JPH0284742A - Pattern formation for charge transfer device - Google Patents

Abstract

PURPOSE:To obtain a forming method for a pattern tough against position aberration at a junction part of pattern subjected to divided exposure by a method wherein, when the patterns of a charge transfer device having a charge transfer part are combined and formed on a substrate while being subjected to divided exposure at the charge transfer part, the end-portion of the pattern subjected to the divided exposure is arranged in a region not storing treated charge. CONSTITUTION:When the patterns of a charge transfer device having a charge transfer part are combined and formed on a substrate 31 while being subjected to divided exposure at the charge transfer part, the end-portion D of the pattern subjected to divided exposure is arranged in a region not storing treated charge. For example, on the P-type silicon substrate 31, an oxide film 32, a nitride film 33 and an oxide film 34 are laminated in order, and a first electrode layer 35 is formed thereon; a positive type resist layer 37 is formed in order to pattern the first electrode layer 35; by applying a line D in figure to a junction line, the divided exposure of a pattern for opening a window of a charge transfer electrode of the charge transfer part is executed; then a mask is formed by developing the resist layer 37; by using said mask, the first electrode layer 35 is patterned on the electrode for charge transfer use, of the charge transfer part.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a method for forming a pattern of a charge transfer device by synthesizing and forming a pattern of a charge transfer device on a substrate while performing divided exposure.

[Summary of the invention]

In a pattern forming method for a charge transfer device that performs divided exposure and synthesizes each pattern, the present invention prevents manufacturing variations and increases the This method makes it possible to easily obtain a charge transfer device of the same size.

[Conventional technology]

In manufacturing a charge transfer device such as a COD line sensor, processing is performed using an exposure device in order to perform fine processing. FIG. 5 is a schematic diagram of a reduction projection optical system for exposure, in which light that has passed through the reticle lOO is projected onto the surface of the wafer 101 after being reduced to a ratio of 5:1.

By the way, if you are trying to get a line sensor with a large size 1, for example 201 or larger, 5:
In the reduction projection optical system of No. 1, the size of the reticle 100 is 10
Since the exposure area of the reticle 100 is not so thick at present, it is not easy to obtain a large-sized line sensor.

Therefore, a large-sized line sensor is formed using an exposure device with a same-magnification projection optical system.

[Problem to be solved by the invention]

However, the above-mentioned same-magnification projection optical system has problems such as deterioration of mask alignment accuracy and uneven exposure, which increases design constraints and increases bit-to-bit variation. linked.

On the other hand, a divisional exposure technique is known in which the exposure for each chip on a wafer is divided into multiple parts.
It is also described in the May 1986 issue, pages 132-134 (published by Nikkei McGraw-Hill).

However, simply exposing the film by dividing it leads to problems with seams at the ends of the divisions. That is, in the step-and-repeat method, misalignment of each pattern may occur, and it is necessary to minimize changes in the characteristics of the charge transfer device when misalignment occurs.

Therefore, the present invention provides the steps for manufacturing a charge transfer device.
It is an object of the present invention to provide a method of forming a pattern that is resistant to positional displacement at the joints of patterns exposed separately.

[Means to solve the problem]

In order to achieve the above object, a pattern forming method for a charge transfer/device according to the present invention involves forming a pattern of a charge transfer device having a charge transfer section on a substrate by combining and exposing the charge transfer section in sections. At this time, the end portions of the pattern to be subjected to divided exposure are provided in the handling charge non-accumulation region.

Here, the charge transfer device is a device having a charge transfer section using a charge-coupled device (COD), and may also be provided with a photoelectric conversion section, regardless of whether it is -dimensional or two-dimensional. Further, it is preferable that the charge transfer device has at least a continuous pattern portion.The shape of the end portions of the pattern is not limited to a linear shape, and any shape may be used as long as the end portions can be overlapped with each other. Further, the handled charge non-storage region can be a transfer section that is on the shallow side of the stepped potential of the charge transfer section.

Further, the pattern subjected to the divided exposure can be, for example, a pattern of a plurality of electrodes for a charge transfer device formed in a plurality of charge transfer portions. In that case, the pattern corresponding to the charge transfer electrode at the end can be made to have the same size as the charge transfer electrode.

[Effect]

In the handled charge non-storage region, no charge is accumulated even when the region is driven for charge transfer, and charges are accumulated in the adjacent storage section. Therefore, even if the size of the handled charge non-accumulation region itself changes, it does not affect the amount of transferred charge, and therefore even if positional deviation occurs during exposure. However, the charge transfer characteristics do not deteriorate.

〔Example〕

Preferred embodiments of the present invention will be described with reference to the drawings.

