Many land and undersea oil operations rely heavily on temperature sensing for improved safety and functionality in harsh environments. Silica-based optical fibers have been increasingly used for distributed temperature and acoustic sensing applications, in which temperatures and/or acoustic signals are recorded as a continuous profile along the length of the optical fiber sensor cable. For fiber interrogation up to 15 km, well and pipeline operators can use fiber-based sensing to measure, in nearly real time, an entire wellbore or pipeline span with a resolution of 1 m or less using Raman-, Brillouin-, or Rayleigh-based backscatter methods. An example of such an application is shown in Figure 1, where steam-assisted gravity drainage (SAGD) technology uses an optical fiber for distributed temperature measurements. Figure 1. Steam-assisted gravity drainage (SAGD) technology for producing heavy crude oil and bitumen includes distributed temperature sensing with an optical fiber. DTS: distributed temperature sensing. Courtesy of OFS. Optical fibers operate under extreme conditions that may include high temperatures and pressures as well as ionizing radiation and aggressive chemicals in the environment. In oil wells, for instance, temperatures can reach 380 °C and pressures can rise to 2000 psi standard atmosphere (atm), while the expected lifetime of the fiber is years or even tens of years1. To be used in such conditions, the fiber must be mechanically robust, while transmitting optical power with a minimum of added attenuation or signal loss. Fiber design and longevity Most of the optical fibers used in these applications have a silica-based core and cladding, taking advantage of the advanced properties of silica, such as high optical transmission, superior thermal stability, and mechanical robustness. The core and cladding are enclosed in a polymer coating, which is designed to protect the fiber mechanically and to minimize bend-induced optical losses. In general, organic polymer materials are less thermally and environmentally stable than silica glass. Therefore, the mechanical stability of optical fibers is primarily dictated by failures of the polymer coating, especially when the fibers are used in the harsh environments mentioned above. Somewhat arbitrarily, failures of optical fiber are grouped in two categories: increased optical loss and mechanical breakage. The main factors responsible for the development of added optical loss are hydrogen ingression into the fiber, the effects of β and γ ionizing radiation, and macrobends and microbends in the fiber. The effects of hydrogen ingression and ionizing radiation have been extensively studied2,3,4. It is more challenging to predict the effects of microbending, which develops over time as a result of the degradation of the polymer coating. For instance, the tendency of the fiber to stick to adjacent hardware while cooling may result in the development of significant microbend-induced loss. Alternatively, microbends can develop due to chemical degradation of the coating material, which leads to axial and radial shrinkage of the coating and changes to its mechanical properties. The mechanical robustness of optical fibers at ambient conditions has also been well-studied5, while the performance of optical fibers after exposure to elevated temperatures and/or liquid media has been studied to a lesser extent. Instead, the latter exposures were performed at ambient pressure, and the studied temperatures did not exceed 100 °C. To the best of our knowledge, no systematic data has documented the mechanical strength of optical fibers during high temperatures and pressures. In recent work, OFS performed an experimental study of fibers immersed in four types of high-temperature/high-pressure fluids: distilled water, seawater, isopropyl alcohol (IPA), and paraffin oil. The choice of these fluids was dictated primarily by undersea and downhole applications, where fibers can be exposed to such (or similar) environments. For example, paraffin oil is composed of saturated hydrocarbons that resemble the basic material of most optical cable gels6. IPA is used for cleaning the interiors of metal tubes in fiber-in-metal-tube (FIMT) constructions and also for the installation of fibers into the tubes. Water is a factor in quite a few geophysical applications. Since optical fibers can be employed with different coatings, seeing the advantages and limitations of each coating system in various environments is of particular interest. The coatings selected for the study belong to the following main