Strength-Probability-Time (SPT) diagrams provide an intuitively pleasing method for presenting reliability data based
on extrapolations from accelerated fatigue testing data. If power-law crack growth kinetics are assumed the calculations
required to generate the SPT diagram are particularly simple. However, if exponential or other more complex forms are
used, this is not the case. If the accelerated data are for dynamic fatigue measurements (strength as a function of
stressing rate) the SPT diagram can only be determined after numerical integration of the crack growth equations,
followed by non-linear regression to the fatigue data. However, we have developed software to perform this task. In
this paper we describe the methods used and show sample results of lifetime predictions using SPT diagrams. Also, the
effect of using different crack growth kinetics models on predicted lifetimes is discussed.
Specialty optical fibers operating in harsh aerospace environments are typically exposed to high temperatures and
elevated humidity. This calls for better performing protective coatings. Recently developed sol-gel derived inorganicorganic
hybrid materials called hybrid glass offered improved protective performance as compared to standard dual
polymer coated fibers [1]. In this paper we examine the effectiveness of online UV curing for the protective ability of
hybrid glass coatings. For this purpose two types of UV-curable hybrid glass candidates representing two different
concentrations of acrylate groups were applied online to silica fibers as single and dual coats. Samples of fibers were
collected and subjected to dynamic fatigue testing by two-point bending. The stress corrosion parameter, n, as well as the
strength of the fibers were determined. Both the strength and n were higher for fibers with two layers of coating as
compared to single coatings even when the thickness of both one and two layer coatings was the same. This may be
caused by the greater degree of cross linking of the inorganic component when the coating is exposed twice to the heat
generated in the UV chamber. Coating materials with reduced acrylate group content had higher values of the fatigue
parameter n but at the same time reduced strength.
The lifetime of an optical fiber depends on its environment. Previous work extensively measured and characterized the separate effects of humidity and temperature on the fatigue parameters using three different kinetics models, but the combined effect has not been determined in detail. In this work, the details of how the fatigue parameters vary with temperature in a humid environment were investigated. It was found that the kinetics model parameters were different from values obtained elsewhere in liquid water. This may be the result of differences in the apparent activation energy for fatigue in liquid and vapor environments.
Often lightguide fiber processing involves steps that may cause degradation of very high strength or flaw-free, perfect fiber. A very obvious type of degradation is the development of abrasion flaws during handling. Also, heating of a fiber to moderate temperatures (~300-600°C), for instance during the soldering of pigtails, has been shown to result in strength degradation of strong fiber. It has been suggested that the use of HF etching may be a reasonable technique for
the elimination of many types of strength-lowering defects. In this paper we discuss early results from the literature on the effects of heating and HF etching on the strength of silica glass and present new results on both.
The strength of optical fiber at low temperature is an important parameter since it approximates the inert strength, i.e. the starting strength of the material before degradation by fatigue. Published data suggest that the fatigue may abruptly slow below some temperature. However, published data are limited to strength vs temperature or fatigue in liquid nitrogen. We report strength and fatigue data for both bare (stripped) and metal coated fused silica optical fiber at temperatures down to 77 K. While fatigue slows as the temperature is reduced (i.e. the stress corrosion parameter increases with falling temperature) fatigue is still measurable at 77 K. This is the case even for hermetic metal coated fiber with extremely low water activity at the glass surface. The results confirm that fused silica exhibits "intrinsic" fatigue, i.e. fatigue in the absence of moisture.
This paper reviews the dependence of strength and fatigue of fused silica optical fiber on the environmental parameters temperature, humidity and pH. It is shown that the stress corrosion parameter, n, is not a constant but depends on the nature of the environment. Further, different kinetic forms for the stress corrosion kinetics lead to different interpretations of experimental results. Since lifetime predictions are very sensitive to the value of n and the kinetic form, it is important to know which form is correct. It is shown that the empirical power law form that is almost exclusively used by the fiber optics industry provides a good fit to fatigue data for high strength fiber, but an exponential form provides a more self-consistent description of fatigue in different environments.
The mechanical reliability of fused silica glass fibers is significantly influenced by the properties of their polymer coatings. The primary coating, which is in contact with the fiber surface, is expected to control the chemistry there, but the secondary coating does have a considerable effect on the strength and aging behavior of the fiber. This observation is confirmed by data obtained for ten fused silica fibers, having the same primary coatings but ten different secondary coatings. These fibers were aged at 85 degree(s)C in both de-ionized water and 85% relative humidity for up to 6 weeks and the residual strength as a function of aging time was measured. Dynamic fatigue measurements were carried out on as-received and aged fibers using two-point bending. The results show that the secondary coating has a notable effect on the aging behavior and the coating strip force, but does not greatly influence the dynamic fatigue parameter.
