In this experiment we used commercially available Fibercore PS 1250/1500 photosensitive fiber, where refractive indices of core and cladding defined at λ = 1,550 nm were 1.44937 and 1.4402, respectively. Gratings based on this fiber show very high pressure sensitivity [6]. Each of our samples contains a 10-cm section of the photosensitive fiber with the ends spliced to Corning SMF28 fibers. A set of LPGs was written with a computer-assisted precision arc-discharge apparatus, described in [4] and [17]. The LPGs were written only in the PS 1250/1500 fiber section. The discharge current was adjusted to be low enough to heat the fiber locally and not to produce any visible tapers due to the axial tension applied to the fiber. The arc discharge time was established at τ = 300 ms. The grating period was Λ = 375 μm and the length of the gratings was L = 20–25 mm. The optical transmission of the fiber in the range of λ = 1,160–1,660 nm was monitored during the LPG fabrication process in order to obtain the desired spectral attenuation notches. We used an Agilent 83437A broadband light source and an Agilent 86142B optical spectrum analyzer for this purpose.

The films were deposited using the Plasmalab 80+ system (Oxford Plasma Technology) working at f = 13.56 MHz [18]. SiN_x films were deposited on the LPG samples suspended 2.5 mm over the RF electrode, and simultaneously on wet oxidized silicon wafers to be used as reference samples which were placed directly on the electrode. The deposition procedure was similar to the one previously reported [19,20]. We employed a high [SiH₄:N₂]/[NH₃] flow ratio (2% SiH₄ diluted in N₂) equal to 285/15 in order to obtain high-n SiN_x films. As for the other parameters for deposition, the RF power was 15 W, the pressure in the chamber was 0.53 mbar, the deposition time was from 3 to 8 minutes and the electrode temperature was 325 °C.

Film parameters such as the n, the extinction coefficient (k) and the thickness were determined by a Horiba Jobin-Yvon UVISEL spectroscopic ellipsometer on the silicon reference samples using a procedure described elsewhere [21]. Thickness of the SiO₂ film deposited on the silicon wafers was measured to be in the range of 385 nm to 400 nm, which was thick enough to behave like the fused silica cladding of the fiber.

For pressure measurements, the LPGs were installed inside a steel housing in a transmission configuration [6]. To keep the fiber at a constant axial tension under different pressures, a fiber loop was formed inside the small inner channel of the housing, which was filled with Exxon instrument oil (n_d = 1.4793) and connected to a hydrostatic pressure standard DWT-35, capable of generating and calibrating pressures up to 100 MPa (accuracy of at least 0.1%). The pressure measurements were performed at room temperature (T = 23 °C). RI (n_d) measurements of the liquids were performed using a VEE GEE PDX-95 refractometer working with an accuracy of ±10⁻⁴ refractive index units (RIU).

The effect of the deposition of two SiN_x films with different thicknesses and thus different n on the spectra of the arc-induced LPGs can be seen in Figure 2. The deposition times were 3 and 8 minutes. The properties of the films, as determined using reference samples, were: film thickness, 28.5 and 74 nm; and n, 2.293 and 2.366 (both at λ = 1,560 nm). The phenomenon of a lower n for SiN_x films thinner than 50 nm has been discussed elsewhere [25]. The k for both films were extremely low (close to 0) in the infrared spectral range.

Both spectra shown in Figure 2 experience a shift induced by deposition of the high-n nanocoatings, with a larger shift observed for the thicker and higher-n coating [Figure 2(b)]. Increasing the thickness or n of the coating can shift the maximum RI sensitivity to its lower values, while decreasing the sensitivity to the higher external RI [26,27]. We successfully shifted the maximum sensitivity of the LPGs from an external RI of 1.458 (for a grating with no film) to one in the range between 1 and 1.33. For both samples surrounded by any of the applied liquids, the coating is thick enough or has a high enough n to induce a transition of each of the cladding modes to lower-order ones. In this process, one of the modes begins to propagate in the coating, thus inducing the shift of other cladding modes to lower-order modes [28]. It can be seen from Figure 2(a) that for the LPGs with a thinner coating (28.5 nm), the high-order cladding modes are hardly guided when water surrounds the structure. When the coating is thicker (74 nm), the spectrum recovers well to below −10 dB for all of the applied liquids [Figure 2(b)]. With oil or glycerine surrounding the LPG, the spectrum shows a similar resonant wavelength to that of the uncoated sample surrounded by air. The results suggest that for these conditions the RI sensitivity is low, compared to that of the LPG with no coating when surrounded by the same oil or glycerine (Figure 1). However, due to the high extinction-coefficient of oil, the resonance is not as deep as for glycerine which has a lower extinction-coefficient and for which observed resonances are deeper for both investigated LPGs.

It must be mentioned here that there are several other nanocoating deposition techniques already applied to grow coatings on LPGs. However, the commonly applied deposition techniques show a series of disadvantages: they are time-consuming (e.g., electrostatic self-assembly), they are poorly controllable in terms of film thickness (e.g., sol-gel deposition) and in some cases the deposited films show low durability (e.g., Langmuir-Blodgett deposition). In order to maintain their long-term stability for such demanding applications as pressure sensors, the nanofilms must be very dense, a feature which is correlated to their high n and to their high optical quality. A higher n in the coating implies higher variations of the effective refractive index of the cladding modes as a function of the external RI and a higher shift in the attenuation bands [28,29]. That is why for the applications discussed here, the high-n coatings can be thinner and still have a comparable effect to thicker coatings with lower n. Moreover, it is known [16] that for some of the nanodeposition techniques mentioned above, the films suffer from a significant extinction-coefficient, which can cause the attenuation bands in the transmission spectrum to vanish. Conversely, plasma-deposited high-n SiN_x films show a very low k in the infrared spectral range [14] and seem to be a perfect choice for LPG-based pressure sensors capable of operating even in harsh environments.

As reported in our earlier paper [6], the resonances of LPGs fabricated in a boron co-doped fiber with a period of Λ = 375 μm experience a red shift. They reach linear sensitivities of 78, 56 and 45 pm/bar under the influence of pressure in a range of up to 240 bar, for the LP₀₇, LP₀₆ and LP₀₅ cladding modes respectively. This is the highest pressure sensitivity reported so far for LPGs. The pressure sensitivity of the LPGs fabricated in the PS 1250/1500 optical fiber can be discussed in terms of the different elastic properties of the silica cladding and the B₂O₃/GeO₂-containing silica glass core of the fiber [6]. The addition of B₂O₃ to SiO₂ decreases the number of network bonds, and also decreases the bulk modulus (K) of the glass [30], i.e., the resistance of the material to a uniform compression. The pressure-optic coefficient is dependent mainly on isothermal compressibility, that is the inverse of K [31]. The incorporation of B₂O₃ into fused silica in turn increases its pressure-optic coefficient, so that the core material has a higher pressure-optic coefficient than the cladding. When the influence of pressure on the period of the grating (Λ) is disregarded [6], Equation (1) shows a relation between the resonance wavelength ( λ res m) of an m^th cladding mode, the effective refractive index of the propagating core mode ( n eff 01) and the effective refractive index of the m^th cladding mode ( n eff 0 m) induced by pressure (P) changes. The fact that the pressure-optic coefficient of the core is higher than that of the cladding explains the positive pressure sensitivity of LPGs written in the B₂O₃-containing fiber. (1) ∂ λ res m ∂ P = Λ ( ∂ n eff 01 ∂ P − ∂ n eff 0 m ∂ P )

The response of the nanocoated LPGs to pressure changes is shown in Figure 3. Due to the thin (28.5 nm) high-n SiN_x coating, the resonances are clearly visible in the output spectrum of the grating, when the structure is operating immersed in a high-RI oil. The shift increases with the order of the cladding modes, just as was observed for the uncoated structure operating in water [6].

When we compare the pressure sensitivity of both uncoated and coated LPGs, where the measurements were performed in water and oil, respectively, it can be seen that the results are in good agreement (Figure 4). A small difference for the LP₀₅ and LP₀₇ modes is most probably caused by a slight lack of repeatability of the grating fabrication process. It has been shown that even a very small tapering of the fiber during the arc-induced fabrication process can have a significant influence on the spectrum of the LPGs [17]. It is known that SiN_x can have a bulk modulus K more than five times higher than fused silica [32]. Following the relation between the pressure-optic coefficient and the bulk modulus of the material mentioned above, the pressure-induced change in the n of the SiN_x nanocoating is much smaller than the change experienced by either the cladding or the core materials. On the other hand, the nanometric thickness of the film makes it elastic and simultaneously does not disturb the pressure effect experienced by both core and cladding.