This paper presents an optical fiber-field access device suitable for use in different in-line fiber-optics’ systems and fiber-based photonics’ components. The proposed device utilizes a thin silica micro-wire positioned in-between two lead-in single mode fibers. The thin micro-wire acts as a waveguide that allows for low-loss interconnection between both lead-in fibers, while providing interaction between the guided optical field and the surrounding medium or other photonic structures. The field interaction strength, total loss, and phase matching conditions can be partially controlled by device-design. The presented all-fiber device is miniature in size and utilizes an all-silica construction. It has mechanical properties suitable for handling and packaging without the need for additional mechanical support or reinforcements. The proposed device was produced using a micromachining method that utilizes selective etching of a purposely-produced phosphorus pentoxide-doped optical fiber. This method is simple, compatible with batch processes, and has good high-volume manufacturing potential.
© 2012 OSA
Effective design of fiber-field access devices and methods for their manufacturing is important for further practical development and usage of various fiber-optic components, such as sensors [1–10], filters [11–15], gratings [16, 17], attenuators/modulators [18–20], micro-resonators [21–25], lasers [26, 27], polarization-control devices [28, 29], and many other photonic structures [30–32].
Access to a fiber’s optical-field is usually achieved by the application of bi-conical fiber tapers, D-fibers or side-polished fibers. Side-polishing  of the fiber is mechanically complex, further complicated by the tight tolerances required when producing practical devices. D-fibers do provide an easier way of accessing the fiber-field however they have proved to be difficult to splice resulting in additional losses to standard fibers , and they still require precisely-controlled etching processes to set the desired dimensions of the field- access region. Both, the side-polished and D-fibers suffer from circular asymmetry that leads to polarization dependence  and to an asymmetric shape of the modal-field within the field interaction zone, thus leading to additional losses and birefringence. Circular asymmetry also limits field interaction cross-section, which results in reduce field interaction strength. On the other hand, bi-conical tapers produced by fiber-heating and pulling [36–38] or direct etching [39–41] do not suffer from the above limitations and are usually produced by the tapering of standard single-mode optical-fibers. Etched tapers, in general, exhibit higher insertion losses than heated/pulled tapers . This increase in loss can be attributed to the non-uniformities of etched fiber-sections , but it is also likely that part of this loss originates from the insufficient length of a transitional (conical) region that leads to non-adiabatic mode conversions [38, 42]. This problem can be partially solved by adding more control to the etching process, as described in , in order to produce slower transitional regions. High-quality in-line devices are therefore often based on bionic tapers produced by the heating and pulling of standard single-mode fibers (SMF). However, the designing and fabricating of fiber tapers introduces other limitations. In order to prevent coupling of the (fundamental) mode with the cladding modes, and consequently cladding mode interactions including losses, the length of the taper needs to be typically of an order of centimeters or more [38, 42], which compromises any miniature dimensions of the device. Long-length and, in particular, the very small waist diameters required, cause the production, handling, alignment, and packaging of fiber tapers to be difficult and, consequently, less practical for usage. Several fiber-taper packaging techniques have been proposed , but all such techniques yield in considerable size of the final devices. Furthermore, a comprehensive overview of micro and nanowires that allow for access to a fiber’s field is provided in ref .
This paper presents an in-line, all-silica, fiber-field access device (FFAD) that is miniature in size, straightforward for manufacturing, and can provide low-insertion losses. The proposed device retains compatibility with standard SMF, including circular symmetry of the evanescent field-access region. Furthermore, the design of the device provides the additional capability of precisely controlling the phase-matching conditions (e.g. effective index) of the mode propagation within the field interaction region.
2 Miniature device for accessing of the fiber field
The proposed fiber field access device (FFAD) is shown in Fig. 1 . It is composed of a thin micro-wire positioned in-between two lead single-mode fibers. The micro-wire acts as a waveguide that interconnects both single-mode fiber cores. The micro-wire’s diameter and its cross-sectional index profile were chosen to provide the desired evanescent field interaction between the guided field and any surrounding material (or another photonic structure). The micro-wire can be homogenous or composed of a core and the thin cladding surrounding that core. In order to prevent the transfer of mechanical stresses caused, for example, by the device’s manipulation towards the thin and otherwise fragile micro-wire, the proposed FFAD is mechanically reinforced by two side-support members positioned at opposite sides of the micro-wire. The void between the thin micro-wire and the side-support members prevents any interaction between the field propagating within the core and the side-supports’ members. The entire structure is fusion-spliced and made from silica. It is thus mechanically, chemically, and thermally stable.
