The influence of system compliance and sample geometry on composite polymerization shrinkage stress

The objective of this study was to use finite-element analysis to model tensilometer tests of polymerizing dental composites. A typical sample in polymerization shrinkage stress tests is shaped like a flat disk, that is, has a high aspect ratio (ratio of diameter to height). In the experimental literature it is implied that the induced stress state in the flat disk composite samples is uniaxial. Three published tensilometer tests of curing dental composite samples with similar high aspect ratios (varying from 3 to 5) were modeled, but with test configurations having low, intermediate and high relative compliance (a tenfold variation). With the use of linear elastic finite element analysis, an instantaneous volumetric shrinkage of 1% was applied to the composite via the thermal analogy and the following questions were addressed: 1. Does the numerically predicted state of stress in composite samples tested in this fashion differ substantially from the uniaxial stress state assumed in the experiments? 2. How do the numerically predicted stresses compare with the experimentally determined nominal stresses? 3. Does compliance of the mountings influence the numerically predicted stress state? The finite-element results predicted a complex triaxial stress state that was strongly influenced by the compliance of the mountings. For the low and intermediate system compliance the model overpredicted the polymerization contraction stress, as would be anticipated due to the inability of the model to account for viscoplastic flow. For high system compliance, the numerical and experimental stress values were in better agreement, mainly because the linear elastic mountings accounted for most of the measured system compliance.