US20060099536A1

US20060099536A1 - Patterned structures of high refractive index materials

Info

Publication number: US20060099536A1
Application number: US11/316,058
Authority: US
Inventors: Elsa Reichmanis; Shu Yang
Original assignee: Individual
Current assignee: Individual
Priority date: 2002-12-17
Filing date: 2005-12-21
Publication date: 2006-05-11
Also published as: US20040115566A1; US7008757B2

Abstract

A process for forming a polymer template includes exposing a photoresist having polymer molecules to a light pattern and baking the photoresist to chemically react polymer molecules in portions of the photoresist that were exposed to light of the light pattern. The reacted polymer molecules have a different solubility in a solvent than chemically unreacted polymer molecules. The process also includes washing the baked photoresist with the solvent to produce a porous structure by selectively solvating one of the reacted polymer molecules and the unreacted polymer molecules. The porous structure can be used as template for forming porous structures of high refractive index materials.

Description

BACKGROUND OF THE INVENTION

1. Field of the Invention
The invention relates to structures with artificially patterned refractive indexes.
2. Discussion of the Related Art
A photonic crystal is a material with a refractive index that varies periodically. In a photonic crystal, the optical response results from a coherent superposition of optical waves propagating therein. Photonic crystals with sufficiently high refractive index contrasts, e.g., a contrast of more than 2, exhibit complete photonic bandgaps. A photonic bandgap prohibits electromagnetic waves from propagating in such a crystal for wave vectors at optical wavelengths.
Herein, a photonic crystal with a complete photonic bandgap will be referred to as a photonic bandgap structure. Photonic bandgap structures have applications in a variety of optical devices. Thus, processes for fabricating such structures are useful.
One process for fabricating a photonic bandgap structure involves producing a periodic array of voids in a base material. In such a void-filled photonic bandgap structure, the refractive index pattern has a high contrast if the base material has a high refractive index.
Fabricating a void-filled structure from a material with a high refractive index typically involves a multi-step template-based process. First, a reaction step produces a template with a periodic array of voids therein. Second, a fill step introduces a material with a high refractive index into the voids of the template. Third, a removal step eliminates the template material from the filled structure to produce a porous structure of the high refractive index material alone.
Such template-based fabrication processes need to account for the properties of both the template material and the filling material. In particular, the filling step must not damage the template material, and the removal step must not damage the filling material. Unfortunately, many photoresists do not withstand the process of being filled with a material having a high refractive index. Similarly, many high refractive index materials do not withstand the treatments for removing templates made of photoresists.

BRIEF SUMMARY

Various embodiments provide polymer templates and processes for fabricating the polymer templates. The polymer templates enable fabrication of porous structures from materials with high refractive indexes. In particular, the template material is compatible with being filled with a high refractive index material, and removal of the template material from the resulting structure is possible without damaging the high refractive index material. Both the filling and removal steps are typically performed at low temperatures to avoid damage to the various materials.
In one aspect, the invention features a process for forming a polymer template. The process includes exposing a photoresist including polymer molecules to a light pattern and baking the photoresist to chemically react polymer molecules in portions of the photoresist that were exposed to light of the light pattern. The reacted polymer molecules have a different solubility in a solvent than chemically unreacted polymer molecules. The process also includes washing the baked photoresist with the solvent to produce voids therein by selectively dissolving away either the reacted polymer molecules or the unreacted polymer molecules.
In another aspect, the invention features a process for forming a template of non-crosslinked polymer molecules. The process includes exposing a polymer photoresist to a light pattern to generate acid in illuminated regions of the photoresist and baking the photoresist to cause polymer molecules that are located in the illuminated regions to react with the acid. The baking step does not crosslink the polymer molecules. The process includes washing the baked photoresist in a nonpolar solvent to produce voids therein by selectively solvating either the reacted polymer molecules or the unreacted polymer molecules.
Herein, various processes involving exposure of a material to a light pattern with light and dark regions produce a void-filled structure. The resulting structures are referred to as positive tone structures if the voids occupy regions that were exposed to light regions of the light pattern. The resulting structures are referred to as negative tone structures if the voids occupy regions that were dark regions of the light pattern.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is flow chart showing a process for fabricating a porous template of non-crosslinked polymer molecules:
FIG. 2 is flow chart showing a process for fabricating a porous structure of a of a high refractive index material from the template of FIG. 1;
FIG. 3 illustrates a process that fabricates a porous structure of a high refractive index material by fabricating a negative tone polymer template;
FIG. 4 illustrates a process that fabricates a porous structure of a high refractive index material by fabricating a positive tone polymer template;
FIG. 5 shows an exemplary polymer molecule for the polymer photoresist used in the processes of FIGS. 3-4:
FIGS. 6A-6C show exemplary dye molecules for the polymer photoresist used in the processes of FIGS. 3-4;
FIG. 7A-7B show exemplary PAG molecules for the polymer photoresist used in the processes of FIGS. 3-4;
FIG. 8 illustrates the sequence of photoreactions in exemplary photoresists used in the processes of FIGS. 3-4;
FIG. 9 shows an acid-catalyzed reaction of polymer molecules in the processes of FIGS. 3-4; and
FIG. 10 shows a reaction that changes the solubility of polymer molecules previously modified by the acid-catalyzed reaction shown in FIG. 9.
In the various Figures, like reference numbers indicate elements with similar functions.