The proposed FFAD concept offers possibilities for tuning or optimizing device performance. Firstly, the refractive index difference between the thin micro-wire and the surrounding medium can be high and thus considerably different from the index contrast encountered in the lead-in single-mode fiber. This can cause significant mismatches between the lead-in fiber and the thin micro-wire’s modal fields, which can further lead to high optical losses of the device and/or multimodal operation. In tapered-devices this field mismatch is handled by gradual/adiabatic transitions [38, 42] between the lead-in fibers and the field-access (micro-wire) regions that make these devices considerable in length. In miniature devices, as in the presented FFAD, alternative approaches are required in order to limit this loss. When FFAD is submerged into a medium with a (considerably) lower index of refraction, such as air or aquatic mediums, there are two general ways that can be used for device-loss reduction:
In a case where the thin micro-wire is composed of a monolithic material (for example from pure silica) FFAD losses can be minimized by a proper setting of the thin micro-wire’s diameter. Since the micro-wire’s diameter directly affects the micro-wire’s fundamental mode-field diameter (MFD), the micro-wire’s diameter adjustment can be used to reasonably match the micro-wire’s and the fiber’s modal-field. The optimum micro-wire diameter therefore depends on the micro-wire-surrounding refractive index contrast, the lead-in fiber MFD, and the operating wavelength. Such monolithic micro-wire design also provides maximum field strength at the micro-wire’s surface.
Good matching between the thin micro-wire and the lead-in fiber’s fundamental mode fields can also be achieved by implementing a core-cladding structure into the micro-wire’s cross-section. In such a case, the cladding thickness affects both the field amplitude at the micro-wire’s outer surface and any insertion losses of the structure. Low local index contrast between the micro-wire’s core and cladding can provide good matching between the micro-wire’s fundamental mode-field and the field of the lead-in fiber. The additional cladding layer, however, reduces the field amplitude at the micro-wire’s outer surface and consequently reduces the field interaction strength.
Numerical modeling was used to investigate both the above cases by applying the commercially-available Optiwave’s simulation software OptiFiber 2.0 (a vector mode solver was used to provide accurate results in high-index contrast cases). The mode-field of the lead-in fiber was overlapped by the fundamental mode-field of the thin micro-wire in order to predict device losses. Figure 2 presents the calculated total FFAD losses in the case of micro-wire composed of a standard single-mode core, surrounded by a thin pure silica cladding. The Fig. 2 shows the device losses as a function of the surrounding medium refractive index (RI) and the thin silica cladding thickness. Calculations were performed at 1550 nm.
In the water (n = 1.33) the total predicted loss somewhat exceeded –2.6 dB when the core was fully exposed to the surrounding medium (no cladding). By adding 1 μm of thick cladding over the core, the total loss of FFAD reduced to −1.65 dB. Cladding with a thickness of about 1.5 μm is required in order to reduce the losses of the FFAD device below −1 dB in water. Lower losses can be obtained when the device is operated in a medium with RI closer to the RI of silica. For example, when the RI of the surrounding medium is 1.4, a loss of about −1 dB can be obtained for cladding thickness, corresponding to about 1 μm. Since any increase in the cladding thickness also exponentially reduces the field amplitude at the outer surface of the micro-wire, a compromise between field interaction strength and acceptable FFAD losses may be found for a specific FFAD application. A core–cladding structure is particularly useful when the device is used in a medium with RI close to the RI of the silica, or when a weak evanescent-field interaction strength is permitted or required (e.g. such as in different sensor applications) or when certain photonic structures, such as for example gratings, are inscribed into or deposited within/onto a field’s access region (e.g. micro-wire).
