Add nanoparticles or change the cross-linking structure of the resin, which is more effective to ? is there any relevant literature available? thank you.
Crosslinking can be effective to reduce the linear coefficient of thermal expansion when you compare it with non-crosslinked system. However, polymerized epoxy resin is already highly crosslinked. Therefore, by changing the crosslink structure, you will not expect significant change from the usual, already crosslinked epoxy resin. Instead, adding an inorganic filler, such as silica will significantly reduce the coefficient beyond the already crosslinked epoxy resin since those materials have fairly low coefficient of expansion. This is the reason why epoxy packaging material for computer chip is made of silica filled epoxy. You can reduce the thermal expansion coefficient to as low as 30-40 ppm for alpha 1 coefficient. There is no reason to use nano filler. Rather, nano filler will increase the interphase and the interphase is typically made of lower density material than the bulk. Thus, it will not be effective. Use micrometer size fillers. Play with the particle size distribution (I recommend trimodal distribution) to reduce the viscosity of the filled epoxy so that you can process the filled epoxy relatively easily.
I think, graphene filler could reduce the thermal expansion instead of inorganic filler. as graphene have better thermal conductivity and low coefficient of thermal expansion , which may optimized with flake size distribution, volume fraction also resulting to reduce the viscosity .