The charge transfer device of this embodiment is a COD line sensor. First, the reticle used in the pattern forming method will be explained. As shown in FIG. 1, the A pattern portion 2.
B pattern part 3. A C pattern 7 portion 4 is provided as a pattern on the reticle l.

The A pattern part 2 has a pattern corresponding to the charge transfer part of the COD line sensor, and is formed continuously in the X direction in the figure, which is the longitudinal direction, so that a large-sized line sensor can be obtained by dividing exposure. has been done. That is, one end 2e of the A pattern section 2 is connected to the other end 2e.
Even if it is formed adjacent to e', consistency will not be lost. As will be described later, when the handle 1 is applied to the first electrode layer, 〃 of the charge transfer electrode is located at one end 2e, and % of the charge transfer electrode is located at the other end 2e'. It can be made to exist in

The B pattern section 3 has a pattern corresponding to the input side of the CCD line sensor, and one side perpendicular to the X direction in the figure is adjacent to the end 2e or end 2e' of the A pattern section 2.

Further, the C putter 7 section 4 has a pattern corresponding to the output side of the CCD line sensor, and one of the sides perpendicular to the X direction in the figure is the end 2e' of the A pattern section 2 or the end 2
Adjacent to e.

In addition, as a reticle, each of the above pattern parts 2°3.4
may be a separate reticle.

Next, the reticle 1 is set in a step type exposure device and used to expose a desired wafer. The exposure apparatus used has a reduction projection optical system (for example, 5:1) in order to improve alignment accuracy and prevent variations from bit to bit.

Then, divided exposure is performed using a step method.
For example, one COD line sensor is divided into five exposures. FIG. 2 shows the pattern on the wafer 5 that has been subjected to the divided exposure.
The pattern portion is formed by performing divided exposure three times and the C pattern portion a total of five times. Generally, in a CCD line sensor, the size of the charge transfer section occupies a large proportion and is a continuous pattern, so in this example, the A pattern section is exposed three times to obtain the required length. A line sensor can be obtained using a reduced projection optical system.

Incidentally, it goes without saying that the method of forming a baguette according to the present invention is not limited to three times.

Next, with reference to FIGS. 3a to 3r and FIGS. 4a and 4, the joint portion between the A pattern portions will be specifically described.

First, as shown in FIG. 3a, a P-type silicon substrate 31
An oxide film 32 on top. Nitride film 33. Oxide films 34 are sequentially stacked, and a first electrode layer 35 is formed on top of the oxide films 34 . Since the nitride film 33 is formed between the oxidized Wis 32 and 34, a MONO3 (metal-oxide film-nitride film-monoxide film-semiconductor) structure is achieved, and so-called pinhole phenomena are prevented. Impurities are introduced into the main surface of the P-type silicon substrate 31 in contact with the oxide film 32 to form an N-type high concentration impurity region 36.

Then, a positive resist M37 is formed to pattern the first electrode layer 35.

After forming the resist layer 37, the wafer is placed in a step-and-repeat type exposure apparatus.

Here, divisional exposure will be performed, and the exposure shown in FIGS. 3a and 4a is first performed on the joint portion. That is, in FIGS. 3a and 4a, if line D is the seam line, light is applied to the resist layer 37 in a pattern of a predetermined width W0 corresponding to the pattern of opening the charge transfer electrode of the charge transfer section. At this time, at the above DI, the width w6/2 is half of the original pattern width W.
A window is opened with the pattern width as . In addition, Figure 3a
In the middle, the shaded area of the layer 38 indicates an area not irradiated with light, and the dotted area of the resist layer 37 indicates an area irradiated with light.

Subsequently, selective exposure using the reticle of the same A pattern portion is performed so as to be continuous along the seam line. As shown in Fig. 3 and Fig. 4, the pattern part with the already formed pattern width W0/2 is continuously changed to the half width W6/2 so that the pattern part with the pattern width W0/2 becomes the original pattern width W0. This is done by supplementing the pattern part. For this reason, the electrodes of the charge transfer section are exposed to light at the seam without the gap between them increasing, and as will be described later, the pattern at the seam where two half widths w0/2 are combined is Since the area is treated as a charge non-accumulation area due to the potential, even if the alignment at the joint is misaligned, variations in the charge transfer characteristics can be minimized.

After such divisional exposure is performed, as shown in FIG. 3C, the resist layer 37 is developed.

Obtain a mass for selectively removing the first electrode [35.

Next, using the mask, for example, by RTE method,
The first electrode layer 35 is patterned into a charge transfer electrode of the charge transfer section. Then, oxidation is performed, and P-type impurities (for example, boron, etc.) are implanted into the N-type high concentration impurity region 36 through the patterned first electrode layer 35 and the self-alignment line. By this ion implantation, the silicon base 131 in the region between the first electrode layers 35 has N
The + type is canceled out and changed to the N- type, and the potential of the N° type low concentration impurity region 39 becomes shallow.

