- Email: [email protected]
Multicore microstructured optical fibre for sensing applications
Optics Communications 344 (2015) 71–76
Contents lists available at ScienceDirect
Optics Communications journal homepage: www.elsevier.com/locate/optcom
Multicore microstructured optical fibre for sensing applications L. Sójka a,c,n, L. Pajewski a, M. Śliwa a, P. Mergo b, T.M. Benson c, S. Sujecki c, E. Bereś-Pawlik a a b c
Telecommunications and Teleinformatics Department, Wroclaw University of Technology, Wybrzeze Wyspianskiego 27, 50-370 Wroclaw, Poland Laboratory of Optical Fibre Technology, Marie Curie-Skłodowska University, Pl. M. Curie-Sklodowskiej 3, 20-031 Lublin, Poland George Green Institute for Electromagnetics Research, University of Nottingham, University Park, Nottingham NG7 2RD, UK
art ic l e i nf o
a b s t r a c t
Article history: Received 17 May 2014 Received in revised form 31 December 2014 Accepted 3 January 2015 Available online 6 January 2015
In this contribution we present the sensing capabilities of a novel N-path Mach–Zehnder interferometer (MZI) that relies on a multicore microstructured optical fibre (MC-MOF) connected between two sections of standard single mode fibre. The modal properties of the MC-MOF structure are analysed experimentally by measuring near field profiles. The dependence of the N-path MC-MOF MZI sensitivity on temperature, tensile strain and bending is investigated. The results suggest that such an interferometer is a good candidate for a tensile strain or bending sensor. Crown Copyright & 2015 Published by Elsevier B.V. All rights reserved.
Keywords: Microstructured fibre Multicore fibre In-line Mach Zehnder interferometer
1. Introduction Optical fibre sensors belong to the most intensively studied subjects in photonics in recent years. The reason for this is that they can find many applications in biological, chemical and environmental industries. Furthermore, optical fibre sensors have many advantages, which include electromagnetic immunity, high sensitivity and simple and robust construction [1–3]. One of the most widely used types of optical fibre sensors is all-fibre Mach– Zehnder interferometers (MZIs) [1–4]. This is because connecting an all-fibre MZI with other optical fibre components is a simple and straightforward task. The operating principle of this device in essence consists in measuring the phase shift between two or more optical arms. The factors that change the phase shift can be related to some useful measurands like the mechanical strain, pressure, elongation, bending or temperature. In the literature a number of all-fibre MZI designs can be found. A specific type of the MZI is an in-line MZI whereby all interferometer arms are integrated into one optical fibre. For example in-line configurations can exploit twin core fibres [5,6], air hole collapsing of microstructured optical fibre (MOF) [7,8], core mismatch [9,10], a multimode fibre segment [11] or fibre tapering [4]. These configurations exhibit good performance when sensing strain, temperature and bending. Furthermore, in-line fibre MZIs are simple, cost effective and easy to assemble. These features are essential for the n Corresponding author at: Telecommunications and Teleinformatics Department, Wroclaw University of Technology, Wybrzeze Wyspianskiego 27, 50-370 Wroclaw, Poland. E-mail address: [email protected] (L. Sójka).
http://dx.doi.org/10.1016/j.optcom.2015.01.005 0030-4018/Crown Copyright & 2015 Published by Elsevier B.V. All rights reserved.
practical application of the MZIs. Recently in the literature the realisation of novel types of inline MZI, based on multicore fibres, has been reported by several research groups [12–15]. The multi-core microstructured optical fibre (MC-MOF) MZI acts as an N-path interferometer. It has been proven that multi-path interferometers have enhanced sensitivity when compared with conventional two path interferometers [16]. However, this is a very new topic and one which in our opinion has not been fully addressed in the literature. A bending sensor that utilises a seven-core MOF has been recently proposed in [15]. These authors used a lateral offset of the coupling fibres to invoke interference between two supermodes having different effective refractive index. In contrast, in this contribution we analyse the modal properties of MC-MOF using an experimental approach. Furthermore, results presented in the literature shows that in order to obtain equal intensity distribution in each core the careful optimisation of the multicore fibre structure is needed [17,18]. We have shown that we were able to achieve the equal intensity distribution for our MC-MOF by coupled the light with a SMF-28. In this paper experimentally investigate the sensing properties of an N-path interferometer based on a multicore silica-based microstructured optical fibre. We also analyse the modal properties of the MC-MOF using an experimental approach. Whilst our fabricated MC-MOF waveguide structure is similar to the structure presented in [15], our waveguide lattice parameters are significantly different from the ones reported in [15]. In the present paper the N-path MZI was realized by inserting a section of a MCMOF between two sections of a standard single mode fibre (SMF28). Sensing capabilities for the temperature, tensile strain and bending of the MC-MOF were characterised. Unlike [15], we do not
72
L. Sójka et al. / Optics Communications 344 (2015) 71–76
offset the coupling fibres; this simplifies the setup. The results obtained show that the proposed N-path MC-MOF MZI can provide good sensitivity for measurement of tensile strain or bending, with the benefit of low sensitivity to operating temperature. It retains the additional benefit of compactness that in-line MZIs offers over conventional MZIs. The results are presented as follows: in Section 2 the procedures used for the fabrication and initial characterisation of the MC-MOF are described. Section 3 presents the experimental evaluation of the sensing behaviour of an N-path MZI based on the MC-MOF fabricated. Conclusions are drawn in Section 4.
