Liquid Metal Embrittlement (LME) during Resistance Spot Welding (RSW) can cause cracks when welding advanced high strength steels. Recent advances in steel metallurgy, resistance spot welding processing and accompanying simulation tools have substantially improved the way that LME can be handled in industrial practice. This article gives a brief overview of easy measures to implement when LME might potentially occur during production.
Introduction
During resistance spot welding of zinc-coated advanced high strength steels (AHSS) liquid metal embrittlement -related cracking may be observed. Since LME is often associated with a reduction of steel’s mechanical properties, it is desired to control its occurrence during production. An exemplary LME crack, forced with increased weld heat and deliberate electrode misalignment, is shown below.
Figure 1: A typical LME crack created under laboratory conditions by deliberately increasing the welding time and introducing 5° electrode tilt
Over the past several years, LME has been a a focus in welding research. It is now well-understood to the degree that it can be predicted and avoided with easy measures. Below is an overview of four key steps to address the potential of LME during automotive production.
Obtain the latest steel grades from your steel supplier
Over the past decade, steel producers have released AHSS with improved chemical compositions, helping to significantly reduce the occurrence of LME iIt is beneficial to talk with steel suppliers and ask about their latest AHSS grades, as these are likely far less sensitive to LME than previously tested grades. A recent study commissioned by WorldAutoSteel[Link to LME component study] demonstrated that all five chosen material stack-ups from current production data did not show any LME even under exacerbated conditions. Only by choosing an especially difficult material stack-up could LME be forced to appear at all to conduct the study.
Read up on the current state of research for LME
WorldAutoSteel has published two studies on liquid metal embrittlement: One focused on lab conditions and the second on real-life stamped components. These studies provide an overview of all aspects of LME and how to manage and avoid LME issues. [Link to both studies]
Establish in-house testing protocols to gauge the sensitivity of your material stack-ups
To investigate LME in-house, it’s critical to establish a testing protocol that forces the cracks to appear and allows for comparison of different steels, stack-ups and welding parameters. as there There is currently no industry-wide agreed-upon testing standard.
Still, there is a good selection of well-documented procedures to choose from. The easiest procedure is to increase the welding time until cracks start to appear – keep in mind that you need to remove the zinc coating before you can observe any cracks on the surface.
Other methods are based on so-called “Gleeble testing” or on deliberately introducing imperfections like tilted electrodes or large gaps into the welds. As you establish a testing procedure in your lab, you can use it to evaluate LME occurrence in the stack-ups that you want to implement into body-in-whites.
Think about implementing LME mitigation strategies in your most difficult welds
Suitable measures should always be adapted to the specific use case. Generally, the most effective measures for LME prevention or mitigation are:
Avoidance of excessive heat input (e.g. excess welding time, current)
Avoidance of sharp edges on spot welding electrodes; instead use electrodes with larger working plane diameter, while not increasing nugget-size
Employing extended hold times to allow for sufficient heat dissipation and lower surface temperatures
Avoidance of improper welding equipment (e.g. misalignments of the welding gun, highly worn electrodes, insufficient electrode cooling)
These measures can be implemented in the planning stage and in an ongoing production environment to increase the LME-free process windows.
In Conclusion
While Liquid Metal Embrittlement may present a challenge when welding AHSS, it’s no longer an unpredictable threat. Thanks to advancements in steel development, welding techniques, and testing methods, manufacturers have the tools they need to reliably mitigate LME during production.
Staying informed, working closely with steel suppliers, implementing smart testing protocols, and applying targeted welding strategies can help automakers maintain both strength and quality in AHSS joints. With this proactive approach, LME doesn’t have to stand in the way of innovation in automotive manufacturing.
This is a summary of a paper of the same title, authored by K. Májlinger, E. Kalácska, and P. Russo Spena, used by permission.M-65
Researchers at the Budapest University of Technology and Economics and the Free University of Bozen-Bolzano tested gas metal arc welding (GMAW) of dissimilar Advanced High-Strength Steel (AHSS) sheets.M-65 The test pieces were 100 x 50 mm samples of 1.4 mm TWIP (TWIP1000) and 0.9 mm TRIP (HCTC800T) sheet steels were welded in a lap joint configuration with 0.8 mm diameter AWS ER307Si austenitic stainless steel wire to determine appropriate GMAW parameters for good quality welds. Quality was determined by external appearance, microstructure, and mechanical properties. Good welds were achieved with linear heat inputs (Q) with ranges from 500-650 kJ/m. The only fractures that occurred appeared within the weld bead by ductile failure modes. The HAZ of the TWIP steel showed grain coarsening and the HAZ of the TRIP steel experienced microstructural changes relative to the distance from the fusion boundary. The ultimate tensile strength (UTS) varies between 73%-84% of the weaker of the two steels.
Welding was conducted with an automated linear drive system with pure Argon (99.996% Ar) shielding gas at 10L/min. Wire feed rates were approximately 3.5 m/min with Direct Current Electrode Positive (DCEP) polarity. Changes in current, voltage, weld speed, and the resulting linear energy are compared in Table 1.
