The paper presents a new approach to developing exposed-core fibers. We designed a new asymmetric structure of suspended core fibers with series of additional air holes in the cladding. Using the standard wet etching method we removed a part of glass, demonstrating that the method allows to open a selected air hole surrounding the suspended core. Such modified of fibers can be used to build sensors and devices dedicated to chemical and biological studies and based on the interaction of light with liquids. We used the developed fiber to develop an interferometric sensor that measures changes in the refractive index with a high accuracy. As a proof of concept, we present the experimental measurement results of the ethanol concentration in water.
© 2018 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
The interaction of an evanescent field of a guided mode with the tested substance is the basis for constructing of many chemical and biological sensors [1,2]. In such sensors, it is important to ensure that the light guided in the fiber core interacts as much as possible with the surrounding substance, gas or liquid. One of the solutions in this case is the use of different types of optical fibers designed such that they allow for real time sensing and the mechanical stability of the sensor.
Several configurations of fiber optic sensors based on evanescent wave have been proposed. First, fibers with the removed coating , D-shaped fibers , and tapered fibers  are easy to make but they are not mechanically resistant and provide a short measuring length up to several dozen micrometers, which affects the accuracy of the measurements. Their main advantage is free access to the tested substance in the proximity of the core where light is propagated, which ensures immediate response of the sensor. Second, selectively liquid-filled photonic crystal fibers [6,7], nanowires , suspended core  or wagon wheel fibers  provide a strong interaction of the evanescent wave with analytes and a long pathway of interaction. However, the measurement requires filling the fiber with analyte, for which it is necessary to drill holes to open access to the core area. Moreover, the filling of the fiber itself is a long-lasting process, which makes fast, on-line measurements impossible. Often, capillary forces and diffusion are insufficient for filling the fiber with liquids  with gases  is, so the fiber needs to be filled under pressure. In addition, there is also a high risk of blocking of the holes in the microstructured fibers . A solution is using exposed-core microstructured optical fibers (MOF) [9,14–16], which combine the advantages of the above mentioned fibers. On the one hand, they provide a strong interaction between the optical field and the tested substance, and on the other hand, they offer free access to the core for in-line measurement.
The first MOF type structure was proposed by Hoo et al. in 2003 . The practical implementation of this type of fiber is possible by direct injection , focused ion beam milling , drilling holes into the cladding of preforms , extrusion of soft glass , drilling and cutting the polymer [14,16], molding the chalcogenide glass , and ultrasonic drilling and cutting of silica glass . The use of a specific technique may depend on the tools available and the characteristics of the final fiber.
Measuring the properties of gasses or liquids can be based on various physical phenomena such as absorption spectroscopy , fluorescence , surface plasmon resonance (SPR) , Raman spectroscopy , multimode interferometry  and multi-core mode-coupling . However, the most common measurement is based on the change in the refractive index . The most accurate methods are based on the SPR and allow to measure the refractive index at the level of 10−7 RIU . However, when fibers with exposed core are used, the accuracy of 5.5 x10−3 RIU  is achieved.
Physical functionalization  or graphiting  is used to increase the change in the refractive index or to 'sensitize' selectively the sensor to a particular measured chemical or biological agent. In addition, interferometer measurements with Bragg gratings  and long-period gratings  are used to increase sensor functionality and measurement accuracy.
In the present paper, we demonstrate a new PCF fiber architecture suitable for opening the side channel with the use of wet etching. This is an alternative method of fabricating exposed-core fibers. As a proof-of-concept example of the use of the fabricated MOF, we built an interferometric sensor in order to measure the concentration of ethanol in water by changing the refractive index of light.
2. Fiber design and fabrication
The cross-section of most optical fibers shows the rotational symmetry. Chemical etching of such fibers would lead to the complete exposing of the core. This, in turn, would significantly decrease the mechanical strength of such a fiber, making the approach unpractical. Therefore, to use chemical etching it is necessary to break the fiber symmetry, and fabricate a fiber with only partial access to the core after the wet etching. In this case a part of fiber cladding will mechanically strengthen the fiber. Producing such a fiber can be done in two ways: Either by using two types of glasses that react to etching at different rates, or by changing the geometry of the MOF made of one type of glass. Thus, in subsequent experiments we explored which of the two methods would be more suitable for our purposes.