2. Characterisation of MC-MOF 2.1. Fabrication of MC-MOF The cross-sectional structure of the manufactured seven-core microstructured silica-based optical fibre is presented in Fig.1. The fibre was made using a stack-and-draw technique. The fibre preform was created by stacking capillaries around a solid rod. Each core was produced using a single rod. The MC-MOF has been polymer coated. Table 1 presents the parameters of the fabricated MC-MOF. 2.2. Near-field measurements The mode field properties in MOFs strongly depend on the lattice parameters. Therefore, it is important to know the exact mode field profile of a MC-MOF. The near field mode intensity of the MC-MOF fabricated was measured experimentally at a signal wavelength of 630 nm. Each core in MC-MOF is endlessly singlemode because the ratio d/Λ o 0.43 [19]. The light from a laser was introduced into a 2 m length of the MC-MOF by connecting both ends of the MC-MOF to a single mode fibre SMF-28 using conventional a FC/PC fibre connectors. The near-field image at the MCMOF output was captured using a CCD camera and the mode field profile obtained is shown in Fig. 2a. The intensity plot of the mode field profile is given in Fig. 2b. These results indicate that the light from SMF-28 can effectively excite all cores of the MC-MOF
Fig. 1. Cross-section of the silica MC-MOF fabricated.
Table 1 Parameters of the MC-MOF fabricated. Core diameter Air-hole diameter, d Lattice constant, Λ (pitch) Filling factor, d/Λ
5.15 mm 1.5 mm 3.6 mm 0.413
without coupling a significant proportion of the input power to the cladding modes. It has been proven in the literature that is possible to achieve equal intensities in each core of 7-core MOF by careful optimisation of the fibre structure [17,18,20]. Each core is slightly different in practice due to manufacturing differences. Therefore this simple configuration can be used to realise a sensor, which exploits the interference that occurs between the core guided modes. Based on this principle the MC-MOF-MZI was implemented. We excite all cores of the multicore fibre with the SMF-28 aligned with the centre of the MC-MOF.
3. The experimental results for N-path MC-MOF MZI 3.1. Principle of operation Schematic and illustrative diagrams of the realised N-path MCMOF MZI are depicted in Fig. 3a and b, respectively. The MC-MOF cores are not exactly identical due to fabrication inaccuracies this results in difference between the optical path length of the interferometers arm. The interferometer was fabricated by inserting 2 m section of MC-MOF between two sections of SMF-28 using conventional FC/PC fibre connectors. An overall MC-MOF length of 2 m was used in all experiments, however the MC-MOF section exposed to external factors (temperature, strain, bending) was slightly altered in each experiment. In each case the MC-MOF length exposed to temperature, strain or bending is specified. The fibre was coated in all experiments. The SMF-28 at the beginning and at the end of MC-MOF acts as an input and output coupler, respectively. The light from the SMF-28 excites several modes in the MC-MOF. A fibre with seven coupled cores supports seven supermodes [21,22]. At the end of the MC-MOF section the light from each mode is coupled back to SMF-28. Multi-mode interference occurring in this fibre. Each supermodes acts as a separate arm of a MZI. The coupling efficiency depends on the relative phase differences between the optical beams propagating in the cores of the MC-MOF, since each mode acts as a separate arm of a MZI. Spectrally, the interference generates a periodic transmission spectrum of the N-path MZ device. The fabricated MC-MOF MZI is simple and compact, in contrast to a conventional MZI assembly where the two optical fibre couplers and two fibre sections are needed. In order to investigate the transmission properties of the MCMOF MZI, broadband light from an erbium amplified spontaneous emission (ASE) source was introduced into the fibre. The transmission spectrum of the MC-MOF MZI was monitored using an Optical Spectrum Analyser (OSA), Yokogawa AQ6370C, with a resolution of 0.02 nm. The experimental set-up is shown in Fig. 4. Since SMF-28 is used at both ends of the sensor further coupling with standard fibre optic devices is straightforward. The insertion loss of the MCF device was obtained as 1.89 dB. Fig. 5 presents the interference fringe pattern observed at the output of the MC-MOF MZI near the wavelength 1550 nm. A maximum contrast of around 10 dB is observed. The measured spacing between the interference peaks was around 0.8 nm. It can be observed that the interference fringe pattern is not uniform like in a conventional two-path MZI. The sidelobes in the spectrum are characteristic of an N-path interferometer, and are due to the superposition of the multiple
L. Sójka et al. / Optics Communications 344 (2015) 71–76
73
Fig. 2. (a) Near-field image obtained for MC-MOF at a wavelength of 630 nm, (b) intensity plot obtained by post-processing the image data presented in (a) using Matlab.
Fig. 3. (a) Schematic diagram of N-path MC-MOF MZI. (b) Illustrative diagram of N-path MC-MOF MZI coupled to SMF.
Fig. 4. MC-MOF MZI experimental set-up.
3.2. Temperature sensitivity
Fig. 5. Transmission spectrum of the MC-MOF MZI at wavelengths around 1550 nm.
beams guided by the interferometer arms [14–16]. In addition, the MC-MOF MZI can also behave like a comb filter and hence it potentially provides an additional wavelength selection mechanism for a fibre laser [5,6,11].
