Modified Specimen Preparation for Applied-Moment Double-Cantilever-Beam Testing

A method is discussed for preparing applied-moment double-cantileverbeam specimens used in fracture toughness (K_Ic) testing. This procedure eliminates the need to precrack the specimens, which often leads to specimen failure before testing. In addition to yielding toughness values which are in agreement with literature values, this method allows the experimenter greater control in selecting the location of failure initiation.     

Thermally Induced Wavy Hertzian Fracture

Reproducible "complexed" Hertzian cones were formed in glass as a result of thermally induced tensile stresses. These cones differ from normal cones by having wavy profile patterns, indicating two modes of fracture operating alternately. One mode is shallow and the other steep, together averaging a crack angle of 18.9 ± 1°. The thermal technique offers new test conditions for studying Hertzian fracture.     

Impact Fracture of Thermally Tempered Glass Helicopter Windshields

In-service fracture of helicopter windshields was studied. Simulated catastrophic fracture tests were conducted by firing alumina and steel spheres onto stationary tempered and as-received glass panels. The results were studied by Hertzian analysis and modified Auerbach's relations. Thermally tempered glass shows much higher impact resistance than that estimated from superposition of residual stresses. Subcritical impact sites exhibit slow crack growth in tempered plates, eventually leading to fracture of the entire plate.     

Discussion of Thermally Activated Approaches to Glass Fracture

A review of the Charles and Hillig approach to thermally activated crack growth mechanisms indicates that this theory may be applicable to cracking in vacuum at high temperatures but not to stress-corrosion cracking in aqueous solutions. For stress-corrosion mechanisms, the ramifications are 2-fold: (1) Thermal activation analysis shows that the true activation energy in soda-lime glasses is closer to 0.55 × 10⁵ J/mol (13.2 kcal/mol) than to the 1.2×10⁵ J/mol (29 kcal/mol) obtained using the Charles and Hillig approach; and (2) other models for slow crack growth controlled by bulk diffusion of H₂, OH^-, or Na⁺-H⁺ are suggested and shown to be consistent with much of the observed phenomena.     

Fracture Features at Internal Fracture Origins in a Commercial Crystallized Glass

Failures in samples of a commercial crystallized glass subjected to intense pulsed electron-beam irradiation resulted information of numerous full (circular) fracture mirrors. These features, resulting from simultaneous fracture initiation at internal sites, are of sizes corresponding to the range of fracture stresses found in extensive flexure tests, suggesting that the strength of this ceramic may be controlled by intrinsic (volume) defects.     

Fracture of Glass in Vacuum

The fracture of 6 glasses was studied in vacuum, <10⁻⁴ torr (10⁻² N/m²), as a function of temperature from 25° to 775°C. Subcritical crack growth was observed in 4 of the glasses. Activation energies for crack motion ranged from 60 to 176 kcal/mol. The glasses which did not exhibit slow crack growth were "anomalous" glasses with abnormal thermal and elastic properties. Critical stress intensity factors for these 2 glasses increased ∼10% as the temperature increased to ∼600°C. It is felt that subcritical crack growth is not the result of alkali-ion diffusion or viscous flow but rather of a thermally activated growth process which depends on the crack-tip structure in the glass. A narrow cohesive region at the crack tip favors subcritical crack growth, whereas a wide region favors abrupt fracture.     

Mixed-Mode Fracture in Soda-Lime Glass

Soda-lime-silica glass was fractured under combined mode I and mode II loading from flaws produced by hardness indentations. Critical stress intensities calculated from K₁ and K₁₁ combined using four analyses, are compared to K_IC measured by a fracture mechanics technique and to values of K_IC determined by measuring fracture-mirror size. The comparison showed that K_IC calculated using a noncoplanar strain-energy release-rate analysis gave the best agreement with values obtained by fracture mechanics techniques over the widest range of crack orientations. K_ICcalculated from mirror sizes was constant regardless of the orientation of the original flaw.     

