Journal of Materials Processing Tech.

288 (2021) 116917

Contents lists available at ScienceDirect

Journal of Materials Processing Tech.

journal homepage: www.elsevier.com/locate/jmatprotec

Occurrence of sub-critical heat affected zone liquid metal embrittlement in

joining of advanced high strength steel
C. DiGiovanni a, *, A. Ghatei Kalashami a, F. Goodwin b, E. Biro a, N.Y. Zhou a
a
Department of Mechanical and Mechatronics Engineering, University of Waterloo, 200 University Ave. West, Waterloo, ON, N2L 3G1, Canada
b
International Zinc Association, Durham, NC, 27713, USA

A R T I C L E I N F O A B S T R A C T

Associate Editor: Jian Cao Advanced high strength steels (AHSS) used in automotive structural components are commonly protected using
zinc coatings. However, the steel/zinc system creates the potential for liquid metal embrittlement (LME) cracking
Keywords: during welding. In this study, the effect of angular welding electrode misalignment as recommended by industry
Resistance spot welding was investigated and resulted in LME cracking in the sub-critical heat affected zone (SCHAZ) which is a region
Liquid metal embrittlement
where LME has not been previously observed. The largest misalignment case of 6◦ resulted in bending of the
Advanced high strength steel
work piece and an angled weld nugget/heat affected zone (HAZ), causing LME cracks in the lower temperature
Weld zone characterization
Scanning electron microscopy HAZ regions. SEM and EBSD analysis characterized the LME crack area as the SCHAZ and below the Ac1
Cracking transformation temperature. Furthermore, dilatometry analysis showed the Ac1 temperature to be approximately
Electron backscatter diffraction 660 ◦ C, meaning LME occurred below the traditional 700 ◦ C minimum threshold. In addition to this novel
observation of LME before transformation and at low temperature, it was also noted that only misalignment
beyond 2◦ promoted LME cracking. The low angle 2◦ case was observed to be statistically the same using a
Student’s t-test (α = 0.05) on 6 welded samples for LME crack severity compared to a standard 0◦ case.

1. Introduction ◦
C was identified and it was concluded LME only occurs at these tem­
peratures. However, it was also noted that the minimum LME temper­
To improve passenger safety and vehicle fuel economy, automotive ature lowers with increasing strain rate, which was attributed to the
companies are using higher strength steels in the body-in-white. Use of higher flow stress of the steel sample, resulting in increased
higher strength steels allows thinner materials to be used for part con­ stress-assisted diffusion of zinc to the steel GBs (Beal, 2012). Addition­
struction without decreasing part strength. Increasing steel strength in ally, Kang et al. (2016) examined twinning-induced plasticity (TWIP)
automotive design has been an on-going trend over the past 20 years steel, press-hardened steel (PHS), and interstitial-free (IF) steel using
which has led to the development of advanced high strength steels thermomechanical testing, and found zinc penetration to be the most
(AHSS). In addition, a zinc coating is used to protect AHSS from corro­ frequent in the TWIP steel, with no evidence of zinc penetration in the IF
sion during service. The underlying AHSS substrate is given cathodic steel. Since the IF steel was tested both below the Ac1 temperature (no
protection by a thin layer of zinc. microstructural transformation) and above the Ac3 temperature
Liquid metal embrittlement (LME) is a cracking phenomenon that (microstructure completely transformed to austenite), it was concluded
results in a loss of ductility because of premature failure along grain that LME is not directly related to crystal structure. It was theorized that
boundaries (GBs). For LME to occur Ashiri et al. (2015) found that three the IF steel’s low stress levels and lack of alloying elements did not allow
factors must be present: a liquid embrittler must be in direct contact with for zinc to penetrate the AHSS GBs. Jung et al. (2016) studied LME in
the solid metal, the substrate must have a susceptible microstructure, AHSS with different microstructures such as TWIP steel, DP steel and
and the substrate must be under tensile stresses. In recent years it has drawing quality steel. Similarly, Jung also concluded the crystal struc­
been determined that AHSS with certain compositions and microstruc­ ture was not directly responsible for LME since the phenomenon was
tures in contact with liquid zinc are highly susceptible to LME. Beal et al. observed in different structures. Similar to the previous work, a tem­
(2012) extensively studied zinc LME in AHSS using thermomechanical perature range was identified for LME, which depended on strain rate
Gleeble testing. A ‘ductility trough’ temperature range of 700 ◦ C – 900 and steel type. Overall, the present literature agrees there is a set

