when the annular illumination B is changed to the normal illumination A but also when the normal illumination A is changed to the annular illumination B. Needless to say, there is. Further, the present invention can be applied to a space between the normal illumination A and the four-division illumination C. For example, the alignment mark 74 for normal illumination has a width of 0.4 μm.
When the isolated space 76 is set as shown in FIG.
According to the four-part illumination alignment inspection mark (not shown)
When the width of the space and the line is set to 0.6 μm and the measuring method S for measuring the width of the central space is selected, the amount of displacement due to the change in illumination is about 0.12 μm.
m, the amount of misalignment due to coma caused by a change between the normal illumination A and the four-division illumination C can be offset.

Note that both the annular illumination B and the four-division illumination C belong to the modified illumination (oblique illumination illumination) and have the same tendency in the amount of positional shift. It is considered that the necessity of applying the inspection principle and the correction principle for canceling the position shift amount due to the coma aberration due to the change between the above is small.

Further, for example, when the active region 34 is an annular illumination B
As in the case where the word line 31 is also exposed by the annular illumination B, there is no positional deviation due to coma between exposure steps in which the illumination method is not changed. It is not necessary to apply the inspection principle and the correction principle for canceling the displacement.

Although the invention made by the inventor has been specifically described based on the embodiment, the invention is not limited to the embodiment, and various modifications can be made without departing from the gist of the invention. Needless to say.

For example, the alignment inspection mark for normal illumination is constituted by an isolated space, and the alignment inspection mark for annular illumination or the four-division illumination is not limited to the combination of space and line. The alignment inspection mark may be composed of a combination of a space and a line, and the annular illumination alignment inspection mark or the quadrant illumination alignment inspection mark may be composed of an isolated space.

The mark formed by a combination of a space and a line is not limited to a combination of three spaces and two lines, but may be constituted by a combination of four or more spaces and three or more lines. Is also good. Further, the space width (W) and the line width (w) are not limited to be equal to each other, but may be different from each other. That is, the width of the space may be varied while keeping the pitch of the space constant.

FIG. 10 shows the relationship between the normal illumination, the annular illumination, the four-division illumination, and the positional deviation with respect to the alignment inspection mark constituted by the pattern composed of the combination of five spaces and four lines. . FIG.
An alignment inspection mark 94 formed by a pattern composed of a combination of a stripe space 96 and four stripe lines 97 is shown. FIG. 10B shows an alignment inspection mark pattern (hereinafter, referred to as a mark pattern) 98 lithographically formed by the alignment inspection mark 94 of FIG. 10A. FIG. 10C shows a waveform diagram of a result of dimension measurement for the mark pattern 98 of FIG. 10B.
The center of the width S of the central mark pattern 98 is determined, and the amount of displacement is measured (hereinafter, referred to as measurement method S).
And the maximum width L of the mark patterns 98 at both ends.
Are determined and the amount of displacement is measured (hereinafter, referred to as measurement method L). FIG. 10D shows that the pitch is 1 μm.
The horizontal axis indicates the relationship between the width value of the space 96 and the amount of displacement of the mark pattern 98 when the width value of the space 96 is changed while keeping the width of the space 96 constant. The vertical axis indicates the displacement amount. In FIG. 10D, a curve 91S is a displacement curve by the measurement method S when the exposure is performed using the normal illumination A,
Curve 92S is a displacement curve by measurement method S when exposed using annular illumination B, curve 93S is a displacement curve by measurement method S when exposed using quadrant illumination C,
Curve 91L is a displacement curve according to measurement method L when exposed using normal illumination A, curve 92L is a displacement curve according to measurement method L when exposed using annular illumination B, and curve 93L is divided into four. 3 shows a displacement curve by the measurement method L when exposure is performed using the illumination C.

The exposure conditions of the stepper when each curve in FIG. 10D is obtained are as follows. The NA of the projection lens is 0.6, the exposure wavelength λ is 365 nm, and the coma aberration at the resolution position of the mark 90 of the projection lens is 0.4λ. The normal illumination A is performed when σ is 0.3, the annular illumination B is performed when the outer diameter σ of the translucent portion Ba (see FIG. 3) is 0.7, and the inner diameter σ is 0.4, and is divided into four. The illumination C has an outer diameter σ of 0.
7, the circular light-transmitting portion Ca was shielded by a cross-shaped light-shielding portion Cb having a width of 0.3.

