Open Access
Add to BibSonomyAbstract
The scatterings of TE- and TM-polarized evanescent wave on the surface of a tilted fiber Bragg grating (TFBG) with a 50 nm thick gold coating were investigated experimentally by observing radiation patterns from discontinuities in the coating. The scattering intensity for TM-polarized light is larger than for TE light when the evanescent wave propagates from the coating towards the discontinuity. The opposite occurs for light propagating from an uncoated section towards the coating edge. However in the latter case the scattering is much weaker. These results confirm that cladding modes with TE and TM polarization can be excited selectively with a TFBG, and that they scatter light differentially at discontinuities. These results are used to propose a simple polarimeter design based on total scattered light intensity monitoring.
© 2014 Optical Society of America
1. Introduction
Tilted fiber Bragg gratings (TFBGs) have been proposed for many sensing applications, ranging from external refractive index (RI), bending, modern biological analysis, and so on [1–6]. Unlike the perturbation of the period or RI of an ordinary fiber Bragg grating (FBG), by tilting the grating planes relative to the perpendicular of the fiber axis, the result of the interaction of the periodic perturbation with the core guided light brings vastly different properties from that of the FBG. In TFBG, light can be coupled from the core to a subset of the large number of modes that can be guided by the cladding of the fiber. In particular, the wavelengths of all the resonances of a TFBG have the same temperature dependence (they shift by about 10 pm/°C). Therefore, the temperature cross-sensitivity of all other sensing modalities can be eliminated by considering relative wavelength shifts instead of absolute spectral measurements. Based on these characteristics, fiber-based surface plasmon resonance (SPR) sensors made from a TFBG with a metal coating in the sensing region are widely developed [1–3]. However, the TFBG spectra are strongly depended on the polarization of the input light, especially as it was used in SPR sensor. So it is very important to clarify the polarization dependent properties of the modes of metal-coated TFBGs. Some investigations on the polarization properties of gold-coated TFBGs indicate that when the cladding mode light is tangential to the surface (TE polarization) it cannot penetrate the metal layer, while TM light can tunnel across it, as expected from conventional electromagnetic theory [4, 5]. However if the metal coating is discontinuous, then TE light should be able to exit through the holes of the coating. These considerations were not directly observed but rather inferred from simulations and analysis of the transmission spectra.
In this paper, near infrared (NIR) imaging of the scattering from thin gold-coated TFBGs was carried out. By using the NIR imaging, and polarization control of the cladding mode evanescent field, the polarization resolved scattering intensity of discontinuities in the gold coating was observed directly. Finally, it is shown that through image processing of the scattered light, the polarization orientation of the core guided light incident on the grating can be monitored.
2. Experiments
Through a phase-mask method, a 10 mm long TFBG with a tilt angle of 10 degree was written in a hydrogen-loaded photosensitive Corning SMF-28 fiber by using a pulsed KrF excimer laser. The Bragg wavelength is around 1610 nm so that the cladding mode excitation is maximum near 1550 nm. A gold coating with a thickness of 50 nm was deposited by sputtering on the surface of the fabricated TFBG. As shown in Fig. 1, we used a broad band source (BBS) (JDS Uniphase), polarization controller (PC) (JDS Uniphase), and an optical spectrum analyzer (OSA) (AQ6317B, ANDO) to launch light towards the TFBG with precise S- and P-polarized input light, as determined by the transmission spectrum of the TFBG (here the S- and P- orientations refer to the orientation of the tilt plane) [1]. The PC contains one polarizer, a half-wave plate and a quarter-wave plate. This combination allows the preparation of arbitrary polarization states at the fiber input, which can compensate for any change of polarization state induced by fiber loops and twists in the optical path leading to the TFBG. As noted in [1], S-polarized input is expected to couple to cladding modes with azimuthally polarized evanescent fields, and P- input to cladding modes with radially polarized evanescent fields. For each input polarization, the NIR-scattering from the fiber surface was imaged with an infrared CCD camera (SUI, GoodRich) mounted on a microscope. The visible light micrographs of the coated TFBG were obtained by using another camera, a visible light CCD (ELMO, TNC4604X).
In order to observe the scattering characteristics of the P- and S-polarized input light on the boundaries of the metal coating, as shown in Fig. 2(a), we scraped off the gold film in a small area to form a hole over the TFBG.
Fig. 2 (a) Micrograph of a hole in a gold-coated TFBG and the corresponding NIR images of scattering for P- (b) and S- (c) polarized input light. Cladding mode light propagates from right to left in these images.
