Open Access
Add to BibSonomyAbstract
In light of recent proposals linking structural change and stresses within regenerated gratings, the details of regeneration of a seed Type-I Bragg grating written in H2 loaded germanosilicate fiber annealed at high temperatures (~900°C) are systematically explored. In particular, the influence of the strength of the grating, the effect of GeO2 doping concentration and the annealing conditions on regeneration are studied. We show that the role of dopants such as Ge and F contribute nothing to the regeneration, consistent with previous results. Rather, they may potentially be detrimental. Strongest regenerated gratings with R ~35% from a 5mm seed grating could be obtained in fibres with the lowest GeO2 concentrations such as standard telecommunications-compatible grade fibre.
©2011 Optical Society of America
1. Introduction
Gratings that can operate at temperatures well above standard telecommunication requirements are critical to the success of many real time sensing applications. Sensing in harsh environments like that in the oil and gas industries, power generating stations, air and space craft engine turbines, or industries that uses high temperature furnaces, requires monitoring of temperatures anywhere from 400°C to in excess of 1000°C. A new type of fiber Bragg grating (FBG), where a periodic index modulation can be regenerated after erasure of the UV induced type-I grating [1–5] through annealing at high temperature (~800°C to 900°C), promises to have stable operable temperature in this regime and is attracting worldwide attention recently. There are several works reported earlier with a common goal to push up the operable temperature of FBGs. It may be through tailoring the glass composition [6,7], or the process of hypersensitisation of the fiber [8,9] and also through formation of type-II and type-IIA gratings by using femtosecond lasers with wavelengths at UV [9,10] or at IR [11,12].
The regenerated grating family has several attractive features. It requires a simple Type-I grating in standard fibers as the seed, which is then annealed out at high temperature before being regenerated – this is akin to latent image intensification in photography and hypersensitisation [13,14], indicating the general phenomenon of stored memory within the glass, the origins of which are not fully identified. These regenerated gratings have been shown to operate at temperatures as high as ~1295°C [15]. Like the type-I seed, the strength has been found to be approximately proportional with the length of the grating; therefore, it is plausible to fabricate strong regenerated gratings [16]. It is also shown to retain precise phase shift information inscribed in the seed at nm resolution [17] opening up the prospect of complex gratings that are stable at high temperature. Such extraordinary preservation of the phase information of a seed structure by an otherwise macroscopic annealing process would appear to question a simple temperature dependent diffusive model as underpinning the mechanism - washing out of some phase information would seem almost unavoidable. Thus in this work we also explore the role of key dopants such as Ge. Finally, in the realm of distributed sensing, the spectral characteristics and the insertion loss of the gratings plays an important role and in this respect regenerated gratings have a distinct edge over Type-II gratings [18,19] where the gratings produced often compromise the ideal properties offered by type I gratings.
However, though the spectral property as well as the high temperature stability of the regenerated grating is reasonably good, the limitation is in realizing post-regeneration high contrast local refractive index modulation. In general, it has been observed that the contrast is on the order of ~10% that of the seed grating [1,2]. Therefore, a strong grating over a small grating length has been difficult to achieve. The prime requirement is now to best understand the underlying principle behind regeneration. It is now established that regeneration takes place in standard germanosilicate-only fibers and therefore any diffusion based on smaller atoms such as F [4,5] should not be the root cause for regenerated grating formation. One of the authors proposed the index change based on glass structural transformation arising from relaxation of relatively high internal pressures (few hundred MPa) at high temperature [15,20,21]. This model essentially recognizes that pure vitreous silica is analogous to the liquid form – this can transform at ambient pressures to the polymorph cristabolite when heated slowly at ~1000°C [22]. In practice, the transformation conditions will be somewhat different in a multi-component glass such as germanosilicate under some tensile stress between core and cladding. Add to this the presence of hydrogen and the internal stress this brings, the actual situation is quantitatively complex in appearance. Generally, it may be assumed given the slow annealing processes involved with regeneration (rapid heating fails to produce regeneration), structural changes are likely to have analogous crystalline polymorphs.
