JP2865685B2 - Method for manufacturing semiconductor device - Google Patents

Description

Description: BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for manufacturing a semiconductor device having a step of transferring a fine pattern formed on a mask substrate onto a semiconductor substrate by a projection exposure apparatus.

[Conventional technology]

A reduction projection exposure apparatus that illuminates a mask (hereinafter, referred to as a reticle) on which a circuit pattern such as a semiconductor element is drawn by an illumination optical system and transfers the circuit pattern onto a wafer includes:
It is required that a transferable pattern can be miniaturized.
As one method of transferring a fine pattern near the resolution limit of a reduction projection exposure apparatus, a method of giving a phase difference to illumination light transmitted through an opening adjacent to the reticle through a fine light-shielding portion is disclosed in No. 173744. Further, a conventional reticle for providing a phase difference is described in IEE, Transaction on Electron Device 29 (1982), p. 1282 and below (IEEE, Tra
ns.on Electron Devices, Vol.ED-29 No.12 (1982) P.
1282 ~) by Mark Dee Levenson (Mar
cDLevenson) and others on “Improving Resolution in Ph
The reticle proposed in the above-mentioned document gives a 180 ° phase difference to the illumination light passing through the opening adjacent to the light-shielding part. This is suitable for improving the resolving power of periodically arranged patterns obtained by the projection exposure apparatus.

However, the above method is only effective when adjacent openings are separated from each other. If the adjacent openings are adjacent at a certain portion, no phase difference can be given because the openings are continuous. Therefore, it is impossible to manufacture a semiconductor element composed of a fine pattern connected at a portion where an adjacent opening exists, using a mask in which the phase difference is introduced.

[Problems to be solved by the invention]

In an actual semiconductor element circuit pattern, for example, adjacent linear opening patterns may be connected at an end.
In the conventional method of giving a phase difference, the phase of the illumination light changes abruptly at this connection portion, so that the pattern is resolved at the portion where the linear pattern is adjacent, but the pattern is also separated at the connection portion. Therefore, there is a problem that a pattern having a desired shape cannot be obtained on the wafer.

An object of the present invention is to provide a method of manufacturing a semiconductor device capable of accurately transferring even a fine pattern in which adjacent opening patterns are connected at a certain portion.

[Means for solving the problem]

The above object is achieved by a configuration in which the phase difference of the illumination light transmitted through the opening pattern on the mask substrate is changed continuously or stepwise inside the opening pattern. There are a method of providing a phase difference of illumination light, a method of providing a transparent thin film, a method of etching a mask substrate, and a method of changing a refractive index by injecting impurities into a mask substrate.

[Action]

By providing a 180 ° phase difference (inverting the phase) to the light transmitted through the adjacent opening on the reticle, the resolution is improved. Further, the portion where the adjacent openings are connected at one end can be formed without separating the pattern by changing the phase continuously or stepwise in the vicinity of the connection.

〔Example〕

Hereinafter, an embodiment of the present invention will be described with reference to FIG. First
The figure shows a U-shaped opening pattern on a reticle to which the present invention is applied. Illumination light transmitted through the aperture pattern 1 has different phase differences in five regions 1-1 to 1-5 as shown in the figure. That is, in the region 1-1, the phase difference is 0 °, and in the region 1-2,
The phase difference is 45 °, the region 1-3 is 90 °, the region 1-4 is 135 °, and the region 1-5 is 180 °. A method for changing the phase difference of the illumination light stepwise in one aperture pattern will be described below.

FIG. 7 is a view showing a cross-section taken along a chain line along the pattern of the opening pattern 1 shown in FIG. Opening portions 10-1 to 10-10 where light shielding film 8 on reticle substrate 7 is removed
-5 corresponds to the five regions 1-1 to 1-5 shown in FIG. The phase shift layer 9 made of SiO ₂ is provided in the region 10-5 with a thickness of 0.38 μm. The value of the thickness d is n, where n is the refractive index of SiO ₂ and λ is the wavelength of the illumination light. Therefore, the phase of the illuminating light transmitted through the region 10-5 is shifted by 180 ° from the phase of the illuminating light transmitted through the region 10-1 where no phase shift layer is provided. I have. The thickness of the phase shift layer 9 provided in the regions 10-2, 10-3, and 10-4 is 1/4, 1/2 of the thickness d of the phase shift layer 9 provided in the region 10-5. , 3/4.

A method for partially changing the thickness of the phase shift layer 9 will be described below. First, an SiO ₂ layer is provided on the entire surface of the opening by vapor deposition or the like. Next, a photoresist is applied and the photoresist on the regions 10-1 to 10-4 is removed by a usual lithography (exposure, development) method. Using the remaining photoresist as a mask
The SiO ₂ is etched away by a quarter thickness of 0.095 μm. Apply photoresist again, this time in areas 10-1 to 10
The photoresist on the -3 removed, only the SiO ₂ remaining photoresist as a mask again 0.095μm removed. SiO ₂
By changing the region from which is removed, a stepwise phase shift layer 9 can be finally formed as shown in FIG.