According to FIG. 10D, a curve 91S according to the measuring method S under normal illumination, a curve 92S according to the measuring method S under annular illumination, and a curve 93 according to the measuring method S under four-division illumination.
It is understood that the amount of the displacement in S is different.
In particular, the difference is remarkable between the curve 91S according to the measurement method S under normal illumination, the curve 92S according to the measurement method S under annular illumination, and the curve 93S according to the measurement method S under four-division illumination. There is no difference between the curve 92S according to the measurement method S under illumination and the curve 93S according to the measurement method S under four-division illumination with a space width of 0.55 μm to 0.65 μm. Similarly, curve 91 according to measurement method L under normal illumination
L, it is understood that the amount of displacement in the curve 92L according to the measurement method L under annular illumination and the curve 93L according to the measurement method L under four-division illumination are different. In particular, there is a remarkable difference between the curve 91L according to the measurement method L under normal illumination, the curve 92L according to the measurement method L under annular illumination, and the curve 93L according to the measurement method L under four-division illumination. Curve 92L according to measurement method L under illumination and measurement method L under quadrant illumination
The difference between the curve 93L and the curve 93L is small.

For example, according to the displacement curve 81S by the measurement method S under the normal illumination in FIG. 8D and the displacement curve 92L by the measurement method L under the annular illumination in FIG.
When the annular inspection mark 64 is composed of a combination of three 0.4 μm spaces and two lines, the alignment inspection mark 94 is formed of five 0.4 μm lines.
In the case where the spaces are arranged at a pitch of 1 μm, the positional deviation amounts are approximately equal at about 0.14 μm. Therefore, these alignment inspection marks can be used as a normal illumination alignment inspection mark and an annular illumination alignment inspection mark.

When the optical system of the stepper is different,
The pitch P and the width W of the space for the alignment inspection mark can be set to optimal values according to the optical system based on the following formulas. Where λ is the wavelength of the exposure light, NA
Is the numerical aperture of the projection lens, P = k ₁ × (λ / NA), where k ₁ is 1.4 to 1.8. W = k ₂ × (λ / NA) · ... However, k ₂ is a 0.8 to 1.3.

In the above-described embodiment, a method for preventing a decrease in alignment accuracy due to a positional shift due to coma aberration of a lens depending on an illumination condition has been described. According to the method for preventing a decrease in alignment accuracy, it is possible to prevent a relative positional shift between alignment layers of alignment inspection marks, and thus it is possible to prevent an increase in alignment shift of an actual pattern. Incidentally, the relative displacement between the actual patterns may also occur depending on the coma aberration. As a method of preventing the positional deviation between the actual patterns depending on the coma aberration, a method of correcting the position of the transfer pattern in advance by adjusting a lens corresponding to each actual pattern can be considered. For example, when the actual pattern is mainly composed of a pattern consisting of simple lines and spaces, the displacement of the pattern is measured in advance, and the lens of a stepper (projection exposure optical device) used in the exposure process is used. By adjusting the position, the positional deviation of the pattern can be corrected.

According to the correction method by adjusting the lens, the correction is usually made concentrically from the center of the lens. For example, the actual pattern is mainly composed of vertical lines (lines extending in the Y direction). In such a case, the displacement caused by the coma of the lens is different between the horizontal direction (X direction) and the vertical direction (Y direction). Therefore, if the position correction of the actual pattern is executed while paying attention to the horizontal position shift, the vertical position shift will be enlarged. Therefore, when using the correction method by adjusting the lens, it is necessary to set a large allowable amount of misalignment in advance with respect to misalignment other than the direction of interest. In addition, the alignment inspection mark used in this case needs to be formed of a pattern that is substantially the same as the pattern used in the direction of interest of the actual pattern, and reflects the positional deviation of the actual pattern.

FIG. 11 is a view for explaining a correction method by lens adjustment according to another embodiment of the present invention.
Is a plan view of a mask, (b) is a plan view of an actual pattern,
(C) is a plan view of the alignment inspection mark.

