Welding and joining of advanced high strength steels (AHSS)

Prévia do material em texto

Welding and Joining of Advanced High Strength 
Steels (AHSS)
Related titles
The Welding Engineer’s Guide to Fracture and Fatigue
(ISBN 978-1-78242-370-6)
Control of Welding Distortion in Thin-plate Fabrication
(ISBN 978-0-85709-047-8)
Thermochemical Surface Engineering of Steels
(ISBN 978-0-85709-592-3)
Woodhead Publishing Series in Welding and 
Other Joining Technologies: Number 85
Welding and Joining of 
Advanced High Strength 
Steels (AHSS)
Edited by
Mahadev Shome and 
Muralidhar Tumuluru
AMSTERDAM • BOSTON • CAMBRIDGE • HEIDELBERG 
LONDON • NEW YORK • OXFORD • PARIS • SAN DIEGO 
SAN FRANCISCO • SINGAPORE • SYDNEY • TOKYO
Woodhead Publishing is an imprint of Elsevier
Woodhead Publishing is an imprint of Elsevier
80 High Street, Sawston, Cambridge, CB22 3HJ, UK
225 Wyman Street, Waltham, MA 02451, USA
Langford Lane, Kidlington, OX5 1GB, UK
Copyright © 2015 Elsevier Ltd. All rights reserved.
No part of this publication may be reproduced, stored in a retrieval system or transmitted 
in any form or by any means electronic, mechanical, photocopying, recording or otherwise 
without the prior written permission of the publisher.
Permissions may be sought directly from Elsevier’s Science & Technology Rights 
Department in Oxford, UK: phone (+44) (0) 1865 843830; fax (+44) (0) 1865 853333; 
email: permissions@elsevier.com. Alternatively, you can submit your request online by 
visiting the Elsevier website at http://elsevier.com/locate/permissions, and selecting 
Obtaining permission to use Elsevier material.
Notice
No responsibility is assumed by the publisher for any injury and/or damage to persons 
or property as a matter of products liability, negligence or otherwise, or from any use or 
operation of any methods, products, instructions or ideas contained in the material herein. 
Because of rapid advances in the medical sciences, in particular, independent verification of 
diagnoses and drug dosages should be made.
British Library Cataloguing-in-Publication Data
A catalogue record for this book is available from the British Library
Library of Congress Control Number: 2014957593
ISBN 978-0-85709-436-0 (print)
ISBN 978-0-85709-858-0 (online)
For information on all Woodhead Publishing publications 
visit our website at http://store.elsevier.com/
Typeset by TNQ Books and Journals
www.tnq.co.in
Printed and bound in the United Kingdom
mailto:permissions@elsevier.com
http://elsevier.com/locate/permissions
http://store.elsevier.com/
http://www.tnq.co.in
Contents
List of contributors ix
Woodhead Publishing Series in Welding and Other Joining Technologies xi
 1 Introduction to welding and joining of advanced high-strength 
steels (AHSS) 1
M. Shome, M. Tumuluru
 1.1 Introduction 1
 1.2 Overview of major welding processes for AHSS 4
References 7
 2 Properties and automotive applications of advanced high-strength 
steels (AHSS) 9
T.B. Hilditch, T. de Souza, P.D. Hodgson
 2.1 The automobile body 9
 2.2 AHSS microstructures and tensile properties 14
 2.3 Formability and fracture of AHSS 22
 2.4 Automotive in-service properties 24
 2.5 Current and future trends in AHSS 25
References 27
 3 Manufacturing of advanced high-strength steels (AHSS) 29
M.-C. Theyssier
 3.1 Introduction 29
 3.2 Key challenges faced in producing AHSS grades 30
 3.3 Future trends 51
References 51
 4 Resistance spot welding techniques for advanced high-strength 
steels (AHSS) 55
M. Tumuluru
 4.1 Introduction 55
 4.2 Characterizing welding behavior 56
 4.3 General considerations in resistance spot welding of AHSS 59
 4.4 Coating effects 64
 4.5 Microstructural evolution in welds 65
vi Contents
 4.6 Weld shear tension strength and cross-tension strength (CTS) 67
 4.7 Summary 69
References 69
 5 Laser welding of advanced high-strength steels (AHSS) 71
S.S. Nayak, E. Biro, Y. Zhou
 5.1 Introduction 71
 5.2 Background 72
 5.3 Laser welding of AHSS 72
 5.4 Microstructure of laser-welded AHSS 73
 5.5 Hardness 78
 5.6 Performance of laser-welded AHSS 84
 5.7 Future trends 90
References 90
 6 High-power beam welding of advanced high-strength steels (AHSS) 93
L. Cretteur
 6.1 Introduction 93
 6.2 Back to basics: fundamentals of high-power beam welding 95
 6.3 Metallurgical phenomena in laser welding of AHSS 100
 6.4 Laser-welded blanks (LWBs): issues related to the use of AHSS 106
 6.5 Body-in-white joining applications 109
 6.6 Conclusions 117
Acknowledgments 118
References 118
 7 Hybrid welding processes in advanced high-strength steels (AHSS) 121
S. Chatterjee, T. van der Veldt
 7.1 Introduction 121
 7.2 Laser–arc hybrid process description 122
 7.3 Laser–arc hybrid process parameters for welding automotive 
AHSS 124
 7.4 Applications in the automotive industry 132
 7.5 Costs and economics 133
References 134
 8 Metal inert gas (MIG) brazing and friction stir spot welding of 
advanced high-strength steels (AHSS) 137
M. Shome
 8.1 Introduction 137
 8.2 MIG brazing 138
 8.3 Friction stir spot welding (FSSW) 146
 8.4 Conclusions 163
Acknowledgements 164
References 164
viiContents
 9 Adhesive bonding techniques for advanced high-strength 
steels (AHSS) 167
K. Dilger, S. Kreling
 9.1 Introduction: the exigency of adhesive bonding of 
high-strength steels 167
 9.2 Challenges in adhesive bonding of AHSS 169
 9.3 Boron–manganese steels: anticinder coatings and their 
influence on adhesive bonds 174
 9.4 Weldbonding of AHSS 176
 9.5 Conclusions 177
References 178
 10 Mechanical fastening techniques for advanced high-strength 
steels (AHSS) 181
C. Hsu
 10.1 Introduction 181
 10.2 The use of drawn arc welding (DAW) for attaching studs to metals 181
 10.3 Assessing the feasibility of DAW for stud welding of AHSS 183
 10.4 Robotic stud welding 184
References 185
Index 187
 
This page intentionally left blank
List of contributors
E. Biro ArcelorMittal Global Research, Hamilton, ON, Canada
S. Chatterjee Tata Steel Research and Development, Joining and Performance 
Technology, Wenckebachstraat, The Netherlands
L. Cretteur ArcelorMittal R & D, Automotive Application Research Center, 
Montataire, France
K. Dilger TU Braunschweig, Institute of Joining and Welding, Braunschweig, 
Germany
T.B. Hilditch Deakin University, Waurn Ponds, Victoria, Australia
P.D. Hodgson Deakin University, Waurn Ponds, Victoria, Australia
C. Hsu Consultant, UK
S. Kreling TU Braunschweig, Institute of Joining and Welding, Braunschweig, 
Germany
S.S. Nayak University of Waterloo, Waterloo, ON, Canada
M. Shome Research & Development, Tata Steel, Jamshedpur, India
T. de Souza Deakin University, Waurn Ponds, Victoria, Australia
M.-C. Theyssier ArcelorMittal R & D Center, Maizières les Metz, France
M. Tumuluru Research and Technology Center, United States Steel Corporation, 
Pittsburgh, PA, USA
T. van der Veldt Tata Steel Research and Development, Joining and Performance 
Technology, Wenckebachstraat, The Netherlands
Y. Zhou University of Waterloo, Waterloo, ON, Canada
 
This page intentionally left blank
Woodhead Publishing Series in Welding 
and Other Joining Technologies
 1 Submerged-arc welding
Edited by P. T. Houldcroft
 2 Design and analysis of fatigue resistant welded structures
D. Radaj
 3 Which process? A guide to the selection of welded and related processes
P. T. Houldcroft
 4 Pulsed arc welding
J. A. Street
 5 TIG and plasma welding
W. Lucas
 6 Fundamentals of welding metallurgy
H. Granjon
 7 Fatigue strength of welded structures
S. J. Maddox
 8 The fatigue strength of transverse fillet welded joints
T. R. Gurney
 9 Process pipe and tube welding
Edited by W. Lucas
 10 A practical guide to TIG (GTA) welding
P. W. Muncaster
 11 Shallow crack fracture mechanics toughness tests and applications
Conference Proceedings
 12 Self-shielded arc welding
T. Boniszewski
 13 Handbook of crack opening data
T. G. F. Gray
 14 Laser welding
C. T. Dawes
 15 Welding steels without hydrogen crackingN. Bailey and F. R. Coe
 16 Electron beam welding
H. Schultz
 17 Weldability of ferritic steels
N. Bailey
 18 Tubular wire welding
D. Widgery
 19 Stress determination for fatigue analysis of welded components: Recommendations of IIW 
Commissions XIII and XV
Edited by E. Niemi
 20 The ‘local approach’ to cleavage fracture
C. S. Wiesner
xii Woodhead Publishing Series in Welding and Other Joining Technologies
 21 Crack arrest concepts for failure prevention and life extension
Seminar Proceedings
 22 Welding mechanisation and automation in shipbuilding worldwide
R. Boekholt
 23 Heat treatment of welded steel structures
D. Croft
 24 Fatigue design of welded joints and components: Recommendations of IIW Joint Working 
