There is a wide-ranging array of applications where static loading at or near
ambient conditions represents the governing design condition. It is often the
case, however, that the operating environment is somewhat more arduous than this
and may involve cyclic stresses, elevated temperatures or exposure to media
other than air. It may be possible in an initial scoping study to carry out
preliminary calculations using data obtained at, say, room temperature and then
to exercise judgement based on past experience, but at some point in the
analysis due account must be taken of the influence of other effects. This is
not to say that composites perform badly in these respects when compared with
alternative materials, as in a number of instances they are selected in
preference to others for reasons concerned with their good behaviour. For
example, GRP is noted for its corrosion resistance and is used in preference to
costly alloy or lining systems and the fatigue response of CFRP is regarded as
being excellent. At the detailed level the mechanisms that control behaviour,
property change and, ultimately, design life are fundamentally different from
those that occur with metals and an appreciation of the issues concerned is
important in the evaluation of component performance.
Chemical media
- The degradation of composite materials under the influence of an
aggressive environment can result from a number of factors:
- Loss of strength of reinforcing fibres by stress corrosion.
- Loss of bond strength through degradation of the interfacial fibre/matrix
bond.
- Chemical degradation of the matrix.
- Accelerated degradation caused by the combined action of temperature and
chemical environment.
As a result of each of these factors, acting singly or in combination,
mechanical properties can be adversely affected. The nature of corrosion
phenomena is entirely different from that observed with metals where
electrolytic effects may predominate. For composites the physical processes of
diffusion and osmosis are often the more important mechanisms to be considered.
With regard to the effects of environment on the reinforcing fibres it has been
demonstrated that under certain circumstances delayed failure under load or
"static fatigue" can have a pronounced effect. This features
particularly with glass fibres where it has been concluded that initially
non-critical cracks grow under the influence of stress and the reactive
environment until they are sufficient to cause failure. The figure below shows
the stress rupture behaviour of E-glass strands in air, water and distilled
water and as can be seen the fall off in strength is considerable, although it
should be noted that at normal design levels (design strains < < 0.5%) it
may not be a significant issue.
Stress rupture of polyester impregnated E-glass
The behaviour of other fibres is shown below.
Time-dependent strength of various fibre types in water
Carbon fibres are essentially unaffected, whereas aramids fall somewhere
between the two extremes. Because of their nature, polymer matrix materials can
have a comparatively rapid uptake of environmental agents such as water due to
diffusion processes. The most common technique for the modelling of diffusion is
to consider Fick's law :
where dc/dx is the concentration gradient and Dx is
the diffusion gradient in direction x. It can be shown that the mass absorbed in
time t, Mt can be expressed as :
where M is the mass absorbed at saturation and h is the
thickness of the plate of concern. The figure below shows the absorption cure
for water uptake for a glass reinforced epoxy indicating classical behaviour as
given by this equation.
Moisture absorption of glass reinforced epoxy
Interfacial bond strength can also be affected by ingress of
moisture and other reagents. In fact, capillary action along the fibres can
account for a significant proportion of uptake in the first instance. In general
terms, the effect of moisture is to cause hydrolytic breakdown of the fibre
matrix bond which will in turn affect the efficiency of load transfer between
the phases. In some cases this effect is reversible, whereupon on drying out,
the composite properties return to their original values. The figure below shows
the load deflection curves for three-point notched bending tests carried out
during/after boiling and dried after boiling.
Load deflection curves for notched unidirectional GRP
The effects of drying are to return the material almost to its
original state. This behaviour is not repeated over a large number of cycles
owing to progressive irreversible damage of the laminate. It is difficult to
generalize with respect to the overall response effects of composites under the
action of chemical media because of the vast number of material and reagent
combinations possible.
In the fig. below the results of stress corrosion tests of GRP
filament wound pipe under acid conditions are plotted.
Stress corrosion of GRP pipe
Results are given for both unidirectional (hoop) and angle-ply
laminates ( ± 55°) and are expressed in terms of the fibre stress sl.
Plotted in this way, there is a common relationship between fibre stress and
time to failure. The displacement of the plots is thought to be due to initial
mechanical damage which allows more rapid ingress of the acid. The figure below
shows times to failure of a different laminate type, chopped strand mat (CSM),
for a range of chemical media and, as can be seen, the picture is not
straightforward. Also plotted is the maximum tensile strain experienced during
the test and it is worth noting that typical design' values lie between 0.2% and
0.4%. The chemical process plant industry, for example, uses 0.2% as a norm.
Effect of applied stress and strain on the time to failure of
chopped strand mat (CSM)
Back...
| 