First, in the preliminary etching test in fluorohydric acid (HF) and nitric acid (HNO3), we examined the etching rate of glass rods made from various heavy metal oxide glasses. At the beginning, the glass rods had a diameter of 1 mm. The diameter of the rods was measured during the etching process at subsequent moments of time. Unfortunately, experiments showed that for glasses of a similar rheology, which can be drawn together on the fiber optic tower, the maximum difference in etching rate reached 1.5. This value was insufficient to fabricate a double glass fiber with open side channel while maintaining mechanical strength of the fiber.
Therefore we opted for developing an air-glass MOF. On the basis of the experimental etching of various glasses, finally glass labeled NC21 with the following mass composition: SiO2 - 55%, Al2O3 - 1%, B2O3 - 26%, Na2O - 9.5%, K2O - 5.5%, LiO2 - 3% and As2O3 – 0.8% was selected for air-glass MOF development. The details of optical and thermo-physical properties were previously published in . The suspended core structure was chosen for further consideration. It provides a strong interaction of the propagating mode in the core with the medium around the core. The use of the stack-and-draw method  for suspended core fiber development allows for a freedom in designing additional holes to break the rotating symmetry of the fibers and finally allows for effectively anisotropic etching of the fiber.
Two types of fibers were designed, with simplified versions of preforms shown in Fig. 1. The goal of these designs was to obtain effectively large anisotropic wet etching of the fiber structure to open one of three air holes surrounding the core and remain large part of the glass cladding to sustain a mechanical strength of the fiber. Several fibers were drawn from each preform, changing the air pressure maintained in the central holes from 40 to 80 Pa. Altogether, we developed 8 types of fibers dedicated to further post-processing with wet etching, as shown in Fig. 2 and Fig. 3. The basic geometrical parameters of the individual fibers are presented in Table 1 and Table 2.
3. Wet etching of suspended core fibers
The next stage in the development of the optical fiber with an open side channel was the etching of the glass in such a way as to open one of the hollows around the core and simultaneously leave as much glass around as possible to ensure mechanical stability. An important parameter taken into consideration was also the time which elapses from the opening of the first hollow around the core to the opening the next hollow. This time should be sufficient to stop wet etching.
In order to estimate how much glass is left after the etching process and how much time elapses from the opening of the first channel to the opening of the next channel, we carried out a numerical analysis of the etching process for each of the produced fibers using COMSOL MULTIPHYSICS modeling environment. The simulation used the fast marching method  which proved to be accurate enough to estimate the time needed for etching various fibers. The etching rate for an 10% concentration of the HF acid was about 1 µm/min and 0.5 µm for a 5% concentration. The only numerical parameter of the simulation was the glass etching rate for a given concentration of HF acid, which was experimentally measured during a preliminary etching test. The results of the simulation, presented in Fig. 4, allowed to select optimal fibers of two types.
In the case of fibers fabricated from preform PR1, we found that the highest asymmetry of the etching process while maintaining as much glass as possible is possible for the F1.5 fiber. The simulation time map of fiber etching in HF acid at 10% concentration and the shape of fiber cross-section at characteristic moments of the etching process are shown in Fig. 5.
The presented results show that approximately after 27 minutes of the etching process [Fig. 5(b)], the outer layer of glass is removed and the acid solution reaches the first air holes in the cladding. In 32 minutes [Fig. 5(c)] the second layer of holes is already etched and in 33 minutes the third layer of air holes is removed. After 34 minutes and 41 seconds [Fig. 5(d)] HF acid reaches one of the holes in the first layer of air holes around the core. This is an optimal moment for breaking the etching process, since only 29 seconds later the fiber will be damaged. At 35 minutes and 10 seconds the first of three struts that support the core is interrupted, and the acid solution fills the next air hole [Fig. 5(e)]. After a further 29 seconds, all struts are etched [Fig. 5(f)]. Such a short time window to finish the etching process proved difficult in the real process. Therefore, we assumed that the final etching would be carried out in HF acid with a concentration of 5%. In this case the respective etching times would be approximately 2 times longer.
Next, we performed a series of experiments to experimentally monitor the wet etching process. In this case we coupled light into the core and continuously measured the intensity of light propagating in the core during the etching process. We observed that the changes in the total light intensity of the core are negligible during the etching process, until the first hole surrounding the core is not opened [Fig. 6(a)]. At the moment when the HF acid penetrates an air hole close to the core, slight changes in light intensity and ‘sparking’ near the core are observed. These phenomena are related to the changes of the guiding conditions in the fiber core where the air in the hole was being replaced by higher effective index HF solution. At this moment, the etching process should be stopped. Otherwise a significant and rapid decrease in the total light intensity is observed. The decrease in the intensity to zero means that the continuity of the core had been broken and the core of the fiber was etched. In conclusion, the implementation of the proposed procedure and the monitoring of the wet etching process allows for a reproducible fabrication of fibers with the open side channel.
