At the previous SPIE conference in San Diego (2008), the authors presented and compared a range of low-cost optical fibre sensors for monitoring the cross-linking process of a thermosetting resin. The same sensor was used subsequently
to monitor and quantify the diffusion of water in the cross-linked polymer.
The current paper presents recent data on the deployment of an array of low-cost fibre-optic sensors to monitor the water
diffusion front. The data obtained from the sensors are compared with conventional gravimetric measurements and
theoretical predictions for the diffusion profile for water ingress in a cross-linked epoxy/amine resin system.
Whilst considerable progress continues to be made on the design and deployment of fibre optic sensors for chemical
process monitoring and structural integrity assessment, the majority of these sensor designs can only impart information
on one or two relevant measurands. For example, in the case of chemical process monitoring of advanced fibrereinforced
composites involving thermosetting resins, it is generally appreciated that cross-linking kinetics can be
influenced by a number of factors including the following: the stoichiometry of the reagents, temperature, surface
chemistry of the substrate and presence or absence of contaminants. Thermosetting resins also shrink during the crosslinking
process. When thermosets are used and processed above room temperature during the production of fibrereinforced
composites, upon cooling back to ambient temperature, residual stress can develop due to the mismatch in
thermal expansions between the reinforcing fibres and the matrix.
This paper reports on recent progress on the design and demonstration of a novel multi-functional fibre optic sensor that
can provide data on (i) temperature, (ii) strain, (iii) refractive index, (iv) transmission infrared spectroscopy and (v)
evanescent wave spectroscopy. A unique and attractive feature of this sensor is that a conventional commercially
available Fourier transform infrared spectrometer is used to interrogate the sensor. The sensor design is based on an
extrinsic fibre Fabry-Perot interferometer.
The authors have demonstrated previously that reinforcing glass fibres can be used as light-guides to facilitate chemical
process monitoring and structural integrity assessment of fibre reinforced composites. In the current paper, the authors
explore concepts for the development of self-sensing, self-healing and crack-arrestor composites.
The first part of the papers presents a brief overview of previously reported technologies for self-sensing, self-healing
and crack-arrestor; the advantages and disadvantages of the various technologies are discussed. The second part of this
paper present the design concept and performance requirements for the self-sensing, self-healing and crack-arrestor
composites. The final part of the paper presents preliminary results on the manufacture and evaluation of this class of
It is generally appreciated that the ingress of moisture in composites can have adverse effects on matrix-dominated
properties such as the glass transition temperature and compressive mechanical properties. Moisture ingress in
composites can also lead to swelling and blistering. A number of excellent studies have been reported on the detection,
modelling and effects of moisture ingress on the properties of thermosetting resins (matrix) and composites. However, it
is generally taken for granted that the quality of the resin and the processing conditions used to cross-link the resin are
identical. Given the recent advances in the design and deployment of optical-fibre sensors in composites, it is now
possible to use the same sensor to facilitate in-situ cure monitoring and structural health monitoring (after processing).
This paper will present recent developments in the design of
low-cost fibre-optic sensor systems for in-situ chemical
process monitoring and the detection of moisture ingress after curing. The cure kinetics derived from three fibre optic
sensor designs is presented as well as those obtained from evanescent-wave spectroscopy using E-glass fibres.
After conducting the in-situ cure monitoring experiments, one of the fibre-optic sensor designs was selected and the
samples (with the embedded sensors) were dried to constant mass at 50°C then transferred to water baths maintained at
70, 50, and 30 °C. The diffusion kinetics for the samples was determined using samples without and with embedded
optical-fibre sensors. The effect of moisture ingress in the resin was also assessed using dynamic mechanical thermal
analysis (DMTA), transmission infrared spectroscopy (FTIR) and differential scanning calorimetry (DSC). Preliminary
results are also presented to demonstrate that the reinforcing fibres in E-glass composites can be used to track the cross-linking
kinetics of a commercial epoxy/amine resin is presented.
