Long-period fiber grating sensor with a styrene ... - OSA Publishing

1 downloads 0 Views 1MB Size Report
C. Tao, X. Li, J. Yang, and Y. Shi, “Optical fiber sensing element based on .... wave optical fiber sensor [6,7], and mode-filtered light optical fiber sensor [8], etc.
Long-period fiber grating sensor with a styreneacrylonitrile nano-film incorporating cryptophane A for methane detection Jianchun Yang,1,* Chuanyi Tao,1 Xueming Li,2 Guangqin Zhu,2 and Weimin Chen1 1

Key Laboratory for Optoelectronic Technology and Systems, Ministry of Education, College of Optoelectronic Engineering, Chongqing University, Chongqing 400044, China 2 College of Chemistry and Chemical Engineering, Chongqing University, Chongqing 400044, China *[email protected]

Abstract: This paper presents a novel sensor design and application of long period fiber grating (LPFG) for detection of methane. A styreneacrylonitrile nano-film incorporating cryptophane A, which is sensitive to methane in close vicinity to the surface, is constructed onto the cladding of long-period grating. For optimal design of the LPFG sensor, the relationship between the resonant wavelength shift and the complex refractive index of sensing film is analyzed based on the coupled-mode theory. The change in refractive index of the sensing film, induced by methane, can easily be obtained as a shift in resonance wavelength. The prepared LPFG sensor with time response of 50 s and good sensitivity (~0.375 nm % 1) suitable for the detection of methane below 3.5 vol. % is demonstrated. The response of the sensor (wavelength shift) is linear with methane concentration within our tested range and a detection limit of about 0.2% is estimated for the new sensor. ©2011 Optical Society of America OCIS codes: (050.2770) Gratings; (060.2370) Fiber optics sensors; (280.4788) Optical sensing and sensors; (120.1880) Detection; (130.6010) Sensors.

References and links A. Cusano, P. Pilla, L. Contessa, A. Iadicicco, S. Campopiano, A. Cutolo, M. Giordano, and G. Guerra, “High sensitivity optical chemosensor based on coated long-period gratings for sub-ppm chemical detection in water,” Appl. Phys. Lett. 87(23), 234105 (2005). 2. Z. Gu and Y. Xu, “Design optimization of a long-period fiber grating with sol–gel coating for a gas sensor,” Meas. Sci. Technol. 18(11), 3530–3536 (2007). 3. S. W. James and R. P. Tatam, “Fiber optic sensors with nano-structured coatings,” J. Opt. A, Pure Appl. Opt. 8(7), S430–S444 (2006). 4. G. Lein, S. Paquette, S. Vadhavkar, L. Fuller, and K. S. V. Santhanam, “Batron P–Si microsensor for methane and its derivatives,” Sens. Actuators B Chem. 142(1), 147–151 (2009). 5. C. Tao, X. Li, J. Yang, and Y. Shi, “Optical fiber sensing element based on luminescence quenching of silica nanowires modified with cryptophane-A for the detection of methane,” Sens. Actuators B Chem. 156, 553–558 (2011). 6. M. Benounis, N. Jaffrezic-Renault, J. P. Dutasta, K. Cherif, and A. Abdelghani, “Study of a new evanescent wave optical fibre sensor for methane detection based on cryptophane molecules,” Sens. Actuators B Chem. 107(1), 32–39 (2005). 7. J. Yang, L. Xu, and W. Chen, “An optical fiber methane gas sensing film sensor based on core diameter mismatch,” Chin. Opt. Lett. 8(5), 482–484 (2010). 8. S. Wu, Y. Zhang, Z. Li, S. Shuang, C. Dong, and M. M. F. Choi, “Mode-filtered light methane gas sensor based on cryptophane A,” Anal. Chim. Acta 633(2), 238–243 (2009). 9. S. Korposh, S. W. James, S.-W. Lee, S. Topliss, S. C. Cheung, W. J. Batty, and R. P. Tatam, “Fiber optic long period grating sensors with a nanoassembled mesoporous film of SiO2 nanoparticles,” Opt. Express 18(12), 13227–13238 (2010). 10. Z. Gu, Y. Xu, and C. Deng, “Optical characteristics of coated long-period fiber grating and their sensing application,” Proc. SPIE 6800, 680013, 680013-8 (2007). 11. N. D. Rees, S. W. James, R. P. Tatam, and G. J. Ashwell, “Optical fiber long-period gratings with LangmuirBlodgett thin-film overlays,” Opt. Lett. 27(9), 686–688 (2002). 12. I. Del Villar, I. R. Matías, and F. J. Arregui, “Influence on cladding mode distribution of overlay deposition on long-period fiber gratings,” J. Opt. Soc. Am. A 23(3), 651–658 (2006). 1.

