Experiments in Vacuum Brazing of Titanium

December 1, 2007

Army Research Laboratory, Adelphi, Maryland

An experimental study of vacuum brazing of titanium and of the effects of changes in brazing alloys and brazing process conditions has been performed. [As used here, “titanium” signifies both commercially pure titanium and an alloy nominally consisting of 90 weight percent of titanium, 6 weight percent of aluminum, and 4 weight percent of vanadium (commonly abbreviated “Ti-6Al-V”).] The knowledge gained in this study is intended to contribute to development of capabilities for fabricating titanium structures in circumstances in which welding — heretofore the typical method of joining titanium — cannot be performed because access is limited or adjacent nonmetallic components would be harmed. There is a particular need for such knowledge to enable fabrication of lightweight, durable titanium- based structures for armored vehicles. Examples of such structures include standard lightweight plate structures, titanium components encapsulating ceramics, and panels that comprise pyramidal frame cores sandwiched between face sheets.

This Sandwich Structure was fabricated by brazing a pyramidal frame core to face sheets. The face sheets and core were made of commercially pure titanium. The face sheets were 1.22 mm thick; the core members were fabricated from 0.61-mm-thick sheets. The brazing alloy was 37.5Ti-37.5Zr-15Cu-10Ni (proportions in weight percent).

The experiments performed in this study are summarized as follows:

Blocks of Ti-6Al-4V having dimensions of 75 by 75 by 50 mm were joined by tungsten/ inert-gas welding around the edges, followed by hot isostatic pressing (HIP) in argon at a temperature of 900 °C and a pressure of 103 MPa.
Blocks of Ti-6Al-4V were joined in a vacuum by diffusion bonding, using a foil of a brazing alloy comprising 70Ti- 15Cu-15Ni (proportions in weight percent) placed around the edges. In each case, the block-and-foil assembly was heated to a temperature of 1,000 °C under a deadweight equivalent to a pressure of about 15 kPa, followed by HIP as in experiment 1 described above.
Blocks of Ti-6Al-4V were brazed and subjected to HIP as in experiment 2, except that prior to heating, the foil was placed over the entire Ti-6Al-4V bonding surface.
Blocks of Ti-6Al-4V were brazed as in experiment 3, omitting the HIP step.
Panels comprising commercially pure pyramidal frame cores sandwiched between commercially pure titanium face sheets (see figure) were fabricated in a process that included a multistep thorough cleaning, application of a zirconium- rich titanium-brazing paste/tape to the face sheets, stacking the face sheets to the cores (with the brazing paste/tape faces in contact with the cores) in the sandwich configuration, and heating in a vacuum furnace at about 900 °C under a deadweight equivalent to a pressure of 50 kPa.

Shear and tensile tests were performed on the specimens thus fabricated to determine relationships among strength, ductility, and the various fabrication processing routes. Optical and scanning electron microscopy were employed to quantify degrees of bonding and to examine microstructural changes (both within the base titanium materials and at the bond surfaces) associated with the process variations.

The conclusions drawn from the test results are summarized as follows:

The combination of diffusion bonding and HIP as in experiment 1 and the combination of brazing and HIP as in experiment 2 were found to be suitable as means of encapsulating a ceramic within a titanium structure. Further research to optimize brazing time and temperature is warranted.
The brazing-only process as in experiment 4 was found not to result in a viable structure. Further study could enable determination of whether increased axial load during brazing could improve bonding and possibly eliminate the need for HIP.
Success was achieved in bonding pyramidal frame cores with the face sheets by use of the zirconium-rich titanium-brazing paste/tape. While measured shear strengths were considerably lower than historical values, they were great enough to sustain stretching of core microstructures. Increases in braze shear strengths were produced by increasing brazing times and temperatures.
Different brazing alloys may result in bonds of similar or greater strength at lower brazing temperatures, thus eliminating risks of creep and of degradation of properties of the titanium parent material. The use of lower-temperature brazing alloys at greater applied pressures during brazing may result in stronger bonds.

This work was done by Kevin J. Doherty, Jason R. Tice, Steven T. Szewczyk, and Gary A. Gilde of the Army Research Laboratory.

ARL-0021