Cure monitoring is a term that is used to describe the cross-linking reactions in a thermosetting resin system. Advanced
fiber reinforced composites are being used increasingly in a number of industrial sectors including aerospace, marine,
sport, automotive and civil engineering. There is a general realization that the processing conditions that are used to
manufacture the composites can have a major influence on its hot-wet mechanical properties. This paper is concerned
with the design and demonstration of a number of sensor designs for in-situ cure monitoring of a model thermosetting
resin system. Simple fixtures were constructed to enable a pair of cleaved optical fibers with a defined gap between the
end-faces to be held in position. The resin system was introduced into this gap and the cure kinetics were followed by
transmission infrared spectroscopy. A semi-empirical model was used to describe the cure process using the data
obtained at different cure temperatures. The same sensor system was used to detect the ingress of moisture in the cured
Optical fibre is often required for long-term applications in harsh environments, eg. underwater cables or aerial
cables in snowy areas. However, stress corrosion or the growth of cracks present on the surface of the glass results in
the failure of the fibres, especially if they are to be deployed in humid areas and bearing a static stress. The presence
of a fatigue limit, a stress under which the crack does not grow has not been established in pure silica glass, in
contrast to borosilicate and the soda-lime glasses. A method to check the presence of a fatigue limit was developed
by Gupta1 by studying the relation between the power and exponential laws of crack growth. Here, dynamic fatigue
tests were carried out to estimate the fatigue parameters. The limiting stress is found and is compared to static fatigue
results from tests carried out under hot water to accelerate the corrosion process. The extrapolation of the static
fatigue stress for a time of 25 years predicts a failure stress that is higher than the fatigue limit.
In this paper we report on investigations of some of the factors that have a bearing on the reliability and repeatability of polymer fibre Bragg gratings. The main issues discussed are the fibre preform composition, the fibre drawing conditions and the thermal history of the fibre grating.
The dynamic fatigue test is often preferred to the static fatigue test to measure stress corrosion parameters due to the short time required to conduct tests. The results can be interpreted using the power law or the exponential law for crack growth. However, the results are usually interpreted using the power law which can be analytically integrated in the dynamic fatigue form in contrast to the exponential law which cannot be integrated analytically. Nowadays, the two-point bend machine is used widely to carry out dynamic fatigue tests. Two loading modes are available for this machine: constant plate speed and constant stress rate. In the constant plate-speed mode the mobile plate moves with constant speed towards the other plate thus increasing the stress at the apex of the bent fibre in a non-linear profile. The constant stress-rate mode, which is the more advanced, requires that the mobile plate moves at a variable speed to maintain a constant rate of stress increase at the apex of the optical fibre. In this paper we examine dynamic fatigue tests on optical fibre in water at different temperature using both the constant stress-rate and the constant plate-speed loading-profiles. Then we analyze the results using the power law and the exponential law for crack growth. The parameters for stress corrosion are then used to predict static fatigue behaviour of the optical fibre and are compared with experimental results of static fatigue carried out in two-point bend under water.
In addition to the numerous assessments presented in literature of the radiation-induced optical property changes in
optical fibres, we investigate the impact of MGy dose levels on the mechanical properties of different single and
multimode fibres. We compare the tensile test and two-point bending test results obtained by two different laboratories
with commercially available optical fibres before and after irradiation up to 15 MGy. Our results show a significant
strength reduction of about 50 % at high dose levels, relevant for particular applications in nuclear power facilities and in
large nuclear physics experiments. This strength reduction seems to depend on both the coating materials and the test
conditions, as suggested by the roughening of the outer glass surface of the optical fibre, observed with atomic force
microscope (AFM) images.
This work presents a study of the mechanical properties of optical fibre sensors based on fibre Bragg gratings for application in the in situ monitoring of displacements in tunnels. Tests were performed on fibres and gratings in both dry and humid environments. The tests encompassed both dynamic fatigue (constant stress rate) in both two-point bend and in tensile loading and static fatigue (constant stress conditions). The results indicate that the fibres and gratings are expected to have a very long lifetime under the likely mechanical and environmental conditions.
The reliability of optical fibre for sensor applications is an important issue since they are embedded in structures. The production of these sensors sometimes requires that the coating is stripped from the optical fibres with hot sulphuric acid. Tests conducted on the static fatigue of optical fibres under water, where the time to failure is measured at different stresses showed a deviation from the power-law equation. This indicates an increase in the stress corrosion parameter, n, at lower stress. Tests done using a two-point bend machine to measure the fracture strength of fibre at different time intervals after previous loading to a fixed stress showed a slight increase with time before the strength starts decreasing. This can be explained by the suggestion that the silica bonds deform by twisting to decrease the stretch in the bonds themselves before cracking. However, this might be a problem for optical fibres strain sensors used at high stress.