classes of chemistries used on optical fibers: Acrylates Silicones Polyimides Hybrid organic-inorganic materials Graphite carbon Determining failure criteria When optical fibers are used at elevated temperatures and/or in aggressive environments, it is important to know the factors limiting their performance in those conditions. The failure mechanism is always of primary interest. For optical fibers, the most frequent failure modes are related to added attenuation or loss of mechanical strength. Failure criteria can be quite different and depend on the type of application. Optical loss requirements, for instance, are sometimes dictated by the attenuation budget of the sensing interrogator being used. By definition, failure defines the lifetime of an optical fiber, and therefore different criteria may correspond to different fiber longevities, even under exactly the same usage conditions. The failure criterion dictates the boundaries of the upper use temperature and the continuous use time. These two terms are also related to each other: At the same failure criterion, higher upper use temperatures correspond to shorter continuous use times7. In this study, strength degradation was the failure criterion. More specifically, it was determined that a fiber should be considered to have failed if at least one data point fell below 2 GPa. It must be emphasized that all observations were made for relatively short 7- to 14-day aging times. If, at a certain temperature/environment, the fiber did not fail the strength test after 10 days of aging, this did not necessarily mean that the same fiber would survive longer exposures, such as months or years. On the other hand, if the failure did happen within these 7 to 14 days, it would not inspire hope for a prolonged use of the fiber at the trialed conditions or at more aggressive conditions of environment, temperature, and pressure. Experimental conditions Of the eight fibers under study and drawn at OFS, seven had a 50-µm graded-index (GI) core and 125-µm diameter silica cladding. The numerical aperture of the fibers was 0.2. The following coating types were used on the GI 50/125-µm fibers: Two types of dual acrylate (DA-1 and DA-2) coating, both telecom grade. A single acrylate (SA) dual coating with a silicone primary layer and a high-temperature acrylate secondary layer (Si/A). A hybrid organic-inorganic coating (Hyb). Polyimide (PI). Carbon/polyimide (C/PI). For dual coating systems, the primary and secondary coating diameters were 190 ± 5 and 250 ± 10 µm, respectively. The diameter of the SA-coated fiber was 243 µm; the diameter of the PI- and C/PI-coated fibers was 155 ± 5 µm. The carbon (C) coating thickness was about 90 nm. In addition, a 200-µm pure silica core fiber was drawn and coated with a fluoroacrylate optical cladding and upbuffered with ethylene-tetrafluoroethylene (ETFE) copolymer. The glass, coating, and buffer dimensions of these polymer-clad fibers were 200, 225, and 500 µm, respectively; the numerical aperture was 0.37. This fiber type is called hard-clad silica (HCS). Among the four immersion fluids — distilled water, artificial seawater, paraffin oil, and IPA — the seawater was modeled by adding sea salt to distilled water. Two separate solutions were made: one with 3.5 percent salinity (the concentration found in seawater) and the other with 7 percent salinity (twice the amount of salt). Test results Table 1 summarizes the upper use temperatures for the different coating types observed in the study. This data corresponds to the criteria for strength loss failure and the few days of continuous observation, as mentioned. The far-right column shows the upper use temperature that had been determined previously for the same coatings in a dry air environment, assuming a failure criterion of 25 percent loss of the initial mass and a lifetime of 20 years7,8. Table 1. Upper use temperatures for optical fibers in different environments. Courtesy of OFS. It is interesting to compare the effects induced on the optical fibers by water, IPA, and paraffin oil. As expected, water was the most aggressive environment for most of the coatings (Figure 2). Water deteriorates the glass cladding of the fibers and, at elevated temperatures, efficiently dissolves polymer coating materials along with the glass. Carbon is the only coating that survived exposures to water up to 300 °C. The addition of sea salt to water makes the environment even more aggressive. Figure 2. Diameters of selected fibers before and after aging in distilled water at 2000 psi. Courtesy of OFS. Paraffin oil