the location on the stress/strain curve where it deviates from linear elastic behavior was 75 MPa. On the other hand, the 0.2% offset yield stress was found to be 89 MPa. The neutron diffraction (ND) experiment was performed on the Neutron Powder Diffractometer (NPD) at the Los Alamos Neutron Scattering Center (LANSCE). Using a hydraulic load frame, the sample was loaded in tension in the presence of the neutron beam as shown in Fig. 1. Data were collected in 20 MPa intervals for approximately one hour at each step. Strain was measured on the sample using an extensometer with a 25 mm gage length. In addition, phase specific elastic strain was determined from the diffraction data via the Rietveld method [2,3]. A 10 mm wide neutron beam struck the entire width of the sample at an angle of 45° resulting in a 14 mm gage length (Fig. 1). Diffraction patterns were collected at +90o and −90° 2θ providing strain information in both the longitudinal and transverse directions. RESULTS AND DISCUSSION Neutron Diffraction Measurements Figs. 2-3 show the applied composite stress vs. axial (longitudinal) strain measured in each phase using ND. Within the error of the ND measurements (±150 µε when specimens are changed and ±25 µε during a single measurement), the thermal residual strains due to the mismatch of the thermal expansion coefficients of the fiber and the matrix were determined to be roughly relaxed. The reference values for the strains in Figs. 2-3 were taken to be the values measured from the monolithic (“stress-free”) versions of the constituents. In other words, the data presented here include the (nearly relaxed) thermal residual strains. Results shown in Fig. 2 suggest plastic deformation in the matrix during the first loading cycle at an applied composite stress around 60 MPa. A jump in the position of the load frame crosshead at around 55 MPa confirms the fiber broke during the first loading cycle. In addition, the plasticity observed from the extensometer (these data are not shown here) supports an assumption of discontinuity in the fiber, as does the change in slope after 60 MPa for both the fiber and matrix. Upon unloading, residual tensile strain is observed in the fiber and residual compressive strain is seen in the matrix. This is likely a result of the development of plastic strains in the matrix. Subsequent X-ray radiography showed separation of the fiber at the notch, which was not observed before loading. In Fig. 2, we see further development of tensile residual strains in the fiber and compressive residual strains in the matrix upon fully unloading after each cycle. We suspect that propagation of localized matrix plastic deformation (around the fiber break) is largely responsible for these residual strains. Furthermore, there is a noticeable change in the composite modulus upon loading and unloading during + 90° Detector Bank Incident Neutron Beam - 90° Detector Bank Tensile Axis Q⊥ Q|| 10 mm 8.2 mm Figure 1. Schematic of the NPD tensile testing geometry. Specimen is at 45° to the incident beam. The scattering (Q) vectors for each detector bank indicate the directions of measured strains.
RkJQdWJsaXNoZXIy MjM0NDE=