1. Introduction

Biosensors are electrical, optical, chemical, or mechanical devices that are capable of detecting various biological and chemical species. During the past several decades, biosensors have been extensively studied for a variety of industrial applications, such as bio-chemical sensing, medical diagnosis, food quality control, and environmental monitoring.

Optical fibers or waveguides has been considered as a very promising platform to build biosensors, as they offer miniaturization, high degree of integration, as well as distributed sensing possibilities. Typically, in such sensors, the spectral properties of the fibers are sensitive to the effective refractive index changes caused by the refractive index changes of the bulk analyte solution, or due to the binding of biomolecules onto the fiber surface. Variations in the effective refractive index of the surrounding medium cause changes in the light propagation conditions, leading to changes in the properties of the reflected or transmitted optical waves. This sensing modality is of particular interest in numerous industrial applications. First, since the refractive index values of materials are related to their compositions, measurement of the analytes refractive indices can be used for the determination of material purity and concentration, as well as physical/chemical process monitoring [1–4]. Particularly, on-line monitoring of the solution concentrations, including heat transfer fluids, coolant, or other dilutions, is of significant importance in many industrial processes [5]. Second, detection of analytes refractive indices is one of the most important chemical/biological analysis methods used in biosensors, because many specimens can be identified by their refractive indices [6–8]. Moreover, binding events occurring on the fiber surface can be detected by analyzing the variations in the effective refractive index of the surrounding medium. This kind of sensing mechanism is especially relevant for biosensing applications, such as precise detection of analyte layers or trace amounts of biomolecules, study of antigen-antibody interactions, and monitoring of surface dynamics, as well as identification of bacteria pathogens [9–11]. Additionally, another highly attractive feature of refractive index sensors is potentially being able to detect various analytes in powder forms including illicit drugs, explosive or hazardous powders, and suspended powder pollutants, because of changes of the refractive index in the vicinity of the fiber surface. This sensing strategy has important applications in the pharmaceutical and food industry, as well as environmental pollution control [12].

Photonic crystal fibers (PCFs), also known as micro-structured optical fibers (MOFs), have enormous potential for chemical and biological sensing. The micron-sized holes running along the length of PCFs enable hosting of the biological/chemical samples in liquid or gaseous forms inside the air holes in the immediate vicinity of the fiber core, thereby significantly enhancing light-analyte coupling and ensuring high sensitivity. Of particular interest is the ability of PCFs to guide in the analyte-filled hollow cores using a photonic bandgap effect. In such fibers, guidance is possible with analytes of any refractive indices. Additionally, PCFs offer a number of other unique benefits compared to the conventional TIR-based (total internal reflection) sensors. First, PCFs naturally integrate optical detection with microfluidic channels, thus, allowing for continuous monitoring of dangerous samples in real-time without exposing the personnel to danger. Second, only a small volume of samples is required for sensing, due to the micro-sized holes of the fluidic channels. Third, PCF-based sensors can be coiled into long sensing cells, thus dramatically increasing the interaction lengths and their sensitivities, while the same is impossible to achieve with traditional TIR-based fiber sensors, as the side-polishing step limits sensor length to only several centimeters. Fourth, PCFs can be mass-produced using a commercial fiber drawing tower in a cost-effective way, while traditional TIR-based fibers require significant post-processing procedures before they can be used for sensing applications.

Generally, PCFs can be divided into two classes, i.e., solid-core PCFs and hollow-core PCFs [13]. Solid-core PCFs typically guide with a modified total internal reflection principle, which is similar to the guidance of traditional step-index fibers. Proposed by Monro et al. [14], the holes in the cladding of an index-guiding PCF can be filled with liquid or gaseous analytes, which are then detected by the evanescent field propagating in these holes. Careful design of the geometry parameters such as core size, air-filling fraction allows an enhanced overlap between the modal field of the fiber and the test analyte. When the cladding holes are filled with liquid analytes with refractive indices higher than that of the fiber material, the guidance of the solid-core PCFs turns into the photonic bandgap guidance. These sensors generally use a spectral-based detection modality. Variations in the refractive index of a liquid analyte filling the fiber would modify the bandgap guidance of the solid-core PCFs, leading to strong spectral shifts in the fiber transmission spectra. Thus, the spectral shifts can be used to extract the changes in the refractive indices of the test analytes. For example, D. K. C. Wu et al. [15] reported a solid-core PCF refractive index sensor, which achieved very high sensitivity of 30100nm/RIU. However, this design requires a complicated selective filling method to introduce the liquid analytes into the porous cladding. Moreover, such sensors can only be used for the detection of analytes with refractive indices higher than the fiber structure material, which limits their applications in chemical/biological fields involving aqueous solution with a refractive index of n~1.33.

