1 Introduction

The general goal of reducing CO2 emissions is being pursued in particular with lightweight structures. In the transport and logistics sector in particular, aluminum materials thus offer potential for reducing dead weight and increasing payload. This is, among other things, the reason for the increasing demand in the aluminum sector and goes hand in hand with the requirement that the processing properties of the alloy allow the material to be used in the individual application. In this context, joining technology and, in particular, welding technology play an essential role. Safe joining conditions are requirements for maximum utilization of the material properties in the subsequent application. The material properties are adapted and specifically influenced by alloying elements. In GMAW welding processes, the chemical composition of the weld metal is influenced in particular by the filler wire. Microalloying elements are added to influence, for example, the mechanical properties or the cracking resistance. The avoidance of hot cracks is a particular challenge in the welding of aluminum materials.

Due to its economic efficiency, the MIG welding process is widely used in the processing of these materials. Various influencing variables have to be considered, which have an influence on the weld seam or the WAAM structure [1]. In addition to the deposition strategy [2] and the travel speed [3], the filler material has a major influence on the resulting material structure. The alloy composition of the filler materials used in this process is adjusted in the molten phase, and the resulting semi-finished products (rod material) are subsequently processed into welding wire electrodes. In addition to the alloy adjustment in the molten phase, cored wires also influence the composition of the weld metal for steel wire electrodes.

Investigations show that thin-film coatings, which can be applied to selected substrate wire electrodes by PVD processes, are also appropriate for introducing microalloying elements into the weld metal [4,5,6,7,8,9,10]. As expected, it is shown that the amount of alloying elements can be modified by adjusting the thickness of the coating. The external coating, however, influences not only the alloy composition and microstructure, but also the welding process.

The grain refining effect of titanium diboride (TiB2) on various aluminum materials has already been researched [11,12,13,14]. In tungsten inert gas welding and laser beam welding, it has been proven that the alloying elements titanium and boron, which are introduced into the welding area by cast inserts, have a corresponding effect [15,16,17,18,19,20]. By increasing the titanium diboride content with casting inserts in the weld seam area, Schempp [16] was able to reduce the grain size of the alloy AlMg4.5Mn by 44%.

The influence of titanium on the microstructure and grain size has also been investigated [21,22,23,24,25]. Wang, Suo et al. have investigated the influence of titanium powder in the interlayers of a WAAM structure. They note a grain refining effect and increasing tensile strength in the weld metal [26].

The state of research shows that titanium and titanium diboride can have a grain refining effect and influence the microstructure. This effect can be seen both in casting tests and in the welding process, into which the elements are introduced in various ways. In contrast, the possibility of microalloying elements being added to the weld metal of aluminum materials via coatings is not yet widely investigated. The aim of this work is to work out general interactions in the processing of thin-film-coated aluminum welding wire electrodes and their influence on the weld metal and the welding process. The effectiveness of the coatings is to be investigated.

2 Experimental

The tests carried out deal with the GMAW welding process of the aluminum alloy S Al 5754 (AlMg3), according to ISO 18273 [27], which tends to be susceptible to hot cracking. Titanium and titanium diboride (TiB2) thin-film coatings are to be used to achieve grain refinement of the weld metal structure, which can potentially reduce susceptibility to hot cracking. Furthermore, the coatings influence the arc length, the energy per unit length, and the fusion penetration.

2.1 PVD-coating

For coating the wire electrodes, a PVD system from CemeCon AG according to industrial standards is used. Figure 1 shows the coating chamber of the system, which is loaded with the substrate wire electrode in batches.

Fig. 1
figure 1

PVD coating system with substrate electrode

Full size image

Titanium and titanium diboride are applied to the substrate wire electrode in three different layer thicknesses each. Table 1 summarizes the coating thicknesses of the thin-film coatings resulting from the selected coating duration and the titanium content in the weld metal determined by optical emission spectrometry. To determine the coating thicknesses, a half-masked and polished base material sample is added to the respective coating process. After the process, the height difference between the coated and uncoated areas respectively the coating thickness can be measured using a confocal laser scanning microscope. As expected, the coating thickness increases with the coating duration resulting in an increased titanium content in the weld metal.

