Objectives: To demonstrate that determination of the depth of cure of resin-based composites needs to take into account the depth at which the transition between glassy and rubbery states of the resin matrix occurs.

Methods: A commercially available nano-hybrid composite (Grandio) in a thick layer was light cured from one side for 10 or 40 s. Samples were analyzed by Vickers indentation, Raman spectroscopy, atomic force microscopy, electron paramagnetic imaging and differential scanning calorimetry to measure the evolution of the following properties with depth: microhardness, degree of conversion, elastic modulus of the resin matrix, trapped free radical concentration and glass transition temperature. These measurements were compared to the composite thickness remaining after scraping off the uncured, soft composite.

Results: There was a progressive decrease in the degree of conversion and microhardness with depth as both properties still exhibited 80% of their upper surface values at 4 and 3.8 mm, respectively, for 10 s samples, and 5.6 and 4.8 mm, respectively, for 40 s samples. In contrast, there was a rapid decrease in elastic modulus at around 2.4 mm for the 10 s samples and 3.0 mm for the 40 s samples. A similar decrease was observed for concentrations of propagating radicals at 2 mm, but not for concentrations of allylic radicals, which decreased progressively. Whereas the upper composite layers presented a glass transition temperature - for 10 s, 55°C (±4) at 1 mm, 56.3°C (±2.3) at 2 mm; for 40 s, 62.3°C (±0.6) at 1 mm, 62°C (±1) at 2 mm, 62°C (±1.7) at 3 mm - the deeper layers did not display any glass transition. The thickness remaining after scraping off the soft composite was 7.01 (±0.07 mm) for 10 s samples and 9.48 (±0.22 mm) for 40 s samples.

Significance: Appropriate methods show that the organic matrix of resin-based composite shifts from a glassy to a gel state at a certain depth. Hence, we propose a new definition for the "depth of cure" as the depth at which the resin matrix switches from a glassy to a rubbery state. Properties currently used to evaluate depth of cure (microhardness, degree of conversion or scraping methods) fail to detect this transition, which results in overestimation of the depth of cure.