Effect of titanium additions to low carbon, low manganese steels on sulphide precipitation

Author

Sima Aminorroaya-Yamini, University of WollongongFollow

Year

2008

Degree Name

Doctor of Philosophy

Department

Engineering Materials Institute - Faculty of Engineering

Recommended Citation

Aminorroaya-Yamini, Sima, Effect of titanium additions to low carbon, low manganese steels on sulphide precipitation, PhD thesis, Engineering Materials Institute, University of Wollongong, 2008. http://ro.uow.edu.au/theses/403

Abstract

Recent developments of high strength line pipe steel have seen a decrease in carbon content from 0.25 wt% to less than 0.05 wt% and manganese to less than 1 wt% in an attempt to reduce centreline segregation. The sulphur content should be less than 0.008 wt% as well. However, recent papers argued that low manganese levels in pipeline steels have the potential to improve weld line toughness and allow a higher tolerance for sulphur. Should this be true it would be possible to eliminate the usual desulphurising treatments thereby decreasing the production cost. The current study was designed to explore the effect of titanium on sulphide precipitation in steels with higher sulphur contents than currently used in the pipeline industry. The mechanical properties of steels are influenced by the type, shape and size of sulphide precipitates. Titanium additions to low carbon steel influence the precipitation behavior and composition of sulphide precipitates and could provide grain refinement, precipitation strengthening and sulphide shape control. In the present study, titanium was added to a low carbon, low manganese steel which contained approximately 0.01 wt% sulphur in order to study the effect of titanium on sulphide formation in the absence of other alloying elements. Microscopic assessment of centreline sulphide precipitates revealed that irontitanium- sulphide co-exists with (Mn,Fe)S. An increase of the titanium content from 0.008 wt% to 0.024 wt%, results in an increase in the titanium content of irontitanium- sulphide phases and a decrease in the iron content of the (Mn,Fe)S phases. In contrast to reports in the literature where it has been suggested that titanium can dissolve in MnS and even that it is possible for TiS to replace MnS. This study has shown that Ti replaces iron in FeS but does not dissolve in manganese sulphides. The presence of FeTiS2 (trigonal CdI2-type structure with a P3m1 space group and lattice parameters of a = 0.34 nm and c = 0.57 nm) precipitates has been verified for the first time in steel. IV A concentric solidification technique in a laser-scanning confocal microscope was employed to observe in situ, solidification and the precipitation of sulphides following segregation of alloying elements into the remaining liquid. Microstructural development of low carbon steel upon solidification has been observed in situ, alloying element distributions were measured experimentally and sulphide precipitates have been compared to the precipitates at the centreline of continuously cast steel slabs. Microstructural development was similar in the two cases and some aspects of alloying element segregation upon solidification of a slab can be simulated experimentally by the use of the concentric solidification technique. There is a remarkable similarity between the sequence of events decreased in the in-situ observations and that occurring in industrially-cast slabs reported in the literature. Sulphide precipitates at the centreline of concentrically solidified specimens are similar in morphology and composition to precipitates in continuously cast steel slabs and TEM analysis confirmed that FeTiS2 as well as MnS (containing iron) can form on prior austenite grain boundaries in both cases. The progression of the solid/liquid interface in plain Fe-C and Fe-Ni alloys was simulated by using Diffusion Controlled TRAnsformation (DICTRA) software which operates in conjunction with Thermo-Calc. Good correlation was found between experimental results and computations. Solute build up in the liquid at the solid/liquid interface was determined experimentally in Fe-Ni alloys and there was good agreement between the calculated concentration gradient of nickel and experimental measurements using an EPMA technique. Therefore, it is possible to determine comparatively the centreline segregation with fairly elementary experimental procedures and mathematical simulation at various steel compositions if DICTRA and Thermo-Calc are provided with sufficiently accurate thermodynamic information. Precipitates which formed in the solid state were assessed by carbon replication techniques, transmission electron microscope studies and atom probe tomography. Manganese and copper sulphides were the dominant sulphide precipitates. Copper sulphides nucleated on manganese sulphides and formed a shell around rod-like and V globular manganese sulphides. Iron did not dissolve in manganese sulphides and therefore the manganese sulphides which formed in solid state have higher melting points than sulphides formed on prior austenite grain boundaries at the centreline of slab. Titanium nitrides were observed in various sizes and distributions. TiN acts as nucleation sites for the precipitation of sulphides. An increase of the titanium content from 0.008 wt% to 0.024 wt%, results in precipitation of a few micrometer TiN and a decrease in the number of small cubic TiN precipitates. Small particles (~ 10nm length and 3 nm thickness in size) that contain Ti, C, S and N have been identified for the first time by the use of three-dimensional atom probe tomography