Discrimination of bending and temperature ... - OSA Publishing

Discrimination of bending and temperature sensitivities with phase-shifted long-period fiber gratings depending on initial coupling strength Young-Geun Han, Ju Han Lee, and Sang Bae Lee Photonics Research Center, Korea Institute of Science and Technology 39-1 Hawolgok-dong, Seongbuk-gu, Seoul 136-791, Korea [email protected]

Abstract: The discrimination of bending and temperature sensitivities based on phase-shifted long-period fiber gratings is discussed. Their spectral evolution during phase-shifted grating formation by UV post-exposure corresponding to their initial coupling strength is also presented. 2004 Optical Society of America OCIS codes: (060.2310) Fiber optics and (060.2370) Fiber optics sensors

References and links 1. 2. 3. 4. 5. 6. 7. 8.

A. M. Vengsarkar, P. J. Lemaire, J. B. Judkins, V. Bhatia, T. Erdogan and J. E. Sipe, “Long-period fiber gratings as band-rejection filters,” J. Lightwave Technol. 14, 58-64 (1996). Y. G. Han, C. S. Kim, K. Oh, U. C. Paek, and Y. Chung, “Performance enhancement of long-period fiber gratings for strain and temperature sensing, IEICE Trans. on Electronics E83-C, 282-286 (2000). H. J. Patrick, C. C. Chang, and S. T. Vohra, “Long period fiber gratings for structural bend sensing,” Electron. Lett. 34, 1773-1775 (1998). Y. G. Han, B. H. Lee, W. T. Han, U. C. Paek, and Y. Chung, “Resonant Peak Shift and Dual Peak Separation of Long Period Gratings for Sensing Application,” IEEE Photon. Technol. Lett. 13, 699-701 (2001). X. Shu, Y. Liu, D. Zhao, B. Gwandu, F. Floreani, L. Zhang, and I. Bennion, “Fiber grating type dependence of temperature and strain coefficients and application to simultaneous temperature and strain measurement,” in Proc. OFS 2002, 83-86 (2002). C. C. Ye, S. W. James, and R. P. Tatam, “Simultaneous temperature and bend sensing with long-period fiber gratings,” Opt. Lett. 25, 1007-1009 (2002). V. Dupray, M. Zeller, W. Zhang, J. Williams, K. Weir, and M. McCall, “Novel UV post-processed fiber Bragg grating sensor for temperature and strain measurements,” in Proc. OFS 2000, 9-12 (2000). M. LeBlanc, S. T. Vohra, T. E. Tsai, and E. J. Friebele, “Transverse load sensing by use of pi-phase-shifted fiber Bragg gratings,” Opt. Lett. 24, 1091-1093 (1999).

1. Introduction Long-period fiber gratings (LPFGs) have a great potential for mass production and the added advantage of low back-reflection and broad bandwidth [1]. LPFGs couple the co-propagating core (HE1,1) mode to the several cladding modes (HE1,m). Since the cladding modes can directly interface with external environments, LPFGs have a high sensitivity to perturbations of external physical parameters and thus have been of much interest for sensing probes for various mechanical quantities like strain, temperature, and bending curvature [1-3]. Recently, simultaneous sensing of multiple perturbations based on LPFGs and fiber Bragg gratings (FBGs) has been developed and reported [4-6]. In this paper, we discuss simultaneous measurement of bending and temperature based on phase-shifted LPFGs. The bending can change the grating period and finally induce the resonant wavelength shift toward the longer wavelength. Two different sections with different photo-induced average indices can induce the peak splitting with temperature change since they have different temperature sensitivities. The resonant peak shift and wavelength spacing change with variation of bending and temperature allows discrimination between the bending and temperature effects, respectively. We also investigate the transmission characteristics of

