U.S. patent number RE28,664 [Application Number 05/546,293] was granted by the patent office on 1975-12-23 for single material optical fiber structures including thin film supporting members.
This patent grant is currently assigned to Bell Telephone Laboratories, Incorporated. Invention is credited to Stewart Edward Miller.
United States Patent |
RE28,664 |
Miller |
December 23, 1975 |
Single material optical fiber structures including thin film
supporting members
Abstract
Optical fibers, for propagating optical radiation in guided
modes, are fabricated in an integral structure. Advantageously, the
fiber structure is made of a single filamentary material, such as
fused silica, with a relatively large cross section at the central
portion of the fiber and with a relatively thin film portion at the
extremities of the fiber. The thin film portion .Iadd.has a
thickness larger than the wavelength of the optical radiation to be
propagated and .Iaddend.serves as a supporting member for the
central portion of the fiber. Such optical fiber structures are
capable of propagating either single mode or multimode guided
optical waves. In addition, the exposed surface of the central
portion (which is not contacted by the thin film supporting member
portion) can be contacted with an optically nonlinear material, in
order to provide suitable interactions with the propagating signal
wave energy and thereby to produce electrooptic effects such as
amplification, modulation, or laser action.
Inventors: |
Miller; Stewart Edward (Locust,
NJ) |
Assignee: |
Bell Telephone Laboratories,
Incorporated (Murray Hill, NJ)
|
Family
ID: |
26976474 |
Appl.
No.: |
05/546,293 |
Filed: |
February 3, 1975 |
Related U.S. Patent Documents
|
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
Issue Date |
|
Reissue of: |
308833 |
Nov 22, 1972 |
03813141 |
May 28, 1974 |
|
|
Current U.S.
Class: |
385/125;
385/122 |
Current CPC
Class: |
G02B
6/02 (20130101); C03B 2203/12 (20130101); C03B
2203/10 (20130101); C03B 2203/14 (20130101) |
Current International
Class: |
G02B
6/02 (20060101); G02B 005/14 () |
Field of
Search: |
;350/96WG,96R |
References Cited
[Referenced By]
U.S. Patent Documents
Foreign Patent Documents
Primary Examiner: Corbin; John K.
Attorney, Agent or Firm: O'Konski; Thomas C.
Claims
What is claimed is:
1. An optical fiber structure for waveguiding optical radiation
which comprises a filament of a unitary optically transparent
structure whose cross section is characterized by a relatively
thick .[.cross section.]. .Iadd.cross-sectional area portion of a
fiber optical material at a central portion thereof and by
relatively thin cross-sectional area portions of the same material
contacting at least two extremities of the central portion, said
thin portions .Iadd.having thicknesses larger than the wavelength
of the optical radiation and .Iaddend.providing mechanical support
for the central portion, and said thin portions extending in a
direction away from the central portion for distances which are at
least an order of magnitude larger than the wavelength of the
optical radiation, whereby the optical radiation can be propagated
in at least one mode through the optical fiber.
2. An optical fiber structure according to claim 1 in which the
central portion has a rectangular cross section.
3. An optical fiber structure according to claim 1 in which the
central portion has a circular cross section.
4. An optical fiber structure according to claim 1 in which the
filament ismade of a single transparent material.
5. An optical fiber structure according to claim 1 in which the
exposed edges of the filament, located distally from the central
portion, are attached to a peripheral cylindrical hollow second
filament which is transparent to the optical radiation.
6. The optical fiber structure recited in claim 5 in which the
cylindrical portion is encased in an optically lossy jacket.
7. The optical fiber structure recited in claim 5 in which at least
a portion of the space between the peripheral cylindrical portion
and the central portion is occupied by an optically nonlinear
material for interaction with the optical radiation propagating
through the optical fiber.
8. The optical fiber structure recited in claim 1 in which at least
a portion of the exposed surface of the central portion is coated
with an optical nonlinear material for interaction with the optical
radiation propagating through the optical fiber.
9. The optical fiber structure recited in claim 1 in which at least
a portion of the exposed surface of the central portion is coated
with an optically linear material in order to provide further
waveguiding of the optical radiation propagating through the
fiber.
