Effect of Sample Preparation on the Glass-Transition of
Thin Polystyrene Films
Z. FAKHRAAI,
1
J. S. SHARP,
2
J. A. FORREST
1
1
Department of Physics and Guelph–Waterloo Physics Institute, University of Waterloo, 200 University Ave. W.,
Waterloo, Ontario, Canada, N2L 3G1
2
School of Physics and Astronomy, University of Nottingham, University Park, Nottingham, NG7 2RD, U.K.
Received 1 June 2004; revised 12 August 2004; accepted 15 August 2004
DOI: 10.1002/polb.20281
Published online in Wiley InterScience (www.interscience.wiley.com).
ABSTRACT We have investigated the effect of sample preparation on the glass-transi-
tion temperature (T
g
) of thin films of polystyrene (PS). By preparing and measuring the
glass-transition temperature T
g
of multilayered polymer films, we are able to assess the
contribution of the spincoating process to the reduced T
g
values often reported for thin
PS films. We find that it is possible to determine a T
g
even on the first heating cycle, and
that by the third heating cycle (a total annealing time of 15 min at T = 393 K) the T
g
value has reached a steady state. By comparing multilayered versus single layered
films we find that the whole T
g
depends only on the total film thickness, and not on the
thickness of the individual layers. These results strongly suggest that the spincasting
process does not contribute significantly to T
g
reductions in thin polymer films. © 2004
Wiley Periodicals, Inc. J Polym Sci Part B: Polym Phys 42: 4503– 4507, 2004
INTRODUCTION
There has been a significant amount of interest in
recent years in the physical properties of very
thin polymer films.
1–3
In particular, in a number
of cases the properties of thin films display signif-
icant deviations from that of the bulk material, or
even that of much thicker films. Of particular
interest is the glass transition temperature (T
g
).
For many different polymers, measurements of
the T
g
in thin films have revealed large differ-
ences from the bulk value. Of these polymers, the
most well studied material is atactic polystyrene
(PS). Since the original study by Keddie et al.
4
showing thin film T
g
values lower than that of the
bulk, there have been many studies aimed at
quantifying this behavior.
5
A recurring concern
about these and other studies of thin polymer
films involves the effect of sample preparation.
6, 7
Such films are typically spin cast from dilute so-
lutions of polymer in a good solvent (such as tol-
uene in the case of PS) and then annealed above
the bulk T
g
value for a time usually ranging from
a few hours to a few tens of hours. During the
process of spincasting, the chains may be in con-
figurations not typical of those in equilibrium
melts. In addition, even though the films are an-
nealed, in many cases (depending on the sub-
strate on which they are prepared) they are very
different from the equilibrium state of a fully
dewetted film. This concern is often dismissed,
although there is not yet any proof that it is valid
to do so. Although a definitive explanation to the
observed T
g
reductions has not yet emerged, there
is compelling evidence that it is related to the
presence of the free surface.
8
One way to distin-
guish between an effect due to the free surface
Correspondence to: J.A. Forrest. (E-mail: jforrest@
uwaterloo.ca)
Journal of Polymer Science: Part B: Polymer Physics, Vol. 42, 4503– 4507 (2004)
© 2004 Wiley Periodicals, Inc.
4503