#147711 - $15.00 USD

(C) 2011 OSA

Received 18 May 2011; revised 24 Jun 2011; accepted 25 Jun 2011; published 15 Jul 2011

18 July 2011 / Vol. 19, No. 15 / OPTICS EXPRESS 14696

13. I. Del Villar, M. Achaerandio, I. R. Matías, and F. J. Arregui, “Deposition of overlays by electrostatic selfassembly in long-period fiber gratings,” Opt. Lett. 30(7), 720–722 (2005). 14. X. Jiang and Z. Gu, “Design of a gas sensor based on a sensitive film coated phase-shifted long-period fiber grating,” J. Opt. 12(7), 075401 (2010). 15. J. Barnes, M. Dreher, K. Plett, R. S. Brown, C. M. Crudden, and H.-P. Loock, “Chemical sensor based on a longperiod fiber grating modified by a functionalized polydimethylsiloxane coating,” Analyst (Lond.) 133(11), 1541– 1549 (2008). 16. I. Del Villar, I. Matías, F. Arregui, and P. Lalanne, “Optimization of sensitivity in long period fiber gratings with overlay deposition,” Opt. Express 13(1), 56–69 (2005). 17. E. Simões, I. Abe, J. Oliveira, O. Frazão, P. Caldas, and J. L. Pinto, “Characterization of optical fiber period grating refractometer with nanocoating,” Sens. Actuators B Chem. 153(2), 335–339 (2011). 18. D. Gloge, “Weakly guiding fibers,” Appl. Opt. 10(10), 2252–2258 (1971). 19. T. Erdogan, “Cladding-mode resonances in short- and long-period fiber grating filters,” J. Opt. Soc. Am. A 14(8), 1760–1773 (1997). 20. C. Tsao, Optical Fibre Waveguide Analysis (Oxford University Press, 1992), Chap. 10. 21. Y. J. Rao, Y. P. Wang, Z. L. Ran, and T. Zhu, “Novel fiber-optic sensors based on long-period fiber grating written by high-frequency CO2 laser pulses,” J. Lightwave Technol. 21(5), 1320–1327 (2003). 22. Y. J. Rao, T. Zhu, Z. L. Ran, Y. P. Wang, J. Jiang, and A. Z. Hu, “Novel long-period fiber grating written by high-frequency CO2 laser pulses and applications in optical fiber communication,” Opt. Commun. 229(1–6), 209–221 (2004). 23. K. J. Laidler, J. H. Meiser, and B. C. Sanctuary, Physical Chemistry, 4th ed. (Houghton Mifflin Co., 2003), Chap. 18. 24. M. Born and E. Wolf, Principles of Optics: Electromagnetic Theory of Propagation, Interference and Diffraction of Light, 7th ed. (Cambridge University Press, 1999), Chap. 2. 25. I. M. White and X. Fan, “On the performance quantification of resonant refractive index sensors,” Opt. Express 16(2), 1020–1028 (2008).

1. Introduction Recently, optical fiber gas sensors have been attracting attention owing to several advantages over conventional electricity-based gas sensors [1–4]. In particular, the combination of optical fibers and cryptophane molecules (Fig. 1) provides a prospect for the fabrication of methane gas sensors with high sensitivity and quick response to targeted methane gas in fields of coalmine production and environmental applications [5–8]. Several approaches based on the interaction of cryptophanes and molecular methane have been developed for methane detection including optical fiber sensing element via luminescence reflection [5], evanescent wave optical fiber sensor [6,7], and mode-filtered light optical fiber sensor [8], etc. However, the signal change of these sensors is dependent on intensity modulation, which is very susceptible to temperature, stability of light source and connector loss.

Fig. 1. Chemical structure of cryptophane A, which is globularly shaped and contain two coneshaped cyclotriveratrylene (CTV) units attached to one another via three O-(CH2)2-O bridges.