Another alternative is using hollow-core PCFs. In this case, the hollow-core is filled with liquid samples. A highly attractive aspect of this configuration is that the modal field is almost completely confined to the liquid samples. The sensing mechanism of these sensors is based on the interrogation of spectral shifts in response to changes in the refractive indices of the liquids filling the fiber core. For example, in [16], the authors experimentally demonstrated a liquid-core refractive index sensor that features a hollow core surrounded by a porous cladding. A sensitivity of ~5000 nm/RIU was reported. However, fabrication of the hollow-core PCFs requires a sophisticated drawing technique, and the fiber is very expensive (thousands of dollars per meter). Moreover, these sensors require a relatively long response time to introduce the test analytes into the micron-sized holes of the PCFs.

Hollow-core photonic bandgap Bragg fibers (or simply hollow-core Bragg fibers) can also be used for sensing the analytes refractive indices. In the cross section of such fibers, a hollow core is surrounded by a periodic sequence of micron-sized layers of different materials in the cladding. This configuration avoids the problems of selective filling, at the same time, achieves almost complete modal overlap with the test analytes. Due to the possibility of having a relatively large core size (~1mm), the response time of introducing the test analyte into the fiber core could be shortened to ~1s, thus allowing for continuous on-line monitoring of liquid samples in a contained, highly integrated manner. In [17], K. J. Rowland et al. reported a hollow-core high-refractive-index-contrast Bragg fiber sensor with a sensitivity ~330nm/RIU. More recently, H. Qu, et al reported a hollow-core low-refractive-index-contrast Bragg fiber sensor operating on a spectral modality [18]. The authors showed that such Bragg fiber sensors offer superior performance in detection of changes in the real part of the analyte refractive index by monitoring the spectral shifts in the transmission spectra. The sensitivity of this sensor, compared to that of the sensor reported in [17], is considerably improved (~1400nm/RIU). In fact, low-refractive-index-contrast Bragg fibers are most suitable for liquid-core sensors, while high-refractive-index-contrast Bragg fibers are most suitable for gas-core sensors [17].

In our previous work, we have proposed and experimentally demonstrated hollow-core photonic Bragg fibers for bulk refractometry of commercial liquids and surface sensing applications including in situ monitoring of surface dynamics, as well as thickness detection of bio-layers and powder analytes [19,20]. In such sensors, we typically rely on a spectral modality, which relates the spectral shift in the fiber transmission spectrum to the changes in the liquid refractive index, or surface layer thickness in the fiber core.

In order to address many practical industrial applications, which involve liquids with different spectral properties, one needs to develop Bragg fibers with their primary bandgap located at different wavelength regions. In this work, we, thereby, demonstrate the wavelength scalability of the Bragg fiber bandgap positions by sampling controlling the fiber diameter and the bilayer thickness in the reflector. Additionally, in order to resolve minute changes in the liquid refractive index or the thickness of analyte layer, one can resort to two approaches. The first one is to employ spectrometers with high spectral resolution. However, such spectrometers are relatively bulky and heavy, which limits their application in many industrial fields where compact configurations are required. The other approach is to further enhance the sensitivity of the fiber sensor. In this work, we propose to enhance the sensitivity by optimizing the Bragg fiber geometry. Both theoretical analysis and experimental demonstrations have been performed to verify the proposed methodology. By designing and fabricating Bragg fibers with optimized bilayer thickness contrast, we have significantly enhanced the sensitivity by more than 32%, and the highest spectral sensitivity achieved experimentally in this work is 1850nm/RIU, which, to the best knowledge of the authors, is the highest value for the Bragg fiber based refractive index sensors. Furthermore, we studied the influence of temperature on the sensor performance. Finally, we indicate the challenges and limitations of the proposed sensors, and suggest future research directions of this project.

