Valeria La Saponara

Composite Structural Health Monitoring Through Use of Embedded PZT Sensors

The need to understand and monitor the integrity of structural components made of composite materials is becoming critical, due to an increase of the use of composites in aerospace, civil, wind energy, and transportation engineering. Off-the-shelf piezoelectric transducers embedded inside the composites or bonded onto the structure surface are a possible solution for on-line structural health monitoring and non-destructive evaluation: they can be used to generate Lamb waves, which are able to detect damage. This article focuses on the behavior of two sets of woven fiberglass/epoxy specimens, one with embedded, one with surface-mounted piezoelectric wafer transducers (lead zirconate titanate). The specimens are tested under axial tensile fatigue at high stress ratio, and the transducers are interrogated in pitch-catch mode at different stages of the specimens’ life, while they are subjected to the mean test load (the testing machine is paused). A novel signal processing technique based on wavelet thresholding/denoising and Gabor wavelet transform is discussed. This technique identifies changes in boundary conditions, loading/unloading prior to damage and during damage. It appears to correlate the contour area changes with the so-called characteristic damage state observed in the literature in composite laminates under tensile fatigue.

Detection of spatially distributed damage in fiber-reinforced polymer composites:

This work describes a novel method of embedded damage detection within glass fiber–reinforced polymer composites. Damage detection is achieved by monitoring the spatially distributed electrical conductivity of a strain-sensitive multiwalled carbon nanotube thin film. First, thin films were spray-deposited directly upon glass fiber mats. Second, using electrical impedance tomography, the spatial conductivity distribution of the thin film was determined before and after damage-inducing events. The resolution of the sensor was determined by drilling progressively larger holes in the center of the composite specimens, and the corresponding electrical impedance tomography response was measured by recording the current–voltage data at the periphery of the monitored composite sample. In addition, the sensitivity to damage occurring at different locations in the composite was also investigated by comparing electrical impedance tomography spatial conductivity maps obtained for specimens with sets of holes drilled at different locations in the sensing area. Finally, the location and severity of damage from low-velocity impact events were detected using the electrical impedance tomography method. The work presented in this study indicates a paradigm shift in the available possibilities for structural health monitoring of fiber-reinforced polymer composites.

Spatial Sensing Using Electrical Impedance Tomography

The need for structural health monitoring has become critical due to aging infrastructures, legacy airplanes, and continuous development of new structural technologies. Based on an updated structural design, there is a need for new structural health monitoring paradigms that can sense the presence, location, and severity with a single measurement. This paper focuses on the first step of this paradigm, consisting of applying a sprayed conductive carbon nanotube-polymer film upon glass fiber-reinforced polymer composite substrates. Electrical impedance tomography is performed to measure changes in conductivity within the conductive films because of damage. Simulated damage is a method for validation of this approach. Finally, electrical impedance tomography measurements are taken while the conductive films are subjected to tensile and compressive strain states. This demonstrates the ability of electrical impedance tomography for not only damage detection, but active structural monitoring as well. This paper acts as a first step toward moving the structural health monitoring paradigm toward large-scale deployable spatial sensing.

Experimental and numerical analysis of delamination growth in double cantilever laminated beams

Delamination crack growth in laminated composites is investigated using experiments and finite element (FE) models. Tests are performed on cross-ply graphite/epoxy specimens under static conditions. The load–displacement response is monitored in the tested coupons along with crack length. The FE models employ a cohesive layer that is used to simulate the debonding and crack propagation. The cohesive parameters are calibrated from the experimental load–displacement curves. Crack growth and strain measurements are compared with those from the FE models. The predicted results from the FE models are in good agreement with the test results. The same modeling approach is also used to simulate crack propagation in the transverse direction of a notched laminate. The proposed FE analysis with cohesive layers can simplify fracture toughness assessment in multilayered specimens. 2002 Elsevier Science Ltd. All rights reserved.

Influence of carbon nanotube on the piezoresistive behavior of multiwall carbon nanotube/polymer composites

The role of the physical properties of multiwall carbon nanotubes on the strain-sensing piezoresistive behavior of multiwall carbon nanotube/polymer composites is systematically studied using three types of multiwall carbon nanotubes as fillers of a brittle thermosetting (vinyl ester) and a tough thermoplastic (polypropylene) polymers under quasi-static tensile loading. Two of the three multiwall carbon nanotubes investigated have similar length, aspect ratio, structural ordering, and surface area, while the third group contains longer multiwall carbon nanotubes with higher structural ordering. The results indicate that longer multiwall carbon nanotubes with higher structural ordering yield higher piezoresistive sensitivity, and therefore are better suited as sensors of elastic and plastic strains of polymer composites. The highest gage factor achieved was approximately 24 and corresponded to the plastic zone of multiwall carbon nanotube/polypropylene composites with the longest nanotubes.

