OSA's Digital Library

Optics Express

Optics Express

  • Editor: Michael Duncan
  • Vol. 14, Iss. 4 — Feb. 20, 2006
  • pp: 1388–1394
« Show journal navigation

Suppression of discrete cladding mode resonances in fibre slanted Bragg gratings for gain equalisation

Emmanuel Kerrinckx, Arif Hidayat, Pierre Niay, Yves Quiquempois, Marc Douay, Isabelle Riant, and Carlos De Barros  »View Author Affiliations


Optics Express, Vol. 14, Issue 4, pp. 1388-1394 (2006)
http://dx.doi.org/10.1364/OE.14.001388


View Full Text Article

Acrobat PDF (189 KB)





Browse Journals / Lookup Meetings

Browse by Journal and Year


   


Lookup Conference Papers

Close Browse Journals / Lookup Meetings

Article Tools

Share
Citations

Abstract

In a slanted Bragg grating, coupling between the fundamental guided mode and the counterpropagative cladding modes result in discrete resonances in the transmission spectrum. These resonances are a drawback when Slanted Bragg Gratings are used for gain flattening of fibres amplifiers. A new method based on a chemical etching of the cladding is proposed leading to an overlap of the resonances and a reduction of the amplitude of the modulation. This method can be applied for any value of photo induced modulation amplitude in the SBG.

© 2006 Optical Society of America

1. Introduction

Gain flattening of optical fibre amplifier is important for obtaining higher bandwith in wavelength division multiplexing (WDM) transmissions. It becomes essential in long haul systems involving cascaded optical fibre amplifiers. The insertion of an optical filter, with a loss spectrum that flattens the amplifier gain spectrum, appears to be the best solution [1

1. C. R. Giles and E. Desurvire, “Modeling erbium-doped fiber amplifiers,” J. Light. Technol. 9, 271 (1991) [CrossRef]

]. The use of a UV photowritten grating in the core of the fibre was already reported for short period (straight or slanted) and for long period gratings [2

2. R. Kashyap, “Fiber Bragg gratings,” in Optics and Photonics, (Academic Press, 1999).

, 3

3. I. Riant, “Fiber gratings for gain equalisation,” in Technical Digest of Optical Amplifiers and their Applications and of Bragg Gratings, Photosensitivity and Poling in Glass waveguides, Invited Papers OAA/BGPP’01, (2001)

]. Slanted Bragg gratings are very attractive candidates for gain equalisation because of cost efficiency and low insertion losses [3–5

3. I. Riant, “Fiber gratings for gain equalisation,” in Technical Digest of Optical Amplifiers and their Applications and of Bragg Gratings, Photosensitivity and Poling in Glass waveguides, Invited Papers OAA/BGPP’01, (2001)

]. Slanted Bragg Gratings (SBG) are photowritten by using a UV pattern where the UV fringes are not perpendicular to the fibre axis. This feature enhances the coupling of the fundamental guided mode to the counter-propagated cladding modes, leading to losses within a spectral range of generally a few tens of nanometres. These couplings lead to discrete cladding modes in the transmission spectrum. The back reflection can be reduced significantly by choosing fibre geometry and/or well-designed fibre photosensitivity profile [3

3. I. Riant, “Fiber gratings for gain equalisation,” in Technical Digest of Optical Amplifiers and their Applications and of Bragg Gratings, Photosensitivity and Poling in Glass waveguides, Invited Papers OAA/BGPP’01, (2001)

, 6

6. T. Erdogan and J. E. Sipe, “Tilted fiber phase gratings,” J. Opt. Soc. Am. A. 13, 296–313 (1996). [CrossRef]

, 7

7. L. Brilland, D. Pureur, J. F. Bayon, and E. Delevaque, “Slanted gratings UV-written in photosensitive cladding fibre,” Electron. Lett. 35, 234–235 (1999). [CrossRef]

]. Therefore, Slanted Bragg Gratings (SBG) can be loss filters with very small residual reflectivity. Moreover, the sensitivity of SBG to the bending is very low as compared to long period gratings. Unfortunately, the discrete cladding modes in the transmission spectrum are unsuitable for gain equalisation: the SBG creates a strong “modulation” in the transmission spectrum due to the “cladding-surrounding medium” interface that creates a resonant cavity. The fibre can be immersed in or coated with a medium of refractive index equal to that of the cladding [8

8. G. Laffont and P. Ferdinand, “Sensitivity of slanted fibre Bragg gratings to external refractive index higher than that of silica,” Electron. Lett. 37, 289–290 (2001). [CrossRef]

]. However, the refractive index value of this surrounding medium could evolve in the future, especially in submarine systems. This could then modify the spectrum. Moreover, if the coating detaches from silica, the slight air gap will lead to formation of the discrete peaks or “modulation”.