* Corresponding author.
E-mail address: [email protected] (C. DiGiovanni).

https://doi.org/10.1016/j.jmatprotec.2020.116917
Received 20 June 2020; Received in revised form 26 August 2020; Accepted 27 September 2020
Available online 1 October 2020
0924-0136/© 2020 Elsevier B.V. All rights reserved.
C. DiGiovanni et al. Journal of Materials Processing Tech. 288 (2021) 116917

Table 1
Chemical composition and tensile properties of the investigated TRIP 1100 steel.
Material Main alloying elements (wt.%)

C Mn Si Cr Al Mo
0.20 2.17 1.61 0.03 – –
Mechanical Properties
TRIP1100
Thickness Yield Strength Tensile Strength
% Elongation Top coating thickness (g/m2) Bottom coating thickness (g/m2)
(mm) (MPa) (MPa)
1.6 861 1100 14.6 58 70

temperature range for LME and identifies the influence of strain rate and
stress for the onset of LME.
LME in steel grades in contact with liquid zinc has been observed in
several high temperature material processing practices such as hot
stamping by Cho et al. (2014) and press-hardening by Lee et al. (2012).
However, the occurrence of LME during resistance spot welding (RSW)
has only received attention in the past several years and research in this
field is still ongoing. The most common joining process in automotive
assembly is RSW. During RSW, steel sheets are overlapped, and elec­
trodes apply a clamping force. By passing a high current through the
steel sheets, Joule heating occurs which generates the required heat
input. The generated heat melts the material at the faying surface,
forming a molten nugget. At the end of the weld cycle, the weld current
ceases and the joint remains fastened during the hold time, promoting
solidification under the force and cooling of the traditionally
water-cooled electrodes. However, the protective zinc coating in­
troduces complications to the welding process. Beal et al. (2012) found
the contact of the liquid zinc (embrittler) and solid AHSS (susceptible
substrate), which occurs at the electrode/sheet interface due to its
moderate temperatures (600 ◦ C–900 ◦ C), leads to zinc penetration into
AHSS grain boundaries, which are also stressed from the welding pro­
cess; this allows the development of LME cracks during RSW.
Much of the recent research in the field of LME during RSW has
focused on welding process modifications for LME reduction during
welding. Wintjes et al. (2019a) used optimized weld current pulse and
Fig. 1. Schematic of angular electrode misalignment with α denoting the angle
pre-current to reduce LME cracking during RSW. Similarly, Ashiri et al. of misalignment.
(2016) used a multi-pulse weld schedule to control thermal stress and
thus LME formation. Finally, Murugan et al. (2018) studied the impact
2.2. Welding process conditions
of RSW electrode geometry on LME severity. However, few studies have
examined the impact of manufacturing process conditions on LME
Resistance spot welding was completed using a trans-gun direct
cracking. Frei et al. (2019) studied LME during externally loaded
current spot welding machine mounted on a 6-axis robot arm. The
welding and found LME cracking to worsen in DP steel when loading
electrode gun is a C-type clamp gun with a servo-controlled electrode
exceeded 60 % of the yield strength. Although restraint conditions are
force and a Rexroth welding controller. The robot arm was programmed
important for examining manufacturing conditions, there is no work
to weld at four electrode angle cases that are relevant to industrial en­
investigating the impact of angular electrode misalignment on LME
vironments including 0◦ , 2◦ , 4◦ , and 6◦ . Fig. 1 shows an illustration of
sensitivity. Previous work by Xing et al. (2018) has shown that welding
the off-axis angled welding. Welding was carried out on 100 × 25 mm
with angular misalignment results in uneven electrode indentation,
coupons of TRIP1100 clamped on both ends, using the same weld
resulting in an angled weld nugget and heat affected zone (HAZ). The
schedule for all four cases. The electrode geometry and two-pulse
present study investigates the impact of electrode angular misalignment
welding schedule was selected based on the American Welding Society
on LME crack development, given the introduction of mechanical
D8.9 standard (American Welding Society, 2012) at a heat input just
stresses and weld sample deformation. Electrode angular misalignment
below the expulsion current (Imax -10 %) as shown in Table 2. This heat
is a common issue in production environments and the impact on LME
input was selected to promote LME cracking for clear observation while
during RSW is not well understood. An LME sensitivity range to angular
not exceeding the maximum expulsion current. All cases produced a
misalignment was determined and a severity progression is presented.
weld nugget size of 6.5 mm with a 95 % confidence interval of 0.2 mm.
Furthermore, LME crack location in the HAZ is characterized, providing
Samples for LME observation were prepared through conventional
key insights into the necessary temperatures for the onset of LME.
procedures, i.e. mounting in conductive epoxy resin (Polyfast), grinding
with SiC sandpaper up to 1200 grit, polishing with 1 μm diamond dust to
2. Materials and methods
allow for clear observation and measurement of LME cracking in the
weld area. LME cracks were observed and measured using an optical
2.1. Materials
microscope (Oxford BX51 M). LME severity was quantified from crack
measurements of the welded samples using the LME crack index
This study was carried out on hot-dip galvanized coated
developed by Wintjes et al. (2019b). Measurements were taken for both
transformation-induced plasticity (TRIP) 1100 AHSS, as it is highly
the crack depth and number of cracks, which have both shown to be an
susceptible to LME and is of interest to industry. The chemical compo­
important factor in governing post-weld strength by DiGiovanni et al.
sition and mechanical properties can be found in Table 1 below.
(2019). The crack index quantifies LME crack severity of a welded