10. An optical fiber structure according to claim 1 in which the
central portion has a rectangularly shaped cross section, and in
which one major surface of each of the thin portions together with
the one major surface of the central portion form a single planar
surface.
Description
FIELD OF THE INVENTION
This invention relates to the field of optical communications
systems and, more particularly, to optical fiber structures for the
propagation of electromagnetic wave enenrgy.
BACKGROUND OF THE INVENTION
In the prior art, optical waveguides in the form of optical fibers
have been utilized for the propagation of optical wave energy in a
single mode or in multimodes from one location to another. A basic
problem arises in these optical fibers in connection with the
attachment of supporting members which are required for mechanical
support uniformly along the whole length of the optical fibers. In
particular, at the juncture of the supporting member with the
optical fiber, an optical disturbance or perturbation is introduced
in the modes of wave energy being transmitted through the fiber.
This perturbation causes various problems in the propagation of the
electromagnetic signal wave energy, such as the unwarranted
conversion of signal energy from one mode to another mode with
consequent distortion problems. Another problem arises from the
need for "cladding" material surrounding the central core of
optical fibers, in order to keep foreign materials (such as dust)
from contacting the central core and thereby causing further
undesired perturbations of the optical modes propagating through
the core. Such cladding must ordinarily be made of a material
having a lower optical refractive index than that of the core. For
such desirable core materials as fused silica, it is difficult to
find such suitable cladding which has a lower refractive index than
the core and, at the same time, presents sufficiently low optical
absorption loss to make it commercially attractive. Moreover,
ordinarily this "cladding" material obstructs any coating of the
optical fiber core with various optical materials which could serve
to provide interaction with the signal wave energy propagating
through the optical fiber, it would, therefore, be desirable to
have available ana optical transmission fiber which is supported in
such a way that the cladding of the fiber does not introduce the
losses and obstructions of the prior art.
SUMMARY OF THE INVENTION
In accordance with this invention, an optical fiber, together with
transparent supporting members, is made in a unitary integral
structure, advantageously formed of a single material. The central
portion of the fiber is relatively thick, and is joined at two or
more edges by thin film supporting members of the same material as
that of the central portion. Such a structure can be designed to
operate in either single mode or multimode propagation of the
optical waves through the fiber. In addition, the exposed periphery
of relatively thick central portion can be contacted in whole, or
at any desired part, by various linear or nonlinear optical
materials, in order to afford linear or nonlinear interaction with
the optical wave energy propagating through the fiber. In this way.
further mode control can be provided, or various devices such as
lasers, amplifiers and modulators can be integrated into the
optical fiber structure.
In a particular embodiment of this invention, a fused silica
optical fiber is fabricated with a rectangular cross section in the
central region together with a pair of thin film fused silica
portions contiguous with two opposite surfaces of the rectangular
cross section. The thickness of the thin film portions of the
optical fiber, which furnish mechanical support for the central
portion, is made relatively large compared with the propagating
optical wavelength, in order to provide sufficient mechanical
strength for supporting the central portion of the optical
waveguide fiber. On the other hand, the dimensions of the
rectangular cross section of the central portion of the fiber are
made larger than the thickness of thin film supporting members, in
order that the optical signal wave energy propagating through the
overall optical fiber structure is confined to the central portion
thereof by reason of waveguiding properties of the structure. The
exposed extreme edges of the thin film supporting members are
advantageously fused to a fused silica glass cylinder which, in
turn, is coated with optically absorbing material. In this way,
unwanted "slab guide" modes, which are not exponentially decreasing
in intensity in the thin films (going away from the central
portion) and which thereby "leak" through the supporting members to
these edges, are absorbed by the coating.