Table 1 Coating time, resulting layer thickness, and element content
Full size table

2.2 Welding and grain size

The coated welding wire electrodes are welded using the parameters summarized in Table 2.

Table 2 Welding parameters
Full size table

To evaluate the microstructures and in particular to determine the grain size, beads on a plate are welded on substrate material EN AW-5754 according to [28]. The substrate plates are preheated to 100 °C for this purpose. At a length of 70 mm, micrographs are taken from the center of the weld beads, cf. Figure 2.

Fig. 2
figure 2

Comparison of the grain size in the micrograph. Left side, welding wire AlMg3 uncoated. Right side, welding wire AlMg3 with TiB2-coating

Full size image

The weld metal of the uncoated reference wire electrode is shown in comparison with a TiB2-coated material. To analyze the microstructures, the microscopic specimens are electrolytically contrasted using the Barker method. For this purpose, the samples are etched 5 times for 40 s at a voltage of 12 V and viewed under polarized light. The influence of TiB2 on the grain size is evident. To quantify the grain refining effect, five images are taken in each case in the globular dendritically solidified region of the microstructure and then evaluated using the linear intercept method.

2.3 Modified varestraint-transvarestraint test (MVT-test)

To determine the influence of the thin-film coatings on the susceptibility to hot cracking, the modified varestraint-transvarestraint test is carried out in accordance to the DVS code of practice 1004–2 [29] and DIN Technical Report 17,641–3 [30]. In contrast to the test procedure described in that documents, the MIG welding process with a consumable electrode is used.

Figure 3 shows the test setup. For the test, specimens with the dimensions 60 × 300 × 10 mm are inserted into the hydraulic test stand. Starting at the edge of the plate, welding is performed in the direction of the center of the plate using the parameters given in Table 2. The samples are also preheated to 100 °C for the MVT test.

Fig. 3
figure 3

Modified varestraint-transvarestraint test

Full size image

The straining process is triggered at the same time as the torch is switched off in the middle of the substrate plate and the samples are stretched to 1%, 2% and 4% with different matrices. At least 3 specimens are welded for each coating and strain. The weld is then cut out, and the hot-cracked end crater and parts of the weld surfaces on which hot cracks are visible are imaged under a scanning electron microscope. Image analysis software then allows manual measurement of the individual cracks (Fig. 4), which can finally be summed up to a total crack length.

Fig. 4
figure 4

Hot cracks at the end of a welding seam (SEM)

Full size image

3 Results and discussion

Figure 5 shows the average grain size as a function of the titanium content, for the titanium diboride and for the pure titanium coating.

Fig. 5
figure 5

Grain size as a function of titanium content

Full size image

It gets obvious that the grain size tends to decrease with increasing titanium content in the studied area, in accordance with the theory and in agreement with the research work described further above.

The results of the modified varestraint-transvarestraint test are shown in Fig. 6 for titanium and in Fig. 7 for titanium diboride. Both coatings reduce the susceptibility to hot cracking. In particular, the welds with the lowest element contents and thus lowest coating thicknesses lead to the lowest hot cracking susceptibilities compared to the reference.

Fig. 6
figure 6

The result of the trans-varestraint test for Ti-coatings

Full size image
Fig. 7
figure 7

The result of the trans-varestraint test for TiB2-coatings

Full size image

The influence on the arc is evaluated by high-speed recordings. For this purpose, the substrate plates are moved under the stationary torch with the aid of a robot and the welding speed of the welding process. In this way, high-speed images can be taken at 5000 frames per second over a period of 1 s using the backlight method. In addition, the welding current and the voltage curves of the welding process are recorded simultaneously. To evaluate the high-speed images of the arc, the individual images of each recording (Fig. 8) are converted into binary images using a thresholding (gray scale) method. The arc length is then measured in the converted images. The angle of the arc is approached at the upper end via the center of the arc 10% below the attachment point at the anode and via the center of gravity of the arc in the two-dimensional image. Figure 9 shows the binary image cropped to the arc. The distance the arc is measured along is shown in red.

Fig. 8
figure 8

The video recording of the welding arc

Full size image
Fig. 9
figure 9

The binary image of the arc with measuring path

Full size image

To determine the average arc length, the arc lengths of the individual images of the high-speed recording are averaged over the measured period. The changes in arc length in percent compared to the reference arc (uncoated wire electrode) are summarized in Table 3.

