# Fileset

[lat_proc2.pdf](https://mdr.nims.go.jp/filesets/c44a9608-947a-4923-8143-d51e731e1681/download)

## Creator

[TODOROKI, Shin-ichi](https://orcid.org/0000-0003-3986-1900)

## Rights

In Copyright[In Copyright](http://rightsstatements.org/vocab/InC/1.0/)

## Other metadata

[In-situ observation of fiber-fuse ignition](https://mdr.nims.go.jp/datasets/fec879cb-88a8-4e9f-b536-bc2e1461838d)

## Fulltext

In-situ observation of fiber-fuse ignitionShin-ichi TodorokiAdvanced Materials Laboratory, National Institute for Materials Science,Namiki 1-1, Tsukuba, Ibaraki 305-0044, JapanABSTRACTUltrahigh-speed video recording of fiber-fuse ignition was reported for the first time. Fiber fuse was initiated in a glassferrule in which the fiber end was in contact with cobalt oxide powder. Optical discharge emerged from a slowly movingdarker radiant at about 300µm distance from the fiber end. Absence of void in the trajectory of the dark radiant impliesthat there is no glass-plasma interface at the radiant. Two other types of fused damage on ignition were also reported.Keywords: Fiber fuse, Ultrahigh-speed videography, Laser-induced damage1. INTRODUCTIONFiber-fuse effect was discovered in late 1980s,2, 3 initiated by(a)(b)(c)Figure 1. Successive captured video images of fiber-fuseignition taken by a normal video recorder.1 The shootingspeed is 30 frames per second.local heating of optical fiber to generate an optical discharge run-ning along the fiber to the light source (∼W) resulting in catas-trophic destruction of core region. Ever since, nearly 40 papershas been published concerning this phenomenon. The list of thesepapers is available elsewhere (The author’s another paper4 in thisvolume, and his web page; its URL is printed below).Recent growth of available laser power (>kW) gives rise toan urgent need for fiber fuse termination.5–8 It is more important,however, to recognize this initiation mechanism. Most of the workson fiber-fuse ignition up to now are based on indirect experimentalresults, such as required pumping power density and the methodsfor initiaton. This is due to the difficulty in direct observation ofthis phenomenon. Usually, the ignition is accomplished by oneof the following ways; bringing the fiber output end into contactwith absorbent materials, or heating the fiber with a flame, a fusionarc, or in a furnace. Each of these makes us hard to observe thephenomenon directly and microscopically.Recently, the author succeeded to take a photograph of the ig-niting moment using a normal video camera (see Fig. 1).1, 9 Thispaper presents further investigation of this topic using ultrahigh-speed videography.2. EXPERIMENTALFigure 2 shows the experimental setup in this study. One end of a commercial single-mode silica glass optical fiber (SMF-28, Corning, core diameter: 9µm) was connected to a Raman fiber laser (PYL-10-1480, IPG Laser, 1.48µm, 9 W). In orderto initiate a fiber fuse in a observable configuration, the other end of the fiber was inserted into a glass ferrule with a smallamount of cobalt oxide powder, as shown on the right of Fig. 2. The ignition was observed through a CCD camera (ultimaAPX-RS, monochrome version, Photron Ltd., sensitivity range: 380-790 nm) with an appropriate zoom lens. Pictureswith a resolution of 256×32 were taken every 10µs with 1-µs-exposure time through ND (neutral density) filters (x64).Damaged sites were examined by an optical microscope.E-mail: TODOROKI.Shin-ichi@nims.go.jp, Facsimile: 81 298 54 9060, URL: http://www.geocities.com/Tokyo/1406/FiberLaser1480nm, 9WND filterSMF-28Zoom lensUltrahigh-speedcameraGlass ferruleCo oxide powderFigure 2. Experimental setup for observing fiber fuse ignition (left) and configuration for self-ignition by laser pumping (right).Figure 3. Photographs of visible light emission around fiber-fuse ignition (upper) and their intensity profiles along the dashed lines onthe photos at every 10µsec (lower). The fiber end is located nearx = 0. The laser pumping started several seconds beforet = 0.0 ms.3. RESULTSFive takes of fiber-fuse ignition were recorded. A typical result is compiled in Fig. 3. The upper half shows photographsof visible light emission around fiber-fuse ignition (t=2.2 ms). The laser pumping started several seconds beforet=0.0 ms.The fiber end is located nearx=0. Thus, the photo att=0.0 ms represents heated area of the Co oxide powder. Lower half ofFig. 