Glass matrix composites for transparent security measures
Transparent glass matrix composites for use in high resistance structures and security windows are being explored by a team at Imperial College London. Aldo Boccaccini, David McPhail and Bo Pang outline the research.
The addition of ceramic fibres or particles – carbon or silicon carbide – to a glassy borosilicate matrix can produce sizable improvements in performance. These materials are called glass matrix composites, and applications include structural components for high temperature and impact resistant panels.
However, composites produced using those fibres by conventional hot-pressing methods are not transparent. The development of transparent glass matrix composites, containing oxide fibres as reinforcement, aims to achieve inorganic composites that combine transparency and enhanced thermomechanical properties, called ‘opto-mechanical composites’. These materials will have applications in security glazing, armour structures, monitoring windows and impact resistant transparent panels.
The work at Imperial College London has specialised in processing and characterising a range of opto-mechanical composites with a focus on alumina fibre-reinforced borosilicate glass matrix composites. To improve the fracture toughness, zirconium dioxide coated fibres are being considered as the reinforcement.
Fibre reinforced glass matrix composites are typically fabricated by the sandwich hot pressing technique. This method involves arranging the reinforcing fibres between glass slides and subjecting the sandwich structure to a thermal treatment, as shown schematically:
The sandwich hot pressing technique for fabricating transparent fibre reinforced glass matrix composites (Diagram adapted from Optomechanical Glass Matrix Composites, by A R Boccaccini, S Atiq, G. Helsch. Comp. Sci. and Technol. 63, 2003, pp779-783)
In a typical experiment, the material is heated at 500°C for two hours to burn out organic impurities. The temperature is then increased again to a value exceeding the glass transition temperature (Tg) of the borosilicate glass. The final temperature and holding time must be carefully determined by trial-and-error to produce composites with sufficient structural integrity.
Viscous flow of the glass should occur around and between the fibre bundles. This will lead to strong bonding of the glass slides without losing the transparency and geometry of the composite structure. The processing parameters are chosen according to the properties of both the fibres and the glass matrix used. In the present developments, composites with different fibre interspacing, such as four, five and six millimetres, are being fabricated for light transmittance and mechanical property measurements.
The method commonly used for the sol-gel synthesis of zirconia uses zirconyl chloride octahydrate, oxalic acid and yttrium (III) nitrate hexahydrate as precursors:
Above: Schematic diagram of the zirconium oxide sol-gel coating process
Below: Borosilicate/zirconium oxide interface sample prepared by focused ion beam secondary ion mass spectroscopy
These chemicals are dissolved separately in water to promote molecular-level mixing between all the constituents of the reaction. In a typical experiment, oxalic acid solution is slowly poured into zirconyl chloride octahydrate solution and mixed under constant stirring. When a sudden increase in viscosity is observed the viscous sol is subjected to intense agitation until a clear sol is formed. Fibre bundles are then immersed.
The air inside the fibre bundle can be evacuated using a vacuum pump. As the air is extracted, the sol is seen to bubble. This enables effective infiltration of the sol into the bundles, which are then removed and dried at room temperature.
Coated fibres are heat treated at 500°C for three hours at a slow heating rate of one degree per minute to diminish surface cracking. Sintering occurs at 1,300°C after this calcination step.
A hybrid sol-gel coating method has been developed to improve the coated surface's quality. In this new technique, zirconium oxide powder is added to the sol and the pH of the mixed sol is controlled at 4.2 to obtain optimal viscosity. This can produce thin and smooth films on the fibres. In addition, the fibres are heated at 950°C for sintering. It was anticipated that the extra zirconium oxide powder would fill cracks formed during the slow drying process and would, therefore, improve the surface condition.