The modulation can be reduced using a careful design of the grating profile. Two methods have been already proposed i.e. shortening the grating length or chirping the grating period [9

9. I. Riant, C. Muller, T. Lopez, V. Croz, and P Sansonetti, “New and efficient technique for suppressing the peaks induced in fiber slanted Bragg grating spectrum,” Tech. Digest of OFC′ 2000 , (2000) pp 118–120.

]. These two methods prove to be efficient but require high photoinduced refractive index for filtering of a few dB [9

9. I. Riant, C. Muller, T. Lopez, V. Croz, and P Sansonetti, “New and efficient technique for suppressing the peaks induced in fiber slanted Bragg grating spectrum,” Tech. Digest of OFC′ 2000 , (2000) pp 118–120.

].

Throughout this paper, we demonstrate the efficiency of a new method for reducing the “modulation” in Slanted Bragg Grating spectrum. This new method is independent of the photo induced refractive index. The main idea consists in inducing a random change in the cladding diameter along the fibre axis in the SBG location. The random change is achieved by means of a chemical treatment of the cladding.

2. Experimental set-up and results

2mm-long SBG were photowritten in H2 loaded single-mode standard telecommunication fibre (Corning SMF28) using a 244 nm frequency doubled Ar+ CW laser (H2 loading conditions: 140 atmospheres at 23°C during one month). The gratings were written using a phase mask (IBSEN Photonics ref: 41235009). The angle between the grating phase mask vector and the fibre axis was equal to 6°. These values have been used in order to get losses close to 6dB in the transmission spectra. The gratings have been written by scanning the UV beam along the fiber axis.

After the photoinscription, a chemical treatment was applied to the claddings. The part of fibre where the SBG was written was placed between two optical cleaning papers impregnated of hydrofluoric acid HF in order to realize a non uniform etching of the cladding surface. SBG transmission spectra were measured using a mono-frequency tuneable infrared laser (TUNICS) source with 0.01 nm step during the chemical treatment.

3. Results

Figure 1 shows the transmission spectrum of a 6° SBG before the chemical treatment in grey line. The discrete peaks (or ripples) due to the coupling between the fundamental guided mode and the counter-propagating cladding modes are maximum at 1510nm. The losses extend from 1470nm to 1535nm. A small reflection peak is observed at 1540nm due to the use of SMF28 fibre which is not optimised for low back reflection. The purpose of this paper is not to demonstrate a reduction of this reflection. Small reflection can be reduced using a photosensitive cladding [2–3

2. R. Kashyap, “Fiber Bragg gratings,” in Optics and Photonics, (Academic Press, 1999).

, 7

7. L. Brilland, D. Pureur, J. F. Bayon, and E. Delevaque, “Slanted gratings UV-written in photosensitive cladding fibre,” Electron. Lett. 35, 234–235 (1999). [CrossRef]

].

The measurements of the amplitude (peak to peak) of the “modulation” (discrete peaks) were recorded in the course of the chemical treatment. Figure 2 shows clearly that the wavelength of each maximum or minimum of the modulation decreases during the chemical treatment. In addition, the Full Width at Half Maximum (HWHM) of the peak in Fig. 2 increases during the chemical treatment. Figure 1 (solid line) shows a typical transmission spectrum at the end of the chemical treatment. The ripples (modulation) of the losses disappear on the transmission spectrum.

Fig. 1. Normalised transmission spectrum of a 2 mm long 6° blazed Bragg grating into a standard singled mode fibre (Corning SMF28). The grating is uncoated and surrounded by air. In grey the transmission spectrum before treatment and in black after treatment. The peaks due to discrete cladding mode coupling can clearly be observed.
Fig. 2: Record of the amplitude of the modulation for different chemical treatment times (t=0 filled diamonds, t=45sec empty diamonds, t=1min45sec filled squares, t=10 min empty squares). Full line with diamonds is the pristine sbg transmission.

The fibre was then cut at the grating region in order to evaluate the impact of the chemical etching. A typical result is shown in Fig. 3 where the diameter of the fibre was observed using a microscope.

Fig. 3. Picture of a fibre diameter after chemical treatment.

As one can see on Fig. 3, the diameter of the fibre is slightly reduced in comparison to the original 125microns. Measurements performed on different locations along the z axis of the fibre using a microscope lead to diameters between 110 microns and 120 microns. This means that the etching of the cladding was not uniform along the grating direction leading to changes in the cladding diameter. The diameter measurements show a random distribution along the z axis. This chemical treatment was applied to several samples and led to the same observations.