Fiber structures of this invention can be fabricated from an
original fused silica fiber of geometrically similar but much
larger cross section than the final desired structure. The original
fiber is cleaned, heated and drawn (stretched) in the longitudinal
direction, in order to reduce the dimensions of the original cross
section to the desired relatively small final cross section. The
exposed portion of the periphery of the central region of the final
optical structure can be, if desired, then contacted at various
locations, uniformly or (spatially) periodic or nonperiodic, with
various optically linear or nonlinear materials. Thereby, these
materials can provide suitable linear or nonlinear interaction
phenomena along the fiber between the optical wave energy
propagating through the central portion of the fiber and the linear
or nonlinear material. In this way, the useful electromagnetic
signal wave energy propagating through the optical fiber is
confined to the central rectangular portion of the fiber; and, at
the same time, access for optically linear or nonlinear materials,
to interact with the propagating signal wave energy, is afforded in
the optical fiber structure of this invention.
BRIEF DESCRIPTION OF DRAWING
This invention together with its features, advantages and objects
can be better understood from the following detailed description
when read in conjunction with the drawing, in which:
FIG. 1 is a longitudinal diagram, partly in cross section, of an
optical fiber structure in accordance with a specific embodiment of
the invention.
FIG. 2 is a cross-sectional view of the optical fiber structure
shown in FIG. 1;
FIG. 3 is a cross-sectional view of the optical fiber structure
shown in FIG. 2 in an initial stage of its manufacture; and
FIG. 4 is a cross-sectional view of an optical fiber structure in
accordance with another specific embodiment of the invention;
and
FIG. 5 is a cross-sectional view of an optical fiber structure
having a circular central portion, in accordance with yet another
specific embodiment of this invention.
DETAILED DESCRIPTION
An optical fiber filament structure 10 (FIG. 1) includes a
transparent central portion 11, and a pair of supporting
transparent thin film portions 12.1 and 12.2 (FIG. 2). The central
portion 11 together with the supporting films 12.1 and 12.2 are
located in a cavity provided by a peripheral hollow cylindrical
portion 13. An optically lossy jacket 14 advantageously encases the
cylindrical portion 13. Advantageously, the central portion 11 and
the thin films 12.1 and 12.2 are all made of the same optically
transmitting material. The peripheral portions 13 typically is
likewise made of the same material as the central portions of the
thin film supporting members 12.1 and 12.2. An optical source 21
and an optical utilization means 22 are located at opposite
longitudinal ends of the optical fiber structure 10 (FIG. 1).
If desired for nonlinear optical interaction with the optical waves
propagating through the fiber structure 10, the space cavity region
15 between the peripheral portion 13 and the central portion 11 may
be filled with an optically nonlinear material, typically a liquid.
In the alternative, the entire exposed surface of the central
portion 11, or various portions of said surface, may be coated with
optically nonlinear material as desired. In this way, the optical
wave energy, propagated from the source 21 to the utilization means
22 through the optical fiber structure 10, will advantageously
interact with this nonlinear material. In addition or as an
alternative, the entire surface, or various portions thereof, of
the central portion 11 may be coated with a linear optical
material, in order to furnish further waveguiding of the optical
wave energy propagating through the central portion 11.
The thickness of the supporting thin films 12.1 and 12.2, indicated
by b in FIG. 2, is fabricated advantageously to be larger than the
propagating optical wavelength furnished by the source 21, in order
to provide mechanical support for the central portion 11. In
addition, the width of the supporting thin films 12.1 and 12.2 in
the X direction is at least an order of magnitude larger than the
wavelength, in order to provide sufficient space for the
exponential decrease of the amplitude of the optical modes in
direction in the .+-.X direction going away from the central
portion 11. Moreover, it is important that the thickness B of the
central portion 11 should be larger than the thickness b of the
thin film portions, in order to provide the desired optical
waveguiding. In this way the useful modes propagating through the
optical structure 10 will be exponentially decreasing with distance
in the -X direction in the supporting member 12.1, and also
exponentially decreasing in the +X direction in member 12.2 (i.e.,
in the directions going away from the central portion 11). The
other optical modes, which are exponentially increasing or periodic
in these respective directions in the supporting members 12.1 and
12.2 ("slab guide modes" ), are not useful in this invention; and
these "slab guide" modes are quickly absorbed by the optically
absorbing material of the outer jacket 14 upon their propagation
through the optical fiber 10 in the Z direction.
In order to fabricate the optical fiber structure 10 shown in FIGS.
