U.S. patent application number 11/635440 was filed with the patent office on 2007-04-05 for process for making crystalline structures having interconnected pores and high refractive index contrasts.
This patent application is currently assigned to Lucent Technologies, Inc.. Invention is credited to Gang Chen, Ronen Rapaport, Elsa Reichmanis, Shu Yang.
Application Number | 20070074540 11/635440 |
Document ID | / |
Family ID | 32927035 |
Filed Date | 2007-04-05 |
United States Patent
Application |
20070074540 |
Kind Code |
A1 |
Chen; Gang ; et al. |
April 5, 2007 |
Process for making crystalline structures having interconnected
pores and high refractive index contrasts
Abstract
Techniques for producing a glass structure having interconnected
macroscopic pores, including providing a polymeric structure having
interconnected macroscopic pores; providing polymerizable glass
precursors; filling pores in the polymeric structure with the
polymerizable glass precursors; polymerizing the polymerizable
glass precursors to yield a filled polymeric structure; and
decomposing the filled polymeric structure to produce a glass
structure having interconnected macroscopic pores. Techniques for
filling pores of such glass structure with a material having a high
refractive index, and for then removing the glass structure.
Structures can be produced having interconnected macroscopic pores
and high refractive index contrasts, which can be used, for
example, as photonic band gaps.
Inventors: |
Chen; Gang; (New Providence,
NJ) ; Rapaport; Ronen; (Chatham, NJ) ;
Reichmanis; Elsa; (Westfield, NJ) ; Yang; Shu;
(North Plainfield, NJ) |
Correspondence
Address: |
THE ECLIPSE GROUP
10605 BALBOA BLVD., SUITE 300
GRANADA HILLS
CA
91344
US
|
Assignee: |
Lucent Technologies, Inc.
Murray Hill
NJ
07974-0636
|
Family ID: |
32927035 |
Appl. No.: |
11/635440 |
Filed: |
December 6, 2006 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
10383150 |
Mar 6, 2003 |
7168266 |
|
|
11635440 |
Dec 6, 2006 |
|
|
|
Current U.S.
Class: |
65/395 |
Current CPC
Class: |
C03B 19/12 20130101;
C03B 2201/58 20130101; C03C 11/00 20130101 |
Class at
Publication: |
065/395 |
International
Class: |
C03B 37/016 20060101
C03B037/016 |
Claims
1. A method of producing a glass structure having, interconnected
macroscopic pores comprising: providing a polymeric structure
having interconnected macroscopic pores; providing polymerizable
glass precursors; filling pores in said polymeric structure with
said polymerizable glass precursors; polymerizing said
polymerizable glass precursors to yield a filled polymeric
structure; and decomposing said filled polymeric structure to
produce a glass structure having interconnected macroscopic
pores.
2. The method of claim 1, further including opening surface pores
on said glass structure.
3. The method of claim 1 further including filling pores of said
glass structure with a material having a high refractive index to
yield an interpenetrating structure
4. The method of claim 1 in which said polymerizable glass
precursors include silicon.
5. The method of claim 3 in which filling pores is carried out by
chemical vapor deposition.
6. The method of claim 3 in which said material is filled into
pores of said glass structure at a temperature of at least about
400.degree. C.
7. The method of claim 3 in which said material has a refractive
index of at least about 3.0.
8. The method of claim 3 in which said material is selected from
the group consisting of: silicon, tellurium, gallium arsenide,
gallium nitride, indium phosphide, aluminum nitride, indium
nitride, gallium antimonide, indium antimonide, aluminum
antimonide, aluminum gallium nitride, aluminum gallium arsenide,
aluminum gallium antimonide, gallium aluminum antimonide, indium
gallium antimonide, gallium arsenic antimonide, indium gallium
phosphide, indium gallium arsenide, indium arsenic antimonide,
indium gallium arsenide phosphide, indium aluminum gallium
arsenide, indium aluminum gallium nitride, indium aluminum gallium
antimonide, lead sulfide, cadmium selenide, tin sulfide, cadmium
sulfide, zinc selenide, bismuth, and selenium.
9. The method of claim 3 further including removing said glass
structure from said interpenetrating structure.
10. The method of claim 6 in which said material is filled into
pores of said glass structure at a temperature of at least about
500.degree. C.
11-20. (canceled)
21. The method of claim 1 in which providing polymerizable glass
precursors includes providing inorganic-organic
silicon(e)-containing precursors in which silicon is bonded
directly or indirectly to reactive organic groups.
22. The method of claim 1 in which providing polymerizable glass
precursors includes providing a member selected from the group
consisting of organically modified silicate monomers and
silsesquioxane monomers.
23. The method of claim 1 in which providing polymerizable glass
precursors includes providing partially condensed silsesquioxane
oligomers having an average molecular weight of about 500-20,000
grams per mole.
24. The method of claim 1 in which providing polymerizable glass
precursors includes providing an inorganic silicon-containing
compound dispersed in a polymerizable material.
25. The method of claim 24 in which providing polymerizable glass
precursors includes providing silicon dioxide dispersed in a
polymerizable material.
26. The method of claim 1 in which providing polymerizable glass
precursors includes providing an alkoxysilane.
27. The method of claim 1 in which providing polymerizable glass
precursors includes providing a silsesquioxane oligomer having the
following formula: ##STR3## in which R can be any saturated or
unsaturated, linear or branched, aliphatic or aromatic hydrocarbon
moiety.
28. The method of claim 1 in which providing polymerizable glass
precursors includes providing polydimethylsiloxane oligomers.
29. The method of claim 1 in which providing polymerizable glass
precursors includes providing vinyl-terminated siloxane base
oligomers and hydride functional siloxane oligomers.
30. The method of claim 1 including providing polymerizable glass
precursors that include silicon.
Description
FIELD OF THE INVENTION
[0001] The present invention generally relates to improvements in
the field of materials having high refractive index contrasts and
open, interconnected pores. These materials are useful for
fabrication of photonic band gap structures and other structures
taking advantage of such porosity. More particularly, the present
invention addresses advantageous techniques for producing negative
template periodic porous glass molds that can withstand the high
temperatures necessary to fill the negative template pores with a
material having a high refractive index, to create a positive final
structure having interconnected macroscopic pores.
BACKGROUND OF THE INVENTION
[0002] A photonic crystal is a periodically structured composite
material, with a unit cell whose dimensions are on the order of a
wavelength of visible to infrared light. Such dimensions may
broadly range from about 50 nanometers (nm) to about 10 micrometers
(.mu.m), and preferably are smaller than one .mu.m. These
dimensions may be suitable, for example, to generate a photonic
band gap useful with wavelengths between about 600 nm and about
1.65 .mu.m, and more particularly, for example, wavelengths such as
1.3 .mu.m and 1.55 .mu.m as typically used in optical
communications. Such a crystal is made from two constituent
materials whose refractive indices greatly differ, such that the
contrast ratio between them generally is at least 2:1.
Three-dimensional (3D) photonic crystals typically consist of
interpenetrating networks of dielectric material and air, the
latter serving as the material of relatively low refractive
index.