Long-period fiber grating (LPFG) is a widely used fiber device with photo-induced periodic modulation of the refractive index (RI) of the core of the fiber [9–13]. A widely exploited characteristic of LPFG is the sensitivity to the RI of the external medium, because the effective indices of the cladding modes show strong dependence on RI of the external medium. The resonant wavelengths will shift in the transmission spectrum when the external refractive index is changed. However, LPFG cannot be used directly for gas detection because the RI of the gas is much smaller than that of the fiber cladding [14]. By coating a transparent polymeric cladding of chemo-sensitive material on the surface of LPFG, the RI-based gas sensors can be fabricated [15]. LPFG modified with Langmuir–Blodgett (LB) organic thin film with a RI of 1.57 was proposed by Rees et al. for chemical vapor sensing. Because the RI

#147711 - $15.00 USD

(C) 2011 OSA

Received 18 May 2011; revised 24 Jun 2011; accepted 25 Jun 2011; published 15 Jul 2011

18 July 2011 / Vol. 19, No. 15 / OPTICS EXPRESS 14697

of the film is higher than that of the cladding of the fiber, the wavelengths of resonance bands show very high sensitivity to the optical thickness of the film, especially when the thickness of the film has an order of few hundred nanometers [11]. A comprehensive theoretical and numerical investigation has demonstrated that high refractive coating was able to favor the transition between cladding guided modes in overlay guided modes, causing a strong redistribution of the cladding modes [13,16]. Based on the triple-clad numerical model and coupled-mode theory, the relationship between sensitivity and the optical parameters of the thin film coated on the surface of LPFG has been analyzed, in order to obtain the maximum resolution [2]. The response characterization of LPFG to variations of RI of the external medium is also studied by Simoes et al. For tricosenoic acid thin film in the RI region around 1.56, values of ~2578 nm RIU 1 for LPFG were obtained [17]. To date, methane sensing film for evanescent wave optical fiber sensor or mode-ltered light optical fiber sensor is mainly constituted by including cryptophane molecules in a transparent polysiloxane cladding with the RI region around 1.42 [6–8]. If the polysiloxane film is directly coated on the LPFG surface, the resonant bands of LPFG are insensitive and cannot be used for detection of methane. In view of the RI of optical material styreneacrylonitrile (SAN) resin close to 1.57, SAN material could be a potential candidate for the sensing film of LPFG methane sensor. In this paper, we report the methane sensing characteristics of LPFG coated with SAN nano-film including cryptophane A. Based on triple-clad model and coupled-mode theory for the LPFG with weakly absorbing coating, the hybrid mode complex eigenvalue equation of the triple-clad waveguide is solved by perturbation method. In the experiment, LPFG methane sensor coated with SAN nano-film including cryptophane A is fabricated, and the response to methane is investigated. The results confirm that the LPFG methane sensor can be highly sensitive to methane with a fast response. 2. Sensing Principle 2.1 Sensor Structure and Theoretical Model The structural diagram of triple-clad LPFG methane-sensitive film sensor model is shown in Fig. 2(a). The sensing film consists of a styrene-acrylonitrile layer incorporating cryptophane A. Figure 2(b) shows the RI profile of LPFG sensor. n1, n2, n3 and n4 are RI of the core, cladding, sensing film and external medium (gas), respectively, where the complex RI n3 = nf + ikf (nf and kf are the real refractive index and extinction coefficient, respectively), and n4 is approx. 1.0. a1, a2 and a3 are radius of the core, cladding and sensing film, respectively. The value (a3 - a2) equals to film thickness h. (a)

(b) Gas  

SAN/cryptophane A sensing film

RI

n3 n1 n2

Core

n4 Cladding Long-period grating

a1

a2 a3

r

Fig. 2. (a) Structural diagram of the triple-clad long-period fiber grating methane sensing film sensor model and (b) refractive index profile.

The LPFG couples the light from the forward propagating mode of the core to a discrete set of co-propagating cladding modes at wavelengths governed by the phase matching condition:

#147711 - $15.00 USD

(C) 2011 OSA

Received 18 May 2011; revised 24 Jun 2011; accepted 25 Jun 2011; published 15 Jul 2011

18 July 2011 / Vol. 19, No. 15 / OPTICS EXPRESS 14698

 res  (nco  ncl( ) ),   1, 2, 3,

(1)

where λres is the resonant wavelength, Λ is the grating period, nco and ncl(ν) are the effective indices of the fundamental core mode and the νth cladding mode of the fiber, respectively. nco could be obtained by solving core mode eigenvalue equation [18]. In a triple-clad fiber grating, the cladding mode propagating constant βcl(ν) = k0ncl(ν) must satisfy the condition k0n4< βcl(ν) < k0n2 (azimuthal order l = 1). The effective indices of the cladding modes ncl(ν) can be obtained from the dispersion equation [2,19,20], which can be expressed as pl22 9122  232  pl23 10342  232 