Group XIII-XV
Edited by A. Hobbacher
 25 Introduction to the non-destructive testing of welded joints
R. Halmshaw
 26 Metallurgy of basic weld metal
T. R. Gurney
 27 Fatigue of thin walled joints under complex loading
T. R. Gurney
 28 Handbook of structural welding
J. F. Lancaster
 29 Quality assurance in adhesive technology
A. W. Espie, J. H. Rogerson and K. Ebtehaj
 30 Underwater wet welding and cutting
TWI/Paton Electric Welding Institute
 31 Metallurgy of welding Sixth edition
J. F. Lancaster
 32 Computer technology in welding
Conference Proceedings
 33 Exploiting advances in arc welding technology
Conference Proceedings
 34 Non-destructive examination of underwater welded structures
V. S. Davey
 35 Predictive formulae for weld distortion
G. Verhaeghe
 36 Thermal welding of polymers
R. J. Wise
 37 Handbook of mould, tool and die repair welding
S. Thompson
 38 Non-destructive testing of welds
B. Raj, C. V. Subramanian and T. Jayakumar
 39 The automotive industry: joining technologies
TWI
 40 Power generation: welding applications
TWI
 41 Laser welding
TWI
 42 Fatigue: welding case studies
TWI
 43 Fracture: welding case studies
TWI
 44 The welding workplace
R. Boekholt
 45 Underwater repair technology
J. Nixon
 46 Fatigue design procedure for welded hollow section joints: Recommendations of IIW 
Subcommission XV-E
Edited by X.-L. Zhoa and J. A. Packer
xiiiWoodhead Publishing Series in Welding and Other Joining Technologies
 47 Aluminium welding
N. R. Mandal
 48 Welding and cutting
P. T. Houldcrof and J. A. Packer
 49 Health and safety in welding and allied processes
J. Blunt
 50 The welding of aluminium and its alloys
G. Mathers
 51 Arc welding control
P. Julian
 52 Adhesive bonding
R. D. Adams
 53 New developments in advanced welding
Edited by N. Ahmed
 54 Processes and mechanisms of welding residual stress and distortion
Edited by Z. Feng
 55 MIG welding guide
Edited by K. Wenem
 56 Cumulative damage of welded joints
T. R. Gurney
 57 Fatigue analysis of welded components: Recommendations of IIW Commissions XIII and XV
E. Niemi
 58 Advanced welding processes
J. Norrish
 59 Fatigue assessment of welded joints by local approaches
D. Radaj
 60 Computational welding mechanics
Edited by L. E. Lindgren
 61 Microjoining and nanojoining
Edited by Y. N. Zhou
 62 Real-time weld process monitoring
Edited by Y. M. Zhang
 63 Weld cracking in ferrous alloys
Edited by R. Singh
 64 Hybrid laser-arc welding
Edited by F. O. Olsen
 65 A quick guide to welding and weld inspection
Edited by S. E. Hughes
 66 Friction stir welding
Edited by D. Lohwasser and Z. Chen
 67 Advances in structural adhesive bonding
Edited by D. Dillard
 68 Failure mechanisms of advanced welding processes
Edited by X. Sun
 69 Advances in laser materials processing
Edited by J. Lawrence and J. Pou
 70 Welding and joining of magnesium alloys
Edited by L. Lui
 71 Fracture and fatigue of welded joints and structures
Edited by K. MacDonald
 72 Minimization of welding distortion and buckling
Edited by P. Michaleris
 73 Welding processes handbook Second edition
K. Weman
xiv Woodhead Publishing Series in Welding and Other Joining Technologies
 74 Welding and joining of aerospace materials
Edited by M. C. Chaturvedi
 75 Tailor welded blanks for advanced manufacturing
Edited by B. Kinsey and X. Wu
 76 Adhesives in marine engineering
Edited by J. R. Weitzenböck
 77 Fundamentals of evaluation and diagnostics of welded structures
A. Nedoseka
 78 IIW recommendations for the fatigue assessment of welded structures by notch stress analysis
W. Fricke
 79 IIW recommendations on methods for improving the fatigue strength of welded joints
P. J. Haagensen and S. J. Maddox
 80 Advances in brazing
Edited by D. P. Sekulic
 81 Advances in friction-stir welding and processing
M.-K. Besharati-Givi and P. Asadi
 82 Self-piercing riveting
Edited by A. Chrysanthou and X. Sun
 83 Control of welding distortion in thin plate fabrication: Design support exploiting computa-
tional simulation
T. Gray, D. Camilleri and N. McPherson
 84 The welding engineer’s guide to fracture and fatigue
P. L. Moore and G. S. Booth
 85 Welding and joining of advanced high strength steels (AHSS)
Edited by M. Shome and M. Tumuluru
Welding and Joining of Advanced High Strength Steels (AHSS). http://dx.doi.org/10.1016/B978-0-85709-436-0.00001-1
Copyright © 2015 Elsevier Ltd. All rights reserved.
Introduction to welding and joining 
of advanced high-strength steels 
(AHSS)
M. Shome1, M. Tumuluru2
1Research & Development, Tata Steel, Jamshedpur, India; 2Research and Technology 
Center, United States Steel Corporation, Pittsburgh, PA, USA
1
1.1 Introduction
Fuel efficiency, lowering carbon emissions and passenger safety have been the main 
drivers in designing automobiles for the past two decades. Vehicle weight reduction was 
identified as a key strategy to minimize fuel consumption. For enhanced passenger safety, 
automotive structures that have a higher energy absorption in a crash situation would be 
ideal. Advanced high-strength steels (AHSSs) were developed to support these strategic 
requirements. A recent report from World Steel Dynamics projected that by 2025 the 
usage of AHSSs would reach 23.7 million tons. This means that a significant part of the 
low-carbon steel parts would be replaced by AHSSs (http://www.autosteel.org, report of 
October 4, 2014). Reductions in automotive mass and government regulations on crash 
requirements seem to have mutually opposing directions: fuel economy is ensured but 
safety can seemingly be endangered by lighter vehicles. However, studies to date using 
AHSSs for automobile designs have shown that reducing the weight of vehicles can be 
achieved without compromising passenger safety.
AHSSs are extensively used in the automobile industry for manufacturing several 
body-in-white parts of vehicles. Auto designers have introduced these steels in critical 
structural parts such as the A, B and C pillars; the roof rails and bow; cross-members; 
door beams; front and side members; and as bumper reinforcement. They also are exten-
sively used in internal panels made of tailor-welded blanks. AHSSs exhibit ultimate 
tensile strengths of 600 MPa or higher, allowing vehicle manufacturers to make major 
strides in terms of the strength and rigidity of thinner-gauge sheets. In addition to high 
tensile properties these steels have good ductility, the capacity for high energy absorption 
and a high work-hardening coefficient over the uniform elongation regime. Dual-phase 
(DP), complex-phase (CP), transformation-induced plasticity (TRIP) and martensitic 
steels are the prevalent AHSS grades that are currently in commercial use. These grades 
are referred to as first-generation AHSSs. The relationship between the strength and 
ductility (as measured by elongation) of these steel grades is shown in Figure 1.1.
AHSS are multiphase steels that contain various concentrations of ferrite, bainite, mar-
tensite and retained austenite phases. The proportion of these phases and their morpholo-
gies are engineered to obtain the functional characteristics of a steel (Bhattacharya, 2011, 
p. 163; Davies, 2012; Galán,Samek, Verleysen, Verbeken, & Houbaert, 2012; Kuziak, 
Kawalla, & Waengler, 2008; Senuma, 2001). DP steels are commercially available 
http://www.autosteel.org
2 Welding and Joining of AHSS
from 500 to 1180 MPa, whereas TRIP and CP steels are available up to 980 MPa 
strength. These steel grades are used in applications that require high strength and 
high ductility (and hence good formability), as well as good weldability. Several stud-
ies have clearly shown the excellent weldability of these steel grades (Radakovic & 
Tumuluru, 2012; Sharma & Molian, 2011; Tumuluru, 2013). Some of the applications 
of these steels include B pillars and body inners. The microstructure of DP steels 
consists of ferrite and martensite, which provide the necessary strength and meets 
elongation requirements. Higher strength implies that there is a larger volume fraction 
of martensite in the steel. DP steels are used in both hot-rolled and cold-rolled con-
ditions. Hot-rolled DP steel is mostly used for the structural parts and wheels of cars. 
Continuous yielding characteristics are a special feature of DP steels that ensures 
a smooth surface after the forming operation. TRIP steels contain ferrite, bainite and 
retained austenite phases. The retained austenite is transformed into martensite under 
a strain-induced deformation effect, absorbing significant amounts of energy; there-
fore TRIP steel is a designer’s choice for making crash-resistant components.
