Please use this identifier to cite or link to this item:
|Title:||Defect-initiated fracture and strength of ceramics|
|Department:||Materials Science and Engineering|
|Keywords:||Materials Science and Engineering;Materials Science and Engineering|
|Abstract:||<p>The strength of ceramics is related to their fracture toughness and the fracture-initiating defects. The present thesis reports investigations of the strength/defect relationships that increase understanding of strength-improvement pedagogies and strength prediction methodologies for ceramics. The strength improvement of a 4.5wt% yttriapartially-stabilized zirconia (Y-PSZ) via defect elimination is demonstrated. Defect types were identified and eliminated via several novel techniques. These procedures were combined with a conventional sintering route. A crack model is developed and used to examine the discovered flaw/strength relationships. Four-point-bend flexural strengths were systematically improved from 880 to 1380 MPa, a value comparable to hot-isostatically-pressed or slip-cast materials. The flaw/strength relationships for various fracture origin types were examined, i.e. fiber inclusions, agglomerates, iron inclusions, pores and alumina inclusions. An order of flaw-type severity was discovered and defect types of the same "size" were found to act in a dissimilar fashion in fracture initiation. The "severity" is related to local residual stresses around defects due to thermal mismatch or differential sintering. The relative defect-severity was quantitatively ranked via the apparent fracture toughness and a "relative defect severity parameter" developed. The temperature dependence of this parameter for the various defect types was investigated and the behavior explained via residual stress relief. Combined high-temperature four-point-bend and chevron-notched-bar fracture toughness (Kio) tests were conducted. It was found that the residual stress associated with alumina defects in tetragonal zirconia was temperature-dependent, that associated with agglomerates, temperature-independent and could be annealed. No residual stress was associated with pores. The residual-stress relaxation temperature (Tg) was determined for tetragonal zirconia. The size equivalence of inclusion defects and the resulting flaws was investigated. The fracture behavior of alumina inclusions was investigated via an Al₂O₃-ZrO₂ composite layer sintered onto the tensile surface of PSZ four-point-bend bars. It was found that inclusion fracture exclusively occurred and the fracture was associated with weak interfaces between the grains of the alumina inclusion or between the inclusions and matrix. Crack initiation in the fracture-toughness chevron-notched-bend bar was studied. Ideal testing conditions for valid Kio values strongly depend on specimen preparation and material. An in-test precracking techniques was developed for Kio testing at room and elevated temperatures. A Y-PSZ ceramic was tested and ideal testing conditions proven. Defect-initiated fracture models for ceramics developed. Fracture initiation is a typical inclusion with a thermal expansion coefficient and fracture toughness lower than the matrix and a Young's modulus higher than the matrix was analyzed and modeled for a spherical inclusion using a weight function method to compute the residual stress intensity of factor and a part-through elliptical crack model to follow the crack extension. The model was used to predict the strength of PSZ containing α-Al₂O₃ inclusions. It was also used to compute the Tg for tetragonal zirconia. The predictions were in good agreement with the experimental data. For agglomerate-initiated fracture, an agglomerate sintering/ residual-stress-retention model is proposed and used to estimate the fracture stress for agglomerate-matrix decohesion. The residual stress levels after elastic and thermal recovery were estimated and, agreed with experimental data.</p>|
|Appears in Collections:||Open Access Dissertations and Theses|
Items in MacSphere are protected by copyright, with all rights reserved, unless otherwise indicated.