When the micro-wire utilizes a monolithic design (e.g. no core), FFAD loss depends on the surrounding RI, the rod’s index, and the rod’s diameter. Figure 3 shows the calculated FFAD’s losses versus the rods’ diameters for two different surrounding mediums, e.g. air and water. During simulation, standard single-mode fibers are assumed to be the lead-in fibers and the pure silica as the rod material. There are two different rod diameters where good match (minimum loss) between the field of the lead-in (standard SMF) fiber and the fundamental mode of the rod can be obtained. In the first case, the micro-wire’s diameter falls within a sub-micron range, more precisely it equals 750 nm for the surrounding medium with n = 1.33 and 430 nm for the surrounding medium with n = 1. The minimum theoretically-achievable losses for these cases are −0.62 dB and −0.60 dB for n = 1 and n = 1.33, respectively. Another minimum in FFAD loss occurs within the micrometer diameter range, more precisely, when the rods’ diameters correspond to 15 and 14.6 µm for n = 1 and n = 1.33, respectively. The corresponding predicted total insertion losses for these two cases are −0.37 dB and −0.35 dB. In the first (nano) case, the rod-surrounding the medium waveguide structure only supports the fundamental mode, while in the second case, the rod supports several modes. Overlap-integral calculations, however, show that under ideal conditions none of the higher-order modes would receive more than −20 dB of the total input power present at the splice between the lead-in fiber and the micro-wire. Since there are two transitions between the lead-in fiber and the rod, multipath interference within the range of –40 dB should be achievable under optimal conditions, even in the ‘micro-wire’ (multimode) case. Another important difference between the ‘nano-wire’ and ‘micro-wire’ ranges is in the effective index of the fundamental mode propagating within the field interaction region. The effective index of the FFAD within a ‘nanometer’ regime is close to the index of the surrounding medium, while in the case of a ‘micrometer’, the effective index of the fundamental mode is close to the rod index.
Furthermore, the proposed FFAD design offers the possibility of fine-tuning the effective refractive index of the fundamental mode’s propagation with the micro-wire. The effective index of the fundamental mode can be controlled by changing the RI of the core and/or the entire rod, which can be achieved by adding dopants to the rod. The effective index of the fundamental-mode within the field interaction region can thus be set within a span that approximately corresponds to the RI range obtainable by silica doping (e.g. by assuming available fluorine doping at the lower-end and titania/germania doping at the higher end, this refractive index range is roughly between 1.415 and 1.515). This capability might be particularly useful in applications where precise tuning of the effective index regarding the device’s fundamental mode is required in order to satisfy the phase-matching conditions required when coupling between the proposed FFAD and, for example, the surrounding medium, the structure, or another photonic device.
3 FFAD manufacturing
The FFAD shown in Fig. 1 was produced by a micromachining process that utilizes the selective etching of phosphorus pentoxide (P2O5)-doped silica glass [45, 46]. When P2O5 is introduced into the silica, the etching rate of the silica in hydrofluoric acid (HF) can be substantially increased . This property can be used to create a large and preferential etching area within a section of specially-designed structure-forming fiber (SFF), which can be further removed very selectively in HF to form a desired microstructure.
Those SFFs used for the production of experimental FFAD devices, are presented in Fig. 4 . The fiber in Fig. 4(a) consisted of a circular GeO2-doped core (the core Δn and diamter were approximately matched to standard SMF, e.g. Δn = 0.0042 and diameter = 10 µm), a circular inner pure silica layer (with thickness of 4.5 µm), a large elliptical P2O5-doped region, and circular non-symmetric pure silica cladding. Figure 4(b) shows the SFF used for the production of a homogenous micro-wire (e.g. silica rod) FFAD. This fiber design is identical to the design shown in Fig. 4(a), except for the inner fiber region consisting entirely of pure silica (no core) and having a diameter corresponding to 20 µm. The P2O5 concentrations in both fibers corresponded to about 8.5% mol. During the fabrication of the SFF preform, a standard modified chemical vapor deposition (MCVD) technique was used to deposit P2O5-doped region, the inner pure silica layer, and the GeO2-doped core. After preform characterization, the outer silica layer consisting of the initial substrate tube was partially removed at the opposing sides by a grinding/polishing process. The drawing of such fiber preform-yielding fibers, as shown in Fig. 4 (reduced viscosity of the P2O5 doped region causing a partial collapse of this region into an elliptical cross-sectional form; the pure silica inner cladding and the core geometry were unaffected).
The designs of the SFF, other than those described above, are also possible but to allow for the formation of side-support regions, it is important to break the circular symmetry of the P2O5-doped region, the outer silica region, or both regions simultaneously.
Experimental FFADs were produced by splicing a short section of the SFF in-between two single-mode fibers (Corning SMF 28e), as shown in Fig. 5 . The lengths of these selections also determined the active lengths of the field-access regions, e.g. micro-wires, and can be typically within the range from a few tens of microns to a few millimeters. The pure silica cladding of the SFF had the same glass transition temperature as the lead-in standard single-mode fiber cladding, which allowed for effortless fusion-splicing amongst both fiber types.