Another category of AHSSs is martensitic steel. These steels are currently available with 
strength from 900 to 1900 MPa. The microstructure of these steels consists essentially of 
martensite. These steels are alloyed with carbon, manganese and chromium to achieve the 
required strength. Martensitic steels have high stiffness and anti-intrusion characteristics for 
passenger safety. Because of their higher carbon content—more than is contained in either 
DP or TRIP steels—martensitic steels are used in applications that generally do not require 
welding. Some examples of their application include door intrusion beams and bumpers.
Cold-rolled DP and TRIP steels are processed in continuous annealing lines. Typi-
cal production methods for DP and TRIP steels are shown in Figure 1.2. For DP steel 
Figure 1.1 The strength versus elongation relationship for first-generation advanced high-
strength steels. HSLA, high strength, low alloy; TRIP, transformation-induced plasticity.
3Introduction to welding and joining of AHSS
production, the cold-rolled, fully hard strip is subjected to intercritical (α + γ) annealing, 
followed by rapid cooling so that the austenite transforms into martensite. A uniform 
distribution of about 10% volume fraction of martensite in the ferrite matrix results in an 
excellent strength–ductility combination, low-yield strength–to–tensile strength ratio 
and a high work-hardening index. TRIP steels are produced by applying a two-stage 
heat treatment process. The cold-rolled sheets are heated to the intercritical temperature 
and held there for a short time, allowing austenite to form. The annealing tempera-
ture and time determine the austenite volume fraction and carbon concentration. In 
the second stage the coils are rapidly cooled and isothermally held at a temperature at 
which a bainitic reaction occurs. The carbon rejected during the bainitic transforma-
tion enriches the remaining austenite and stabilizes it. For galvanized and galvannealed 
steels, the same heat treatment concept is followed. For coating purposes, the sheets are 
cooled from the intercritical temperature and passed through a galvanizing bath kept 
at 460 °C (Liu et al., 2012). The silicon content in AHSSs is kept very low to avoid 
adhesion problems in galvanizing baths. Strength improvements for coated AHSSs are 
achieved through alloying with elements such as manganese and chromium.
While these steels have a combination of superior mechanical properties, their 
application in terms of forming and welding requires a different approach than the 
one used for low-carbon steels. During welding, the heat produced alters the micro-
structure of the base material and therefore the mechanical properties. The heating 
and cooling rates are extremely rapid in all welding processes during automotive 
body manufacturing. The peak temperature observed in the fusion zone (FZ) is above 
the melting point of steel and is somewhat lower in the heat-affected zone (HAZ). In 
the HAZ there is significant austenite grain growth followed by phase transformation; 
consequently, the microstructure formed is different from that of the base metal. The 
task, therefore, is to control the thermal conditions by applying appropriate welding 
parameters. Solid-state welding and alternative joining techniques have recently been 
tested to preserve the functional properties of AHSSs without worrying much about 
temperature-related damage to the microstructure caused by conventional welding.
Ferrite
α
α
αM
γ
Pearlite
Bainite
M
α
γ
Ferrite
Pearlite
Bainite
M
Hold
Soak
α
αM
(a) (b)
Figure 1.2 Typical production methods for dual-phase (a) and transformation-induced 
plasticity steels (b).
4 Welding and Joining of AHSS
Welding is an integral part of automobile manufacturing and is carried out through 
various processes. There are advantages and disadvantages of each process. An over-
view of the most common processes with respect to welding and joining AHSSs are 
briefly discussed.
1.2 Overview of major welding processes for AHSS
1.2.1 Resistance spot welding
DP steels are easily weldable and have been commercially implemented in current automo-
tive designs (Radakovic & Tumuluru, 2012). The typical requirement for spot welds is to 
have a minimum load-bearing capacity equivalent to or greater than that of the base metal. 
The load capacity formula includes the thickness of the sheet, the weld nugget diameter 
and the ultimate tensile strength of the steel (Radakovic & Tumuluru, 2008, 2012). The 
nugget diameter depends on the welding parameters and is critical for AHSS because it 
largely controls the type of weld failure under quasi-static and dynamic loading conditions.
Spot welds can fail in any of the following three modes: interfacial failure, in which 
the fracture propagates through the nugget; pull-out failure, in which the weld nugget 
separates from the parent metal; and partial interfacial failure, in which the fracture 
initially propagates through the nugget and then deviates through the sheet thickness, 
similar to pull-out failure. Pull-out failure is preferred because it is associated with 
high-load bearing capacity and high energy absorption. Recent work has shown that 
interfacial fractures are the expected mode in AHSSs and that welds that fail with the 
interfacial fracture mode have a load-bearing ability that is 90% of welds that fail with 
a pull-out mode (Radakovic & Tumuluru, 2008; Tumuluru, 2006b).
Work done on the entire range of DP steels of 590-, 780- and 980-MPa strength 
shows a certain pattern of nugget failure during shear tensile tests. Full-button pull-out 
fracture occurs when the weld nugget size is large, and interfacial fracture occurs when 
the nuggets are small (Tumuluru, 2008; Radakovic & Tumuluru, 2012). A separate study 
of DP600 spot welds showed that thicker sheets are more prone to interfacial failure 
(Tumuluru, 2006a). The crack is initiated at the edge of the weld nugget and at the inter-
face between the two sheets because of strain localization (Ma et al., 2008; Dancette 
et al., 2012). However, the load-bearing capacity of the nuggets with interfacial failure 
was high and acceptable. For AHSSs, the strength of the spot weld is given prominence 
over the type of fracture while qualifying the welds (Radakovic & Tumuluru, 2012).
In the case of spot-welded TRIP780 steel, the hardness of the FZ depends on the 
composition of the steel. Carbon (C)–manganese (Mn)–aluminium, C–Mn–aluminium– 
silicon (Si)or C–Mn–Si steel welds have varying proportions of ferrite, bainite and 
martensite. The weld nugget of the first steel contains a mixture of ferrite, bainite and 
martensite, whereas in the second it is mostly martensite with some bainite. C–Mn–Si 
steel welds contain only martensite, and therefore the hardness was the highest among 
all the TRIP compositions (Nayak, Baltazar Hernandez, Okita, & Zhou, 2012).
Properly designed welds rarely fail under actual conditions, and confirmation 
tests indicate that the failure loads are on par with the base metal strengths. Still, 
any attempt to weld high-strength steels calls for special attention in terms of nugget 
5Introduction to welding and joining of AHSS
diameter, defects in the FZ and the type of microstructure. While large nugget diame-
ters may seem to be the panacea for the problem, they have to be viewed in the context 
of HAZ softening, zinc loss in coated steels and electrode life.
1.2.2 Gas metal arc welding
Gas metal arc welding (GMAW) is mostly applied in chassis parts, where it is import-
ant to secure the strength and rigidity of the joint. The process also has the freedom 
to join parts of various shapes to structural members such as pipes and brackets. Long 
fatigue life of the weld joint is a prerequisite. Spatter, fit-up and gap issues need to be 
dealt with in parts formed during welding. Certain component designs preclude the 
use of resistance spot welds. Further, there are closed parts that cannot be reached with 
resistance spot welding guns. For such applications, the GMAW process is preferred.
The GMAW process is also known as metal inert gas or metal active gas weld-
ing. Carbon dioxide is the active shielding gas in the latter process. Consumables with 
matching strengths are preferred to meet the mechanical property requirements of the 
joint, but lower-strength wires have been used to attain mechanical properties by depos-
iting extra material. One can refer to auto steel partnership program reports in which 
welding parameters and weld joint properties for various AHSS combinations have been 
reported (A/SP Joining Technologies, 2004). Consumable ER70S3 wires and shielding 
gas comprising 90% argon and 10% carbon dioxide produced acceptable welds.
Higher heat input GMAW causes the HAZ to soften in DP steels, which in turn affects 
the fatigue properties. Studies have been carried out to correlate the effect of weld geome-
try and microstructure on the fatigue properties of AHSS butt welds. Some showed that the 
bead geometry and microstructure could act as a notch for the initiation and propagation 
of cracks under fatigue conditions. The lowest hardness point is in the subcritical HAZ of 
DP590 steel, and most samples fail in this location during tensile testing, regardless of the 
bead geometry. Specimens with large beads (convex profile with higher height to width 
ratio) show a significantly shorter fatigue life, with fractures initiated at the toe of the 
weld. A shallow bead, that is, lower height/width ratio, with appropriate microstructure can 
improve fatigue performance in GMAW welds (Ahiale & Jun Oh, 2014).
For welding galvannealed AHSS, wires with chemical composition of low silicon 
to manganese ratio have typically been used with a welding angle less than 30°. The 
weld pool flows in the direction of the arc and prevents the formation of blow holes and 
porosities, which is a major issue during arc welding of zinc-coated sheets. The beads 
are flatter with a smooth curvature at the toe region. In fact, a welding wire with low 
silicon content and a base metal with higher silicon content gives the best bead profile.