KEYWORDS: Diffusion, Polymers, Optical coatings, Finite element methods, Humidity, Glasses, Optical fibers, Systems modeling, Fiber coatings, Chemical elements
In earlier work, diffusion of moisture through polymer coatings was modeled by using an analytical solution to the diffusion equation and so was only applicable to the simplest cases, e.g. cylindrically symmetric Fickian diffusion. In this work the limitation of the analytical approach is avoided by the use of finite element analysis. However, finite element programs do not usually implement matter diffusion, and therefore it has been modeled by analogy with thermal conduction.
It is now well known that fused silica optical fiber can suffer from enhanced strength degrathtion after prolonged exposure to
aggressive environments. This is caused by corrosion of the glass surface by moisture leading to roughening, strength loss,
and, potentially, problems with handleability. It has been found that addition of nanosized silica particles to the polymer
coating can improve the long term mechanical reliability by slowing corrosion and delaying the onset of strength loss.
However, previous studies have shown that addition of these particles can lead to unacceptably high added optical loss, when
measured using the "basketweave" test. In this work, it is shown that the added loss caused by coating additives can be
reduced by improving the mixing and dispersion ofthe silica powders in the polymer. It is further shown that well dispersed
powders still substantially improve the long term fatigue and aging behavior. This clearly shows that coating additives can
improve the mechanical reliability without significantly degrading the optical performance.
Optical fiber may experience cyclic stresses at frequencies ranging from a few hertz in aerial cables to over a kilohertz due to vibration of machinery. The fatigue behavior of brittle materials typically gives times to failure that correspond to a suitably time-averaged applied stress and is independent of the frequency. Previous studies have been limited in the frequencies used but generally show agreement with this simple model. In this paper we describe results for the cyclic fatigue behavior of high strength fused silica optical fibers as a function of stress amplitude and frequency in the range of zero to 100 Hz. The results confirm that fatigue of this material is indeed accurately described by the subcritical crack growth model and the results are shown to be frequency independent in the range studied.
Optical fibers have been found to exhibit an accelerated rate of strength reduction during static fatigue and zero stress aging for long times in aggressive environments. This phenomenon has been commonly referred to as the fatigue and aging `knee'. The onset of the knee has been found to be highly variable and is sensitive to the polymer buffer coating. In past work we have shown that moisture vapor penetrates most polymer coatings on the time scale of minutes, which implies that the diffusion rate of small molecules is not the rate-determining step for aging. On the other hand, the diffusion of large molecules through the polymer coatings can take anywhere from weeks to years to reach the polymer/glass interface. The implication of this result is that large molecule diffusion might be the rate- determining step in aging. In the work presented here the diffusion of moisture and pH buffer solutions through various optical fiber coatings will be discussed. These results are correlated with the zero stress aging behavior of the same fibers.
Subcritical crack growth in fused silica can be modeled as a stress assisted chemical reaction between water and strained bonds at the crack tip. The stress influences the crack growth rate by reducing the free energy of the activated complex. In principal, the stress changes both the activation enthalpy (energy) and entropy; however, the influence of stress on entropy has generally been ignored. The dynamic fatigue behavior of `pristine' optical fiber can be used to determine the fatigue kinetics parameters with unprecedented precision. It is shown that the entropy contribution is at least as significant as the enthalpy and therefore should not be ignored.
The mechanical reliability of optical fiber used in certain biomedical applications is extremely important because failure of the fiber during use might be fatal for the patient. Therefore, prediction of the lifetime of the fiber both in storage and during service is necessary before the fiber can be safely used. In this paper we study two commercially available optical fibers designed specifically for high power laser delivery. The fatigue parameters calculated from static fatigue data are used to estimate the maximum allowed stress that ensures survival for the deign life of the fiber. This work properly accounts for uncertainty in the predictions; uncertainty which arises not only from scatter in the experimental data, but also from uncertainty in the form of kinetics model to use for extrapolation (i.e. power law, exponential, etc.). This paper thus provides an outline for making lifetime predictions for a critical applications involving relatively short lengths of fiber, that does not bind in any questionable assumptions.