3. Results and discussions
Figure 2 shows the micrographs of a gold-coated TFBG with a ~38 μm diameter hole, as well as NIR images of the same area when P- and S- polarized input light is launched in the fiber core. From the NIR images, we can see that regardless of polarization, the evanescent field is only visible on the edges of the hole, because the imaging system can only “see” radiated light (and not the evanescent field). In these images, the core guided light propagates from left to right, and the cladding modes from right to left (the TFBG is a short period grating, and hence a contra-directional coupler). In particular, Fig. 2(b) shows that for P-polarized input, the scattering is much brighter for evanescent waves leaving the metal coated area (on the right hand side) than it is when the air guide evanescent wave (in the hole) re-enters the metal coating on the left. For S- polarized light, as shown in Fig. 2(c), the overall scattering intensity is much weaker but in contrast to the P-polarized case it is slightly stronger in going from air to metal than when the cladding mode leaves the metal coated area.
It is well known that the transfer of energy from the core of a single mode fiber into cladding modes is well explained by coupled-mode theory. The reflectivity R of the grating resonances depends on the coupling coefficient between the core mode and the cladding mode and on the length L of the grating [7],
In this expression, the coupling coefficient is evaluated as follows from the transverse components of the electric fields of the modes [8],
where c is a proportionality constant related to the normalization of the transverse mode fields (Ecore and Er). For TFBG, has the following form,In Eq. (3), is the grating period, is the tilt angle of the grating planes, and Δn is the amplitude of the RI modulation. From Eqs. (1)–(3), and some previous reports [9–11], it has been determined that when the polarization state of the light incident on the TFBG is linear and aligned parallel (P-) or perpendicular (S-) to the tilt plane (two extreme cases), the electrical field pattern of the excited high order cladding modes also has radically different polarization properties. S-polarized light can only couple into high order cladding modes that have their electrical field tangential to the cladding boundary, while P-polarized light excites cladding modes with predominantly radial electrical fields. Therefore, it is expected that the polarization of the evanescent wave that propagates along the fiber surface (both in the coated and uncoated regions) can be selected to be either parallel (TE) or perpendicular (TM) to the surface.
In order to correlate the scattering observations with the state of the evanescent field, we first note that for TM light, the evanescent wave can tunnel across the metal coating when it propagates in the coated section of the grating, while TE light is confined to the glass underneath the gold. In those circumstances, upon arriving from the coated region into the un-coated one the overlap between TM light and the discontinuity (the edge of the gold film) will be larger, hence causing more scattering. At the other side of the hole (going from air clad to metal clad), both TE and TM evanescent fields of the cladding in air have similar extent in air prior to the discontinuity, but the effect of the discontinuity is larger for TE light as it must return under the metal coating. Because of the larger discontinuity of the field extent, the TE wave scatters more than the TM one at that location, as shown in Fig. 2(c). To support this explanation, the intensity distributions of the evanescent fields of TE and TM polarized modes were calculated (using the commercial software FIMMWAVE, by Photon Design) for the coated and uncoated fiber. The relative mode intensity (RMI) patterns in the vicinity of the fiber surface are shown in Fig. 3. It is clear that in the 50 nm thick layer just above the fiber surface, the TM wave coming from the film side has three orders of magnitude more intensity (RMI = 0.023) than on the air side (RMI = 0.00003) and this much light will scatter very efficiently off the boundary. The corresponding TE wave has essentially zero power in that layer (RMI = 4 × 10−8) and therefore no overlap with the edge of the film. At the air to film boundary, the TE (RMI = 4.6 × 10−5) and TM (RMI = 3.1 × 10−5) waves have comparable but relatively low power densities incident on the edge of the film, implying some scattering but not as much as in the first case discussed (and a little more for the TE wave, as observed experimentally).
Fig. 3 Simulated intensity distributions of the cladding modes for bare fiber and fiber with 50 nm of gold, in TE and TM polarizations, as a function of distance from the fiber surface. The grey shaded rectangle indicates the location of the coating.
We also studied the scattering of a particle (with a diameter of about 1.8 μm) on the surface the gold-coated TFBG to further characterize the different scattering characteristics of the TE and TM polarized light. As shown in Fig. 4, the scattering light intensity is changing with the polarization of the input light. For S-polarized input light, the excited cladding modes are TE-polarized and no scattering can be observed (Fig. 4(b)). Upon rotation of the input light towards P-polarization, the scattering intensity by this particle increases gradually, reaching maximum value at 90° from the S-polarized minimum. These are consistent with the fact that TE waves do not penetrate across the gold layer and that TM light can tunnel across and scatter off the particle deposited on the outside gold surface. Based on this property, we can detect the polarization direction of the input light by image processing of the scattered light, as will be demonstrated below.
Fig. 4 Micrographs of (a) a particle (1.8 μm) on the surface the gold-coated TFBG and its NIR scattering ((b) to (k) denote the S (0°) - to P (90°) - polarized input light with a step of 10°, respectively).
