Thermal expansion mismatch consideration

Thermal expansion mismatch between the reinforcement and the matrix is an important consideration. Thermal mismatch is something that is difficult to avoid ia any composite, however, the overall thermal expansion characteristics of a composite can be controlled by controlling the proportion of reinforcement and matrix and the distribution of the reinforcement ia the matrix. Many models have been proposed to predict the coefficients of thermal expansion of composites, determine these coefficients experimentally, and analy2e the general thermal expansion characteristics of metal-matrix composites (29-33). 

The calculation of residual stresses requires consideration of the thermal expansion mismatch between the liquid and the glass when the actual structural state is to be determined as described above. Assuming equibiaxial stress field... 

Another consideration is the difference in thermal expansion between the matrix and the reinforcement. Composites are usually manufactured at high temperatures. On cooling any mismatch in the thermal expansion between the reinforcement and the matrix results in residual mismatch stresses in the composite. These stresses can be either beneficial or detrimental if they are tensile, they can aid debonding of the interface if they are compressive, they can retard debonding, which can then lead to bridge failure (25). 

Similar to the end sealing issues mentioned above, some module packing considerations are mismatching of the thermal expansion coefficients and chemical compatibility. Shown in Table 9.5 is the comparison of the values of the thermal expansion coefficient... 

Many fectors must be taken into consideration in designing an adhesive. The requirements include low level of ionic impurities, no voids under the chip caused by evaporation of solvent or other volatiles, no resin bleed during cure, and thermal expansion properties that match those of the substrate and chip. A significant mismatch in the thermal expansion coeflScient can lead to development of thermal stresses that can result in cracking or distortion of the chip. This problem is becoming more and more important as die sizes continue to increase. 

Given that the membranes and modules are often complex engineered structures, it is not immediately obvious if design problems exist, and, therefore, careful consideration is necessary. For instance, a very thin dense metal membrane must be supported by an underlying porous (or other) structure if the membrane is to be used under significant pressure differentials. This is true for thin metal foils as well as for thin permselective metal layers that are deposited onto a porous support (such as by vapor deposition or plating processes). In the latter case, it is important that the coefficient of thermal expansion of the membrane and of the porous support be similar - a large mismatch will result in membrane failure after a relatively few start-stop cycles. 

There is a considerable mismatch between the thermal expansion coefficients of metals and those of organic polymers. The metal particles can sometimes shrink away from the surrounding polymer as the temperature falls from the moulding temperature. Improved bonding between metal and polymers would be beneficial. It is possible to obtain filled plastics with thermal conductivities up to 95% of those of metals, simply by using high aspect ratio aluminium flakes. 

Thermal property evaluation took place with respect to thermal expansion, diffusivity, specific heat, and ultimately thermal conductivity. Instantaneous CTE data at 20°C is presented in Figure III. The thermal expansion of these materials is an important consideration to take account of, as many applications require matching CTE s to help reduce thermal mismatch stresses during cycling. Results are plotted with a rule of mixture model as well as the Turner and Kemer models for CTE. These are shown in Equations 5, 6, and 7 respectively. These predictions were also based on the SiCrSi system, with the property inputs provided in Table II. 

---