Nonisothermal Fusion Bonding in Semicrystalline
Thermoplastics
Christopher J. G. Plummer, Pierre-Etienne Bourban, Jean-Emile Zanetto, Gregory D. Smith,
Jan-Anders E. Månson
Laboratoire de Technologie des Composites et Polyme `res, Ecole Polytechnique Fe ´de ´rale de Lausanne,
CH-1015, Switzerland
Received 11 November 2001; accepted 16 May 2002
ABSTRACT: The development of the interfacial bond
strength as a function of bonding conditions has been inves-
tigated in two representative semicrystalline thermoplastics,
isotactic polypropylene and polyamide 12. If one side of the
interface is well above the melting point immediately before
contact, more rapid effective bonding is obtained for a given
estimated interface temperature than under isothermal con-
ditions. This is discussed in terms of a simple two-parameter
model for the critical strain energy release rate associated
with crack propagation along the interface, which incorpo-
rates the rate of establishment of intimate contact at the
interface. The model provides a self-consistent phenomeno-
logical description of the time and temperature dependence
of the bonding kinetics in polyamide 12 joints, although
questions remain regarding the detailed mechanisms of
bonding. © 2002 Wiley Periodicals, Inc. J Appl Polym Sci 87:
1267–1276, 2003
Key words: polyamides; poly(propylene) (PP); fracture;
toughness
INTRODUCTION
Industrial fusion bonding falls into three broad
classes, depending on whether the heat required to
provoke melting at the interface is (1) introduced from
an external source, (2) generated in situ by friction or
radiation, or (3) available from a prior forming oper-
ation.
1–6
This last category is of particular relevance to
integrated processing, currently under development
in our laboratory for the production of multicompo-
nent parts that combine the advantages of composites
and neat resins.
7–9
The rational design of a sequence of
forming operations will involve optimization of the
total power consumption. Any heat (or mechanical
work) introduced at a given stage of the process
should, therefore, be exploited to the greatest possible
extent during subsequent operations. If an injected
component is to be fusion-bonded to a pre-existing
component, for example, it may be advantageous to
combine the two operations in a single overinjection
step, rather than reheat the interface at a later stage. In
the former case, joining occurs under nonisothermal
conditions; that is, the surfaces to be joined are not at
the same temperature immediately before contact.
7
This provides the motivation for this study, which
focuses on the specific problem of nonisothermal fu-
sion bonding, as previously defined, by comparing
experimental results for two representative systems
and introducing a new phenomenological model for
data interpolation. The model is then used to derive
simple processing maps for the optimization of over-
injection within an integrated process. In the experi-
mental investigation, both nonisothermal fusion bond-
ing and fusion bonding under isothermal conditions
have been considered. The latter is of less relevance to
integrated processing, but it provides a useful refer-
ence point when we consider nonisothermal bonding
because isothermal bonding conditions are relatively
well defined.
10 –13
Moreover, it is hoped that such a
comparison may further our understanding of the
factors that dominate the development of joint
strength in each case and, therefore, prepare the
ground for controlled experimental studies of the un-
derlying mechanisms.
In the past, fundamental studies of the thermally
induced bonding of thermoplastics under well-de-
fined conditions have tended to focus on amorphous
systems and bonding temperatures just above the
glass-transition temperature, T
g
.
14 –16
Such studies
now often involve the characterization of the resis-
tance to mode I crack propagation along the locus of
the original interface with a linear elastic fracture me-
chanics (LEFM)-based test. The results are then ex-
pressed as a variation of the critical stress intensity
factor, K
c
, or the critical strain energy release rate, G
c
,
with the bonding time and temperature. It is usually
Correspondence to: J.-A. E. Månson (jan-anders.manson@
epfl.ch).
Contract grant sponsor: Swiss Progamme Prioritaire en
Mate ´riaux.
Journal of Applied Polymer Science, Vol. 87, 1267–1276 (2003)
© 2002 Wiley Periodicals, Inc.