A single-layer UV-curable polyacrylate-coated telecommunications grade fused silica fiber was found to have a significant reduction in two-point bending strength after immersion in acetone. The two-point bending and tensile strengths of this fiber as a function of immersion time in acetone were determined, and this strength loss was not seen for 0.5-m gauge length tensile specimens. SEM and optical fractography was performed on the weak specimens, and the cause of the strength reduction is proposed to arise from particles smaller than 3 micrometers in the coating. These particles could cause surface flaws by sliding contact damage incurred during relative motion between the coating and the glass. This sliding could occur either while flexing the fiber in preparation for a bending strength measurement or due to coating elongation. While it is not clear which mechanism is operating, both are consistent with the observation that degradation is only observed for bending strength measurements.
Hermetic aluminum-coated fused silica fibers can withstand high stress levels without failure for prolonged periods of time in water-containing environments. Aluminum-coated fibers from several sources exhibit differences in strength. The aluminum and silica surfaces have been examined using SEM and AFM in order to understand this variation. Differences in the interfacial interaction between aluminum and glass and in the microstructure of the coatings were considered, but were not unequivocally identified as being responsible for the differences in strength observed for the various aluminum-coated specimens.
Two formulations based on perfluorinated polymer were prepared for use as UV-curable optical cladding for silica fibers. In the first formulation an adhesion promoting agent based on fluoroacrylate resin was synthesized and mixed with the experimental product Defensa 7702++ in order to promote wetting and chemical adhesion to the silica fibers. In the second formulation, wetting and physical adhesion between the liquid coating and the silica fibers were achieved by increasing the viscosity of the starting coating by addition of unsaturated perfluorinated polymer into Defensa. Both formulations were used as primary coatings on dual coated silica optical fibers. The mechanical behavior of the formulations was characterized by the strip test, the pull-out test and zero stress aging in 90 degrees Celsius pH 7 buffer. The results show that both formulations exhibit better wetting-adhesion characteristics than unmodified starting coating and that the strength degradation during zero- stress aging was lower for the fiber coated with the formulation of higher viscosity.
The two-point bending technique has been used to measure the strength of both polymer coated and bare fibers in liquid nitrogen after the fibers were first aged in an aggressive environment followed by a drying process. The results show that some strength recovery occurs upon drying of polymer coated samples while continuing degradation was seen when drying bare samples. The healing process observed for coated fiber is thought to be caused by condensation of the hydrated surface layer formed during aging.
The strength degradation of lightguide fibers has been studied over a range of elevated temperatures and at room temperature. Using these data we show that accelerated testing can be used to predict ambient temperature behavior. An activation energy of approximately 90 kJ.mol-1 describes the shift in corresponding times.
Vickers indentation has been used to introduce controlled flaws in fused silica optical fiber in order to model the behavior of 'weak' defects encountered in practice. Novel techniques are used in order to conveniently examine indentations over a broad range of residual strength; these include the use of flat fiber which facilitates specimen alignment during indentation and subsequent strength measurement in bending. The strength of indented silica fiber measured in pH 7 buffer reveals a bimodal behavior at the threshold for radial crack formation which is related to the 'pop-in' of radial cracks after indentation or during strength testing. An unusually low value of the stress corrosion parameter for subthreshold indentations of n approximately equals 11 is observed in pH 7 buffer. This suggests that under some conditions the usual assumption of n approximately equals 20 may lead to an overestimation of fiber reliability.
The most critical variable effecting the strength of silica lightguides is the
availability of water to the fiber surface. At very low temperatures (T < 77 °K) and at
high vacuum (PH2O < i07 torr) the thermodynamic activity of water is so low that
mechanical failure of the fiber occurs by the direct breaking of Si-O-Si bonds. In this
case very high strength (- 12-14 GPa) and very slight time dependence (formula available in paper) are expected. Also an activation energy comparable to the SiO
bond energy is observed (- 100 kcals/mol). On the other hand, under normal
conditions (T - 0- 100°C and normal relative humidity), the strength and time
dependence are controlled by the combination of stress and the reactivity of water with
the fiber surface: SiO2 + H2O = SiOH HOSi. In this case the time dependence of
strength is very much greater (n 20) and the activation energy is approximately 30
kcaI/mol.2 Because of this rather extreme time dependence, the short time tensile
strength (say tf = 10 sees) is only about 5.5 GPa and will be reduced again by a factor
of 2 (to = 2.8 GPa) in about one week. A subject which continues to be discussed
and studied is the proper analytical description of this time dependence. In this regard,
Bubel and Matthewson3 have studied the behavior of several proposed models for
time dependence. They find that the differences in predicted lifetimes from the models
differs significantly. In particular they suggest that the universal use of the optimistic
power law is not appropriate.