When the external temperature experienced by the MC-MOF is increased the fibre can elongate due to thermal expansion. Also the refractive index changes due to the thermo-optic effect. Consequently the optical beam path for each MC-MOF core can change [6]. Any resulting change in relative phase will manifest itself as a wavelength shift in the transmission spectrum [1–4]. The experimental setup is composed of the ASE erbium source working at 1550 nm, the N-path MC-MZI and a Peltier element. A 80 mm section of the MC-MOF was placed on the Peltier element. The temperature was controlled in the range spanning from 7 °C up to 85.6 °C with a step resolution of 10 °C. The temperature variations were achieved by changing the current that passes through the Peltier element. The output spectrum of the interference fringe pattern was observed using the OSA. The temperature sensitivity was measured for a coated fibre. The dependence of the wavelength of a selected dip in the transmission spectrum is given as a function of temperature in Fig. 6a. Fig. 6b shows the dependence of the output power on the wavelength at selected temperatures. It can be seen that the wavelength at which the selected peak in the transmission spectrum occurs remains almost unchanged. This behaviour can be explained by the low thermal expansion coefficient of the MC-MOF since the MC-MOF is made from silica and air only [23]. Pure silica has a low thermo-optic
74
L. Sójka et al. / Optics Communications 344 (2015) 71–76
Fig. 6. (a) Wavelength of a selected dip in transmission versus temperature (the red line indicates the best fit to measured data), (b) wavelength variation of optical power of the selected transmission dip for various temperatures. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
Fig. 7. (a) Wavelength of selected dip in transmission versus tensile strain (the red line indicates the best fit to measured data). (b) Spectral dependence of transmitted power around the selected dip in transmission spectrum as a function of tensile strain. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
higher (typically 10 pm/ oC ) than for our MC-MOF MZI [1,2]. 3.3. Tensile strain sensitivity
Fig. 8. Experimental set-up for measuring the bending properties of MC-MOF MZI.
coefficient [3]. The experimental results allow the temperature sensitivity of the MC-MOF to be estimated as 1.6 pm/ oC .This low sensitivity to temperature feature is an advantage in many fibre sensor applications because it makes tensile strain and bending measurement relatively independent of the ambient temperature. For comparison, standard fibre Bragg grating (FBG) sensors are very sensitive to the temperature changes. For example the temperature sensitivity for standard FBG is one order of magnitude
When a tensile strain is applied to the MC-MOF, the length of the MC-MOF and the effective refractive indices of the modes supported by the MC-MOF are changed, which results in changes of the relative phase velocities of the beams in the MC-MOF MZI arms. In order to determine the tensile strain sensitivity of the MCMOF MZI one end of the MC-MOF was fixed and the other one was attached to a translation stage. Subsequently the MC-MOF was stretched along the fibre axis by moving the translation stage in steps of 50 mm. The tensile strain was calculated from
ΔL (1) L where L ¼110 mm was the MC-MOF length used in the experiment S=
L. Sójka et al. / Optics Communications 344 (2015) 71–76
75
Fig. 9. (a) Wavelength of selected dip in transmission versus curvature (the red line indicates the best fit to the measured data). (b) Spectral dependence of transmitted power around the selected dip in transmission spectrum for various curvatures. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
and ΔL ¼0.350 mm was the maximum displacement of the translation stage. Fig.7a presents the dependence of the wavelength of a selected dip in the transmission spectrum on the applied strain. The dip has undergone a blue-shift of around 0.35 nm when the strain was increased from 0 to 3500 mε (micro strains). Fig. 7b shows the wavelength dependence of the selected dip in the transmission spectrum on the applied strain. It is important to note that the dependence of the wavelength dip position on the applied strain is linear. This is unlike many standard resistive strain gauges [24]. The measured sensitivity to the tensile strain is around 0.08695 pm/mε. This sensitivity is one order of magnitude less than that of fibre sensors based on FBGs, which typically is equal to 1.0 pm/mε. However, it is only two times less than the sensitivity of fibre sensors that exploit highly birefringent microstructured optical fibre (HB-MOFs) [24].
transmission spectrum for several values of curvature within the range from 2 up to 7 m 1.
4. Conclusions In this paper we have investigated the sensing capabilities of a MC-MOF. The N-path Mach Zehnder interferometer was fabricated using a section of a MC-MOF and two single mode fibre sections (SMF-28). The experimental results show that the MC-MOF sensor is insensitive to the temperature changes. A linear response to the applied tensile strain and bending was observed. These results demonstrate that the MC-MOF MZI is robust. Furthermore, such a sensor overcomes the cross-sensitivity problem between tensile strain or bending and temperature. In order to obtain better results we plan in the future to focus on improvement of the multicore fibre waveguide structure.