DETAILED DESCRIPTION OF ILLUSTRATIVE EMBODIMENTS

FIG. 1 illustrates a process 10 for fabricating a template from a polymer photoresist. The template is a homogeneous material of non-crosslinked polymer molecules with a pattern of voids therein. The voids of exemplary templates form periodic patterns in one-, two-, or three-dimensions.
The process 10 includes exposing a slab of photoresist to a light pattern to produce a corresponding pattern of acid in the photoresist (step 12). The photoresist is a homogenous mixture of polymer molecules, dye molecules, and photo-acid generator molecules. The light of the exposure causes the photo-acid generator molecules to produce acid, i.e. hydronium ions (H⁺). The distribution of H⁺ions corresponds to the light intensity pattern during the exposure.
Herein, a light pattern includes both an array of dark regions and an array of light regions. Exemplary light patterns are periodic in one, two, or three dimensions and have periods of about 0.1 microns and about 10 microns. i.e., mesoscopic scale periods suitable for photonic bandgap structures at telecommunications wavelengths. Such periodic light patterns result from interfering two, three, or four mutually coherent light beams. Processes for producing such light patterns are described in pending U.S. patent application Ser. No. 10/040,017, filed Jan. 4, 2002, ('017), which is incorporated herein by reference in its entirety.
The light exposure does not stimulate chemical reactions that significantly change the refractive index of the photoresist. In particular, the photochemical reactions do not polymerize, decompose, or crosslink molecules in the photoresist. Instead, the photochemical reactions produce acid, which does not significantly change the refractive index. The absence of refractive index changes insures that the light patterning will not lead to material changes that feedback and could otherwise interfere with the acid pattern formed in the photoresist. The final acid distribution forms a pattern that corresponds to the light pattern used to expose the photoresist.
The process 10 includes baking the exposed photoresist to cause polymer molecules, which are located in regions exposed by the light, to undergo a solubility change (step 14). In some embodiments, the solubility change results from generation of polar groups on the polymer molecules. In other embodiments, the solubility change results from generation of nonpolar groups on the polymer molecules. The new groups are generated by a heat-induced reaction of polymer molecules that are located in portions of the photoresist where acid was generated during the light patterning. The acid catalyzes the reaction.
The bake step proceeds at a temperature that is above the glass transition temperature of the organic polymer molecules of the photoresist and is lower than the decomposition temperature for the molecules. Exemplary bake temperatures for polymer resists are below about 350° C. and are preferably below about 200° C.
Various embodiments of the process 10 use a photoresist whose polymer molecules are initially soluble in a nonpolar solvent. For polymer molecules initially soluble in a nonpolar solvent, the acid-catalyzed reaction produces polymer molecules with more polar functional groups. The reacted polymer molecules are thus, less soluble in the nonpolar solvent and more soluble in polar solvents. After the bake, the photoresist includes a pattern of polymer molecule solubility type that tracks the light pattern used during the exposure at step 12.
In some embodiments, further chemical treatments invert the solubilities of polymer molecules located in respective light and dark regions of the light pattern as described below.
The process 10 also includes washing the baked photoresist with solvent to produce voids therein by selectively dissolving away polymer molecules of one solubility type (step 16). Exemplary washes use a nonpolar liquid as the solvent. In such solvents, the unreacted and reacted polymer molecules of the baked photoresist have different solubilities. For that reason, the wash generates a pattern of voids that tracks the light pattern originally used to exposure the photoresist. The resulting void-filled photoresist is a polymer template.
To enable the wash to produce voids, the conditions of the bake are selected to not crosslink polymer molecules. Crosslinking would significantly lower solubilities of the polymer molecules and would impede a solvent from removing the polymer template after subsequent filling with high index materials.
FIG. 2 illustrates a process 18 for fabricating a porous structure of a high refractive index material. The process 18 includes providing a void-filled template of non-crosslinked polymer molecules, e.g., a template fabricated by process 10 of FIG. 1 (step 20). The process 18 includes forming a filled structure by filling the voids of the template with a high refractive index material (step 22). The filling is performed under low temperature conditions that neither damage the template nor crosslink polymer molecules of the template. Exemplary filling temperatures are below about 350° C. Next, the process 18 includes washing the structure with a solvent that solvates polymer molecules of the template (step 24). The solvated polymer molecules are washed away to produce a void-filled structure of the high refractive index material alone. Finally, the porous structure of the high refractive index material is dried to remove remaining solvent from the voids therein (step 26).