[0003] The defining characteristic of a photonic crystal is a range
of frequencies within which no propagating electromagnetic modes
exist. Multiple interference between waves scattered from each unit
cell of the crystalline structure can open a photonic band gap. A
photonic band gap is a range of frequencies, analogous to the
electronic band gap of a semiconductor, within which no propagating
electromagnetic modes exist. Structural defects in a photonic
crystal may give rise to spatially localized electromagnetic modes,
or microcavity-confined modes, at energies within the gap.
Waveguides are formed by coupling such defects together. A
waveguide operating at a frequency within a photonic band gap
cannot leak, because there are no propagating electromagnetic modes
in the surrounding photonic crystal capable of carrying energy
away. In principle, this absence of leakage allows the fabrication
of waveguides that turn corners in a distance on the order of the
optical wavelength, requiring two orders of magnitude less space
than semiconductor ridge waveguides currently used in integrated
optics, which typically have a minimum bend radius greater than 100
.mu.m.
[0004] Photonic band gap crystals can form the basis for
miniaturized integrated optical circuits with length scales
comparable with those of integrated electronics. Such crystals can
serve, for example, as waveguides, splitters, optical insulators,
optical filters, microcavity lasers, optical switches, routers, and
in other photonic band gap applications. They can be designed to
optically act on either a one dimensional (1D), two dimensional
(2D) or 3D level. Such engineering applications require fabrication
technologies for the cheap and rapid production of periodic
structures that have the potential to incorporate engineered
structural defects to create microcavities and waveguides. Other
applications for these structures include uses as filters,
catalysts, and biocompatible materials.
[0005] Despite the broad potential utility of materials having
interconnected porous structures having sub-micron periodicity and
high refractive index contrasts, conventional technologies are
found difficult to provide cost effective methods of making them.
According to one conventional method referred to as "log-piling,"
layers of uniformly spaced grid elements formed of a high
refractive index material are painstakingly stacked together by
serial lithography and etching. Although this method can produce a
functional photonic band gap structure, such processing is labor
and time intensive and therefore impractical for commercialization.
Moreover, sufficiently accurate mutual registration of the layers
is difficult to achieve, and control of the contours of the grid
elements is limited. A variation of this method, involving fusion
of successive prefabricated grid wafers, generates similar
problems. Another tedious method for fabricating band gap
structures involves drilling holes in a solid block of high
refractive index material, by using a laser for example.
[0006] Further methods used to make crystalline structures having
interconnected macroscopic porosity have involved creating a
negative template mold into which a material having a high
refractive index, or its precursor, is then filled. The negative
template mold is then decomposed, for example by oxidation, to
produce a positive final porous structure with high index contrast.
These methods include various processes involving exposure of a
material to a light pattern with light and dark regions, producing
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.
[0007] One group of methods used to make negative template molds
for crystalline structures having macroscopic porosity has depended
on chemical self-assembly techniques. For example, one of these
methods involves self-assembly of colloids by sedimentation,
forming a face-centered cubic lattice. Drawbacks to such a
methodology include the prevalence of undesired lattice defects
such as stacking faults, and the inability to obtain lattice types
other than face-centered cubes and to otherwise control lattice
parameters other than the colloid cell diameter. Another chemical
self-assembly method involves selective decomposition of one block
in a block copolymer to leave controlled porosity after processing.
Cylinder and gyroid lattices can be produced by this method.
However, lattice defects are prevalent, and pore size typically is
less than 100 nm. Two-photon polymerization can be used to write
periodic structures with different lattice constants. However, it
is a rather slow point-wise writing process.
[0008] Holographic lithography is a method that has been
successfuilly used to make a polymeric negative template mold
suitable for producing a crystalline structure having
interconnected macroscopic porosity. According to this method, a
photosensitive material, for example, a film of a desired
thickness, is subjected to an optical interference pattern
resulting from multiple beam interference. The material can then be
selectively polymerized or deprotected in regions where the film is
exposed to the interfered optical signals. After subsequent
development using a suitable solvent, a porous 2D or 3D template is
obtained. Among the advantages of holographic lithography are an
ability to select the porous structure's lattice constants, an
ability to produce crystalline structures free of unintended
defects, and the availability of inexpensive commercial means for
implementation.
[0009] However, efforts to effectively use such a negative template
polymeric mold to produce a positive final crystalline structure
having a high refractive index contrast have not been entirely
successful. Titanium tetraethoxide has, for example, been filled
into such molds in sol-gel form. However, complete filling of such
a mold may be problematic due to pore-clogging. Moreover, the
refractive index of titanium dioxide upon decomposition of the
negative- and positive-template organic components is only 2.0 to
2.4 on a scale in which air has a refractive index of 1.0, which is
near the low end of about 2.0 for photonic band gap utility. In a
related method, materials having refractive indices of less than 3,
such as cadmium sulfide and cadmium selenide, have been
electroplated into a polymer mold. However, the range of compounds
suitable for electroplating is limited.
[0010] High temperature methods involving gas phase deposition,
such as chemical vapor deposition (CVD), would be effective for
filling a material of high refractive index, such as elemental
silicon, into a mold. However, a polymer mold such as those
discussed above clearly cannot withstand the temperatures of
400.degree. C. or more, and often 500.degree. C. or more, necessary
to create the chemical vapor. In one effort to resolve this
problem, sea urchin skeletons were used as negative template molds
and filled with polydimethylsiloxane (PDMS) oligomers, which can be
polymerized to solid form and then oxidized to silicon dioxide
glass. However, this method necessarily depends on the irregular
structure of a natural sea urchin skeleton to determine the mold
structure, allowing no control over the uniformity, size, contours
or interconnectivity of the pores. In general, the porosity
resulting from such sea urchin based methods is on the order of
about 20 to about 100 .mu.nm, far above the micron range. Moreover,
photonic band gap structures require introduction of precisely
positioned point defects to provide waveguide pathways through the
otherwise non-propagating material, the introduction of which
cannot be controlled by using sea urchin skeletons as negative
template molds.
[0011] Accordingly, there is a need for a process for the
production of negative template porous molds that can be used to
produce positive final crystalline structures fabricated from
materials having a high refractive index and having interconnected
macroscopic pores.
SUMMARY OF THE INVENTION
[0012] The present invention provides methods for producing
negative template porous glass molds suitable for production of
positive final structures having an interconnected macroporous
crystalline structure. Macroscopic pores are defined as pores that
are larger than about 500 angstroms in diameter. The negative
template glass molds can withstand the high temperatures necessary
to fill the negative template pores by CVD or an alternative high
temperature process, with a material having a high refractive index
to create the positive final porous structure. In one embodiment,
the negative template mold so produced is constituted by silicon
dioxide (SiO.sub.2) or other glass. The present invention provides
methods for making such negative template molds either directly
from glass precursors, or indirectly from a positive template
polymeric mold. The present invention further provides methods for
using the negative template molds to produce positive final porous
structures having high refractive index contrasts and
interconnected macroscopic pores. The present invention further
provides crystalline structures having interconnected macroscopic
pores and high refractive index contrasts, substantially uniform
pore lattices in predefined patterns, and point defects at
predetermined locations to provide optical propagation
pathways.