An actual pattern area 102 is set at the center of the mask 101, and alignment inspection mark areas 103 are set outside the four corners of the actual pattern area 102, respectively. The real pattern 104 formed in the real pattern region 102 is, for example, a pattern constituting a word line, and is configured as shown in FIG. That is, the actual pattern 104 has a width of 0.35 μm.
Are formed in a pattern in which both ends are aligned in parallel with a pitch P of 0.7 μm. In each alignment inspection mark area 103, FIG.
The alignment inspection marks 105 shown in FIG. 1 (c) are respectively formed. The alignment inspection mark 105 has a width of 0.
A plurality of 35 spaces are arranged in parallel with both ends being aligned at a pitch P of 0.7 μm and arranged in parallel, and this space group is constituted by a pattern arranged in a square frame shape. That is, the alignment inspection mark 105 is set to be as equal as possible to the actual pattern 104 in the vertical direction (Y direction). Incidentally, in the alignment inspection mark 105, lines and spaces of horizontal lines (lines extending in the X direction) used for position detection in the Y direction are also patterned in the same manner as vertical lines and spaces.

When the mask 101 configured as described above is transferred to a wafer, the positional deviation due to the coma of the lens becomes substantially the same between the actual pattern 104 and the alignment inspection mark 105. Therefore, the displacement of the pattern of 0.35 μm can be measured in advance, and the projection lens of the stepper can be adjusted to correct the displacement due to the coma of the lens.

FIG. 12 shows 0.3 before and after the lens adjustment.
FIG. 5 is a diagram showing a displacement of a 5 μm pattern with respect to an absolute lattice.

In FIG. 12, reference numeral 106 denotes a curve before lens adjustment, and 107 denotes a curve after lens adjustment. As is clear from the comparison between the curve 106 before the lens adjustment and the curve 107 after the lens adjustment, the positional deviation of the actual pattern and the alignment inspection mark can be reduced.

For example, lens adjustment for other layers having actual patterns other than word lines is performed by adjusting each correction condition in accordance with the illumination conditions to be used and the dimensions and arrangement pitch of the line and space patterns to be resolved. Decide each.
The target value of the correction is set so that the arrangement of the line and space patterns approaches the absolute grid. When the actual pattern is formed by an isolated pattern such as a hole, the alignment inspection mark is formed by an isolated pattern having substantially the same influence of the coma of the lens. Also in this case, it is a feature that the alignment inspection mark is configured so as to have substantially the same positional deviation as the actual pattern, and both need not be configured to have the same pattern shape.

Further, as a method of preventing the displacement of the alignment test mark due to the influence of coma aberration, the alignment test mark is arranged in an area where the coma aberration is small in one transfer shot, that is, an area where the pattern transfer position error is small. Accordingly, an error due to coma aberration in alignment measurement can be minimized. For example, by arranging the alignment inspection mark so as to be substantially located at the point M in FIG. 12, the position error can be suppressed to 0.01 μm or less irrespective of the pattern composed of lines and spaces.

In the above description, the invention made mainly by the present inventor is described in terms of the DRA which is the application field in which the invention is based.
Although the description has been given of the case where the present invention is applied to the manufacturing technique of the M memory cell, the present invention is not limited thereto, and the present invention can be applied to the entire method of manufacturing a semiconductor integrated circuit device such as a gate array.

[0067]

The effects obtained by typical aspects of the invention disclosed in the present application will be briefly described as follows.

By using the normal illumination mark pattern and the modified illumination mark pattern, the amount of displacement caused by the difference between the normal illumination and the modified illumination can be taken out of consideration. By executing the correction for improving the amount of misalignment with the deformed illumination mark pattern, the true amount of misalignment of the actual pattern can be corrected, and as a result, the yield of the semiconductor integrated circuit device manufacturing method can be increased. Can be.

[Brief description of the drawings]

FIG. 1 is a process chart showing main steps of an IC manufacturing method according to one embodiment of the present invention.

FIG. 2 is a schematic view showing a stepper used in a method of manufacturing an IC according to an embodiment of the present invention.

3A and 3B are schematic diagrams showing illumination shapes, where FIG. 3A shows normal illumination, FIG. 3B shows annular illumination, and FIG. 3C shows quadrant illumination.

4A and 4B show the relationship between the alignment inspection mark and the positional deviation for three types of illumination, wherein FIG. 4A is a schematic diagram of the alignment inspection mark, and FIG. 4B is a lithography using the alignment inspection mark of FIG. Sectional view of sample, (c) is (b)
FIG. 7D is a waveform diagram of the measurement results of the dimensions of the concave portion of the sample, and FIG. 9D is a diagram showing the relationship between the space width of the alignment inspection mark and the displacement.

FIG. 5 shows a memory cell of a DRAM, and (a)
Is a bottom view, and (b) is a front sectional view along line bb in (a).

6A and 6B show a word line mask having an annular illumination alignment inspection mark, wherein FIG. 6A is a bottom view and FIG. 6B is a front sectional view taken along line bb of FIG.