1.2.3 Laser welding
In the past two decades laser welding has become popular because lasers have high 
power density (108 W/cm2) and hence are able to weld steels at high speeds to meet 
stringent productivity targets. It provides a narrow HAZ compared with that in con-
ventional arc welding processes. This feature augers well for AHSSs. Carbon dioxide 
lasers are the most common lasers used for sheet metal fabrication, particularly for 
6 Welding and Joining of AHSS
manufacturing tailor-welded blanks involving combinations of AHSS and low-carbon 
formable steel. However, high-power fibre and disc lasers are being extensively used 
to weld AHSSs by several automotive manufacturers.
With high heating and cooling rates, AHSSs normally form martensite in the weld 
metal. The small HAZ, even if softened, has a minimal effect on the overall mechan-
ical properties. Welded samples usually fail in the parent metal, indicating that the 
joints are sound (Nemecek, Muzik, & Misek, 2012). Laser welding of galvanized 
steel in a zero-gap lap joint configuration is challenging because of the vigorous gen-
eration of zinc vapour at the faying surface, which causes porosity. Dual-beam laser 
welding has recently been successfully used to join galvanized sheets in a zero-gap lap 
joint configuration. In the first beam a defocused laser is used to burn the zinc on the 
top surface and the interface, which prepares the surface for better absorption during 
the second pass. In the second pass a stable keyhole is formed to help vent any zinc 
vapour produced. Welding of coated DP980 steel resulted in porosity-free, partially 
penetrated lap joints without any spatter or blow holes. The welding process was sta-
ble. During tensile shear testing, the joints failed in the HAZ zone, with satisfactory 
mechanical properties (Maa, Konga, Carlsonb, & Kovacevica, 2013).
1.2.4 Adhesive joining and weld bonding
Adhesives serve the purpose of enhancing the stiffness of a member by providing a 
continuous joint. As there are concerns regarding the durability of adhesive joints 
under different environmental conditions, the weld bonding process is preferred by 
several manufacturers. This process involves a combination of spot welding and adhe-
sive joining, wherein the benefits of durability provided by spot welding and stiffness 
provided by adhesives are leveraged.
For AHSSs, high-strength structural adhesives with good wettability and flow char-
acteristics have been used. They are spread over the overlap area and cured to obtain 
a suitable bond strength. In the case of weld bonding, spot welding is done soon after 
the adhesive is applied. Hence the adhesive thickness must be kept small to allow spot 
welding to happen. A thick and dense adhesive may either impede the passage of a 
current or cause heavy expulsion, neither of which is acceptable.
Hybrid joints have several advantages such as reduced stress concentration around 
the nugget in spot welds, enhanced strength and higher energy absorption for failure 
and improved stiffness (Bartczak, Mucha, & Trzepiecinski, 2013). In weld-bonded 
joints of DP600 and DP800 steels the shear strength has been reported to be greater 
than that of a spot-welded joint (Bartczak et al., 2013; Hayat, 2011). Epoxy-based, 
high-strength structural adhesives have provided the requisite shear strength value. 
During shear tensile testing, a high level of shearing stress exists at the outer and 
inner edges of the overlap. Due to of the presence of the adhesive layer, a lower stress 
exists at the notch of the weld nugget in weld-bonded joints as compared to spot 
welds. The weld-bonded joint strength of DP590 steel was 40% higher than that of a 
spot-welded joint and 15% higher than an adhesive joint. For DP780 weld bonds, the 
strength was higher by 58% over spot-welded joints and 39% higher than adhesive 
joints. An identical adhesive was used for both the weld bond and adhesive joints 
(Sam & Shome, 2010).
7Introduction to welding and joining of AHSS
1.2.5 New technology
Over the years there have been major advances in machine technology. The use of inverter- 
based medium-frequency direct current spot and seam welding processes has become 
common, especially in countries where power costs are high. This technology has addi-
tional benefits when weldingAHSS because of the low and sustained energy input.
Metallurgical alterations and burn through caused by high heat input are common 
problems during arc welding of thin AHSS sheets. Therefore, low welding currents 
using small-diameter wires (e.g. 0.8 mm) are preferred to ensure low heat input. The 
recently developed cold metal transfer technology does provide low heat input and 
low spatter compared with direct current metal active gas systems. The cold metal 
transfer wire feeder unit can control the forward and backward movement and syn-
chronize it with the current wave form. By doing so it can shorten the arcing time and 
hence the welding heat input. As a result, shallow beads with a low wetting angle at 
the toe region are obtained (Kodama et al., 2013).
The alternating current GMAW process has been recently developed to overcome the 
burn through problem of sheet metal. In this process the advantage of the arc stability of 
direct current electrode-positive mode and the high melting rate of direct current elec-
trode-negative mode are combined. In the latter mode, the wire melting rate is high and 
therefore penetration is limited. Consequently, there is a better gap-bridging effect (Arif & 
Chung, 2014). The control of drop size and drop transfer governs the gap-bridging ability 
in the alternating current GMAW process and is significant for obtaining defect-free welds.
In the laser welding space the introduction of high-power disk lasers and fibre 
lasers have recently had a widespread impact. These lasers are available in powers 
exceeding 5 kW in continuous wave mode, and they have high efficiency and excel-
lent beam quality that enable deep penetration at high welding speeds. Ytterbium:yt-
trium–aluminium–garnet disk lasers produce excellent beam characteristics. Major 
automotive companies are using these lasers for three-shift production at lower 
operating costs than conventional lasers (Sharma & Molian, 2011). Application of 
ytterbium:yttrium–aluminium–garnet lasers on prestrained, cold-rolled DP980 and 
TRIP780 steels created butt welds without any porosity, undercut, burn through or 
convexity. Though HAZ softening in DP980 steel continues to be an issue, such 
softening was highly localized and narrow. Therefore, the impact of HAZ softening 
was minimized and did not affect the overall mechanical properties of the welded 
coupons.
References
A/SP Joining Technologies Committee Report. (2004). Advanced high strength steel (AHSS) 
weld performance study for auto body structural components.
Ahiale, G. K., & Jun Oh, Y. (2014). Microstructure and fatigue performance of butt-welded 
joints in advanced high-strength steels. Material Science Engineering A, 597, 342.
Arif, N., & Chung, H. (2014). Alternating current-gas metal arc welding for application of thin 
sheets. Journal of Materials Processing Technology, 214, 1828.
Bhattacharya, D. (2011). Metallurgical perspectives on advanced sheet steels for automotive 
applications. In Advanced steels (p.163). Berlin: Springer.
8 Welding and Joining of AHSS
Bartczak, B., Mucha, J., & Trzepiecinski, T. (2013). Stress distribution in adhesively-bonded 
joints and the loading capacity of hybrid joints of car body steels for the automotive indus-
try. International Journal of Adhesion & Adhesives, 45, 42.
Dancette, S., Fabrègue, D., Massardier, V., Merlin, J., Dupuy, T., & Bouzekri, M. (2012). Inves-
tigation of the tensile shear fracture of advanced high strength steel spot welds. Engineer-
ing Failure Analysis, 25, 112.
Davies, G. (2012). Material for automobile bodies. London: Butterworth-Heinemann.
Galán, J., Samek, L., Verleysen, P., Verbeken, K., & Houbaert, Y. (2012). Advanced high 
strength steels for automotive industry. Revista de Metalurgia, 48(2), 118.
Hayat, F. (2011). Comparing properties of adhesive bonding resistance spot welding and adhe-
sive weld bonding of coated and uncoated DP 600 steel. JISR International, 18(9), 70.
Kodama, S., Ishida, Y., Furusako, S., Saito, M., Miyazaki, Y., & Nose, T. (2013). Arc welding 
technology for automotive steel sheets. Nippon Steel Technical Report, 103, 83.
Kuziak, R., Kawalla, R., & Waengler, S. (2008). Advanced high strength steels for the automo-
tive industry. Archives of Civil & Mechanical Engineering, VIII.
Liu, H., Li, F., Shi, W., Swaminathan, S., He, Y., & Rohwerder, M. (2012). Challenges in hot-dip 
galvanizing of high strength dual phase steel: surface selective oxidation and mechanical 
property degradation. Surface Coating & Technology, 206, 3428.
Maa, J., Konga, F., Carlsonb, B., & Kovacevica, R. (2013). Two-pass laser welding of gal-
vanized high-strength dual-phase steel for a zero-gap lap joint configuration. Journal of 
Materials Processing Technology, 213, 495.
Ma, C., Chen, D. L., Bhole, S. D., Boudreau, G., Lee, A., & Biro, E. (2008). Microstructure and 
fracture characteristics of spot-welded DP600 steel. Material Science Engineering A, 485, 334.
Nayak, S. S., Baltazar Hernandez, V. H., Okita, Y., & Zhou, Y. (2012). Microstructure-hardness 
relationship in the fusion zone of TRIP steel welds. Material Science Engineering A, 551, 73.
Nemecek, S., Muzik, T., & Misek, M. (2012). Differences between laser and arc welding of 
HSS steel. Physics Procedia, 39, 67.
Radakovic, D. J., & Tumuluru, M. (2008). Predicting resistance spot weld failure modes in 
shear tension tests of advanced high-strength automotive steels. Welding Journal, 87, 
96-s–105-s.