[0013] According to one embodiment, the present invention provides
a method of producing a glass structure having interconnected
macroscopic pores. The method comprises the steps of: providing a
polymeric structure having interconnected macroscopic pores;
providing polymerizable glass precursors; filling pores in said
polymeric structure with said polymerizable glass precursors;
polymerizing said polymerizable glass precursors to yield a filled
polymeric structure; and decomposing said filled polymeric
structure to produce a glass structure having interconnected
macroscopic pores.
[0014] According to another embodiment, the present invention
provides a method of producing a glass structure having
interconnected macroscopic pores comprising the steps of: providing
a photosensitive medium comprising glass precursors; exposing said
medium to an optical interference pattern; polymerizing or
deprotecting the portions of said medium exposed to said optical
interference pattern, leaving other unpolymerized or protected
portions; removing unpolymerized or deprotected portions of said
medium; and decomposing said medium to produce a glass structure
having interconnected macroscopic pores.
[0015] In further embodiments, such methods further comprise the
step of filling pores of said glass structure with a material
having a high refractive index to yield an interpenetrating
structure. In additional embodiments, such methods further comprise
the step of removing said glass structure from said
interpenetrating structure.
[0016] A more complete understanding of the present invention, as
well as other features and advantages of the present invention,
will be apparent from the following detailed description and the
accompanying drawings.
BRIEF DESCRIPTION OF THE DRAWINGS
[0017] FIG. 1 is a flow chart of a method for producing structures
having periodic interconnected macroscopic pores according to the
present invention;
[0018] FIG. 2 shows various silicon alkoxides that can be used
according to the methods of the present invention;
[0019] FIG. 3 shows various silsesquioxanes that can be used
according to the methods of the present invention;
[0020] FIG. 4 shows the reaction process for sol-gel polymerization
of a tetraalkoxysilane, which can be used according to the methods
of the present invention;
[0021] FIG. 5 shows a reaction process for formation of an oligomer
by condensation of two silicone-containing alkoxide (ormosil)
monomers, such oligomers being useful according to the methods of
the present invention;
[0022] FIG. 6 shows a hydrosilylation reaction process for
polymerization of vinyl-terminated siloxane base oligomers and
hydride functional siloxane oligomers, the resulting polymerizates
being useful according to the methods of the present invention;
[0023] FIG. 7 is a flow chart of another method for producing
structures having interconnected macroscopic pores according to the
present invention;
[0024] FIGS. 8A-C show various sensitizer compounds that can be
used according to the methods of the present invention;
[0025] FIGS. 9A-B show various ionic initiator complexes that can
be used according to the methods of the present invention;
[0026] FIG. 10 shows various silicon(e)-containing acrylate and
methacrylate polymers that can be used according to the methods of
the present invention;
[0027] FIG. 11 shows various silicon(e)-containing styrenic
polymers that can be used according to the methods of the present
invention;
[0028] FIG. 12 shows various silicon(e)-containing epoxy oligomers
that can be used according to the methods of the present
invention;
[0029] FIG. 13 shows various silicon(e)-containing acrylate and
methacrylate oligomers that can be used according to the methods of
the present invention;
[0030] FIG. 14A shows a setup for producing an optical intensity
pattern in order to carry out the methods of the present
invention;
[0031] FIG. 14B shows a starting medium and supporting substrate
disposed between glass prisms in order to carry out the methods of
the present invention;
[0032] FIG. 15 shows photo-reactions that are initiated by exposing
a starting medium to light in carrying out the methods of the
present invention; and
[0033] FIG. 16 shows a sequence of polymerization reactions
catalyzed in a starting medium containing epoxy functional groups
in carrying out the methods of the present invention.
DETAILED DESCRIPTION
[0034] The present invention provides methods of producing a
negative template mold, having an interconnected macroscopic porous
structure that can withstand the high temperatures necessary to
fill the pores by CVD or other high temperature process, with a
material having a high refractive index to create a positive final
porous structure. The present discussion will principally address
these product structures in the context of their utility as
photonic band gap structures, but it will be understood that these
structures have other end-use applications as well.
[0035] According to one embodiment, a negative template mold is
prepared by filling an organic polymeric positive template mold
having interconnected macroscopic pores, with a polymerizable
precursor to the formation of SiO.sub.2 or other glass. After
polymerization of the precursor, the positive template organic
polymer mold is removed, and the precursor is converted to glass to
constitute the desired negative template mold.
[0036] In order to produce the organic polymeric positive template
mold for this embodiment, a photosensitive starting medium
(starting medium), for example, is first provided. The starting
medium is a photoresist, and includes photo-sensitizer molecules
and a photosensitive material, uniformly dispersed in a solvent.
The photosensitive material can be in the form of monomers,
oligomers, polymers or mixtures. The photo-sensitizer molecules
initiate polymerization or deprotection of the starting medium,
directly or indirectly, upon exposure to light having an effective
wavelength. For example, upon such exposure to light, the
photo-sensitizer molecules can initiate sequences of photo-chemical
reactions. During post-exposure baking, the products of the
photo-chemical reactions then catalyze chemical reactions including
polymerization of the monomers and oligomers, as well as
deprotection of reactive sites and crosslinking of functional
groups in polymers.
[0037] A suitable starting medium is provided, for example by spin
coating or casting a film of the starting medium on a rigid
substrate. The film may have a thickness, for example, of about 1
.mu.m to about 100 .mu.m or more. The film is then prebaked to
evaporate out the solvent, at a temperature too low to prematurely
induce substantial polymerization or decomposition. For example,
temperatures below about 100.degree. C. can be used.
[0038] The starting medium is then exposed to a 2D or 3D optical
interference pattern produced by combining a plurality of mutually
coherent beams of light of a particular wavelength having
predefined periodic intensity variations. Such variations are
programmed to define regions of interconnected macroscopic porosity
having a defined periodic lattice structure to yield a crystalline
structure having photonic band gap properties. A 3D optical
interference pattern can be used to produce a structure having 3D
optical band gap properties. A 2D pattern yields a crystalline
structure having 2D optical band gap properties. The periodic
regions of intensity generated by the interference pattern cause
localized activation of the dispersed photo-sensitizer molecules,
thereby causing, directly or indirectly, localized polymerization
or photo-deprotection of the photosensitive material.
[0039] In one embodiment, a photo-sensitizer is locally activated
by such 3D interference pattern, but does not cause such localized
polymerization or deprotection of starting medium until a separate
step later occurs, such as heating of the material in the presence
of the activated photo-sensitizer. In this manner, the 3D
interference exposure can be completed before any polymerization or
deprotection occurs. Such polymerization or deprotection during
exposure changes the localized density of the starting medium,
changing its refractive index and distorting incident light. Hence,
by delaying the polymerization or deprotection to a step separate
from exposure to the optical interference pattern, distortions in
the desired pattern of localized activation of the photo-sensitizer
can be avoided. Alternatively, the 3D interference exposure can be
delivered in a pulsed, non-continuous mode, which can also reduce
such distortions.
[0040] A two-photon point-writing process can then be used to
introduce desired point defects. Next, depending on the type of
polymerizable material and photo-sensitizer employed, the selective
polymerization or deprotection of the intended portions of the
exposed starting medium is completed.
[0041] Following selective polymerization or deprotection of the
exposed starting medium, the unpolymerized or deprotected portions
of the monomers, oligomers and polymers are removed by a solvent
exchange step, yielding the desired positive template mold.