8 pl23u312342  23 8u2 1034 23    2  6122 342   2 a1a2 n1n3u22  2 a2 a3 n2 n4 u32

  5 Ku3  6   7 Ku3 8    Ju2  2    n n2 n2 n2 32   Ju2  1  Ku32   3  Ku34    4 n2 3   4 n2 3    4 a a 2 a n12n n34 n u u 2 1 1 2 3 1 2 3 4 2 3 (2)       2 2  Ku   8p u     pl22 pl23 232 342 122   1  Ku3  2   25  32 6  122  2l 2 2 23 34 12  n n  a2 a3 n2 n4 u32 3  4   Ju    8u3 9 2312  9 10 232   Ju2  2   4   22 6  28  342  2 a1a2 n1n3u22 n1   n2

where u 2j  k 2 n2j   2   2j

( j  1, 2, 3, 4),

J  J l'  u1a1  / u1 J l  u1a1  , K  Kl' 3 a2  / 3 Kl 3 a2  , pl 2  pl 3 

J l (u2 a2 ) J l (u2 a1 ) J (u a ) J l(u2 a1 ) J (u a ) J l (u2 a1 ) J (u a ) J l(u2 a1 ) ,q  l 2 2 ,r  l 2 2 ,s  l 2 2 , Yl (u2 a2 ) Yl (u2 a1 ) l 2 Yl (u2 a2 ) Yl (u2 a1 ) l 2 Yl (u2 a2 ) Yl (u2 a1 ) l 2 Yl (u2 a2 ) Yl (u2 a1 ) J l (u3 a3 ) J l (u3 a2 ) Yl (u3 a3 )

Yl (u3 a2 )

, ql 3 

J l (u3 a3 ) J l(u3 a2 ) J (u a ) J l (u3 a2 ) J (u a ) J l(u3 a2 ) ,r  l 3 3 ,s  l 3 3 , Yl (u3 a2 ) l 3 Yl (u3 a3 ) Yl (u3 a2 ) l 3 Yl (u3 a3 ) Yl (u3 a2 )

Yl (u3 a3 )

1  u2 pl 2 sl 3  u3 rl 2 rl 3 ,  2  u2 pl 2 ql 3  u3 rl 2 pl 3 ,

 3  u2 ql 2 sl 3  u3 sl 2 rl 3 ,

 4  u2 ql 2 ql 3  u3 sl 2 pl 3 , 5  u2 pl 2 sl 3 / n  u3 rl 2 rl 3 / n ,  6  u2 pl 2 ql 3 / n22  u3 rl 2 pl 3 / n32 , 2 2

 7  u2 ql 2 sl 3 / n22  u3 sl 2 rl 3 / n32 ,

2 3

8  u2 ql 2 qv 3 / n22  u3 sl 2 pl 3 / n32 ,

9  (rl 3  Ku3 pl 3 )(rl 3 / n42  Ku3 pl 3 / n42 ), 10  ( Ju2 pl 2  ql 2 )( Ju2 pl 2 / n22  ql 2 / n12 ),

122   02124 / a12 n12 n22u14u22 ,  232   02234 / a22 n22 n32u22u32 , 342   02344 / a32 n32 n42u32 w44 , 122  k 2 (n12  n22 ), 232  k 2 (n22  n32 ), 342  k 2 (n32  n42 ),  02   2 l 2 / k 2 . Here, k=2π/λ is the free space wave number; Jl, Kl, and Yl are Bessel functions. The effective index of the cladding mode is related to the RI of sensing film. While the LPFG sensor coated with styrene-acrylonitrile films including cryptophane A is exposed to methane gas, the tiny variation of sensing film RI will result in a change of ncl(ν), which produces a shift in resonant wavelength of the transmission spectrum. Accurate computation of the effective index according to the complex eigenvalue equation is very complicated owing to the complex index of refraction of weakly absorbing film (n3) on the surface of the LPFG. Thus, it must be solved by the perturbation approach. The complex root of the cladding mode RI is described by

ncl( )  Re  ncl( )   i Im  ncl( ) 