The final devices were obtained after about 12-20 min of etching in 30% HF. The HF first uniformly etched the pure silica cladding, but once it came into contact with the P2O5-doped region, it preferentially removed this region at a higher rate, leaving behind the desired structure. The inner pure silica region also acted as a stop-layer (due to the considerably lower etching-rate of pure silica relative to the P2O5-doped silica), which contributed to the final uniformity of the micro-wire.
Besides the removable fiber-clamp that allowed for the holding and immersion of the spliced fiber assembly into HF, no other external mechanical support was used during the device production. A tensile strength of about 1.5 N was obtained for the finished devices, which is well within the limits required for most device manipulation and packaging applications.
The low etching-rate of the inner pure silica region also provides extra etching time, which can be used for performing the fine-tuning regarding the final micro-wire thickness. The possibility of precisely manipulating the micro-wire’s diameter and refractive index profile can be used to fine-tune the optical properties, as described further below.
When the P2O5 region is selectively removed during etching in HF, the thinning of the cladding region compresses the mode-field into the core and thus causes change in the device’s transmission loss. This change in loss can be used as an indication of the micro-wire thickness and can thus serve as a feedback parameter for the etching process termination. A simple setup, which allows for on-line observation of the device’s transmission during etching has thus been added, and is shown in Fig. 6 . This transmission observation-assisted etching termination eliminated the need for precise control of the etching conditions, such as acid concentration and temperature, while providing a high reproducibility of the FFAD production process.
The transmission change during etching is different for the core-cladding and pure silica rod micro-wire FFAD designs, as shown in Fig. 7 . Figure 7(a) shows typical loss vs. time obtained during etching of the FFAD device that utilizes the core cladding design. In this case the modal fields of the lead-in fiber and the micro-wire region are (about) the same before etching, while etching of the cladding compresses the FFAD’s field into the core, which causes increased loss. This loss increase can be used as an indication of actual silica cladding thickness, and thus as a parameter for terminating the etching process. For example, Fig. 7(a) indicates three typical termination points (measures were performed at 1550 nm). Termination point 1 corresponds to the conditions when the transmission becomes affected by the etching process. The cladding thickness corresponds to about 2.5 μm at this point (here the field interaction strength is minimal). The termination point 1 was achieved after approximately 12 min of etching in 30% HF. Termination point 2 indicates the etching time when the loss increases to −1 dB. At this point the cladding thickness corresponds to about 1.1 μm. The transmission loss of about −2 dB (Termination point 3) is reached when the cladding thickness reduces to 0.3 μm, e.g. after about 21 min of etching in 30% HF. These values were determined experimentally (using loss measurements and an electron microscope) and are also consistent with Fig. 2. Furthermore, Fig. 7(b) shows typical transmission vs. time obtained during the etching of the FFAD device utilizing monolithic silica rod micro-wire design. The transmission is low at the beginning of the etching process, since there is no waveguide structure between both lead fibers. When the P2O5 region is gradually removed during etching in HF, the HF and silica rod form a waveguide structure that results in structure reduction loss. The minimum loss indicates the point when the fundamental mode of the rod is best matched to the lead-in fiber MFD. This local maximum also represents an ideal moment for etching termination in the case where FFAD is to be used within an aquatic medium (if FFAD is to be used in air, slightly earlier termination is optimal, as indicated by the curves in Fig. 3). While further etching should theoretically result in a second transmission peak (e.g. when the silica rod’s diameter is reduced to within the nanowire range, as predicted in Fig. 3), we were unable to observe this second loss reduction. This was most likely due to the lack of proper FFAD- forming fiber for this application, e.g. the FFAD-forming fibers available for testing all had relatively large (>15 μm) pure silica cores. Further efforts towards creating a very thin pure silica region in the center of the SFF could perhaps allow for the formation of a FFAD nano-rod.
The proposed manufacturing method is also very versatile in terms of possible micro-wire lengths. Figure 8(a) demonstrates an almost 2 mm-long FFAD, while Fig. 8(b) shows only an 18 µm-long device. Any practical length between a few tens of micrometers and several millimeters can be easily achieved.