Fused silica optical fiber tested in aggressive environments can exhibit a 'knee' in both the zero-stress aging and the fatigue under stress; degradation proceeds at an accelerated rate beyond the knee. This behavior leads to shorter lifetimes than predicted from short term data and to strength degradation even in the absence of an applied stress which can result in handleability problems. While the first observation of this behavior was for a humid environment, later work only reported the knee in liquid aqueous environments. This paper reports the observation of a pronounced fatigue and aging knee for a fiber tested in 85 degree(s)C, 85% relative humidity, clearly indicting this phenomenon can occur in more benign environments. Surface roughness measurements using atomic force microscopy also show an abrupt increase in roughness indicting that, for this fiber at least, the development of surface roughness before the knee can not be used as a precursor for predicting the position of the knee.
Three kinds of UV-curable organically modified silicates have been prepared to be used as protective coatings for optical fibers. The synthesis involves the reaction of the thiol group of 3-mercaptopropyl-trimethoxysilane with a C equals C bond in one of the acrylic groups of three commercially available aliphatic triacrylates. The methoxysilyl groups of the synthesized diacrylate methoxysilanes were subjected to hydrolysis and condensation to form Si-O-Si units. Transparent, viscous, solvent-free resins were obtained that hardened in seconds when exposed to UV radiation. The coating derived from the reaction with glycerol propoxy triacrylate (GPTA) proved to adhere the best of the three to both plastic and glass substrates. It was then tested as a protective coating for silica fibers. Reliability tests were carried out including bending strength and fatigue tests at pH 7 and 10. The results show improved water resistance of the coated fiber in neutral conditions.
We have shown in previous work that the addition of small quantities of colloidal silica to the UV- curable polymer coating of fused silica optical fiber causes a dramatic improvement in the fatigue and aging resistance both in aqueous and in constant humidity environments. The presence of silica in the coating inhibits the mechanisms responsible for the surface roughening that causes the fatigue knee and strength degradation during zero-stress aging. This work presents results which show the effect of higher concentrations of the silica additive (6 wt%) and of an adhesion promoting agent on both the rheological properties of the polymer coating and the fatigue and zero-stress aging behavior of the fiber. Viscosity measurements show thixotropic behavior which indicates that the silica particles tend to form a network structure in the prepolymer. Filtration of the prepolymer to remove large particles is hampered by this phenomenon. The fiber coated with the silica-containing polymer exhibits substantial improvement in the long term mechanical reliability compared to a reference fiber without additive in the coating.
Examination of the surface profile of silica optical fiber using the atomic force microscope (AFM) has proved a useful technique for understanding strength degradation of the fiber upon aging in aggressive environments in terms of the production of surface roughness. However, before AFM examination it is necessary to remove the polymer protective coating and this is usually achieved by dipping the fiber sample in methylene chloride (MeCl) or hot (approximately 200 degree(s)C) sulfuric acid. This raises the possibility that the stripping technique modifies the fiber surface. In this work it is shown that hot acid stripping does not affect the fiber strength. It does, however, remove a surface layer from the aged fiber, probably of hydrated silica, which does not contribute to the strength. Therefore, treatment with hot acid is necessary in order to reveal the strength controlling surface profile, even if there is no polymer coating requiring removal. MeCl does not remove the surface layer and does not reveal the strength controlling surface.
Current models for optical fiber reliability are mostly based upon power law growth kinetics of sharp, stress-free cracks. The physical basis of these models is critically examined and is found to have limitations in describing the behavior of both high strength `pristine' fiber and weak fiber. In particular, the models do not account for the abrupt strength loss sometimes observed in harsh environments for both types of fiber. Recent advances in understanding the behavior of such fibers are discussed. In particular the addition of colloidal silica particles to the coating material is shown to dramatically improve reliability.
Systems containing optical fiber have design lives on the order of decades so that models for assessing the mechanical reliability of the fiber must rely on extrapolations from accelerated short term testing. Such extrapolations are only valid if all relevant mechanisms are fully understood. The physical processes giving rise to mechanical degradation are reviewed and it is shown that no single model describes all situations. In particular, strong “pristine” fiber can behave quite differently from weaker fiber. Additionally, two degradation regimes are identified, one which is stress assisted (fatigue) and one which can occur even in the absence of applied stress (aging). Recent advances in understanding these phenomena are discussed and promising areas for future work are proposed.