Table 3 Arc length change, welding current, and arc voltage depending on the coating
Full size table

It can be demonstrated that the arc length decreases significantly with increasing coating thickness. In the tests carried out, the electrode with the largest TiB2 layer thickness shows the highest process influence. When the electrode is processed, the arc shortens by 44% compared with the reference. At the same time, the welding current increases by 17% with nearly unchanged arc voltage.

The conductivity of the arc is significantly depending on the degree of ionization and in particular on the metal vapor, since the metallic components in the arc plasma to a large extent carry the arc current. Even traces of metal vapor change the ionization state in the arc column [31]. The coatings can lead to an increase in the metal vapor content and thus to increased electrical conductivity of the arc respectively to a decrease in the length-specific arc resistance. Due to the constant voltage characteristic used in gas metal arc welding, the measured voltage remains nearly constant. The changed operating point of the arc characteristic results in a shorter arc. As the length-specific arc resistance decreases, the welding machine increases the welding current due to the characteristic curve, as also shown by the measurements. Macroscopically, this is manifested by a shorter arc. Figure 10 shows images of the arc with different coatings. The decreasing arc length with increasing coating thickness is visible in the figures.

Fig. 10
figure 10

Influence of coatings on the arc length

Full size image

Another consequence of the increased metal vapor content is the decrease in arc column temperature associated with a wider arc core. The penetration thus becomes flatter and wider. The influence on the fusion penetration shape and the fusion penetration depth of the TiB2 coatings is clearly shown by micrographs. Figure 11 compares the reference wire electrode with the two higher coating thicknesses. It is noticeable that the penetration decreases with increasing coating thickness and excess weld increases at the same time.

Fig. 11
figure 11

Fusion penetration and excess weld of the welding seam depending on the TiB2-coating

Full size image

Changes in the penetration shape can be attributed to the fact that the electrically conductive area of the arc expands due to the low ionization energy of the metal vapor and the decreasing current density [32]. It is noticeable that the pore content increases as the TiB2 content rises. With increasing coating thickness, an increasing hydrogen content in the wire electrodes was observed. However, only very low film thicknesses are required to achieve a significant grain refinement effect, as demonstrated here. Further research is challenged to find the optimum point in terms of film thickness, grain refinement effect, and porosity. In addition, the process parameters of the PVD process can be influenced, possibly leading to a reduction in the hydrogen content of the wire electrodes.

4 Conclusions

The investigations carried out show that thin-film coatings can be an interesting tool for influencing the weld metal and the welding process in the welding of aluminum wire electrodes. It was demonstrated that nucleating elements can be introduced into the weld metal via a coating of the filler metals and that a grain-reducing effect can be achieved in this way. By influencing the weld metal and the grain refinement caused by the microalloying elements, the susceptibility to hot cracking can be reduced with a coating of titanium diboride and also with a titanium. In particular, low coating thicknesses show a significant hot crack-reducing effect at high strains. The weld metal with a titanium content of 0.103 wt.% brought in by a titanium diboride coating and the weld metal with a titanium content of 0.119 wt.% brought in by a titanium coating have the lowest hot cracking susceptibility.

In addition, there are clear effects with regard to the welding process, which were illustrated by considering the arc length and the welding process energy. At almost constant arc voltage, the arc length decreases, and the welding current increases with increasing coating thickness. Micrographs show that TiB2 coatings make the fusion penetration flatter and wider, which can be explained by the influence of the arc plasma via a higher metal vapor content.

Depending on the coating element, high coating thicknesses lead to a correspondingly higher hydrogen input into the welding process and thus to higher porosities in the weld metal. For some elements, however, the coating thicknesses required for positive weld metal and process effects are far below a critical range in terms of hydrogen-induced porosity.

5 Outlook and future research

In addition to titanium and titanium diboride, the effect of other coating elements on the weld metal and the process is worth investigating. It is also worthwhile to extend the investigations to other substrate wire electrodes. The substrate wire electrode chosen here is a naturally hard alloy. Research should be extended into the area of heat-treatable alloys. Initial results show an influence on the mechanical-technological properties and thus the need for further research in this respect. In terms of the coating process, further investigations can be made to reduce the inherent hydrogen content of the coatings.