1 and 2, it is convenient to start with optically polished fiber
optic segments 31.1, 31.2, 32 and 33 in the fiber structure 30, as
shown in FIG. 3. Typically, all of these segments are made of the
same optically transmitting material such asa fused silica. In
order to have clean optical surfaces, the exposed surfaces of these
segments are cleaned successively with solutions of
trichlorethylene, acetone, nitric acid (1:1 diluted with deionized
water) and deionized water. Alternatively, the known hot fire flame
cleaning technique may be used to clean the surface.
Advantageously, the overall cross section of the segments in the
structure 30 as initially arranged (FIG. 3) constitutes a
geometrically similar, but greatly enlarged, cross section of the
finally desired cross section shown in FIG. 2. These segments 31.1,
31.2, 32 and 33 are heated to a temperature sufficient to fuse them
together and to enable them to be drawn (stretched) in the
longitudinal Z direction, in order to reduce their cross sections
to the finally desired value for the optical fiber 10. Thus, it is
to be understood that FIG. 3 is not drawn to scale with respect to
FIG. 2, but that ordinarily the structure 30 is many times larger
in cross section than the structure 10.
It should be understood that the relative values of A, B, and b
determine the number of modes which can be guided by the optical
fiber structure 10 (substantially independent of optical
wavelength). The mode supporting efficiency e of guidance of the
optical fiber structure 10 is defined as the ratio of this number
of possible guided modes to the number of possible modes which can
be guided by a similar optical fiber structure but with b = 0.
EXAMPLE 1
(Multimode Fiber):
An optical fiber structure, with a mode efficiency e of about 10
percent or more, can be afforded by the following choices of
parameters. The material for the optical fiber members 11, 12.1,
12.2 and 13 is selected to be of used silica (refractive
.[.inded.]. .Iadd.index.Iaddend., n = 1.46), for propagating
optical radiation from the source 21 (wavelength of about one
micron) to the optical detector 22. The space 15 is filled with air
or vacuum (n = 1.00) and the thickness b is selected to be about
1.4 micron .[.or less.]. . For propagating a suitable number of
optical modes (multimodes), the dimensions of A and B are typically
selected to be at least several times larger than b, but are
otherwise arbitrary. For example, A and B can be selected in the
range of about 5 to 25 micron. It should be remarked, however, that
there is an advantage of using a square cross section (A = B),
namely, that splicing is made easier in that the unavoidable minor
alignment errors in any splicing procedures are not so crucial as
for other cross sections (in which A and B are not equal).
EXAMPLE 2
(Multimode Fiber):
For an optical fiber structure 10 with an optical dispersion of no
more than 10 manoseconds per kilometer, the following design can be
used. Again, as in Example 1, the optical fibers 11, 12.1 and 12.2
are all made of fused silica (n = 1.46); the space 15 is filled
with air or vacuum (n = 1.00); and the optical source 21 provides a
beam of radiation having a wavelength of about 1 micron. For this
case, b is selected to about 5.4 microns, in order to achieve the
desired multimode operation with the desired dispersion. This
choice of parameters will then also provide a numerical aperture
("N.A.") of 0.065 radians in the optical fiber structure 10, and a
tolerable radius of curvature ("R") of approximately 19
millimeters. By "numerical aperture" is meant the maximum angle of
obliqueness in the optical propagation vector which will be
radiated from the output end of the fiber, or which will be
accepted by the fiber at the input end; and by "tolerable radius of
curvature" is meant the minimum radius of curvature for the fiber
(going around bends for example) consistent with losses below one
percent per centimeter of length. Typically, for optical
propagation in a suitable number of multimodes, both A and B are
selected to be at least several times larger than b, for example,
in the range of about 25 to 50 micron.
In this multimode case (for which b = 5.4 micron) moreover, the
cross section of the central portion alternatively can be circular
as indicated in FIG. 5, thereby providing a cross section which is
easier and less critical for splicing one longitudinal section of
the optical fiber with the next adjacent section. The diameter of
this central portion 51 can be about 75 micron, and the periphery
portion 53 can have an inner diameter of 100 micron and an outer
diameter of about 150 micron. Supporting films 52.1 and 52.2, as
well as an optically lossy jacket 54, serve the same function as
the films 12.1 and 12.2 and the lossy jacket 14 in FIG. 2. A
portion of the exposed surface of this central portion can be
coated with an optically linear material 56 in order to provide
further waveguiding of the optical radiation propagating through
the fiber. In addition, or alternatively, a portion of the exposed
surface of this central portion can be coated with an optically
nonlinear material for interaction with the optical radiation
propagating through the fiber.