The electrical response of carbon nanotube-based thin film sensors subjected to mechanical and environmental effects

Fiber-reinforced polymer composites are a popular alternative to traditional metal alloys. However, their internally occurring damage modes call for strategies to monitor these structures. Multi-walled carbon nanotube-based polyelectrolyte thin films were manufactured using a layer-by-layer deposition methodology. The thin films were applied directly to the surface of glass fiber-reinforced polymer composites, with the purpose of structural monitoring. This work focuses on characterizing the sensitivity of the electrical properties of the film using time- and frequency-domain methods under applied quasi-static and dynamic mechanical loading. In addition, environmental effects such as those of temperature and humidity are varied to characterize the sensitivity of the electrical properties to these phenomena. (Some figures may appear in colour only in the online journal)

Crack branching in cross-ply composites: an experimental study

Abstract The objective of this research is to discuss experimental data regarding a type of failure called crack branching. The specimens tested are layered glass/epoxy and graphite/epoxy cross-ply composites, manufactured with an initial interlayer crack. Experiments were carried out under static conditions. A designed two-level two-variable experiment based on an 8×8 Hadamard matrix was performed in order to identify the key parameters of the phenomenon. Moreover, a smoothing technique (smoothing by running median of 3 repeated) was used to interpret the crack growth rate in terms of branching angles. The results indicate that there is a critical branching angle (39–40°). When the crack branches with an angle greater than this critical value, the crack growth rate increases with the branching angle. Moreover, the branching angle increases when the initial delamination decreases, for branching angles greater than the critical value. The position through the thickness does not seem to significantly affect the branching angle.

Static and dynamic strain monitoring of GFRP composites using carbon nanotube thin films

Fiber-reinforced polymers (FRP) composites are widely used in aerospace and civil structures due to its unique material properties. However, damage can still occur and typically manifests itself from within the composite material that is invisible to the naked eye. So as to be able to monitor the performance of FRPs, numerous sensing systems have been proposed for embedment within FRP composites. One such methodology involves the embedment of carbon nanotube-based thin films within FRP laminates for strain monitoring and potentially even damage detection. Unlike other sensors, these piezoresistive thin films possess small form factors (and thus do not serve as stress concentration or damage initiation points) and can be easily integrated during composite manufacturing. In this study, a series of laboratory tests have been conducted to characterize the static and dynamic strain sensing performance of these nanocomposites for monitoring glass fiber-reinforced polymer (GFRP) components. Specifically, monotonic uniaxial, cyclic, and fatigue tests have been conducted, while both time- and frequency-domain measurements have also been obtained. The characterization results obtained from this study indicates bi-functional strain sensitivity to monotonic loading until failure, which is found to be reproducible in cyclic dynamic loadings to amplitudes in both functional ranges.

Characterizing the self-sensing performance of carbon nanotube-enhanced fiber-reinforced polymers

The increased usage of fiber-reinforced polymers (FRP) in recent decades has created a need to monitor the unique response of these materials to impact and fatigue damage. As most traditional nondestructive evaluation methods are illsuited to detecting damage in FRPs, new methods must be created without compromising the high strength-to-weight aspects of FRPs. This paper describes the characterization of carbon nanotube-polyelectrolyte thin films applied to glass fiber substrates as a means for in situ strain sensing in glass fiber-reinforced polymers (GFRP). The layer-by-layer deposition process employed is capable of depositing individual and small bundles of carbon nanotubes within a polyelectrolyte matrix and directly onto glass fiber matrices. Upon film fabrication, the nanocomposite-coated GFRP specimens are mounted in a load frame for characterizing their electromechanical performance. This preliminary results obtained from this study has shown that these thin films exhibit bilinear piezoresistivity. Time- and frequency-domain techniques are utilized to characterize the nanocomposite strain sensing response. An equivalent circuit is also derived from electrical impedance spectroscopic analysis of thin film specimens.

Statistical Considerations in the Analysis of Data From Fatigue Tests on Delaminated Cross-Ply Graphite/Epoxy Composites

The objective of this paper is to analyze the results of compressive fatigue experiments performed on a set of delaminated Graphite/Epoxy cross-ply composites. Crack branching, the failure mode we are interested in, occurred during the tests. Due to scatter, it is somewhat difficult to draw conclusions on the values of the branching angles, the key parameter of the problem, unless the tools of statistical and exploratory data analysis (EDA) are used. Here, a brief discussion on some of these techniques is presented, and their application to the set of obtained test data is carried out. The results seem to indicate that the crack grows faster when it is not self-similar, with a higher rate of growth for cracks that branch with a greater angle out of the interface.