2
C. DiGiovanni et al. Journal of Materials Processing Tech. 288 (2021) 116917

Table 2
Welding parameters used for the study.
Material Current (kA) Time (ms) Force (kN) Hold Time (ms) Electrode Type Electrode Dia (mm) Flow Rate

TRIP1100 9.5 200-33-200 5.0 167 F1-7 (Ball Nose) 7 6 L/min

Fig. 2. Plots of LME crack severity and nugget size for each electrode angle case.

cross-section so an accurate comparison can be made between welding angle of 4◦ and 6◦ . The results are summarized in Fig. 2 with a 95 %
conditions. The crack index is defined as: confidence interval. LME severity appears to worsen as electrode angle
increases, as an increasing trend is shown. However, it is important to
nL
CI = (1) note that no increase in LME was observed in the 2◦ case compared to
τ
the 0◦ case. The results of a Student’s t-test using α = 0.05 on the 6
Where n is the mean number of cracks per sample, L is the lognormal welded samples showed there to be no statistical difference between the
median crack depth and τ is the sheet thickness. The crack index was cracking severities resulting from both conditions. The weld nugget
calculated using six samples for each condition. diameter was measured for each case, which showed a consistent nugget
size of 6.5 ± 0.2 mm. This is expected given the heat input for each case
remained constant. It is interesting to note that in previous work nugget
2.3. Heat affected zone characterization
size was stated to be directly related to LME development during RSW
(Ashiri et al., 2016) but here there is a separation of nugget size and LME
Samples for microstructural observation were prepared in the same
severity as shown in Fig. 2. Having electrode misalignment introduces
fashion as for LME observation with addition of etching with 2 % Nital
bending into the work piece, which results in external mechanical stress.
(2 % HNO3 + 98 % ethanol). Microstructures were characterized using a
Stress has been described by both Kang et al. (2016) and Beal (2012) as a
field emission scanning electron microscope (FESEM, UltraZeiss).
critical condition for the onset of LME. The increase in LME severity is
Electron back scattered diffraction (EBSD) analysis carried out by
due to the increased mechanical distortion from welding with an elec­
field emission SEM (JEOL7000 F). The EBSD samples were prepared
trode angle and was observed to be localized to the shoulder region. A
through conventional sample preparation procedure and a final vibra­
progression of increasing LME crack size with each angle case is shown
tory polishing step for 4 h. A high-speed Hikari CCD camera was used for
in Fig. 3.
pattern acquisition at binning of 2 × 2 which provided proper indexing.
Similar to the crack index plot, an increasing trend in crack size is
Data were recorded at 0.2 μm step size.
observed but there is no discernable difference between the 0◦ and 2◦
case (see Fig. 3). These results show that the change in stress and thermal
3. Results and discussions fields of the 2◦ case was not sufficient to promote a detectable difference
in LME. Therefore, it can be said that LME cracking during RSW is not
3.1. Liquid metal embrittlement severity sensitive to low angular electrode misalignment. However, a steep in­
crease in LME is observed once the threshold has been crossed, leading
The results show a clear sensitivity of LME cracking to an electrode

Fig. 3. Typical weld shoulder LME cracks for each case showing a progression of crack size.

3
C. DiGiovanni et al. Journal of Materials Processing Tech. 288 (2021) 116917

Fig. 4. Example cross-sections of the (a) 0◦ case (b) 2◦ case (c) 4◦ case (d) 6◦ case showing the weld nugget and HAZ shape with large LME cracks highlighted in
yellow (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article).

from the weld shoulder, LME cracks still form at that location. The
consistency of LME at the weld shoulder shows the stresses experienced
during welding are still sufficient for the onset of LME. Additionally, the
distortion from the misaligned electrodes leads to a progressively larger
crack in this region (as seen in Fig. 3) even though peak temperature in
this location decreased with increased misalignment. In the extreme
case of 6◦ , a large LME crack has occurred far outside the UCHAZ
(Fig. 4d). To confirm if LME cracking has occurred outside the UCHAZ or
ICHAZ a full characterization of the HAZ is required.

