WO1997034200A1

WO1997034200A1 - Process for developing a positive photoresist

Info

Publication number: WO1997034200A1
Application number: PCT/US1997/003092
Authority: WO
Inventors: William C. Nelson; James Neville Eilbeck
Original assignee: Hoechst Celanese Corporation
Priority date: 1996-03-12
Filing date: 1997-02-28
Publication date: 1997-09-18

Abstract

A process for improving the contrast, resolution and photospeed of a positive photoresist, consisting essentially of coating a positive photoresist film on a substrate, exposing the photoresist film to actinic radiation, baking the photoresist film after exposure to actinic radiation and developing the photoresist film with a single developer, wherein for a constant Normality developer, the contrast and resolution are improved by increasing the development temperature and/or increasing the time of development, whereas the photospeed is improved by decreasing the development temperature and/or decreasing the time of development.

Description

Description Process for Developing a Positive Photoresist

Field of the Invention The present invention relates to the aqueous alkali development of positive photoresist materials used for patterning a positive photoresist coated on a semiconductor wafer, micromachining, printed wire board and the like.

ffackground of the Invention Photoresists are well known to those skilled in the art. Generally speaking, they must be sensitive to light so that patterns can be formed in them and they must selectively resist subsequent etching or other processing so that the pattern can be transferred to the underlying substrate. The dominant photoresist system employed in integrated circuit manufacturing today is the novolak/diazonaphthoquinone combination. Diazonaphthoquinones of the class employed in positive photoresists are typically formed by the reaction of a naphthoquinone diazide sulfonyl chloride with a phenolic compound. Reaction products thus formed might include the naphthoquinone (1,2) diazide (5)- sulfόnyl, naphthoquinone (1,2) diazide (4)-sulfonyl, naphthoquinone (2,1) diazide (5)- sulfonyl, or naphthoquinone (2,1) diazide 4-sulfonyl radicals or mixtures thereof. Of course, the sulfonate esters thus synthesized might include the residue of any suitable phenolic compound. These diazides are fairly nonpolar organic molecules that are soluble in organic solvents, but not very soluble in water. Upon exposure to light, diazonaphthoquinone photoactive compounds form a polar, base-soluble carboxylic acid in accordance with mechanisms known to skilled artisans. Thus, using an aqueous base as the developer, the exposed photoactive compound is relatively soluble, while the unexposed photo active compound is relatively insoluble; creating a solubility difference which forms the basis of image formation.

It is not sufficient, however, to simply change the solubility of the photoactive compound, rather, the entire photoresist mixture must change its solubility. Thus, the interaction of the photoactive (photosensitive) compound with the binder resin is likewise an important consideration, as is its absorption spectrum. Accordingly, positive photoresist compositions generally include a phenol-formaldehyde resin of the Novolak class or sometimes a hydroxystyrene polymer such as poly(4-hydroxystyrene). Other polymeric components, including styrene, methyl styrene, styrene-maleic anhydride components in combination with the foregoing may also be employed. See, generally,

United States Patent Nos. 4,732,836 and 4,863,827 to Hoechst Celanese Corporation, Somerville, New Jersey for further information relating to positive photoresists.

Of considerable importance to the use of photoresists is the selection of a developer composition and developer process since development is a key aspect of the integrated circuit production process. Typically, optimizing the overall production is a trial and error procedure. This is true whether development is by immersion, by puddle or by spray techniques. United States Patent No. 4,628,023 to Cawston et al discloses and claims an aqueous metal ion-free developer composition including metal-free alkali and surfactants. The developer described reportedly results in a reduction in the energy necessary to expose the photoresist without a deleterious effect on image quality and resolution. In United States Patent No. 5,039,595 to Schwalm et al there is described another aqueous developer with a heterocyclic hydroxyalkyl compound as the hydroxy source. It is noted in the ' 595 patent that the following surfactants may be used: nonylphenoxypoly(ethyleneoxy)-ethanol, octylphenoxypoly(ethyleneoxy)-ethanol or commercial fluorinated surfactants (col. 4, lines 45-51). Of course, commercial fluorinated surfactants could include a vast number of candidate surfactant compounds of different types, including ionic or non-ionic surfactants.