This paper reviews the common techniques for mechanical testing of optical fiber specimens and compares and contrasts their attributes. Any technique must be able to grip the specimens without causing failure at the grips. The techniques generally fall into two categories; uniaxial tensile testing in which the fibers must be gripped carefully, and bending techniques which reduce the local stress at the grips. The former techniques generally give results that are easier to interpret due to the homogeneous stress field but are experimentally the least convenient. In contrast, bending techniques are experimentally convenient, but due to the inhomogeneous nature of the stress field developed and the short fiber test lengths, may not be as useful as tension for some purposes.
Recent advances in the understanding of reliability of silica optical fiber indicate that the chemical durability of the fiber can control the long duration lifetime both under stress- induced fatigue and zero-stress aging conditions. In particular, dissolution of surface material produces strength degrading surface roughness. These mechanisms are discussed and strategies for improving reliability by inhibiting dissolution are examined. As an example, a modified polymer coating formulation is described that is shown to increase the lifetime of the fiber by up to a factor of thirty-fold. Strategies for improving the strength and durability of non-oxide fibers using a similar approach are also discussed.
A computer program is described that can analyze optical fiber fatigue data from a variety of fatigue experiments and for a variety of crack growth kinetics models and can then predict long term static fatigue behavior. The key feature of the program is its’ ability to analyze the statistical uncertainty in the predictions due to scatter in the data to which the models are fitted. It is shown that uncertainty in the lifetime predictions is often dominated by the uncertainty in the choice of the appropriate crack growth model. It is therefore concluded that a range of models should be considered when making lifetime predictions.
The effect of high temperature aqueous solutions of various pH values on the mechanical properties of polymer coated optical fibers of an aluminum fluoride-based composition are examined. It was found that such fibers retain much more strength when aged in these aqueous environments than fibers of the more common zirconium fluoride-based composition. The aging is not affected by pH unless the fiber is under stress, in which case a low pH solution decreases the time to failure of the fiber. In static fatigue, the time to failure of the aluminum fluoride-based fibers is 20 times greater than that of the zirconium fluoride-based fibers.
Prediction of long-term static fatigue for optical fibers under stress requires a model to relate short-term accelerated test results to long-term behavior. The dependence of crack growth on stress intensity is the most fundamental model for this reliability prediction process. Statistical uncertainty for fatigue testing is shown to be significant, but typically smaller than the model uncertainty, which has been neglected in the literature. More research is needed to determine the most appropriate model. It is shown that the differences in allowed stress predictions between models become quite large at long times, especially for fiber of the same strength as that used in the fatigue test. Data-independent conversions of allowed stress from the common power model to other models provide an assessment of the difference between models for various situations. In many applications, the differences are in the range of typical safety factors. However, since model differences are quite large in other applications, universal use of the optimistic power law model is not appropriate, given the limited understanding of fatigue in optical fibers.
The strength and fatigue behavior of bulk fused silica is well understood in terms of the growth of microcracks under the combined influence of stress and environmental attack. The behavior of high strength, flaw free silica optical fiber shows significant differences from the bulk material for poorly understood reasons making long term predictions unreliable. It is known that silica fiber strength and fatigue are sensitive to such environmental parameters as temperature, humidity and pH. However, this paper presents results which also show a sensitivity to ionic species in the environment. These results are interpreted in terms of possible models for the fiber behavior.
Kilometer lengths of optical fiber have a much lower strength than short lengths due to occasional defects of an extrinsic nature. The fatigue properties of these defects are hard to study due to their rarity. Subthreshold indentation flaws in silica optical fibers (/. e. Vickers indentations produced under sufficiently low load to avoid radial crack formation) have been shown to exhibit environmental fatigue similar to "pristine" silica fiber. Thus the indentation technique may be used to introduce controlled flaws into the fiber that model the strength limiting defects found in long length specimens. This paper presents the results of fatigue studies on subthreshold indentation flaws that have strengths of up to ~ 1 GPa (typical of proof stress levels).
This course describes the mechanical properties of optical fibers. It includes a description of the effect of environmental moisture on strength and the subsequent implications for reliability. Mechanical testing techniques are described and compared, with a summary of data analysis techniques.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.