Fiber Gets All Wet
Conduct a poll anywhere in the developed world and people will probably associate optical fiber with telecommunications, medicine, and decorative lighting. Despite a history nearly as long as that of optical communications, fiber-optic sensing is unlikely to get so much as a mention, but perhaps things are about to change. Telecom may have absorbed most of the resources and talent during the boom, but the subsequent downturn released a flood of specialist individuals, components, and ancillary technologies that are now breathing new life into fiber-optic sensing. Technologies such as fiber-optic gyroscopes (FOGs) and fiber versions of the underwater microphones known as hydrophones are now coming to the forefront.
Fiber-optic sensors can be classed as intrinsic and extrinsic. In extrinsic sensors, the fiber simply conducts light from a sensing head to a detector; the interaction between light and the environment takes place outside of the optical fiber. The intrinsic sensors that are the focus of this article rely on the interaction between the environment, the fiber, and the light itself to generate information about a specific measurand. Temperature, pressure, electromagnetism, rotation, vibration, stress, and strain all interact with an optical fiber to create subtle changes in its transmission characteristics. This property has led many researchers to despair that the challenge is not so much making a fiber-optic sensor as working out exactly what it is sensing.
The key advantage of intrinsic fiber-optic sensors is the fundamental ability of a fiber to guide light around bends and over large distances, which enables long optical path lengths to be confined within small physical volumes (i.e., by coiling the fiber) and magnifies the effects of subtle environmental changes to a level at which they may be measured accurately and quantified.The Proof Is in the Fiber
Of all intrinsic fiber sensors, the FOG has enjoyed the greatest commercial success to date. The FOG is based on the Sagnac Effect, which is the differential phase shift induced by rotating an optical system around which light travels both clockwise and counterclockwise, simultaneously (see figure 1).
In a FOG, light from a broadband source, typically an 830-nm superluminescent diode or a 1550-nm erbium-doped fiber-amplified spontaneous emission source, is divided into two waves of equal intensity. One of these waves passes through the coil of polarization-maintaining fiber in a clockwise direction, the other in a counterclockwise direction. After propagation through the coil, the two waves recombine and interfere. If there is no rotation of the fiber coil, the two lightwaves remain in phase, recombining to reconstruct the original waveform. Rotating the coil introduces the aforementioned phase shift, because the effective path length of the lightwave propagating in the direction of rotation is greater than that of the counter-propagating wave. The resulting interference generates variations in light intensity that are detected by the photodiode and converted into an electrical signal that is proportional to the rotation rate of the gyroscope.
The development of low-loss optical fibers in the 1970s enabled the development of FOGs; however, it was the introduction of specialty fibers in the early 1980s that quickened the pace of development. The sensing coils of typical FOGs vary from 20 to 60 mm in diameter and contain between 100 and 3000 m of fiber. Hydrophones tend to have much shorter coils, with a maximum of perhaps 100 m, but coil diameters may be as small as 10 mm. If standard telecommunications fiber were deployed in such coils two things would happen -- light launched into one end of the coil would be radiated, dissipated by bending loss before it reached the other end, and the fiber itself would be liable to fracture due to the phenomenon of static fatigue.
We can enhance the strength of guidance in an optical fiber by raising the refractive index of the core glass relative to that of the cladding through increased germania content. Increasing the core index of a telecom optical fiber by only 0.4% would make its transmission characteristics suitable for FOG use and provide effectively lossless transmission in a 20-mm-diameter coil; a 10-mm-diameter coil would require an increase of 1.7%.
Static fatigue is a phenomenon by which optical fibers can fracture spontaneously if subjected to bending stress or invariant tensile loads. The stress induces growth of intrinsic and microscopic flaws located on the fiber surface, causing fracture as soon as any flaw grows beyond the critical limit of the material. Fibers are made more resistant to static fatigue by increasing proof-test levels and/or reducing glass diameters.