4 Experimental results
The proposed FFAD devices were firstly experimentally-evaluated for achievable insertion losses. A considerable number (over 20) of FFADs utilizing monolithic micro-wire designs were produced for determining the minimum and typical insertion losses. The minimum insertion loss using SFF, as shown in Fig. 9 , was 0.385 dB when submerged in water, which is close to the theoretically-predicted value of 0.35 dB (see Fig. 2). However, the more typical insertion loss was around 0.5 dB, which can likely be attributed to the residual core eccentricity of the SFF (SFF preform was hand-polished on its sides to break its circular symmetry, which resulted in limited concentricity of the preform). This and all further measurements described below were performed at 1550 nm.
Furthermore, six FFADs were produced utilizing the core-cladding micro-wire design and submerged into mediums with different refractive indexes (the SFF shown in Fig. 4(a) was used to produce all six devices). All devices had the same length, e.g. 400 µm, but different thicknesses of the cladding surrounding the GeO2-doped core (the thicknesses of the produced FFAD’s micro-wires were measured under a scanning electron microscope). Figure 9 shows the experimentally-measured losses of the produced FFAD as a function of cladding thickness when submerged into mediums with different indexes of refraction. These experimentally- measured losses were in reasonable agreement with the theoretical predictions shown in Fig. 2, especially when considering the slight core-eccentricity of the experimentally-produced SFF, and only an approximate match to the SMF-28 (core inner-cladding index difference was 0.31% in SFF as opposed to 0.36% core-cladding difference in SMF 28, while the core diameter corresponded to 10 μm in SFF instead of 8.2 μm as in SMF-28). Further improvements are thus likely with further SFF fiber optimization.
Further tests related to an observation of FFAD’s field interaction performance, more precisely to the FFAD response to the external medium refractive index variation. Figure 10 presents the transmission of the 450 µm-long core-cladding type FFAD as a function of the surrounding medium refractive index (cladding thickness was about 1.5 μm). The experimental device was submerged into the refractive index liquid (Cargille Inc., USA), while controlling the liquid temperature to vary the liquid refractive index (exact liquid’s refractive index was calculated from the liquid’s temperature and the manufacturer’s thermo-optic and dispersion data for the liquid at 1550 nm). During determination of the response, shown in Fig. 10 (and subsequent similar measurements), we neglected change of FFAD’s (silica’s) refractive index that occurred during temperature cycling. While this induces some offset in refractive index curves measured at elevated temperatures, this offset is small as the dn/dT of the refractive index fluid is about 40 times larger than dn/dT of silica (other factors, such as dn/dT non-linearity, can have comparable or even larger effects on the index determination while varying the fluid temperature). A smooth and almost completely oscillation-free decoupling of the field from the core was observed when the index of refraction exceeded the effective index of the core.
Figure 11 shows the responses of the FFAD devices with different active lengths to the surrounding medium refractive index changes. Three FFADs were produced, which had micro-wires composed of homogenous (pure silica) rods with identical diameters, but different lengths (120 μm, 310 μm, and 650 μm). The device with the longest micro-wire exhibited the most abrupt and the greatest change in the transmission as a consequence of refractive index change. Such behavior can be intuitively expected, since shortening of the device reduces the field interaction length. The micro-wire length, which is well controlled during the FFAD production process, can thus be used to set devices’ maximum insertion losses, even when conditions for the field-decoupling from the micro-wire region are achieved. This capability might be, for example, useful in the design of multiplexed sensors based on FFADs or FFAD arrays, where the transmission loss of each device must be limited regardless of local conditions present on the device, in order to allow the interrogation of all devices within the array. These results were further confirmed by a beam propagation method simulation as shown in Fig. 11(b) (Optiwave’s OptiBMP simulation package was used for this modeling). The modeled and experimental results are in good agreement.
Finally, tuning of FFADs modes’ effective indexes is demonstrated by producing an additional up-doped SFF, as shown in Fig. 12 . The up-doped SFF included the core cladding structure, but both the core and the cladding were additionally titanium up-doped to raise the refracted indexes of both layers by 0.017. This fiber was further used to produce an up-doped FFAD.