3.4. Bending sensitivity The experimental set-up used to investigate the bending sensitivity of the MC-MOF MZI is shown in Fig. 8. A 99.8 mm section of the MC-MOF was placed on two translation stages. One of the stages was fixed, while the other one was able to move inwards to change the separation distance between the stages. In such a way, different curvature could be applied to on the MC-MOF section. The change of the curvature alters the optical path, resulting in different propagation conditions for each mode that propagates in the MC-MOF. Therefore the variation of the curvature will lead to a wavelength shift of the dips in the transmission spectrum. The fibre curvature was approximated using the expression [25]:
C=
1 ≅ r
24x L3
(2)
where r is the radius of the bent fibre, x is the displacement value of the moveable fibre end and L is the fibre length between the translation stages. Fig. 9a presents the shift in the wavelength of a specific dip in the transmission responses of the MC-MOF MZI for different values of the curvature C. It should be noted that again the dependence of this wavelength on the bending curvature can be approximated using a linear function. The measured curvature sensitivity is 0.02465 nm/m 1. Fig. 9b shows the spectral dependence of transmitted power around the selected dip in the
Acknowledgements This research has been partly supported by the statutory funds of Telecommunications and Teleinformatics Department, Wroclaw University of Technology, Poland. One of the authors (L.S.) is very thankful to the Wroclaw University of Technology for the fellowship co financed by European Union within European Social Fund.
References [1] A.D. Kersey, Opt. Fiber Technol. 2 (1996) 291. [2] A.D. Kersey, M.A. Davis, H.J. Patrick, M. LeBlanc, K.P. Koo, C.G. Askins, M. A. Putnam, E.J. Friebele, J. Lightw. Technol. 15 (1997) 1442. [3] O. Frazão, J.L. Santos, F.M. Araújo, L. Ferreira, , Laser Photon. Rev. 2 (2008) 449. [4] B.H. Lee, Y.H. Kim, K.S. Park, J.B. Eom, M.J. Kim, B.S. Rho, H.Y. Choi, Sensors 12 (2012) 2467. [5] G. Yin, Lou S.,H. Zou, Laser Phys. Lett. 10 (2013) 045103. [6] L. Shuhui, L. Ningliang, W. Ying, G. Jiangtao, L. Zhihua, L. Peixiang, L. IEEE, Photon. Technol. Lett. 24 (2012) 1768. [7] J.M. Sierra-Hernandez, R. Rojas-Laguna, E. Vargas-Rodriguez, J.M. EstudilloAyal, R.I. Mata-Chavez,, D. Jauregui-Vazquez, J.C. Hernandez-Garcia, J. A. Andrade-Lucio, J.C. Gutierrez-Gutierrez, Laser Phys. 23 (2013) 055105. [8] W.G. Chen, S.Q. Lou, S.C. Feng, L.W. Wang, H.L. Li, T.Y. Gou, S. Jian, Laser Phys. 19 (2009) 2115. [9] L.V. Nguye, D. Hwang, S. Moon, D.S. Moon, Y. Chung, Opt. Express 16 (2008) 11369. [10] J. Zhou, C. Liao, Y. Wang, G. Yin, X. Zhong, K. Yang, B. Sun, G. Wang, Z. Li, Opt.
76
L. Sójka et al. / Optics Communications 344 (2015) 71–76
Express 22 (2014) 1680. [11] L. Ma, Z. Kang, Y. Qi, S. Jian, Laser Phys. 24 (2014) 045102. [12] P.M. Blanchard, J.G. Burnett, G.R. Erry, A.H. Greenaway, P. Harrison, B. Mangan, J.C. Knight, P.StJ. Russell, J. Gander, R. McBride, J.D. Jones, Smart Mater. Struct. 9 (2000) 132. [13] W.N. MacPherson, M.J. Gander, R. McBride, J.D. Jones, P.M. Blanchard, J. G. Burnett, A.H. Greenaway, B. Mangan, T.A. Birks, J.C. Knight, P.St. J. Russell, Opt. Commun. 193 (2001) 97. [14] Z. Zhao, M. Tang, S. Fu, S. Liu, H. Wei, Y. Cheng, W. Tong, P.P. Shum, D. Liu, Appl. Phys. B. 112 (2013) 491. [15] Z. Ou, Y. Yu, P. Yan, J. Wang, Q. Huang, X. Chen, C. Du, H. Wei, Opt. Express. 21 (2013) 23812. [16] G. Weihs, M. Reck, H. Weinfurter, A. Zeilinger, Opt. Lett. 21 (1996) 302. [17] A. Mafi, J.V. Moloney, IEEE Photon. Technol. Lett 17 (2005) 348. [18] C. Yi, P. Zhang, S. Dai, X. Wang, Y. Wu, T. Xu, Q. Nie, Opt. Commun. 311 (2013)
270. [19] P. St J. Russell, J. Lightw. Technol. 12 (2006) 4729. [20] Ch. Tonglei, D. Zhongchao, G. Weiqing, K. Asano, L. Meisong, D. Dinghuan, T. Suzuki, Y. Ohishi, J. Lightw. Technol. 31 (2013) 1793. [21] B.M. Shalaby, V. Kermene, D. Pagnoux, A. Desfarges-Berthelemot, A. Barthélémy, Appl. Phys. B105 (2) (2011) 213–217. [22] C. Jollivet, A. Mafi, D. Flamm, M. Duparré, K. Schuster, S. Grimm, A. Schülzgen, Opt. Express. 22 (2014) 30377–30386. [23] X. Dong, H.Y. Tam, P. Shum, Appl. Phys. Lett. 90 (2007) 151113. [24] E.O. Doebelin, Measurement Systems, McGraw-Hill, New York, USA, 1990. [25] Y. Wang, D. Richardson, G. Brambilla, X. Feng, M. Petrovich, M. Ding, Z. Song, Opt. Lett. 36 (2011) 558.
