Inversion of spatially and temporally resolved temperature measurements yields thermal-diffusivity maps.
A variant of the flash diffusivity technique has been devised for determining the thermal diffusivities, and thus the thermal conductivities, of individual aligned fibers. The technique is intended especially for application to nanocomposite fibers, made from narrower fibers of polyphenylene benzobisthiazole (PBZT) and carbon nanotubes. These highly aligned nanocomposite fibers could exploit the high thermal conductivities of carbon nanotubes for thermal-management applications.
In the flash diffusivity technique as practiced heretofore, one or more heat pulse(s) is (are) applied to the front face of a plate or disk material specimen and the resulting time-varying temperature on the rear face is measured. Usually, the heat pulse is generated by use of a xenon flash lamp, and the variation of temperature on the rear face is measured by use of an infrared detector. The flash energy is made large enough to produce a usefully high temperature rise on the rear face, but not so large as to significantly alter the specimen material. Once the measurement has been completed, the thermal diffusivity of the specimen is computed from the thickness of the specimen and the time dependence of the temperature variation on the rear face.
Heretofore, the infrared detector used in the flash diffusivity technique has been a single-point detector, which responds to a spatial average of the thermal radiation from the rear specimen surface. Such a detector cannot distinguish among regions of differing diffusivity within the specimen. Moreover, two basic assumptions of the thermal diffusivity technique as practiced heretofore are that the specimen is homogeneous and that heat flows one-dimensionally from the front to the rear face. These assumptions are not valid for an inhomogeneous (composite) material.
In the present variant of the flash diffusivity technique, one uses an infrared electronic camera fitted with a microscope lens to record the spatial as well as the temporal variations in thermal radiation emitted from the rear face of the specimen (see figure). In the recorded image of a composite-material specimen, it is possible to distinguish between individual fibers, or between a fiber and the surrounding matrix material. Hence, it is possible to measure the rear-face temperature variations of individual fibers. These variations can be correlated with predictions of a computational model of heat transfer in the composite specimen to obtain a diffusivity map of the specimen.
The technique was demonstrated on a specimen containing pure PBZT fibers, one nanocomposite PBZT/carbon-nanotube fiber, and one copper fiber mounted longitudinally in an epoxy matrix. The copper fiber, having known thermal conductivity, was included for qualitative comparison. The temperature transients of the pure PBZT fibers were not distinguishable from that of the matrix, and the thermal diffusivity of the matrix and PBZT fibers was found to be 0.0032 cm2/s. The thermal diffusivity of the PBZT/ carbon-nanotube composite fiber was found to be 0.049 cm2/s; the true bulk diffusivity of the PBZT/carbon-nanotube composite could be higher than the value computed from the measurements because the heat-transfer model used in the computations does not account for thermal coupling between the fibers and the matrix.
This work was done by Brian Mayeaux, Leonard Yowell of Johnson Space Center, and Hsin Wang of Oak Ridge National Laboratory. For further information, contact the Johnson Innovative Partnerships Office at (281) 483-3809.