3 and Fig. 4 show time-varying intensity profile of the radiant along the dashed line shown in the upper photos. Thesetrajectories clearly show that a dark radiant emerged att = 1.55 ms andx= 90µm (see upper arrow in Fig. 4) and movedslowly along the fiber. Then, an fiber fuse occurred from the radiant att = 2.2 ms nearx= 300µm. Averaged speeds of thedischarge and radiant are calculated to be 1.2 m/s and 0.37 m/s, respectively. Other four cases showed similar tendency.Bottom of Fig. 4 shows the voids generated by the ignition. Periodic voids appears in the region ofx >300µm, atwhich the optical discharge appeared. A thin void extends from the fiber end to the depth of about 100µm, where thedark radiant appeared. Thus, it is reasonable to conclude that the thin void and the periodic voids were generated afterthe optical discharge emerged and before the dark radiant emerged, respectively. For all the five cases recorded, similardamages are generated. It is, however, not the only mode of ignition. In some rare cases, different types of damage wereFigure 4. Upper: Time-varying intensity profile of visible light emission before fiber-fuse ignition (t=2.2 ms) along the dashed lines onthe photos shown in Fig. 3. Lower: Optical micrograph of damaged fiber.(a)(b)∼ 6cm⇐=Figure 5. Optical micrographs showing two types of ignited tail.observed as shown Fig. 5(a) and (b). In either case ultrahigh-speed video shooting have not succeeded yet.4. DISCUSSIONA trajectory of the dark radiant is also recorded in the captured video image shown in Fig. 1(b), as an absence of lightfilament. After emerging the optical discharge, light filament appears because the light intensity is large regardless of itshigh speed. Before emerging the dark radiant, it appears due to long staying time in spite of lower light intensity.The speed of dark radiant is as fast as that of optical discharge pumped by 1.5 W light,4 0.33 m/s, where otherexperimental conditions are the same as this study. These two phenomena are, however, completely different becauseonly the latter leaves a thin continuous void. Absence of void in the trajectory of the dark radiant implies that there is noglass-plasma interface at the radiant. Thus, it can be called a transient propagation mode of energy without gas plasma.It is interesting to find such a void-free section in another case of damage (see Fig. 5(b)) at about 6-cm-distance fromfiber end. The 6-cm-long thin void looks like thin voids generated by less than 2.0-W-pumping.4 The pumping energyin this case was, however, more than 2.0 W. It seems to be also different from the thin void shown in Fig. 4 because ofthe presence of large hole on the fiber end surface. Thus, it must be another propagation mode of energy with gas plasma.Further investigation including ultrahigh-speed videography is needed.The last case of damage shown in Fig. 5(a) is likely to be brought about by an ignition without any fore-running phe-nomenon. In other words, light-induced heating of Co powder could accidentally give enough heat to make an immediateignition. A huge hole at the fiber end and a gradual increase of void size toward the fiber end are likely to be due to theheat generated there. Similar void expansion has been reported in an ignition by fusion arc.105. CONCLUSIONFiber-fuse ignition was observed by an ultrahigh-speed video camera. Optical discharge was found to be generated after afore-running phenomenon in which a darker radiant moved slowly along the fiber and left no void. It must be a trangentpropagation mode of energy without gas plasma.ACKNOWLEDGEMENTThe author is grateful to Mr. Kazuhide Hanaka and Mr. Akira Sakamaki (Photron Ltd.) for helping the experiment ofultrahigh-speed videography and Dr. Satoru Inoue (National Institute for Materials Science) for continuous support.REFERENCES1. S. Todoroki, “In-situ observation of fiber-fuse propagation,” inProc. 30th European Conf. Optical CommunicationPost-deadline papers, pp. 32–33, Kista Photonics Research Center, (Stockholm, Sweden), Sept. 2004. (Th4.3.3).2. R. Kashyap and K. J. Blow, “Observation of catastrophic self-propelled self-focusing in optical fibres,”Electron.Lett.24, pp. 47–9, Jan. 1988.3. D. P. Hand and P. S. J. Russell, “Solitary thermal shock waves and optical damage in optical fibers: the fiber fuse,”Opt. Lett.13, pp. 767–769, Sept. 1988.4. S. Todoroki, “Ultrahigh-speed videography of fiber fuse propagation: a tool for studying void formation,” inthisvolume of SPIE proceedings.5. D. P. Hand and T. A. Birks, “Single-mode tapers as ’fibre fuse’ damage circuit-breakers,”Electron. Lett.25, pp. 33–34, Jan. 1989.6. K. Seo, N. Nishimura, M. Shiino, R. Yuguchi, and H. Sasaki, “Evaluation of high-power endurance in optical fiberlinks,” Furukawa Review, pp. 17–22, July 2003.7. S. Yanagi, S. Asakawa, M. Kobayashi, Y. Shuto, and R. Naruse, “Fiber fuse terminator,” inThe 5th Pacific RimConference on Lasers and Electro-Optics, 1, p. 386, July 2003. (W4J-(8)-6, Taipei. Taiwan, 22-26 Jul. 2003).8. E. M. Dianov, I. A. Bufetov, and A. A. Frolov, “Destruction of silica fiber cladding by the fuse effect,”OpticsLetters29, pp. 1852–1854, Aug. 2004.9. S. Todoroki, “In-situ observation of fiber-fuse propagation,”Jpn. J. Appl. Phys.44, 2005. (in print).10. R. Kashyap, “High average power effects in optical fibers and devices,” inReliability of Optical Fiber Components,Devices, Systems, and Networks, H. G. Limberger and M. J. Matthewson, eds.,SPIE Proceedings4940, pp. 108–117,SPIE, Apr. 2003. (Brugge, Belgium, 28 Oct. 2002).