After surface treatment, the composite exhibits qualitatively good transparency (see image (a) below). Light transmittance (LT) measurements (see graph (b) below) indicate that composites containing fibres in two different orientations can retain at least 50% of the LT compared with the glass matrix.
a) Optomechanical composite sample showing optical transparency (the length of the sample is 50mm).
b) Light transmittance versus wavelength for different window areas
Based on the structure of the present composites, a shadow area is defined in the form of the opaque region caused by the presence of alumina fibres. Conversely, the matrix area contributing to the LT is defined as the optical window area:
The ratio of the total shadow area to the total composite area, Sc, can be used to calculate the decreasing LT caused by opaque fibres, as follows:
The light transmittance (LT) of the composites is obtained from the
optical window area by equation 1 (Ref: Fail-safe light transmitting SiC
fiber reinforced spinel matrix optomechanical composite, by A F
Dericioglu and Y J Kagawa. Mater. Sci, 2002, 37, p523).
Where Tc is the light transmittance of the composite, Tm is the light
transmittance of the matrix, Sw is the optical window area, Sc is the
composite area, Ss is the shadow area. Then, the normalised LT of
composites containing fibres in perpendicular directions can be
calculated as follows –
Where Rf and ds are the fibre radius and the size of the optical window (distance between adjacent fibres), respectively.
In the detail
The image, right, shows a scanning electron micrograph (SEM) image of the borosilicate glass fibre composite cross-section, indicating that, in this particular specimen, viscous flow of the glass has not led to complete infiltration of the spaces between fibres in the bundle.
Fibre bundles, coated using the hybrid sol-gel method, have been observed by focused ion beam secondary ion mass spectroscopy (FIB-SIMS). Typical results at two magnifications are shown (below left). Cross-sections of a fibre bundle sectioned by FIB have confirmed that the zirconium oxide sol has successfully infiltrated the bundles, with the material connecting adjacent fibres, thus eliminating or reducing residual porosity.
The average tensile stress values of the fibres, and of the fibres coated using the hybrid sol-gel method, are found to be similar at 1,215MPa and 921MPa, respectively. The temperature limitation of the fibres is 1,300°C, according to the manufacturer. The main reason for the strength reduction is the fine grained structure of alumina, which changes at temperatures above 1,200°C, with grains growing from approximately 65nm to 160nm. The improved zirconium oxide coating system developed has, therefore, led to moderate strength reduction, due to the relatively low sintering temperature used.
The flexural strength of the oxide coated fibre-reinforced borosilicate glass composites has been measured by the four point bending strength test. Composites with different fibre arrangements and varying fibre content (fibre spacing) have been tested, and their flexural strength has been compared to that of the unreinforced matrix fabricated under the same process.
The average strength of the composites is found to be in the range 30-40MPa, with decreased spacing and increased fibre content resulting in a slightly higher average flexural strength.
The flexural strength of 0/90° composites is mainly influenced by the 0° oriented fibre bundles, which are parallel to the direction of the tensile force. A measure of the work of fracture (WOF) has been obtained from the characteristic area under the load–displacement curve, divided by the cross-sectional area of the specimen. It has been found that the composite WOF increases, on average, by a factor of six compared to the glass matrix value, by introducing zirconium oxide coated fibres.
Alumina fibre-reinforced borosilicate glass matrix composites have been developed and characterised. Observations through SEM confirm that the selected process parameters are effective in densifying the composites without significant loss of transparency. It has also been confirmed that a homogeneous zirconium oxide coating deposited on the fibres by a hybrid sol-gel method leads to composites with improved work of fracture.
Although the composites exhibit a small optical window area, the composite’s ability to transmit light in the visible wavelength region is not significantly impaired. The LT of composites with different optical window areas can be predicted by a geometry-based equation.
This is a new family of inorganic transparent composites with potential applications in security glazing and armour structures, as well as impact and high temperature resistant transparent panels. Impact and thermal shock resistance will be the focus of further investigations.
Aldo R Boccaccini, Department of Materials, Imperial College London, London SW7 2AZ, UK, and the Institute of Biomaterials, University of Erlangen-Nuremberg, 91058 Erlangen, Germany. Email: email@example.com
Thanks go to co-authors Mr Richard Charter and Dr Mahmoud Ardakani. Also to Professor J R Taylor and Mr J C Travers for the use of their facilities in the Physics Department, Imperial College London, UK. We acknowledge collaborators and colleagues in the Department of Materials, Imperial College London.