US 20030168154 A1
A composition of phosphate-based glass fiber capable of being fused directly to a silica-based glass fiber includes from about 60 mole % to about 75 mole % P2O5, from about 8 mole % to about 30 mole % X203, from about 0.01 mole % to about 25 mole % of R2O. X is selected from the group of Al, B, La, Sc, Y and combinations thereof. R is selected from the group of Li, Na, K and combinations thereof. Optionally, the phosphate-based glass fiber may include 0.5 to 10 mole % MO, where M is selected from Mg, Ca, Sr, Ba, Zn, and combinations thereof. The composition preferably further includes from about 0.5 mole % to about 15 mole % of an additional component from the group of Si, Ge, Pb, Te and combinations thereof.
1. A method of fusing a silica glass fiber to a phosphate glass fiber, comprising the steps of:
providing a phosphate glass fiber capable of amplifying a signal and having a composition comprising from about 60 mole % to about 75 mole % P2O5, and from about 8 mole % to about 30 mole % X2O3, from about 0.01 mole % to about 25 mole % of R2O, and from about 0.01 to 8.0 mole % of a lasing ion, and wherein X is selected from Al, B, La, Sc, Y and combinations thereof, and R is selected from Li, Na, K and combinations thereof, and
fusing said phosphate glass fiber component directly to said silica glass fiber.
2. The method set forth in
3. The method set forth in
4. The method set forth in
5. The method set forth in
6. The method set forth in
7. The method set forth in
8. The method set forth in
9. The method set forth in
10. The method set forth in
11. The method set forth in
12. The method set forth in
13. A method of fusing a silica glass fiber to a phosphate glass fiber, comprising the steps of:
providing a phosphate glass fiber capable of amplifying a signal and having a composition comprising from about 60 mole % to about 75 mole % P2O5, from about 8 mole % to about 30 mole % X2O3, from about 0.01 mole % to about 25 mole % of R2O, from about 0.01 to 8.0 mole % of a lasing ion, from about 0.5 mole % to about 15 mole % of a component comprising Si, Ge, Pb, Te, and combinations thereof, wherein X is selected from Al, B, La, Sc, Y and combinations thereof, and R is selected from Li, Na, K and combinations thereof;
and fusing said phosphate glass fiber component directly to said silica glass fiber.
14. The method set forth in
15. The method set forth in
16. The method set forth in
17. The method set forth in
18. The method set forth in
18. The method set forth in
19. The method set forth in
20. The method set forth in
21. The method set forth in
22. The method set forth in
 This application claims priority to U.S. Provisional Patent Application No. 60/375,456 filed on Feb. 15, 2003.
 The present invention relates generally to a glass fiber used for amplifying a signal transmitted through a glass fiber. More specifically, the present invention relates to a phosphate glass fiber that can be fused directly to a signal transmitting fused silica glass fiber.
 Glass fibers have been widely used in the telecommunications industry to transmit high volumes of optical signals at high speeds. Traditionally, these glass fibers have been fashioned from fused silica-based compositions. Optical signals transmitted through fused silica glass fibers, however, eventually weaken when transmitted across long distances. Therefore, it is necessary to amplify the transmitted optical signals at various stages along the length of these silica glass fibers.
 One method of amplifying the transmitted optical signal has been to insert a doped, fused silicate or a doped fluoride glass to amplify the strength of the optical signal being transmitted through the silica-based glass fibers. One such doped glass is disclosed in U.S. Pat. No. 5,322,820, filed on Dec. 8, 1992, the content of which is incorporated by reference herein. Doped, phosphate-based glass is also known to amplify the signal transmitted through the fused silica glass, however, is very difficult to fuse or splice phosphate glass to the silica glass fibers because the two glasses have very different physical properties. For example, a typical phosphate-based glass exhibits a glass transition temperature of about 400° C. and a thermal expansion value greater than about 100×10−7/° C. By way of contrast, a fused silica glass typically has a glass transition temperature of about 1,000° C. and a thermal expansion of generally 5×10−7/° C. Phosphate glasses are generally known to have the desirable characteristics of good chemical durability, high ion exchange activity, a high gain per length coefficient, wide gain spectrum, and low conversion characteristics. These properties make it desirable to develop a simple and reliable method for splicing phosphate glasses to fused silica glass.