#147711 - $15.00 USD

(C) 2011 OSA

(3)

Received 18 May 2011; revised 24 Jun 2011; accepted 25 Jun 2011; published 15 Jul 2011

18 July 2011 / Vol. 19, No. 15 / OPTICS EXPRESS 14699

The radii of the core and the cladding (a1 and a2) in this work are 4.15 μm and 62.5 μm, respectively. The refractive indices n1, n2 and n4 are 1.4581, 1.4628 and 1.0, respectively. The incident wavelength λ is 1550 nm, and the grating period Λ is 480 μm. The relationship between resonant wavelength shift (Δλres = λres – λres0) and the RI of sensing film (nf and kf) was analyzed by numerical simulation (see Figs. 3 and 4).

(a)

(b)

Fig. 3. Effects of the RI nf and extinction coefficient kf of sensing film on (a) the real part and (b) the imaginary part of the effective indices of the EH12 (ν = 4) cladding mode ncl (h = 0.5 μm).

2.2 Analyses of the Effective Index of Cladding Mode and Resonant Wavelength Shift From Eq. (1), it can be seen that the resonant wavelength shift Δλres is directly determined by the effective index ncl(ν) of cladding mode. Figure 3 shows the effect of the RI nf and extinction coefficient kf of sensing film on the real part and imaginary part of the effective index ncl of the EH12 cladding mode (ν=4) for the film thickness of 500 nm. A sudden change in the real part of the effective index can be observed in Fig. 3(a) while the RI nf ranging from 1.565 to 1.570. The variation of the cladding mode effective index will result in resonant wavelength shift. Thus, at 1.565~1.570 of RI range, the resonant wavelength shift will be immense. However, it can be seen that the extinction coefficient kf has little impact on the real part of the effective index. On the other hand, the dependence of nf and kf on imaginary part of the effective index is presented in Fig. 3(b), which indicates that, in the nf range of

#147711 - $15.00 USD

(C) 2011 OSA

Received 18 May 2011; revised 24 Jun 2011; accepted 25 Jun 2011; published 15 Jul 2011

18 July 2011 / Vol. 19, No. 15 / OPTICS EXPRESS 14700

1.565~1.570, imaginary effective index exhibits very slight variation with the order of 10 5 when kf ranges from 0.0 to 0.005.

(a) Wavelength shift (nm)

30

20

10

0

-10 1.53

(b)

1.55 1.56 1.57 1.58 Refractive index of the film

1.59

1.6

-3

0 Wavelength shift (nm)

1.54

x 10

-2 -4 -6

-8

0

1 2 3 4 Extinction coefficient of the film

5 -3

x 10

Fig. 4. Effects of (a) the RI nf and (b) extinction coefficient kf on resonant wavelength shifts Δλres of the EH12 mode (h=500 nm, Λ=480 μm).

The effect of the film RI nf on resonant wavelength shift (Δλres) of the EH12 mode for the film thickness of 500 nm is shown in Fig. 4(a), from which we can see a slow decline in wavelength shift (i.e., red-shift of resonant wavelength) when nf increases from 1.53 to ~1.56. While, there exist the jumping of Δλres value with the RI between 1.562 and 1.565, and when the film RI is greater than 1.565, increasing the RI can decrease the shift of resonant wavelength quickly. For sensing film in the RI region around 1.57, the resonant bands have a wavelength red-shift when the film RI decreases, and the sensor sensitivity is approximated to be ~1.5 × 103 nm RIU1. As shown in Fig. 4(b), Δλres varies slowly from 0.0 to 0.006 nm as kf increases from 0.0 to 0.005, revealing a very small shift of the resonant wavelength as a function of kf. Therefore, for LPFG methane sensor, the effect of extinction coefficient of the sensing film is almost negligible, and the tiny variation of the film refractive index nf near 1.57 when exposed to methane gas, should cause a very considerable shift in the resonant wavelength of transmission spectrum. In conclusion, a change in RI of sensing film is induced by methane in close vicinity to the surface, and the change in RI can therefore easily be measured as a shift in resonance wavelength. 3. Experimental 3.1 Fabrication of LPFG LPFGs inscription was performed according to the method previously reported [21,22]. Briefly, the fiber (Corning SMF-28) was scanned by means of a 2-D optical scanner attached to a CO2 laser (10 W full power, 5 kHz frequency, Synrad Inc.) with a computer control. The laser beam was focused to a spot with a diameter of ~50 μm. The focused spot was stepped by

