Fracture behavior at partially miscible polymer interfaces

Polymer Engineering and Science, May, 2004 by Russell E. Gorga, Balaji Narasimhan

INTRODUCTION

Interfaces and interfacial properties of polymers have been the focus of much academic and industrial research for several decades. The properties of polymerpolymer interfaces are important in many industrial applications, especially with the prevalent use of composites, adhesives, coextruded, and laminated materials (1-3). These materials are considerably affected by their interfacial characteristics. A detailed understanding of interfacial molecular phenomena and how that affects the mechanical properties at the interface in partially miscible polymer systems can have significant impact on the design of many technologically relevant materials. The underlying hypothesis of this work is that molecular properties influence interface performance, and meaningful relationships can be obtained between molecular properties (interdiffusion coefficient, segmental interactions) and macroscopic properties (interfacial fracture energy).

Extensive research has been done in the area of polymer interfaces relating structure to strength (1-6). Most of the studies have focused on either miscible ([chi] < [[chi].sub.s]; [chi] is the Flory-Huggins interaction parameter; [[chi].sub.s] = [chi] at the spinodal) or immiscible interfaces (7-14). Relatively few researchers have examined interdiffusion and fracture energy at annealing conditions such that the polymers undergo phase transitions, i.e., [chi] [approximately] [[chi].sub.s] (15-18). When a partially miscible A/B bilayer is heated above the glass transition temperatures of the two polymers, interdiffusion commences and proceeds until equilibrium is reached. During this annealing process, the interfacial region is an A/B blend (i.e., "interphase") with particular phase morphology. An understanding of this morphology is critical in order to develop structure-strength relationships in partially miscible systems.

When two immiscible polymers are joined together at an interface and annealed, interdiffusion is very limited (i.e. the interfacial width is small). Since the chains do not extend far into the foreign matrix, entanglements are also limited. When the interface is subjected to tensile opening stress (Mode I), the chains will easily "pull out" of the matrix they diffused into (a mechanism called chain pullout), causing the fracture energy to be low (1, 19). When two miscible polymers are joined together at an interface and annealed, significant interdiffusion occurs and the density of entanglements is increased; a combination of chain scission and chain pullout is necessary to fracture the interface, and the fracture energy is high (1, 20-22).

Fracture behavior also depends on the maximum stress the interface can withstand relative to the crazing stress of each polymer (22-25). Crazing is the phenomenon of fibril formation crossing the interface and the crazing stress is the minimum amount of stress needed for this phenomenon to occur (26). Crazing is also seen in bulk materials subjected to stress fields (e.g. elongation or fracture) (1). If the maximum stress the interface can sustain is lower than the crazing stress of both polymers, the fracture energy, [G.sub.c], is low, and brittle fracture occurs. If, however, the stress sustained by the interface is higher than the crazing stress of one of the polymers, plastic deformation mechanisms occur in that polymer, and crazes propagate ahead of the crack tip (27).

A tremendous amount of research has been devoted to polymer-polymer interfacial fracture and the mechanisms mentioned above. Kramer (19) showed that polymers with M < 2[M.sub.e] (where [M.sub.e] is the entanglement molecular weight) will craze, and proposed that since no entanglements exist, the craze is spanned by single molecules. Researchers have modeled fracture mechanisms based on interfacial properties such as the number of bridges across the interface (28) and annealing conditions, or healing time for the interface (29). Several models for chain pullout, derived independently, suggest [G.sub.c] [approximately] n and [G.sub.c] [approximately] [L.sup.2], where n and L are the number of chains per unit area and the length of penetration, respectively (1, 10, 30-32). Most theories use reptation as a basis for interdiffusion and disentanglement (1, 33-35).

Even more extensive research has been done on crazing, with a focus in the last two decades on crazing at incompatible amorphous polymer interfaces (36, 37). Most current thrusts have focused on using block copolymers to reinforce the incompatible polymers (10-12, 14). Xu et al. (12) proposed a deformation map for block copolymers at an interface that shows possible deformation and failure mechanisms based on stress and block copolymer chain density. The copolymers were manipulated (via the symmetry of the blocks) to obtain optimum toughening at the interface. Forward recoil spectroscopy (FRES) was used to characterize chain interpenetration by identifying deuterated block species on either side of the fracture surface and transmission electron microscopy (TEM) was used to identify mechanisms of crazing (13, 14). Some attention has also been given to the use of random copolymers to reinforce incompatible interfaces (38).