[0042] The above discussion is exemplary of methods that can be
used to produce a positive template polymeric structure having
interconnected macroscopic pores. Further details regarding
suitable methods for producing such positive template polymeric
crystalline structures are disclosed in Megens et al., U.S. patent
application Ser. No. 10/040,017, filed on Jan. 4, 2002, entitled
"Fabricating Artificial Crystalline Structures," the entirety of
which is hereby incorporated by reference.
[0043] Such a positive template polymeric structure having
interconnected macroscopic pores, a substantially uniform pore
lattice in a predefined pattern, and point defects at predetermined
locations, can then be used to prepare a negative template mold
according to the methods of the present invention.
[0044] According to one embodiment illustrated in FIG. 1, the
present invention provides a method of using such a positive
template polymeric structure to produce a negative template mold,
having an interconnected macroscopic porous crystalline structure,
that can withstand the high temperatures necessary to fill its
pores by CVD or other high temperature process with a material
having a high refractive index. Such CVD filling then creates a
positive final crystalline structure having both the desired
porosity and desired high refractive index contrast.
[0045] More particularly, FIG. 1 illustrates a method 100 for using
such a positive template polymeric structure, provided at step 110
and having interconnected macroscopic pores, according to the
present invention to produce first a negative template mold and
then a positive final crystalline structure having interconnected
macroscopic pores with high refractive index. Although FIG. 1
broadly relates to polymerizable glass precursors, the discussion
below is directed to the exemplary use of silicon or silicone
(silicone(e))-containing precursors.
[0046] According to this method, polymerizable silicon-containing
precursors are prepared at step 120. Such precursors can either
themselves be polymerizable, or they can be dispersed in a
polymerizable material. Polymerizable silicon-containing materials
broadly include silicon (e)-containing monomers and oligomers
having two or more functional groups for potential reaction to form
polymerization bonds. Monomers having only one potential reactive
site can also be bound into polymerizable materials if desired, but
they themselves cause chain termination. Polymerizable materials
may, if desired, contain only silicon, carbon, hydrogen and oxygen.
In general, the polymerizable material needs the capability of
polymerization to a solid state.
[0047] Although this discussion is directed to exemplary precursors
to formation of SiO.sub.2, it is to be understood that precursors
to other glass compositions can be substituted. For example,
precursors to a doped SiO.sub.2 composition, or to a glass
composition excluding silicon, can be used. The glass composition
to be produced preferably is thermally stable at least to about
500.degree. C. to facilitate CVD or other high temperature process
for deposition of the high refractive index material, and
preferably can be attacked and removed by etching with a suitable
solvent such as hydrofluoric acid (HF).
[0048] Suitable polymnerizable silicon-containing materials
include, for example, inorganic-organic silicon(e)-containing
monomers in which silicon is bonded directly or indirectly to
reactive organic groups. These monomers, when condensed together
into oligomers, form organic-inorganic siloxane materials.
Exemplary types of inorganic-organic silicon-containing monomers
include organically modified silicate (ormosil) monomers, and
silsesquioxane monomers. Ormosil monomers include or are typically
produced from trifunctional silicone alkoxides such as
tetraethylorthosilicate (TEOS), also known as tetraethoxysilane,
shown in Formula 1. ##STR1##
[0049] As shown in FIG. 2, organic modifications of TEOS may take
place at any one or more of its four reactive alkoxide ligands. In
FIG. 2, f is representative of the number of reactive alkoxy
ligands connected to the silicon atom, and n is representative of
the number of non-reactive organic moieties connected to the
silicon atom. R can be any saturated or unsaturated, linear or
branched, aliphatic or aromatic hydrocarbon moiety including, for
example, methyl or benzyl. For a mono-functional silicon alkoxide
in which fr=1 and n=3, the bound monomer terminates the polymer
chain since there is only one reaction site. A di-functional
silicon alkoxide in which f=2 and n=2, behaves as a bridging agent,
connecting monomers in a linear fashion. A tri-functional silicon
alkoxide in which f=3 and n=1, behaves as an crosslinker, allowing
for branching in the polymerization network. A tetra-functional
silicon alkoxide in which f=4 and n=0, also behaves as a networking
agent, allowing for maximized connectivity through all four
functional groups of the monomer. Ormosils are discussed here in
the context of exemplary reactive ethoxide ligands. However, it
will be recognized that other reactive groups, including other
alkoxide groups, can be used.
[0050] The second common exemplary type of inorganic-organic
silicon-containing precursors are silsesquioxanes. Exemplary
general structures of silsesquioxane-type precursors are shown in
FIG. 3. The R group is a hydrocarbon substituent that behaves as a
spacer unit residing between two reactive mono-, di- or
tri-alkoxysilane end groups. Exemplary spacer R groups are methyl
and benzyl groups, but other saturated or unsaturated, linear or
branched, aliphatic or aromatic hydrocarbon moieties can be used.
In one embodiment, partially condensed silsesquioxane oligomers
having an average molecular weight of about 500-20,000 grams per
mole are used. Suitable silsesquioxanes include GR630S,
commercially available from Techneglas, Perrysburg, Ohio. Further
background information is disclosed in Chemical Review, vol. 95,
pages 1409-1430, (1995), the entirety of which is hereby
incorporated by reference.
[0051] In a further embodiment, inorganic silicon-containing
compounds may be dispersed in a polymerizable material selected
from monomers and oligomers having two or more ligands for
potential reaction to form polymerization bonds. For example,
particulate SiO.sub.2 or other glass can be dispersed in such a
polymerizable material. The inorganic silicon-containing compound
is physically immobilized in the solid polymer upon its
polymerization.
[0052] Referring again to FIG. 1, at step 130 the pores of the
positive template polymeric structure are filled with the
polymerizable silicon-containing precursor material. In one
embodiment, the viscosity of such precursor material is low in
order to facilitate its infiltration into the positive template
polymeric mold. For example, the precursor material viscosity may
be less than 100 centipoises (CPS) at 25.degree. C. The viscosity
of such precursor material can also be reduced by raising its
temperature so long as such temperature does not prematurely induce
polymerization. Alternatively, pressure or a vacuum may be applied
to induce the precursor material to infiltrate the positive
template polymeric mold.
[0053] Next, the polymerizable silicon-containing precursor
material is polymerized to a solid state at step 140. Conventional
processes suitable for the chosen precursor material can be used.
Elevated temperatures, for example, about 60.degree. centigrade
(C.) to about 300.degree. C., or between about 100.degree. C. and
about 200.degree. C., maintained for about 1-4 hours, are typically
employed.
[0054] In one embodiment, ormosil monomers are polymerized by a
sol-gel process. FIG. 4 schematically illustrates this process as
applied to a tetra-alkoxysilane 410 in which group R represents,
for example, methyl or ethyl moieties. R can also be an aromatic
group such as, for example, a phenyl moiety. According to this
process, an ormosil monomer having two or more reactive alkyloxy
ligands 420, such as the illustrated tetra-afkoxysilane 410, is
mixed with water. Hydrolysis of alkoxy ligands results in
replacement of alkyl moieties by hydrogen. Accordingly, affected
alkoxy ligands are converted to hydroxyl groups 430 yielding
compound 440, and alkyl alcohols 450 are formed as a reaction
byproduct. Next, condensation reactions occur between such hydroxyl
groups 430 and further alkoxy ligands 420, yielding ormosil
oligomers 460 and further alkyl alcohol byproducts 450. FIG. 5
further illustrates the condensation of two ormosil monomers to
produce a silicate oligomer. As shown in FIG. 4, further cycles of
hydrolysis and polycondensation then occur, resulting in long chain
polymerization of the ormosil monomers yielding polycondensate 470.
