JPS647652B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a dry development method, and more particularly to a dry development method for obtaining high resolution of electron beam/X-ray resists developed for microfabrication.

As the miniaturization of devices such as LSI and Josephson micro bridges progresses, high-resolution electron beam and
Linear and deep UV resist materials have been developed. Among these, so-called positive resists, in which the molecular weight of the areas irradiated with various radiation sources decreases, take advantage of the fact that the difference in molecular weight between the areas irradiated with electron beams and the areas that are not irradiated causes the dissolution rate to change greatly depending on an appropriate organic solvent, etc. Generally, patterning is performed by

Conventional developing methods using such solvents have major drawbacks when developing fine patterns, such as swelling of the resist polymer in the solvent and peeling off of fine patterns. Therefore, a few so-called dry development methods that do not use solvents have been developed, but
Since both of these methods are carried out in a dry etching device, which generally involves heat, the resist pattern may sag, and is therefore not necessarily optimal as a dry developing method for fine patterns.

The object of the present invention is to eliminate such drawbacks, and more specifically, to provide a dry development method that can form fine fine patterns without deforming the pattern.

Therefore, the dry development method according to the present invention is characterized in that after exposing an electron beam resist coated on a substrate in a pattern with an electron beam, the entire surface is irradiated with ultraviolet rays to remove the portions not exposed to the electron beam. be.

According to the present invention, unlike development using conventional solvents, there is no pattern swelling phenomenon during development, and development with good pattern accuracy can be realized. Further, although heat is used in conventional dry development, according to the present invention, since no heat is used, it is possible to develop a resist pattern with ultimate high precision.

The present invention will be explained in more detail.

According to the dry development method according to the present invention, an electron beam resist coated on a substrate is exposed in a pattern with an electron beam. The electron beam resists used in this dry development method are mainly positive resists, such as positive resists for electron beams, X-rays, and deep ultraviolet rays. PMMA; trade name Elvacite 2041), polytetrafluoropropyl methacrylate (trade name
A methacrylate resin such as FPM) can be used.

When these methacrylate resins are irradiated with electron beams, their molecular weights change. Figure 1 shows the change in molecular weight in response to electron irradiation. In FIG. 1, a, b, and c represent changes in the molecular weight of the resist material shown below.

a...Polymethyl methacrylate b...Polyhexafluoromethacrylate c...Polytetrafluoropropyl methacrylate As is clear from Figure 1, the main chain of each resist is mainly cut in the area where the irradiation dose is low. When the molecular weight decreases and the amount of electron beam irradiation is further increased, a crosslinking reaction occurs and the molecular weight increases and becomes insoluble in the solvent. However, unlike the cross-linked state of so-called negative resists, this cross-linked state is subject to side reactions such as side chain cleavage, and does not necessarily maintain the original monomer structure, so the reactivity and physical properties are lower than those of the non-electron beam irradiated area. are different.

After electron beam irradiation, the entire surface is irradiated with ultraviolet rays to remove unexposed or low-exposed areas.

When an unexposed area or a low exposure area is irradiated with ultraviolet rays, the methacrylate resist undergoes a decomposition reaction due to the absorption of ultraviolet rays by the carbonyl group in the side chain, and its molecular weight is reduced. Since the lower molecular weight has a lower boiling point, some of it evaporates even at room temperature, the volume (film thickness) decreases, and it becomes polarized and removed.

Next, examples of the present invention will be described.

Comparative Example 1 A polyhexafluoromethacrylate resist is uniformly applied onto a 30 mmφ silicon substrate by spinto coating. The thickness at this time was 3500 Å. Heat this in an electric furnace at 180℃ for 30 minutes. This was irradiated with an electron beam exposure device at different locations and at different irradiation times from 1 × 10 ^-6 coulombs/cm ² to 5 × 10 ^-3 coulombs/cm ² (one irradiation area was 50 μm × 500 μm,
Irradiation conditions were acceleration voltage 20KV, 1×10 ^-9 A).

