Elektronenstrahlschweißen von Titanaluminid-Legierungen

Holk, Jens; Reisgen, Uwe (Thesis advisor); Bobzin, Kirsten (Thesis advisor)

Aachen / Publikationsserver der RWTH Aachen University (2015) [Dissertation / PhD Thesis]

Page(s): XII, 146 S. : Ill., graph. Darst.

Abstract

TiAl alloys offer a great potential to substitute high temperature resistant nickel-base alloys due to their lower specific weight. For increasing oxidation and creep resistance these alloys contain high amounts of niobium and other elements, which may influence the propensity for segregation. Furthermore these alloys generally exhibit low ductility and fracture toughness, which makes them prone to cracking during heat treatment. To avoid atmospheric influence the electron beam process has been chosen for joining two different TiAl alloys by welding. The GE microstructure consists of gamma (NG), the V5 microstructure of a gamma/alpha2 (FL) structure, which both show different properties. The experiments carried out have shown a potential for crack-free welding of TiAl up to 5 mm thickness with the electron beam. The cooling rates are highest in electron beam welding, thus a great reduction of welding speed down to either 1 mm/s or 2 mm/s in combination with preheating of at least 500°C is necessary to obtain crack-free welds. Preheating was carried out efficiently by deflecting a defocused electron beam. While preheating mainly influenced the width of the weld seam and thus reducing stress peaks by homogenizing stress distribution, comparing the two alloys did not show great differences in welding parameters and cracking rates respectively, but in their microstructure. The alloy GE exhibited a peritectic solidification behavior, combined with minor segregations of aluminium. Due to their higher amount of aluminium high amounts of massive gamma-m have been witnessed mainly on weld seam boundaries. With reducing welding speed the amount of gamma-m decreased. The alloy V5 was found to solidify via solid state beta, which could reduce and abolish stresses in the higher temperature range, i.e. directly after solidifying. The lower ductility at room temperature due to the lamellar microstructure could thus be compensated. It should be mentioned that, a certain TiB2 coagulation was found in the HAZ of alloy V5 which could influence the high temperature behavior. Both alloys exhibited a slight loss of aluminium in the weld seam, whose influence could not be figured out exactly, but should be considered during afterwards heat treatment. Nevertheless, hardness increase in both alloys was distinct, but only GE alloy showed a reduction with decreasing cooling rate. This is considered due to the potential of dislocation movement. A final heat treatment had no significant effect on the hardness and only small influence on the microstructure, hypothesizing that parameters have to be adjusted.To evaluate the mechanical properties tensile tests and stress measurements by x-ray and neutron diffraction have been carried out. Due to problematic dimensions and shape the tensile test specimen showed strong scattering of results. Although high values of up to 485 MPa were achieved, strain values were below 1% and results are thus considered only to be trend-setting. The stress analysis, however, showed a significant reduction in welding residual stresses when preheating of at least 500°C or massive reduction of welding speed was applied, this being sufficient for crack free joining. Neutron diffraction analysis thereby showed general applicability, although strong texture and extremely fine microstructure could deteriorate the measurement results.

Identifier

  • URN: urn:nbn:de:hbz:82-opus-52997
  • REPORT NUMBER: RWTH-CONV-145407