If ormosil monomers having three or four reactive alkoxy ligands
are present, crosslinking through branched chains can also
occur.
[0055] In another embodiment, silsesquioxane monomers are
polymerized by a sol-gel process. Formula 2 illustrates a
silsesquioxane oligomer. Upon initial sol-gel network formation,
porosity is limited due to the presence of the organic groups.
Subsequent decomposition of the polymer yields a well defined
porous system including removal of the organic groups. R can be any
saturated or unsaturated, linear or branched, aliphatic or aromatic
hydrocarbon moiety including, for example, methyl or benzyl.
##STR2##
[0056] In another embodiment, PDMS oligomers are used. PDMS is
thermally cured or cured by an organometallic crosslinking
reaction. Referring to FIG. 6, vinyl-terminated siloxane base
oligomers 610 are provided in which n indicates, for example, about
60 siloxane moieties, and hydride functional siloxane oligomers 620
are provided in which n indicates, for example, about 10 siloxane
moieties. R is usually --CH.sub.3 and sometimes --H. The
crosslinking oligomers 620 each contain at least three silicon
hydride bonds, in which R represents hydrogen. The crosslinking
oligomers 620 also contain a dispersed platinum-based catalyst 630
that catalyzes the addition of the SiH bond across the vinyl groups
640, forming Si--CH.sub.2--CH.sub.2--Si linkages 650. For example,
the platinum catalyst may be a
platinum-divinyltetramethyldisiloxane complex having the formula
Pt1.5[(CH.sub.2.dbd.CH(CH.sub.3).sub.2Si].sub.2O). Such catalyst is
commercially available as SIP6831.1 from Gelest Inc., 612 William
Leigh Drive, Tullytown, Pa. The multiple reaction sites on both the
base oligomers 610 and crosslinking oligomers 620 allow for 3D
crosslinking. One advantage of this type of addition reaction is
that no waste products such as water are generated. If the ratio of
crosslinking oligomer 620 to base oligdmer 610 is increased, a
harder, more cross-linked elastomer results. One exemplary
commercially available product, including separate base and
crosslinking agents, is Dow Corning Sylgard elastomer.
[0057] High temperatures are not required for formation of these
polymers, so that the portions of the molecules of the organic
reagents not directly affected by the hydrolysis and condensation
reactions can be conserved during polymerization. Hence, organic
reagents can be chosen to provide additional control over the
flexibility of the resulting polymer. Monomers of lower
functionality can be included to allow flexible control over
chemical densification by creation of specific free volume or
porosity in the sol-gel network. Preferably, the polymer has
adequate flexibility to avoid cracking, and adequate rigidity to
maintain its structural integrity. Excessive flexibility can lead
to shrinkage of the structure upon decomposition of the organic
components.
[0058] The resulting polymerized negative template
silicon-containing material is embedded within the positive
template silicon-free organic polymer mold. Referring to FIG. 1,
oxidation of this structure at step 150 will decompose it and
eliminate all of the carbon and hydrogen, leaving behind the
silicon in the form of silicon dioxide, or SiO.sub.2. Such
oxidation can be carried out, for example, by applying to the
structure resulting from step 140 either an oxygen plasma, or an
elevated temperature in the presence of either oxygen or a mixture
of air and oxygen. The robust resulting negative template
structure, constituted entirely of refractory SiO.sub.2,
accordingly has a melting point above 1,000.degree. C. and can
easily withstand the temperatures necessary for its use as a mold
for CVD or other high temperature deposition of other materials
having a high refractive index, to yield a positive final
crystalline product structure having interconnected macroscopic
pores.
[0059] The process of filling the pores of the positive template
polymeric structure with silicon-containing precursors at step 130
typically leaves excess silicon deposited at the exterior surfaces
of the structure. Following removal of the positive template
polymeric structure by its decomposition at step 150, this excess
silicon in the form of SiO.sub.2 blocks the pores at the exterior
surface of the negative template structure. This excess SiO.sub.2
is preferably removed at step 160, for example by slightly etching
the exterior of the negative template structure. Any composition
effective for etching SiO.sub.2 can be used. In one embodiment, HF
diluted in ethanol or water to a weight percent of about 0.1-10%
can be used.
[0060] Referring to FIG. 1, the SiO.sub.2 negative template porous
structure is used as a template at step 170 for CVD or other high
temperature filling with a material having a high refractive index.
CVD is generally carried out by heating the material to be
deposited, to a selected temperature effective for its
vaporization. The vaporized material is then directed by suitable
means into the negative template structure in order to fill the
macroscopic pores. Alternatively, the desired material having a
high refractive index can be melted and filled into the porous
SiO.sub.2 negative template structure. Other suitable high
temperature or other processes for filling the negative template
structure with the high refractive index material may also be
used.
[0061] The refractive index of air is 1.0. The material used to
create the final positive crystalline structure preferably has a
refractive index relative to air of at least about 2.0, more
preferably of at least about 2.5, even more preferably of at least
about 3.0, and most preferably of at least about 3.4. The
difference between the refractive indices of air and the chosen
material accordingly is maximized, so that the final positive
structure including air-filled interconnected macroscopic pores
provides a high refractive index contrast. Where an etching
solution is used to dissolve out the negative template mold, the
high index material used to create the final positive crystalline
structure needs to be resistant to the solution. For example, in
the case of a SiO.sub.2 negative template mold, the deposited
positive final high index material preferably is resistant to HF
and ethanol. In one embodiment, the material to be deposited is
elemental silicon, having a refractive index of about 3.4 to about
3.6. In another embodiment, the material to be deposited is
elemental tellurium, having a refractive index of about 4.0 to
about 6.0. In further exemplary embodiments, the material to be
deposited is selected from gallium arsenide, gallium nitride,
indium phosphide, aluminum nitride, indium nitride, gallium
antimonide, indium antimonide, aluminum antimonide, aluminum
gallium nitride, aluminum gallium arsenide, aluminum gallium
antimonide, gallium aluminum antimonide, indium gallium antimonide,
gallium arsenic antimonide, indium gallium phosphide, indium
gallium arsenide, indium arsenic antimonide, indium gallium
arsenide phosphide, indium aluminum gallium arsenide, indium
aluminum gallium nitride, indium aluminum gallium antimonide, lead
sulfide, cadmium selenide, tin sulfide, cadmium sulfide, zinc
selenide, bismuth, and selenium. Other semiconductors, for example
those comprising two or more of the elements selected from the
group consisting of gallium, arsenic, indium, phosphorus, aluminum,
nitrogen, antimony, lead, sulfur, cadmium, selenium, tin, zinc, and
bismuth; or otherwise comprising Group 3, 4, or 5 elements, can
also be used.
