Design and test aspects of dental restorations in bonded substrates

Abstract

The recent revolution of dental restorations includes metal-free restorative materials, the digital workflow facilitated by computer-aided design / computer-aided manufacturing (CAD/CAM), and the involvement of artificial intelligence (AI) in restoration design. Among dental restorative materials, lithium disilicate (LDS) and resin-based direct and indirect materials are prevalent for their improved mechanical and wear properties, excellent aesthetics, and ease of application. Indeed, pre-clinical laboratory tests are essential for a comprehensive understanding of material behaviour. There are two vital issues determining the quality and interpretability of laboratory test designs, namely, simulating the oral environment and reproducing clinically significant degradation and failure modes. Furthermore, the quality of AI-generated and CAD-designed prostheses by different personnel is rarely assessed.

This PhD project included a series of studies regarding material properties, design factors and test approaches. The research attempted to conduct in clinically relevant test settings and environments so as to provide a comprehensive evaluation of non-metallic dental restorative materials, laying emphasis on lithium disilicate ceramics and resin-based materials.

The first study presented the mechanical properties, wear resistance, water sorption and solubility, elemental release, degree of conversion, and optical properties of flowable composite materials, such that the nanocomposite resins performed better than the compomer. The study also discovered the correlations among test parameters such as water sorption, solubility, colour stability, and the degree of conversions, and attributed the material performances to their composition features.

The second study intended to evaluate the useability and reliability of dentine analogue materials to substitute human dentine in laboratory fatigue tests. Through a series of experimental (three-point bending test, indentation hardness, and wet cyclic loading), analytical (survival analysis and Weibull statistics) and numerical analysis, the practicability to adopt dentine analogue materials in laboratory fatigue tests was validated, and one material (glass fibre reinforced polyamide) was found to be a suitable dentine analogue material that can (re)produce a high reliability on fatigue test results and a close stress distribution to dentine.

The third study investigated the fatigue behaviour of direct and CAD/CAM indirect restorative materials cemented to a dentine analogue substrate. Flexural properties, energy dissipation parameters, and resistance under wet cyclic fatigue loading were determined. LDS and resin-based materials presented distinct fatigue behaviours; the CAD/CAM indirect composite resin showed the best fatigue resistance among tested materials. LDS and resin-based materials had significantly different energy dissipation capacities that correlate with their fatigue resistance, suggesting that fatigue properties can be predicted by energy dissipation parameters.

The fourth study evaluated knowledge-based AI-generated and human CAD-designed crowns via occlusal morphological discrepancies and fracture resistance. Both knowledge-based AI and human CAD-designed crowns can achieve clinically acceptable fracture resistance, but human can design crowns with higher conformity to the original teeth.

Ultimately, under clinically relevant test methods, settings and loading schemes, laboratory tests of dental restorations are valuable in revealing material behaviour and predicting clinical service conditions, degradation, and lifetime.