In the case of fibers obtained from PR2 preform, we determined that F2.1 fiber would be the optimal choice. The key criterion for fiber selection was the thickness of the struts supporting the core. We selected the fiber structure for which the struts were the thickest, which resulted in a longer time needed to etch them. The simulation time map of fiber etching in 5% concentration of HF acid and the cross-sections at the characteristic moments of the etching process are shown in Fig. 7. The results demonstrate that after 24 minutes of etching the outer layer of glass in the cladding was removed [Fig. 7(c)]. After 27 minutes and 8 seconds, the first hole surrounding the core [Fig. 7(d)] was opened and 17 seconds later the other holes surrounding the core [Fig. 7(e)] were filled with acid. Such a short time between opening the access to the core and the moment when the struts supporting the core were etched was due to the very small thickness of the struts supporting the core, which was less than 600 nm (Table 2). After an additional 17 seconds, the core supported by the struts was finally etched [Fig. 7(f)].
The etching process should be terminated before the glass struts supporting the core are etched. To terminate the etching process we applied pure water and next removed the fiber form the solvent. However, it is impossible to stop the wet etching process immediately. The monitoring of light intensity in the core, and the attempt to terminate the etching process when the decrease of intensity had been detected were insufficient [Fig. 6(b)]. We observed that light was transmitted for a few more seconds, and next, the core was etched since the transmission was reduced to zero.
Therefore, the procedure for monitoring the etching process was modified. We lighted one side of the fiber and measured the light intensity in 4 areas, as presented in Fig. 6(c). The stopping of the etching process at the moment when the light intensity began to decrease in area 2 or 3 allowed to control the etching process. However, minimal differences in the fiber geometry and material inhomogeneities caused that the obtained lengths of the open side channel fiber section varied from a few hundred micrometers to several millimeters [Fig. 8(a)]. The same figure also shows microscopic images of the cross sections of the etched fibers. [Fig. 8(b,c)]. In conclusion, also for this fiber, the implementation of the proposed procedure and the monitoring of the wet etching process allows for reproducible fabrication of fibers with the open side channel.
The etching process does not only selectively open access to the fiber core, but also removes a part of the glass from the whole fiber. This weakens its construction. Therefore, during the etching process and when the fiber was tested as a sensor, the fiber was glued to the glass substrate strengthening the whole structure. However, in order to check to what extent the etching process weakened the fiber, stress tests were carried out. In the case of F1.5 fiber, before etching, the fiber broke at a load of 1.35 kg. After etching, the fiber broke at a load of 190 gr. This shows that the fiber weakened more than seven times. In the case of F2.1 fiber, before etching the fiber broke at a load of 0.95 kg. After etching, the fiber broke at a load of 0.40 kg. In this case, the weakening was less than 2.5 times, which confirms that the optimized structure had better mechanical properties. However, an approximate comparison of the cross-sectional area of the fibers before and after the etching process shows that at least a part of the glass that was preserved after the etching process in HF was further weakened. This could have resulted from micro-cracks in the glass .
The presented method of exposed-core fiber fabrication is an alternative to the other existing methods presented in the introduction [9–21]. In our method, we use standard techniques known from the fabrication of photonic fibers. In the first stage, a glass preform is stacked, then the fiber is drawn out in the optical tower. A non-standard procedure is the etching of the fiber. On the one hand, etching requires precise monitoring of the process but does not require specialized equipment that is often expensive. Therefore, depending on the equipment available, it can be an alternative method to fabricate fibers with an open side channel.
4. Numerical modeling of light propagation in fiber with open side channel
To exemplify how the fabricated fiber with an open side channel could be used in practice, we have constructed an interferometric sensor. First we carried out computer simulations in order to determine modal properties of the fabricated fibers and potential properties of the interferometric sensor. For modelling we used geometrical parameters of the developed fiber F2.1 obtained with the SEM photos [Fig. 6]. The commercially available finite element modelling (FEM) COMSOL package was used for the analysis.
When one of the holes is infiltrated with liquid, a rotational symmetry of the suspended core fiber is broken, so the fiber becomes birefringent. We have calculated the effective refraction index for the fundamental mode for various concentrations of water-ethanol mixture and intensity distribution for both polarization components of the fundamental mode [Fig. 9].
