The strength of adhesive joints or bulk polymers is proportional to the work G required to propagate a fracture over a unit area. Thermodynamics provides a lower bound for G > 2g, where g is the interfacial free energy. Typical values of G for polystyrene and plexiglass are several thousand times larger than this bound. The reason is that weak van der Waals bonds are broken throughout a large volume around the crack tip. This produces an intricate network of fibrils and voids called a craze. The craze is stronger than the undeformed polymer, so more and more material is deformed into the craze.
Research in the group has used simulations in different regions (see below) to calculate the macroscopic fracture energy . Other simulations have analyzed the growth of crazes in detail to determine how they grow, the stress required to grow them and how they fail . Work in progress is analyzing the evolution of the entanglement network during crazing to determine its role. You will find movies of craze formation here.
Other research has examined what determines whether adhesives fail at interfaces or in their bulk  and found that yield stresses not free energies determine the location of failure. Work in progress is studying welding of polymer surfaces that are held together at an elevated temperature to produce diffusion.