3.2. Heat affected zone characterization

Fig. 5. Schematic showing the HAZ moving away from the weld shoulder with Although etching contrast implied that LME cracking occurred in the
increasing angular misalignment while the LME crack location is stationary. SCHAZ, considering the past work by Beal et al. (2012) which showed
LME sensitivity required material to be heat between 700 ◦ C – 900 ◦ C, a
to critical LME, despite being below the expulsion condition. Examples microstructural analysis of the HAZ was completed. Fig. 6 shows the
of the welded cross-sections are shown in Fig. 4, which shows the optical microscope (OM) micrograph and detailed SEM images of the
severity of the 6◦ case. various microstructures post-welding. The microstructure of the base
Fig. 4 highlights deep LME cracks are in the weld shoulder area, on material (BM) in Fig. 6a is comprised of a ferrite matrix with embedded
the side bent in and receiving higher clamping pressure, when large bainite and martensite. Additionally, the microstructure of the SCHAZ
misalignment is introduced. However, a more detailed inspection of the (Fig. 6b) where the temperature during welding did not exceed Ac1 is
weld in Fig. 4d shows LME cracks present in the sub-critical heat affected similar to that of BM structure but includes tempered martensite. No
zone (SCHAZ) in addition to the upper-critical heat affected zone significant changes were observed in terms of the volume of bainite but
(UCHAZ) and inter-critical heat affected zone (ICHAZ). In the present tempered martensite is a clear characteristic of the SCHAZ. The micro­
literature, LME cracks during RSW have only been observed in the structure of the ICHAZ (Fig. 6c), where the temperature during welding
UCHAZ or ICHAZ above the Ac1 temperature. However, in this case, ranged between Ac1 and Ac3, consists of fine martensitic islands in a
cracks can be observed in the UCHAZ and in the outer lighter HAZ re­ ferritic matrix. Also, in this region, the bainite phase has virtually dis­
gions, indicating that LME cracks may also form in the SCHAZ. Clearly, appeared and transformed into martensite due to the intercritical
the weld nugget develops on an angle due to the electrode angle. annealing and subsequent quenching. The UCHAZ region, where the
Additionally, the UCHAZ (shown as the darker shade of grey) is an un­ peak temperature exceeded the Ac3 temperature, transformed to coarse
usual shape. martensitic blocks with small fraction of ferrite (Fig. 6d). As illustrated,
It was observed as electrode angle increases the UCHAZ also becomes the UCHAZ possessed higher fraction of martensite, but lower amount of
angled and moves away from the weld shoulder on the indented side. ferrite compared to the ICHAZ region. Finally, in the fusion zone (FZ),
However, the location of the LME shoulder crack remains the same in all columnar martensite structure with large and thick lath martensite is
cases, which is consistent with previous work carried out by DiGiovanni shown (Fig. 6e). With the microstructural features of each region iden­
et al. (2020) showing that LME cracks in the shoulder are initiated by tified, the area near the crack can be characterized. EBSD analysis and
thermal stresses during RSW. If the cracks were initiated by the stresses SEM images of the LME crack and surrounding microstructure were
from electrode clamping, their location would change with misalign­ taken to identify and match the HAZ region.
ment. A schematic of the lower right weld shoulder showing the To investigate if the region surrounding the crack is the SCHAZ, the
retreating of the UCHAZ boundary, with the stationary LME crack, is grain structure was first analyzed. EBSD maps of the BM and the area
shown in Fig. 5. Despite the shifting of the high temperature zone away around LME crack are compared in Fig. 7. As illustrated in the inverse
pole figure (IPF) map, the BM has a finer ferrite grain distribution

4
C. DiGiovanni et al. Journal of Materials Processing Tech. 288 (2021) 116917

Fig. 6. OM micrograph and SEM microstructure of the (a) base metal (BM), (b) sub-critical heat affected zone (SCHAZ), (c) inter-critical heat affected zone (ICHAZ),
(d) upper-critical heat affected zone (UCHAZ) and (e) fusion zone (FZ).

compared to the area around LME crack. In addition, the image quality consists of ferrite matrix accompanied with tempered martensite and
(IQ) map with the GB misorientations is presented in Fig. 7. In the figure, bainite. Fig. 9 compares the martensite islands in BM and tempered
the dark lines show misorientations above 15 degrees, representing high martensite of the SCHAZ (near LME crack). As can be seen in Fig. 9d,
angle grain boundaries (HAGBs) and the green lines show mis­ cementite has precipitated into martensite blocks which is an estab­
orientations between 5–15 degrees representing low angle grain lished characteristic of tempered martensite from previous works by
boundaries (LAGBs). As shown, the area around the observed LME crack Furuhara et al. (2004). From the work by Furuhara et al. (2004) it is well
possesses a lower fraction of LAGBs compared to the BM. This obser­ known that after tempering martensite below Ac1 temperature, super­
vation shows that the applied heat in the crack region caused grain saturated carbon previously trapped in the martensite precipitates as
growth and reduction of LAGBs but was not high enough to cause a cementite on martensitic block, packet, and lathe boundaries. The
transformation (below Ac1). martensite observed in Fig. 9d showed this structure and can therefore
As a final test to confirm that cracking did occur in the SCHAZ, im­ be characterized as tempered martensite. The tempered martensite in
ages of the microstructure were taken. Fig. 8 shows the microstructure the crack region is clear evidence for considering the occurrence of LME
surrounding the LME crack, resembling the SCHAZ. The microstructure in the SCHAZ region. Occurrence of LME in the SCHAZ has not yet been

5
C. DiGiovanni et al. Journal of Materials Processing Tech. 288 (2021) 116917

Fig. 7. EBSD maps of the base metal (BM) and the area around the LME crack.

recorded in literature as stated by the review done by Bhattacharya propagation.