FIGS. 7A and 7B show a mask for an opening having an alignment inspection mark for normal illumination, wherein FIG. 7A is a bottom view and FIG.
FIG. 4 is a front sectional view taken along line bb of FIG.

FIGS. 8A and 8B show the relationship between the alignment inspection mark for annular illumination and the positional deviation for three types of illumination, wherein FIG. 8A is a schematic diagram of the alignment inspection mark for annular illumination, and FIG. Cross-sectional view of the annular illumination alignment inspection mark pattern lithographically formed by the annular illumination alignment inspection mark,
(C) is a waveform diagram of the measurement result of (b), and (d) is a diagram showing a relationship between a space, a line width, and a positional shift of the alignment inspection mark for annular illumination.

9A is a front cross-sectional view showing an alignment inspection mark pattern for annular illumination, and FIG. 9B is a front cross-sectional view showing the relationship between the alignment inspection mark pattern for annular illumination and the alignment inspection mark pattern for normal illumination. , (C) is also a plan view, (d)
Is a measurement waveform diagram.

FIG. 10 shows a relationship between an alignment inspection mark and misalignment according to another embodiment of three types of illumination;
(A) is a schematic view of an alignment inspection mark according to another embodiment,
(B) is a cross-sectional view of a mark pattern lithographically formed by the alignment inspection mark of (a), (c) is a waveform diagram of the measurement result of (b), and (d) is a space, line width and position of the alignment inspection mark. FIG. 4 is a diagram illustrating a relationship with a shift.

11A and 11B are diagrams for explaining a correction method by lens adjustment according to another embodiment of the present invention, wherein FIG. 11A is a plan view of a mask, FIG. 11B is a plan view of an actual pattern, and FIG. It is a top view of an inspection mark.

FIG. 12 is a diagram showing a positional shift of a pattern with respect to an absolute grating before and after lens adjustment.

[Description of sign]

DESCRIPTION OF SYMBOLS 1 ... Alignment inspection mark, 2 ... Light shielding part, 3 ... Transmissive part (space), 4 ... Sample, 5 ... Concave part, 6 ... Width, 7, 8, 9 ... Curve, 10 ... Stepper (projection exposure optical device), 11 ... light source, 12 ... fly-eye lens, 13 ... illumination shape adjustment aperture, 14 ... first condenser lens, 15 ... mirror,
16: second condenser lens, 17: mask table,
18: Projection lens, 19: Sample table, 20: Z table,
21 ... Z table driving device, 22 ... XY table, 23
... XY table drive, 24 ... controller, 25 ...
Mirror, 26: laser length measuring device, 27: mask table driving device, 28: wafer, A: normal illumination (aperture for normal illumination), B: annular illumination (aperture for annular illumination), C
... Four-division illumination (four-division illumination aperture), G: misregistration amount, 30: DRAM memory cell, 31: word line, 32: data line, 33: capacitor, 33a: lower electrode, 33b: dielectric film, 33c upper electrode, 34 active region, 35 plug electrode, 36 opening, 37 gate insulating film, 38 gate electrode, 39 source, 40 drain, 41, 42 silicon oxide film, 43 interlayer Insulating film, 44 wiring, 51 mask manufacturing process, 52 exposure process, 53 developing process, 54 etching process, 55 alignment test process, 60 mask for word line, 61 mask blank, 62 light shielding film , 63 ... word line pattern, 6
4 ... alignment inspection mark for annular illumination, 65 ... light shielding part, 66
... space, 67 ... line, 68 ... annular illumination mark pattern, 69 ... center, 70 ... opening mask, 71 ... mask blank, 72 ... light shielding film, 73 ... opening pattern,
74: alignment inspection mark for normal illumination, 75: light shielding part, 7
6 ... isolated space 78 ... normal illumination mark pattern
79: center, 81S: misalignment curve by measuring method S under normal illumination, 82S: misalignment curve by measuring method S under annular illumination, 83S: misalignment curve by measuring method S under quadrant illumination, 81L ... Displacement curve by measuring method L under normal illumination, 82L: misalignment curve by measuring method L under annular illumination, 83L: misalignment curve by measuring method L under four-division illumination, 90: alignment inspection mark, 91S ... A curve according to the measurement method S under normal illumination, 92S a curve according to the measurement method S under annular illumination, 93S a curve according to the measurement method S under four-division illumination,
91L: Curve by measurement method L under normal illumination, 92L: Curve by measurement method L under annular illumination, 93L: Curve by measurement method L under four-division illumination, 94: Five spaces and four lines , 96 ... space, 97 ... line, 98 ... alignment inspection mark pattern,
101: mask, 102: actual pattern area, 103: alignment inspection mark area, 104: actual pattern, 105: alignment inspection mark, 106: curve before lens adjustment, 107
… Curves after lens adjustment.