2. Design of the hollow-core Photonic Bragg fiber sensor

As a low-refractive-index-contrast Bragg reflector, we use polystyrene (PS)/polymethyl-methacrylate (PMMA) multilayers, which have refractive index of 1.58 and 1.49 at 589nm [19], respectively. Such materials are chosen owing to their good thermo-mechanical compatibility, and they can be co-drawn into a multilayer structure in a heated oven without cracking or delamination. The schematic of hollow core Bragg fiber is shown in Fig. 1.

 

Fig. 1 A Bragg fiber featuring a large hollow core surrounded by a periodic sequence of high and low refractive index layers.

Download Full Size | PDF

Bragg fibers confine light in the hollow cores by photonic bandgap effect. When a broadband light is launched into the analyte-filled core of a Bragg fiber, only the light with frequencies within the fiber bandgap of Bragg reflector will be confined and guided in the fiber core, while the light with frequencies outside of the bandgap will irradiate out in the first several centimeters of the fiber. From the basic theory of the Bragg fibers [21], the resonant center wavelength $\lambda$ of the fiber fundamental bandgap can be written as:

As shown in Eq. (1), the transmission bandgap shows a blue shift as the refractive index of the fiber core increases. This demonstrates that the transmission property of the Bragg fiber can be modified by the refractive index of the fiber core, which constitutes the spectral based sensing modality of this Bragg fiber for the detection of analyte refractive index.

Additionally, we note that, the bandgap center is dependent on the geometry of the Bragg reflector, and the transmission bandgap (or transmission window) shifts towards longer wavelength region, as the thickness of the bilayer in the Bragg reflector increases. Therefore, by adjusting the thickness of the multilayers in the Bragg reflector, we can target specific bandgap positions. In fact, as we will see in Section 4, by simply controlling the fiber diameter during the fiber drawing process, we can target specific bilayer thickness and bandgap positions. In this work, we fabricate Bragg fibers with their primary bandgaps located in the visible range (600nm-750nm), because many aqueous solutions or commercial liquids are relatively transparent in this wavelength range.

In many practical industrial-sensing applications, sensitivity and detection accuracy is of particular importance in order to resolve minute changes in the liquid refractive index or the thickness of analyte layer. Therefore, we analyze the factors that could influence the sensitivity of a Bragg fiber sensor using the spectral-based detection modality. We, therefore, derive the analytical expression of the spectral sensitivity of a Bragg fiber sensor, and thus, we have:

From Eq. (2), one finds that the spectral sensitivity of a Bragg fiber sensor is a function of the real part of the refractive index of the liquid analyte filling the fiber core, as well as the refractive indices and thicknesses of the individual layers in the Bragg reflector.

Apparently, the spectral sensitivity of a Bragg fiber sensor increases with the refractive index of the analyte filling the fiber core. Consequently, the shift of the resonant wavelength should have a polynomial dependence on the analyte refractive index. However, since our Bragg fiber sensors operate within a relatively small dynamic range ($n_c$: 1.33-1.36), the spectral shift of the resonant wavelength is considered to be virtually linear to increment of the refractive index of the fiber core.

Besides, Eq. (2) indicates that the closer the value of refractive index of the fiber core to those of the individual layers in the reflector, the more sensitive the sensor will be. As a result, low-refractive-index-contrast Bragg fibers are generally more sensitive than their high-refractive-index-contrast counterparts in sensing of the liquid analytes refractive indices.

Additionally, Eq. (2) suggests that the spectral sensitivity of the fiber sensor increases with the thicknesses of the high- and low-refractive index layers. As a result, Bragg fibers with thicker bilayers in the reflector generally have higher spectral sensitivity. However, it is important to realize that simultaneously increasing the thicknesses of both the high- and low-refractive index layer would also shift the bandgap position of the Bragg fiber towards longer wavelengths, according to Eq. (1). Currently, we routinely fabricate Bragg fibers with their primary bandgaps located in the visible range, because aqueous solutions are relatively transparent in this range.

