JPH0344918A - Patterning method for resist layer - Google Patents

Abstract

PURPOSE:To stabilize a residual film of an unexposed part and to obtain a satisfactory profile by exposing a resist layer on a semiconductor substrate, patterning it, and then adding steps of cleaning it with pure water on the way of developing and drying. CONSTITUTION:After a substrate is covered with novolac positive resist as an electron beam lithography resist, a micro pattern is written with an electron beam. Then, it is dipped in solution diluted with pure water from alkali developer to be processed, and then cleaned with water and dried. Thus, a resist film of an unexposed part of the pattern is not varied, and it is presumed that an etching resistant film is formed on an unexposed part of the resist by the cleaning with the pure water and the drying. Then, a stepwise developing step is conducted, and original washing and drying steps are eventually executed. Thus, a micro pattern having satisfactory profile can be formed.

Description

DETAILED DESCRIPTION OF THE INVENTION [Object of the Invention] (Industrial Application Field) The present invention relates to lithography technology, and is particularly suitable for forming fine patterns by electron beam exposure.

(Prior art) It is a well-known fact that as the degree of integration of semiconductor devices improves, the demand for fine pattern patterns increases.As a result, precise pattern dimensions are also required for resist patterns, and the degree of precision is increasing. is also getting bigger. By the way, the novolak-based positive resist (MP-2400, trade name of Cypre Co., Ltd.) layer patterned by electron beam exposure is as follows:
The film is developed with an alkaline developer for a certain period of time, for example, 30 seconds to 90 seconds. In order to fix this development time, an appropriate dose must be set so that the design dimensions are equal.
Currently, we have decided on 4. However, since the pattern dimensions vary slightly depending on the condition (ztb) of the electron beam exposure apparatus and the environment during development, such as temperature and humidity, the dimensions of the completed pattern also vary from predetermined values. Moreover, if the pattern is completely formed with resist remaining at the bottom of the pattern, additional development is required.

(Problem to be solved by the invention) Novolac positive resist (MP-2400 Sibley)
FIG. 2 shows the percentage of resist remaining film obtained by performing a continuous development process using an alkaline developer such as choline after exposure to an electron beam. This resist residual film rate refers to the rate at which the applied positive resist layer is pre-baked (Pre-baked).
bake) L, the film thickness at the end is 1OO1 exposure.
Let the film thickness of this positive resist layer at the end of development be α/
Express as 100 x 100.

In Figure 2, the vertical axis shows the resist remaining film rate and the horizontal axis shows the alkaline development time (seconds), and the relationship between the two is shown as the result of an investigation with an initial value of about 1 μm. A typical positive resist pattern has a thickness of 0.5 μm to 0.8 μm. By the way, in electron beam exposure, 0.1 μm to 0.5 μm
To form dimensions smaller than μm, a dilute alkaline developer is used for 60 to 90 seconds, sometimes 120 seconds to
A rather long development process is performed for 240 seconds. For this reason, as can be seen in FIG. 1, the remaining film rate after 90 seconds is poor, and the remaining film becomes 0.65 μm after about 240 seconds.

Furthermore, in Fig. 1, the vertical axis shows the aperture size unit μm, and the horizontal axis shows the dose m (nc/c+n), which shows the aperture size after development, that is, the resist dimension conversion difference, with respect to the dose during exposure. did. That is, when a novolac positive resist pattern formed under the same electron beam exposure conditions is developed in 90 seconds and 150 seconds, there will be a difference in the opening dimensions if the dose is obviously large. In other words, 0 in 60 seconds
．． The opening size becomes large by 15 μm to 0.2 μm. The results of this investigation revealed that the dimensions significantly fluctuate due to a slight increase in development time. When a novolak positive resist for electron beam is developed with a calkaline developer, the remaining film rate in the unexposed area where additional development is required for a certain period of time may deteriorate due to changes in the conditions of the electron beam exposure equipment, etc., and the aperture size may deteriorate. becomes large, making it difficult to form a pattern equal to the design dimensions with the target resist remaining film rate.

The present invention was made under these circumstances, and in particular,
The purpose is to suppress film loss in unexposed areas and form a fine pattern with a good profile.

[Structure of the Invention (Means for Solving the Problems) Stage development in which after exposing and patterning a resist layer covering a semiconductor substrate, a rinsing process and a drying process are performed one or more times during the development process, and then development is performed. The resist layer patterning method of the present invention is characterized by stabilizing the remaining film in the unexposed areas through the process.