Monitoring the movement of existing tunnels when new tunnel construction or other construction activities occur in close proximity is important to the tunnel owners. Existing manual monitoring systems, although considered most reliable, require access to the tunnel outside of passenger traffic hours and thus only provide measurements during a limited time of the day and under non-operational conditions. Remote monitoring systems, enabling 24 hour monitoring, are currently available based on electrolevels and automatic motorised theodolites. However, it is hoped that the proposed optical fibre sensor system will be more versatile and economic to install and operate.
The underlying concept is based on a fibreglass rod containing optical fibres, with Fibre Bragg Gratings (FBG) written into them, which is fixed at discrete points to the tunnel lining. The movement of the tunnel induces a deformation of the rod, and hence strains the FBGs positioned at different points along the optical fibres. The FBGs work by reflecting narrow bands of light propagating along the optical fibres, the wavelength of the reflected light being a function of the strain at the position of the grating.
This paper will discuss one experimental arrangement designed to explore the challenges of installing the monitoring system and interpreting the results. Results from tests are presented and discussed together with the sensitivity and accuracy achievable with the proposed system. Further, one method of interpreting the strain measurements and hence determining the displacement of the structure will be presented.
The effects of the fabrication procedure for fibre Bragg gratings (FBG) on the mechanical properties of fibres were investigated by dynamic fatigue tests in two-point bending in a “wet” environment. The results show that both chemical stripping and continuous wave UV irradiation reduced the mechanical resistance of the fibre.
Effect of the hot acid stripping of optical fiber polymeric coating on the dynamic and static fatigue is studied for assessing the impact of the stripping process on the life expectancy of embedded sensors.
A study is made into the effect of stripping the polyimide coating by hot acid immersion on the tensile fracture stress of polyimide- coated optical fibre. It is shown that the strength does not degrade for one or multiple lengths of stripped fibre, nor for long (lengths up to 150 mm) stripped areas. Recoating with polyimide is performed and it is shown that, although strength degradation can occur due to the increased handling, it does not necessarily do so. Strength distributions of pristine, chemically-stripped, laser- ablated and recoated fibres indicate the magnitude of flaws introduced by handling, which are likely to be the cause of strength degradation.
In recent years optical fibers have been developed increasingly for sensor applications. In many of these applications optical fiber Bragg gratings are embedded in a matrix material as temperature or strain gauges. In effect, embedded optical fiber sensors are single-fiber composite (SFC) materials. The mechanical properties of fiber-reinforced composite materials are an important area of research, as well as the mechanical reliability of fibers used for sensors. The multiple fragmentation, with increasing load, of the fibers in SFCs is controlled by transfer of stress from the matrix to the fiber. The number of fiber fragments increases rapidly with increasing strain after the initial break with the fragment length decreasing until a limiting fragment length is reached. Analysis of the fragmentation process allows information to be obtained on the strength of the optical fiber on different length scales and on the transfer of stress from the matrix to the fiber. In this paper we present the issues surrounding mechanical tests on an optical flber/matrix SFC and the mechanical reliability of fibers for sensors. We consider the experimental difficulties and issues which must be addressed for a full understanding of interface effects and their effects on mechanical reliability.
In this paper we present results obtained from computational studies on a set of experimental data concerning the strength of commercially available silica optical fibres. We report on the distribution of estimated values of the Weibull modulus using a variety of Weibull estimators and sample sizes. We compare the distribution of the estimated Weibull parameters from experimental data to analogous distributions obtained using Monte Carlo methods.
We show that for small sample sizes <40 the maximum likelihood method produces a skewed distribution of estimated Weibull parameters. There is a small but finite probability that the Weibull modulus will be overestimated by a factor of 2. We also show that the experimental data are in reasonable agreement with our Monte Carlo simulation for linear regression methods. However, for maximum likelihood methods the peak of the distribution of estimated Weibull modulus can be seen to shift for the strength data but the peak is more stable for the Monte Carlo data
The increased bias of the distribution of estimated Weibull modulus from experimental data must be considered when inferring the reliability of optical fibers from strength data.
In this paper we present the results of computational studies on a set of experimental data concerning the strength of commercially available silica optical fibers. We compare various statistical methods for characterizing the distribution of fiber strength and assess their validity by comparing random sampling from a set of experimental data with results generated using a Monte Carlo computer routine.