[0062] Referring to FIG. 1, the negative template glass structure
is removed at step 180, leaving behind the positive final periodic
crystalline structure composed of a high refractive index material
containing interconnected macroscopic pores. According to preferred
embodiments of the present invention, the macroscopic pores of the
final periodic crystalline structure have a pore size between about
50 nanometers and about 10 microns. More preferably, the
macroscopic pores of the final positive periodic crystalline
structure have a pore size between about 100 nanometers and about 5
microns. In embodiments where the negative template structure is
constituted by SiO2, the negative template structure can be
completely removed, for example, by etching the structure with HF
diluted in ethanol or water to a weight percent of about 1-10%. In
the event that elemental silicon is the desired composition for the
final positive crystalline structure, HF etching is feasible due to
the greater resistance of elemental silicon to HF. Other
combinations of compositions for positive and negative template
structures will accordingly dictate suitable etching solutions.
[0063] FIG. 7 illustrates an alternative method 700 according to
the present invention. According to method 700, a negative template
periodic glass or oxide structure is produced that can withstand
high temperatures and having utility as a mold. Such negative
template structure can then be used for producing a positive final
crystalline structure composed of a material having a high
refractive index and having interconnected macroscopic pores. In
this embodiment, the negative template structure is directly
produced from a photosensitive medium, without the need to first
prepare a positive template polymeric mold for its fabrication.
Although FIG. 7 broadly relates to glass precursors, the discussion
below is directed to the exemplary use of silicon-containing
precursors.
[0064] According to this method, a starting medium comprising
silicon-containing precursors and including photosensitive reagents
is prepared at step 710. The silicon-containing precursors can
either themselves be photosensitive, or they can be dispersed in a
photosensitive material. Such photo-exposure induces either
photopolymerization or photodecomposition of the material. Mixtures
of silicon-containing photosensitive materials and other
photosensitive materials can also be used. The starting medium
further includes a photosensitive reagent capable of being
activated by electromagnetic radiation of a suitable wavelength,
such as visible light, that can thereby photo-chemically yield
reaction products themselves capable of directly or indirectly
inducing localized polymerization or decomposition of the starting
medium. This photosensitive reagent may itself also, if desired,
comprise silicon. Products of the photo-chemical reactions induce
localized polymerization of the starting medium, deprotection of
reactive sites, or crosslinking of functional groups of reagents in
the starting medium. The starting medium components are uniformly
dispersed in a suitable solvent, such as, for example,
tetrahydrofuran.
[0065] In one embodiment, a polymerizable silicon-containing
compound, oligomer or polymer is used for preparation of a starting
medium. Such an exemplary starting medium further includes a
non-nucleophilic solvent and a uniform density of dispersed
photo-sensitizer molecules and initiator complexes. The classes of
polymers to be produced from such polymerizable starting media can
include, for example, acrylates, methacrylates, random or block
copolymers, and epoxides. Other random or block copolymers can also
be used.
[0066] FIGS. 8A, 8B and 8C respectively show various exemplary
sensitizer compounds 810, 820 and 830. The sensitizer compounds
810-830 are xanthene dyes that are activated by visible light. In
FIG. 8B, "OBu" represents a t-butoxy group. The dyes 810 and 820
are optimally activated by light having wavelengths of about 535 nm
and 470 nm, respectively. These dyes 810 and 820 are available from
Spectra Group Limited, Inc., 1722 Indian Wood Circle, Suite H,
Maumee, Ohio 43537, under respective product names HNu-535 and
HNu-470. HNu-535, for example, is
2,4,5,7-tetraiodo-6-hydroxy-3-fluorone. The dye 830, known in the
art as Rose Bengal, is optimally activated by light having a
wavelength of about 560 nm and is commercially available from the
Aldrich Company, P. O. Box 2060, Milwaukee, Wis. 53201.
[0067] FIGS. 9A and 9B show exemplary ionic initiator complexes 910
and 920, which are photoacid generators (PAGs). The ionic initiator
complex 910 is diaryliodionium hexafluoroantimonate, and is
commercially available from Sartomer Inc., Oaklands Corporate
Center, 502 Thomas Jones Way, Exton, Pa. 19341, under the product
name SarCat.RTM. SR1012. The ionic initiator complex 920 is
commercially available from UCB Chemicals Corp., 2000 Lake Park
Drive, Smyrna, Ga. 30080, under the product name OPPI.
[0068] In one embodiment, silicon-containing multi-functional
acrylate, methacrylate, styrene or epoxy polymers having an average
molecular weight of about 5,000 to about 20,000 g/mole are
used.
[0069] Exemplary silicon-containing acrylate and methacrylate
polymers are shown in FIG. 10. In the illustrated formulas, R may
be, for example, CH.sub.3--, --CH.sub.2CH.sub.3, --H, --OCH.sub.3,
--OCH.sub.2CH.sub.3, --OSi(R).sub.3, -phenyl, or a higher
functional hydrocarbon optionally with further silicon
substituents. X is 0, 1, 2, 3 or a higher integer. P is an acid
protection group such as t-butyl, tetrahydropyranyl,
trimethylsilane, or hexamethyldisilane. The variables n and m
represent the relative proportions of the indicated moieties
present, each ranging from 0-100% as desired.
[0070] Exemplary silicon-containing styrenic polymers are shown in
FIG. 11. In the illustrated formulas, R may be, for example,
CH.sub.3--, --CH.sub.2CH.sub.3, --H, --OCH.sub.3,
--OCH.sub.2CH.sub.3, --OSi(R).sub.3, -phenyl, or a higher
functional hydrocarbon optionally with further silicon
substituents. X is 0, 1, 2, 3 or a higher integer. P is an acid
protection group. The variables n and m represent the relative
proportions of the indicated moieties present, each ranging from
0-100% as desired.
[0071] In another embodiment, silicon-containing multi-functional
epoxy, acrylate, methacrylate, or styrene oligomers having an
average molecular weight of about 200 to about 1,000 g/mole are
used.
[0072] Suitable exemplary silicon-containing epoxy oligomers are
shown in FIG. 12. In the illustrated formulas, R may be, for
example, CH.sub.3--, --CH.sub.2CH.sub.3, --H, --OCH.sub.3,
--OCH.sub.2CH.sub.3, --OSi(R).sub.3, -phenyl, or higher functional
hydrocarbon optionally with further silicon substituents. M, n, p
and p' each independently are 0, 1, 2, 3 or a higher integer. One
suitable epoxy oligomer derivative of a bisphenol-A novolac is
commercially available from Resolution Performance Products, 1600
Smith Street, 24th Floor, P.O. Box 4500, Houston, Tex. 77210-4500,
under the product name EPONTM SU-8.
[0073] Exemplary silicon-containing acrylate and methacrylate
oligomers are shown in FIG. 13. In the illustrated formulas, R may
be, for example, CH.sub.3--, --CH.sub.2CH.sub.3, --H, --OCH.sub.3,
--OCH.sub.2CH.sub.3, --OSi(R).sub.3, -phenyl, or higher functional
hydrocarbon optionally with further silicon substituents. N is 1,
2, 3 or a higher integer. P and p' each independently are 0, 1, 2,
3 or a higher integer.