Retaining a qualitative interpretation, in the first instance the use of hydrogen helps to alleviate this stress and enable compaction to a denser glass. But this may imply that the effects are localised more to the interface region around the original seed grating induced change – that is, spatially distinct despite the interdependence. Some circumstantial evidence for this was observed by the application of tension which saw a wavelength separation between the original seed grating and its regenerated counterpart [2]. Again, this suggests that the GeO2 component, whilst important for the seed grating generation and therefore the stored information of a periodic structure in the memory of the glass, is not so important for the subsequent regeneration process. A very crude back-scattered x-ray analysis provided some tentative support to this argument [21] but clearly more work is needed. To verify this more detailed study of different fibres with different GeO2 concentrations is necessary. It is also observed [21] that the regeneration threshold is influenced by the wavelength of radiation used for seed grating writing, consistent with some dependence on irradiation history (analogous to thermal history of glasses) and may be tuned by the seed grating strength.
Hence, in this work, to extend the research towards understanding the underlying principle of regeneration we investigate and report regeneration characteristics of Type-I gratings written using 248nm pulsed laser in H2 loaded Ge-doped fibers in the following section and show in detail (2.1) the influence of the annealing schedule, (2.2) the strength of the seed gratings, and (2.3) the Ge concentration and fibre geometry on regeneration characteristics. Regenerated gratings with reflectivity as high as ~35% could be obtained in standard single mode fiber (SMF-28) for a seed of length 5mm, for example.
2. Experimental results and discussion
In this section, the results of a systematic study of regeneration characteristics of type-I seed gratings written in germanosilicate fibers are presented as described above.
2.1 Regeneration through different annealing schedule
In general, when a Type-I grating is annealed isochronally, with the rise in temperature, the grating gradually loses its strength; i.e. the transmission loss at the Bragg wavelength is gradually reduced and at some temperature the transmission dip is submerged in the noise and, within the noise floor of the experiment, the grating is said to be erased. The reflection spectrum, however, is still measurable at this point raising the question as to whether the original periodic structure can at all be fully erased or not. Further annealing causes the reflection peak at Bragg wavelength to go down further, in some cases also into the noise floor, until the onset of regeneration. A typical evolution of the peak of the reflection spectrum of a type-I grating and also its wavelength excursion from erasing point to stabilized regeneration is depicted in Fig. 1 where the progression is subdivided into three regions demarcated by the vertical lines. The plot shows that the grating is annealed isothermally at erasing temperature (region A) where the time zero is the start of isothermal annealing at erasing temperature. The grating temperature is then increased to a higher value (region B) and then is stabilized at that higher temperature (region C). It has been found that the nature of this evolution is similar irrespective of annealing the seed isochronally beyond the erasing temperature or isothermally at the point of erasure. However, the ultimate regeneration strength has been found to be greater if annealed isothermally at erasing temperature of a seed grating. To show this we used three FBGs (samples named S1 to S3)with typical grating strength ~47 dB. Gratings of 5mm length were inscribed into a H2 loaded (1500psi, 100°C for 24 hrs) photosensitive fiber (NM-113) developed in-house containing [GeO2] ~10 mol% in the core. Batch of fibers required for a particular experiment, after taking out from the H2 loading chamber, were put in a freezer (−40°C). Gratings were written immediately after taking fibers out from the freezer and then were annealed. We followed this procedure for all the individual samples in the subsequent experiments presented in this paper. The energy density (~45-50 mJ/cm2) used for writing was identical for all the samples. Annealing schedule-1 as shown in Fig. 2 was followed for sample S1 where the temperature of the furnace was raised uniformly from room temperature to ~1050°C in about two hours and is kept at that temperature for 10 minutes. The erasing temperature has been found to be ~900°C for the seed grating under consideration. Subsequently, annealing schedule-2, also shown in Fig. 2, was followed for the sample S2 where the temperature of the furnace was kept constant at the erasing temperature for about 20 minutes and then raised to ~1050°C where the grating was further annealed isothermally at that temperature for about 10 minutes. This is to mention at this point that increasing the holding temperature more than around 20 minutes did not accrue any advantage in terms of regeneration; i.e. the peak reflectivity and also the Bragg wavelength has found to remain unchanged. The rate of change of furnace temperature in schedule-2 as compared to schedule-1 was adjusted to make the total dwell time of both the samples in the furnace nearly equal; i.e. about 120 minutes for this case. The third sample S3 was inserted in the furnace which was pre-heated at ~900°C and was kept for about 20 minutes and then, like in previous cases, was taken up to ~1050°C and kept there for about 10 minutes. Results are summarized in Table 1 .The measured reflection spectra of gratings at room temperature regenerated from respective samples are shown in Fig. 3 . It is evident that different annealing conditions produced regenerated gratings of different strength in spite of similar seed gratings. The regeneration was the worst in the case in which pre-annealing before regeneration threshold was skipped. It proves that thermal stabilization through pre-annealing of the host structure, as proposed earlier [21], before the regeneration threshold plays a significant role for grating regeneration. In general, rapid isochronal annealing leads to very little regeneration, the first indication that a standard glass transformation from the vitreous state to a crystalline polymorph may be involved. However, the fact that a weak regenerated grating is observed means that the possibility that a weak polyamorph, potentially from a continuum spectrum of densities particularly within such a complex multi-component environment.