In order to examine the effect of the concentration of the water-ethanol mixture on the fundamental mode's interaction with the liquid we calculated the effective mode area (Aeff) and modal power fraction (PF) within the sensing region, defined as :Fig. 10]. A relatively low value of the modal power fraction results from the large diameter of the core (1.75 μm) and the large contrast of the refractive indices between the glass and the tested liquid. On the other hand, the obtained modal power fraction was more than two times higher than in the case of D-fibers previously reported by Ebendorff-Heidepriem et al. .
The results obtained suggest that the use of the presented fiber to measure the concentration of water-ethanol mixture requires the use of the interferometric method. On the other hand, the use of methods based on light absorption measurements will not be appropriate and because such measurement is not sensitive enough since most of the light is guided within the solid core.
5. Experimental verifications
The fibers with the etched side channel (overall length of approximately 50 cm) were used to build an interferometric sensor of the water-ethanol mixture concentration. Importantly, the interferometer measurement was intended only to exemplify the usage of such fibers, but was not intended to achieve a new sensor of high accuracy. In order to protect the fiber against the penetrating of the test liquid thanks to capillary forces, the ends of the etched area were secured with glue. The 4 mm of the open side channel core was exposed to the measured liquid. In the Mach-Zehnder interferometer [Fig. 11], a light beam is split into two parts by beamsplitter 1. One is the reference beam propagating in air and the second one is the test beam propagating through the etched fiber. After recombining the beams by beamsplitter 2, the result of the interference of both light beams is registered on the detector. Temperature compensation is not included in the system at this stage. However, in order to achieve higher measurement accuracy and independence from temperature fluctuation, temperature compensation can be achieved by adding a reference fiber in the second interferometer path, which has the same temperature as the measuring fiber.
As our source we used a temperature stabilized fiber pigtailed laser diode with wavelength 634 nm, and a photodiode as the detector. The measurement was made for several water-ethanol mixture concentration values in two series. In series S1, every several seconds pure ethanol portions were added to the pure water, while in series S2, pure water portions were added to the pure ethanol (97%), in both cases finally reaching the concentration of 50%. Water and ethanol droplets were administered in portions of approximately 50 μl with mini-pipettes. As the reference level (minimum light intensity on the detector), a fiber immersed in pure water was selected. The experimental results are presented in Fig. 12(a). Their comparison with the theoretical results is shown in Fig. 12(b) and Table 3.
As shown in Fig. 12(b), the presented sensor can be used to unambiguously determine the ethanol concentration in water in the range of 5 to 25% where the dependence of light intensity on the concentration is linear. However, the results obtained show that the active open area of 4 mm fiber length is too long to match the full range of variability of ethanol concentration in water from 0% to 100% into a single slope of interferometer characteristics. The full measurement range includes 2 minima and two maxima, which makes it impossible to read the results unambiguously. In order to obtain a monotonic dependence of the measured signal on concentration in the whole range, the active area of the optical fiber should be reduced to approximately 1 mm. However, this would require more control by limiting the etching area. This can be achieved by coating the fiber before the etching process with a protective layer, leaving only the appropriate part of the fiber exposed. The largest concentration measurement error of 4% was recorded for concentrations of 50% and 57%. The errors mainly resulted from a few percent inaccuracy with which the next drops were measured. The obtained results differ from the best results obtained in the measurement of ethanol concentration in water where the accuracy of 0.02%  was obtained. However, the aim of the construction of the sensor was to show the possibility of using the fabricated fiber for measurement, and not to optimize the sensor. The presented sensor, due to its wide access to the core, is characterized by a short reaction time to a change in the concentration of the tested liquid of 1 s. This result is comparable to the results obtained for similar fibers fabricated by other methods .
This paper presented the theoretical and experimental verification of novel design of the suspended core fiber with open side channel. The designed internal fiber structure allows the use of isotropic wet etching to selectively open one of the air holes surrounding the suspended core while maintaining the mechanical stability of the fiber. On the one hand, wet etching requires precise monitoring of the process, but does not require any specialized (and often expensive) equipment that is necessary for many other methods of fabrication of this type of fibers. Therefore, depending on the available equipment, our method may be an alternative for fabricating fibers with exposed-core, compared to the existing methods. However, because it is difficult to stop the etching process at a precise moment, it is best to use fibers where the struts supporting the core are relatively thick (370 nm for fiber F1.5 and 580 nm for fiber F2.1). This, in turn, results in weaker light interaction with the liquid infiltrating the core and does not allow to use techniques based on light absorption.