The evolution of the normalised amplitude is reported on Fig. 4. We can observe a rapid decrease in the amplitude of the ripples in the first minute of chemical treatment then the amplitude of the ripples reduces to a value close to zero after 11 minutes of chemical treatment. The FWHM and peak to peak amplitude were also recorded carefully during the chemical treatment of several SBG. A typical result is reported on Fig. 4.

Fig. 4. Normalised peak to peak amplitude of the ripple and resonances FWHM as a function of the chemical treatment

It shows that the FWHM of the peak increases exponentially with the chemical treatment time. After 10 minutes of chemical treatment, the FWHM is close to 0.7nm for a separation of 0.9nm of the peaks. Then, the overlap between the peaks increases with the chemical treatment time leading to a strong reduction of the amplitude of the “modulation”.

Figure 3 shows clearly that the shape of the fiber remains close to a cylindrical geometry. Estimation of the influence of a change in the cylindrical symmetry was performed by side polishing the fibre at the grating location. The amplitude was reduced due to side polishing at the grating place. Figure 5 shows the reduction of the modulation amplitude when the fibre is side polished for 14 microns (black line). The transmission spectrum of the pristine fibre is shown in grey for comparison.

Fig. 5. Normalised transmission spectrum of a 2 mm long 6° blazed Bragg grating into a standard singled mode fibre (Corning SMF28). The grating is uncoated and surrounded by air. In grey the transmission spectrum before polishing and in black after polishing.

As it can be seen, the contrast of the fringes is only reduced by a factor of about 2 despite the fact that the cylindrical symmetry does not exist any more (as shown in the inset of Fig 5). Therefore, the diminution of the ripple amplitude is not due mainly to a modification of the cylindrical symmetry. Another explanation can be suggested: the resonance wavelength is a function of the diameter of the cladding. The evolution of the resonance position was calculated as a function of the cladding radius, supposing that the fibre remains cylindrical. The calculation was performed using the coupling mode theory. Figure 6 shows the calculated shift (full line) in a resonance wavelength as the cladding diameter was reduced. The evolution of the spectral shift was recorded during the chemical treatment. At the end of the chemical treatment, the fibre was cut at the grating place and the diameter was measured equal to 110µm with a precision of 4 microns using an optical microscope. This measurement was then used to calibrate the curve of the Fig. 6.

Comparison between these two evolutions permits to determine the speed of the chemical effect. By this way, we find that the cladding diameter reduction speed was approximately of 0.82 μm/min. This value represents a mean estimation. The good agreement between simulation and measurements gives credit to our explanation. The shift is equal to 0.43 nm/µm. This huge value indicates that the resonance wavelength shift has a strong influence on the reduction of the ripple in the transmission spectrum. Indeed, a different pattern of modulation corresponds to each diameter of the cladding. Accordingly, the random change of the cladding diameter after the chemical treatment (a few microns) leads to a jamming of the resonance wavelength and a reduction of the modulation in the spectral range of the SBG. Continuous variations of the cladding diameter may also result in the same spectrum. For instance, writing the SBG into a taper will results in a reduction of the ripples.

Fig. 6. Evolution of the wavelength of a peak as a function of the etching depth

4. Conclusion

A new experimental method is reported for suppressing the peaks induced by the discrete cladding mode coupling in slanted Bragg gratings. HF chemical treatment was applied to the cladding of the fibre at the grating region. This chemical treatment changes the diameter of the cladding not uniformly along the z axis of the fiber. This non uniformity broadens the peak of each resonance. Moreover, the reduction of the cladding diameter during the chemical treatment is responsible of a shift in the resonance wavelengths leading to a reduction of the amplitude of the ripples to the profit of a continuous value of the transmission coefficient. The combination of these two effects results in the disappearance of the discrete peaks. This method proves its efficiency and can be applied for any value of modulation amplitude photoinduced in the SBG.

Acknowledgment

This work has been partially supported by the “Ministère Chargé de la Recherche”, the “Région Nord/Pas de Calais” and the “Fonds Européens de Développement Economique des Régions”.

References and links

1.

C. R. Giles and E. Desurvire, “Modeling erbium-doped fiber amplifiers,” J. Light. Technol. 9, 271 (1991) [CrossRef]

2.

R. Kashyap, “Fiber Bragg gratings,” in Optics and Photonics, (Academic Press, 1999).

3.

I. Riant, “Fiber gratings for gain equalisation,” in Technical Digest of Optical Amplifiers and their Applications and of Bragg Gratings, Photosensitivity and Poling in Glass waveguides, Invited Papers OAA/BGPP’01, (2001)

4.

