hyperbranched polymers used in this study. The evolution of the
distribution has previously been described and the emergence of
“shoulders” in the molecular weight distribution (MWD) is
commonly observed in branched polymers [29]. This is rationalised
by the presence of multiple MWDs with equivalent radii of gyration
due to the random nature of the free radical reaction used to synthesise
these polymers; the macromolecules formed will contain a
range of branching and molecular weights, which cannot be
differentiated by chromatography [29]. Triple detection GPC measurements
confirm the branched nature of the hyperbranched
polymers when compared to linear PMMA, Fig. 1b. In all cases the
MarkeHouwink a-values as determined in MEK are significantly
lower for all three hyperbranched polymers compared with linear
PMMA 310 kDa, a ¼ 0.4 compared with a ¼ 0.69, indicating a more
compact structure and higher molecular density. Branching was
calculated using a Zimm-Stockmeyer model first reported by Frechet
and co-workers which relates the intrinsic viscosities of the
branched fractions to that of the linear reference materials [30],
with the highest molecular weight fractions for the hyperbranched
polymers being the most highly branched through successive
generation. As can be seen in Fig. 1b, HB being more highly
branched compared with both LB and MB over the whole molecular
weight range as evidenced by the lower branching ratio, g’. A full
explanation of the branching calculations is given in S4.
Increasing molecular weight for linear PMMA in MEK from Mw
of 90 kDae468 kDa results in an increase intrinsic viscosity from
0.22 to 0.77 g/dl and a decrease in overlap concentration (c*) from
5.2 to 2.9 g/dl [20]. In contrast LB, MB and HB all have significantly
lower intrinsic viscosities ranging from 0.1 to 0.5 and c* values up to
8.5 g/dL. This has consequences for the viscosity concentration
profiles of hyperbranched polymers when compared to linear
polymers.