In fact, even when the bandgap positions of the Bragg fibers are fixed, one can still possibly enhance the spectral sensitivity by optimizing the bilayer thickness contrast in the reflector, which, in this work, is define as $d_h/d_l$. To verify this theoretically, we investigate the dependence of the fiber spectral sensitivity on the bilayer thickness contrast based on Eq. (2). In order to keep the bandgap center of the Bragg fiber constant (e.g. 680nm), when we change the thickness of the high-RI layer, the thickness of the low-RI layer is also changed simultaneously according to Eq. (1), and the bilayer layer thickness contrast is, thus, determined. In Fig. 2, we plot the spectral sensitivity of the Bragg fiber sensor as a function of the bilayer thickness contrast. Our simulation results suggest that by using a Bragg reflector with smaller bilayer thickness contrast, higher spectral sensitivity could be achieved. As a matter of fact, this observation can be easily rationalized by noting that coefficient of $d_l$ in Eq. (2) is somewhat larger than that of $d_h$ because $n_l$ is closer to $n_c$ than $n_h$.

 

Fig. 2 Dependence of the spectral sensitivity of the Bragg fiber on the bilayer thickness contrast in the Bragg reflector.

Download Full Size | PDF

3. Fabrication of the hollow-core photonic Bragg fibers

In order to fabricate hollow-core Bragg fibers, we use a stack-and-draw technique [22]. Specifically, we first fabricate a preform by co-rolling commercial PS/PMMA films (each film has a thickness of 50μm). The preform is hosted with a PMMA tube in order to enhance their mechanical stability. The preform is then consolidated in oven at 80°C under vacuum for one week. Then, hollow-core Bragg fibers can be manufactured by heating and drawing of the preform in a drawing tower. Fiber geometry can be controlled during the drawing process, by adjusting the parameters such as temperature, drawing speed, and preform feed speeds, as well as the air pressure of the hollow core. In our experiment, we find that by simply controlling the fiber diameter, we can control the bilayer thicknesses, thus targeting specific bandgap positions. Additionally, by controlling the thicknesses of PMMA and PS layer during the preform fabrication, we can target specific bilayer thickness contrast in the Bragg fiber, and thereby achieve optimized sensitivity. The Bragg fibers used in this work feature relatively a large hollow core with a diameter (from 800μm and 1000μm) surrounded by periodic sequence of PS/PMMA Bragg reflector.

In order to verify the spectral scalability of the Bragg fiber Bandgap position, we fabricate Bragg fibers with different diameters in the same batch. We note that the bilayer thickness contrast for each sample in this section is almost the same (~1). However, the thickness of the bilayer for different fiber sample is different, which is relatively easy to rationalize because different bilayer thickness corresponds to different bandgap positions.

In order to verify the sensitivity enhancement of the Bragg fibers, we also fabricate Bragg fibers with different bilayer thickness contrast. All the fibers are specially designed to have the bandgap position (filled with water) at around 680nm. As a demonstration, we fabricate two fiber samples, named fiber 1 and fiber 2. The cross section of the Bragg fiber reflectors under scanning electron microscopy (SEM) is shown in Fig. 3. Bragg fiber 1 has a bilayer thickness contrast of ~1 ($d_h$ = $d_l$ = 225nm), in comparison, Bragg fiber 2 has an average bilayer thickness of ~0.3 ($d_h$ = 110nm, $d_l$ = 365nm), which is expected to have an enhanced sensitivity. In both cases, the Bragg reflector geometry is specifically designed, in order to ensure they have the same or very close bandgap positions. The two fibers are ready for subsequent experimental characterizations of their sensitivities.

 

Fig. 3 Cross section of the Bragg reflector taken by a scanning electron microscope (SEM), which features alternating polystyrene (PS) /poly-methacrylate (PMMA) layers. (a) Bragg fiber sample 1 with a bilayer thickness contrast of ~1. (b) Bragg fiber sample 2 with an average bilayer thickness contrast of ~0.3.