(2018) on LME in AHSS. This is likely due to the use of conventional With the LME cracking established to occur in the SCHAZ, the
welding setups, which have not observed SCHAZ in the weld shoulder. determination of the Ac1 temperature will uncover the upper limit for
Additionally, LME in AHSS has been reported by Beal et al. (2012) and the temperature range that LME occurred. A dilation study of the ma­
Jung (2015) to occur at temperatures above a typical Ac1, which made terial was carried out and the results are plotted in Fig. 10a. The results
the idea of SCHAZ crack seem unlikely. However, a detailed look at show that the Ac1 transformation starts to occur at approximately 660
Fig. 8b and c conclusively shows the LME crack in the SCHAZ and ◦
C, which was characterized using a method developed by Mesplont
branching out along the ferrite-prior martensite GBs. et al. (2001). Additionally, Fig. 10b shows the transformation percent­
In literature, LME is widely reported in austenitic structures and age as temperature increases according to the methodology by Mesplont
scarcely talked about in ferritic structures. One of the mechanisms for et al. (2001), clearly indicating the temperature required for trans­
LME crack initiation in the steel/zinc system is stress-assisted zinc formation. This upper limit temperature of the SCHAZ is below the re­
diffusion as described by both Ling et al. (2019) and Kang et al. ported temperature range for LME in AHSS as described in by both Beal
(2016). As reported by Kang et al. (2016), the GBs diffusivity is much et al. (2012) and Jung (2015). The occurrence of LME not only in the
higher than the bulk zinc diffusivity in both ferritic and austenitic SCHAZ but also at unusually low temperatures adds to the novel ob­
structures, therefore zinc diffuses along the GBs prior to lattice servations of this study.
diffusion. Additionally, Kang et al. (2016) also describes austenitic LME cracking in the SCHAZ and below 700 ◦ C has not been discussed
structures to have a higher zinc GB diffusivity than ferritic structures, extensively in the literature. Given the bending induced by the angled
but a lower bulk zinc diffusivity which promotes zinc GB diffusion electrode, the results of this study show the minimum temperature for
further. Since the LME crack is intergranular, austenitic structures the onset of LME is dependent on mechanical factors. This notion has
have commonly been reported to be more susceptible than ferritic been discussed in earlier studies but results showing evidence of low
structures. However, as presented in Figs. 8 and 9, the LME crack temperature/SCHAZ LME cracking with mechanical assistance has been
propagated along ferrite-prior martensite interfaces. One possible sparsely reported. Furthermore, in the case of the TRIP1100 steel, it is
explanation is the increased GB energy due to lattice distortion at the possible for LME to occur outside the traditionally reported range.
ferrite-martensite interface. A study carried out by Razmpoosh et al.
(2019) shows LME occurs in high-energy grain boundaries, in which 4. Conclusion
case, the increased energy due to lattice distortion may have
decreased the necessary stress for the onset of LME. Therefore, it Under angular electrode misalignment conditions, LME cracking was
seems the lattice distortion of the ferrite-martensite interface at observed for the first time in the SCHAZ and below the Ac1 temperature
relatively low temperature in SCHAZ region has assisted LME crack of 660 ◦ C, making this a novel occurrence of LME. Significant LME

6
C. DiGiovanni et al. Journal of Materials Processing Tech. 288 (2021) 116917

Fig. 8. SEM images of the microstructure near a sub-critical LME crack.

7
C. DiGiovanni et al. Journal of Materials Processing Tech. 288 (2021) 116917

Fig. 9. SEM micrographs of the martensite islands in (a,b) base metal (BM) and (c,d) subcritical heat affected zone (SCHAZ) regions.