Figure 13 demonstrates the responses of three different FFADs to the surrounding medium refractive index change. The first device utilized micro-wire made of a pure silica rod (e.g. using the SFF shown in Fig. 4(b)), the second device employed a micro-wire composed of a GeO2-doped core and pure silica cladding (using the SFF shown in Fig. 4(a)), and the third device used an up-doped micro-wire (using the SFF shown in Fig. 12). All three devices were simultaneously immersed into the same refractive index liquid, while varying the liquid’s index using temperature control. The decoupling of the optical-field from individual FFAD devices happened at different values of surrounding medium refractive indexes: decoupling from the device utilizing pure silica rod micro-wire occurred at about n = 1.444, the decoupling from the device utilizing core-cladding micro-wire design occurred at approximately n = 1.4486, and the decoupling from the device utilizing up-doped core-cladding design happened at around 1.4660. For comparison, the effective index of the fundamental mode in the pure silica rod was close to the pure silica level, e.g. 1.444, the effective index of the mode in SMF compatible GeO2 core was about 1.449, and the effective index of the mode in SMF compatible up-doped fiber should be equal to the effective index of the SMF mode increased for the up-doping level, i.e. to about 1.449 + 0.017 = 1.466 in this particular experimental case. The effective refractive index difference between those modes propagated in core-cladding and the up-doped core cladding micro-wire, thus corresponded to 0.017 which matched well the measured decoupling refractive index difference of 0.0174 in the case of the core cladding and up-doped core cladding FFADs, as indicated by Fig. 13. This experiment thus demonstrated the possibility of fine-tuning the effective index of the fundamental mode of the FFAD’s micro-wire through adjustment (doping) of the SFF. This capability adds to the versatility of the proposed FFADs since it allows for precise and environmentally-stable phase-matching between a FFAD field and any potential surrounding medium, device, or another micro structure.
This paper presented the designing and introduction of an effective method for producing all-fiber, all-silica fiber-field access devices that can form a versatile base for creating a variety of miniature, in-line, fiber optic micro-photonic systems and devices. The proposed device is miniature in size, while retaining fully circular-symmetrical geometry and low insertion loss. The proposed device is versatile and can be configured to accommodate the various needs and requirements of particular photonic designs. In particular it provides an opportunity to precisely set an effective index for mode propagation within field-access regions, through fiber-doping.
The proposed fiber field-access devices were created through a micromachining process based on selective etching using purposely-designed phosphorus-doped silica fibers. Since the device production is accomplished through specialty fiber manufacturing (a single fiber production can yield a large number of devices) it also presents a potentially cost-effective production process suitable for high-volume production.
We would like to thank Borut Lenardic and Stanislav Campelj from OptaCore d.o.o., Slovenia who prepared the MCVD recipes, and supplied the experimental SFFs. This work was sponsored by Slovenian Research Agency conducted under grant. no. P2-0368 and J2-3623.
References and links
1. R. Bharadwaj, V. V. R. Sai, K. Thakare, A. Dhawangale, T. Kundu, S. Titus, P. K. Verma, and S. Mukherji, “Evanescent wave absorbance based fiber optic biosensor for label-free detection of E. coli at 280 nm wavelength,” Biosens. Bioelectron. 26(7), 3367–3370 (2011). [CrossRef] [PubMed]
2. L. K. Chau, Y. F. Lin, S. F. Cheng, and T. J. Lin, “Fiber-optic chemical and biochemical probes based on localized surface plasmon resonance,” Sensor. Actuat. Biol. Chem. 113(1), 100–105 (2006).
3. J. M. Corres, F. J. Arregui, and I. R. Matias, “Design of humidity sensors based on tapered optical fibers,” J. Lightwave Technol. 24(11), 4329–4336 (2006). [CrossRef]
5. H. S. Haddock, P. M. Shankar, and R. Mutharasan, “Evanescent sensing of biomolecules and cells,” Sensor. Actuat. Biol. Chem. 88(1), 67–74 (2003).