Radakovic, D. J., & Tumuluru, M. (2012). An evaluation of the cross-tension test of resistance 
spot welds in high strength dual phase steels. Welding Journal, 91, 8S–15S.
Sam, S., & Shome, M. (2010). Static and fatigue performance of weld bonded dual phase steel 
sheets. Scientific World Journal, 15, 242.
Senuma, T. (2001). Physical metallurgy of modern high strength steel sheets. Iron and Steel 
Institute of Japan International, 41, 520.
Sharma, R. S., & Molian, P. (2011). Weldability of advanced high strength steels using an 
Yb:YAG disk laser. Journal of Materials Processing Technology, 211, 1888.
Tumuluru, M. D. (August 2006a). Resistance spot welding of coated high strength dual-phase 
steels. Welding Journal, 31.
Tumuluru, M. (2006b). A comparative examination of the resistance spot welding behavior of 
two advanced high strength steels. In SAE Technical Paper No. 2006-01-1214, presented 
at the SAE Congress, Detroit, MI.
Tumuluru, M. (2008). Some considerations in the resistance spot welding of dual phase steels. In 
Paper presented at the 5th International Seminar on advances in resistance welding, Septem-
ber 24–26, 2008, Toronto, Canada. Weston, Ontario, Canada: organized by Huys Industries.
Tumuluru, M. (2013). Evolution of steel Grades, joining Trends and Challenges in the automo-
tive Industry. In Invited keynote presentation, American welding society FABTECH Weld-
ing Show and Conference, Chicago IL.
Welding and Joining of Advanced High Strength Steels (AHSS). http://dx.doi.org/10.1016/B978-0-85709-436-0.00002-3
Copyright © 2015 Elsevier Ltd. All rights reserved.
Properties and automotive 
applications of advanced 
high-strength steels (AHSS)
T.B. Hilditch, T. de Souza, P.D. Hodgson
Deakin University, Waurn Ponds, Victoria, Australia
2
2.1 The automobile body
The automobile body is a highly complex structure that must simultaneously meet 
numerous functional, cost and aesthetic requirements. These requirements range from 
being a simple fixture to which other key subsystems are attached, such as the power-
train and suspension, to providing controlled crush zones for crashworthiness. These 
functions must generally be low cost and suitable for mass production. The automo-
bile body also establishes its unique style, an extremely critical design and marketing 
function. Vehicle styling is often one of the more dominant design factors and often 
the first point at which the vehicle’s form is developed. A typical mass-produced, 
passenger vehicle body is a large assemblyof stamped sheet metal components. Each 
component can serve a variety of specific structural and functional requirements. As 
a result, the geometry, material type and grade of the components vary significantly.
2.1.1 Body structure design requirements
The key performance requirements of an automobile body structure include structural 
static stiffness, durability, safety or crashworthiness and noise vibration and harsh-
ness. While the entire body structure must meet these requirements, at a basic level 
the many individual components can be categorised into two classes (Malden, 2011):
 1. Parts that react to loads with minimal deformation
 2. Parts that react to loads with significant deformation, which enhances the functions of the part.
It is, therefore, important to distinguish between these two functions. The first is 
dominated by stiffness properties, and the second depends on the strength and energy 
absorption characteristics of the structure.
Stiffness: The stiffness of a structural member is a function of the material’s modulus 
of elasticity and the geometry of the component, in particular its moment of inertia. 
Most components require a suitable amount of stiffness to meet loading requirements, 
in particular body components that support chassis/suspension components, and provide 
suitable reductions in noise vibration and harshness. Furthermore, the automobile body 
itself must have high levels of static bending and torsional stiffness to accommodate 
10 Welding and Joining of AHSS
road input loads and allow the ride and handling to be tuned. The elastic modulus of all 
steel grades is constant; therefore, component geometry is the primary design parameter. 
Substituting a steel grade with a higher strength or advanced high-strength steel (AHSS) 
does not improve component stiffness; however, the added formability of AHSS allows 
additional geometric form to be added to improve component stiffness. The additional 
component stiffness allows for reductions in the sheet thickness to reduce mass.
Strength: The strength of a component depends on its geometry and the material 
yield and tensile strengths. Strength-dominated components may be required to sup-
port a significant load with a controlled level of deformation to the structure. Other 
components may require a high level of energy absorption for very little deformation. 
For these strength-dominated components there is an obvious advantage of applying 
higher-strength materials, such as AHSS.
2.1.2 Body structure types
Numerous body structure types have been explored, all with their own level of success 
for specific applications. The most common forms are described below and shown in 
Figure 2.1. A summary of the advantages and disadvantages of each architecture type 
is also provided.
 1. Body on frame: the upper body structure is separated from a lower frame. The frame consists 
of a series of longitudinal and lateral closed-profile beams forming a ladder structure. This 
frame is the major load-bearing member. The body-on-frame architecture progressed from 
coach building and was one of the first vehicle architectures. Its use these days is limited to 
light trucks and niche vehicles.
 2. Spaceframe: a three-dimensional network of constant cross-sectional beams connected by 
shared nodes. These nodes are often welded intersections of the beams or cast joints or 
sometimes are adhesively bonded. The spaceframe structure separates itself from the styling 
surface and, as a result, can be optimised towards a structural and lightweight solution. The 
complexity of the joining methods, however, often limits production volumes to small num-
bers and high-performance niche vehicles.
 3. Central tunnel: dominated by a large, closed-profile structural member situated along the 
symmetrical axis of the vehicle. This closed tunnel integrates with suspension loading points 
and provides the majority of the vehicle’s structural integrity. The large tunnel is obtrusive to 
the occupant compartment and limited to applications in two- and four-seater vehicles.
Body on frame Spaceframe Central tunnel Monocoque
(d)(c)(b)(a)
Figure 2.1 Comparison of various automobile body architectures: body on frame (a), 
spaceframe (b), central tunnel (c) and monocoque (d).
Adapted from Malden (2011).
11Properties and automotive applications of AHSS
 4. Monocoque construction: integration of the vehicle’s exterior body and structural frame. 
The monocoque is the most common body structure type. It consists of stressed thin-wall 
panels, which form the exterior styling surface, integrated with closed-profile members. The 
combination of stressed skins and beam sections forms the major load-bearing members. 
The vehicle’s style dictates the initial form of the body; therefore a trade-off between the 
most structurally efficient solutions is made early. However, the monocoque construction 
provides a good balance in meeting this trade-off. Automated stamping lines and robotic 
spot-welding facilities make it cost-effective at high production volumes. Being the most 
common automobile body architecture, the requirements of the sheet material have driven 
the development of steel over the years, in particular the development of AHSS. The design 
approaches for monocoque construction of automobile bodies are the primary focus of the 
subsequent sections.
2.1.3 The elements of an automobile body
The typical passenger vehicle body makes up approximately 20% of the total mass of 
the vehicle (Davies, 2012), yet is the largest physical subsystem. The body in white 
(BIW) is the primary subassembly of the vehicle’s body and is often described as the 
‘skeleton’ of the vehicle. The BIW can be segregated into the ‘body-less doors’ 
and the ‘hang-on’ skin panels.
At a high level, the automobile body has two distinct safety features: an impene-
trable safety cell or occupant compartment and dedicated crumple zones. The safety 
cell must withstand extremely high loads with minimal deformation or intrusion. The 
dedicated crumple zones, however, are optimised to collapse in a controlled manner, 
absorbing the maximum amount of energy possible. Numerous crash-testing scenarios 
and standards are continually being introduced to ensure occupant safety in vehicles 
is improved.
For conventional monocoque construction, a combination of thin-sheet panels, 
closed-profile beams, joints and supporting brackets is used to perform its many func-
tions. The primary structure typically consists of longitudinal and lateral floor mem-
bers and a three-dimensional frame (safety cell). The safety cell consists of vertical 
pillars (A, B, C/D pillars), lateral roof beams, corner joint supports and the roof panel 
itself, as shown in Figure 2.2. Each of these structural elements is designed to suit 
various loading requirements and meet specific functional requirements.
 • Exterior body panels, such as door skins, bonnets and the roof panel, require high levels 
of stiffness and resistance to dent for out-of-plane loads. They require an ‘A’ class surface 
finish and their geometry is usually complex, requiring highly formable materials.
 • The main floor and front/rear bulkheads react to in-plane loading, requiring moderate 
strength levels and rigidity. Some complex geometric form is required for stiffness; therefore 
suitable material formability is needed.
 • Longitudinal and lateral beam sections provide tensile/compressive/bending stiffness and 
controlled impact resistance. These components often consist of an inner, outer and internal 
reinforced stamped profile spot-welded together along a common flange. A variety of 
strength grades are needed for these components, depending on their function. Structural 
elements in the safety cell require very high strength but little or no deformation, whereas 
elements in the crushable zones undergo significant deformation to absorb maximum energy.
12 Welding and Joining of AHSS
While geometryplays a pivotal role in each of these component categories, the 
materials applied are equally if not more important. Therefore, to meet these per-
formance requirements, a number of different material types and grades are used in 
the automobile body. Furthermore, demands by government legislation and consumer 
requirements for improved safety and vehicle efficiency have become more stringent 
over the past two decades. These performance changes have challenged automotive 
designers and material suppliers to develop new technologies to meet these growing 
needs.