Fig. 1 Evolution of Bragg wavelength and reflection peak during grating regeneration and stabilization.
Table 1. Results of grating regeneration for different annealing schedules
Fig. 3 Reflection spectra of regenerated grating obtained through different annealing schedule (measured at room temperature).
It is also apparent that among the pre-annealed samples S1 and S2, isothermal annealing at the regeneration threshold further enhanced the regeneration process. Based on this finding annealing schedule 2 has been followed throughout rest of the experiments.
2.2 Influence of strength of seed gratings on regeneration
In the experiments to evaluate the role of the seed grating strength, four FBG samples (S4 - S7) which were written in H2 loaded NM-113 fiber, were used. Average fluence set for writing gratings was f av ~50 mJ/cm2. The cumulative fluencies were altered to vary the seed grating strength. Details of the seed gratings and the corresponding regeneration data has been summarized in Table 2 . It has been observed that lower the strength of the seed grating, the lower was its erasing temperature. This was found to be ~820°C for the seed with transmission loss, T ~20dB (S4) as compared to ~900°C for the grating with T ~54dB (S7). As per annealing schedule-2, all the seeds were isothermally annealed at respective erasing temperatures for about 20 minutes and then raised to ~1050°C where the regenerated gratings were stabilized for about 10 minutes. The results clearly show (Table 2) that for the same fiber and for seed type-I gratings of similar length, the strength of the regenerated gratings depends on the strength of the corresponding type-I seed grating. Measured spectra of the seeds and the regenerated gratings have been shown in Fig. 4 and 5 respectively.
Table 2. Results of grating regeneration from seed gratings of different strengths
Fig. 5 Reflection (a) and Transmission (b) spectra of regenerated gratings obtained from Type-I seed gratings as shown in Fig. 4.
The peak reflectivity, R, evolution for all the samples recorded from the onset of regeneration until the end of stabilization was found to be similar. For every sample we observed start of regeneration i.e. the reflection peak started growing after reaching a minimum value within <2 minutes of isothermal annealing at the erasing temperature.. The evolution of the peak reflection at the Bragg wavelength as shown in Fig. 6 depicts that observed for samples #S4 and #S7 which had seed grating strengths ~20dB and ~54dB respectively. This is to mention at this point that there is certainly some experimental limitation to exactly pinpoint the moment of grating erasure and therefore, there will be always some error in terms of the time scale in switching over to isothermal heating. However, it is interesting to observe that irrespective of this error the process appears to follow the same trend for all the samples once regeneration has begun and the error induced does not seem to have significant influence on the ultimate regeneration strength. The difference is that the onset of regeneration occurs at higher peak reflectivity for gratings with stronger seed strengths. It is apparent that the latent image of the seed at the regeneration threshold has some structural dependence with the initial seed grating strength. To be more precise, we may recall that for this particular experiment all the grating parameters that contribute towards varying the strength of a standard Type-I grating remain similar for all the samples we considered, apart from the cumulative fluence which were f cum ~0.6 kJ/cm2 and ~2.2 kJ/cm2 for samples #S4 and #S7 respectively. Therefore, it may be concluded that the structural difference observed at the regeneration threshold is due to the difference of cumulative energy with which the host structure is exposed. Considering similar host material, the higher the energy, the greater would be the induced stress in the periodic structure and therefore, the higher would be the corresponding stress relaxation at high temperature annealing which contributes towards forming regenerated gratings of higher strength.
Fig. 6 Evolution of reflection peak of type-I gratings of different strength during grating regeneration.