is the most inert environment among those studied (Figure 3). Still, at 300 °C, polymer coatings exhibit some dissolution in the oil. Figure 3. Strength of optical fibers before and after 10-day exposure to paraffin oil at ambient pressure (at 150 °C) and 2000 psi (at other temperatures). Error bars show the range of the observed values. Courtesy of OFS. For most of the coatings, IPA displays intermediate properties between paraffin oil and water (Figure 4). There are two exceptions: DA-1 and DA-2. These coatings swell strongly in IPA. There is no doubt that the soft primary coatings are the most vulnerable to swelling, which eventually leads to cracking of the dual coatings, followed by their ultimate separation from the glass cladding. Figure 4. Strength of optical fibers before and after 7-day exposure to IPA at 1500 psi and different temperatures. Error bars show the range of the observed values. Courtesy of OFS. From the data obtained, the coatings can be rated according to their thermal stability in different environments. For applications in high-temperature water, for example, the best coating system is C/PI, but it should be noted that at temperatures above 200 °C, PI will sooner or later be dissolved in water. Although carbon protects the glass from dissolution, the remaining coating is not mechanically robust. Therefore, the C/PI optical fiber is expected to survive in water at 200 °C or higher only if it is not challenged mechanically. Four coatings survived the highest temperature aging in IPA: SA, PI, C/PI, and Hyb. Three of these coatings (SA, C/PI, and Hyb) also survived the highest temperature exposure in paraffin oil. As expected, the PI coating displayed the most consistent thermal stability in different environments, while the DA-1 and DA-2 coatings showed the lowest thermal stability. Other than that, no direct correlation exists between the data obtained using different failure criteria and that collected for different environments. Conclusion The survivability of optical fibers under the harsh conditions found in oil operations on land and undersea depends upon multiple factors, including type of coating, environment, temperature, pressure, and usage time. Aging optical fibers with eight different coatings in high-temperature water, seawater, IPA, and paraffin oil — and using the mechanical strength of the fibers as the lifetime criterion — enabled the ranking of the fiber coatings with respect to their stability in different environments. Fibers with telecom grade DA-1 and DA-2 coatings were found to be the most vulnerable in all studied environments, while C/PI-coated fibers were found to be the most stable. In IPA and paraffin oil, fibers with SA and Hyb coatings were found to have the highest thermal stability. Polymer coating and silica glass were both found to dissolve in water at temperatures above 200 °C, while carbon coating was found to protect the glass fiber from full dissolution in water. Meet the author Andrei Stolov joined Lucent Technologies, now OFS, in 2000. His focus has been the development and characterization of polymer coatings for specialty optical fibers. He has authored multiple publications in the fields of molecular spectroscopy, polymer science, and photonics; email: stolov@ofsoptics.com. References 1. T. Reinsch and J. Henninges (2010). Temperature-dependent characterization of optical fibers for distributed temperature sensing in hot geothermal wells. Meas Sci Tech, Vol. 21, p. 094022. 2. P.G. Lemaire and E.A. Lindholm (2007). Hermetic optical fibers: carbon-coated fibers. In Specialty Optical Fibers Handbook. New York: Elsevier, pp. 453-490. 3. S.L. Semjonov et al. (2006). Fiber performance in hydrogen atmosphere at high temperature. Proc SPIE, pp. 6193, 61930N. 4. M.N. Ott (2002). Radiation effects data on commercially available optical fiber: database summary. Radiation effects data workshop, https://doi.org/10.1109/REDW.2002.1045528. 5. C.R. Kurkjian and M.J. Matthewson (2007). Mechanical strength and reliability of glass fibers. In Specialty Optical Fibers Handbook. New York: Elsevier, pp. 735-743. 6. B.G. Risch et al. (2000). Water blocking gels compatible with polyolefin optical fiber cable buffer tubes and cables made therewith. U.S. Patent 6,085,009. 7. A.A. Stolov et al. (2008). Thermal stability of specialty optical fibers. J Lightwave Tech, Vol. 26, Issue 20, pp. 3443-3451. 8. A.A. Stolov et al. (2016). Thermal stability of specialty optical fiber coatings: observation of kinetic compensation effect. J Therm Anal Calorim, Vol. 124, Issue 3, pp. 1411-1423.