In part 1 of this series, the concept of a critical material thickness
was introduced and demonstrated experimentally using polystyrene (PS)
as a test material. Below the critical thickness, brittle polymers be
come ductile. The value of the critical thickness is material-dependen
t and related to the entanglement density. The dependence of the criti
cal thickness on the entanglement density was investigated using the m
iscible system polystyrene-poly(2,6-dimethyl-1,4-phenylene ether) (PS-
PPE). PS possesses a low entanglement density and PPE a high entanglem
ent density, and the system's entanglement density can be varied depen
ding on the relative ratio of PS and PPE in the mixture. Equivalent to
the experimental procedure developed in our previous paper, the thick
ness was set by either changing the PS-PPE layer thickness in stratifi
ed PS-PPE/PE tapes (polyethylene (PE) is present to separate the PS-PP
E layers) or by adjusting the volume fraction of non-adhering core-she
ll rubbery particles in the PS-PPE blend, i.e. the ligament thickness.
The experimentally determined critical thickness (ID(c)) proved to in
crease continuously from 0.06 mum for PS-PPE 80-20 to 0.18 mum for PS-
PPE 40-60 blends. This compares well with the value of 0.05 mum found
for pure PS. Under the (moderate) testing conditions used, the PS-PPE
20-80 blend was always tough. The maximum macroscopic strain to break
(lambda(macr)) of the PS-PPE blends correlated with the theoretical va
lue (lambda(max)) based on stretching the entanglement network to its
full extension. The transition from a macroscopic, brittle-to-ductile
deformation behaviour is associated with a change in type of deformati
on mechanism from void formation (e.g. crazing) to shearing, except fo
r the PS-PPE 20-80 blend, which always deforms by shear deformation. A
simple model based on an energy criterion could explain the occurrenc
e of a critical material ligament thickness as well as its dependence
on the entanglement density.