As an example of application of the new fiber, we proposed to measure the concentration of water-ethanol mixture in the Mach-Zehnder interferometer system. The presented sensor system, despite the non-temperature compensation, made it possible to measure the concentration with an accuracy of several percent. Thanks to the wide access to the core, the reaction time of the sensor to concentration changes was estimated at 1 s. Overall, we have proved that the proposed fabrication method can be a cheaper alternative to other exposed-core fibers fabrication techniques such as drilling or focused ion beam milling.
Foundation for Polish Science Team Programme (Project TEAM TECH /2016-1/1); European Regional Development Fund, FP7-ICT-2013-11 (619205 ACTPHAST); Polish National Science Centre (NCN) (2014/13/B/ST7/01742).
1. I. R. Matias, S. Ikezawa, and J. Corres, Fiber optic sensors: current status and future possibilities (Springer, 2016).
2. S. Yin, P. B. Ruffin, and F. T. S. Yu, Fiber optic sensors (CRC, 2008).
3. K. Newby, W. M. Reichert, J. D. Andrade, and R. E. Benner, “Remote spectroscopic sensing of chemical adsorption using a single multimode optical fiber,” Appl. Opt. 23(11), 1812–1815 (1984). [CrossRef] [PubMed]
4. T. L. Lowder, J. D. Gordon, S. M. Schultz, and R. H. Selfridge, “Volatile organic compound sensing using a surface-relief D-shaped fiber Bragg grating and a polydimethylsiloxane layer,” Opt. Lett. 32(17), 2523–2525 (2007). [CrossRef] [PubMed]
5. T. K. Yadav, R. Narayanaswamy, M. H. Abu Bakar, Y. M. Kamil, and M. A. Mahdi, “Single mode tapered fiber-optic interferometer based refractive index sensor and its application to protein sensing,” Opt. Express 22(19), 22802–22807 (2014). [CrossRef] [PubMed]
7. J. Jensen, P. Hoiby, G. Emiliyanov, O. Bang, L. Pedersen, and A. Bjarklev, “Selective detection of antibodies in microstructured polymer optical fibers,” Opt. Express 13(15), 5883–5889 (2005). [CrossRef] [PubMed]
8. H. Ebendorff-Heidepriem, S. C. Warren-Smith, and T. M. Monro, “Suspended nanowires: fabrication, design and characterization of fibers with nanoscale cores,” Opt. Express 17(4), 2646–2657 (2009). [CrossRef] [PubMed]
9. S. C. Warren-Smith, H. Ebendorff-Heidepriem, T. C. Foo, R. Moore, C. Davis, and T. M. Monro, “Exposed-core microstructured optical fibers for real-time fluorescence sensing,” Opt. Express 17(21), 18533–18542 (2009). [CrossRef] [PubMed]
11. S. Afshar V 5th, S. C. Warren-Smith, and T. M. Monro, “Enhancement of fluorescence-based sensing using microstructured optical fibres,” Opt. Express 15(26), 17891–17901 (2007). [CrossRef] [PubMed]
13. Y. Ruan, E. P. Schartner, H. Ebendorff-Heidepriem, P. Hoffmann, and T. M. Monro, “Detection of quantum-dot labelled proteins using soft glass microstructured optical fibers,” Opt. Express 15(26), 17819–17826 (2007). [CrossRef] [PubMed]
14. E. P. Schartner, G. Tsiminis, A. Francois, R. Kostecki, S. C. Warren-Smith, L. V. Nguyen, S. Heng, T. Reynolds, E. Klantsataya, K. J. Rowland, A. D. Abell, H. Ebendorff-Heidepriem, and T. M. Monro, “Taming the light in microstructured optical fibers for sensing,” Int. J. Appl. Glass Sci. 6(3), 229–239 (2015). [CrossRef]
16. R. Kostecki, H. Ebendorff-Heidepriem, C. Davis, G. McAdam, S. C. Warren-Smith, and T. M. Monro, “Silica exposed-core microstructured optical fibers,” Opt. Mater. Express 2(11), 1538–1547 (2012). [CrossRef]
17. C. M. B. Cordeiro, E. M. Dos Santos, C. H. Brito Cruz, C. J. S. de Matos, and D. S. Ferreiira, “Lateral access to the holes of photonic crystal fibers - selective filling and sensing applications,” Opt. Express 14(18), 8403–8412 (2006). [CrossRef] [PubMed]