2.3 Influence of Ge doping concentration and the fiber geometry on regeneration
To study the role of GeO2 concentration, [GeO2], the following fibers were used: (1) SMF-28 of Corning Inc. with [GeO2] ~3-4 mol%; (2) enhanced photosensitive fiber UVS-EPS from Coractive Inc., and (3) NM-113 and NM-41 developed in-house with [GeO2] ~10mole %. None of the fibers contain boron in the core. In NM-41 there is a depressed inner clad having fluorine doping. It was not possible to get any quantitative data regarding [GeO2] in UVS-EPS, but from the measured value of the core-clad refractive index difference (Δn ~0.0264) it can be estimated that [GeO2] ~20 mole % or more (based on Δn ~1.3 x 10−3 corresponds to [GeO2] = 1 mole %) [23]. Seed gratings have been written in each type of fiber with average fluence f av ~45-50 mJ/cm2. Fibers were H2 loaded under similar condition (1500psi, 100°C for 24 hrs). The erasing temperatures of the seed gratings of strength ~47 dB for individual fiber types were first recorded. Seed gratings in NM-41 erase at a much lower temperature (~800°C), perhaps in part due to the F content in the inner cladding, whereas the seed grating written in SMF-28 could be erased at ~935°C. The erasing temperature of the seed in UVS-EPS, however, was similar to that of NM-113. Regenerated gratings were then produced in individual fiber types from another set of four similar seed gratings via annealing schedule 2. The measured data of the seed and corresponding regenerated grating are summarized in Table 3 . Figure 7 shows R spectra of the regenerated gratings at room temperature.
Table 3. Results of grating regeneration from the seeds written in fibers with different [GeO2].
Fig. 7 Reflection spectra of regenerated gratings as obtained in different fibers from similar seed gratings
These results suggest there is an inverse dependence of regeneration strength with GeO2 doping concentration – the lower [GeO2] the higher the strength of the regenerated grating. This indicates that GeO2 does not contribute directly to the regeneration process, although it is important for the seed grating fabrication. In turn, this supports the notion that the main transformation responsible at the temperatures being discussed is that of silica, potentially driven by the core cladding interface with the highest change located there. This is consistent with previous reports [21].The role of other dopants such as fluorine are negated further since even poorer regeneration is obtained within fibre NM-41 which contained fluorine in the inner cladding. An important question is why the regeneration process is poorer in fibers with higher [GeO2]. At these elevated temperatures diffusion may be occurring and this is washing out the grating index contrast. However, if this were true, then the decay of the gratings might be expected to be rapid at the much higher temperature stabilisation process at ~1100°C [21] used to stabilise the gratings. Further, if this process was occurring, much larger variations in regeneration threshold should be observed with varying concentration. Significant variation in the regeneration thresholds of the seed gratings of similar strengths was not observed even though the fibers had different [GeO2] from 3 to 20 mol%. Another possibility is that there is phase separation as the silica transformation takes place, since the free energy of mixing would be increased between the GeO2 and a cristabolite-like SiO2 form, given the larger material mismatch. Once this has occurred, the process can be stabilised. Whilst these effects cannot be entirely ruled out, there remains one more immediate relationship to consider.
The fibre geometry and optical field overlap needs consideration, especially with changes in stress through the core cladding. It is straight forward to estimate that a seed grating of ~47 dB has to have a coupling coefficient κ ac ~1.2x10−3. This number is independent of fiber type, any doping concentration and also fiber geometry. Now κ ac, for similar λBragg, depends on the product of the parameters η, the fractional mode power contained in the grating; i.e. the mode overlap factor, and δn the UV induced index modulation. To get a desired κ ac it is then obvious the parameters η and δn have respective contributions in inverse form; i.e. for a high NA fiber the required δn would be less to get a desired κ ac for a grating. For example, the parameter η for SMF-28 and UVS-EPS was found to be η = 0.76 and η = 0.89 respectively. We can then estimate that to get a grating ~-47dB, the index modulation should be δn ~7.9 x 10-4 and ~6.5 x 10-4 for the respective fibers. The cumulative fluence were f cum ~2.43 kJ/cm2 and f cum ~0.7 kJ/cm2 respectively. Therefore, the larger fluence would suggest that UV-induced stress variations are larger in SMF-28 than that in UVS-EPS. It is interesting to note that if the same index modulation δn ~6.5 x 10−4 induced in UVS-EPS would have been induced in SMF-28, the seed grating would be of strength ~-35 dB. As shown earlier, the strength in regenerated grating increases with seed grating strength so a 35dB seed would produce a weaker regenerated rating than a 47dB seed. Thus, we could come back to our original proposition that the UV induced periodic stress distribution and its stabilization through high temperature annealing is central to enabling structural transformations behind grating regeneration. Moreover, as this stress distribution is predominantly concentrated around the core-clad boundary high NA fibers, having lesser fractional mode power around the core-clad boundary, would therefore be expected to have reduced interaction with the regenerated index modulation in this region. As a consequence of the high confinement of power in the core, these would produce regenerated gratings of weaker strength as compared to a low NA fiber having appreciably more power at the core-cladding interface.