Figure 1: Overview of Dissimilar AHSS GMAW Welding.M-65
Table 1: Results of the preliminary welding tests in terms of TWIP-TRIP joint quality.M-65
After welding, transverse sections were cut from the welds and etched to show the microstructure. Vickers hardness testing was conducted on the weld samples based on the ASTM E384 standard. Tensile tests were performed on the samples according to the EN ISO 6892-1 standard. Tests were also conducted on unwelded TWIP and TRIP steels for comparison. Scanning electron microscopy (SEM) examinations were made of fracture surfaces to determine failure modes and examine for microscopic weld defects.
The study concluded that dissimilar welds between AHSS steels with the GMAW process can be achieved with consistent results desired for automotive applications.
Quenching and partitioning (QP) steels are one of several third generation advanced high strength steel formed by controlled martensite phase fractions and retained austenite. Researchers from the University of Shanghai Jiao Tong tested the effect of HAZ softening in a QP1180 lap joint with the GMAW cold metal transfer (CMT) process.W-1 The steel was welded with ER130s electrode. The fusion zone consisted of chiefly acicular ferrite. The supercritical zone consists of martensite, which is harder than the base metal. There is a drop in hardness (100 HV) in the subcritical zone, and there is a noted lack of retained austenite present in the microstructure. Precipitates are also present in the subcritical zone. The intercritical HAZ only experiences mild softening; where fresh martensite has formed. The softening in the subcritical HAZ presents room for failure that must be accounted for when planning welding using CMT.
Figure 1 shows the different microstructures in detail throughout the weld and the base metal. In the softening zone shown in Figure 2 correlates to the tempered sorbite region in Figure 1 (part f). The hardness maps shows that the fusion zone is approximately as hard as the base metal, with the supercritical HAZ having the highest hardness values before rapidly softening in the intercritical and subcritical HAZ zones. The softened region must be accounted for in designing CMT or other high heat input welded components.
Figure 1: a) Base metal with retained austenite between martensite lathes, b) fusion zone with acicular ferrite and ferrite, c) coarse martensite grains in supercritical HAZ, d) fine martensite grains and ferrite in supercritical HAZ, e) intercritical HAZ with fresh martensite and untransformed martensite, f) softest region in subcritical HAZ; shows tempered sorbite, g)zoomed in region of subcritical zone with precipitates, h) subcritical HAZ with 350-360HV hardness, i) hardness with 380-390HV.W-1
Figure 2: a) macrostructure of the CMT-welded QP1180 joint, b) hardness chart through joint, c) hardness map in upper sheet, d) hardness map in lower sheet. 1) fusion zone, 2) supercritical HAZ, 3) intercritical HAZ, 4) subcritical HAZ, BM = base metal.W-1
A common issue when welding Advanced High-Strength Steels (AHSS) is with protective coatings causing weld defects. A group of researchers at the NMAM Institute of Technology and Dong-Eui University studied common issues with gas metal arc welding (GMAW) in the cold metal transfer (CMT) mode on a zinc-coated steel.V-2 The study used infrared thermography to observe the welds as they were created, helping to get detailed observations on some defects appearing in real time. With GMAW in CMT mode, the prevailing defect with welding a zinc-coated steel was porosity from metal vapors escaping through the weld. This issue could be addressed by adjusting the heat input and travel speed to provide more time for metal gases to escape.
In Figure 1, it shows that with a higher heat input, more heat is in the weld puddle. In low and medium heat inputs, the puddle is above melting temperature, but not as high as the high heat input. Figure 2 shows that the low heat input also has the fastest solidification rate, and the high heat input has the slowest solidification rate. Figure 3 shows where the zinc vapors from the molten coating evaporate through the weld. In the left picture, at low heat input, the nucleation is contained inside of the weld, and the fusion zone would collect in the fusion zone. In the middle picture, at medium heat input, the zinc vapors bubble out just as the metal starts to solidify. In the right picture, at high heat input, the zinc bubbles out in the weld puddle while it is still molten.
Figure 2: Variation of temperature during CMT for High, Medium, and Low Heat Input.V-2
Figure 3: Variation of Zinc Porosity Position vs Low, Medium, and High Heat Input.V-2
These factors combined indicate several factors that influence zinc porosity in GMAW CMT weldments. The researchers concluded that at low heat inputs, the zinc collects in the fusion zone. At medium heat inputs, the solidification rate and temperature gradient through the weld puddle traps the zinc in the fusion zone but also allows some to bubble out through the weld puddle. This caused the worst material properties of the three weldments for the researchers. At high heat inputs, the zinc bubbles out through the weld puddle, before solidification occurs. This condition is optimal, to reduce porosity with zinc metal vapors, the heat input should be increased so that the weldment temperature increases and solidification rate decreases.