The use of surfactants, desirable in some respects, frequently makes the developing process difficult to control, as is illustrated in United States Patent No. 4,710,449 to J. M. Lewis and A. J. Blakeney. In the ^x449 patent it is illustrated that 1%

(10,000 ppm) of a fluorinated alkyl alkoxylate in aqueous base developer is simply too aggressive, resulting in poor wall angles, poor contrast and unacceptable film loss. These problems may apparently be obviated by pre-treating the photuresist substrate with a pre-dip of aqueous base developer solution containing a cationic surfactant. It is, however, undesirable from a materials, productivity and logistical perspective to increase the number of processing steps. It is also well known to those skilled in the art that the performance behavior of a developer may be altered by variation in the normality of the developer. That is, more dilute developers will generally increase the contrast of a photoresist through higher selectivity. On the other hand, concentrated developers decrease the contract while increasing the apparent photosensitivity (photo speed), as compared to more dilute developers, so that production throughput is increased with the use of concentrated developers.

In modern production environments, developer solutions are supplied to various developing stations from a central, large volume storage tank. The use of a variety of different developers containing various surfactants or varying normalities to alter the development characteristics of the photoresist is, thus, undesirable.

Summary of the Invention

A method of developing a positive photoresist film subsequent to exposure is disclosed and claimed. The method includes applying the developer to the photoresist on the substrate at a specified and desired normality, and developing the exposed photoresist at a specified and desired temperature for a specified and desired development time to achieve the desired performance characteristics of the photoresist developer system. The basic and novel characteristics of the process are that the performance characteristics, e.g. contrast, for a given photoresist composition developed using a given developer and a given development process, are overwhelmingly the function of the developer normality, the development temperature and the development time. A single developer, such as that supplied from a central supply system, can be used to produce a variety of performance parameters, extremely high contrast or resolution on the one hand, or high throughput (high photospeed) on the other hand. A suitable temperature range for development is from about 15°C to about 40°C perferably from about 20°C to about 30°C, preferably for from about 20 seconds to about 180 seconds, more preferably from about 40 seconds to about 80 seconds.. Use of the present process allows one to control developer performance characteristics, without the addition of a surfactant. The present invention specifically relates to a process for improving the contrast and resolution of a positive photoresist, the process consisting essentially of coating a positive photoresist film on a substrate, exposing the photoresist film on the substrate to actinic radiation, baking the photoresist film after exposure to actinic radiation (post exposure baking) at a temperature of from about 120°C to about 160°C, more preferably from about 130°C to about 150°C, for a period of time of from about 5 seconds to about 30 seconds, more preferably from about 5 seconds to about 20 seconds, and developing the photoresist film with a single developer, wherein for a constant Normality developer, the contrast and resolution are improved by increasing the development temperature, while maintaining the development temperature within the range of from about 15°C to about 40°C, preferably from about 20°C to about 30°C, and/or increasing the time of development while mamtaining the time of development within the range of from about 20 seconds to about 180 seconds, preferably from about 40 seconds to about 80 seconds. Preferably, the developer utilized in this process has a Normality of from about 0.15 to about 0.5, iore preferably from about 0.2 to about 0.35.

The present invention also relates to a process for improving the photospeed of a positive photoresist, the process consisting essentially of coating a positive photoresist film on a substrate, exposing the photoresist film on the substrate to actinic radiation, baking the photoresist after exposure to actinic radiation (post exposure baking) at a temperature of from about 120°C to about 160°C, preferably from about 130°C to about

150°C, for a period of time of from about 5 seconds to about 30 seconds, preferably from about 5 seconds to about 20 seconds, and developing the photoresist film with a single developer, wherein for a constant Normality developer, the photospeed is improved by decreasing the development temperature, while maintaining the development temperature within the range of from about 15°C to about 40°C, preferably from about 20°C to about 30°C while maintaining the time of development within range of from about 20 to about 180 seconds, preferably from about 40 seconds to about 80 seconds. The developer utilized in the present process preferably has a Normality of from about 0.15 to about 0.5, preferably from about 0.2 to about 0.35. Description of the Preferred Embodiments