2.1.4 Material usage trends
Steel is the primary material used for automobile body structures because of its versa-
tility and cost. Sheet steels used in the automotive industry have conventionally been 
chosen for their good formability characteristics, allowing them to be conveniently 
stamped at room temperature into the designed component shapes. These initial steels 
had a predominantly ferrite microstructure, resulting in relatively low strength levels 
and high ductility. The strengthening of steel using mechanisms such as solid solution 
strengthening, grain refinement and precipitation strengthening all typically result in 
a decrease in formability. This trade-off in formability had previously limited the use 
of higher-strength steel (and hence thinner-gauge steel) in the automotive industry. 
Figure 2.3 shows the reduction in elongation with increasing yield strength for a range 
of steels, including conventional high-strength steel (HSS) currently used in automo-
tive body structures, such as high-strength low-alloy (HSLA) and bake-hardenable 
steels. These traditional steels have an ultimate tensile strength less than 600 MPa. In 
the 1980s, low-strength, drawing quality steel dominated the automobile body, with 
only a small fraction of hot-rolled higher-strength steel used.
Figure 2.2 A brief overview of some key body structure elements.
Adapted from AISI (1998).
13Properties and automotive applications of AHSS
In the past several decades, the importance of safety and vehicle emissions has 
increased. This has led to the need for higher-strength steels to improve the crash per-
formance of automotive structures while allowing for a reduction in thickness. There 
was also an imperative to introduce higher-strength sheet steels while maintaining 
formability to allow automotive manufacturers to retain existing manufacturing pro-
cesses and equipment, as well as to maintain design flexibility.
The Ultralight Steel Auto Body (ULSAB) project (AISI, 1998) provided a launch-
ing pad for the introduction of AHSS for use in automobile bodies, to the point that 
current passenger vehicles have a significant amount of AHSS in key structural and 
crash-optimised areas. The first-generation AHSS introduced in sheet metal struc-
tures, such as dual-phase (DP) and transformation-induced plasticity (TRIP) grades, 
had a yield strength range similar to that of conventional HSS grades, with higher 
work-hardening rates and formability. This was achieved using a ferrite matrix to pro-
vide ductility with a harder second phase to provide strength. In the case of TRIP 
steels, metastable austenite was retained in the microstructure to provide further 
ductility and work hardening to higher strain levels. This improved strength–ductility 
combination for these grades is shown in Figure 2.3 as a deviation above the normal 
elongation–strength curve that the lower-strength sheet steel grades follow.
Increasingly stringent emissions laws between 2000 and 2010 led to the need for 
increasingly higher strength grades. The introduction of higher-strength DP and TRIP 
grades (tensile strength of 800 MPa) posed greater challenges; while they can be cold 
Figure 2.3 Elongation versus yield strength for the different classes of sheet steel. AHSS, 
advanced high-strength steel; BH, bake-hardenable; CMn, carbon–manganese; CP, complex phase; 
DP, dual phase; HSLA, high strength, low alloy; HSS, high-strength steel; IF, interstitial-free; MS, 
martensitic; TRIP, transformation-induced plasticity; TWIP, twinning-induced plasticity.
14 Welding and Joining of AHSS
stamped, the significantly higher press forces cause difficulties in terms of dimen-
sional control because of springback and curl, as well as tool wear. The introduction 
of 1000-MPa tensile strength DP and TRIP grades and 1000–1500-MPa martensi-
tic grades has resulted in the need for changes in the forming processes from tradi-
tional cold stamping, with alternative processes such as roll forming, hydroforming 
and hot stamping. This has led more recently to a range of alternative methods to 
produce martensitic sheet grades based on stamping the material, either at high 
temperatures or while in a softened state and heat treating afterwards, to achieve 
necessary strength levels.
In addition to the more widespread usage of DP, TRIP and martensitic steels in 
automotive structures, several more specialised grades have also been developed. 
Ferrite–bainite (FB) grades were developed and introduced in response to the poor 
performance of steel containing martensite in stretch-flange or hole expansion appli-
cations, whereas complex-phase (CP) steels were developed for components requiring 
low strain to form and for which a high yield strength is beneficial.
There are a number of newer-generation steels being developed with the automo-
tive sheet steel industry in mind. Twinning-induced plasticity (TWIP) steels are part 
of the second-generation AHSS that cover the same tensile strength ranges as DP and 
TRIP but have significantly higher ductility. This greater ductility comes at a large 
alloying cost, which is a major barrier to widespread usage. Other variations of multi-
phase microstructures also are being developed, particularly based on austenite as the 
dominant microstructural constituent.
2.2 AHSS microstructures and tensile properties
There are a number of different steel grades typically classified as either first- or 
second-generation AHSSs that are available or being developed for use in automotive 
applications. The Future Steel Vehicle Program (WorldAutoSteel, 2011), designed to 
highlight the potential of steels for use in 2015–2020 vehicles, used a number of these, 
including DP, TRIP, CP, FB, martensitic, hot-forming and TWIP. AHSS grades are 
commonly designated by their nominal tensile strength (e.g. DP600 has a nominal 
minimum tensile strength of 600 MPa).
2.2.1 Dual phase
DP steels have a microstructure that typically consists of martensite islands 
surrounded by a ferrite matrix, as shown in Figure 2.4. Basic steel processing involves 
a short annealing time in the intercritical (ferrite and austenite) region of the phase 
diagram to produce a structure of ferrite and austenite. Partitioning occurs during 
annealing, causing the austenite to become enriched with carbon. Sufficiently rapid 
cooling follows to transform the austenite to martensite. The chemical composition 
is based on an approximate weight percentage of 0.1 carbon and 1.5 manganese, 
though this varies slightly depending on the grade. The higher carbon and manga-
nese contents compared with conventional sheet steels are important to obtain the 
15Properties and automotive applications of AHSS
necessary hardenability, which assists in preventing pearlite or bainite from forming 
during processing. Silicon can be added to promote the partitioning of carbon to 
austenite.
DP steels are typically characterised by continuous yielding and a low yield-to-tensile 
strength ratio that results in high initial work-hardening rates. The continuous, low yield 
strength is related to the soft ferrite phase, whereas the high tensile strength is related to the 
hard martensite regions. The volume change associated with the austenite-to-martensite 
transformation during processing generates dislocations in the ferrite matrix, which fur-
ther contribute to the low yield strength and thecontinuous yielding behaviour. The 
increased strain-hardening rate of DP steels compared with conventional low-carbon 
and HSLA steels is shown in Figure 2.5. A higher work-hardening rate means that after 
forming, the DP steel has a higher flow stress/strength level than the HSLA grade. This 
higher flow stress is beneficial in part for both fatigue and crash behaviour, and it allows 
thinner-gauge material to be used.
The common range of DP grades designed for the automotive industry is DP500 
to DP1000. The tensile strength of DP steels increases with an increase in the volume 
fraction of martensite, as shown in Figure 2.6 (Davies, 1978). Lower-strength grades 
have approximately 20% martensite. There is a reasonably linear trade-off between 
tensile strength (volume fraction of martensite) and ductility for DP grades (Sakuma, 
2004). The ductility of DP steels is comparable to or better than HSLA grades with a 
similar tensile strength.
Figure 2.6 shows that the yield strength of a DP steel also increases linearly with 
increasing martensite volume fraction (Davies, 1978). Because the soft ferrite matrix 
is primarily responsible for the deformation that occurs within the materials, the dis-
tribution or ‘island’ size of martensite has an effect on the uniform elongation and the 
tensile strength (Llewellyn & Hudd, 1998), whereas the ferrite grain size can influence 
the yield strength for a given martensite volume fraction (Ramos, 1979). It has been 
suggested that the optimum combination of strength and formability is obtained by a 
very fine distribution of martensite islands and a very fine ferrite grain size (Llewellyn & 
Hudd, 1998). DP600 and DP780/800 are widely used in the automotive industry.
Figure 2.4 Scanning electron image of a dual-phase steel microstructure showing ferrite (F) 
and martensite (M) (Kang, Han, Zhao, & Cai, 2013).
16 Welding and Joining of AHSS
9000
8000
7000
6000
5000
4000
3000
2000
1000
0
0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16 0.18 0.2 0.22
HSLA 350
DP 600
IF steel
True strain
St
ra
in
-h
ar
de
ni
ng
 ra
te
 (M
Pa
)
Figure 2.5 Increase in strain-hardening rate of dual-phase (DP) steels compared with 
conventional steels for a range of true strains. HSLA, high strength, low alloy; IF, interstitial-free.
Tensile strength
Percent martensite
10 20 30 40 50 60 70 80 90 100
Flow stress
(ε = 0.002)
30
0
20
0
10
00
0
20
00
10
0
0
730
740
760
780
800
820
840
Quench temp. °C
K
Si
St
re
ss
 (M
Pa
)
Figure 2.6 Effect of martensite volume on both the yield and tensile strength of dual-phase 
steel (Davies, 1978).