(Function) Electron beam exposure using novolak positive resist
In the development process, after processing with an alkaline developer, the developer is washed away with pure water and a drying process is performed. However, in the present invention, a novolak positive resist drawn by an electron beam is immersed in a diluted alkaline developer for a minimum of the same amount of time, and then a step of washing with pure water and a drying step are added. It is clear that the novolac-based positive resist pattern that has undergone such processing steps is more advantageous than the conventional method because the resist film in the unexposed areas hardly changes. The reason for this is that it is presumed that an etching-resistant film is formed in the unexposed portions of the resist through this washing step with pure water and the drying step, so it is thought that it is not necessary to completely remove water in the drying step. That is, the vertical axis represents the resist residual film (unit: μm), and the horizontal axis represents the alkaline development time (
In addition, Figure 3 shows the difference in electron beam resist dimension conversion, with the vertical axis representing the aperture size (in μm) and the horizontal axis representing the dose (in nc/cm). As mentioned above, according to the method of the present invention, there is almost no change in the opening size, and the present invention was completed based on this fact.Moreover, the method of the present invention has been completed based on this fact. This is the basis for limiting what can be done.

(Example) An example according to the present invention will be described with reference to FIGS. 2 and 3. The resist layer covering the semiconductor substrate described in the present invention may be applied to an insulating layer normally deposited on the semiconductor substrate, or an active layer or passive layer formed by introducing and diffusing impurities into the semiconductor substrate. This includes the case of laminating electrically connected electrodes and wiring layers, and further includes interlayer insulating films, electrodes, and wiring layers for so-called multilayer wiring elements.

After coating a silicon semiconductor substrate with a novolac-based positive resist MP-2400 (product name of Sibley) as a resist for electron beam drawing, a fine pattern with an opening diameter of 0.2 μm to 0.3 μm is formed using an electron beam device. Drawing was performed using an electron amount of about 1 nanoclone/cm. Next, use an alkaline resist specially designed for novolac positive resists.
The liquid was developed using a solution diluted with pure water. As for the development conditions, after the first development for 15 seconds, washing with pure water (running water) was performed for 30 seconds and a drying step was performed. This drying step may be performed by spin drying for 10 seconds or by spraying clean gas for 10 seconds, and even if the moisture is not completely removed, it is sufficient to form an etching-resistant film on the unexposed areas. At this time, as conditions for forming such a film, a drying step is inevitable, and washing with pure water is also a necessary condition. Subsequent development processing at 0°1.5, 30.45, 75,
An additional 100 seconds was added, and finally the original washing process and drying process were performed. 0 seconds in this additional development is a means taken to obtain comparative data.

Furthermore, patterning was performed using a resist layer of the same quality, a developer of the same concentration, and the same exposure conditions, but the first development was performed for 90 seconds, and after washing with pure water for 30 seconds, a drying process was performed. , furthermore, carry out additional development for 30 seconds,
In the end, the original washing and drying steps were added. As described above, in the present invention, a novolak positive resist is subjected to stepwise development, and a washing step with pure water and a drying step are carried out in the middle of the development.

As a result, unprecedented good results were obtained as shown in FIGS. 2 and 3, which will be explained below.

Figure 2 shows the residual film difference due to stepwise development of a novolac positive resist using electron beam exposure, with the vertical axis representing the novolac positive resist residual film (in μm) and the horizontal axis representing the alkaline development time (in seconds). It is shown. In this figure, the square marks indicate continuous development, and the circles indicate the results of t5 seconds development + pure water washing step and drying step + X seconds additional development. As is clear from this, it can be confirmed that in the development process indicated by the circle for one continuous phenomenon, the resist remaining film in the unexposed area after a total of 90 seconds hardly fluctuates.

In addition, in Figure 3, the dimensional conversion difference that occurs when a novolak positive resist is exposed and developed with an electron beam is shown, with the vertical axis representing the opening dimension (unit: μm) and the horizontal axis representing the dose amount (unit: nc/cm).
cm). The square marks show the results of 90 seconds development, and the circles show the results of 90 seconds development, rinsing step and drying step + 60 seconds additional development. That is, it is clear that the division of the openings does not change even if additional development is performed for 60 seconds.

[Effects of the Invention] As described above, in the present invention, when developing a novolak positive resist, the bottom part of the pattern drawn with a certain amount of dose is completely developed by additional development, and the scum is also eliminated by the cleaning process. Since it is considered that an etching-resistant film is formed in the unexposed portions by the drying process after removal, a good profile can be completed without any variation, which is extremely effective in mass production.

[Brief explanation of drawings]

FIG. 1 is a diagram showing the characteristics of a resist pattern obtained by the conventional method, and FIGS. 2 and 3 are diagrams showing the characteristics of the resist pattern obtained by the method of the present invention.

Claims (1)

[Claims] After exposing and patterning the resist layer that covers the semiconductor substrate, a rinsing process and a drying process are performed at least once during the development process, and then the development process is performed to stabilize the remaining film in the unexposed areas through the development process. Characteristic resist layer patterning method.