18. A. van Brakel, C. Grivas, M. N. Petrovich, and D. J. Richardson, “Micro-channels machined in microstructured optical fibers by femtosecond laser,” Opt. Express 15(14), 8731–8736 (2007). [CrossRef] [PubMed]
19. S. C. Warren-Smith, R. M. André, C. Perrella, J. Dellith, and H. Bartelt, “Direct core structuring of microstructured optical fibers using focused ion beam milling,” Opt. Express 24(1), 378–387 (2016). [CrossRef] [PubMed]
20. P. Toupin, L. Brilland, C. Boussard-Pledel, B. Bureau, D. Mechin, J.-L. Adam, and J. Troles, “Comparison between chalcogenide glass single index and microstructured exposed-core fibers for chemical sensing,” J. Non-Cryst. Solids 377, 217–219 (2013). [CrossRef]
21. E. P. Schartner, A. Dowler, and H. Ebendorff-Heidepriem, “Fabrication of low-loss, small-core exposed core microstructured optical fibers,” Opt. Mater. Express 7(5), 1496–1502 (2017). [CrossRef]
22. J. B. Jensen, L. H. Pedersen, P. E. Hoiby, L. B. Nielsen, T. P. Hansen, J. R. Folkenberg, J. Riishede, D. Noordegraaf, K. Nielsen, A. Carlsen, and A. Bjarklev, “Photonic crystal fiber based evanescent-wave sensor for detection of biomolecules in aqueous solutions,” Opt. Lett. 29(17), 1974–1976 (2004). [CrossRef] [PubMed]
24. A. Khetani, V. S. Tiwari, A. Harb, and H. Anis, “Monitoring of heparin concentration in serum by Raman spectroscopy within hollow core photonic crystal fiber,” Opt. Express 19(16), 15244–15254 (2011). [CrossRef] [PubMed]
25. L. V. Nguyen, K. Hill, S. C. Warren-Smith, and T. M. Monro, “Interferometric-type optical biosensor based on exposed-core microstructured optical fiber,” Sens. Actuat. B Chem. 221, 320–327 (2015). [CrossRef]
27. S. Silva, P. Roriz, and O. Frazao, “Refractive index measurements of liquids based on microstructured optical fibers,” Photonics 1(4), 516–529 (2014). [CrossRef]
30. R. Kostecki, H. Ebendorff-Heidepriem, S. Afshar V, G. McAdam, C. Davis, and T. M. Monro, “Novel polymer functionalization method for exposed-core optical fiber,” Opt. Mater. Express 4(8), 1515–1525 (2014). [CrossRef]
32. C. Silva, J. Coelho, P. Caldas, and P. Jorge, “Fibre sensing system based on long-period gratings for monitoring aqueous environments,” Fiber Optic Sensors (Intech, 2012), pp. 317–342.
33. R. Buczynski, D. Pysz, R. Stepien, A. J. Waddie, I. Kujawa, R. Kasztelanic, M. Franczyk, and M. R. Taghizadeh, “Supercontinuum generation in photonic crystal fibers with nanoporous core made of soft glass,” Laser Phys. Lett. 8(6), 443–448 (2011). [CrossRef]
34. J. Pniewski, R. Kasztelanic, J. M. Nowosielski, A. Filipkowski, B. Piechal, A. J. Waddie, D. Pysz, I. Kujawa, R. Stepien, M. R. Taghizadeh, and R. Buczynski, “Diffractive optics development using a modified stack-and-draw technique,” Appl. Opt. 55(18), 4939–4945 (2016). [CrossRef] [PubMed]
35. J. A. Sethian, Level set methods and fast marching methods evolving interfaces in computational geometry, Fluid mechanics, Computer vision, and Materials science. (Cambridge University, 1999).
36. H. Park, J. H. Cho, J. H. Jung, P. P. Duy, A. H. Tuan Le, and J. Yi, “A review of wet chemical etching of glasses in hydrofluoric acid based solution for thin film silicon solar cell application,” Curr. Photovolt. Res. 5(3), 75–82 (2017).
37. S. C. Warren-Smith, S. Afshar V, and T. M. Monro, “Theoretical study of liquid-immersed exposed-core microstructured optical fibers for sensing,” Opt. Express 16(12), 9034–9045 (2008). [CrossRef] [PubMed]
38. H. Saad, M. T. Ali, and M. K. Abd-Rahman, “High Sensitivity Optical POF Sensor for Detecting Low Ethanol Concentration in Water,” Adv. Mat. Res. 1107, 693–698 (2015). [CrossRef]