Proof testing involves straining the fiber to destruction to screen out intrinsic flaws -- the higher the proof-test level, the smaller the size of intrinsic flaw that will remain in the surviving fiber and the lower the probability that the fiber will fail under static fatigue. The telecommunications industry standard for proof testing is 1% strain. Deploy such a fiber in a 20-mm-diameter FOG coil and it could fracture within months; deploy it in a 10-mm-hydrophone coil and failure could be immediate. For this reason, manufacturers generally proof test specialty fibers to 2% or 3% strain.
A more direct way to enhance lifetime is to limit bending stress levels by reducing the outside diameter of the fiber itself. A typical FOG or hydrophone fiber has a diameter of 80 µm -- a little more than two-thirds that of a standard telecommunications fiber (125 µm). When bent, the induced stress within these fibers is around 40% less than that of the larger fiber, slowing growth of intrinsic flaws and boosting lifetimes from mere months to 20 years or more.
The use of polarization-maintaining (PM) fibers in interferometric sensors greatly simplifies sensor design by avoiding the phenomenon of signal fade induced by polarization drift. When two light pulses recombine, as they do in a FOG, constructive interference is maximized when they have the same state of polarization. The PM fiber design most commonly used in fiber sensors is the so-called bow-tie fiber, developed by the University of Southampton (Southampton, UK) in 1982. In this design a pair of wedge-shaped (bow-shaped) stress-applying parts located at either side of the core induces birefringence that preserves the polarization state of the light guided within the fiber.Dipping an Ear in the Water
The founding principles of the fiber gyro can be traced back to the early 20th century, but the principles for underwater acoustic sensing can be traced back more than 400 years. Since that time, considerable advances have been made in the science of underwater acoustics and the hardware used for the detection of faint signals. There are two primary types of underwater sound systems. The first is a direct listening system, which passively detects sounds made by vessels and marine life passing near the hydrophone. The second type uses a sound source for echo ranging. These systems are called active detection systems, of which sonar is perhaps the best known example.
In an active detection system, the source makes a sound that propagates through the water and reflects from moving objects. Hydrophones detect the reflected signals, and signal processing performed on the signal extracts information about a variety of characteristics of the reflecting object. Typical sonar systems used throughout the world today rely on piezo-ceramic hydrophones and DC-powered signal conditioning and data telemetry. The hydrophones are configured in three basic formats: bottom-mounted lines or grids of hydrophones, hydrophones arranged in a pattern on a submerged surface, and long flexible hose-like hydrophone arrays (streamers) towed through the water.
All of these systems tend to be expensive and fairly delicate, particularly streamers, making more robust and/or economical technologies very attractive. Fiber-optic hydrophone arrays have the potential to fulfill both of these needs, with reduced diameter fibers accommodating tight bend diameters and innovative coating packages countering the effects of long periods in salt water. The physical size of the hydrophones themselves is of particular importance in towed applications because of fluid drag and turbulence/noise.
Over the past several years, the U.S. Navy has worked to develop fiber-optic sensor and array development programs for electrically passive underwater acoustic sensing. Fiber-optic sensor multiplexing topologies have been developed for applications such as towed arrays, hull arrays, and bottom-mounted arrays. In such electrically passive topologies, the lasers and receivers are located remotely from the sensor arrays and interrogate the sensors via fiber-optic links. The sensors themselves typically consist of compliant mandrels that are wrapped with specialty optical fiber (see figure 2). The mandrels respond to the acoustic signals and introduce a phase shift in the light traveling through the fiber. These signals are then transmitted over optical fibers back to the signal processing location.
Although world events have pointed out the desirability of intrusion detection for homeland defense applications, a variety of commercial applications exist for large arrays of multiplexed hydrophones. Two of the largest applications of hydrophone arrays are oceanic geophysical exploration and oil field monitoring. Historically, the commercial sector has used arrays of ceramic hydrophones towed by research ships for these applications; however, as fiber-optic sensing technology matures, it will provide an attractive alternative.