Fig. 10. Curves showing (a) TRIP1100 dilation and (b) percentage transformation during heating.

cracks consistently formed at the same location reinforcing previous CRediT authorship contribution statement
literature’s claim of thermal stresses for crack initiation, but the
increased misalignment led to deeper LME cracks. However, a minimum C. DiGiovanni: Conceptualization, Methodology, Investigation,
threshold of 2◦ misalignment was established for increasing LME Writing - original draft, Writing - review & editing, Visualization. A.
severity as measured by the Crack Index. In the highest 6◦ case, Ghatei Kalashami: Methodology, Investigation, Writing - review &
distinctive distortion of the work piece was observed which resulted in editing. F. Goodwin: Resources, Writing - review & editing, Funding
an angled weld nugget/HAZ and large SCHAZ cracks. The results of this acquisition. E. Biro: Writing - review & editing, Supervision. N.Y. Zhou:
study show that the temperature range and microstructure required for Writing - review & editing, Supervision, Funding acquisition.
the onset of LME depend on various aspects and are not static, which is
something that must be considered in material processing design for
AHSS since LME cracks are detrimental to part performance. The tem­ Declaration of Competing Interest
perature and stress state must be considered together as the results of
this study show that lower temperatures alone are not sufficient to The authors declare that they have no known competing financial
suppress LME. Additionally, an AHSS possessing a ferritic structure will interests or personal relationships that could have appeared to influence
not be immune to LME, and the coupled temperature and stress state the work reported in this paper.
must be considered during material processing to avoid LME.
Acknowledgements