 These large differences in thermal and physical properties have rendered the common fusion splice method used to connect separate strands of silica fibers impractical when mating a silica glass fiber to a phosphate glass fiber. One method used to attach an amplifying phosphate glass fiber to a silica glass fiber system is to insert a splicing glass similar to that disclosed in U.S. Pat. No. 6,277,776. The splicing glass includes thermal and physical properties somewhere between the properties exhibited by phosphate-based glasses and silica-based glasses. In the absence of a splicing glass, the differences in glass transition temperature between the phosphate and silica based glasses is known to cause a weak fusion between the glasses resulting in breakage at the fusion joint during regular usage. However, the introduction of a spliced glass between a phosphate glass fiber and a silica glass fiber is known to be expensive and difficult to perform in a mass production environment. Therefore, it would be desirable to introduce a phosphate-based glass fiber capable of being fused directly to a silica-based glass fiber using conventional glass fiber splicing methods.
 The present invention provides phosphate glass fibers having modified thermal and physical properties enabling the use of a common fusion splice method for connecting phosphate-based glass fibers directly to silica-based glass fibers. The composition of the inventive glass includes from about 60 to about 75 mole percent of P2O5, from about 8 to about 30 mole percent X2O3, and from about 0.5 to about 25 mole percent of R2O. X is preferably selected from the group comprising Al, B, La, Sc, Y, and mixtures thereof. R is preferably selected from the group comprising Li, Na, K and mixtures thereof. The composition may also comprise from about 0 to 15 mole percent of MO, wherein M is selected from Mg, Ca, Sr, Ba, Zn, and mixtures thereof. The index of refraction and the thermal expansion of the phosphate-based glass is modified by including from about 0.5 to about 15 mole percent of a component selected from the group of Si, Ge, Pb, Te, and mixtures thereof.
 The combination of elements listed above provides a phosphate glass with a glass transition temperature approximately 100° C. higher than conventional phosphate-based glasses. Further, the thermal expansion value of the inventive phosphate-based glass is significantly lower than typical phosphate-based glasses. In addition, the phosphate glass formulation set forth above exhibits greater chemical durability and higher gain when doped with active lasing ions for use as fiber amplifiers as known to those of skill in the art.
 The modified thermal and physical properties of the phosphate-based glass unexpectedly provide the ability to form a durable fusion joint between phosphate and silica-based glass fibers. This can be accomplished by using low loss fusion splicing, with commercial fusion splicing equipment, which provides high gain phosphate glass fiber amplifiers and fiber lasers directly fused to silica glass fibers. This replaces the more expensive and inefficient technology presently being used to insert phosphate fiber amplifiers into a silica-based fiberglass strand.
 Other advantages of the present invention will be readily appreciated as the same becomes better understood by reference to the following detailed description when considered in connection with the accompanying drawings wherein:
FIG. 1 is photograph of the inventive phosphate fiber spliced to a Corning SMF-28™ silica glass fiber;
FIG. 2 is a second photograph of the inventive phosphate fiber spliced to a Corning SMF-28™ silica glass fiber;
FIG. 3 is a photograph of the inventive phosphate fiber spliced to a Corning SMF-28™ silica glass fiber after having been subjected to destructive testing.
 The present invention is directed toward a composition of glass capable of being fused directly to a silica-based glass fiber strands, which are widely used to transmit optical signals. The composition of glass includes from about 60 to about 70 mole percent of P2O5, from about 8 to about 30 mole percent of X2O3, and from about 0.01 to about 25 mole percent of R2O. X is preferably selected from the group-comprising Al, B. La, Sc, Y and combinations thereof. R is preferably selected from the group comprising Li, Na, K, and combinations thereof.
 The index of refraction and the thermal expansion of the phosphate-based glass is preferably modified by including from about 0.5 to about 15 mole percent of a component selected from the group of Si, Ge, Pb, Te, and combinations thereof.
 Additions of MO from about 0 to about 10 mole percent may also be made to the glass composition, where M is selected from the group of Mg, Ca, Sr, Ga, Zn and combinations thereof.
 The cores and cladding of the phosphate-based glass fibers are preferably doped up to and including the limits of solubility with lasing and sensitizer ions from the Lanthanide series as displayed in the Periodic Table of Elements. Preferably, the amount of lasing or sensitizer ion is from 0.01 to 8.0 mole percent. Preferably, the lasing and sensitizing ions are used in the tri-oxide form and comprise: cerium (Ce), praseodymium (Pr), neodymium (Nd), promethium (Pm), samarium (Sm), europium (Eu), gadolinium (Gd), terbium (Tb), dysprosium (Dy), holmium (Ho), erbium (Er), thulium (Tm), ytterbium (Yb), lutetium (Lu), copper (Cu), and chromium (Cr). Alternatively, combinations of elements listed in the Lanthanide series of the Periodic Table may be included in the cores and claddings of the phosphate-based glass strands to dope the phosphate glasses as is known to those of skill in the art. Preferably, combinations of erbium and ytterbium are used.