FIGS. 3 and 4 illustrate specific embodiments 40, 40′ for fabricating porous structures 41, 41′ of a high index material according to the processes 10, 18 of FIGS. 1 and 2. The two porous structures 41 and 41′ are positive and negative tone structures, respectively. Exemplary structures 41, 41′ include photonic bandgap structures with high refractive index contrasts.
Processes 40 and 40′ start with a layer of photoresist 42. The photoresist is supported on a substrate 44, e.g., a glass substrate. The photoresist 42 is a homogeneous mixture whose components include soluble polymer molecules, dye molecules, photoacid generator (PAG) molecules, and solvent. Exemplary components for the photoresist 42 are shown in FIGS. 5, 6A-6C, and 7A-7B.
An exemplary polymer component for the photoresist 42 includes the polymer molecule 30 shown in FIG. 5. The polymer molecule 30 is a chain that includes M units of nonpolar fluorinated monomers 31 and N units of acid-sensitive monomers 32. The integers N and M are typically in the range of 10-1000 so that the polymer molecule 30 is soluble in nonpolar solvents. The fluorinated monomers 31 increase the solubility of the polymer molecule 30 in nonpolar solvents like fluid and supercritical carbon dioxide.
An exemplary dye component for the photoresist 42 includes one of the dye molecules 33, 34, 35 shown in FIGS. 6A-6C. The dye molecules 33, 34 are activated by visible light having wavelengths of about 535 nanometers (nm) and 470 nm, respectively. These dye molecules 33, 34 are available from Spectra Group Limited, Inc., 1722 Indian Wood Circle, Suite H Maumee, Ohio 43537 USA, under respective product names HNu-535 and HNu-470. The dye molecule 35 is activated by light with a wavelength of about 560 nm. The dye molecule is known as Rose Bengal and is available from Aldrich Company, P.O. Box 2060, Milwaukee, Wis. 53201 USA.
An exemplary PAG component for the photoresist 42 includes one of the molecules 36, 37 as shown in FIGS. 7A-7B. The PAG molecule 36 is available from Sartomer Inc., Oaklands Corporate Center, 502 Thomas Jones Way Exton, Pa. 19341 USA, under the product name SarCat® SR1012. The PAGM 37 is available from UCB Chemicals Corp., 2000 Lake Park Drive, Smyrna (Atlanta), Ga. 30080 USA, under the product name OPP1.
Exemplary solvents for the photoresist 42 include hydrocarbons, fluorinated hydrocarbons, and liquid and supercritical carbon dioxide.
Referring to FIG. 3, the fabrication process 40 includes exposing the photoresist 42 to a light pattern 50 (step 12). The light pattern produces an array of illuminated regions 45 in the photoresist 42. In the illuminated regions 45, a photoreaction sequence produces acid.
FIG. 8 shows the photoreaction sequence 38 that occurs in illuminated regions 45 of the photoresist 42. In the photoreaction sequence 38, photons excite dye molecules (DYE*), which then excite PAG molecules (Ar₂I⁺X⁻). The excited PAG molecules Ar₂I⁺X^−*subsequently dissociate to produce free radicals Ar—I^+*and Ar*X⁻. The free radicals Ar—I^+*react with solvent molecules (Sol-H) to produce hydronium ions H⁺. The photoreaction sequence 38 is also described in the above-referenced '017 patent application.
The photoreaction sequence 38 does not produce changes to the refractive index in the illuminated regions 45 of the photoresist 49. Thus, the light pattern does not produce feedback, which could otherwise interfere with formation of the light pattern in the photoresist 42. For this reason, the exposure cm be performed over an extended time period and produce an acid pattern that tracks a selected light pattern.
Referring again to FIG. 3, the fabrication process 40 includes baking the photoresist 42 to cause a reaction that changes the solubility of polymer molecules that are located in the previously illuminated regions 45 of the photoresist 42 (step 14). Photo-produced acid catalyzes the reaction, which generates polar functional groups on the polymer molecules located in the regions 45. The polar functional groups make the reacted polymer molecules of the baked photoresist 43 less soluble in nonpolar solvents.
FIG. 9 shows an exemplary solubility changing reaction for the polymer molecule 30 shown in FIG. 5. The reaction is driven by H⁺ ions and generates polar carboxylic acid groups, i.e., —CO₂H groups, on the polymer molecules. Due to the carboxylic acid groups, the reacted polymer molecules 39 have lower solubilities in fluid or supercritical carbon dioxide.
Referring again to FIG. 3, the fabrication process 40 includes producing a void-filled template 47 by washing the baked photoresist 43 with a solvent that selectively solvates nonpolar polymer molecules (step 16). The washing step involves placing the photoresist 43 and supporting substrate 44 in a cell and filling the cell with liquid or supercritical carbon dioxide. The liquid or supercritical carbon dioxide diffuses into the photoresist 43 to replace the nonpolar solvent therein without causing drying. Since liquid or supercritical carbon dioxide is a nonpolar solvent, the diffusing carbon dioxide dissolves nonpolar, unreacted, polymer molecules without dissolving the polar, reacted, polymer molecules. Thus, the diffusion of the carbon dioxide selectively removes the unreacted polymer molecules to produce a void-filled template 47. Since the nonpolar polymer molecules are located outside of the previously illuminated regions 45, the template 47 is a negative tone structure.