3. Conclusion
In this paper, we report the results of our investigation on regenerated fiber Bragg grating in H2 loaded germanosilicate fibers. Our main concern was to achieve a quantitative understanding of all aspects that influence grating regeneration. In this study, standard single mode fibers like SMF-28 and also fibers having enhanced photosensitivity with different Ge doping concentration were considered. It has been shown that thermal stabilization of stress relaxation process of the induced periodic structure is important and regeneration efficiency can be enhanced by following an appropriate annealing schedule. By optimising this process somewhat, we could achieve improved regeneration (~35% reflectivity) in SMF-28 fiber as compared to the work reported earlier [1]. We have also shown that the GeO2 component does not contribute directly to grating regeneration. Further, the results indicate the presence of other dopants such as fluorine along with the Ge worsens the situation. On the other hand, for the fibres reported here GeO2 is important for their photosensitivity and as well in inducing the original core-cladding stresses both in the fibre and periodically in the seed grating. These stresses are typically tensile and often significant – their exact role is compounded in fibre form by the geometry and optical field overlap, which as we point out needs careful consideration.
Hydrogen is equally important in reducing these stresses overall, enabling regeneration conditions that are strikingly similar to the normal vitreous glass transformation to cristabolite under ambient pressure conditions and temperatures only a little higher to that reported here. An important observation was that the strength of regeneration maintains an almost linear relationship with the modulation depth induced during the seed grating inscription (corresponding to the amount of induced stress) has been identified. These and other results from this investigation shed considerable light on the process of regeneration, whilst simultaneously raising more interesting questions both of an applied and fundamental nature. Ways of further optimising and improving regenerated grating fabrication have been reported, for example, which will help towards developing low loss FBGs with superior spectral quality suitable to use in high temperature applications. The proposed and demonstrated use of an optical fibre as a micro-processing chamber able to be interrogated optically, here with novel interferometric in-line fibre devices such as Bragg gratings, and with relative ease for studying glass transformations in multi-component systems, is a new approach of general significance to glass studies. For example, the regeneration process is yet another latent image intensification process, along with hypersensitisation, type IIa grating formation and so on. This is reflective of the general diversity capable within complex amorphous materials with potentially access to many polyamorphous and polymorphous states. The approach reported is therefore not limited to the silicate system studied here - with the increasing trend towards complex glass systems in fibre form, can have a more substantive contribution to make in the elucidation of the glassy state more broadly.
Acknowledgments
This work is supported by the Council of Scientific & Industrial Research, India under 11th Five Year Plan. Funding from Australian Research Council has also been acknowledged. The authors thank Dr. M. C. Paul and S. Das for design and development of photosensitive fibers NM-113 and NM-41 and for providing the fibers for experimentation.