17Properties and automotive applications of AHSS
2.2.2 Transformation-induced plasticity
TRIP steels are a multiphase steel typically containing ferrite, retained austenite, and 
bainite and/or martensite, as shown in Figure 2.7. Austenite can be made stable at 
room temperature by accumulating enough carbon during processing to depress the 
temperature at which austenite transforms to martensite below room temperature. 
TRIP steels are intercritically annealed in the ferrite and austenite region, similar to 
DP steels, but are cooled to a temperature of approximately 400 °C (austempering 
temperature) to develop the transformation product (bainite). During intercritical 
annealing, the carbon partitions to the austenite. During austempering, the bainite 
formation rejects more carbon into the austensite. The austempering time and tem-
perature thus are a trade-off between the amount of bainite formed, the stability of the 
austenite (and hence the martensite formed upon cooling) and the amount of austenite 
retained at room temperature.
The chemical composition of a TRIP steel is based on carbon–manganese, with 
either silicon or aluminium additions. Both carbon and manganese concentrations are 
higher in TRIP steels than in DP steels, with a weight percentage of approximately 
0.1–0.15 carbon and ∼2.0 manganese for a TRIP600 steel. The addition of silicon or 
aluminium assists in suppressing carbide formation during bainitic transformation, 
thus making it easier for carbon to be rejected into the austenite.
Available TRIP steel grades cover the same approximate ultimate tensile 
strength range as DP steels (500–1000 MPa). Higher-strength grades of TRIP steel 
typically retain more austenite and other transformation products (bainite and/or 
martensite) and hence less low-strength ferrite. While TRIP steels have higher 
yield strength than DP steels for a given ultimate tensile strength, they also have 
greater elongations. The instantaneous work-hardening rate of TRIP steels is lower 
Figure 2.7 Scanning electron image of a transformation-induced plasticity steel micro-
structure showing bainite (B), martensite (M) and retained austenite (RA) in a ferrite matrix 
(Chiang, Lawrence, Boyd, & Pilkey, 2011).
18 Welding and Joining of AHSS
than DP steels at low strains, but the improved ductility results in an increase in 
work-hardening rate with increasing strain, whereas that of DP steel decreases 
(Figure 2.8). In addition to the higher relative yield strength, TRIP steels also 
tend to show discontinuous yielding, although the amount of yield point elonga-
tion is generally low. The superior strength/ductility combination of TRIP steels, 
as shown in Figure 2.3, compared with conventional steels grades is due to the 
transformation of the retained austenite to martensite during room temperature 
deformation. This transformation occurs as the austenite that is retained at room 
temperature is metastable and transforms into martensite if subjected to enough 
strain. The transformation delays the onset of necking, leading to a high uniform 
elongation.
Not all austenite transforms during straining (Streicher, Speer, & Matlock, 2002), 
however, and this is because the carbon content of the retained austenite is too high 
or, alternatively, the orientation, size or morphology of the retained austenite regions 
are unsuitable for transformation. The volume fraction of austenite transformation 
is also dependent on the strain path, though it does not appear to vary significantly 
between different forming modes (Sugimoto, Kobayashi, Nagasaka, & Hashimoto, 
1995).
2.2.3 Complex phase
CP steels have an ultimate tensile strength of around 800–1200 MPa while retain-
ing a reasonable level of ductility (approximately 7–15%). They have higher carbon 
Figure 2.8 Comparison of instantaneous work hardening (n-value) between high-strength, 
low-alloy (HSLA), transformation-induced plasticity (TRIP) and dual-phase (DP) steels.
19Properties and automotive applications of AHSS
and manganese concentrations than DP and TRIP steels (around 0.15 wt% carbon and 
2 wt% manganese) and a microstructure that often contains ferrite and bainite with 
small amounts of pearlite, martensite and retained austenite. These steels have a very 
fine microstructure obtained via alloying additions such as titanium, vanadium and 
niobium that form precipitates to assist in preventing grain growth during processing. 
The highly refined grain size, combined with the presence of precipitates and harder 
phases such as martensite, ultimately results in a material with a high yield strength. 
The high yield strength (approximately 600–1000 MPa) means that these steels have 
a lower work-hardening rate compared with DP and TRIP steels, although they are 
still formable using conventional cold stamping processes (though to less complex 
geometry).
2.2.4 Ferrite–bainite
FB steels are a variation of DP steel that combines ferrite with bainite as a sec-
ond phase instead of martensite. The properties of FB steels are thus similar to 
ferrite–martensite DP steels, though with slightly lower strength values for a given 
second-phase volume fraction due to bainite being a lower-strength phase than 
martensite. FB steels cover a tensile strength range of approximately 500–900 MPa, 
with a corresponding total elongation of 30–10%. The FBsteels were primarily 
developed for edge-stretching applications where DP and TRIP steels can result in 
poor formability because of premature cracking (particularly for sheared edges). 
The improved performance of FB steels in these applications is due to the decreased 
likelihood of cracks forming in bainite during shearing operations (such as hole 
punching or blanking).
2.2.5 Martensitic steels
Martensitic steels have a predominantly lath martensite microstructure, as shown 
in Figure 2.9, and are formed by continuous annealing in the austenite region 
followed by rapid quenching of the steel. Higher hardenability is achieved by 
increased carbon content, typically of the order of 0.25 weight percent carbon. The 
manganese content is also reasonably high, around 1.5 Mn, and small additions 
of boron may also be included to further increase hardenability. The increased 
hardenability via alloying reduces the quenching time necessary to achieve a fully 
martensitic structure.
Martensitic sheet steels have tensile strengths ranging from 900 to 1600 MPa, 
with total elongations typically around 4–7%. The yield stress ranges from 800 to 
1350 MPa, meaning that these steels have very low work-hardening behaviour. The 
strength is also related to the carbon content of the microstructure, with increasing 
carbon resulting in increasing strength.
Forming martensitic steels is difficult because of the exceptionally high yield stress 
and low ductility. At room temperature, roll forming is the primary shaping method. 
Roll forming limits the complexity of part design using martensitic steels, thus limit-
ing the potential uses of martensitic steels.
20 Welding and Joining of AHSS
2.2.6 Hot-formed steels
Hot-formed steels are a variation of martensitic steels that are designed to be formed 
at temperatures in excess of 850 °C and quenched at rates faster than 50 °C/s. The 
preheated sheet steel typically has an aluminide coating to limit oxidation and is 
formed at temperatures in the austenite region where the formability is high. The 
component is then rapidly quenched via water jets while still under load in the die/
press (Vaissiere, Laurent, & Reinhardt, 2002). Hot-formed steels typically have 
small boron additions (between 0.002% and 0.005%) for high hardenability to 
ensure that martensite is formed during quenching. There are two alternative pro-
cesses: the direct process, where all forming is done at elevated temperatures, and 
the indirect process, where some forming is initially done at room temperature with 
steel in a softened state before final forming at elevated temperatures. The processes 
have much lower productivity than conventional cold stamping, but are achieving 
increased usage as the only way to produce complex shapes at strength levels of 
approximately 1000–1300 MPa.
2.2.7 Post-forming heat-treated steels
Post-forming heat-treated steels are fully formed at room temperature while the steel 
is in a softened state. After forming, the component is heat treated and quenched to 
obtain a high-strength microstructure. Fast quenching to obtain a martensitic micro-
structure often requires fixtures to prevent distortion, although some air-hardening 
grades that show a bainitic and/or martensitic microstructure are currently available. 
These air-hardened steels are based on a composition of approximately 0.15% carbon, 
with additions of silicon, manganese, chromium, molybdenum, vanadium and nitro-
gen. Yield and tensile strengths are similar to those of the martensitic sheet steels, with 
ductility similar to HSLA steels.
20kV ×3000 2 µm
Figure 2.9 Scanning electron image of a martensitic microstructure (M1200) (Wang 
et al., 2013).
21Properties and automotive applications of AHSS
2.2.8 Twinning-induced plasticity
TWIP steels have an austenitic structure at room temperature that is stable because of 
a high manganese content (ranging from 15% to 30%) and a carbon content of 0.6%. 
As TWIP steels are strained, deformation-nucleated twins (Frommeyer, Brux, & 
Neumann, 2003; Prakash, Hochrainer, Reisacher, & Reidel, 2008) typically form 
in the austenitic structure because of a low stacking fault energy (SFE). The twins 
act as dislocation barriers and reduce the effective mean free path of dislocations, 
which increases the flow stress. These twins are quite thin, and there is a continuous 
nucleation of new, increasingly smaller deformation twins. These steels also have a 
high rate of dislocation accumulation independent of twin formation due to reduced 
cross-slip resulting from the low SFE. The resulting tensile strength ranges of TWIP 
steels are similar to those of DP and TRIP steels (600–100 MPa); however, the tensile 
ductility range is significantly higher at 40–80% (Figure 2.10). Ductility and strength 
are related to the manganese content; smaller manganese additions typically show 
higher strength and lower ductility. The high total elongation levels present a chal-
lenge for automotive usage because significantly higher strains need to be imparted to 
the component than those that normally occur in traditional stamping processes to get 
similar final component strength levels in TRIP and DP steels.