The present invention is described in detail below with reference to the examples provided for purposes of illustration and not by way of limitation. Modifications to specific embodiments will be readily apparent to those skilled in the art, which modifications are within the spirit and scope of the present invention as herealter claimed. In general, following coating, soft baking, exposing, and post-exposure baking a photoresist film on a suitable substrate, the image is developed by applying an aqueous alkali solution to the coated substrate at a specific temperature to thereby produce a relief image. The base is from about 0.15 to about 0.5 Normal (N) in the hydroxide component, about 0.2 to about 0.35 N being more typical, with about 0.261 N being preferred as an industry standard. The development temperatures are between about 15°C and about 40°C, preferably from about 20°C to about 30°C, typically about 25°C. The time for development is preferably from about 20 seconds to about 180 seconds, more preferably from about 40 seconds to about 80 seconds. The classical remedy for improving developer performance (wetting characteristics) is to in corporate a surfactant into the developer composition. This has the effect of lowering the surface tension at the interface between the developer and the surface of the exposed photoresist film deposited on the substrate. Good wetability results in reduced amounts of developer being required to develop the image. In addition to the clear cost benefits,, there is also an improvement in the uniformity of the development process, especially important when large wafers are being utilized as the substrate; and an improvement in CD (critical dimensions) uniformity by a factor of from about 2 to 3 times, as compared to the same developer without surfactant. Good wetability also contributes significantly to improvements in line definition, resolution, depth of focus and focus latitude.

The process of the present invention allows the use of non-surfactant developers to increase the wetability of the surface of the photoresist to be developed. Without surfactant, the development of fine lines is not acceptably achieved, normally because of the formation of web-like configurations between the lines (webbing) at the end of the development. This results in whole areas appearing to either have not started to develop or to be in the initial stages of the development process. The present process provides a substantial reduction in the surface tension at the developer/photoresist interface which results in the substantial absence of webbing, and vastly improved CD control. In controlling the temperature in the present process, it is also preferred to utilize a short term (5 to 30 seconds, preferably 5 to 20 seconds) post exposure bake (PEB) at a higher temperature than normal (120 to 160 degrees C, preferably 130 to 150). A PEB cycle of, e.g. 140 degrees for 10 seconds, after exposure on a contact hotplate substantially enhances the uniformity of the development process, while at the same time enhancing the thermal stability of the photoresist formulation. This can, suφrisingly, provide a zero bias for dense/isolated lines and spaces at varying focal points, which is highly desired in the present-day production of semiconductor having extremely fine dimensions of 0.30 microns (micrometers) or less.

Among the other parameters that, to a substantially less extent, improve photoresist performance are: speed enhancers in the developer, single vs. multiple puddle development, hot water rinse between multiple puddles, develop wash before exposure of the photoresist, solvent treatment (e.g. toluene), optimized softbake temperature, temperature gradient and/or time, softbake method (oven, hotplate, microwave, etc.), optimized humidity and/or temperature of the atmosphere, and process delays after sofibake exposure. However, the process of the present invention is an effective method for substantially improving the performance of the photoresist by controling the normality of the developer, and the development temperature and time.

The ammonium hydroxide component may be tetramethylammonium hydroxide (TMAH), trimethylethanol ammonium hydroxide (choline), tetra(2-hydroxyethyl) ammonium hydroxide, mixtures thereof or like compounds. TMAH is particularly preferred, while other ammonium hydroxide compounds may be employed if so desired. While any positive photoresist substrate πu ^e suitable, the inventive process is particularly applicable to photoresist comp .aons containing a phenol-formaldehyde resin known as a Novolak resin and a diazonaphthoquinone photoactive compound.

As used herein, the terminology "consists essentially of refers to the fact that the developing process of the present invention does not require substantial steps other than applying the developer at a suitable temperature and rinsing it away after a suitable time interval. This is contrasted with the process disclosed by Japanese Application 05181286-A which describes a two-step developing process using a developer first at a higher temperature then at a lower temperature. Similarly, IP 95021640 teaches the use of various developers at a single temperature. Other patents such as JP 58102942 and US 5342738 teach that constant temperature is important. Finally, US 5474877 teaches that heating the substrate while applying the developer.

By contrast, the present invention teaches using a single developer, such as would be used in a modem semiconductor factory, supplied from a central storage tank, and that the process contrast of the developer can be altered by merely changing the developer temperature during the develop cycle, in such a manner as to give performance similar to using multiple concentrations of developers at varying normalities.