The fabrication process 40 includes raising the pressure and temperature in the cell to values corresponding to the critical point of carbon dioxide. At the critical point conditions, the carbon dioxide is allowed to diffuse out of the template 47 thereby drying the template 47. The drying does not produce internal stresses, which could otherwise cause cracks, because critical carbon dioxide is a surface tension free fluid. Drying the template 47 in critical carbon dioxide minimizes the risk of damage due to internal surface tension forces.
The fabrication process 40 includes filling voids in the template 47 with a high refractive index material (step 22). The filling step is performed at a temperature below 350° C. and preferably below about 300° C. to avoid damaging polymer molecules, cross-linking of polymer molecules, or damaging the high refractive index material. The filling step produces filled structure 48, which includes non-crosslinked polymer molecules in regions 45 and high refractive index material elsewhere.
A variety of processes are available to fill the voids of the template 47 with a high refractive index material. Exemplary fill steps include melting a high refractive index material with a low melting temperature, e.g., selenium or bismuth, and filling the melted material into the voids of the template 47 under pressure. Other exemplary fill steps include subliming a high refractive index material having a low vaporization temperature into the voids of the template 47. Other exemplary fill steps include performing a gas-phase reaction that deposits a high refractive index byproduct it, the voids of the template 47.
The fabrication process 40 also includes washing the filled structure 48 to dissolve away the polymer molecules of the original template 47 to produce a porous structure 41 of the high refractive index material and then drying the structure (steps 24 and 26). The washing step produces a porous structure of the high refractive index material alone. For the polar polymer molecules 39 shown in FIG. 7, a basic aqueous solution is a suitable solution for the washing step.
FIG. 4 illustrates the alternate process 40′ for fabricating porous structure 41′ of high index material. The process 40′ includes performing steps 12 and 14 to produce the baked photoresist 43 as already described with respect to process 40 of FIG. 3. In the baked photoresist 43, the reacted polymer molecules in regions 45 are insoluble in a nonpolar solvent. As an example, the reacted polymer molecule 39 of FIG. 8 is insoluble in nonpolar solvents due to the polar carboxylic acid groups produced by the chemical reaction of step 14.
The process 40′ includes performing a vapor-phase reaction to change the solubility of a portion of the polymer molecules of the photoresist 43. The vapor-phase reaction produces a photoresist 43′ in which the polymer molecules located in the regions 45 have a changed solubility. The reaction causes these previously reacted polymer molecules to become more soluble in nonpolar solvents such as liquid or supercritical carbon dioxide.
FIG. 10 shows an exemplary vapor-phase reaction for changing the solubility of the polymer molecule 39 of FIG. 9. The reaction is a hydrosilylation in which (CH₃)₃SiOSi(CH₃)₃(HMDS) reacts with carboxylic acid groups on the polymer molecule 39 to produce less polar functional groups. The hydrosilylation reaction changes the solubility of the previously reacted polymer molecules. In particular, the reaction caps the polar carboxylic acid groups to make molecule 39′ significantly more soluble in nonpolar solvents than the original molecule 39.
Referring again to FIG. 4, the process 40′ includes uniformly illuminating the photoresist 43′ to generate acid in regions 46 via the photo-reaction sequence 38 of FIG. 8. Then, a rebake step causes the photo-produced acid to catalyze a reaction in the polymer molecules that are located in the regions 46 of the photoresist 43″. The reaction changes the solubility of said polymer molecules, which were not previously exposed by the nontrivial light pattern. From the exemplary nonpolar polymer molecules 30, shown in FIG. 9, the rebake produces the polar molecules 39. The rebake does not crosslink polymer molecules or cause hydrosilylated polymer molecules in the previously reacted regions 45 of the photoresist 43′ to further react.
The process 40′ also includes producing a void-filled polymer template 47 by washing the rebaked photoresist 43″ with fluid or supercritical carbon dioxide as already described with respect to above process 40 (step 16). The wash dissolves away the nonpolar polymer molecules that located in regions 45 thereby producing a positive tone structure in polymer template 47′. Combining the vapor-phase reaction, uniform illumination, and rebake inverted the spatial distribution of molecules soluble in polar and nonpolar solvents, respectively, to produce a positive tone template 47′ rather than the negative tone template 47 of FIG. 3.
The fabrication process 40′ also includes filling voids in the template 47′ with a high refractive index material to produce filled structure 48′ in a manner already described with respect to process 40 (step 22). The filled structure 48′ is then washed with a solvent that solvates the non-crosslinked polymer molecules of the template 47′ to produce the porous structure of the high refractive index material (step 24).
The porous structure is dried to produce the final void-filled structure 41′ of the high refractive index material (step 26). The final void-filled structure 41′ of the high refractive index material is a positive tone structure.
From the disclosure, drawings, and claims, other embodiments of the invention will be apparent to those skilled in the art.

