Fiberoptic polarizer and method of making the same

Details Fiberoptic polarizer and method of making the same Fiberoptic polarizer and method of making the same

2000-12-21

2003-03-18

Healy, Brian (Department: 2874)

Optical waveguides

Polarization without modulation

C385S014000, C385S065000, C385S073000, C385S031000

Reexamination Certificate

active

06535655

ABSTRACT:

FIELD OF THE INVENTION
The invention relates to optical functional devices that are integrated directly into optical fibers for use in telecommunication systems. More particularly, the invention relates to a fiber-optic polarizer device and a method of making the device.
BACKGROUND OF THE INVENTION
In optical communication systems, several kinds of optical functional devices, such as isolators, switches, filters, and amplifiers must be inserted between optical fibers. Most in-line optical devices use lens elements to collimate a light beam from an incoming single-mode fiber (SMF) and focus it on an out-going SMF, where functional elements are placed in between two fibers. As a result, precise alignment (within 0.1-0.5 &mgr;m) between SMF and lens is required and subsequent alignment between the two facing lenses is required to make collimate system. Those alignments are extremely complicated and troublesome.
The use of very thin elements makes possible the production of optical devices without the need for expensive fiber collimator system using lens elements to lessen alignment problems in order to maintain high light throughput. Coupling loss between two optical fibers mainly depends on optical distance between the two fiber-ends, and the core diameter of each fiber. For any integrated fiberoptic device, one of the goals is to shorten the optical path-length in order to decrease coupling loss caused by diffraction. These and other aspects of vertical integration technology, as it is known are described by Shiraishi et al., in
Vertical Integration Technology for Fiber Optic Circuit,
OPTOELECTRONICS, Vol. 10, No. 1, pp.55-74, March 1995. Shiraishi's article discusses an approach that focuses on making fiber-integrated isolators, where the optical components are relatively thick, having several hundred microns. Therefore, employment of a fiber having a large core diameter, such as a TEC (Thermally Expanded Core)-fiber, is essential to suppress coupling loss in such a device.
Generally, two types of fiber-optic polarizers are known. A first type consists of a thick (>0.1 mm thick) polarizer material that is placed in between fiber collimator system. In this first type of polarizer, many optical components that require high structural preciseness are used, so that the cost is increased. In such a polarizer, one of the lenses collimates a light beam from an incoming fiber and the other lens focuses the light on an out-going fiber. Because of the rather bulky arrangement and large thickness of the optical elements, such a polarizer design cannot avoid large coupling losses between the two fibers without the use of a fiber collimator system that contains lenses. Precise alignment (0.1-0.5 &mgr;m) between lens and fiber is required and subsequent precise alignment between the two facing lenses that is fixed with each fiber is also required to produce the collimator system for this polarizer. Moreover, costly packaging of such a polarizer is essential to maintain optical alignment, as well as performance reliability.
In order to avoid difficult alignment problems, workers proposed a second type of fiberoptic polarizer in which lenses are not used. This second type of lens-free device uses a laminated polarizer material, known commercially as LAMIPOL by Sumitomo Osaka Cement (SOC). In an embodiment of this kind of fiberoptic polarizer, LAMIPOL is placed in between the end facets of two optical fibers. LAMIPOL has a structure in which metal and dielectric layers are alternately laminated with periodicity, and is made by alternatively depositing Al (or Ge) and SiO
2
films with RF sputtering. LAMIPOL has a thickness of typically about 30 &mgr;m. Due to the thinness of the material, lens elements are not necessary since the coupling loss is negligibly small.
Use of LAMIPOL in lens-free devices, however, has several problems. One of the major concerns is difficulty in handling for alignment during the fabrication process. As Sasaki et al., in Japanese Patent No. 99-23845, points out, inserting LAMIPOL into a gap formed in a waveguide results in low yield because of the inherent difficulties associated with handling and breakage of LAMIPOL. In particular, due to its relatively small physical size (1.6 mm×1 mm square or 1.6 mm×4.6 mm square) and its brittleness, handling pieces of LAMIPOL in the alignment process leads to unacceptable levels of breakage.
Although Sasaki et al. improved the handling aspect of fabrication by increasing the physical size of a piece of LAMIPOL, they failed to solve another inherent problem of LAMIPOL, that is, it has a very small optical aperture (0.1 mm×1 mm square or 0.1 mm×4.6 mm square). This optical aperture is located at an end of a lateral side of a piece of LAMIPOL. Therefore, vertical alignment is essential for LAMIPOL to be inserted into a gap. As mentioned before, LAMIPOL has a structure where metal and dielectric layers are alternately laminated with periodicity. Since the absorbing cross-section of a laminated structure depends strongly on the incident angle, another problem with Lamipol is its inherently small acceptance angle. The width of a gap needs to have a relatively tight tolerance for insertion of the LAMIPOL. When the width of a gap between fiber ends is larger than the thickness of a piece of LAMIPOL, LAMIPOL can be accidentally tilted into an incorrect angle relative to the normal of waveguide orientation. This tilting has a significant effect upon return and insertion losses. Ordinarily, arranging optical components at a tilted angle against the normal of the optical axis in fiber-optic applications has been an effective configuration to improve return loss. But, because the acceptance angle of LAMIPOL is inherently small, a tilted configuration results in an increase in insertion loss. Theoretical calculation shows that an angle greater than approximately 2.5° is required to attain a return-loss of up to approximately 55 dB when LAMIPOL is inserted in between two ends of SMF. Given the inherently small acceptance angle of LAMIPOL, a trade-off must be made between either improving return loss or reducing insertion loss. Hence, in a LAMIPOL fiber polarizer when the LAMIPOL is placed at an angle to improve the return loss, insertion loss increases. Thus, it may be impossible to use LAMIPOL to reduce return loss, because it may be impossible to insert LAMIPOL at an angle in a tilted configuration.
An alternate example of this second type of polarizer is disclosed by J. Stone in European Patent Application, EP 0751410A2. Stone proposed sandwiching a polished piece of prefabricated dichroic glass polarizer using a glue, having thickness of less than 50 &mgr;m, in between end facets of two fibers. The fabrication process for the Stone example entailed thinning a piece of polished dichroic glass polarizer, affixing the polarizer to one end facet of a first fiber and in a subsequent step optically coupling an end facet to a second fiber to the polarizer. This process is accomplished using the aid of a rotary connector or alignment sleeve, which has an inherent loss due to residual misalignment. Although residual misalignment may be slightly suppressed with rotation of each fiber, other complications would arise. In order to rotate the each fiber, end facets of each fiber need to be perpendicular to the optical axis, which means that the polarizer material needs to be sandwiched perpendicularly to the optical axis. This configuration forecloses the use of an angled placement of the polarizer material to reduce return loss. Therefore. since Stone's polarizer need to be set perpendicular to the optical axis, this configuration has inherent problems of return loss. Further, the refractive index of the glue depends on the temperature difference that would result in index mismatch. In addition, more numerous optical components are required, and Stone's process is rather slow and can make only one polarized fiber at a time.
In view of the foregoing discussion, a new design for vertical integration technolog

Affiliated with

Also associated with

Rating

Fiberoptic polarizer and method of making the same does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Fiberoptic polarizer and method of making the same, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Fiberoptic polarizer and method of making the same will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-3012185