The authors would like to acknowledge the International Zinc

8
C. DiGiovanni et al. Journal of Materials Processing Tech. 288 (2021) 116917

Association (IZA) for their financial support and donation of material of Technol. Weld. Join. 24, 624–633. https://doi.org/10.1080/
13621718.2019.1582203.
this study, the Natural Sciences and Engineering Research Council of
Furuhara, T., Kobayashi, K., Maki, T., 2004. Control of cementite precipitation in lath
Canada (NSERC) and Canada Research Chairs (CRC) program for their martensite by rapid heating and tempering. ISIJ Int. 44, 1937–1944. https://doi.org/
support. 10.2355/isijinternational.44.1937.
Jung, G., 2015. Liquid Metal Embrittlement of High Mn TWIP Steel. Pohang University of
Science and Technology.
References Jung, G., Woo, I.S., Suh, D.W., Kim, S.J., 2016. Liquid Zn assisted embrittlement of
advanced high strength steels with different microstructures. Met. Mater. Int. 22,
American Welding Society, 2012. American Welding Society: Test Method for Evaluating 187–195. https://doi.org/10.1007/s12540-016-5579-7.
the Resistance Spot Welding Behavior of Automotive Sheet Steel Materials (AWS D Kang, H., Cho, L., Lee, C., De Cooman, B.C., 2016. Zn penetration in liquid metal
8.9M). embrittled TWIP steel. Metall. Mater. Trans. A 47, 2885–2905. https://doi.org/
Ashiri, R., Haque, M.A., Ji, C.W., Shamanian, M., Salimijazi, H.R., Park, Y.Do, 2015. 10.1007/s11661-016-3475-x.
Supercritical area and critical nugget diameter for liquid metal embrittlement of Zn- Lee, C.W., Fan, D.W., Sohn, I.R., Lee, S.J., De Cooman, B.C., 2012. Liquid-metal-induced
coated twining induced plasticity steels. Scr. Mater. 109, 6–10. https://doi.org/ embrittlement of Zn-coated hot stamping steel. Metall. Mater. Trans. A Phys. Metall.
10.1016/j.scriptamat.2015.07.006. Mater. Sci. 43, 5122–5127. https://doi.org/10.1007/s11661-012-1316-0.
Ashiri, R., Shamanian, M., Reza, H., Haque, A., Bae, J., Ji, C., Chin, K., Park, Y., 2016. Ling, Z., Chen, T., Kong, L., Wang, M., Pan, H., Lei, M., 2019. Liquid metal embrittlement
Liquid metal embrittlement-free welds of Zn-coated twinning induced plasticity cracking during resistance spot welding of galvanized Q&P980 steel. Metall. Mater.
steels. Scr. Mater. 114, 41–47. https://doi.org/10.1016/j.scriptamat.2015.11.027. Trans. A 50, 5128–5142. https://doi.org/10.1007/s11661-019-05388-6.
Beal, C., 2012. Mechanical Behaviour of a New Automotive High Manganese TWIP Steel Mesplont, C., Zhao, J.Z., Vandeputte, S., De Cooman, B.C., 2001. An improved method