Subsequently, the second electrode layer 40 is formed, and as shown in FIGS. 3d and 3e, a resist layer 41 is formed on the second electrode layer 40, and then divided exposure is performed again. will be held. The divided exposure for patterning the second electrode layer 40 is similar to the step of patterning the first electrode layer, as shown in FIG. 3d, one of the seam lines is exposed first, The other side of the seam line is then exposed, as shown in Figure 3e. In the figure, for example, the pattern width W
The figure shows a case where exposure is performed at step 1. At this time, when exposing the second electrode N40, since the pattern of the first electrode layer already exists on the seam line, a pattern that straddles the seam is not formed. Therefore, variations in the pattern due to misalignment at the seam do not occur.

These resist layers 41 are subjected to divided exposure and developed to obtain a mask for the second electrode layer 40. Etching is performed using the mask to obtain a charge transfer section in which charge transfer electrodes consisting of first and second electrode layers 35 and 40 are provided as shown in FIG. 3r. In this charge transfer section, the lower part of the first electrode layer 35 becomes a storage part with a deep potential, and the lower part of the second electrode N40 becomes a transfer part with a shallow potential. Thereafter, the necessary wiring, formation of the glabella insulating layer, etc. are performed to complete the device.

As described above, in the pattern forming method of the charge transfer device of this embodiment, N
The ゛-type low concentration impurity region 39 is used as a seam during divisional exposure, and at the same time, the N°-type high concentration impurity region 36, which is a deep potential storage region, is not located on the seam. Therefore, even if a positional shift occurs, the N0 type high concentration impurity region 36 where handling charges are accumulated
In this case, the electrode length does not change, so the amount of charge handled does not change and stable transfer is realized. In addition, the positional shift is
In this embodiment, stable transfer can be realized, including not only positional deviation in the charge transfer direction but also in other directions. In addition, as in this embodiment, the pattern that is the handling charge non-accumulation region over the seam can be made by combining a pattern portion with half the original width w0/2, and in this case, the This means that it can be patterned into a continuous shape similar to the pattern of the part.

In the above embodiment, the structure of the charge transfer section is M
Although the ONO3 structure is used, an MO3 structure may also be used. Furthermore, although the resist layers 37 and 41 have been described as being of positive type, they can also be of negative type.

Also, I explained about the seams between A pattern parts,
In a COD line sensor, the joint between the B pattern part and the A pattern part on the input side, and the C on the output side.
Similarly, the joint between the pattern part and the A pattern part can be formed using the handled charge non-accumulation region.

Further, the present invention is not limited to the above-described embodiments, and various modifications can be made without departing from the gist thereof.

〔Effect of the invention〕

In the pattern forming method of the charge transfer device of the present invention, the handled charge non-storage region is used as a seam of divided exposure, and no seam is formed in the region where the handled charge is accumulated and transferred. Therefore, even if the mask is misaligned during divided exposure, there is no change in the amount of charge handled, and stable transfer characteristics can be obtained. Further, by such divisional exposure, it is possible to employ a reduction projection optical system, and problems at the time of equal-magnification projection exposure can be solved at the same time.

[Brief explanation of the drawing]

FIG. 1 is a schematic diagram showing an example of a reticle used in the pattern forming method of a charge transfer device of the present invention, and FIG. 2 is a schematic diagram showing an example of a pattern on a wafer exposed by the pattern forming method of a charge transfer device of the present invention. The schematic diagrams, FIGS. 3a to 3f, are process sectional views for explaining an example of the pattern forming method of the charge transfer device of the present invention in the order of the steps, and FIGS. FIG. 5 is a plan view for explaining a part of the above-mentioned example of the pattern forming method for a charge transfer device according to the steps, and FIG. 1 electrode layer 36...N° type high concentration impurity region 39...N' type low concentration impurity region 40...Second electrode layer 37.41...Resist layer Patent applicant Sony Corporation Company Representative Patent Attorney Akira Kobu (2 others) l...Reticle 2...A pattern section 3...B pattern section 4...C pattern section 5...Wafer reticle - # tsu Fig. 1 Fig. 3a Fig. 3b Example of wafer mounting circuit width Fig. 2 Fig. 3C Fig. 3d Fig. 3e Fig. 3f Fig. 4a Fig. 4b

Claims (1)

[Claims] When a pattern of a charge transfer device having a charge transfer section is synthesized and formed on a substrate while being subjected to divided exposure using the charge transfer section, the end portion of the pattern to be subjected to divided exposure is provided in a handling charge non-accumulation region. A method for forming a pattern of a charge transfer device characterized by: