Polymer-fiber grating sensors
For more than 15 years, research groups around the world have been exploiting the innate photosensitivity of the core material of many optical fibers to fabricate Bragg grating sensors. These devices comprise a submicron, spatially periodic modulation of the core refractive index, typically produced by exposing the fiber to the interference pattern produced by two intersecting beams of UV light (see Figure 1). An important property of such devices is that they reflect back only light of a specific wavelength, determined by the spacing of the index modulation in the fiber. Because the grating period and refractive index, and hence the reflected wavelength, are changed when the fiber is strained, the gratings can be used as strain sensors.1 This technology has matured to the point where many companies now sell complete sensing systems to the oil, gas, and wind energy industries, for example.
To date, activity on Bragg grating sensors has focused on silica fiber. Now, however, we at Aston University and other research centers are working to transfer the technology to polymer optical fiber (POF) to take advantage of some of the very different material properties of polymers. One of the features of POF is that it is much more elastic than silica, with a Young's modulus about 25 times smaller. This becomes important when using the fiber strain sensors to monitor materials that are themselves very compliant and where a stiff silica fiber may act to reinforce the material, thereby returning an artificially low strain value. As an example, in one project with Southampton University (UK) we have been investigating the potential of optical fibers to monitor strain in tapestries as an aid to their conservation.2 Figure 2 shows results from digital image correlation (DIC) on a strained fabric monitored with POF and silica gratings. The POF sensor, fixed using the flexible conservation adhesive dimethyl cyclosiloxane, clearly distorts the strain field much less than the other devices, including a second POF sensor fixed with much stiffer Araldite adhesive.
A second, rather graphic, example of this behavior comes from a European project—PHOSFOS (PHOtonic Skins For Optical Sensors)3—where we are seeking to develop a variety of flexible skins with embedded fiber sensors. Figure 3 compares one very elastic skin instrumented with POF and silica fiber Bragg gratings (FBGs). The stiff silica fiber is unable to recover the strain applied to the skin, and it displays considerable hysteresis.
Some polymers, notably poly(methyl methacrylate) which is commonly used for POF, possess an affinity for water, which causes the material to swell and raises the refractive index. Both reactions lead to an increase in the wavelength reflected from a Bragg grating in the fiber: see Figure 4(a). For some applications, this sensitivity would be problematic (although there are other polymers available for fiber manufacture that do not suffer in this way). However, we are exploiting this phenomenon to detect small quantities of water in aviation fuel. The water content must be kept to a minimum or, in the worst case, it may freeze and lead to engine failure, as appears to have happened to the British Airways Boeing 777 that crashed at Heathrow Airport on 17 January 2008.4 Figure 4(b) shows the significant changes in the reflected wavelength that result when one of our sensors is swapped between fuels with differing water contents.
One final area of current research concerns the realization of a low-cost POF FBG sensing system. Much of the expense of current Bragg grating sensor systems is linked to the fact that they require picometer-resolution wavelength measurement and employ single-mode fiber, and consequently must use costly single-transverse-mode optical sources. We are currently developing Bragg grating sensors based on 50μm-core multimode POF. As shown in Figure 5, the gratings are typically several nanometers wide, as opposed to FBGs in single-mode fiber, which are an order of magnitude narrower. The increased reflection bandwidth reduces the precision with which the central wavelength can be determined and, hence, the strain resolution achievable from such a sensor. However, we believe this will not be a significant drawback in applications that make use of the very large strain-sensing ranges achievable with POF, which can be in excess of 10%.5
In conclusion, polymer-FBG sensors represent a recent development that seeks to build on the unique—and controllable—material and chemical properties of polymers. The technology has just matured to the point where we are looking at applications outside of the optical laboratory.
This work has been supported by the EU (PHOSFOS) and the UK Engineering and Physical Sciences Research Council (EPSRC). Ian Johnson was awarded an EPSRC studentship. The authors acknowledge contributions from their colleagues Chi Zhang and Xianfeng Chen at Aston University, Janice Barton at Southampton University, and Bram van Hoe and Geert van Steenberge of Ghent University, Belgium.
Ian Johnson works in the photonics research group. His particular interests concentrate on polymer optical fiber, taking the technology from the laboratory to the real world. He has an engineering background, with a BEng(Hons) in electronic engineering and a DisplayMasters MSc from the University of Dundee. He will be presenting an invited paper to SPIE Photonics Europe on multiplexed FBG sensors in microstructured POFs.
