No abstract

The mechanical integrity of a structural composite is strongly affected by the strength and toughness of the fiber-matrix interface/interphase [1], with interfacial shear strength (IFSS) being generally accepted as the best quantifying metric. The value of the IFSS is not directly measurable, but it can be approximated by several micromechanics based test methods with the value obtained being dependent on the choice of the model. The most popular of these test methods is the embedded single fiber fragmentation test (SFFT) which provides the experimental data needed to estimate the IFSS: (a) mean fragment length at saturation and (b) fiber strength at the critical fragment length.Because the IFSS is used in unidirectional composite models to predict strength and failure behavior, where the interaction between fibers can be important, the validity of extrapolating from test results based upon the repeated failure of a single isolated fiber has often been questioned. In this paper, the spatial distribution of fiber breaks in a 2-D array of glass fibers is compared with break locations observed from SFFT specimens. In both cases, the break locations in each fiber were found to evolve to a uniform distribution, thereby confirming that the ordered fragment lengths from the repeated fracture process conforms for both SFFT and multi-fiber fragmentation test (MFFT) specimens to a cumulative distribution function (CDF) derived by Whitworth [2][3][4][5]. The array break density was also observed to be less than the break density in isolated fibers, and break locations across array fibers were observed to be highly coordinated and mostly aligned.

Precise measurements of fiber break regions have been made during the single fiber fragmentation test (SFFT) procedure on E‐glass/diglycidyl ether of bisphenol‐A (DGEBA)/meta‐phenylenediamine (m‐PDA) test specimens. From these measurements, the location and size of each fiber fragment was determined, and the resulting information was used to construct fragmentation maps of the tested fiber. By comparing these maps, the fragmentation process supports random fragmentation along the length of the fiber. Since the interfacial shear strength (IFSS) or the interfacial shear stress transfer coefficient (I‐STC) is obtained from the fragment length data at the end of the test (saturation), frequency histograms of the fragment length data were constructed to determine the repeatability of the fragmentation process. Since the SFFT is performed by sequential step‐strains of the test specimen, test protocols were developed by controlling the step size of each strain increment and the time between each step‐strain (dwell time). For the testing protocols used in this research, the E‐glass/DGEBA/m‐PDA frequency histograms of the fragment lengths were found to be generally repeatable. However, when the effective strain rate of the test was altered by changing the dwell time between strain increments, the fragment distribution at saturation of the E‐glass/DGEBA/m‐PDA SFFT specimens changed. The direction of the change was found to be inconsistent with the effect one might expect when only the nonlinear viscoelastic behavior of the matrix is considered. However, the magnitude of the change observed in the E‐glass/DGEBA/m‐PDA SFFT specimens is not universal. Fragmentation data obtained on E‐glass/polyisocyanurate SFFT specimens revealed a much smaller change in fragment length distributions with the same change in testing protocols. Consistent with the results obtained on the E‐glass/DGEBA/m‐PDA, fiber fragmentation occurs when the polyisocyanurate matrix exhibits nonlinear viscoelastic behavior. The implication of these results for interfacial shear strength measurements is discussed.

No abstract

A minimally invasive testing methodology was developed to monitor the in-service performance of soft body armor. In the spirit of the single fiber test standard, ASTM C 1557-03, the fiber diameter was measured at five equally spaced locations along a 6-cm gauge length specimen. In addition, the single fiber test specimen was modified by placing reflecting tape just outside the specimen gauge length to allow the use of a laser extensometer for directly measuring fiber displacement and hence, strain. This modified testing methodology was found to be reproducible and provide data that was normally distributed based on descriptive statistics when variations in the fiber diameters along the length of the specimen were considered. The abnormality in the data, identified using Pierce's outlier criterion, may be associated with processing variations during the fiber manufacture and/or the woven fabrics and not associated with the testing methodology. Furthermore, the failure strain was found to be drastically reduced when the minimum fiber diameter along the length of the test specimen was less than 11 lm. This suggests that a methodology that accurately profiles fiber diameter changes along the gauge length of the fiber may be useful in analyzing single fiber test results.