[0074] In yet another embodiment, polymerizable silicon-containing
monomers are used. In general, the silicon-containing monomers
discussed earlier are suitable. In another embodiment, an inorganic
silicon-containing compound such as ground SiO.sub.2 is dispersed
in a starting medium.
[0075] Referring next to step 720 of FIG. 7, a volume of the
starting medium is prepared, for example by spin coating or casting
a film of the starting medium on a rigid transparent substrate.
After the spin coating or casting, the film is prebaked to
evaporate out the solvent, at a temperature too low to induce
substantial polymerization or deprotection.
[0076] In one embodiment, two separate steps 730 and 740 are next
provided to define the polymerized regions of the negative template
structure. In the first of these steps 730, an optical interference
pattern produced by combining a plurality of mutually coherent
beams of light of a particular wavelength exposes the starting
medium, catalyzing localized photo-chemical reactions without
causing polymerization or deprotection and hence without causing
significant changes to the starting medium's refractive index. This
exposure can be controlled to deliver, for example, about 1 joule
per square centimeter for about 1 second. The exposure step 730 is
carried out at a moderate temperature that inhibits or prevents
polymerization. In the second step 740, a heat treatment induces
generation of a catalyst which then initiates a localized chemical
reaction in the starting medium, thereby polymerizing or
deprotecting and thereby changing the refractive index of the
starting medium in the localized regions that were exposed by the
optical intensity pattern. The catalyst may be, for example, highly
acidic free H+ ions produced by the earlier exposure. Since the
first step 730 does not initiate refractive index changes, the
initial clarity of the negative template pattern in the starting
medium is not degraded during the exposure step 730.
[0077] FIG. 14A shows a setup 2 for producing the optical intensity
pattern needed in exposure step 730, with multiple laser light
beams 3, 4, 5, and 6. The beams 3-6 are mutually coherent and thus
coherently interfere to produce an optical pattern with periodic
intensity variations in three independent directions. Exemplary
beams 3-6 pass through (1,1,5), (1,5,1), (5,1,1), and (3,3,3)
directions in a coordinate system where the starting medium 7 is
located at the coordinate origin (0,0,0).
[0078] Referring to FIG. 14B, to ensure that the 3D optical
intensity pattern is defined by the four beams 3-6, the starting
medium 7 and supporting substrate 8 are preferably disposed between
thick transparent glass prisms 9. The prisms 9 are shaped and
positioned to stop reflected light rays 3.sub.R-6.sub.R from
returning to the starting medium 7 and thereby further exposing the
starting medium 7 in unintended regions. Alternatively, the
refractive indices of the starting medium 7 and the supporting
substrate 8 may be matched, and a film of hydrocarbon oil can be
interposed between the starting medium 7 and the supporting
substrate 8 to prevent reflection by the substrate.
[0079] Referring to FIG. 7, the exposed regions of the starting
medium are then polymerized in step 740. FIG. 15 shows exemplary
photo-reactions 30 that are initiated by exposing the starting
medium to light with a wavelength of 514 nm. The exemplary starting
medium includes the sensitizer compound 810 shown in FIG. 8A and
the ionic initiator complexes 910 shown in FIG. 9A.
[0080] The reactions start with the production of an activated
molecule of dye* by absorption of a photon of 514 rn light by the
molecule of dye, at reaction 32. The activated molecule of dye*
transfers energy to an ionic initiator complex,
Ar.sub.2I.sup.+X.sup.-, to produce an activated ionic initiator
complex, Ar.sub.2I.sup.+X.sup.-*, shown as reaction 34. The
activated ionic initiator, Ar.sub.2I.sup.+X.sup.-*, subsequently
decays to produce a pair of free radicals ArI.sup.+X* and
Ar.sup.-*, shown as reaction 36. One of the free radicals,
ArI.sup.+X.sup.-*, reacts with a hydrogen atom in a solvent
molecule indicated as Sol-H, to produce a complex ArI.sup.+XH and a
free radical, Sol*, shown as reaction 38. The complex ArI.sup.+XH
is unstable and decays to produce a highly reactive acidic free
hydrogen cation H+, shown as reaction 40. The sequence of
photo-reactions 30 thermodynamically favors rapid production of H+
ions. These ions are then available for reaction with polymerizable
reagents as discussed with regard to step 740.
[0081] During exposure step 730 of FIG. 7, the temperature of the
starting medium is kept relatively low to prevent and inhibit
polymerization reactions between polymerizable reagents.
Temperatures that are lower than the glass-like to rubber-like
transition temperature of the selected starting medium, for
example, below 65.degree. C., are low enough for this purpose.
Below the glass transition temperature, there is little free volume
for polymerizable oligomers to perform rotational movements that
are needed for their polymerization. At such temperatures, the
H+-ions produced by the photo-reactions 30 of FIG. 15 are also
thereby prevented from diffusing through the starting medium. As a
result, the H+-ions do not significantly catalyze polymerization of
the polymerizable oligomers at such temperatures, and such ions
remain locally distributed in the regions of the starting medium
positively exposed by the optical interference pattern.
[0082] FIG. 16 shows an exemplary sequence of polymerization
reactions 42 that are catalyzed by H+-ions in a starting medium
that includes the epoxide oligomers 21 shown in FIG. 12. The
polymerization reactions are favored under the elevated temperature
conditions present during polymerization step 740 in FIG. 7. A
cycle of the sequence of polymerization reactions 42 starts when an
H+-ion attacks the epoxide ring 52 of an oligomer 54 shown as
reaction 44. After being protonated, the oxygen of the epoxide ring
52 undergoes nucleophilic attack by a hydroxide moiety of R--OH
molecule 56, shown as reaction 46. Exemplary R--OH molecules are
trace alcohol or water molecules absorbed in the starting medium.
The nucleophilic attack produces a complex 58 in which the R-moiety
from the second molecule is chemically bonded to the oxygen of the
original epoxide ring 52. Subsequently, the complex 58 decomposes
to regenerate a free H+-ion and an alcohol molecule 60 with a
hydroxide moiety, shown as reaction 48.
[0083] The sequence of polymerization reactions 42 generates
molecule 60, which has a hydroxide moiety, and regenerates an
H+-ion. These two molecular entities are available-for the next
cycle of the sequence of reactions 42. In the next cycle, the R--OH
molecule is the molecule 60 produced by the previous cycle. A
single H+-ion is thus capable of catalyzing many cycles of the
sequence of reactions 42. Each cycle adds an additional oligomer to
molecule 60 from the last cycle, thereby producing a growing
polymer.
[0084] In each cycle, the protonated epoxy ring 52 of an oligomer
54 must properly align with the growing polymer, and the acidic H+
must be close to the epoxy ring for reactions 46 and 44 to proceed.
To produce the alignment, chain segments of the oligomers need
great freedom to perform rotational motions. Such rotational
motions are possible when the starting medium is in a rubber-like
phase, that is when the temperature is above the temperature of
transition to such a phase. In the rubber-like phase, there is more
free volume in the starting medium, and the acidic H+ cations are
able to move close to the epoxy ring. Thus, polymerization only
proceeds at temperatures significantly above the glass-rubber
transition temperature.