It should be emphasized that the above Examples 1 and 2 provide
multimode optical propagation. For some fiber optical system
applications, however, single mode propagation may be desired. In
single mode operation, as known in the art, only the fundamental
modes (with both polarizations) are propagated by the optical
waveguide, which just cuts off for the next ("second") higher order
mode. A further desirable, though not necessary, condition for
single mode operation is that the thickness b (in the Y direction)
is much greater than the wavelength of the fundamental wave energy,
just as for multimode operation as described in Examples 1 and 2.
The cutoff condition, that is the requirement for single mode, is
given by the relation ##EQU1## where the quantity .beta..sub.x2
refers to .beta..sub.x for the "second" order mode, in which
.beta..sub.x is zero outside the range x = .+-. A/2, and in which
.beta..sub.x inside this rang is given by the solution of the
following simultaneous equations with "unknowns" .beta..sub.x,
k.sub.a, and k.sub.d.
k.sup.2.sub.a + k.sup.2.sub.d = (.epsilon..sub.r
-1).beta..sup.2.sub.o(2)
tan(k.sub.d a/2) =.epsilon..sub.r k.sub.a k.sub.d (3)
k.sup.2 .sub.a = .beta..sup.2.sub.x - .beta..sup.2.sub.o +
.beta..sup.2.sub.y. (4)
In Equations (2) - (4), .lambda..sub.o and .beta..sub.y are given
by
.beta..sub.o = 2.pi./.lambda..sub.o (5)
.beta..sub.y = .pi.16 (6)
where .lambda..sub.o is the vacuum wavelength of the optical
radiation and where .epsilon..sub.r is the corresponding dielectric
constant of the optical fiber portion 11 relative to the region
15.
Whereas Equations (2) - (6) apply only to the rectangular cross
section in the range x = .+-. A/2, Equation (1) is perfectly
general for any cross section provided the contour in the region x
=.+-. A/2 is sufficiently slowly varying so that substantially no
optical reflection occurs whithin this region (except at the
extremities). In this general case, the .beta..sub.x2 in Equation
(1) is the transverse wave number of the "second" order mode.
It should be noted that for single mode operation, asymmetrical
configurations (in the y direction) for the optical fibers in this
invention are preferred. As illustrated in FIG. 4, an asymmetrical
optical fiber 40 is illustrated which can advantageously be
produced, as previously indicated in connection with FIG. 3, except
that central segment 31.2 is omitted entirely. In the embodiment
illustrated in FIG. 4, it should be noted that the bottom surface
of the optical fiber 40 is completely planar and has the advantage
of fewer intial segments in the fabrication process, as well as the
advantage of simpler calculations for predicting the optical modes
which can be supported in the fiber. In the optical fiber 40, as
finally produced, the central portion 41 has a total thickness in
the y direction denoted by D. The central portion is supported by
the relatively thin film members 42.1 and 42.2, both of thickness
b, on either side thereof, respectively.
EXAMPLE 3
(Single Mode Fiber):
Again assuming that the rectangular optical fiber 40 is selected to
be made of fused silica, and that the optical radiation to be
propagated therethrough has a wavelength of approximately 1 micron,
and selecting a width A which is equal to the thickness D, it
follows from Equation (1) that b = A/.sqroot.2, approximately, for
the case where b is at least several times greater than the optical
wavelength. While this latter condition simplifies the calculation,
it is not essential to the invention. In an illustrative case, b is
selected to be about 5 micron, with A and D selected to be about 7
micron.
EXAMPLE 4
(Single Mode Fiber):
Referring to FIG. 4, a single-mode, single-material fiber 40 can
support propagating optical energy of wavelength approximately 1
micron. In an illustrative case, the thicknesses of the thin
portions 42.1 and 42.2 are both about 7 micron, whereas the
thickness D of the central portion 41 in the Y direction is about
10 micron. Likewise, the width A in the X direction of the central
portion 41 is also about 10 micron. Finally the inner diameter of
the periphery portion 43 is about 60 micron, and the outer diameter
thereof is about 100 micron.