Download Full Size | PDF

4. Experimental demonstration of sensitivity enhancement of the Bragg fiber

4.1 Experimental setup

The schematic of the experimental setup for the characterization of the liquid-core Bragg fiber sensors is shown in Fig. 4. To integrate the hollow-core Bragg fiber (length: ~15cm) in the setup, we use two opto-fluidic blocks, which enables both optical coupling and the flow of the target analytes. Specifically, the Bragg fiber tip is sealed hermetically in the horizontal channels of the opto-blocks with tread seal tape (or polytetrafluoroethylene tape). A thin glass window is attached at the other end of the horizontal channel to facilitate optical coupling into the fiber core. Each block also features a vertical channel, which is connected to the horizontal channel, thus allowing for a continuous fluidic injection of liquid analyte into the Bragg fiber. A syringe is used to pump the liquid analytes into the Bragg fiber core. During the experiments, both ends of the fluidic channels are submerged into a water-filled beaker in order to avoid the existence of air bubbles in the sensing system, which would suppress the fiber transmission. After pumping the liquid into the hollow-core Bragg fiber, a broadband supercontinuum beam is launched into one end of the liquid-core Bragg fiber using an objective, and the output spectrum of the fiber sensor is registered by a grating monochromator. The Bragg fiber used in the sensor has a core diameter of sub-mm. Such a large core results in a short response time of the sensor, since the flow resistance decreases rapidly as the core radius increases.

 

Fig. 4 Experimental setup for characterizing the liquid-core Bragg fiber sensors.

Download Full Size | PDF

4.2 Scalability of the Bragg fiber bandgap position

In order to verify the scalability of the Bragg fiber bandgap position during the fiber drawing process, we simply control the bilayer thickness via adjusting the fiber diameter. During the fiber drawing process, the fiber diameter is monitored using a laser micrometer. In Fig. 5, we present the bandgap scalability of the hollow-core Bragg fibers (filled with water, refractive index: 1.33), which are manufactured in the same batch. The only difference between them is the fiber outer diameter (i.e., 800μm, 860μm and 950μm). As shown in Fig. 5, with the increase of the fiber outer diameter, the bandgap position of the Bragg fiber shifts toward longer wavelengths. Therefore, by changing the periodicity, or equivalently the overall scale to which the fiber is drawn, we can target the bandgap positions at selectable wavelength regions. We note that the spectral shape (or linewidth) of the Bragg fiber spectra varies from each other. This is probably due to the following reasons. First, the multilayers in the Bragg reflector are not perfectly uniform, which may lead to broadening in the fiber transmission band. Second, the length of the fiber samples is not identical, the optical confinement of each Bragg fiber sample is different. Third, the optical coupling condition might be different when we characterize their transmission spectra, and many high-order modes are excited and propagated. These high order modes tend to have different bandgap positions. Nevertheless, we clearly observe the scalability of the fiber transmission window by simply controlling the fiber diameter. This is of particular importance in practical industrial applications, because the commercial liquids used in different industrial fields typically have different transmission windows and spectral properties. By designing and fabricating Bragg fibers with their bandgap positions located in different spectral regions, one can target specific industrial applications.

 

Fig. 5 Scalability of the bandgap positions of the water-filled Bragg fibers by controlling the outer diameter.

Download Full Size | PDF

4.3 Sensitivity optimization of the hollow-core photonic Bragg fiber sensor

In order to characterize the sensitivity of the two Bragg fibers with different bilayer thickness contrast as fabricated in Section 3, we register the transmission spectra of the fibers when they are filled with water ($n_c$ = 1.33) and salt water ($n_c$ = 1.35), respectively. As shown in Fig. 6, both fibers have almost identical bandgap positions (centered at around 680nm when $n_c$ = 1.33). Moreover, we find that when the fiber core refractive index is changed from 1.33 to 1.35, the spectrum of Bragg fiber 1 with bilayer thickness contrast of ~1 features a blue shift of 28nm, while the spectral shift in the spectrum of the fiber 2 has been significantly increased to 37nm. The projected spectral sensitivities for fiber 1 and fiber 2 are 1400nm/RIU and 1850nm/RIU, respectively. We note that, to the best of our knowledge, the spectral sensitivity of 1850nm/RIU is the highest value achieved experimentally for refractive index sensors based on Bragg fibers, which is also more than 32% enhancement compared to the values reported in previously publications [18,19].

 

Fig. 6 (a) Experimental characterization of the spectral sensitivity of a Bragg fiber with bilayer thickness contrast of ~1. (b) Experimental characterization of the spectral sensitivity of a Bragg fiber with a bilayer thickness contrast of ~0.3.

Download Full Size | PDF

Therefore, by optimizing the geometry parameters in the Bragg reflector, one can enhance the spectral sensitivity of the Bragg fiber sensor, which, in turn, improves its detection accuracy for refractive index measurement and concentration prediction of commercial liquids. This is of particular importance in many industrial sensing applications. In our future work, we will further study and optimize the factors that could influence the fiber sensor sensitivity, such as bilayer thickness, refractive index contrast, Bragg reflector geometry in the fiber cross section, as well as core diameter, and we will release the findings in our future publications.