[0085] Referring again to FIG. 7, method 700 exposes the starting
medium at step 730 to three or four interfering light beams to
produce the optical intensity pattern. The three or four light
beams are not all coplanar,.as best seen in FIG. 14A. Thus, the
three or four light beams do not have completely coplanar
polarizations. Due to the absence of coplanar polarizations, the
beams produce patterns with nonzero constant background optical
intensities when combined. If a starting medium is exposed with
such a combination of light beams, the constant background optical
intensity will produce a constant background polymerization
structure, upon polymerization at step 740. Constant background
polymerization structures are typically undesirable, because they
are closed crystalline structures without interconnected pores. It
is not possible to fill the pores of a negative template closed
porous structure in order to produce a positive final crystalline
structure, because the pores are not interconnected.
[0086] In order to avoid the nonzero background polymer density
that results from nonzero constant background optical intensities
generated in step 730, an appropriate concentration of neutralizer
molecules preferably is dispersed in the starting medium prepared
at step 710. The initial concentration of the neutralizer molecules
is selected to be sufficient to neutralize the background density
of polymerization catalyst, such as acidic free H+ ions, that will
be produced by the multi-beam interference pattern, without
completely stopping the polymerization reaction. If the
polymerization catalyst is an acid such as H+ ions, a base such as
triethyl amine or N,N,2,4,6-pentaniethylaniline is an appropriate
neutralizer.
[0087] At step 740, the exposed starting medium is heated under
conditions that favor polymerization of the starting medium in the
vicinity of the acidic H+ cations. This step is completed during
the considerable time period during which the density pattern of
reaction products resulting from step 730 continues to track the
original light intensity pattern. Exemplary heating steps involve
baking the exposed starting medium at a temperature higher than a
temperature at which the.starting medium 7 makes a transition from
a glass-like phase to a rubber-like phase. For example, the exposed
starting medium can be heated to a temperature between about
60.degree. C. and about 300.degree. C., or preferably between about
100.degree. C. and about 200.degree. C., for a time period of about
1 minute to 4 hours.
[0088] In the higher temperature rubber-like phase, rotational
motions by oligomers or polymers of the starting medium 7 and by
groups of said oligomers or polymers significantly increase. This
increased motion produces larger free volumes in the starting
medium 7 and allows polymerization reactions to proceed. The high
temperatures and larger free volumes enable the reaction products
that were earlier produced when the sensitizer molecules were
exposed to the optical intensity pattern, to catalyze or stimulate
polymerization or crosslinking reactions. Thus, the distribution of
polymerized oligomers tracks the original optical intensity pattern
that exposed the starting medium. These polymerized products of the
catalyzed or stimulated reactions change the refractive index of
the starting medium in regions where the polymerizations occur.
Thus, polymerization step 740 produces a refractive index pattern
with a 2D or 3D crystalline symmetry in the exposed starting
medium.
[0089] In another embodiment, step 730 desirably includes,
following exposure to the optical interference pattern, the use of
a focal region of a converging light beam to sequentially expose
one or more desired points and lines in the starting medium. The
light of the converging beam for this desirable additional
treatment has a longer wavelength than the light used in the
interference pattern exposure of step 730. The longer wavelength
enables the converging light beam to only activate photo-sensitizer
molecules through two-photon or other multiple-photon processes.
Such processes only occur at significant rates in the high
intensity focal region of the converging light beam. For this
reason, the converging beam functions as a writing instrument that
enables exposing small points and fine lines with sub-wavelength
resolution in the starting medium. Such points and lines induce
corresponding defects in the periodic pattern of the final positive
macroscopically porous crystalline structure. When the final
product is used as a band gap structure, such predetermined defects
can serve as optical channels for transporting optical signals
through the structure or as point-defects for providing couplings
between various optical channels.
[0090] Referring now to step 750 of FIG. 7, the starting medium is
then washed with a solvent that dissolves away the unpolymerized
portions of the starting medium. Any solvent suitable to dissolve
the unpolymerized reagents but to which the polymerized structure
is resistant, can be used. A suitable solvent for epoxide reagents,
for example, is propylene glycol methyl ether acetate. This step
effectively replaces unpolymerized reagents with solvent in the
positive regions of the crystalline structure. The washing may be
facilitated by agitation, for example by using an ultrasonic
bath.
[0091] The washed polymerized structure preferably is then dried.
The drying conditions are selected to limit surface-tension
stresses that the solvent applies to the porous polymeric structure
during drying. Such surface-tension stresses could otherwise
destroy the fragile macroscopic porous crystalline structure. One
method for limiting such internal stresses during drying takes
advantage of critical properties of liquids. In this method, the
absorbed solvent is first replaced by liquid carbon dioxide.
Replacing the solvent involves placing the washed polymerized
structure in liquid carbon dioxide under pressure adequate to
maintain the liquid state. The liquid carbon dioxide diffuses into
the structure thereby replacing absorbed solvent without drying the
porous structure. After replacing the solvent with liquid carbon
dioxide under pressure, the temperature of the porous structure is
raised to carbon dioxide's critical point at 31.06.degree. C., and
the pressure is gradually reduced to atmospheric pressure. The
carbon dioxide is then allowed to diffuse out of the porous
structure. Since carbon dioxide has no surface tension at its
critical point, the diffusion of carbon dioxide out of the
interconnected porous structure does not produce internal
surface-tension stresses.
[0092] Referring to FIG. 7, decomposition of this structure by
oxidation at step 760 eliminates all of the carbon and hydrogen,
leaving behind the silicon in the form of silicon dioxide, or
SiO.sub.2. Such decomposition can be carried out in the same manner
as discussed in connection with step 150 of FIG. 1. The resulting
structure is constituted by negative template porous Sio.sub.2 as a
reversed structure of the desired positive final crystalline
structure having interconnected macroscopic pores with high
refractive index contrast. As indicated at step 770, the external
pores of the structure are preferably then opened in the same
manner as discussed in connection with step 160 of FIG. 1.
[0093] This negative template structure can be utilized to produce
a positive final porous structure with a high refractive index
contrast. As indicated by steps 780 and 790 of FIG. 7, this may be
carried out in the same manner as discussed above in connection
with steps 170 and 180 of FIG. 1. The resulting product is a
positive final crystalline structure composed of a high refractive
index material containing interconnected macroscopic pores.
[0094] The final positive crystalline structure is useful for a
variety of photonic applications, including waveguides, splitters,
optical insulators, optical filters, microcavity lasers, optical
switches, routers, and other photonic band gap applications.
Particular design considerations for photonic band gap materials
are known in the art, as reflected for example in J. D.
Joannopoulos et al., Photonic Crystals, Princeton University Press
(1995), the entirety of which is hereby incorporated by reference.
See also, Russell, P. S. J., "Photonic Band Gaps," Physics World,
Vol. 37, August 1992; Amato, I., "Designing Crystals That Say No to
Photons," Science, Vol. 255, p. 1512 (1993); and U.S. Pat. Nos.
5,600,483 and 5,172,267, the entireties of which are hereby
incorporated by reference. Other applications include filters,
catalysts, and biocompatible materials.
[0095] While the present invention is disclosed in the context of
presently preferred embodiments, it will be recognized that a wide
variety of implementations may be employed by persons of ordinary
skill in the art consistent with the above discussion and the
claims which follow below.
* * * * *