Types of composites in use today include Metal matrix composites, Ceramic matrix composites, Carbon matrix composites, Polymer matrix composites and Hybrid composite materials. There may be potential for other types of composites to be used in a carabiner, however, most successful aluminium to composite transitions, where weight reduction is the aim, have made use of fibre-reinforced polymer matrix composites (e.g. bicycles, sports rackets, aerospace structures etc.). It is not feasible to cover every option in the scope of this project; therefore the focus will be on fibre-reinforced polymer matrix composites only.
This section will explore the material options for the fibres and the matrix, these options will be measured against the criteria set out in the previous section. The ideal material combination will have the highest strength to weight ratio, a good stiffness to weight ratio and will satisfy the requirements above. Properties of aluminium 7075 T6 are given below (table 3) as a benchmark for comparison. The properties of a composite depend on the combined properties and interaction of the matrix and reinforcement; it is not possible to predict the properties of a composite based on the material properties of the constituent parts alone. This is because the final properties of the composite also depend on the processing conditions, the method of manufacture, the geometry and preparation of the fibres (this dictates the properties of the matrix/fibre interface) however, some aspects of the contribution of each part to the overall properties can be separated.
Table 3: Properties of aluminium 7075 T6 [20-22].
Fracture toughness , K IC
MPa . m 1/2
Price (approximate for 7075-T651)
In a fibre reinforced polymer (FRP) the reinforcement fibres are generally very strong, stiff materials. Their high tensile strength comes from their molecular orientation and small cross section, typically carbon fibres have a diameter between 4 and 11 Ám . Due to their small diameter the maximum defect size in fibres is much smaller than in their monolithic counterparts. In very stiff materials, where failure is caused by brittle fracture, strength is largely determined by the size of defects present. Therefore reducing the maximum defect size will significantly increase strength; however, as they are so small and brittle they are vulnerable to damage. During processing and handling fibres can become damaged due to abrasion. In an FRP the fibres are protected from abrasion and environmental effects by the surrounding matrix. The matrix gives the composite its durability, shape and appearance, whilst the fibres dictate the overall stiffness and strength.
Table 4: Properties of composite reinforcing fibres . In order; tensile modulus, failure stress, failure strain, density, specific modulus, specific strength.
Figure 11: Comparing specific strength and specific modulus for the most common fibre types .
Fibres with a large strain to failure improve impact resistance for high energy impacts . However, a trade-off must be met between stiffness and ductility. Fibres that have greater strain to failure have a lower modulus and therefore impart less stiffness to the composite structure. It is important for a carabiner that it remains functional under load (see requirements), in order to remain functional a carabiner must not deform excessively or else the gate will no longer open.
From table 4 and figure 11; glass fibres are inexpensive, have high strain to failure and moderate specific strength but very poor specific stiffness. Aramids have good specific strength, high strain to failure but poor specific stiffness. Boron is very expensive, has moderate strain to failure, moderate specific strength and good specific stiffness. Carbon fibres have the broadest spectrum of properties - they can be made to have a very high modulus at the expense of strength, or vice versa, however the very high modulus fibres also have very low strain to failure. High strength carbon fibres have good specific stiffness, moderate strain to failure, good specific strength and are significantly less expensive than high modulus types. Ultra-high molecular weight polyethylene fibres are shown in figure 11 but not in table 4, they can have a specific strength and stiffness significantly superior (3.1 MJ/kg and 177 MJ/kg respectively ) compared to high strength carbon fibres, however they also have a low melting point (135°C ). The temperatures needed to cure epoxy resins range from 120-180°C , typically, high performance applications will use epoxies that require curing in the top end of that range. Commonly used thermoplastic matrices require even higher processing temperatures, for example polyimide has a melting point of around 375°C, polyamide of around 240°C and PEEK of 322°C . Although they have remarkable properties, polyethylene fibres are not compatible with the high temperatures required for most polymer based composite processing.