Before this however, it is worth mentioning that as shown in Fig. 5, the transmission spectra of the gold-coated TFBG with different input light polarizations are also strongly dependent on polarization. But the correlation is much less clear since the maximum attenuation of the transmission resonances, here indicative of P-polarization and penetration of the cladding mode field in the metal layer, can also be due to S-polarized light leakage through gaps in the coating, as was shown in [4]. Therefore, the quantitative relationships between the polarization direction and the intensity of the resonance are difficult to obtain. On the other hand, image processing of the NIR scattering provides a direct indication of the orientation of the linearly polarized input light that provides the strongest tunneling across the metal layer, i.e. P-polarization. Through comparison of the scattering light in the same region (with no holes) of the TFBG surface, the polarization direction of the input light can be determined. Figure 6 shows the normalized brightness level of the same group of pixels on the NIR images for the input light with different input polarization directions. There is a one to one relationship, as expressed by the second order polynomial fit indicated on the figure, between relative brightness and polarization orientation. This information can be used to measure the orientation of the polarization inside the fiber by an outside measurement.
Fig. 6 Relative brightness levels of the pixels on the NIR images with different polarization angle, insets are the scattering images for polarization angles of 0°, 30°, 60°, 90°.
4. Conclusion
In conclusion, we present a NIR imaging technique to investigate the dominant polarization state of the evanescent light from high order cladding modes as it relates to the polarization orientation of the input light incident on an in-fiber TFBG. Results show that TM light tunnels across continuous metal coatings and also scatters preferentially at coating discontinuities, as long as the direction of propagation is from the metal coated region to the metal-free region. Finally a quantitative relationship between average scattering intensity and input light linear polarization angle is presented, which allows for the determination of the light polarization inside the fiber core from a measurement carried out outside of the fiber.
Acknowledgments
This work was supported by the Natural Science and Engineering Research Council of Canada, the Canada Foundation for Innovation, Canada Research Chairs program and National Natural Science Foundation of China (No. 51374188) and Zhejiang Provincial Natural Science Foundation for Distinguished Young Scientists (No. LR13E040001).
References and links
1. J. Albert, L.-Y. Shao, and C. Caucheteur, “Tilted fiber Bragg grating sensors,” Laser Photonics Rev. 7(1), 83–108 (2013). [CrossRef]
2. C. Caucheteur, Y. Shevchenko, L.-Y. Shao, M. Wuilpart, and J. Albert, “High resolution interrogation of tilted fiber grating SPR sensors from polarization properties measurement,” Opt. Express 19(2), 1656–1664 (2011). [CrossRef] [PubMed]
3. L.-Y. Shao, Y. Shevchenko, and J. Albert, “Intrinsic temperature sensitivity of tilted fiber Bragg grating based surface plasmon resonance sensors,” Opt. Express 18(11), 11464–11471 (2010). [CrossRef] [PubMed]
4. W. Zhou, D. J. Mandia, M. B. E. Griffiths, A. Bialiayeu, Y. Zhang, P. G. Gordon, S. T. Barry, and J. Albert, “Polarization-dependent properties of the cladding modes of a single mode fiber covered with gold nanoparticles,” Opt. Express 21(1), 245–255 (2013). [CrossRef] [PubMed]
5. Y. C. Lu, R. Geng, C. Wang, F. Zhang, C. Liu, T. Ning, and S. Jian, “Polarization effects in tilted fiber Bragg grating refractometers,” J. Lightwave Technol. 28(11), 1677–1684 (2010). [CrossRef]
6. L.-Y. Shao and J. Albert, “Compact fiber-optic vector inclinometer,” Opt. Lett. 35(7), 1034–1036 (2010). [CrossRef] [PubMed]
7. R. Kashyap, Fiber Bragg Gratings, 2nd ed. (Academic, 2011).
8. T. Erdogan and J. E. Sipe, “Tilted fiber phase gratings,” J. Opt. Soc. Am. A 13(2), 296–313 (1996). [CrossRef]
9. Y. Shevchenko, C. Chen, M. A. Dakka, and J. Albert, “Polarization-selective grating excitation of plasmons in cylindrical optical fibers,” Opt. Lett. 35(5), 637–639 (2010). [CrossRef] [PubMed]
10. C. Caucheteur, C. Chen, V. Voisin, P. Berini, and J. Albert, “A thin metal sheath lifts the EH to HE degeneracy in the cladding mode refractometric sensitivity of optical fiber sensors,” Appl. Phys. Lett. 99(4), 041118 (2011). [CrossRef]
11. M. Z. Alam and J. Albert, “Selective excitation of radially and azimuthally polarized optical fiber cladding modes,” J. Lightwave Technol. 31(19), 3167–3175 (2013). [CrossRef]