Conversely, a method of gradually increasing the thickness of the phase shift layer 9 while repeating deposition and etching of SiO ₂ is also conceivable. That is, first, 0.095 of SiO _{2 is applied} to the entire surface of the opening.
μm, then apply photoresist, expose, develop,
Eighth using the usual lithography technique of etching
The phase shift layer 9-1 shown in the figure is formed. Again Si
The O ₂ only deposition 0.095 m, to form a phase shift layer 9-2 in the same manner. Thereafter, layers 9-3 and 9-4 shown in FIG. 8 are formed by repeating deposition and etching. As a result, the same mask as that shown in FIG. 7 can be formed. The layers 9-1 to 9-4 do not necessarily need to be made of the same transparent material, and the refractive index and the thickness of the material can be appropriately selected so that a phase difference of 45 ° occurs. The phase shift layer 9 is formed of a resist, for example, PMMA
The resist for electron beam lithography itself may be used. In this case, the steps of PMMA application, drawing, development, and curing may be repeated, and the etching step is unnecessary. Here, each layer is formed with a phase difference of 45 ° as one step.
The phase difference is not limited as long as the boundary between the layers does not transfer as a dark line portion.

As means for giving a different phase difference to the mask opening, the mask substrate itself may be etched to a predetermined depth as shown in FIG. Assuming that the refractive index of the mask substrate is n, the phase of the illumination light is shifted by 180 ° if the etching is performed to a depth of λ / {2 (n−1)}. Changing the etching depth changes the phase difference.

As another means for providing a different phase difference to the mask opening, ions such as Pb, P, and B may be implanted into the mask substrate to locally change the refractive index. In this case, the thickness of the mask substrate does not change but the refractive index changes, so that the phase changes correspondingly.

When a reflective mask is used, the phase can be changed by providing a concave portion or a convex portion on a predetermined reflecting surface and giving a difference in optical path length to reflected illumination light. The phase shift amount can be controlled by changing the depth of the concave portion or the height of the convex portion.

FIG. 10 shows a projection exposure apparatus according to an embodiment of the present invention.
The mask 13 is a reticle having a pattern 1 shown in FIG. 1, and is illuminated by an illumination optical system including an illumination light source 11 and a condenser lens 21. The pattern on the mask is transferred onto a wafer 16 through a projection lens 15. . The wafer 16 is mounted on an XY stage 17, and the position of the wafer 16 is measured by a laser length measuring device 18, and the XY stage 17 is fed in a fixed size to repeat pattern exposure. The reticle pattern 1 shown in FIG. 1 is converted to a reduced projection exposure apparatus (wavelength λ = 0.
FIG. 2 shows the calculated values of the light intensity distribution obtained when the image is transferred onto the wafer using 365 μm and numerical aperture NA = 0.4). The minimum dimension of reticle pattern 1 is 0.3 μm in terms of the dimension on the wafer. As can be seen from the contour line 2 showing the light intensity distribution, a light intensity distribution having a desired shape is obtained although there is a difference in light intensity level.

On the other hand, if a pattern having the same shape as the pattern 1 shown in FIG. 1 is to be realized by a conventional method of giving a phase difference, the pattern becomes, for example, a pattern 3 shown in FIG. That is, the pattern 3 is divided into two regions 3-1 and 3-2, and the phase of the illumination light in each region is set to 0 ° and 180 °. The light intensity distribution on the wafer at this time is as shown by a contour line 4 in FIG. Since the phase difference is given,
Adjacent linear patterns are separated and resolved. However, since the portions to be connected by the two patterns are also separated, a light intensity distribution of a desired shape cannot be obtained. In addition, as shown in FIG. 5, when an opening pattern 5 having the same shape as that of FIG. 1 is used without giving a phase difference, the light intensity distribution on the wafer becomes like contour lines 6 shown in FIG. The pattern is not resolved at all. According to the embodiment shown in FIG. 1, a fine pattern can be transferred without causing the above-mentioned poor resolution.

FIG. 11 is an example of a wiring pattern for manufacturing a semiconductor device using the mask of the present invention. This pattern includes a light-shielding pattern 19 having a fine dimension inside a wide opening 21.
Is what exists. The aperture area 20 has an illumination light phase of 18
The aperture region 22 is a region where the phase of the illumination light is shifted by 90 °. The other opening area 21 is an area where the amount of phase shift is zero. The linear light-shielding pattern 19
The phase difference between the illumination light on both sides is 180 °, so that the light is reliably transferred. The openings 20, 21, and 22 have a phase difference of 90 ° on a boundary line contacting each other, and are sufficiently smaller than 180 °, so that they are transferred without separation. The region 22 in which the phase of the illumination light shifts by 90 ° is disposed at the tip of the light-shielding pattern 19 having a fine dimension, but the shape and size thereof are not particularly limited, and the opening region 20 and the opening region 21 are in contact with each other. What is necessary is just to arrange so that it may not. Also, here, a region for providing a phase difference of 90 ° is provided by dividing the phase difference of 180 ° into two stages, but a region for providing a phase difference of 60 °, 120 ° or the like is provided in three or more stages. Good.