The lines towed by these research ships are each connected to a large in-line array of hydrophones. A ship can tow as many as 16 streamers simultaneously, each of which can contain more than 1000 separate hydrophone channels. In addition to the streamers of hydrophones, a geophysical exploration ship also tows an acoustic source. The source generates loud pulses of low-frequency sound energy. These pulses penetrate the sea floor, reflecting off of various sub-bottom layers, as well as oil reservoirs, for example. The reflected pulses are detected by the towed streamers and processed to construct a picture of the strata below the sea floor in the survey area. This exploration and mapping technique allows oil companies to be very precise in the location of wells, minimizing unnecessary drilling.
Fiber-optic sensing has been an active area of research and development for nearly 30 years. After years spent starved of both attention and investment, the fields of fiber sensors and fiber sensing are once more beginning to flourish. FOGs are now widely available commercially. Fiber-optic hydrophone arrays show good potential for commercial applications as well. Specific developments in specialty fiber design, performance, and qualification are enabling the fundamental benefits of ruggedness, reliability, and precision to be realized. oe
By Brian Rogers
Fiber-coupled spectrometers began a revolution in low-cost instrumentation. Many people do not realize that there are significant differences between fiber used for spectroscopy and that used for typical telecom applications. Yes, both transmit light and both have a higher refractive index in the core than the cladding, but that is where the similarities end. The fibers differ in size, materials (not only in the glass but in the strength-enhancing outer buffer layer), and the process by which they are made.
One of the most notable distinctions of spectroscopy-type fiber is that it is much larger than telecom fiber. Singlemode telecom fiber typically has a core diameter of 8 µm and an outer cladding diameter of 80 µm or 125 µm. Spectroscopy applications generally feature fibers with core diameters of 400 µm to 1000 µm, sizes that far exceed singlemode operation. This is obvious by looking at the normalized frequency parameter, or the V-number, for the fiber:
V = [2π(NA)a]/λ
where NA is numerical aperture, a is fiber radius in micrometers, and λ is wavelength in micrometers; a V-number below 2.405 represents singlemode fiber, for example.
Singlemode operation is not critical, and perhaps not even desirable, for most spectroscopy applications because spectroscopy generally focuses more in the frequency domain and is not ususally concerned with temporal dispersion. Note that the equation relates directly to the fiber NA, which is determined by the refractive index of the glass, but which is a general measure of the light-gathering ability of the fiber; we define the full acceptance angle of a fiber by 2arcsin(NA).
Standard singlemode fibers typically feature NAs of about 0.12, whereas those of spectroscopy fibers are typically 0.22, 0.39, and in some cases even higher. For most applications, the spectrometer should be designed for optimal performance with fiber of a specific NA, allowing designers to get more photons to the detector and increase the signal-to-noise ratio of the overall application; applications such as fluorescence and photo-luminescence, for example, produce very low signal levels.
Spectroscopy fibers consist of different glass types than telecom fibers. Typical singlemode fiber consists of a germanium-doped silica core surrounded by a cladding. In spectroscopy, we surround a pure silica core with a fluorine-doped cladding, or an outer strengthening buffer layer of polymer acts as a cladding to yield high-NA fibers. The presence of hydroxyl ions (OH) also plays an important role. Low-OH fiber performs best for applications in the visible and near-IR spectral regions (400 nm to 2200 nm), whereas high-OH fiber performs best for UV visible applications (200 nm to 1000 nm).
Fibers for spectroscopy typically have different buffer materials than singlemode fibers. The buffers of telecom-grade fibers typically consist of UV-cured acrylates. Although these buffers allow the fibers to be drawn very quickly (typically 1200 m per minute), they are limited in functional temperature ranges (up to 100°C) and chemical compatibility. Spectroscopy fibers feature heat-cured coatings such as polyimide, which can perform up to 300°C but drops the draw rates to the order of several meters per minute. Other, more robust coatings exist, such as Teflon and metals that can operate up to 600°C.
When specifying fiber for your next application, remember that not all optical fiber is created equal. It varies by NA, material and transmission wavelengths, and buffer type. Specify carefully.
Brian Rogers is manager of fiber optics at Ocean Optics, Dunedin, FL.
Aileen Sansone is a senior fiber-optic systems engineer at Chesapeake Sciences Corp., Millersville, MD.