in the Presence of Liquid Zinc [Dissertation]. Insa Lyon.. for determining the continuous cooling transformation diagram of C-Mn steels.
Beal, C., Kleber, X., Fabregue, D., Bouzekri, M., 2012. Embrittlement of a zinc coated Mater. Technol. 72, 263–270.
high manganese TWIP steel. Mater. Sci. Eng. A 543, 76–83. https://doi.org/ Murugan, S., Mahmud, K., Park, Y.-D., 2018. The influence of electrode geometry on
10.1016/j.msea.2012.02.049. liquid metal embrittlement cracking in resistance spot welding of advanced high
Bhattacharya, D., 2018. Liquid metal embrittlement during resistance spot welding of strength steel. In: International Insititute of Welding Annual Meeting. Bali, Indonesia
Zn-coated high-strength steels. Mater. Sci. Technol. (United Kingdom). https://doi. pp. III- 1876–18.
org/10.1080/02670836.2018.1461595. Razmpoosh, M.H., Biro, E., Zhou, N.Y., 2019. Liquid metal embrittlement during laser
Cho, L., Kang, H., Lee, C., De Cooman, B.C., 2014. Microstructure of liquid metal welding of Zn-coated steel: role of random and coincidence site lattice grain
embrittlement cracks on Zn-coated 22MnB5 press-hardened steel. Scr. Mater. 90, boundaries. 72nd IIW Annu. Assem. Int. Conf. 7.
25–28. https://doi.org/10.1016/j.scriptamat.2014.07.008. Wintjes, E., DiGiovanni, C., He, L., Bag, S., Goodwin, F., Biro, E., Zhou, Y., 2019a. Effect
DiGiovanni, C., Han, X., Powell, A., Biro, E., Zhou, N.Y., 2019. Experimental and of multiple pulse resistance spot welding schedules on liquid metal embrittlement
numerical analysis of liquid metal embrittlement crack location. J. Mater. Eng. severity. J. Manuf. Sci. Eng. 141, 1–9. https://doi.org/10.1115/1.4044099.
Perform. 28 https://doi.org/10.1007/s11665-019-04005-2. Wintjes, E., DiGiovanni, C., He, L., Biro, E., Zhou, N.Y., 2019b. Quantifying the link
DiGiovanni, C., He, L., Pistek, U., Goodwin, F., Biro, E., Zhou, N.Y., 2020. Role of spot between crack distribution and resistance spot weld strength reduction in liquid
weld electrode geometry on liquid metal embrittlement crack development. metal embrittlement susceptible steels. Weld. World 63. https://doi.org/10.1007/
J. Manuf. Process. 49 https://doi.org/10.1016/j.jmapro.2019.11.015. s40194-019-00712-5.
Frei, J., Biegler, M., Rethmeier, M., Böhne, C., Frei, J., Biegler, M., 2019. Investigation of Xing, B., Yan, S., Zhou, H., Chen, H., Qin, Q.H., 2018. Qualitative and quantitative
liquid metal embrittlement of dual phase steel joints by electro-thermomechanical analysis of misaligned electrode degradation when welding galvannealed steel. Int.
spot-welding simulation electro-thermomechanical spot-welding simulation. Sci. J. Adv. Manuf. Technol. 629–640.