FIG. 12 is an example in which an odd number of minute-sized light-shielding patterns 19 exist inside the wide opening 21. When the phase shift amount of the illumination light transmitted through the opening 21 is 0, the opening 20 is an area where the phase of the illumination light is shifted by 180 °, and the opening 22 is an area where the phase of the illumination light is shifted by 90 °. It is. The linear light-shielding pattern 19 has a phase difference of 18
Since it is 0 °, it is reliably transferred. Also, openings 20 and 22
Or the openings 21 and 22 have a phase difference of 90 ° at their boundary and are sufficiently smaller than 180 ° so that they are not separated from each other. The phase difference of 180 ° may be divided into three or more stages, and regions giving a phase difference of 60 °, 120 °, etc. may be arranged.

〔The invention's effect〕

According to the present invention, since the phase difference of the illumination light transmitted through the opening is gradually changed stepwise from 0 ° to around 180 °, the light intensity becomes zero during this period and the pattern is separated. I will not do it. Therefore, even a fine pattern in which adjacent opening patterns are partially connected to each other can be transferred by using the projection exposure apparatus and the mask that provides a phase difference to the illumination light according to the present invention. is there. Further, there is an effect that a semiconductor element including the fine pattern as described above can be manufactured.

[Brief description of the drawings]

FIG. 1 is a view showing an opening pattern in which the present invention is implemented on a reticle, FIG. 2 is a view showing a light intensity distribution when the pattern shown in FIG. 1 is transferred onto a wafer, and FIG. FIG. 4 is a diagram showing an opening pattern formed by a method of giving a phase difference, FIG. 4 is a diagram showing a light intensity distribution obtained from the opening pattern shown in FIG. 3, FIG. FIG. 6 is a diagram showing a light intensity distribution obtained from the opening pattern shown in FIG. 5, and FIGS. 7, 8, and 9 are diagrams each showing a cross section along the opening pattern shown in FIG. FIG. 10 is a diagram showing a reduced projection exposure apparatus embodying the present invention, and FIGS. 11 and 12 are diagrams showing an embodiment of the present invention in the case where a fine light shielding pattern exists inside a wide opening. It is. Description of reference numerals 1-1, 1-2, 1-3, 1-4, 1-5... So that the phase shift amounts of the illumination light become 0 °, 45 °, 90 °, 135 °, and 180 °, respectively. , A contour line representing a light intensity distribution obtained when an opening pattern composed of the opening portions 1-1 to 1-5 is transferred onto a wafer, 7 a reticle substrate, 8 a light-shielding film , 9 ... Phase shift layer, 11 ... Illumination light source, 12 ... Condenser lens, 13 ... Mask giving phase difference, 15 ...
A reduction projection lens, 16 wafers, 20 apertures configured so that the phase shift of the illumination light is 180 °, 21 apertures
... An opening configured so that the phase shift of the illumination light is 0 °, 22... An opening configured such that the phase shift of the illumination light is 90 °.

──────────────────────────────────────────────────続 き Continuing on the front page (72) Inventor Shigeo Moriyama 1-280 Higashi-Koigabo, Kokubunji-shi, Tokyo Inside the Central Research Laboratory, Hitachi, Ltd. (56) References JP-A-61-292643 (JP, A) JP-A-62-189468 (JP, A) JP-A-57-62052 (JP, A) (58) Fields studied (Int .Cl. ⁶ , DB name) G03F 1/00-1/16

Claims (5)

(57) [Claims] A first light-transmitting region; a second light-transmitting region in which a phase of the transmitted light is substantially inverted from a phase of light transmitted through the first light-transmitting region; and a first light-transmitting region. And between the first light transmitting region and the second light transmitting region, and is in contact with the first light transmitting region and the second light transmitting region without passing through the light shielding portion, and the phase of the transmitted light is transmitted through the first light transmitting region. Irradiating light on a mask substrate having a third light transmission region such that the phase becomes the phase between the transmitted light and the light transmitted through the second light transmission region, and the first, second, and A method for manufacturing a semiconductor device, comprising a step of forming a pattern by transferring one opening pattern formed of a third light transmitting region. 2. The method according to claim 1, wherein the third light transmitting region is formed such that the phase of the transmitted light changes continuously. 3. The method according to claim 1, wherein the third light transmitting region is formed such that the phase of the transmitted light changes stepwise. . 4. A phase difference between a first light transmitting region and light passing through the first light transmitting region, which is adjacent to the first light transmitting region via a light shielding portion, and substantially equal to a light transmitted through the first light transmitting region.
A second light transmitting region that is 180 degrees, and both the first light transmitting region and the second light transmitting region are in contact with each other without passing through the light blocking portion, and the transmitted light passes through the first light transmitting region. Having a step of transmitting light to a mask substrate having a third light transmitting region having a phase between the transmitted light and the light transmitting through the second light transmitting region, and transferring the pattern of the semiconductor element onto the semiconductor substrate. A method for manufacturing a semiconductor device, comprising: 5. The semiconductor device according to claim 4, wherein the openings formed by the first, second, and third light transmitting regions form a U-shaped opening pattern. Method.