#147711 - $15.00 USD

(C) 2011 OSA

Received 18 May 2011; revised 24 Jun 2011; accepted 25 Jun 2011; published 15 Jul 2011

18 July 2011 / Vol. 19, No. 15 / OPTICS EXPRESS 14701

1 μm to achieve multiple exposures of the high-frequency CO2 laser pulses at a fixed point on the fiber. In our work, the obtained grating period Λ is around 480 μm. 3.2 Dip Coating of LPFG with the Styrene-Acrylonitrile Nano-Film Including Cryptophane A Cryptophane A (Fig. 1) was synthesized from vanillin using a three-step method [5,6]. Optical styrene-acrylonitrile (SAN) resin was purchased from Ashley Polymers, Inc. A SAN nano-film including cryptophane A was deposited on the region of the optical fiber containing the long-period grating (around 20 mm length) by the dip coating technique [6,9]. Firstly, the cladding area of the fiber grating was cleaned by distilled water, pure ethanol and acetone in turn. The fiber was then put into a vacuum drying closet at 60°C for 20 min. The solution of SAN including cryptophane A was obtained by dissolving 150 μmol of cryptophane A and 1.0 g of SAN with RI of ~1.57 in 15 mL of dichloromethane. The thin layer was deposited onto the optical fiber surface by immersing the fiber into the SAN/cryptophane A solution with a dipping rate of about 2 mm s 1. The thickness values obtainable ranged from hundreds of nanometers to a few microns, depending upon the number of SAN/cryptophane A deposition times. The fiber was rinsed with distilled water, and dried by flushing with nitrogen gas after each deposition step. Figure 5 shows a typical scanning electron microscope (SEM; TESCAN, VEGA II LMU) micrograph of the LPFG coated with SAN nano-film including cryptophane A, which indicates that the coating on the cladding of the LPFG is rather uniform with a thickness of about 500 nm.

Optical fiber ca. 500 nm

SAN/cryptophane A film

Fig. 5. SEM micrograph of the cross-section of the fiber coated with SAN/cryptophane A.

SLD light source

OSA

Tee joint T LPFG sensor Outlet

Methane Nitrogen

SMF-28

Gas chamber Helical tubing Mass flow controller PC Fig. 6. Experimental apparatus for the detection of methane.

3.3 Experimental Setup The experimental setup for the sensing system is depicted schematically in Fig. 6. Two mass flow controllers, individually controlled from an electronic unit, were employed to precisely

#147711 - $15.00 USD

(C) 2011 OSA

Received 18 May 2011; revised 24 Jun 2011; accepted 25 Jun 2011; published 15 Jul 2011

18 July 2011 / Vol. 19, No. 15 / OPTICS EXPRESS 14702

control the flow of methane and nitrogen carrier gas. A stainless steel helical tube was fixed to keep gases well mixed. A stainless steel gas chamber, with an inlet and an outlet that allow the gases to flow in and out, was constructed to accommodate the LPFG sensor in the center of the gas chamber. The ends of the long-period grating optical fiber were connected to a light source and a detector, respectively. The light source is a 2 mW SLD with a central wavelength of ~1550 nm and a bandwidth of ~40 nm (Dense Light CO. Ltd.). The optical spectrum analyzer (OSA; Agilent 86140B) with a resolution of 10 pm was used as the detector. The flow of methane was switched to different value every approx. 2 min, and the transmission spectra were recorded every 12 s. The total mass flow rate of carrier gas and sample gas was set at 200 sccm. All spectral measurements were made at atmospheric pressure and room temperature of 25°C. 4. Results and Discussion To evaluate sensing properties of the methane sensor, LPFGs coated SAN nano-film including cryptophane A were prepared and exposed directly to various concentrations of methane gas. According to theoretical analysis, it is possible to select suitable RI of sensing film to obtain a greater shift of the resonant wavelength for the LPFG methane sensor. The RI of SAN nano-film including cryphophane A deposited on a Si wafer substrate was measured using an M-2000U-Xe spectroscopic ellipsometer (J. A. Woollam Co., Inc., US) at 25°C. The RI of the nano-film was estimated to be 1.5695 + i 0.0003 (at a wavelength of 633 nm). 4.1 The Relationship Between Resonant Wavelength Shift and Methane Concentration

Transmission (dB)

-20

-25

-30

-35

CH4: 0%

1540

1545

1550

3.5%

1555 1560 1565 Wavelength (nm)

1570

1575

1580

Fig. 7. Transmission spectra of the sensor before and after exposing to methane. 1.4

Wavelength shift (nm)

1.2

y = 0.375*x - 0.0304 r2=0.992

1 0.8 0.6 0.4 0.2 0

0

0.5

1 1.5 2 2.5 Methane concentration (%)

3

3.5

Fig. 8. Calibration curve between Δλres and the methane concentration.

