The limit of strength and toughness of steel

The ideal structural steel combines high strength with high fracture toughness. This dissertation discusses the governing principles of strength and toughness, along with the approaches that can be used to improve these properties and the inherent limits to how strong and tough a steel can be.; Chapter I provides the background on fracture mechanics and describes the factors that affect cleavage fracture from atomic, microstructural and macroscopic levels. From such considerations, the inherent resistance to cleavage fracture is apparent.; Chapter II discusses the ideal fracture strength and the critical fracture strength in high strength steels. Recent ab intio computations predict an ideal cleavage strength of Fe near 14 GPa on {lcub}100{rcub} planes, indicating that the normal fracture mode of high strength steel, characterized by a yield strength greater than about 4 GPa is brittle. A preliminary study of the effect of alloying with Ni shows a trend of decreasing brittle behavior with increasing Ni content. More sophisticated methods to study this effect are proposed.; Chapter III examines the practical methods for achieving properties that approach the limits of strength and toughness of steel. The discussion focuses on grain refinement through thermal treatments in lath martensitic steels. The objective is to limit the crystallographic coherence length for transgranular cleavage propagation. A new method is presented for directly disrupting the crystallographic orientation alignment within a martensitic packet, and thereby achieving submicron effective grain size. This has been accomplished in AerMet 100 and 9Ni steel by an alternate “intercritical annealing” (L) and “reversion” (Q) treatment. An “L” treatment creates a dual-phase microstructure containing fine parallel laths with different alloy contents. Subsequent austenite reversion (“Q” treatment) leads to a two-step martensitic transformation during cooling, disrupting the laths within a packet and generating a large misorientation between the adjacent laths.; Chapter IV explains why this grain refinement can be accomplished, as described by martensitic transformation theory. It is shown that all KS and NW relationships can be divided into three Bain variant groups based on their transformation strain matrix. Only the variants from different Bain groups will have a large angle of misorientation. Furthermore, multiple variants from different Bain groups might appear if the geometric constraint prevents the relaxation of transformation strain.