IFFPI Structures Based on Air Holes
Besides introducing in-line FBGs or reflective films as mirrors, IFFPI sensors can be constructed by using an air hole as the reflective element. The Fresnel reflection between the fiber material and air offers a reflection of about 4%, which is sufficient to form a high-performance mirror for the IFFPI sensor based on two-beam interference.
The fabrication process of the micro-air-hole-based IFFPI sensors includes three steps: drilling a micro-hole at the fiber end face, fusion splicing the fabricated fiber with another cleaved fiber to form the air gap, and cleaving, as shown in Figure 2.2. The last cleaving step makes another Fresnel reflection for the two-beam interference for the IFFPI sensor.
There are several methods to make a micro-hole at the fiber end face, including chemical etching, laser micromachining, focused ion beam (FIB) milling, etc. All these methods have been used for
Figure 2.2 Fabrication process of the air-gap mirror for the IFFPI sensors. (a) Splicing etched fiber with a cleaved fiber, (b) the fabricated air-gap mirror.
fabricating a wide air gap to form an extrinsic IFFPI sensor, which will be introduced in Section 2.1.2. For the intrinsic IFFPI sensor, the air gap serves as one reflective mirror so that the width should be much narrower than that used in the EFFPI sensor. The most convenient and cost-effective way is by using wet chemical etching. It is also good for mass production. It was used in the 1990s for fabricating sharp fiber tapers at nanoscale for near-field scanning optical microscopy [23,24].
By using the hydrofluoric (HF) acid, the etching rate is faster for the Ge-doped silica fiber core, compared with that of the F-doped fiber cladding. Therefore, a micro-air-hole can be fabricated at the fiber end due to the etching rate difference. The etching rate depends on the concentration of HF acid. So a cover layer of nonvolatile reagent that is insoluble in water can be used to prevent the concentration changes due to volatilization of HF. Buffered HF acid, that is, a mixture of HF acid and ammonium fluoride (NH4F), was often used to control the etching rate. However, NH4F is not necessary, and only changing the concentration of HF acid is more than enough to control the etching rate. Note that higher concentration of HF acid leads to a rougher surface, along with a higher etching rate. Therefore, a better choice is to use low concentration of HF acid and long etching time to obtain a smooth-etched surface.
Tso and Pask investigated the influence of the concentration of HF acid on the etching rate and found that the reaction rate increases in a linear fashion when the concentration is lower than 10 M or higher than 25 M . The diffusion process influences the etching rate and agitation during the etching process would increase the etching rate. The dissolution rate, and thus the etching rate, also increases with an increasing temperature .
The second step is to fusion splice the etched fiber to a cleaved fiber. An air gap is formed during the splicing and the whole air gap acts as one of the reflective mirrors for the IFFPI sensor. Anbo Wang and his coworkers investigated the influence of the shape, including the width and the diameter, of the air gap on the transmittance characteristics . They suggested that the width of the air gap of d < 3 p,m and a diameter of more than 20 |!m are required for the highest coupling coefficient, as shown in Figure 2.3. They demonstrated two structures for the IFFPI sensors. The first is by using one air-gap mirror and one Fresnel reflection of a cleaved fiber. It is good for single-point
Figure 2.3 (a) Microscopic image of the air-gap mirror and (b) schematic of the IFFPI sensor based on air-gap mirrors.
measurement. The second is by using two air-gap mirrors. It is better for multi-points measurement via sensor multiplexing. Both temperature and strain measurement were carried out by these air-gap-based IFFPI sensors.
Fernando and his coworkers investigated the chemical etching of optical fibers in detail by using a 3D surface profiler, optical microscopy, and scanning electron microscopy . Although the authors claimed that the sensor was an intrinsic FFPI sensor, it is classified as a EFFPI sensor in this book, as the medium in the cavity is air rather than a section of fiber. However, the method described in their work is very helpful to fabricate the air-gap mirror for intrinsic FFPI sensors. The concentration of 48% HF acid was used for the etching. No NH4F was used. The etched fiber diameter, etch depth, and the cone angle of the etched hole were recorded versus etching time. A decrease of the fiber diameter in a linear fashion at about 3 p,m/min was observed. The depth of the etched hole increased linearly with a rate of 1 |lm/min. Their experiment was performed at a temperature of 20°C with a stable water bath. The roughness was around 21-44 nm with an etching time of 5-16 min, but increased up to 147 nm when etched for 22 min.
The FFPI sensor was then fabricated by fusion splicing two etched fibers. One advantage of splicing two etched fibers is that there is no difference between the diameter of the two etched fibers so that the automatic alignment during the splicing process is more stable, and moreover, the mechanical strength is better. In the case of splicing one etched fiber with a cleaved fiber, the fiber diameter is different and may cause weak mechanical properties. However, the fiber diameter is small with an etching time longer than 20 min and it may require precise alignment and careful choice of the fusion splicing parameters. Generally speaking, after chemical etching, the melting point of the fiber may decrease. Deformation and expansion may occur at the splicing part.