 The composition of the phosphate-based glass set forth above has produced the following physical properties that have proven beneficial to splicing the phosphate-based glass directly to the silica-based glass. The refractive index at the sodium D-line (nD) is from about 1.53 to about 1.55 whereas silica-based glass has a refractive index of 1.46. The phosphate-based glass composition has an Abbe number (measurement of optical dispersion), determined from the equation (nd−1)/(nf−nC), ranging from about 63.5 to about 65.0, whereas the silica-based glass has an Abbe number of about 70.4. An optical dispersion that is equivalent to silica-based glass is preferable when fusing the phosphate-based glass to the silica-based glass. The non-linear refractive index of the phosphate-based glass is from about 1.19×10−13 esu to about 1.23×10−13 esu, whereas the silica-based glass has a non-linear refractive index of 2.4×10−13 esu. The thermal expansion co-efficient of the phosphate glass is from about 70×10−7/° C. to about 84×10−7/° C. whereas the silica-based glass is 5.5×10−7/° C. The glass transition temperature of the phosphate-based glass ranges from about 440° C. to about 515° C., whereas the silica-based glass has a glass transition temperature of about 1042° C. Notably, a typical phosphate glass exhibits a glass transition temperature of generally about 400° C. The increase in glass transition temperature shown by the inventive phosphate-based glass has provided a significantly stronger fusion joint to the silica-based glass. The deformation temperature (Td) of the phosphate-based glass is from about 480° C. to about 535° C., whereas the deformation temperature of the silica-based glass is generally 1585° C. The thermal conductivity of the phosphate-based glass is from about 0.8 w/mk to about 0.9 w/mk, whereas the thermal conductivity of the silica-based glass is 1.30 w/mk. The density of the phosphate-based glass is from about 2.6 g/cc to about 3.0 g/cc, whereas the density of the silica-based glass is generally 2.2 g/cc.
 The physical and thermal properties set forth above have unexpectedly provided the ability to make a “forgiving” fusion splice of a phosphate-based glass fiber to a silica-based glass fiber. Most notably, increasing the glass transition temperature by up to 100° C. over traditional phosphate-based glass transition temperatures of generally 400 ° C. and reducing the thermal expansion values to well below the general value of 100×10−7/° C. has resulted in the ability to form a durable spliced joint between phosphate-based glass fibers and the silica-based glass fibers in the absence of using a splicing glass.
 Experiment Description and Results
 Single mode fiber samples of Corning SMF-28™ (125 μm OD) fused silica glass and a sample of the inventive phosphate-based glass composition having the characteristics listed in Table 1 below were cleaved by using an Amherst Ericsson model #EFC 11-4 electronic cleaver or a Fitel Furukawa model #S323 manual cleaver.
 Single mode (rectangular) double-clad and round multi-mode fibers were also evaluated having the following properties disclosed in Table 2:
 Good quality cleaved fiber faces were achieved with both cleaving units on various test fibers. Initial successes with the Ericsson electronic cleaver unit on the single mode fibers were surprisingly surpassed with quality and consistency of the cleaves produced by a simpler Fitel manual cleaver unit. Test cleaving optimization studies with the Ericsson unit presented a trend indicating a requirement for higher tension settings for the production of optimized (flat and smooth) cleaved fiber faces. Continued studies with the Fitel manual cleaver (which provides no fiber tension unlike the Ericsson unit, which does provide fiber tension) surpassed all earlier work with very consistent high quality cleaved fiber end faces.
 Initial results of the fiber splicing experiments using the single mode phosphate-based glass composition spliced to standard fused silica-based glass were evaluated. Strong splices were produced that exhibited mechanical strengths of 0.5 Gpa (equivalent to 6 newtons). This unexpected result is comparable to a standard silica/silica splice. Initially, a strong splice with a measured loss of approximately 2.5 dB was established.
 A Sumotomo type 36 unit arc circuit board was modified to increase the heat control resolution and limit the range to the lowest 10% of the unit's standard scale. The power was set to 1, the gap to 10, the overlap to 15, the pre-fusion to 0.0, the duration to 0.15, and the arc spattering duration to 0.05s.
