Trends in Welding Home      Event Overview      ASM Homepage
Back to "Session 13 - Microstructure II" Search
    Back to Main Search

Wednesday, June 4, 2008 - 10:25 AM

Direct observation of inclusion formation in low-alloy-steel welds

H. Terasaki, T. Yamada, Y. I. Komizo, Osaka University, Ibaraki, Osaka, Japan

When the weld have a wt-%Al/wt-%O ratio of ~0.6 , a little amount of titanium and boron in low-alloy steel, the microstructure is dominated by the acicular ferrite, which posses the ideal mechanical properties: strength and toughness. In this chemical range, it is well-known that the spinel-type oxide is formed. The spinel-type oxide is anticipated to serve as preferred nucleants rather than other types of inclusion due to low crystallographic misfit with ferrite. However, the formation mechanism of spinel-type oxide during welding and the role of titanium are unclear. In the present work, the inclusion formation in 0.09C-0.35Si-1.72Mn-0.014Ti-0.0035B -0.046Oxygen mass% steel welds was in-situ observed by using time-resolved x-ray diffraction system and the formed inclusion was analyzed in detail by using transmission electron microscopy. Re-melting by gas tungsten arc was conducted in the beam line of synchrotron radiation source (SPring-8). The diffraction patterns detected in weld pool, during solidification and in solid state were analyzed in the time resolution of 0.05 seconds. It was directly verified that the spinel-type oxide was formed in the solid state (austenite). Furthermore, TEM micrograph of inclusion showed that the spinel-type oxide was formed in the amorphous phase (Mn-Si-O) and the titanium-rich layer existed at the periphery of the inclusion. Summarizing the experimental results for direct observation of inclusion formation and the formed inclusion analysis, it was suggested that glassy phase (Mn-Si-O) was formed in the liquid pool and the part of the glassy phase transformed to spinel-type oxide in the solid state, followed by the formation of titanium-rich phase at the periphery of the inclusion. The results suggested that possible nucleant for acicular ferrite was the Ti-rich phase even if spinel-type oxide was formed in the inclusion.

Summary: The spinel-type oxide formation in low-alloy steel welds was in-situ analyzed by using TRXRD system. The results were related to the direct analysis of inclusion by using TEM. Summarizing those results, the sequence of the inclusion formation and the new candidate for the ferrite nucleant were suggested instead of spinel-type oxide.