Page 61 - Handbook of Adhesives and Sealants

An Introduction to Adhesives and Sealants 29

Since nearly every adhesive application is unique, the adhesive
manufacturers often do not have data concerning the aging charac-
teristics of their adhesives in specific environments. Thus, before any
adhesive is established in production, a thorough evaluation should
be made in either a real or a simulated operating environment.
With most structural adhesives, strength is more directional than
with mechanical fasteners. Generally, adhesives perform better when
stressed in shear or tension than when exposed to cleavage or peel
forces. Residual stresses inside the joint can also present serious prob-
lems. Such stresses arise from shrinkage due to cure or aging, from
different coefficients of thermal expansion between substrates, and
from other circumstances.
The adhesive material itself should never be used as a structural
substrate. Very heavy bondlines with uneven joint thickness result in
undesirable concentrations of stresses. Many adhesives cure by an ex-
othermic chemical reaction whose intensity is dependent on the mass
of material. Adhesives are generally formulated to cure in thin sec-
tions. Therefore, certain epoxy adhesives, when applied in significant
bulk, could over-heat due to their own crosslinking reaction and, in
fact, burn or degrade when cured in thick sections.

1.4.7 Design limitations
The adhesive joint must be carefully designed for optimum perform-
ance. Design factors include the type of stress, environmental influ-
ences, and production methods that will be used. Many rigid adhesives
do not work well when external stresses act to peel or cleave the sub-
strates from one another. These stresses can often be reduced or elim-
inated by careful joint design. Seldom can a joint, which is designed
for mechanical fastening, be used successfully for adhesive bonding
without revision. Sometimes such revisions result in added expense
or manufacturing steps.
There are no standards to guide the user with regard to design lim-
its or to provide a safe design margin. These will depend on the ad-
hesive and substrate, on the production methods, on the specific
end-use environment, and on many other factors that are often not
foreseen at the time of design development. Therefore, it is very dif-
ficult to predict the useful life of a bonded joint. Simple life estimation
processes used in other industries (such as Arrhenius plots to predict
the aging of electrical wire insulation) are not effective with adhesives
because of the numerous and sometimes competing reactions that can
take place within a bonded joint. The only effective method of esti-
mating the useful life of an adhesive bond is to test prototypes under
environmental conditions that will accelerate the stress on the bond.