Comparative Example A diazonaphthoquinone photoresist, AZ ®7800 (available from Hoechst Celanese Corporation, AZ Photoresist Products, 70 Meister Ave., Somerville, NJ 08876) was coated on silicon wafers to yield a film thickness of 1.09 μm (micrometers), after softbaking at 90'°C for 60 seconds on a hotplate. This photoresist film was image-wise exposed on a Nikon® 1755i7B Nikon Stepper equipped with a 0.54 NA lens using a reticule with varying line-space dimensions. After post-exposure baking at 110°C for 60 seconds on a hot plate, the photoresist film was developed using AZ® 300 MIF developer (0.261 N tetramethylammonium hydroxide solution, also available from AZ Photoresist Products), using a spray puddle process outlined in Table 1 in which the developer temperature was held at 20°C. The results are outlined in Table 2 below.

Table 1

Step Step Description Time (seconds^

1 Spray 3 sec @ 300 rpm

2 Build Puddle 2 see @ 0 rpm

3 Puddle 55 sec @ 0 rpm

4 Rinse 15 sec @ 300 rpm

5 Spin 15 sec @ 4000 rpm Examples 1 and 2 The Comparative Example was repeated, except the developer temperature was held at 25°C (Example 1) and 30°C (Example 2) throughout the develop cycle. The results are outlined in Table 2 below.

Examples 3 and 4 The Comparative Example was repeated, except the developer was AZ® 312 MIF (1:1.2) tetramethylammonium hydroxide solution. The results are outlined in Table 2 below.

Table 2 (55 second puddle developed)

Example Developer Developer Resolution (μπ Photospeed Normality Temp. (C.) fmJ/cm²

Comparative 0.261 20 0.34 175

1 0.261 25 0.30 245

2 0.261 30 0.28 290

3 0.245 20 0.36 195

4 0.220 20 0.34 265

The examples and Figure 1 show that, for constant normality developers, e.g. 0.261 N, the photospeed decreases with increasing developer temperature and resolution improves because of increased process contrast. Reducing developer normality increases photospeed, but does not improve resolution illustrating the developer temperature, and not photospeed, causes the improved performance.

Claims

Clai s:

1. A process for improving the contrast and resolution of a positive photoresist consisting essentially of coating a positive photoresist film on a substrate, exposing the photoresist film to actinic radiation, post exposure baking said photoresist film on said substrate at a temperature of from about 120°C to about 160°C for a period of time of from about 5 seconds to about 30 seconds, and developing the photoresist film with a single developer, wherein for a constant Normality developer, the contrast and resolution of the photoresist are improved by increasing the development temperature, while maintaining said development temperature within the range of from about 10°C to about 40°C, and/or increasing the time of development, while maintaining said time of development within the range of from within about 20 seconds to about 180 seconds.

2. The process of claim 1, wherein said photoresist film is post exposure baked at a temperature of from about 130°C to about 150°C for a period of time from about 5 seconds to about 20 seconds.

3. The process of claim 1, wherein the development temperature ranges from about 20°C to about 30°C.

4. The process of claim 1, wherein the time development ranges from about 40 seconds to about 80 seconds.

5. The process of claim 1, wherein said single developer has a Normality of from about 0.15 to about 0.5.

6. The process of claim 1, wherein said single developer has a Normality of from about 0.2 to about 0.35.

7. A process for improving the photospeed of a positive photoresist consisting essentially of coating a positive photoresist film on a substrate, exposing the photoresist film to actinic radiation, post exposure baking said photoresist film on said substrate at a temperature of from about 120°C to about 160°C for a period of time of from about 5 seconds to about 30 seconds, and developing the photoresist film with a single developer, wherein for a constant Normality developer, the photospeed of the photoresist is improved by decreasing the development temperature, while maintaining said development temperature within the range of from about 10°C to about 40°C.

8. The process of claim 7, wherein said photoresist film is post exposure baked at a temperature of from about 130°C to about 150°C for a period of time from about 5 seconds to about 20 seconds.

9. The process of claim 7, wherein the development temperature ranges from about 20°C to about 30°C.

10. The process of claim 7, wherein the time development ranges from about 40 seconds to about 80 seconds.

11. The process of claim 7, wherein said single developer has a Normality of from about 0.15 to about 0.5.

12. The process of claim 7, wherein said single developer has a Normality of from about 0.2 to about 0.35.