#147711 - $15.00 USD

(C) 2011 OSA

Received 18 May 2011; revised 24 Jun 2011; accepted 25 Jun 2011; published 15 Jul 2011

18 July 2011 / Vol. 19, No. 15 / OPTICS EXPRESS 14703

Figure 7 shows the transmission spectra of LPFG sensor when exposed to the gas with the methane concentrations of 0.0% and 3.5%, respectively. The resonant band of the LPFG exhibits a wavelength red-shift of ca. 1.11 nm (from 1562.64 nm to 1563.75 nm). Figure 8 presents resonant wavelength shifts (i.e., sensor signal) as a function of the concentration of methane. The resonant wavelength shifts to the longer wavelength when the methane concentration ranging from 0.0 to 3.5% by volume. The plot of calibration curve for the methane sensor exhibits good linearity, with a slope of 0.375 (nm % 1) and a correlation coefficient (r2) of 0.992. In order to explain the methane sensing behavior observed in the LPFG sensor, a simple model based on the Langmuir adsorption isotherm was proposed. For the SAN film incorporating molecules of cryptophane A, there is a limited number of available binding sites (cryptophane A molecules). Suppose that, after equilibrium (the complexation of methane by cryptophane A) is established, a fraction θ of the binding sites is occupied by adsorbed methane molecules; a fraction 1-θ will not be occupied [23]. The equilibrium constant can be written as K; then



K[CH4 ] 1  K[CH4 ]

(4)

so that at very low concentrations (5)   K[CH4 ] On the basis of the theoretical analysis and numerical simulation results, the shift of the resonant wavelength depends strongly on the film RI nf. Thus, the wavelength shift due to the chemical interaction between the sensing film and the methane can be presented as:  δ Δλres   res  δnf

   Δnf ( )  S n  Δnf ( ) 

(6)

where Sn describes the sensitivity against the variations of the effective RI, which is a multivariable function of the film and grating optical parameters, such as nf, h, and Λ, etc. We assume that methane molecule adsorption occurs at constant film thickness h when very low methane concentrations (3.5% in this work) are considered. Furthermore, for given h, Λ and λ, when sensing film RI nf is in the limited range (around 1.57), it can be safely assumed that Sn is an approximate constant, not dependent on h, Λ and λ (see Fig. 4(a)). In particular, as a consequence of the methane molecule adsorption within the sensitive film, the variations of RI are expected due to the film density change as described by the Lorentz-Lorentz equation [24]. Moreover, based on the definition of detection limit made by White and Fan [25], it is anticipated that this methane sensor should be sensitive enough to detect methane at levels below 0.2% v/v. 4.2 Response Time and Recovery Time To measure the sensor response time (defined as the time required for a sensor to reach 90% of signal change), the successive exposure of the sensor to a fixed methane concentration of was investigated. Figure 9 shows the wavelength shifts versus time of the LPFG sensor when it was exposed repeatedly to 3.5% of methane at a mass flow rate of 200 sccm. It can be seen that the sensor absorbs methane rapidly but it desorbs the gas slowly. We can also observe that the signal reaches a stable value when the sensor is exposed to the fixed concentration of 3.5%. The response time of the sensor was found to be 50 s for the concentration of 3.5%. The recovery time (the time needed for a sensor to achieve 90% of the corresponding signal change for desorption of methane) was nearly 70 s. It must be point out that the sensor response was affected after various exposures to methane due to the possible instability of the

#147711 - $15.00 USD

(C) 2011 OSA

Received 18 May 2011; revised 24 Jun 2011; accepted 25 Jun 2011; published 15 Jul 2011

18 July 2011 / Vol. 19, No. 15 / OPTICS EXPRESS 14704

SAN nano-film incorporating cryptophane A. The signal drift was estimate to be less than 1%, indicating a fair reliability of the LPFG sensor. 1.5

Wavelength shift (nm)

3.5% 1

0.5 0% 0 0

2

4 6 Time (min)

8

10

Fig. 9. Sensor signal when exposed repeatedly to pure nitrogen and to 3.5% methane.