6. Z. M. Hale, F. P. Payne, R. S. Marks, C. R. Lowe, and M. M. Levine, “The single mode tapered optical fibre loop immunosensor,” Biosens. Bioelectron. 11(1–2), 137–148 (1996). [CrossRef]
7. K. Q. Kieu and M. Mansuripur, “Biconical fiber taper sensors,” IEEE Photon. Technol. Lett. 18(21), 2239–2241 (2006). [CrossRef]
9. S. M. Lee, S. S. Saini, and M. Y. Jeong, “Simultaneous Measurement of Refractive Index, Temperature, and Strain Using Etched-Core Fiber Bragg Grating Sensors,” IEEE Photon. Technol. Lett. 22(19), 1431–1433 (2010). [CrossRef]
10. P. Polynkin, A. Polynkin, N. Peyghambarian, and M. Mansuripur, “Evanescent field-based optical fiber sensing device for measuring the refractive index of liquids in microfluidic channels,” Opt. Lett. 30(11), 1273–1275 (2005). [CrossRef] [PubMed]
11. M. I. Zibaii, A. Kazemi, H. Latifi, M. K. Azar, S. M. Hosseini, and M. H. Ghezelaiagh, “Measuring bacterial growth by refractive index tapered fiber optic biosensor,” J. Photochem. Photobiol. B 101(3), 313–320 (2010). [CrossRef] [PubMed]
12. H. Choi, Y. Jeong, and K. Oh, “Wide, tunable band rejection filter based on micro-optical waveguide on microactuating platform covering O, E, S, C, L, and U bands,” Opt. Lett. 36(4), 484–486 (2011). [CrossRef] [PubMed]
13. C. A. Millar, M. C. Brierley, and S. R. Mallinson, “Exposed-core single-mode-fiber channel-dropping filter using a high-index overlay waveguide,” Opt. Lett. 12(4), 284–286 (1987). [CrossRef] [PubMed]
14. K. R. Sohn and J. W. Song, “Tunable in-line fiber optic comb filter using a side-polished single-mode fiber coupler with LiNbO3 overlay and intermediate coupling layer,” Opt. Commun. 203(3–6), 271–276 (2002). [CrossRef]
15. M. Wilkinson, A. Bebbington, S. A. Cassidy, and P. Mckee, “D-fibre filter for erbium gain spectrum flattening,” Electron. Lett. 28(2), 131–132 (1992). [CrossRef]
16. C. L. Lee, Z. Y. Weng, C. J. Lin, and Y. Y. Lin, “Leakage coupling of ultrasensitive periodical silica thin-film long-period grating coated on tapered fiber,” Opt. Lett. 35(24), 4172–4174 (2010). [CrossRef] [PubMed]
17. K. H. Smith, B. L. Ipson, T. L. Lowder, A. R. Hawkins, R. H. Selfridge, and S. M. Schultz, “Surface-relief fiber Bragg gratings for sensing applications,” Appl. Opt. 45(8), 1669–1675 (2006). [CrossRef] [PubMed]
18. V. K. S. Hsiao, Z. Li, Z. Chen, P. C. Peng, and J. Tang, “Optically controllable side-polished fiber attenuator with photoresponsive liquid crystal overlay,” Opt. Express 17(22), 19988–19995 (2009). [CrossRef] [PubMed]
19. S. Pu, X. Chen, Y. Chen, Y. Xu, W. Liao, L. Chen, and Y. Xia, “Fiber-optic evanescent field modulator using a magnetic fluid as the cladding,” J. Appl. Phys. 99(9), 093516 (2006). [CrossRef]
20. X. Tian, X. Cheng, W. Wu, Y. Luo, Q. Zhang, B. Zhu, and G. Zou, “Reversible All-Optical Modulation Based on Evanescent Wave Absorption of a Single-Mode Fiber With Azo-Polymer Overlay,” IEEE Photon. Technol. Lett. 22(18), 1352–1354 (2010). [CrossRef]
21. M. Cai and K. Vahala, “Highly efficient optical power transfer to whispering-gallery modes by use of a symmetrical dual-coupling configuration,” Opt. Lett. 25(4), 260–262 (2000). [CrossRef] [PubMed]
22. M. K. Chin and S. T. Ho, “Design and modeling of waveguide-coupled single-mode microring resonators,” J. Lightwave Technol. 16(8), 1433–1446 (1998). [CrossRef]
23. G. Griffel, S. Arnold, D. Taskent, A. Serpengüzel, J. Connolly, and N. Morris, “Morphology-dependent resonances of a microsphere-optical fiber system,” Opt. Lett. 21(10), 695–697 (1996). [CrossRef] [PubMed]