Contents lists available at ScienceDirect
Optics Communications journal homepage: www.elsevier.com/locate/optcom
Multicore microstructured optical fibre for sensing applications L. Sójka a,c,n, L. Pajewski a, M. Śliwa a, P. Mergo b, T.M. Benson c, S. Sujecki c, E. Bereś-Pawlik a a b c
Telecommunications and Teleinformatics Department, Wroclaw University of Technology, Wybrzeze Wyspianskiego 27, 50-370 Wroclaw, Poland Laboratory of Optical Fibre Technology, Marie Curie-Skłodowska University, Pl. M. Curie-Sklodowskiej 3, 20-031 Lublin, Poland George Green Institute for Electromagnetics Research, University of Nottingham, University Park, Nottingham NG7 2RD, UK
art ic l e i nf o
a b s t r a c t
Article history: Received 17 May 2014 Received in revised form 31 December 2014 Accepted 3 January 2015 Available online 6 January 2015
In this contribution we present the sensing capabilities of a novel N-path Mach–Zehnder interferometer (MZI) that relies on a multicore microstructured optical fibre (MC-MOF) connected between two sections of standard single mode fibre. The modal properties of the MC-MOF structure are analysed experimentally by measuring near field profiles. The dependence of the N-path MC-MOF MZI sensitivity on temperature, tensile strain and bending is investigated. The results suggest that such an interferometer is a good candidate for a tensile strain or bending sensor. Crown Copyright & 2015 Published by Elsevier B.V. All rights reserved.
Keywords: Microstructured fibre Multicore fibre In-line Mach Zehnder interferometer
1. Introduction Optical fibre sensors belong to the most intensively studied subjects in photonics in recent years. The reason for this is that they can find many applications in biological, chemical and environmental industries. Furthermore, optical fibre sensors have many advantages, which include electromagnetic immunity, high sensitivity and simple and robust construction [1–3]. One of the most widely used types of optical fibre sensors is all-fibre Mach– Zehnder interferometers (MZIs) [1–4]. This is because connecting an all-fibre MZI with other optical fibre components is a simple and straightforward task. The operating principle of this device in essence consists in measuring the phase shift between two or more optical arms. The factors that change the phase shift can be related to some useful measurands like the mechanical strain, pressure, elongation, bending or temperature. In the literature a number of all-fibre MZI designs can be found. A specific type of the MZI is an in-line MZI whereby all interferometer arms are integrated into one optical fibre. For example in-line configurations can exploit twin core fibres [5,6], air hole collapsing of microstructured optical fibre (MOF) [7,8], core mismatch [9,10], a multimode fibre segment [11] or fibre tapering [4]. These configurations exhibit good performance when sensing strain, temperature and bending. Furthermore, in-line fibre MZIs are simple, cost effective and easy to assemble. These features are essential for the n Corresponding author at: Telecommunications and Teleinformatics Department, Wroclaw University of Technology, Wybrzeze Wyspianskiego 27, 50-370 Wroclaw, Poland. E-mail address: [email protected] (L. Sójka).
http://dx.doi.org/10.1016/j.optcom.2015.01.005 0030-4018/Crown Copyright & 2015 Published by Elsevier B.V. All rights reserved.
practical application of the MZIs. Recently in the literature the realisation of novel types of inline MZI, based on multicore fibres, has been reported by several research groups [12–15]. The multi-core microstructured optical fibre (MC-MOF) MZI acts as an N-path interferometer. It has been proven that multi-path interferometers have enhanced sensitivity when compared with conventional two path interferometers [16]. However, this is a very new topic and one which in our opinion has not been fully addressed in the literature. A bending sensor that utilises a seven-core MOF has been recently proposed in [15]. These authors used a lateral offset of the coupling fibres to invoke interference between two supermodes having different effective refractive index. In contrast, in this contribution we analyse the modal properties of MC-MOF using an experimental approach. Furthermore, results presented in the literature shows that in order to obtain equal intensity distribution in each core the careful optimisation of the multicore fibre structure is needed [17,18]. We have shown that we were able to achieve the equal intensity distribution for our MC-MOF by coupled the light with a SMF-28. In this paper experimentally investigate the sensing properties of an N-path interferometer based on a multicore silica-based microstructured optical fibre. We also analyse the modal properties of the MC-MOF using an experimental approach. Whilst our fabricated MC-MOF waveguide structure is similar to the structure presented in [15], our waveguide lattice parameters are significantly different from the ones reported in [15]. In the present paper the N-path MZI was realized by inserting a section of a MCMOF between two sections of a standard single mode fibre (SMF28). Sensing capabilities for the temperature, tensile strain and bending of the MC-MOF were characterised. Unlike [15], we do not
72
L. Sójka et al. / Optics Communications 344 (2015) 71–76
offset the coupling fibres; this simplifies the setup. The results obtained show that the proposed N-path MC-MOF MZI can provide good sensitivity for measurement of tensile strain or bending, with the benefit of low sensitivity to operating temperature. It retains the additional benefit of compactness that in-line MZIs offers over conventional MZIs. The results are presented as follows: in Section 2 the procedures used for the fabrication and initial characterisation of the MC-MOF are described. Section 3 presents the experimental evaluation of the sensing behaviour of an N-path MZI based on the MC-MOF fabricated. Conclusions are drawn in Section 4.