#4460 - $15.00 US

(C) 2004 OSA

Received 27 May 2004; revised 22 June 2004; accepted 3 July 2004

12 July 2004 / Vol. 12, No. 14 / OPTICS EXPRESS 3204

the phase-shifted LPFGs depending on the initial coupling strength during the process of fabrication by the UV post-exposure on the half section of the grating. 2. Formation of phase-shifted LPFG by exposing a half of the grating region to UV beam Recently, strain and temperature sensors using FBGs with a spectral hole fabricated by UV post-exposure of a half of the grating have been reported [4]. We applied this technique to LPFGs and observed that a phase-shifted LPFG was formed due to the accumulated phase shift induced by the overall increase of the refractive index. Fig. 1(a) shows the schematic of the fabrication of the phase-shifted LPFGs by inducing the refractive index change at the half section of the grating through UV beam exposure. Exposure of the half section to UV beam induces the increase of the photo-induced refractive index in that region and results in the accumulated phase shift and thus the formation of the phase-shifted LPFG. The amount of the phase shift depends on the total fluence of the UV beam exposure that determines the amount of the photo-induced refractive index change. The grating fabrication parameters were: the UV laser energy = 150 mJ/pulse, grating length = 2 cm, and grating period = 400 µm, respectively. The fiber used in the experiment is an ordinary photosensitive fiber with a step index profile and co-dopants of B and Ge. In LPFGs, the mode coupling between the forward propagating core mode (HE11) and the forward propagating cladding modes (HE1m) occurs because of the large grating period more than a few hundred [1]. The initial coupling strength of LPFGs before the UV post-exposure determines the evolution of the transmission spectrum during the formation of the phase-shifted LPFGs. The photo-induced refractive index change consists of the “AC” part as the index modulation and the “DC” part as the average index [4]. The index modulation is directly related to the coupling strength and thus the transmission spectrum of the LPFG. The average index is related with the resonance peak shift of the LPFG. During the grating formation process, the resonance peak shifts toward the longer wavelength due to the increase of the average index while the increase of the index modulation amplitude causes the peak depth. Since the photoinduced refractive index change increases rapidly at the beginning of the grating formation and gradually slows down, the rate of increase of ∆n decreases with the UV fluence. Therefore, the UV post-exposure process results in reduction of the index modulation because the index increase of the unexposed region is higher than that of the exposed region. The average index, however, increases with the UV fluence during the post-exposure process. Decrease of the coupling constant κ with the UV post-exposure can either increase or decrease the peak depth of the LPFG depending on the initial coupling strength since its transmittance is proportional to cos2(κL), where L is the grating length. The phase-shifted LPFGs fabricated by UV post-exposure of a half of the grating are theoretically approximated to two cascaded different grating sections with different photoinduced refractive index changes. Based on the co-directional coupled mode theory, the modal amplitudes of the core and cladding modes after passing through a LPFG with two different photo-induced refractive indices units can be written in matrix form as [4]  a1    a   2 Out

 a1 (0)  ,  a ( 0 )  2 

= T1 ⋅ T2 

where aco and acl represent the modal amplitudes of the core and cladding modes, respectively. Ti is the transfer matrix of LPFGs with the photo-induced refractive index of ∆ni and coupling coefficient of κi, respectively. The transmission characteristics of LPFG during UV postexposure on the half section of the grating can be simply analyzed by changing the photoinduced average index of the section. The full length of gratings including two different sections is 2 cm. Figures 2 and 3 show the theoretical and experimental results of the transmission spectra of the phase-shifted LPFGs by the UV post-exposure of a half of the grating corresponding to their initial coupling strength. For the first saturated LPFG in region A of Fig. 1, the UV post-exposure induces accumulated phase shift in the exposed region of #4460 - $15.00 US

(C) 2004 OSA



the LPFG while decreasing the coupling strength. This effect is shown in Fig. 2. As the total fluence of the UV post-exposure increases, the original loss peak of the spectrum shifts to the longer wavelength due to the increase of the average index. In comparison, for the second saturated LPFGs in region C of Fig. 1(b), the resonant wavelength still shifts to the longer wavelength, but the loss peak depth initially decreases and then increases after passing the point κL = π. This effect is depicted in Fig. 3 as the increase of the depth of the loss peak begins near the wavelength of 1510 nm. The small discrepancy between theoretical and experimental results is caused by the dispersion effects even if their transition characteristics with the UV fluence change are similar. We measured PDL of LPFGs before and after exposing a half of grating to UV beam by using Tektronics PDL measurement. The PDL values of the pristine grating and the phase-shifted L PFG were 0.8 dB and 0.82dB, respectively. There is no critical variation of PDL before and af ter exposing a half of grating to UV beam. (a)