24. A. B. Matsko and V. S. Ilchenko, “Optical resonators with whispering-gallery modes - Part I: Basics,” IEEE J. Sel. Top. Quantum Electron. 12(1), 3–14 (2006). [CrossRef]
26. Y. W. Song, S. Yamashita, C. S. Goh, and S. Y. Set, “Carbon nanotube mode lockers with enhanced nonlinearity via evanescent field interaction in D-shaped fibers,” Opt. Lett. 32(2), 148–150 (2007). [CrossRef] [PubMed]
27. Y. J. Zhang, F. F. Zhong, W. B. He, Y. Zhang, Y. Wang, J. Xu, and J. L. Ju, “A long uniform taper applied to an all-fiber Tm3+ doped double-clad fiber laser,” Laser Phys. 20(11), 1978–1980 (2010). [CrossRef]
28. A. Diez, M. V. Andres, and D. O. Culverhouse, “In-line polarizers and filters made of metal-coated tapered fibers: Resonant excitation of hybrid plasma modes,” IEEE Photon. Technol. Lett. 10(6), 833–835 (1998). [CrossRef]
29. S. G. Lee, J. P. Sokoloff, B. P. McGinnis, and H. Sasabe, “Fabrication of a side-polished fiber polarizer with a biref ringent polymer overlay,” Opt. Lett. 22(9), 606–608 (1997). [CrossRef] [PubMed]
32. L. Su, T. H. Lee, and S. R. Elliott, “Evanescent-wave excitation of surface-enhanced Raman scattering substrates by an optical-fiber taper,” Opt. Lett. 34(17), 2685–2687 (2009). [CrossRef] [PubMed]
33. M. H. Cordaro, D. L. Rode, T. S. Barry, and R. R. Krchnavek, “Precision fabrication of D-shaped single-mode optical fibers by in situ monitoring,” J. Lightwave Technol. 12(9), 1524–1531 (1994). [CrossRef]
35. T. L. Lowder, B. R. Tebbs, R. H. Selfridge, S. M. Schultz, K. H. Smith, and T. D. Monte, “Polarization analysis of surface-relief D-fiber Bragg gratings,” Appl. Opt. 46(13), 2387–2393 (2007). [CrossRef] [PubMed]
36. F. Bilodeau, K. O. Hill, S. Faucher, and D. C. Johnson, “Low-loss highly overcoupled fused couplers: Fabrication and sensitivity to external pressure,” J. Lightwave Technol. 6(10), 1476–1482 (1988). [CrossRef]
37. Y. Takeuchi and J. Noda, “Novel fiber coupler tapering process using a microheater,” IEEE Photon. Technol. Lett. 4(5), 465–467 (1992). [CrossRef]
38. D. Donlagic, “In-line higher order mode filters based on long highly uniform fiber tapers,” J. Lightwave Technol. 24(9), 3532–3539 (2006). [CrossRef]
39. H. S. Haddock, P. M. Shankar, and R. Mutharasan, “Fabrication of biconical tapered optical fibers using hydrofluoric acid,” Mat. Sci. Eng. B-Solid 97(1), 87–93 (2003). [CrossRef]
40. J. P. Laine, B. E. Little, and H. A. Haus, “Etch-eroded fiber coupler for whispering-gallery-mode excitation in high-Q silica microspheres,” IEEE Photon. Technol. Lett. 11(11), 1429–1430 (1999). [CrossRef]
41. E. J. Zhang, W. D. Sacher, and J. K. S. Poon, “Hydrofluoric acid flow etching of low-loss subwavelength-diameter biconical fiber tapers,” Opt. Express 18(21), 22593–22598 (2010). [CrossRef] [PubMed]
42. J. D. Love, W. M. Henry, and W. J. Stewart, “Tapered single-mode fibres and devices. I. Adiabaticity criteria,” IEEE Proc.-J 138(5), 343–354 (1991).
43. L. M. Xiao, M. D. W. Grogan, S. G. Leon-Saval, R. Williams, R. England, W. J. Wadsworth, and T. A. Birks, “Tapered fibers embedded in silica aerogel,” Opt. Lett. 34(18), 2724–2726 (2009). [CrossRef] [PubMed]
44. G. Brambilla, F. Xu, P. Horak, Y. Jung, F. Koizumi, N. P. Sessions, E. Koukharenko, X. Feng, G. S. Murugan, J. S. Wilkinson, and D. J. Richardson, “Optical fiber nanowires and microwires: fabrication and applications,” Adv. Opt. Photon. 1(1), 107–161 (2009). [CrossRef]
45. S. Pevec, E. Cibula, B. Lenardic, and D. Donlagic, “Micromachining of Optical Fibers Using Selective Etching Based on Phosphorus Pentoxide Doping,” IEEE Photon. J 3(4), 627–632 (2011). [CrossRef]