5. Temperature stability of the Bragg fiber sensor

The operation of most refractive index sensors can be influenced by the temperature variations because of the thermo-optic effect of the fiber materials [26]. In this section, we discuss how the thermal variations would affect the performance of the demonstrated Bragg fiber sensors. We note that the glass transition temperature of PS and PMMA is in the range of 75°C-110°C [27,28], as a result, our Bragg fiber sensor cannot operate beyond this range. Within the glass transition temperature, an increase in the ambient temperature leads to expansion of the multilayers in the Bragg reflector and the refractive index changes of the fiber materials. Both of them would result in shift in the transmission peak (i.e., bandgap position) of the Bragg fiber. In order to estimate the effect of the temperature changes on the fiber transmission spectrum, we, therefore, calculate the normalized transmission spectra of the fundamental HE₁₁ mode in a 10cm long Bragg fiber at different temperatures. The temperature-dependent refractive indices of PMMA and PS are retrieved in [27,28]. The linear thermal expansion coefficient for PMMA and PS are 7 × 10⁻⁵m/m·°C and 9 × 10⁻⁵ m/m·°C, respectively [29]. The fiber core is considered to be water-filled, which has a temperature dependent refractive index of −1 × 10⁻⁴/°C [30]. The thickness of each individual layer in the Bragg reflector is extracted from the SEM graph of Bragg fiber (Fig. 3(a)). The calculated normalized transmission spectra of the Bragg fiber operating in different ambient temperatures are shown in Fig. 7(a). The bandgap center position shifts towards longer wavelength with the increase of the operating temperature. In Fig. 7(b), we plot the bandgap center positions as a function of the ambient temperatures, and a linear dependence is found. The temperature response of our Bragg fiber sensor with aqueous solutions in the fiber core is only 45pm/°C, which is comparable or superior to those to those recently reported fiber-based refractive index sensors summarized in Table 1.

 

Fig. 7 (a) Simulated transmission spectra of a water-filled Bragg fiber sensor at different temperatures. (b) Spectral positions of the transmission peak at various temperatures.

Download Full Size | PDF

Table 1. Temperature response of some of the recently reported fiber-based refractive index sensors

View Table

6. Limitations of the liquid-core Bragg fiber sensors and future directions

While our Bragg fiber sensors offer many advantages such as high sensitivity, fast response, and low cost, they also have some limitations. First, since the Bragg fiber sensors use all-polymer structure, they are less resistant to aggressive chemicals and high temperature than silica fibers. However, we note that this does not affect their primary applications for fast refractometry of various commercial oils and other aqueous solutions. Moreover, since biomolecules can be directly bound to the PMMA surface without any further chemical functionalization [23], the presented Bragg fiber sensors could also be conveniently used as a promising platform for a wide range of bio/chemical sensing applications.

Another drawback of the Bragg fiber sensors is related to their relatively broad transmission bands, compared to sensors based on Bragg gratings. This may lead to difficulty in determination of minute spectral shift caused by small changes in the analyte refractive indices or the coated biolayer thicknesses. To address this problem, we would like to explain why the Bragg fibers have relatively broad transmission windows. First, we find that the multilayers in the Bragg reflector fabricated by the commercial fiber-drawing technique are not perfectly uniform. The variations in the bilayer thicknesses may lead to broadening in the fiber transmission band. Second, the Bragg fibers used in this work are only several centimeters long; as a result, the transmission spectra are relatively broad due to reduced fiber attenuation at the bandgap edges because of short fiber lengths. Third, the presented Bragg fiber sensors operate in a multimode regime, and many high-order modes with different bandgap positions have also been excited and propagated.

Despite the relatively broad transmission windows of the Bragg fibers, the spectral features (peaks or dips) in their transmission spectra can still be easily differentiated under our experimental conditions. Clear shifts of spectral peaks can also be observed in response to changes in the refractive index of the fiber core or the biolayer thickness, even when the Bragg fiber sensor is greatly squeezed [20].