It should be understood that although the invention has been
described in terms of detailed embodiments, various modifications
can be made by the worker of ordinary skill in the art without
departing from the scope of the invention .[.x range.].. For
example, various glass materials, in addition to fused silica, can
be used for the transparent materials in the portions 11, 12.1,
12.2 of optical fiber structure 10. These portions need not all be
of the same transparent material, so long as they can be fused
together. Moreover, to furnish optical interaction wiwth the
propagated modes, suitably optically nonlinear material to be
placed in contact with the central portion 11 can be selected from
such well-known materials as Rhodamine 6-G in water, Rhodamine 6-G
in methanol, ethyl alcohol, chlorobenzene, and carbon disulphide.
Alternatively, or in addition, optically linear material for
cladding the central portion 11 can be selected of known optically
linear dielectric materials. Also, the cross section of the central
portion of the optical fiber need not be rectangular, but other
contours can be used, such as circularly or semicircularly
cylindrical, for both single and multimode operation.
Finally, an optical cable, containing many similar optical fibers
of this invention, can be fabricated by incorporating these fibers
in a single peripheral structure having a circular, elliptical or
rectangular cavity for containing these fibers (all of which are
joined to the peripheral structure at the tips of the tin film
supporting members).
THEORY
Assuming, in the structure shown in FIG. 2, that both A, B and b
are all much larger than the propagating optical wavelength, by at
least an order of mangitude, the mathematical solution of the
optical boundary value problem presented by this cross section
shows that the modes which are exponentially decreasing in
intensity in the .+-. X direction, going away from the central
portion, can be supported by the structure. Moreover, in discussing
these modes, it is convenient to introduce a quantity V defined
as
V = (.pi. b/.lambda.) .sqroot.n.sup.2.sub.c - n.sup.2 (7)
wherein n.sub.c is the common refractive index of the central
portion 11 and its members 12.1 and 12, at wavelength .lambda.
where n is the refractive index of the space 15 contacting the
exposed surfaces thereof. In terms of this quantity V, it can be
shown that the number of guided modes is equal to N given
approximately by (for large numbers thereof only):
N = .pi. AB/2b.sup.2 /1 + (.pi./2V).sup.2 (8)
The mode supporting efficiency e (the ratio of the number of modes
guided by this structure to the same structure except with b = 0)
is given by
e = 1/1 + (2Y/.pi.).sup.2. (9)
It is further convenient to define quantity .delta. as follows:
.delta. = (e/2)(1-n.sup.2 /n.sup.2.sub.c). (10)
It should be noted that the electromagnetic boundary value problem
presented by the structure shown in FIG. 2 can be approximated as a
one-dimensional optical fiber problem with a central rectangular
slab portion, of refractive index n, contacted at only two opposite
sides by rectangular slabs of refracted index n.sub.e and by vacuum
on the other two sides in which
n.sub.e = m(1-.delta.) (11)
and in which n.sub.c is the refractive index of the material in the
central portion 11 in FIG. 2. It can be further shown that in terms
of this equivalent problem:
.delta. = n.sub.c -n.sub.e /n.sub.c. (12)
Furthermore, the numerical aperture, ("N.A."), that is, the maximum
acceptance angle with respect to the axis of the filament in FIG.
2, at which obliquely directed optical radiation can be propagated
through the fiber, is given by:
N.A. = n.sub.c .sqroot.2.delta.. (13
Also, for this structure, the tolerable radius of curvature, R (for
one percent loss per centimeter), is given by:
R = A/2.delta.. (14)
And the dispersion, in terms of time delay between lowest and
highest order modes per unit longitudinal length of the optical
filter, is given by
T = n.sub.c .delta./c (15)
where c is the speed light in vacuo. The important operating
parameters given by Equations (8) through (15) are thus simply
calculated in advance, in order to design and obtain the
structure's desired operational characteristics.
* * * * *