The Influence of Friction Stir Welded Process Parameters of AA2519- T62 on Joint Quality Defined by Non-destructive Laser Amplified Ultrasonic Method and by Microstructure Analysis

The paper contains the results of a non-destructive laser ultrasound internal structure analysis made on specially prepared friction stir welded joints made of aluminum samples. The process was conducted using four different groups of welding parameters. Different tool rotation speeds and tool traverse speeds were used. The main purpose of the research was to analyze the joint quality using a non-destructive laser amplified ultrasound method. A microstructure analysis was also conducted to compare the results of both tests.


Introduction
FSW (eng. Friction Stir Welding) is a very promising method in terms of joining aluminum alloys, which are difficult to weld using conventional methods [1][2][3][4]. This method is a solid-state welding process, where the joint is formed by plasticizing and mixing two workpieces during the action of a specially design tool on the material to be welded [1,[5][6][7]. An example of aluminum alloy difficult to weld by conventional means is AA2519 -armor grade alloy used for light military constructions [8,9]. Due to the relatively high concentration of copper (above 5.3%), solidification of this alloy in traditional welding causes a problem of low melting phase Al2Cu (548oC) and as a result, a high risk of hot cracking [10][11][12]. Although the low temperature of the FSW process (400-500oC) allows this problem to be avoided, it is still important to properly select welding parameters determining the quality of the obtained joint [13][14][15][16].
It is very important to determine the joint quality not only using destructive but also non-destructive methods to analyze joints right after the process. In connection with the growth in usability of FSW technology it is very important to determine the proper joint quality check method to ensure the required welded material properties. Ultrasound testing methods are useful in determining the internal quality of the material including internal and surface inclusions and defects [17][18][19]. This method is also useful for analyzing material properties by measuring the shear wave velocities [20,21]. The ability to use a laser to amplify the signal ensures the possibility of analyzing elements whose thickness is greater than 10 millimeters [22].

Experimental Background
The workpiece to be joined was a 5 mm thick AA2519-T62 extrusion with the chemical composition presented in Table 1. The friction stir welding process was performed a using ESAB FSW Legio 4UT machine with an axial force equal to 17 kN and the tilt angle of the MX Triflute tool set to 2o. The used welding parameters together with a designation of the samples are presented in Table 2. Table 2. T41  400  100  T81  800  100  T82  800  200  T84 800 400

Sample designation Tool rotation speed [rpm] Tool traverse speed [mm/min]
The joint was determined along the entire length of the weld and also perpendicularly to the traverse transition direction. This kind of analysis were made using optoacoustic equipment and a measuring technique basing on the generation of an ultrasonic signal with ultrashort, high power pulses amplified by an Nd: YAG laser [23]. Additionally, the samples were sectioned perpendicularly to the welding direction and were metallurgically examined. The microstructure investigation was performed using an Olympus LEXT OLS 4100 digital light microscope. As part of the metallographic sample preparation, samples were mounted in resin, ground with abrasive paper of 80, 320, 600, 1200, and 2400 gradations, and polished using diamond pastes (3 and 1 μm gradation). The samples were etched using Keller reagent (20 mL H2O + 5 mL HNO3 + 2 mL HF + 1 mL HCl) with an etching time equal to 5s.

Results and Discussion
The internal structure of the samples was shown in Figure 1. In both samples where the traverse speed was 100 mm/min (T41 & T81) some imperfections can be observed visible as different colored structures and wave shape distortion. This could also be connected with the density change which is caused by the welding process. As can be seen, the wave shape distortions are present in the samples obtained with the lowest tool traverse speed, which entails the longest affecting time of the tool on welded AA2519-T62.
Front view of the joint This leads to phenomena such as substantial dissolution and coarsening of the strengthening phase, as well as fragmentation of the remaining Al2Cu precipitates in the stir zone, and it can partly explain the wave distortions in Fig. 1 [14].
In the course of scanning, local velocities are determined. The velocities dependent on density are related to local modules of elasticity; they can be calculated as follows: ( 1 ) ( 2 ) where: E is Young's modulus. Note that an S-wave pulse is recorded in the interval between the first and second reflections of P-waves from the back side; the time delay of the S-wave's arrival can be used to calculate its velocity.
Density was determined by hydrostatic weighing of the samples in distilled water. Mean of density AA2519 alloy. The average obtained values of the elastic moduli for the aluminum alloy were E=67,5GPa and G=28,5GPa.
Distinctly distinguishable zones of recrystallization and thermomechanical affection relative to the base material. During the scan, it was found that the elastic modules in the thermo-mechanically affected zone and the recrystallization zone are reduced by 15%, relative to the base material. The changes in elastic modules in these areas can be explained by far-reaching changes in the welded material microstructure. The mechanical properties of AA2519 are mostly determined by the presence of the strengthening phase and during the FSW process this phase undergoes disadvantageous evolutions mostly in the recrystallization zone and thermo-mechanically affected zone [1,4].
To verify the results from the ultrasound method, light microscope observations were performed. The macrostructure of the T81 joint is presented in Figure 2 with the retreating and advancing side situated on the right and left side, respectively. The macrostructure consists of zones typical for the FSW process: the dynamically recrystallized stir zone (SZ), thermo-mechanically affected zone (TMAZ), heat-affected zone (HAZ) and base material (BM). The microstructure analysis of the T81 sample did not reveal any imperfections in the joint.  In the case of the T41 sample, the low value of the tool rotation speed (400 rpm) causes the formation of imperfections close to the flash on the advancing side of the joint as a result of insufficient material plasticization (Fig.  3a). The investigation of the T82 and T84 samples revealed imperfections in the form of voids localized in the upper part of the stir zone (Fig. 3c,d). The number of voids increases together with increasing tool traverse speed value. As can be observed, the same area in the T81 sample is characterized by a lack of visible voids (Fig. 3d). Higher values of tool traverse speed result in decreasing time in which the tool affects the workpiece by friction, and this can cause the presence of imperfections in the joint. It is a noteworthy fact that all imperfections identified by microstructure analysis are localized on the advancing side of the joints, which corresponds to lower heat input than the retreating side. Although insufficient heat input is the main reason for low joint quality, the character of the imperfections differs depending on the welding parameters.

Conclusions
As a result of the work, it was possible to diagnose the technology of friction stir welding with the help of laser-ultrasonic structuroscopy and the subject of the presence of continuity defects. A method for determining Young's modulus and shear modulus is presented. The study showed that the decrease in the value of the elastic moduli of the recrystallization region and thermo-mechanically affected zone was on average 15%.
During the analysis of the microstructure, it was determined that high values of tool traverse speed lead to the appearance of defects in the joint. Most of the specific defects are localized on the advancing side of the joint.