Strength Degradation of Thermally Tempered Glass Plates

An earlier theory of contact-induced strength degradation of brittle materials is extended to include plates in residual surface compression. The scale of the strength-controlling flaw is predicted by indentation fracture mechanics, with the modifying effect of the residual field incorporated into both indentation and strength equations. Experimental verification of the predictions is obtained from diamond-pyramid indentation tests on thermally tempered glass plates. As with untempered plates, the theory accounts for the load dependence of the strength loss; it also explains the insensitivity of the degradation characteristics to initial flaw distribution and identifies toughness as the controlling material parameter. Most significant, however, is the demonstration that surface strengthening can produce dramatic improvements in degradation resistance. The possibility of obtaining all parameters necessary for a complete degradation analysis of a given tempered inaterial entirely by routine indentation/strength testing is discussed.     

Fracture of Tempered Glass

The fracture of thermally tempered glass is discussed in terms Of both the stored elastic strain energy in the glass due to tempering and the elastic energy release rate Of crack extension, 9. The latter is used to obtain an analytical correlation between the maximum tensile stress and the average particle size at time of fracture. The theoretical predictions are supported by experimental data obtained for various glass thicknesses and temper levels.     

Proposed Fracture Theory of a Dispersion-Strengthened Glass Matrix

A fracture theory is proposed for a composite system based on a continuous glass matrix. It is hypothesized that hard crystalline dispersions within the glass matrix will limit the size of Griffith flaws and strengthen the composite. Quantitative relations are derived for the effect of a dispersed phase on composite strength. At low volume fractions of the dispersed phase, the average flaw size is statistically reduced independent of the size of the dispersed particles. At high volume fractions of the dispersed phase, the average flaw size is governed by the average distance between particles dispersed in the matrix. The strength of a composite should, therefore, be a function of the volume fraction of the dispersed phase at low volume fractions and dependent on both the volume fraction and particle size of the dispersed phase at high volume fractions. For verification of the theory, cross-bending strengths were measured on a sodium borosilicate glass containing varying volume fractions of spheroidized alumina over a range of particle sizes. The average distance between dispersed particles ranged from approximately 15 to 500μ. Good agreement with theory was found. Values of glass surface energy calculated from the experimental data agree well with literature data.     

Double-Cantilever-Beam Testing of a Transversely Isotropic Fibrous Silica Material

The double-cantilever-beam (DCB) geometry was used to study the fracture behavior of a transversely isotropic fibrous silica material that is used in the thermal protection system of the Space Shuttle. For this testing geometry, the evaluation of K_c requires measurement of the elastic constants of the material. Determination of the compliance behavior of the material allowed the various analytical beam models that have been put forward to describe the DCB geometry, to be distinguished. The Timoshenko beam on a generalized foundation was found to be the best model to describe the experimental compliance behavior and the K_c measurements. The measured K_c values were found to be higher than for previous work that had used the double edge-notched tensile geometry. Crack propagation in the material was found to be characterized by variations in the crack-resistance force, probably reflecting microstructural variations. The subcritical crack-growth behavior was studied using the constant-load-mode DCB technique, and the results were in agreement with the predictions of an earlier study.     

Fracture Surface Energy of Glass

Fracture surface energies of six glasses were measured using the double-cantilever cleavage technique. Values ranged from 3.5 to 5.3 J/m² depending on the chemical composition of the glass and the temperature of the test. The fracture surface energy increased with decreasing temperature and increasing Young's modulus; however, exceptions to this behavior were noted. The magnitude of the values obtained is discussed with respect to the theoretical strength of glass and possible irreversible effects at the crack tip such as stress corrosion and plastic deformation are considered.     

Thermally Induced Delamination of Symmetrically Graded Multilayers

A fracture mechanics model is used for calculating the steady-state energy release rate of symmetrically graded ceramic multilayers. Results from the present study and previous results on asymmetric multilayers are generalized to simple design rules for avoidance of delamination of graded multilayers: (A) For asymmetric multilayers, delamination can be suppressed by inserting a few additional interlayers. (B) For symmetric multilayers, multilaying is not efficient; instead materials should be selected which have a sufficiently low thermal expansion mismatch. The later can be judged by making bilayers of identical thickness as the desired symmetric multilayers. If the bilayer delaminates it is unlikely that symmetric multilayers can be processed, and other materials (having a smaller thermal expansion coefficient mismatch) should be tested.     

Morphology of Creep Fracture of a Phase-Separated Borosilicate Glass

A phase-separated borosilicate glass with interconnected microstructure was stretched in the viscoelastic temperature range. When elongated, significant necking and subsequent fracture occurred. SEM study of the fractured samples revealed an enormous amount of crazing on the surface and void formation in the bulk of the samples near the fractured edges.     

