April AMP_Digital

A D V A N C E D M A T E R I A L S & P R O C E S S E S | A P R I L 2 0 1 9 1 5 lambda plate was varied. Images were captured using a digital camera. RESULTS AND DISCUSSION Figure 1 shows a cross section through a friction stir butt weld of two pieces of AA5754 of different thickness- es etched with voltage applied imme- diately after submerging the sample into the solution. Grain flow contrast (Fig. 1a) and grain definition at the weld root transition from the parent metal to the stir zone (Fig. 1b) are excellent, but it is difficult to see the oxide stir line. After significant trial and error in seek- ing a satisfactory modification to the etching technique, it was determined that the optimal method was to sub- merge the sample in the etchant for 20 s prior to applying voltage. Shorter times did not provide significant contrast for the oxide line, and longer times pitted the precipitates in the microstructure more severely (although there is still some limited pitting of the precipitates using the optimal time as well). F riction stir welding (FSW) is a solid- state welding technique developed to join various alloys [1,2] and compo- nents. For example, the automotive in- dustry uses the technique to fabricate aluminum alloy tailor welded blanks (TWBs) used in the manufacture of au- tomobiles. There is a demand for these components in various material thick- nesses and alloy compositions. This re- quires a metallographic analysis method that enables thorough, rapid analysis of FSW joints to support R&D and keep pace with industry demand. This article discusses the development of a metal- lographic characterization technique to identify defects in the weld fusion area to enable fast optimization of weld process- ing conditions. Defects in FSW joints have been examined using various methods [3] . However, most cross-sectional analy- sis techniques only allow examination of either grain structure and flow in the FSW stir zone or defects such as ox- ides in the joint. Oxides in the stir line can lead to lower strength values [4] , so process optimization and tool design are used to prevent entrainment of ox- ides. However, the oxide line needs to be analyzed in conjunction with grain flow and grain structure to properly op- timize the process. The authors devel- oped an etching technique that reveals grain structure and grain flow and iden- tifies lines of oxides, voids, and other exogenous materials on the same sam- ple. Using optical microscope imag- ing, the etching technique reveals all of these microstructural features in one step for 5000 and 6000 series aluminum alloys. Results are based on examining samples produced using high-speed FSW with >2 m/min linear travel speed at TWB Co., Monroe, Michigan, with process parameters developed for cost- effective mass production of aluminum alloy TWBs. Nodata is presented for pro- duction FSW conditions and no welds fromoptimized production samples are shown due to proprietary restrictions. EXPERIMENTAL PROCEDURES Metallographic samples for cross- sectional examination were cut from welded blanks or blanks, mounted in epoxy, and resectioned using a high-speed pre- cision saw. A small piece of aluminum was left protruding from the backside of the mount for easy electrical con- nection during etching. Grinding and polish- ing were done by hand. Final polishing was performed immediate- ly before etching using 0.05-µm colloidal silica on a pre-wet (with water) low-nap syn- thetic polishing cloth at 150 rpm for 30 s. Samples were thor- oughly washed with a micro-organic soap, rinsed with water, and transferred to the etch- ing station immediate- ly after rinsing. (Con- tact the corresponding author for more details on the sample prep procedure.) All samples were etched using Barker’s reagent (4% aqueous solution of fluoroboric acid, or HBF). The counter electrode was alu- minum alloy 6111. Because all samples (one sample per mount) exhibited rel- atively consistent areas, current was not monitored and voltage was set at 20 V. Some samples were etched by submerging the polished face in the acid for 120 s with the voltage on. Oth- ers were etched using a modified pro- cedure consisting of submerging in Barker’s for 20 s with no voltage ap- plied, then applying 20 V for 100 s. Samples were rinsed with flowing wa- ter, dried, sprayed with ethanol, and dried for at least 5 min in warm air prior to examination. Optical microscopy was per- formed using an invertedmetallograph- ic microscope with polarized light and a full-wave (lambda plate) sensitive tint filter, which alters the polarization state of the light, producing color contrast between grains in the samples. The lev- el of polarization prior to entering the Fig. 1 — Optical micrographs of a friction stir butt-weld joint fabricated with two aluminum alloy 5754 sheets of dissimilar thickness, etched with Barker’s reagent following standard procedures where voltage is applied immediately: (a) image of the entire FSW stir zone, magnification 12.5x; and (b) transition from parent metal to stir zone at the weld root, magnification 50x. 500 µm 2 mm