Fibre reinforced composites, a combination of reinforcing fibre and matrix, offer many
advantages over traditional materials and therefore found wide application in the aerospace
and construction industry. Among the advantages that traditional thermosetting composites
offer, the most citied are weight saving, high modulus, high strength-to weight ratio, corrosion
and fatigue resistance. Recently, the development of high performance thermoplastics
composites has been accountable for increasing their demand. Whereas the developed
thermosetting composites have created environmental concerns because of hazardous fumes
generated during fabrication of composites and, most importantly, they are non-recyclable.
Lack of previous experience on the durability of thermoplastic composites and the demand for
more understanding of their properties in the planning and design stage is increasing. Such
advances have created in inescapable need for establishing an understanding and comparison
of the mechanical behaviour of thermoplastic and thermosetting composites.
The main aim of this research is to investigate and explore the potential advantages and
feasibility of thermoplastic composites in typical applications that are traditionally deploying
thermosetting composites. In addition, the study also widens the scope of designing
composites in a simple and direct way by industry persons, without the need for complex
mathematical analysis. This study also covers issues of environmental aging, durability and
cost analysis for thermoplastic and thermosetting composites that can serve as a ready
reference for the industry.
The objectives of this research are:
1) to develop a class of thermoplastic-based Boron, Kelvar-49 and Carbon reinforced fibre
composite laminates;
2) to study the mechanical behaviour of the developed composites and explore the
application of theoretical models for predicting the mechanical behaviour of these
composites, and
3) to study the value, benefit and potential application of thermoplastic composites for the
industry.
The selection of fibres for this study was based on an initial study using some simple
theoretical models for predicting the mechanical properties of a large number of fibres with
combination of several thermoplastic and thermosetting matrixes. The fibre and matrix
combinations are chosen such that low, medium and high strength composites are developed
for further study. The three types of fibres selected (Boron, Kevlar-49 and Carbon) are
combined with three types of thermoplastics (Styrene Acrylonitrile (SAN), Acrylonitrile
Butadiene Styrene (ABS) and Polyethylene (PE)) and two types of thermosetting matrices
(Polyimide (PI) and Low Modulus Polyester).
Firstly, the prediction of mechanical behaviour of composites with these chosen combinations
of constituents has been made. Theoretical prediction of mechanical properties of
thermoplastic and thermosetting composites from their constituent's properties has been made
by using the following micromechanical models: (1) Rule of Mixtures (ROM), Halpin-Tsai,
Eshelby's Inclusion model, Modified Rule of Mixtures (Modified ROM) and Self-Consistent
model for uniaxial longitudinal tensile strength; (2) Rosen - extensional mode, Rosen - shear
mode and Xu-Reifsnider Model for longitudinal compressive strength and; (3) Inverse Rule
of Mixture (IROM) and Modified Inverse Rule of Mixtures (Modified IROM) for transverse
strength.
All the tested specimens were prepared in sheet form by modified pultrusion process which is
being widely used in the manufacture of thermosetting composites. The fibre volume fraction
was maintained at 61% which is commonly used for most structural composites. The
experimental work to evaluate the mechanical properties of the developed composites
included uniaxial tension test following British Standards BS 2782 Part 3 Method 321:1994,
compression test based on BS 2782 Part 3 Method 345A:1993 standard, and three-point
bending test following BS 2782 Part 3 Method 335A:1993 standard.
The consolidated results indicate that all experimentally obtained strength values are the
highest for Boron composites when compared with Carbon and Kevlar-49 thermoplastic and
thermosetting composites. The tensile strengths are nearly the same for all composites
prepared with the same fibre but with different matrix materials, indicating that they are fibre
dominated. Their respective compressive strengths are lower than tensile strengths as they
appear to be matrix dependent. The flexural strengths are between the tensile and compressive
values, whereas the flexural modulus values are the lowest among the three types of loading,
except Boron-Polyimide composite having its flexural strength higher than tensile strength.
In assessing the standing of theoretical models which predict the mechanical properties of
Boron, Carbon and Kevlar-49 thermoplastic and thermosetting composites in terms of their
constituents, it is often difficult to discriminate between the validity of assumptions made in
the models and constructional imperfections within the specimens. Results indicate the
theoretically estimated tensile strengths using ROM model are dramatically higher by an
order of magnitude than the experimentally obtained tensile strengths. The tensile strengths
are controlled and dominated by fibre alone and the matrix contribution to their strengths is
negligible. From the compression tests, it is found that micro-buckling approach has made
little success in predicting the compressive strength by Rosen's model whereas Xu-Reifsnider
model has shown good agreement with the experimental values which are only about 10%
lower than theoretically obtained compressive strengths. The matrix properties seem to be the
major contributors of the prepared composites. It is found that the flexural strength values
obtained from the three-point flexural tests are about 14-17% lower than the corresponding
theoretical values using IROM models. These properties have more complex dependency with
respect to the properties of participating fibre, matrix and manufacturing.
Systematic and numerical costing analysis of respective thermoplastic and thermosetting
composites has been made for managing their application. The total product cost is
established based on factors such as material, labour, overhead, selling expenses, general
administrative expenses, taxes as well as net profit. The results indicate that the production
cost is highest for Boron based thermoplastic composites among all the selected composites
whereas thermosetting composites with Carbon are the lowest. From the comparison of cost per unit strength analysis, it is found that Boron-Polyethylene, Carbon-Polyimide and
Boron-SAN gave the best results in tensile properties, flexural properties and compressive
properties, respectively. The outcomes can assist the fabricator to choose suitable composites
in order to suit their specific loading requirement during design stage.
In conclusion, the micromechanical models can hardly be used for exact practical analysis of
composite materials and structures because even the most vigorous models cannot describe
adequately the real composites evolved during manufacturing. Many features such as voids,
microcracks, and randomly damaged and misaligned fibres cannot be formally reflected in
mathematical models. In such circumstances, micromechanical models may be used for initial
qualitative analysis, providing an understanding of how the constituent parameters affect the
mechanical properties followed by experimental determination of these properties.
Unintentional variations in fibre alignments provide unforeseeable complications and
variations to the experimental results and the problems are enhanced in fabric reinforced
blocks. Even with these reservations and complications, the results obtained in the present
investigation fall into a clearly identified pattern and the results obtained are readily
applicable in the composites industry.
| Date of Award | 15 Jul 2011 |
|---|
| Original language | English |
|---|
| Awarding Institution | - City University of Hong Kong
|
|---|
| Supervisor | Pitcheswara Rao KAMINENI (Supervisor) |
|---|