References and links
1. B. Zhang and M. Kahrizi, “High temperature resistance fiber Bragg grating temperature sensor fabrication,” IEEE Sens. J. 7(4), 586–591 (2007). [CrossRef]
2. S. Bandyopadhyay, J. Canning, M. Stevenson, and K. Cook, “Ultrahigh-temperature regenerated gratings in boron-codoped germanosilicate optical fiber using 193 nm,” Opt. Lett. 33(16), 1917–1919 (2008). [CrossRef] [PubMed]
3. E. Lindner, C. Chojetzki, S. Brückner, M. Becker, M. Rothhardt, and H. Bartelt, “Thermal regeneration of fiber Bragg gratings in photosensitive fibers,” Opt. Express 17(15), 12523–12531 (2009). [CrossRef] [PubMed]
4. S. Trpkovski, D. J. Kitcher, G. W. Baxter, S. F. Collins, and S. A. Wade, “High-temperature-resistant chemical composition Bragg gratings in Er3+-doped optical fiber,” Opt. Lett. 30(6), 607–609 (2005). [CrossRef] [PubMed]
5. M. Fokine, “Formation of thermally stable chemical composition gratings in optical fibers,” J. Opt. Soc. Am. B 19(8), 1759–1765 (2002). [CrossRef]
6. Y. Shen, J. He, Y. Qiu, W. Zhao, S. Chen, T. Sun, and K. T. Grattan, “Thermal decay characteristics of strong fiber Bragg gratings showing high-temperature sustainability,” J. Opt. Soc. Am. B 24(3), 430–438 (2007). [CrossRef]
7. O. V. Butov, E. M. Dianov, and K. M. Golant, “Nitrogen doped silica-core fibres for Bragg grating sensors operating at elevated temperatures,” Meas. Sci. Technol. 17(5), 975–979 (2006). [CrossRef]
8. M. Åslund and J. Canning, “Annealing properties of gratings written into UV-presensitized hydrogen-outdiffused optical fiber,” Opt. Lett. 25(10), 692–694 (2000). [CrossRef]
9. J. L. Archambault, L. Reekie, and P. St. Russell, “100% reflectivity Bragg reflectors produced in optical fibres by single excimer laser pulses,” Electron. Lett. 29(5), 453–455 (1993). [CrossRef]
10. N. Groothoff and J. Canning, “Enhanced type IIA gratings for high-temperature operation,” Opt. Lett. 29(20), 2360–2362 (2004). [CrossRef] [PubMed]
11. D. Grobnic, C. W. Smelser, S. J. Mihailov, and R. B. Walker, “Long-term thermal stability tests at 1000◦C of silica fibre Bragg gratings made with ultrafast laser radiation,” Meas. Sci. Technol. 17(5), 1009–1013 (2006). [CrossRef]
12. Y. Li, M. Yang, D. N. Wang, J. Lu, T. Sun, and K. T. V. Grattan, “Fiber Bragg gratings with enhanced thermal stability by residual stress relaxation,” Opt. Express 17(22), 19785–19790 (2009). [CrossRef] [PubMed]
13. T. Tani, “A study of intensification of latent images in reduction-sensitized emulsions though delayed development,” J. Imaging Sci. 30(2), 41–46 (1986).
14. J. Canning, “Fibre gratings and devices for sensors and lasers,” Laser Photon. Rev. 2(4), 275–289 (2008) (and refs therein.). [CrossRef]
15. J. Canning, M. Stevenson, S. Bandyopadhyay, and K. Cook, “Extreme Silica Optical Fibre Gratings,” Sensors (Basel Switzerland) 8(10), 6448–6452 (2008). [CrossRef]
16. J. Canning, M. Stevenson, J. Fenton, M. Aslund, and S. Bandyopadhyay, “Strong regenerated gratings,” Proc. SPIE 7503, 750326 (2009). [CrossRef]
17. S. Bandyopadhyay, J. Canning, P. Biswas, R. Chakraborty, and K. Dasgupta, “Regeneration of Complex Bragg Gratings,” Proc. SPIE 7503, 750371 (2009). [CrossRef]
18. Y. Li, C. R. Liao, D. N. Wang, T. Sun, and K. T. V. Grattan, “Study of spectral and annealing properties of fiber Bragg gratings written in H2-free and H2- loaded fibers by use of femtosecond laser pulses,” Opt. Express 16(26), 21239–21247 (2008). [CrossRef] [PubMed]
19. M. L. Åslund, N. Nemanja, N. Groothoff, J. Canning, G. D. Marshall, and S. D. Jackson, “A. Fuerbach, and M.J. Withford, “Optical loss mechanisms in femtosecond laser-written point-by-point fibre Bragg gratings”, Opt. Express 16, 14248–14254 (2008).
20. J. Canning, “Fibre gratings and devices for sensors and lasers,” Laser Photon. Rev. 2(4), 275–289 (2008). [CrossRef]
21. J. Canning, S. Bandyopadhyay, P. Biswas, M. Aslund, M. Stevenson, and K. Cook, “Regenerated Fibre Bragg Gratings,” pp. 363–384, in Frontiers in Guided Wave Optics and Optoelectronics, Ed. Bishnu Pal, INTECH, ISBN 978–953–7619–82–4, February (2010)
22. P. J. Heaney, Silica: Physical Behavior, Geochemistry & Materials Applications, Reviews in Mineralogy, Chapter 1, Vol 29 (Ed. P.J. Heaney) (1994)
23. J. A. Kurki, “Development of fabrication technology and measurement system for multimode, single mode and polarization maintaining optical fiber,” Helsinki University of Technology, Doctor of Technology Thesis, Chapter 2, Table 2.4, 1983.