In order to achieve narrower transmission bands of the Bragg fibers, one could pursue the following approaches in future work. To begin with, one could use longer fibers with more precisely controlled bilayer thicknesses and increase the number of bilayers in the Bragg reflector, thus enabling a stronger confinement of the optical modes and achieving narrower bandgaps; alternatively, one could also use fibers operating in a single mode regime or few modes regime. To do this, one could use Bragg fibers with smaller diameters to eliminate the influence of higher order modes, or coil fibers into circles with diameter of several centimeters to strip out the high-order modes. Moreover, one could achieve a single TE01 mode operation in the large-core Bragg fibers with the modal filter effect [24,25]. Finally, another attractive approach would be modifying the broad transmission window with resonant dip with narrow linewidth via the introduction of a defect in the Bragg reflector, as we have demonstrated in our previous publications [12].

7. Practical applications of the liquid-core Bragg fibers

Due to the advantages of the liquid-core Bragg fibers such as high sensitivity, short response time, simplicity in structure, small footprint, and the possibility of cost-effective mass production, we believe that the proposed Bragg fiber sensors could be a very attractive platform for a variety of scientific and industrial sensing applications.

We now propose several potential applications of the liquid-core Bragg fiber sensors. Firstly, the Bragg fiber sensor can be used for online monitoring of the concentrations of many industrial fluids, such as heat transfer fluids, sawing fluids, and other industrial dilutions. This is of significant importance in various industrial processes. Second, the Bragg fiber sensor can be used in food industry and medicine. For example, the refractive indices of glucose solutions (or syrup) are monotonically dependent on the glucose concentration. Considering the resolution of the demonstrated Bragg fiber sensor to be 7 × 10⁻⁵ RIU, this fiber sensor is able to resolve variations in the glucose concentration of 0.05% by weight. Thirdly, the sensor could also be used as a platform to detect the bio-layer thickness, study the surface dynamics and molecular interactions, as well as antigen-antibody conjugations, thus making the proposed all-plastic, hollow core Bragg fibers a promising component for the development of a new generation of the fiber-based biosensors. One advantage of using PMMA based structure is that many biomolecules can be directly attached to the surface without any chemical functionalization, which means that the fabricated Bragg fiber sensor could be conveniently used as a promising platform for a wide range of biochemical sensing applications. The demonstrated “one fiber” solution for both bulk and surface sensing applications will open up important commercialization opportunities for sensor-system instrumentation. Finally, the demonstrated Bragg fiber sensors can be simply used as absorption-based sensors for bio-chemical detection. Compared to those absorption-based sensors which frequently use “leaky modes” of fiber capillaries, the Bragg fibers in this work offer much lower propagation loss, thus enabling longer sensing length and, therefore, higher sensitivity.

8. Summary

In summary, in this work, we experimentally verify the spectral scalability of Bragg fibers by simply controlling the diameters during the fiber drawing process. In order to enhance the sensitivity and the detection accuracy of the fiber sensor, we first study the factors that influence the spectral sensitivity of the Bragg fiber sensor, and we propose to enhance the sensitivity by optimizing the Bragg fiber geometry. Both theoretical analysis and experimental demonstrations have been performed to verify and characterize the proposed methodology. The highest spectral sensitivity achieved experimentally in this work is 1850nm/RIU, which, to the best knowledge of the authors, is the highest value that for the Bragg fiber based refractometers. Such sensitivity is enhanced by more than 32% compared to the experimental values reported before. The presented fiber sensor can inherently integrate optical detection with microfluidics, thus allowing for online monitoring of the refractive index/concentration of many industrial fluids, trace amount of biomolecules, real-time detection of binding and affinity, study of kinetics, with enhanced accuracy. Additionally, we studied the influence of the operating temperature on the performance of the fiber sensor. Finally, we indicate the challenges and limitations of the proposed sensors, and suggest future research directions of this project. The Bragg fiber sensors operates on a spectral-based detection modality, which is relatively immune to negative influences such as intensity fluctuations of the optical source, deviations in the mechanical/optical alignment when loading and unloading the analytes, bending of the fiber, and existence of micro-particles or air bubbles in the system. Therefore, such sensors offer higher accuracy and repeatability in industrial sensing applications. The presented fiber sensor can inherently integrate optical detection with microfluidics, thus allowing for online monitoring of the refractive index/concentration of many industrial fluids, trace amount of biomolecules, real-time detection of binding and affinity, study of kinetics, with enhanced accuracy.