 Fiber fusion splicing was performed with the modified Sumotomo type 36 and an Ericsson model FSU 995 FA unit. As the optimization experiments proceeded, it was found that the Ericsson model FSU 995 FA provided more flexibility when adjusting the system parameters. Optimum settings for the Ericsson unit were as follows: pre-fuse time 0.2 sec., pre-fuse current 2.0 mA, Gap 50.0 mm, overlap 13.0 mm, segment #1 fusion time 0.1 sec., segment #1 fusion current 8.0 mA, Segment #2 fusion time 1.0 sec., segment #2 fusion current 2.0 mA, segment #3 fusion time 1.2 sec., segment #3 fusion current 2.0 mA, and set center (offset form 255 arc center) 225.
 Processes for cleaving and fusion splicing single mode phosphate fibers to fuse silica fibers were independently developed. Using 180 μm diameter cladding, single mode core, amplifier fibers produced a fusion splice where the strength of the bond between the single mode phosphate fibers and the fused silica fibers were unexpectedly good even though the chemical and thermal properties of the two fibers were radically different. Ten splices were examined for breakage characteristics. In each case, the fractures occurred in the fused silica fibers generally within 3 mm of the fusion splice. Therefore, the tensile strength of the fusion splice proved to be unexpectedly stronger than the fused silica fibers.
 A first order quantitative measurement of the tensile strength of the first sample of the phosphate-based glass composition was determined by a loop test and the value compared to that of other fibers. A loop of fiber was drawn by hand. The tensile strength of the fiber was calculated from the radius or diameter of the loop when the fiber broke. Five separate tests were performed-and all broke at the same diameter, 4 mm±0.5 mm. Assuming that the modulus of elasticity of the glass is 10.2×106 psi (67 GPa), and the fiber diameter was 125 μm, then the strength of the fiber was calculated as:
 A value of 382,000 psi is considered to be a very high strength for phosphate-based glass fibers. For example, pristine E glass fibers designed for reinforcement purposes are generally found to have a strength of about 500,000 psi and have a modulus of elasticity of 10.5×106 psi. The inventive phosphate-based glass fiber sample that was tested showed a ratio of σ/E of 0.0375, or about 75% of about what would be expected under ideal conditions.
 Referring to FIGS. 1 and 2, photographs that are representative of the phosphate-based glass fiber indicated in Tables 1 and 2 above being spliced to Corning SMF-28™ (125 μm OD) are shown. The phosphate-based glass fiber has a larger diameter than the Corning SMF-28™. The interface loss is estimated to be approximately 0.25 dB in FIG. 1 and 0.20 in FIG. 3. FIG. 3 shows a typical result of a destructive test use to evaluate the strength of the fused joint between the two different types of glass fibers. The photograph in FIG. 2 indicates the break occurred in the Corning SMF-28™ approximately 3 mm from the fused joint. Therefore, it is believe that the fused joint is stronger than the Corning SMF-28™ glass fiber sample that was tested.
 Optical probing is a straightforward method to measure the loss contribution of a fusion splice. There are three major contributors to the loss through a section of spliced fiber, Fresnel loss, splice loss, and optical absorption. The refractive index difference between the fused silica (1.460) and the phosphate-based glass composition (1.540) glass fibers introduces a Fresnel loss at the splice boundary. The calculated Fresnel loss at the fusion interface for these materials is 0.003 dB. The Fresnel loss at the air interface of the unspliced end of the phosphate fiber was calculated to be 0.2 dB. The material absorption of the phosphate-based glass fiber at 1318 nm was determined to be 0.47 dB/cm.
 The loss measurements were taken on several samples of single-end fusion splice phosphate glass fibers. Two foot lengths of the SMF-28 ™ were fusion spliced to samples of the phosphate glass in Tables 1 or 2 and a high quality APC connector was installed on the loose end of the SMF-28 ™ fiber. Single mode 1318 nm light, with known power, was injected into the APC connector. The output power of the open end of the phosphate fiber was measured with a broad area detector. The length of the splice samples was about 5.0 cm. The total throughput loss at 1310 nm for a 5 cm spliced phosphate fiber was expected to be 2.6 dB. However, the measured throughput loss of the single end spliced phosphate glass fiber was determined to be much lower than expected. The throughput loss ranged from between 2.54 dB to 0.03 dB, with a mean of about 0.6 dB across 12 samples.
 The invention has been described in an illustrative manner, and it is to be understood that the terminology which has been used is intended to be in the nature of words of description rather than of limitation.
 Obviously, many modifications and variations of the present invention are possible in light of the above teachings. It is, therefore, to be understood that within the scope of the appended claims, the invention may be practiced otherwise than as specifically described.