Fracture Velocity and Fracture Energy of Glass in the Fatigue Range

A method is developed for determining crack velocities from the stress-time curve of fracture. Velocities of glass broken in air and in vacuum converge at a value between 1 and 10 mm. per second. This convergence is considered to be the upper limit of the fatigue range. Fracture energy has been computed in terms of strain energy release rates. For glass broken in air under low stresses this energy is about equal to the surface energy of the glass, but when in vacuum it is fifteen times greater. At the upper limit of the fatigue range it is thirty times greater, whereas at the terminal velocity of fracture it is of the order of fifty times greater. It is concluded that surface energy must constitute only a small part of the energy absorbed in the fracture process. This excess energy has a pronounced influence on the fracture process and on the measured strength of glass.     

Thermally Induced Reversible Coagulation in Ceramic Powder-Polymer Liquid Suspensions

It was observed that slurries of oxide powders in oxidized polybutene fluids can be caused to change reversibly between fluid, nearly Newtonian behavior and plastic behavior by modest changes in temperature. This phenomenon was believed to result from changes in the dispersion vs association among the particles. The rheological effects of temperature, polymer oxidation, and particle size were observed for 30 vol% slurries of TiO₂, Al₂O₃, and ZrO₂ powders in polybutene fluids. Elasticity (in oscillation) and low-shear-rate viscosity (in steady shear) were observed to increase with increasing temperature for TiO₂ and Al₂O₃ particles in oxidized polybutene fluids. This behavior was attributed to the creation of interparticle structures. The attainment of this structure on heating was observed to be inhibited by increased oxidation of the polymer and increased particle size. It was concluded that the adsorption of oxidized molecules from the polymer liquid, along with the high viscosity of the bulk polymer, resulted in suspensions that were metastable against coagulation. Increased temperature resulted in lower viscosities of the liquid, allowing coagulation on a short time scale. The presence of the adsorbed polymer, however, prevented intimate contact among the particles so that the coagulated structure was easily destroyed upon subsequent cooling and shearing.     

Crack-Front Curvature and Glass Slow Fracture

Fracture-surface markings on glass samples broken under liquid decane indicate that region II constitutes a transition between environmentally assisted and environmentally free fracture modes. Region II itself is subdivided into two stages, both characterized by negative crack-front curvature. This reversed curvature develops initially to allow more time for diffusion of the environmental agent to the middle portion of the fracture front and leads to increased local stress which activates the environmentally independent fracture mode at the center of the fracture front. Most of region II occurs in the latter mixed-mode fracture condition.     

Acoustic Emission in the Fracture of Glass Plates

The ring-down counting method was used to measure acoustic emission (AE) during the subcritical cracking of float-glass plates. It was found that AE emanated primarily from surface flaws and that a flaw =4 μm deep was capable of producing one AE count. The dependence of AE count rate on experimental parameters such as system gain and specimen size was determined and the use of AE simulators for calibrating AE systems was examined. The results were used to show how data on different ceramic materials might be compared.     

Mixed-Mode Fracture in Soda-Lime Glass Using Indentation Flaws

Mixed-mode failure of soda-lime glass under inert and fatigue test conditions was studied using Knoop indentation flaws. For annealed cracks (residual stress-free) crack extension (catastrophic or subcritical) is by an abrupt transition from the initial crack plane to a noncoplanar crack plane followed by a reorientation of the crack normal to the applied stress. Although fatigue strength of these inclined flaws increased linearly with respect to orientation of the flaws to the applied stress up to an angle of 60°, this increase was considerably less than what was predicted by existing theories. It is believed that subcritical crack growth causes the crack to be realigned perpendicular to the applied stress before failure for all orientations; hence, fatigue strength does not show the dramatic increase at orientation angles as predicted by theory. For as-indented cracks the contact residual stress causes the crack extension to be less inclined to the initial crack plane than for annealed cracks, but in this case also, the crack realigns itself perpendicular to the applied stress. Again, fatigue strength is relatively insensitive to the orientation angle as predicted by theory and subcritical crack growth is believed to play a primary role in determining this strength dependency.