2. Characterisation of MC-MOF 2.1. Fabrication of MC-MOF The cross-sectional structure of the manufactured seven-core microstructured silica-based optical fibre is presented in Fig.1. The fibre was made using a stack-and-draw technique. The fibre preform was created by stacking capillaries around a solid rod. Each core was produced using a single rod. The MC-MOF has been polymer coated. Table 1 presents the parameters of the fabricated MC-MOF. 2.2. Near-field measurements The mode field properties in MOFs strongly depend on the lattice parameters. Therefore, it is important to know the exact mode field profile of a MC-MOF. The near field mode intensity of the MC-MOF fabricated was measured experimentally at a signal wavelength of 630 nm. Each core in MC-MOF is endlessly singlemode because the ratio d/Λ o 0.43 [19]. The light from a laser was introduced into a 2 m length of the MC-MOF by connecting both ends of the MC-MOF to a single mode fibre SMF-28 using conventional a FC/PC fibre connectors. The near-field image at the MCMOF output was captured using a CCD camera and the mode field profile obtained is shown in Fig. 2a. The intensity plot of the mode field profile is given in Fig. 2b. These results indicate that the light from SMF-28 can effectively excite all cores of the MC-MOF
Fig. 1. Cross-section of the silica MC-MOF fabricated.
Table 1 Parameters of the MC-MOF fabricated. Core diameter Air-hole diameter, d Lattice constant, Λ (pitch) Filling factor, d/Λ
5.15 mm 1.5 mm 3.6 mm 0.413
without coupling a significant proportion of the input power to the cladding modes. It has been proven in the literature that is possible to achieve equal intensities in each core of 7-core MOF by careful optimisation of the fibre structure [17,18,20]. Each core is slightly different in practice due to manufacturing differences. Therefore this simple configuration can be used to realise a sensor, which exploits the interference that occurs between the core guided modes. Based on this principle the MC-MOF-MZI was implemented. We excite all cores of the multicore fibre with the SMF-28 aligned with the centre of the MC-MOF.
3. The experimental results for N-path MC-MOF MZI 3.1. Principle of operation Schematic and illustrative diagrams of the realised N-path MCMOF MZI are depicted in Fig. 3a and b, respectively. The MC-MOF cores are not exactly identical due to fabrication inaccuracies this results in difference between the optical path length of the interferometers arm. The interferometer was fabricated by inserting 2 m section of MC-MOF between two sections of SMF-28 using conventional FC/PC fibre connectors. An overall MC-MOF length of 2 m was used in all experiments, however the MC-MOF section exposed to external factors (temperature, strain, bending) was slightly altered in each experiment. In each case the MC-MOF length exposed to temperature, strain or bending is specified. The fibre was coated in all experiments. The SMF-28 at the beginning and at the end of MC-MOF acts as an input and output coupler, respectively. The light from the SMF-28 excites several modes in the MC-MOF. A fibre with seven coupled cores supports seven supermodes [21,22]. At the end of the MC-MOF section the light from each mode is coupled back to SMF-28. Multi-mode interference occurring in this fibre. Each supermodes acts as a separate arm of a MZI. The coupling efficiency depends on the relative phase differences between the optical beams propagating in the cores of the MC-MOF, since each mode acts as a separate arm of a MZI. Spectrally, the interference generates a periodic transmission spectrum of the N-path MZ device. The fabricated MC-MOF MZI is simple and compact, in contrast to a conventional MZI assembly where the two optical fibre couplers and two fibre sections are needed. In order to investigate the transmission properties of the MCMOF MZI, broadband light from an erbium amplified spontaneous emission (ASE) source was introduced into the fibre. The transmission spectrum of the MC-MOF MZI was monitored using an Optical Spectrum Analyser (OSA), Yokogawa AQ6370C, with a resolution of 0.02 nm. The experimental set-up is shown in Fig. 4. Since SMF-28 is used at both ends of the sensor further coupling with standard fibre optic devices is straightforward. The insertion loss of the MCF device was obtained as 1.89 dB. Fig. 5 presents the interference fringe pattern observed at the output of the MC-MOF MZI near the wavelength 1550 nm. A maximum contrast of around 10 dB is observed. The measured spacing between the interference peaks was around 0.8 nm. It can be observed that the interference fringe pattern is not uniform like in a conventional two-path MZI. The sidelobes in the spectrum are characteristic of an N-path interferometer, and are due to the superposition of the multiple
L. Sójka et al. / Optics Communications 344 (2015) 71–76
73
Fig. 2. (a) Near-field image obtained for MC-MOF at a wavelength of 630 nm, (b) intensity plot obtained by post-processing the image data presented in (a) using Matlab.
Fig. 3. (a) Schematic diagram of N-path MC-MOF MZI. (b) Illustrative diagram of N-path MC-MOF MZI coupled to SMF.
Fig. 4. MC-MOF MZI experimental set-up.
3.2. Temperature sensitivity
Fig. 5. Transmission spectrum of the MC-MOF MZI at wavelengths around 1550 nm.
beams guided by the interferometer arms [14–16]. In addition, the MC-MOF MZI can also behave like a comb filter and hence it potentially provides an additional wavelength selection mechanism for a fibre laser [5,6,11].