4.3 Interference Test The selectivity of the LPFG sensor was investigated by exposing it to N2(99.99%), dry air, O2(99.99%), CO(99.95%), CO2(99.95%), and H2(99.99%), respectively, under equal measuring conditions. The resonant wavelength shifts for all investigated gases are assembled in Table 1. It was found that common potential interferents such as N2, dry air, O2, CO, CO2, and H2 did not interfere significantly with the response of the sensor, which implied perfect discrimination ability of methane among different analytes gas. The change of relative humidity (RH) may interfere with the response of the sensor due to the water absorption of SAN. In order to clarify the effect of relative humidity, the sensing properties of the sensor were measured at 45%RH and 90%RH, respectively. Table 2 shows the variation of the resonant wavelength. The resonant wavelength has a slight blue-shift with the increase of relative humidity. Fortunately, this effect of small change in humidity is not significant. Table 1. Effect of Potential Interferents on the LPFG Sensor Analyte gas (% v/v) N2(99.99) Dry air O2(99.99) CO(99.95) CO2(99.95) H2(99.99) CH4(1.5)/N2(98.5)

Resonant wavelength (nm) (an average of 6 determinations) 1562.60 (as reference value) 1562.56 1562.60 1562.62 1562.62 1562.56 1563.12

Resonant wavelength shift (nm) 0.0 0.04 0.0 0.02 0.02 0.04 0.52

Relative standard deviation of the shift (%) 1.0 3.0 1.5 1.8 1.8 2.7 3.2

Table 2. Effect of the Relative Humidity on the LPFG Sensor Relative humidity (%) (in the air at 25°C) ~0 (dry air) 45 90

Resonant wavelength (nm) (an average of 6 determinations) 1562.57 (as reference value) 1562.49 1562.45

Resonant wavelength shift (nm) 0.0 0.07 0.12

Relative standard deviation of the shift (%) 2.7 1.5 2.4

In addition, to get the maximum shift of the resonant wavelength, it is not only necessary to optimize the RI of SAN nano-films including cryptophane A, but also to improve the methane-sensing sensitivity of the styrene-acrylonitrile nano-film including cryptophane A. #147711 - $15.00 USD

(C) 2011 OSA

Received 18 May 2011; revised 24 Jun 2011; accepted 25 Jun 2011; published 15 Jul 2011

18 July 2011 / Vol. 19, No. 15 / OPTICS EXPRESS 14705

Additionally, the film thicknesses are closely related to the dip coating technique, so it is necessary to control the coating conditions, such as the dip time, temperature, etc. Thus, these experiments are in progress. 5. Conclusions A long-period fiber grating film sensor for methane detection was designed and implemented successfully. Based on the triple-clad LPFG model, the effect of the complex refractive index of sensing film on the resonant wavelength shift is analyzed. Numerical simulation shows that, for given film thickness, the changes of the resonant wavelength shifts are particularly evident within some refractive index range, whereas the influence of extinction coefficient is very small and negligible. The long-period fiber grating methane sensor was fabricated by the inclusion of cryptophane A in the styrene-acrylonitrile cladding, deposited on the long-period grating. The sensing film could be used for sensitive adsorption of methane in surrounding gases, which induced the changes in the refractive index of the film, with a concomitant effect on the transmission spectrum in the LPFG region. The prepared fiber sensor is highly sensitive to 0.2~3.5% of the methane and the response time is very fast within 50 s. The wavelength shifts demonstrate a good linear dependence on the methane concentration in this range. There is almost no interference from dry air, O2, CO, CO2, and H2 on the detection. The new sensor is ideal for the detection of methane in coal mine. Acknowledgments This work was carried out with the financial support of the National Natural Science Foundation of China (No. 60871039), the Natural Science Foundation Project of Chongqing (No. 2010BB4246), the Fundamental Research Funds for the Central Universities (No.CDJXS10122217), and the Science and Technology Development Project of Chongqing (No. 2009AC6157).

#147711 - $15.00 USD

(C) 2011 OSA

Received 18 May 2011; revised 24 Jun 2011; accepted 25 Jun 2011; published 15 Jul 2011

18 July 2011 / Vol. 19, No. 15 / OPTICS EXPRESS 14706