Representative failure data for structural epoxies can be very difficult to find for the optomechanical engineer. Usually,
test data is only available for shear configuration at room temperature and fast pull rate. On the other hand, the slowly
induced stress at extreme temperature is for many optical systems the worse-case scenario. Since one of the most
referenced epoxy for optical assembly is the 3M™ Scotch-Weld™ Epoxy Adhesive EC-2216 B/A Gray, better
understanding its behavior can benefit a broad range of applications.
The objective of this paper is two-fold. First, review data for critical parameters such as Young's modulus and
coefficient of thermal expansion. Secondly, derive failure criteria from correlation between a thermal stress experiment
and a finite element model.
Instead of pulling out a standard tensile specimen, it is proposed to test thin bondline geometry to replicate an optical
device usage. Four test plates are assembled at the Institut National d'Optique (INO) in Quebec City, Canada with
bondlines of 50 μm and 133 μm. To detect the failure of the epoxy, the low level vibration signature of a cantilever Invar
plate is monitored as temperature changes. Following the finite element analysis, a failure criterion is found to better
match the experimental results than generic lap shear data.