JOINING OF CERAMIC COMPOSITES AND ADVANCED CERAMICS

Ceramics and Ceramic Matrix Composites (CMC) are well known as useful materials for harsh environment applications. Ceramic components have been widely used as abrasives, cutting tools, nuclear fuel elements, catalyst supports and astronomical telescope mirrors. Generally speaking, ceramics and CMC are excellent candidates for high-temperature applications because of their good mechanical properties and thermal stability at elevated temperatures. However, to produce large-size and complex components, development and testing of joints for ceramics and CMC are fundamental. Joining materials and techniques currently available to join ceramic and CMC include diffusion bonding using various active fillers, transient eutectic phase methods such as nano-infiltration and transient eutectic-phase (NITE), laser joining, selected area chemical vapor deposition , glass-ceramic joining , solid state displacement reactions, preceramic polymer routes, reaction forming, brazing. Adhesive joining materials (e.g. epoxy resins) are used to join ceramics only for applications at temperature lower than 150 °C, i.e. adhesively bonded joints are widely used for automotive, aerospace, electronic and packaging applications. Together with the need of a reliable joining method, a widely accepted standard to test the mechanical strength of joined ceramics and composites is still unavailable and measuring the shear strength of the same joining material with different test methods could lead to different results. The aim of this Thesis is to compare and discuss several different shear strength tests used to join ceramics and CMC, with the final goal to provide designers and scientific community a widely accepted, reliable test method. A homogeneous pure shear stress state is not obtainable with most of the currently used single or double lap tests, which give apparent and not pure shear strength of the joined samples; in addition, the presence of several different apparent shear strength tests in the literature makes comparison quite impossible. The asymmetrical four point bending test (ASTM C1469-10) is recommended as a standard test by the ASTM to test joined ceramics and composites, but notches have to be cut in the joined area when the joining material strength is high. When the shear strength of the joining material approaches that of the substrate to be joined, ASTM C1469-10 is not suitable. With torsion tests a pure shear loading strength can be measured without using notches. One of the very first proposals for torsion test for epoxy bonded aluminum alloy samples was reported by M. Ouddane et al.: as thoroughly discussed there, torsion results were considered more reliable than those obtained by standard lap tests, the improvement mainly due to the fact that lap tests induce non uniform stress concentrations that affect the reliability of results. Recent papers deal with torsion tests on joined hourglass shaped samples : preliminary results suggest that the torsion test method with a miniature specimen has a potential to evaluate the shear properties of the joint interface, provided that the fracture occurs in the joined area. Torsion tests have been thoroughly analyzed in this Thesis: a pure non-uniform shear stress distribution is obtained with torsion tests; together with asymmetrical four point bending test (ASTM C1469-10), torsion test is the only one able to measure pure shear strength of joined ceramics and composites. The torsion test method is proposed in some ASTM standards but none of them is directly applicable to joined ceramics: the main result of this Thesis is to demonstrate the reliability of torsion as a method to measure the shear strength of joined ceramics and CMC. Two torsion standards have been adapted to joined C/C and ceramics, first by preparing square section samples (TS), rods (TC), tubes (TT), then by mechanically shaping the joined samples as a hourglass in different shapes. The choice of an epoxy resin (Araldite AV119) to join SiC in this PhD thesis was done in order to have a “model” brittle joining material to obtain a statistically relevant number of joined samples in a reasonable time, to compare shear strength results . More than one hundred joined samples have been prepared and tested at room temperature by asymmetrical four point bending (A4PB) (ASTM C1469-10), torsion on square section samples (TS), torsion on circular section rods (TC), torsion on hourglass shaped samples (THG, TDHG, TRHG), torsion of tubes (TT), single lap in compression (SL), single lap off set in compression (SLO), double lap off set in compression (DLO), Brazilian test (BT), double notch (DN) methods. A modified ASTM B898 standard has been used as a further example of single lap test in compression to complete the comparison work (B898).