When the external temperature experienced by the MC-MOF is increased the fibre can elongate due to thermal expansion. Also the refractive index changes due to the thermo-optic effect. Consequently the optical beam path for each MC-MOF core can change [6]. Any resulting change in relative phase will manifest itself as a wavelength shift in the transmission spectrum [1–4]. The experimental setup is composed of the ASE erbium source working at 1550 nm, the N-path MC-MZI and a Peltier element. A 80 mm section of the MC-MOF was placed on the Peltier element. The temperature was controlled in the range spanning from 7 °C up to 85.6 °C with a step resolution of 10 °C. The temperature variations were achieved by changing the current that passes through the Peltier element. The output spectrum of the interference fringe pattern was observed using the OSA. The temperature sensitivity was measured for a coated fibre. The dependence of the wavelength of a selected dip in the transmission spectrum is given as a function of temperature in Fig. 6a. Fig. 6b shows the dependence of the output power on the wavelength at selected temperatures. It can be seen that the wavelength at which the selected peak in the transmission spectrum occurs remains almost unchanged. This behaviour can be explained by the low thermal expansion coefficient of the MC-MOF since the MC-MOF is made from silica and air only [23]. Pure silica has a low thermo-optic
74
L. Sójka et al. / Optics Communications 344 (2015) 71–76
Fig. 6. (a) Wavelength of a selected dip in transmission versus temperature (the red line indicates the best fit to measured data), (b) wavelength variation of optical power of the selected transmission dip for various temperatures. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
Fig. 7. (a) Wavelength of selected dip in transmission versus tensile strain (the red line indicates the best fit to measured data). (b) Spectral dependence of transmitted power around the selected dip in transmission spectrum as a function of tensile strain. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
higher (typically 10 pm/ oC ) than for our MC-MOF MZI [1,2]. 3.3. Tensile strain sensitivity
Fig. 8. Experimental set-up for measuring the bending properties of MC-MOF MZI.
coefficient [3]. The experimental results allow the temperature sensitivity of the MC-MOF to be estimated as 1.6 pm/ oC .This low sensitivity to temperature feature is an advantage in many fibre sensor applications because it makes tensile strain and bending measurement relatively independent of the ambient temperature. For comparison, standard fibre Bragg grating (FBG) sensors are very sensitive to the temperature changes. For example the temperature sensitivity for standard FBG is one order of magnitude
When a tensile strain is applied to the MC-MOF, the length of the MC-MOF and the effective refractive indices of the modes supported by the MC-MOF are changed, which results in changes of the relative phase velocities of the beams in the MC-MOF MZI arms. In order to determine the tensile strain sensitivity of the MCMOF MZI one end of the MC-MOF was fixed and the other one was attached to a translation stage. Subsequently the MC-MOF was stretched along the fibre axis by moving the translation stage in steps of 50 mm. The tensile strain was calculated from
ΔL (1) L where L ¼110 mm was the MC-MOF length used in the experiment S=
L. Sójka et al. / Optics Communications 344 (2015) 71–76
75
Fig. 9. (a) Wavelength of selected dip in transmission versus curvature (the red line indicates the best fit to the measured data). (b) Spectral dependence of transmitted power around the selected dip in transmission spectrum for various curvatures. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
and ΔL ¼0.350 mm was the maximum displacement of the translation stage. Fig.7a presents the dependence of the wavelength of a selected dip in the transmission spectrum on the applied strain. The dip has undergone a blue-shift of around 0.35 nm when the strain was increased from 0 to 3500 mε (micro strains). Fig. 7b shows the wavelength dependence of the selected dip in the transmission spectrum on the applied strain. It is important to note that the dependence of the wavelength dip position on the applied strain is linear. This is unlike many standard resistive strain gauges [24]. The measured sensitivity to the tensile strain is around 0.08695 pm/mε. This sensitivity is one order of magnitude less than that of fibre sensors based on FBGs, which typically is equal to 1.0 pm/mε. However, it is only two times less than the sensitivity of fibre sensors that exploit highly birefringent microstructured optical fibre (HB-MOFs) [24].
transmission spectrum for several values of curvature within the range from 2 up to 7 m 1.
4. Conclusions In this paper we have investigated the sensing capabilities of a MC-MOF. The N-path Mach Zehnder interferometer was fabricated using a section of a MC-MOF and two single mode fibre sections (SMF-28). The experimental results show that the MC-MOF sensor is insensitive to the temperature changes. A linear response to the applied tensile strain and bending was observed. These results demonstrate that the MC-MOF MZI is robust. Furthermore, such a sensor overcomes the cross-sensitivity problem between tensile strain or bending and temperature. In order to obtain better results we plan in the future to focus on improvement of the multicore fibre waveguide structure.
