Teams Try New Approaches to Fiber Sensors

Breck Hitz

A pair of unrelated papers published recently in Optics Letters illustrates the diversity of interesting and promising new approaches to fiber optic sensors that scientists currently are investigating. In one, the principle of cavity ringdown spectroscopy (see sidebar) is applied in a fiber loop to create a novel pressure sensor. In the other, a sensitive Mach-Zehnder interferometer is created in a single photonic crystal fiber. Neither approach is beyond the proof-of-concept stage, but the work is indicative of the potential of new fiber techniques in sensing applications.

Figure 1. A pulse of light injected into the loop through the bottom coupler "rings down" as it travels around the loop.

Recently, a team at Mississippi State University in Starkville applied the cavity ringdown principle to a fiber optic loop (Figure 1). The researchers injected a laser pulse in-to the 99:1 coupler at the bottom of the loop and sampled a small portion of the pulse through another 99:1 coupler at the top of the loop, observing an exponentially decaying signal.
Several other scientists previously considered fiber ringdown loops for spectroscopic applications, but the Mississippi investigators have demonstrated that such a setup can be used as a pressure sensor. They stripped the jacket off a short length of the fiber and applied pressure of several hundred pounds per square inch to the bare section. Pressure changes the refractive index of glass, and this small change in the fiber's refractive index was observed as a change in the loop's ringdown time. Figure 2 shows the ringdown time as the pressure increases from zero to approximately 595 psi, indicating the repeatability of the measurement.

Figure 2. The loop's ringdown time decreases from approximately 3.94 µs when no pressure is applied to the fiber to 2.38 µs when 595 psi is applied. It returns to 3.94 µs when the pressure is removed.

The scientists are optimistic that the technique will engender a family of sensors for gauging temperature and strain, as well as pressure. They believe that these devices will have greater dynamic range, will be less expensive and will be capable of surviving more extreme environments than conventional Fabry-Perot and fiber-Bragg-grating fiber sensors.

Holey fiber interferometer

A different approach to fiber optic sensing involves fabricating a Mach-Zehnder interferometer in a length of photonic crystal fiber. In a conventional, free-space Mach-Zehnder interferometer (Figure 2, top), one beamsplitter divides the incoming light between the interferometer's two arms, and a second recombines the beams. The amount of phase shift in one arm relative to the other is revealed by the degree of constructive/destructive interference in the combined beam. Similarly, in a single-fiber Mach-Zehnder interferometer (Figure 3, bottom), a long-period grating divides the light between the two arms. One arm propagates in the cladding, and the other in the fiber core. A second long-period grating combines the arms.

Figure 3. A conventional, free-space Mach-Zehnder interferometer uses beamsplitters to separate and combine the interferometer's two arms (top). A single-fiber Mach-Zehnder interferometer uses a long-period grating for the same task (bottom).

Other laboratories have demonstrated single-fiber Mach-Zehnder interferometers in conventional single-mode fiber, but, recently, a collaboration at SungKyunKwan University in Suwon, and at Kwangju Institute of Science and Technology in Gwangju, both in South Korea, has demonstrated such a device in a photonic crystal fiber.

Photonic crystal fibers are drawing increasing interest for their unconventional properties. The optical characteristics of these fibers -- such as birefringence, dispersion and nonlinearity -- can be tailored by modifying their design, so a sensor based on a photonic crystal fiber could have unique and interesting capabilities. The scientists in Korea believe that their device will have promising applications in both sensing and communications.

Figure 4. Because photonic crystal fibers are pure silica and not photosensitive, researchers mechanically induced a long-period grating in the fiber by applying periodic pressure along the fiber.

Fabricating a Mach-Zehnder interferometer in single-mode fiber is relatively straightforward, but doing so in photonic crystal fiber is more challenging. Unlike Ge-doped single-mode fiber, pure-silica photonic crystal fibers are not photosensitive. The long-period grating that acts as a beamsplitter cannot be written into the fiber with ultraviolet light, as it can in a single-mode fiber.

Instead, the researchers created the grating mechanically. They pressed two identical, 2-cm-long groove plates against the fiber at locations separated by 5.5 cm. The period of the grooves in the plates was 600 µm, and the resulting pressure on the fiber induced a refractive-index modulation that formed the long-period grating (Figure 4).

Mach-Zehnder interferometers in single-mode fiber are well-understood, but it will be more difficult to analyze the sensing data from one in a photonic crystal fiber because the physics of light propagation in the airhole cladding of a photonic crystal fiber has not been thoroughly studied. Indeed, the members of the collaboration believe that their device may find its first application as a tool to study cladding modes in these fibers.