Thereafter, the film was developed with methyl isobutyl ketone for about 20 seconds, and a sensitivity curve showing the residual film of the resist against the exposure amount as shown in FIG. 2 was obtained. In this way, polyhexafluoromethacrylate can be used in a normal developer solution (isopropanol: methyl isobutyl ketone =
150:1), the exposed area becomes a positive resist whose film wears off, but if a strong solvent such as methyl isobutyl ketone is used, not only the low electron beam area such as the positive area but also the unexposed area will be developed and dissolved. This results in a so-called negative type image in which only the highly irradiated area remains.

Example 1 A sample exposed to electron beam in the same manner as in Comparative Example 1 was exposed to a deuterium lamp (D ₂ lamp, 200W,
1.5A) (distance from lamp to board 15cm)
As a result, as the irradiation time increases, the amount of film loss increases in the non-electron beam irradiated area and the low electron beam irradiation area, and after about 100 minutes of irradiation, the film in the non-electron beam irradiated area and the low electron beam irradiation area completely disappears, and compared to Comparative Example 1. Similarly, only the part of the electron beam high irradiation area remained. Negative development was possible by irradiating with a deuterium lamp in this manner.

Similarly, a sample exposed to electron beam in the same manner as Comparative Example 1 was exposed to a deuterium lamp ( _D2 lamp,
200W, 1.5A) was irradiated at varying times. The sensitivity curve is shown in FIG. In FIG. 3, d, e, f, and g indicate sensitivity curves representing the amount of resist remaining film relative to the exposure amount for different irradiation times.

d... Deuterium lamp No irradiation e... Deuterium lamp 15 minutes f... Deuterium lamp 65 minutes g... Deuterium lamp 105 to 140 minutes Example 2 In the same manner as Comparative Example 1, a polyhexafluoromethacrylate resist was applied. After coating on two substrates, each sample was exposed to an electron beam exposure device with a line width of 1 μm.
0.5 μm and 0.2 μm were exposed at an exposure dose of 2×10 ⁻³ C/cm ² . Thereafter, one of the samples was developed with methyl isobutyl ketone as in Comparative Example 1, and the other sample was also developed with a deuterium lamp as in Example 1. As a result, when developed with methyl isobutyl ketone, a pattern with a width of 1 μm was developed without any meandering, whiskers, or bridges, but with pattern widths of 0.5 μm and 0.2 μm, the pattern meandered significantly due to swelling during development with methyl isobutyl ketone. In some cases, pattern peeling occurred.
On the other hand, those developed with a deuterium lamp have a diameter of 1 μm.
No meandering was observed in either the 0.5μm or 0.2μm patterns.
Moreover, unlike solvent development, there is no swelling of the pattern, so it was found that the pattern accuracy was good.

As explained above, when using a positive resist in the negative reversal irradiation area, developing with ultraviolet rays such as a deuterium lamp results in good pattern accuracy because unlike conventional solvent development, there is no swelling of the pattern during development. It has the advantage of being able to be developed.

Further, while conventional dry development methods that do not use solvents involve heat, the present invention does not involve heat, making it possible to develop resist patterns with ultimate precision.

[Brief explanation of drawings]

Figure 1 is a graph showing the change in the amount of resist with respect to the amount of electron beam irradiation, Figure 2 is the sensitivity curve of polyhexafluoromethacrylate resist developed by methyl isobutyl ketone, and Figure 3 is the deuterium lamp development curve of polyhexafluoromethacrylate resist. This is a sensitivity curve due to irradiation and development.

Claims (1)

[Claims] 1. An electron beam resist material whose molecular weight decreases under low electron beam irradiation and increases under high irradiation is applied to the substrate, and after pattern exposure with electron beam, the unexposed and low exposed areas are exposed to ultraviolet rays. A dry development method characterized by irradiation and removal.