3.4. Bending sensitivity The experimental set-up used to investigate the bending sensitivity of the MC-MOF MZI is shown in Fig. 8. A 99.8 mm section of the MC-MOF was placed on two translation stages. One of the stages was fixed, while the other one was able to move inwards to change the separation distance between the stages. In such a way, different curvature could be applied to on the MC-MOF section. The change of the curvature alters the optical path, resulting in different propagation conditions for each mode that propagates in the MC-MOF. Therefore the variation of the curvature will lead to a wavelength shift of the dips in the transmission spectrum. The fibre curvature was approximated using the expression [25]:
C=
1 ≅ r
24x L3
(2)
where r is the radius of the bent fibre, x is the displacement value of the moveable fibre end and L is the fibre length between the translation stages. Fig. 9a presents the shift in the wavelength of a specific dip in the transmission responses of the MC-MOF MZI for different values of the curvature C. It should be noted that again the dependence of this wavelength on the bending curvature can be approximated using a linear function. The measured curvature sensitivity is 0.02465 nm/m 1. Fig. 9b shows the spectral dependence of transmitted power around the selected dip in the
Acknowledgements This research has been partly supported by the statutory funds of Telecommunications and Teleinformatics Department, Wroclaw University of Technology, Poland. One of the authors (L.S.) is very thankful to the Wroclaw University of Technology for the fellowship co financed by European Union within European Social Fund.
References [1] A.D. Kersey, Opt. Fiber Technol. 2 (1996) 291. [2] A.D. Kersey, M.A. Davis, H.J. Patrick, M. LeBlanc, K.P. Koo, C.G. Askins, M. A. Putnam, E.J. Friebele, J. Lightw. Technol. 15 (1997) 1442. [3] O. Frazão, J.L. Santos, F.M. Araújo, L. Ferreira, , Laser Photon. Rev. 2 (2008) 449. [4] B.H. Lee, Y.H. Kim, K.S. Park, J.B. Eom, M.J. Kim, B.S. Rho, H.Y. Choi, Sensors 12 (2012) 2467. [5] G. Yin, Lou S.,H. Zou, Laser Phys. Lett. 10 (2013) 045103. [6] L. Shuhui, L. Ningliang, W. Ying, G. Jiangtao, L. Zhihua, L. Peixiang, L. IEEE, Photon. Technol. Lett. 24 (2012) 1768. [7] J.M. Sierra-Hernandez, R. Rojas-Laguna, E. Vargas-Rodriguez, J.M. EstudilloAyal, R.I. Mata-Chavez,, D. Jauregui-Vazquez, J.C. Hernandez-Garcia, J. A. Andrade-Lucio, J.C. Gutierrez-Gutierrez, Laser Phys. 23 (2013) 055105. [8] W.G. Chen, S.Q. Lou, S.C. Feng, L.W. Wang, H.L. Li, T.Y. Gou, S. Jian, Laser Phys. 19 (2009) 2115. [9] L.V. Nguye, D. Hwang, S. Moon, D.S. Moon, Y. Chung, Opt. Express 16 (2008) 11369. [10] J. Zhou, C. Liao, Y. Wang, G. Yin, X. Zhong, K. Yang, B. Sun, G. Wang, Z. Li, Opt.
76
L. Sójka et al. / Optics Communications 344 (2015) 71–76
Express 22 (2014) 1680. [11] L. Ma, Z. Kang, Y. Qi, S. Jian, Laser Phys. 24 (2014) 045102. [12] P.M. Blanchard, J.G. Burnett, G.R. Erry, A.H. Greenaway, P. Harrison, B. Mangan, J.C. Knight, P.StJ. Russell, J. Gander, R. McBride, J.D. Jones, Smart Mater. Struct. 9 (2000) 132. [13] W.N. MacPherson, M.J. Gander, R. McBride, J.D. Jones, P.M. Blanchard, J. G. Burnett, A.H. Greenaway, B. Mangan, T.A. Birks, J.C. Knight, P.St. J. Russell, Opt. Commun. 193 (2001) 97. [14] Z. Zhao, M. Tang, S. Fu, S. Liu, H. Wei, Y. Cheng, W. Tong, P.P. Shum, D. Liu, Appl. Phys. B. 112 (2013) 491. [15] Z. Ou, Y. Yu, P. Yan, J. Wang, Q. Huang, X. Chen, C. Du, H. Wei, Opt. Express. 21 (2013) 23812. [16] G. Weihs, M. Reck, H. Weinfurter, A. Zeilinger, Opt. Lett. 21 (1996) 302. [17] A. Mafi, J.V. Moloney, IEEE Photon. Technol. Lett 17 (2005) 348. [18] C. Yi, P. Zhang, S. Dai, X. Wang, Y. Wu, T. Xu, Q. Nie, Opt. Commun. 311 (2013)
270. [19] P. St J. Russell, J. Lightw. Technol. 12 (2006) 4729. [20] Ch. Tonglei, D. Zhongchao, G. Weiqing, K. Asano, L. Meisong, D. Dinghuan, T. Suzuki, Y. Ohishi, J. Lightw. Technol. 31 (2013) 1793. [21] B.M. Shalaby, V. Kermene, D. Pagnoux, A. Desfarges-Berthelemot, A. Barthélémy, Appl. Phys. B105 (2) (2011) 213–217. [22] C. Jollivet, A. Mafi, D. Flamm, M. Duparré, K. Schuster, S. Grimm, A. Schülzgen, Opt. Express. 22 (2014) 30377–30386. [23] X. Dong, H.Y. Tam, P. Shum, Appl. Phys. Lett. 90 (2007) 151113. [24] E.O. Doebelin, Measurement Systems, McGraw-Hill, New York, USA, 1990. [25] Y. Wang, D. Richardson, G. Brambilla, X. Feng, M. Petrovich, M. Ding, Z. Song, Opt. Lett. 36 (2011) 558.