R. Kashyap, R. Wyatt, and R. J. Campbell, “Wideband gain flattened erbium fibre amplifier using a photosensitive blazed grating,” Electron. Lett. 29, 154–156 (1993) [CrossRef]

5.

R. Kashyap, R. Wyatt, and P. F. McKee, “Wavelength flattened saturated erbium amplifier using multiple side-tip Bragg gratings,” Electron. Lett. 29, 1025–1026 (1993). [CrossRef]

6.

T. Erdogan and J. E. Sipe, “Tilted fiber phase gratings,” J. Opt. Soc. Am. A. 13, 296–313 (1996). [CrossRef]

7.

L. Brilland, D. Pureur, J. F. Bayon, and E. Delevaque, “Slanted gratings UV-written in photosensitive cladding fibre,” Electron. Lett. 35, 234–235 (1999). [CrossRef]

8.

G. Laffont and P. Ferdinand, “Sensitivity of slanted fibre Bragg gratings to external refractive index higher than that of silica,” Electron. Lett. 37, 289–290 (2001). [CrossRef]

9.

I. Riant, C. Muller, T. Lopez, V. Croz, and P Sansonetti, “New and efficient technique for suppressing the peaks induced in fiber slanted Bragg grating spectrum,” Tech. Digest of OFC′ 2000 , (2000) pp 118–120.

OCIS Codes
(050.2770) Diffraction and gratings : Gratings
(060.0060) Fiber optics and optical communications : Fiber optics and optical communications
(060.2340) Fiber optics and optical communications : Fiber optics components

ToC Category:
Diffraction and Gratings

History
Original Manuscript: December 13, 2005
Revised Manuscript: February 3, 2006
Manuscript Accepted: February 3, 2006
Published: February 20, 2006

Citation
Emmanuel Kerrinckx, Arif Hidayat, Pierre Niay, Yves Quiquempois, Marc Douay, Isabelle Riant, and Carlos De Barros, "Suppression of discrete cladding mode resonances in fibre slanted Bragg gratings for gain equalisation," Opt. Express 14, 1388-1394 (2006)
http://www.opticsinfobase.org/oe/abstract.cfm?URI=oe-14-4-1388


Sort:  Author  |  Year  |  Journal  |  Reset  

References

  1. C. R. Giles, and E. Desurvire, "Modeling erbium-doped fiber amplifiers," J. Lightwave Technol. 9, 271 (1991) [CrossRef]
  2. R. Kashyap, "Fiber Bragg gratings," in Optics and Photonics, (Academic Press, 1999).
  3. I. Riant, "Fiber gratings for gain equalisation," in Technical Digest of Optical Amplifiers and their Applications and of Bragg Gratings, Photosensitivity and Poling in Glass waveguides, Invited Papers OAA/BGPP’01, (2001)
  4. R. Kashyap, R. Wyatt and R. J. Campbell, "Wideband gain flattened erbium fibre amplifier using a photosensitive blazed grating," Electron. Lett. 29, 154-156 (1993) [CrossRef]
  5. R. Kashyap, R. Wyatt and P. F. McKee, "Wavelength flattened saturated erbium amplifier using multiple side-tip Bragg gratings," Electron. Lett. 29, 1025-1026 (1993). [CrossRef]
  6. T. Erdogan and J. E. Sipe, "Tilted fiber phase gratings," J. Opt. Soc. Am. A. 13, 296-313 (1996). [CrossRef]
  7. L. Brilland, D. Pureur, J. F. Bayon and E. Delevaque, "Slanted gratings UV-written in photosensitive cladding fibre," Electron. Lett. 35, 234-235 (1999). [CrossRef]
  8. G. Laffont and P. Ferdinand, "Sensitivity of slanted fibre Bragg gratings to external refractive index higher than that of silica," Electron. Lett. 37, 289-290 (2001). [CrossRef]
  9. I. Riant, C. Muller, T. Lopez, V. Croz, and P Sansonetti, "New and efficient technique for suppressing the peaks induced in fiber slanted Bragg grating spectrum," Tech. Digest of OFC'  2000, (2000) pp 118-120.

Cited By

Alert me when this paper is cited

OSA is able to provide readers links to articles that cite this paper by participating in CrossRef's Cited-By Linking service. CrossRef includes content from more than 3000 publishers and societies. In addition to listing OSA journal articles that cite this paper, citing articles from other participating publishers will also be listed.


« Previous Article  |  Next Article »

OSA is a member of CrossRef.

CrossCheck Deposited