(b) 1.0

UV beam Amplitude mask AC

Transmittance (Cos2(κL))

DC

n core UV beam

DC AC

n core DC : Photo - induced average index AC : Photo - induced index modulation

C

A

0.5

B

0.0

Photo - induced index by UV uniform exposure

0

2

4

6

Detuing Factor (κL)

(a)

(a)

0

0

-5

2

0 J/cm 2 20 J/cm 2 51 J/cm

-10 1400

1450

1500

1550

(b) 0

-5

2

0 J/cm 2 20 J/cm 2 51 J/cm

-10 1400

1450

1500

1550

-10

2 0 J/cm 2 20 J/cm 2 51 J/cm 2 71 J/cm

-15 1480

1500

1520

1540

1560

1580

1600

1 520

15 40

15 60

1 580

160 0

(b) 5 0 -5 -10 -15

2 0 J/cm 2 20 J/cm 2 51 J/cm 2 71 J/cm

1 500

W avelength [nm ]

Fig. 2. (a) The theoretical and (b) experimental results of the transmission characteristics of the phaseshifted LPFG fabricated with the first saturated LPFG (region A of Fig. 1(b)) when the UV fluence increases.

(C) 2004 OSA

-5

148 0

Wavelength [nm]

#4460 - $15.00 US

Transmission [dB]

Transmission [dB]

Fig. 1. Figure 1(a) Schematic of the refractive index change during the grating formation and UV post-exposure and (b) the peak transmittance of LPFG in terms of the product κL of the coupling constant κ and grating length L.: A-the first saturated grating, B-overcoupled grating, and C-the second saturated grating, respectively.

Fig. 3. (a) The theoretical and (b) experimental results of the transmission characteristics of the phase-shifted LPFG fabricated with the second saturated LPFG (region C of Fig. 1(b)) when the UV fluence increases.



3. Simultaneous measurement of bending and temperature using the second saturated and phase-shifted LPFG The phase-shifted gratings are very applicable to the physical sensors [8]. We applied the second saturated and phase-shifted LPFG to the simultaneous measurement of bending and temperature. The applied strain can change the transmission properties like resonant wavelength due to change of the grating period and the stress-optic effect. The dependence of the transmission characteristics of the second saturated and phase-shifted LPFG on the bending curvature was investigated by using the phase-shifted LPFG pair fixed on the translation stage using epoxy with no twist [3, 4]. The fiber was first held flat horizontally with no curvature other than slight sagging due to the weight. The translation stage was then moved inward in the z-direction, and the curvature was estimated from the position of the stage. The scotch tape flags attached to both ends of the fiber indicate bending of the fiber in the y-direction [3, 4]. The flags are pointing in the x-direction to reduce the effect of the undesired bending in the y-direction. When bending is applied to the phase-shifted LPFG, the primary effect is the increase of the grating period due to the induction of the strain and consequently the resonant wavelength shifts to the longer wavelength. Figure 4(a) shows the experimental results of the transmission characteristics of the phase-shifted LPFG with bending curvature change. The left and right resonant wavelengths shifted to the longer wavelength while the wavelength spacing was not changed as seen in Fig. 4(b). After fitting the measured data with a linear function, the slope of the wavelength shift was estimated to be 28.183 nm/m-1 in a bending curvature range from 0.2 m-1 to 1.3 m-1. The grating strength becomes stronger since the structural bending reduces the coupling coefficient (κ) and it increases the transmittance of the overcoupled grating. Currently, we are going to analyze those effects to improve and optimize the coupling strength for the enhancement of the proposed sensing system. On the other hands, the temperature sensitivity of fiber grating strongly depends on grating types since total average indices of FBGs corresponding to grating types are different. It has been reported that the temperature dependence of fiber Bragg gratings (FBGs) can be changed depending on grating types like type I, IA, and IIA [5]. Since the phase-shifted LPFG has two different sections with different photo-induced average indices, the wavelength spacing between left and right peaks decreases as the temperature increases. Figure 5(a) shows the transmission characteristics of the phase-shifted LPFG with temperature change. These results are summarized in Fig. 5(b). The peak wavelength spacing is a linearly decreasing function of temperature change as shown in Fig. 5(b). The peak separation slope was estimated to be 0.013 nm/oC in the temperature range from 20oC to 150oC after fitting the data with a linear function. Therefore, it is clearly evident that the discrimination between bending and temperature sensitivities is straightforward with a single sensing head based on the proposed sensing system 4. Conclusion In summary, we experimentally investigated the discrimination of bending and temperature sensitivities using the phase-shifted LPFG based on the resonant wavelength shift and wavelength spacing change by the bending and temperature change, respectively. The phaseshifted LPFGs was fabricated by UV post-exposure on a half of the grating region. The UV post-exposure has the effect of inducing a positive phase shift for the core mode while decreasing the coupling strength between the core and cladding modes. The spectral evolution of the transmission thus depends on the initial coupling strength of the grating. For the unsaturated or first saturated gratings, the relatively weak index modulation depth is rapidly erased by the UV post-exposure and the spectral evolution shows the resonant wavelength shift to the longer wavelength. Decrease and eventual disappearance of the loss peak depth is also observed. For the second saturated gratings, despite similar trends in the initial spectral evolution, the loss peak disappears and appears again as the coupling strength continues to decrease. This characteristic allows significantly larger phase shift in the second saturated