──────────────────────────────────────────────────の Continued on the front page (51) Int.Cl. ⁶ Identification code FI H01L 27/108 H01L 21/30 522B 21/8242 27/10 621C (72) Inventor Takeshi Kato 5-chome, Kamizuhoncho, Kodaira-shi, Tokyo No. 20 In the Semiconductor Division, Hitachi, Ltd. (72) Inventor Masayuki Hiranuma 5-2-1, Kamisumihonmachi, Kodaira-shi, Tokyo In the Semiconductor Division, Hitachi, Ltd. (72) Inventor Tsuneo Terasawa Kokubunji, Tokyo 1-280, Higashi-Koigakubo, Hitachi, Ltd.Central Research Laboratory, Hitachi, Ltd. 2350 Nippon Texas Instruments Co., Ltd. (72) Inventor Yasuhiro Homma Miura Muraki, Inashiki-gun, Ibaraki 2350 Japan Texas vinegar Instruments within Co., Ltd.

Claims (16)

[Claims] A photomask manufacturing step of manufacturing a plurality of photomasks each having a plurality of practical mask patterns and alignment inspection marks associated with each other;
An exposure step of simultaneously exposing each practical mask pattern and alignment test mark of each photomask on the same semiconductor wafer, and an exposure step of exposing each practical mask pattern and alignment test mark exposed on the semiconductor wafer A method of manufacturing a semiconductor integrated circuit device, comprising: a step of inspecting a misalignment between an alignment test mark pattern previously exposed after the exposure step and an alignment test mark pattern subsequently exposed after the exposure step. In the photomask manufacturing process, a photomask having a normal illumination alignment inspection mark as each of the photomasks, and a photomask having a modified illumination alignment inspection mark are manufactured, respectively. A photomask having a normal illumination alignment inspection mark is A semiconductor integrated circuit, wherein a normal illumination is used when exposing to c, and a modified illumination is used when exposing the photomask having the deformed illumination alignment inspection mark to the semiconductor wafer. Device manufacturing method. 2. The alignment mark for normal illumination and the alignment inspection mark for modified illumination have a relationship that substantially cancels a positional shift caused by a change between the ordinary illumination and the modified illumination. 2. The method of manufacturing a semiconductor integrated circuit device according to claim 1, wherein the settings are made as follows. 3. The alignment mark for normal illumination and the alignment mark for modified illumination make the influence of coma of the exposure optical system substantially the same between the ordinary illumination and the modified illumination. 3. The method of manufacturing a semiconductor integrated circuit device according to claim 1, wherein the settings are made so as to be related to each other. 4. The alignment mark for normal illumination is formed of one of an isolated pattern and a plurality of patterns, and the alignment mark of modified illumination is formed of the other of an isolated pattern or a plurality of patterns. 4. The method for manufacturing a semiconductor integrated circuit device according to claim 1, wherein 5. The isolated pattern width is W for the pattern width, NA for the numerical aperture of the projection lens, λ for the wavelength of the exposure light beam,
5. The method according to claim 4, wherein, when 0.5 ≦ n1 ≦ 0.9, W = n1 · λ / NA is satisfied. 6. A pattern pitch of the plurality of patterns, wherein the pitch is P, the pattern width is W, the numerical aperture of the projection lens is NA,
5. The method according to claim 4, wherein, when the wavelength of the exposure light beam is λ and 0.5 ≦ n2 ≦ 0.9, P = W = n2 · λ / NA. A method for manufacturing a semiconductor integrated circuit device. 7. In the inspection step, when inspecting the alignment inspection mark pattern that has been exposed by exposing the plurality of patterns to the semiconductor wafer, the center of the alignment inspection mark is included in the alignment inspection mark pattern. 6. The method for manufacturing a semiconductor integrated circuit device according to claim 4, wherein the width of the pattern corresponding to the pattern located in the step (a) is measured, and the positional deviation of the alignment test mark pattern is inspected. 8. In the inspection step, when inspecting the alignment inspection mark pattern that has been exposed by exposing the plurality of patterns to the semiconductor wafer, both ends of the alignment inspection mark are included in the alignment inspection mark pattern. 7. The method for manufacturing a semiconductor integrated circuit device according to claim 4, wherein a width between a pair of patterns respectively corresponding to the patterns located at the position is measured, and a positional shift of the alignment test mark pattern is inspected. . 9. The alignment inspection mark for normal illumination is formed of a plurality of patterns, and the alignment inspection mark for modified illumination is formed of a plurality of patterns having a different number of lines from the alignment inspection mark for normal illumination. The method for manufacturing a semiconductor integrated circuit device according to claim 1, 2 or 3, wherein 10. The pitch of the plurality of patterns is P, P is the pattern width, and N is the numerical aperture of the projection lens.