In production, part geometry or joint application requires the use of gas metal arc welding (GMAW) to weld the joint. A commonly used Advanced High-Strength Steel (AHSS) is Dual Phase (DP) 600 which contains a hard martensite phase in a ferrite matrix (approximately 5-20% martensite). Under the heat input from GMAW, this microstructure near the weld [in the Heat Affected Zone (HAZ)] is destroyed, and a new microstructure develops. Researchers from RWTH Aachen University in GermanyR-24 used representative volume elements (RVE) in tandem with electron probe microanalysis and micromechanical finite element (FE) modeling to develop flow curves for 2.5 mm hot rolled DP 600 steel sheet. This can be used to help predict mechanical properties in the HAZ. The researchers observed bainite, coarse grain ferrite, and tempered martensite in the HAZ.
The researchers developed an engineering stress-strain curve for the GMA welded DP 600 steel as depicted in Figure 1. The stress distribution concentrates to the outside edge of the HAZ as shown in Figure 2, primarily in the softest regions of the HAZ (Figure 3). The soft region is where ductile failure is observed as seen in Figure 4. The soft region is a result of a loss of bainite and an increase in ferrite grain growth. Because ferrite is steel’s softest phase, it results in this soft region where plastic strain accumulates. The increasing presence of tempered martensite starts to raise hardness after this region.
Figure 1: Stress-Strain Curve for GMA Welded DP600 Steel.R-24
Figure 2: Hardness Map through FZ and HAZ with emphasis on softened zone.R-24
The HAZ is composed of a variety of microstructures that vary depending on their distance from the centerline. Close to the middle of the fusion zone, the microstructure is almost 100% bainite with small amounts of ferrite and martensite. Bainite is harder than the tempered steel/ferrite combination, which accounts for the hardness of the fusion zone. Away from the fusion zone, the bainite decreases, and the ferrite increases to where the microstructure is roughly 90% ferrite and 10% martensite, with no bainite in the microstructure (Figure 5).
Figure 5: Phase fractions relative to weld centerline.R-24
This article summarizes a paper by W. Mohr and N. Kapustka, EWI, entitled, “Fatigue of GMAW-P Lap Joints in Advanced High-Strength Steels.”M-13
EWI has performed fatigue tests on welds from four Advanced High-Strength Steels (AHSS) in the uncoated condition. The materials were provided in three thicknesses as follows, 2.0-mm DP 780, 1.8-mm 590 SF, 2.0-mm DP 980, and 2.8-mm CP 800. Referring to Figure 1(a), welding parameters were selected to meet the weld profile requirements listed below:
The travel speed to achieve such combinations was 23 mm/s for three of the sheets and 13 mm/s for the 2.8-mm-thick CP 800. Figure 1 shows a completed panel.
Specimens were cut from the lap-welded panels in a configuration recommended by Z 2275, with minimum reduced sections of 20-mm wide, with 20-mm radii on both sides to a full width of 30 mm, as shown in Figure 2.
Fixtures for the bend testing had eight, 6.3-mm radius rollers, four on top and four on the bottom, with offsets of the roller centers to accommodate the lap-joint configuration and the differing sheet thicknesses. The interior span was 120 mm, while the exterior span was 210 mm. The full bending fixture, with a specimen inserted, is shown in Figure 3.
Weld profiles were achieved that met the weld profile requirements for each sheet material type. These weld profiles are shown for the four sheet materials in Figure 4. Fatigue testing results in tension at R = 0.3 gave lifetimes between 30,000 and 9 million cycles, with run-outs at 10 million cycles, as shown in Figure 5.
Figure 4: Cross Sections of Lap Joints (etched with 2% Nital).M-13
Figure 5: Results of Fatigue Testing in Tension at R= 0.3.M-13
Weld root cracking dominated in the 590 SF, as well as the DP 780 and DP 980, with an example shown in Figure 6. Weld toe cracking was observed on the 2.8-mm-thick CP 800, with an example shown in Figure 7.
Figure 6: Example for a Root Crack Breaking Through the Weld Metal on DP 980.M-13
Figure 7: Example of a Toe Crack Breaking Through the Base Metal.M-13
Fatigue testing in bending at R = -1 gave lifetimes between 30,000 and 2 million cycles, with run-outs on tests that continued to up to 7 million cycles, as shown in Figure 8.
Taking the differing thicknesses, minor variations in minimum width, and the stress concentrations from the radii into account, the concentrated stress range was calculated to compare the four materials on a common basis, as shown in Figure 9.
Figure 9: Concentrated Stress Range versus Lifetime for Tension Tests.M-13
The fatigue cracks initiated at the root for the 1.8-mm 590 SF on both tension and bending testing. The fatigue cracks initiated at the weld toe for the 2.8-mm CP 800 on both tension and bending testing. The fatigue cracks initiated from the weld root in the tension testing and from primarily the weld cap in bending testing, for the 2.0-mm-thick DP 780 and 2.0-mm-thick DP 980.