#4460 - $15.00 US

(C) 2004 OSA



LPFGs than the first saturated or unsaturated LPFG. We applied the phase-shifted LPFG fabricated with the second saturated LPFG to the simultaneous measurement of bending and temperature. The left and right resonant peaks shifted into the longer wavelength due to the variation of the grating period as the bending curvature increased and the wavelength spacing between two peaks was not changed by the bending. However, the wavelength spacing was reduced by the temperature change due to the different photo-induced average indices of two sections. A single sensing probe with phase-shifted LPFGs can be easily applied to the simultaneous measurement of bending and temperature. For the case of sensing scheme using two concatenated LPFGs, the sensing system is so complex that it is difficult to demodulate the sensing signals and the special LPFGs are also required for the discrimination between bending and temperature sensitivities. Therefore, the proposed sensing scheme is advantageous since it can simultaneously measure the temperature and bending with a single sensing head. The presented experimental results could also be useful for tailoring the gainflattening filters and adjusting the resonance wavelength, the loss peak depth, and the passband. (b)

(a) -1

0m

Left Peak Right Peak

1590 -1

0.39 m

-5

Wavelength [nm]

Transmission [dB]

0

-1

0.61 m -10

-1

0.87 m

-15 -20

-1

1.22 m

1530

1575 1560 1545 1530

1560 1590 Wavelength [nm]

1515 0.2

1620

0.4

0.6

0.8

1.0

1.2

-1

Curvature [m ]

Fig. 4. (a) The transmission characteristics and (b) peak wavelength shift of the phase-shifted LPFG fabricated with the second saturated LPFG as the bending curvature increases. Linear fitting was 28.183 nm/m-1. (b) 36.0

(a)

Wavelength [nm]

Transmission [A.U.]

35.5

Linear Fitting o = -0.336 nm/ C

1560 1550

1520 1510

Linear Fitting o = -0.349 nm/ C

1540

1500 1490

1530

35.0

1530

Wavelength [nm]

Wavelength Spacing [nm]

1570

30

60

90

120

150

o

Temperature [ C]

34.5

o

Linear Fitting = -0.013 nm/ C 34.0

1470

1500

1530

1560

Wavelength [nm]

1590

20

40

60

80

100

120

140

160

o

Temperature [ C]

Fig. 5. (a) The transmission characteristics and (b) peak wavelength spacing change of the phase-shifted LPFG with the temperature change. The left and right resonant wavelength shifts were shown in the inset. The wavelength spacing between left and right resonant peaks was reduced by the temperature change (-0.013 nm/οC).

#4460 - $15.00 US

(C) 2004 OSA