Claims

1-16. (canceled)

17. A process comprising:

exposing a polymer photoresist to a light pattern to generate acid in illuminated regions of the photoresist;

baking the photoresist to cause polymer molecules located in the illuminated regions to react with the acid without being crosslinked;

washing the baked photoresist in a nonpolar solvent to produce voids in the baked photoresist by selectively solvating one of the reacted polymer molecules of the photoresist and the unreacted polymer molecules of the photoresist; and

producing a filled structure by filling the voids in the photoresist with a material; and

removing unsolvated ones of the polymer molecules from the filled structure to produce a photonic bandgap structure.

18. The process of claim 17, wherein the removing further comprises washing the filled structure with a polar solvent to produce a porous structure by dissolving the polymer molecules of the filled structure.

19. A process, comprising:

exposing a photoresist comprising polymer molecules to a light pattern;

baking the photoresist to chemically react ones of the polymer molecules exposed to light of the light pattern, the reacted ones of the polymer molecules having a different solubility in a solvent than chemically unreacted ones of the polymer molecules;

washing the baked photoresist with the solvent to produce voids in the baked photoresist by selectively solvating one of the reacted ones of the polymer molecules and the unreacted ones of the polymer molecules;

filling the voids with a material to produce a filled structure; and

forming a photonic bandgap structure from to filled structure.

20. The process of claim 19, further comprising washing the filled structure with a solvent that solvates polymer molecules to produce a porous structure.

21. The method of claim 19, wherein the filling includes filling the voids with a melted material.

22. The method of claim 19, wherein the material comprises selenium or bismuth.

23. The method of claim 19, wherein the filling includes subliming the material into the voids.

24. The method of claim 19, wherein the filling includes performing a gas-phase reaction to deposit the material in the voids.