The Relation Between Alloy Chemistry and Hot-Cracking

Hot cracking is a problem in welding 2195 aluminum-lithium alloy. Weld wire additives seem to reduce the problem. This study proposes a model intended to clarify the way alloying elements affect hot-cracking. The brittle temperature range of an alloy extends wherever the tensile stress required to move the meniscus of the liquid film at the grain/dendrite boundaries is less than the bulks flow stress Sigma(sub B) of the grains: 2gamma/delta <= sigma(sub B) + P where gamma is boundary film surface tension delta= boundary film thickness P = gas pressure (Some alloys outgas.) If the above condition is not met, the grains deform under stress and the liquid film remains in place. Curves of 2gamma/delta and sigma(sub B) vs. temperature in the range just below the melting temperature determine the hot cracking susceptibility of an alloy. Both are zero at onset of solidification. sigma(sub B) rises as the thermal activation of the slip mechanism is reduced. 2gamma/delta rises as the film thickness delta which can be estimated from the Scheil equation, drops. But, given an embrittled alloy, whether the alloy actually cracks is determined by the strain imposed upon it in the embrittled condition. A critical strain is estimated, Epsilon(sub C) on the order of Epsilon(sub C) is approximately delta/l where L = grain size and where the the volume increment due to the strain, concentrated at the liquid film, is on the order of the liquid film volume. In the early 80's an empirical critical strain cracking envelope Epsilon(sub C)(T) was incorporated into a damage criterion to estimate the effect of welding parameters on the formation of microfissures in a superalloy with good results. These concepts, liquid film decoherence vs. grain bulk deformation and critical strain, form the key elements of a quantitative theory of hot-cracking applicable for assessing the effect of alloying elements on hot-cracking during welding.