In a composite the matrix acts to transfer the load to the reinforcing fibres, it tends to be far less stiff than the reinforcement (usually by more than an order of magnitude), therefore it does not carry a significant part of the load. The matrix affects interlaminar shear strength which is crucial for applications where there will be bending loads . In discontinuous fibre composites the ability of the matrix to transfer loads is more critical. The matrix determines the environmental resistance, friction/wear properties and has a significant effect on the toughness of a composite . According to the design criteria the carabiner matrix material should be tough, low friction and resist water, salt water, UV radiation, mild acid, alkali and other chemical attacks. Ideally it should also be reasonably inexpensive.
When it comes to polymer matrices there are two categories to choose from, thermoset or thermoplastic (rubbers are inappropriate for a carabiner). By definition the difference between the two is simply that, due to cross linking between molecules, thermosets cannot be re-melted and reprocessed once they have been cured. Thermosets generally have a low viscosity before they are cured, whilst thermoplastic melts have very high viscosity. This makes it difficult to properly impregnate the reinforcement fibres - usually requiring high pressure and temperature. Broadly speaking thermosets are stiffer, are more resistant to chemical attack and can maintain their properties at higher temperatures (table 5) due to cross-linking. However, cross-linking raises their glass transition temperature above room temperature - making them brittle.
The choice of matrix strongly affects the threshold energy for impact damage ; furthermore the threshold energy is not affected by the type or layout of the fibres. As a result it is very important to use a tough matrix. Thermoset matrices, such as epoxy, can be toughened with rubber particles. However epoxies are susceptible to UV degradation and also adsorb moisture - which decreases their physical properties.
Table 5: Comparing properties of thermosets and thermoplastics. Reproduced from .
Young's Modulus (GPa)
1 . 3 - 6 . 0
1 . 0 - 4 . 8
Tensile Strength (MPa)
20 - 180
40 - 190
Kic (MPa . m 1/2 )
0 . 5 - 1 . 0
1 . 5 - 6 . 0
Gic (kJ/m 2 )
0 . 02 - 0 . 2
0 . 7 - 6 . 5
Maximum service temperature (°C)
50 - 450
25 - 230
Thermoplastics have a longer shelf life (in prepreg form) and generally have a shorter processing cycle time , for high volume production the cycle time is one of the major cost factors. Thermoplastics can be amorphous or semi-crystalline; crystallinity affects the strength, stiffness, toughness, and resistance to solvents and environmental attack. The degree of crystallinity is determined by processing conditions and the cooling time after processing. Slow cooling imparts higher crystallinity and therefore greater stiffness. It is very difficult to cool a mould completely evenly; in amorphous polymers the cooling rate has less effect on shrinkage, whilst in crystalline polymers warpage is more likely to occur because crystallinity, and therefore shrinkage, will be greater where cooling is slower. The polymer will crystallise less where it cools more quickly, resulting in differential shrinkage - which causes warping. Careful control of processing temperatures and cooling times can be used to maximise damage tolerance and impact resistance , faster cooling rates make a crystalline polymer more ductile, impact resistant and damage tolerant. However, this comes at the expense of stiffness and other properties, therefore a compromise must be met.
To aid the material selection process a material selection database such as Cambridge Engineering Selector (CES Edupack 2007) is helpful. The CES database includes extensive information on most commercially available materials and provides access to CAMPUS data for specific brands and grades of polymer. However, the database is by no means comprehensive; some grades are not included and some property data is incomplete or is provided as an estimate only. Nonetheless, CES is useful as a guide to aid the selection process.
CES allows limits to be applied to a large range of material properties. In addition it can compare materials by graphical means such as Ashby charts. Property limits can be used to systematically reduce the polymer database to a suitable shortlist of materials. By varying the values of important criteria a general idea of what is a 'high' or 'low' value can be obtained. This process involves raising the requirement of a property, whilst leaving all other properties without any requirements, to the point where the number of materials that pass the limit become small (one or two materials). In this way the highest possible expectation of each desired property can be found. Once the highest possible expectations are found, they can be entered into CES and gradually relaxed until a suitable number of materials passed the limits. An appropriate material can then be chosen from this shortlist. Specific details of the selection criteria and process are described below.