A, wherein, when the wavelength of the exposure light beam is λ, and 0.5 ≦ n2 ≦ 0.9, P = W = n2 · λ / NA is set so as to satisfy: A manufacturing method of the semiconductor integrated circuit device according to the above. 11. In the inspection step, when inspecting the normal illumination alignment inspection mark pattern that has been exposed by exposing the normal illumination alignment inspection mark to the semiconductor wafer, the semiconductor wafer is first exposed to the normal illumination alignment inspection mark pattern. When the width of the pattern corresponding to the pattern located at the center of the deformed illumination alignment inspection mark in the deformed illumination alignment inspection mark pattern that has been revealed by The width of a pair of patterns respectively corresponding to the patterns located at both ends of the alignment inspection mark for normal illumination is measured, and the positional deviation of the alignment inspection mark pattern is inspected. 3. The method for manufacturing a semiconductor integrated circuit device according to 1. 12. In the inspection step, when inspecting the normal illumination alignment inspection mark pattern exposed and exposed by the normal illumination alignment mark, the semiconductor wafer is first exposed to the semiconductor wafer. When the width between patterns corresponding to a pair of patterns respectively positioned at both ends of the deformed illumination alignment inspection mark in the exposed deformed illumination alignment mark pattern is measured, the normal illumination alignment inspection mark pattern 11. The method according to claim 9, wherein a width of a pattern corresponding to a pattern located at the center of the alignment inspection mark for normal illumination is measured, and a positional shift of the alignment inspection mark pattern is inspected. Of manufacturing a semiconductor integrated circuit device. 13. The method of manufacturing a semiconductor integrated circuit device according to claim 1, wherein the modified illumination is annular illumination or quadrant illumination. 14. A photomask manufacturing step for manufacturing a plurality of photomasks each having a plurality of mutually associated practical mask patterns and alignment inspection marks formed thereon, and each of the practical masks of the respective photomasks on the same semiconductor wafer. An exposure step in which the pattern and the alignment inspection mark are simultaneously exposed; an exposure step in which the respective practical mask patterns and the alignment inspection mark exposed in the semiconductor wafer are exposed; and an exposure step that is first exposed after the exposure step. A method for manufacturing a semiconductor integrated circuit device, comprising: an inspection step of inspecting a positional deviation between an alignment inspection mark pattern and an alignment inspection mark pattern that is to be revealed later, wherein the projection exposure used in the exposure step The amount of displacement of the transfer pattern due to the coma of the lens optical system of the optical device is The method of manufacturing a semiconductor integrated circuit device which is characterized substantially be adjusted to be identical between the turn and the alignment inspection mark. 15. The arrangement pitch of a main pattern constituting the alignment inspection mark is set so as to be substantially the same as the arrangement pitch of the main pattern of the practical mask pattern. A manufacturing method of the semiconductor integrated circuit device according to the above. 16. A photomask manufacturing step for manufacturing a plurality of photomasks each having a plurality of mutually associated practical mask patterns and alignment inspection marks formed thereon, and each practical mask of each of said photomasks being formed on the same semiconductor wafer. An exposure step in which the pattern and the alignment inspection mark are simultaneously exposed; an exposure step in which the respective practical mask patterns and the alignment inspection mark exposed in the semiconductor wafer are exposed; and an exposure step that is first exposed after the exposure step. A method of manufacturing a semiconductor integrated circuit device, comprising: an alignment inspection mark pattern; and an inspection step of inspecting a positional deviation between the alignment inspection mark pattern that is to be revealed later, and the arrangement of the alignment inspection mark, Arranged in the area where the coma of the lens optical system of the projection optical system used in the exposure process is small. The method of manufacturing a semiconductor integrated circuit device which is characterized in that.