First, the CES database was set to use polymeric materials only, this includes both pure polymers and polymer based composites (in both continuous fibre laminate and short random fibre forms). Initially, no limit was set on toughness, in this way the maximum feasible value can be ascertained when all other criteria are set to minimum acceptable values. CES uses five descriptive classes to judge the environmental resistance and durability of polymers; the classes are: very poor, poor, average, good and very good. In agreement with the design guidelines set out in the design criteria section a 'very good' requirement was set on the following durability criteria: fresh water, salt water, weak acid, weak alkali, organic solvents and UV radiation. In addition, the maximum service temperature was set to be at least 60°C and the minimum service temperature was set to be at most -40°C. Table 6 summarises these selection criteria. These requirements cut down the number of passable materials in the CES database from 637 to 25.
Table 6: Selection criteria used in CES to make a shortlist of passable materials.
Fresh water resistance
Salt water resistance
Weak acid resistance
Weak alkali resistance
Organic solvent resistance
UV radiation resistance
Maximum service temperature
= 60 °C
Minimum service temperature
= - 40 °C
The next step was to optimise the toughness by setting further limits. In CES both fracture toughness and notched impact energy (Izod) are used to define room temperature toughness. Izod energy gives a way of approximately ranking materials by critical strain energy release rate , but the ratio of and depends on stiffness (eq. 1) .
Where E is Young's modulus and v is Poisson's ratio. Polymers in general have a relatively low stiffness, therefore using (based on Izod) alone may not be a good indication of toughness. For this reason was also included in the selection criteria. This will ensure that polymers with a high Izod toughness but disproportionately low fracture toughness (due to low modulus) will not be able to pass the limits.
First, by trial and error, the limit was found for fracture toughness that allowed through only the top 50% (i.e. 12 results) of the 25 previously passable materials - the value found was 5.5 MPa.m 1/2 . This limit was then removed and, by the same means, the corresponding value for Izod impact strength was found to be 25 kJ/m 2 . Finally, both of these limits were applied simultaneously. The final seven passing results (table 7) consisted of various forms of PEEK based composite, two ETFE based materials, unfilled ECTFE and random oriented glass fibre filled polyester liquid crystal.
Table 7: The final 7 results that passed the CES selection criteria set out in table 6, plus these additional; a K IC of at least 5.5 MPa.m 1/2 and a notched Izod of at least 25 kJ/m 2 .
Further increases in the toughness requirements led to all materials being excluded apart from the carbon fibre PEEK composite laminates. However, it is not necessarily fair to exclude the other materials from consideration because they are in the unfilled form or filled with random short fibres, whilst the PEEK is filled with unidirectional continuous fibres. It is not obvious whether the excluded materials would be better or worse as there are no laminate forms of ETFE, ECTFE or PLC available in the database. The fact that they don't exist in that form may suggest that either, due to incompatibility with fibres, it is difficult to produce, or the resulting properties are not very desirable. Otherwise it is possible that it has simply been overlooked.
The list of materials in table 7 has been produced using criteria for the matrix only, it is also important to consider the strength and stiffness requirements. Those materials in the list that are unfilled and those with short random oriented fibres would have to be combined with continuous reinforcements to attain the necessary strength and stiffness. However, using any of the listed matrix materials in combination with a high volume fraction (50-60%) of intermediate modulus carbon fibres will produce a material that exceeds the specific strength and stiffness of current carabiner material by a large margin. The challenge comes in finding the optimum combination of materials to maximise toughness. This section has studied the available materials with a view to finding materials that satisfy the toughness, strength, durability and other requirements defined previously.
Material combinations in relation to the available manufacturing processes will be considered in the design section later. The next section describes the range of available composite manufacturing methods and relates them to the design criteria and requirements.