Aluminium and silicon are other alloying additions that are often made to TWIP 
steels. Aluminium increases the SFE, which suppresses martensite transformation, 
whereas silicon decreases the SFE, which sustains the martensite transformation 
(Frommeyer et al., 2003). As a result, steels with larger silicon additions have a 
higher strength and tend to transform to martensite rather than twin (behaving 
more like a TRIP steel), whereas aluminium additions lower tensile strength and 
work hardening.
Figure 2.10 Strength versus ductility ranges for common twinning-induced plasticity steels. 
Al, aluminium; C, carbon; Fe, iron; Mn, manganese; N, nitrogen; Si, silicon; TS, tensile 
strength; YS, yield strength (DeCooman, Chin, & Kim, 2011).
22 Welding and Joining of AHSS
2.3 Formability and fracture of AHSS
There are a range of issues related to the manufacturing and performance of automo-
tive structures. The forming mode in particular can have a major effect on the way a 
steel is able to meet shape requirements without extensive thinning or cracking. Sheet 
formability is generally expressed through a forming limit diagram (FLD), whereas 
other forming issues relate to hole expansion and fracture behaviour. Elastic recovery 
after a part is formed is a major issue with AHSS, and this is reflected in springback 
behaviour.
2.3.1 Forming limits
There are a range of different forming modes that are relevant in sheet metal stamp-
ing of AHSS. The forming limits are often characterised as the critical necking strain 
for a strain mode using an FLD. In general, increased formability is related to the 
distribution of strain during forming, and as such work hardening is a key parame-
ter. The higher work hardening of DP and TRIP steels compared with conventional 
HSLA steels results in higher forming limits in most forming modes (Keeler & 
Brazier, 1975). TWIP steels have better stretch-forming properties than AHSS of simi-
lar strength. DP, TRIP and TWIP steels are fairly isotropic, however, and r values tend 
to be around 1. Lower r values limit the deep drawability of AHSS.
When stamping mild and conventional HSS, failure is typically localised neck-
ing, followed by splitting. Failure behaviour of DP and TRIP steels can be accurately 
described using an FLD in cases where localised necking occurs. When stamping an 
AHSS, such as DP980, with a large amount of bending under tension or on stretched 
edges, fracture can occur where there is limited thinning. While there is a ductile frac-
ture surface, there are very large cracks and a large amount of elastic energy release. 
This fracture initiation can be difficult to predict. This type of fracture has occurred in 
DP steel in biaxial stretching rather than normal thinning and splittingand was poten-
tially related to differences in the strength of the different phases, that is, soft ferrite 
and hard martensite (Nikhare, Hodgson, & Weiss, 2011), as shown in Figure 2.11.
2.3.2 Hole expansion
Shearing processes, such as blanking and punching, are commonly used in sheet metal 
stamping either before or during forming. In many instances the sheared edge of the 
sheet may be exposed to a subsequent forming process that causes it to elongate, such 
as stretch-flanging or hole expansion. While the hole expansion limit tends to decrease 
with an increase in material tensile strength, inhomogeneous microstructures, partic-
ularly those containing martensite, have poorer hole expansion limits (such as DP 
and TRIP steels). The implication means that many of the AHSS grades that contain 
martensite are not suited to stamping processes such as stretch-flanging (AISI, 2003a). 
This particular limitation is the main impetus in the development of FB steels; bainitic 
or ferrite–bainitic microstructures are significantly better in this particular forming 
mode than those containing martensite.
23Properties and automotive applications of AHSS
2.3.3 Dimensional accuracy
Dimensional accuracy is of vital importance in sheet-formed body structure compo-
nents because of subsequent assembly requirements. Elastically driven shape changes 
such as springback and curl can occur once the component is released from the die. 
Since these shape changes increase in magnitude with increasing flow strength and 
decreasing material thickness, they are a significantly greater issue in the use of AHSS 
compared with conventional drawing steels (Davies, 1984). Considerable recent 
research is directed toward both quantifying and predicting the amount of springback 
in a given forming process. Tooling design can compensate for these shape changes 
using methods such as increasing the blank holder force; however, this often requires 
accurately predicting the shape change using numerical simulations. Hot-formed and 
post-forming heat-treated steels are advantageous in this instance because forming 
is completed while the steel is in a softened state; hence shape change due to elastic 
recovery processes is not a significant issue.
2.3.4 Bake hardening
After an automotive body structure has been assembled it typically is painted. The rela-
tively high dislocation density present in both the unstrained (as received) and strained 
(after forming) ferrite means that DP and TRIP steels typically show bake-harden-
ing properties (Fredriksson, Melander, & Hedman, 1989). This means that there is 
a significant increase in flow stress after exposure to times and temperatures similar 
Figure 2.11 Forming limit diagram (FLD) of dual-phase (DP) versus transformation-induced 
plasticity (TRIP) versus high-strength, low-alloy (HSLA) steels, showing the higher forming 
strains before necking for the advanced high-strength steels and lower fracture limits (Nikhare 
et al., 2011).
24 Welding and Joining of AHSS
to that encountered in the automotive paint-baking process (approximately 170 °C 
for 20 min). The flow stress increase is due to the diffusion of soluble carbon (and 
nitrogen) to mobile dislocations within the structure that results in the dislocations 
being pinned and resistant to further motion. This strength increase is beneficial in 
most aspects of automotive applications, including dent resistance (Dicello & George, 
1974), fatigue and crash behaviour (AISI, 2003b).
2.4 Automotive in-service properties
Automotive body structures often are designed to provide stiffness to the vehicle, 
withstand vibration and fatigue loading during normal use, as well as provide pas-
senger safety in crash events. While in most cases it is assumed that fatigue failure 
will be dominated by the joints in the structure, crash safety is provided by both 
the geometry of the component and the properties of the metal. Components that 
are used in crash applications require either deformation resistance or high energy 
absorption during deformation. These are both required at high strain rates, often 
up to 200/s.
2.4.1 Deformation resistance
The passenger safety cage requires high deformation resistance; hence the higher 
strength the better. AHSSs are better than conventional steels for deformation-resistance 
applications because of their higher yield strength; martensitic steels have the high-
est yield strength, in excess of 800 MPa. Because of the low yield strength-to-tensile 
strength ratio of DP steels, these steels require higher levels of strain during forming to 
achieve higher strength levels in the final part/application. As with all steels, an increase 
in strain rate results in an increase in both the yield stress and tensile strength of AHSSs 
(AISI, 2003b; Choi et al., 2002). The magnitude of the increase depends on the initial 
strength of the steel; higher-strength steels are generally less sensitive to strain rate 
(AISI, 2003b). This effect is shown in Figure 2.12, where the ultimate tensile strength 
ratio at 100/s to 10−3/s is plotted against the tensile strength of various steel grades. 
Thus, while AHSSs do not achieve the same increase in strength as conventional sheet 
Figure 2.12 Reduction in the increase in 
tensile strength as a result of an increase in 
strain rate with increasing tensile strength 
for a range of different steels (ULSAB, 
2001). BH, bake-hardenable; CP, complex 
phase; DP, dual phase; HSLA, high 
strength, low alloy; IF, interstitial free; 
Mart, martensite; UTS, ultimate tensile 
strength.
25Properties and automotive applications of AHSS
steels, there is still a noticeable positive increase. While the n value tends to decrease 
with increasing strain rate for conventional steels, for many AHSSs the n value remains 
fairly constant with increasing strain rate.
2.4.2 Energy absorption
The motor compartment is predominantly designed for crush and hence is suited to 
steels that allow higher energy absorption. The strength and thickness of these com-
ponents typically need to be balanced with section size such that under axial load-
ing the sections collapse in a stable and uniform manner (Horvath & Fekete, 2004). 
Energy absorption is the area under a stress–strain curve and thus highly depends on 
the tensile strength of a material. Energy absorption can be calculated either at neck-
ing, showing the total energy that can be absorbed by a material, or, alternatively, at a 
specified strain level to allow materials to be compared for a given strain. In crash sit-
uations it has been suggested that the majority of energy is absorbed at plastic strains 
of up to 10% (ULSAB, 2001), and this strain level is often used to compare the energy 
absorption capability of different materials. There can be a significant difference in the 
energy absorption ability of a particular steel grade, depending on the strain level spec-
ified (AISI, 2003b; Bleck, Larour, & Baumer, 2004). When examining AHSS, higher 
elongation grades such as TRIP600 and TRIP780 have the highest energy absorption 
at necking, as shown in Figure 2.13, whereas higher-strength grades such as TRIP980 
have a higher 10% strain energy absorption, as shown in Figure 2.10. The higher yield 
strength means that CP steels have very high energy absorption capacity in the elastic 
and low plastic strain range, which is useful for applications where very little forming 
strain is introduced into the material. Martensitic steels typically have only around 6% 
elongation and are thus not considered for energy absorption applications.
The work-hardening exponent of a material is also an important factor in energy 
absorption during a crash. Higher work hardening distributes strain more evenly and 
causes a greater volume of material within a part to undergo deformation, hence 
greatly increasing the total energy absorption. TWIP, TRIP and DP steels have higher 
n values compared with conventional steels, whereasboth CP and martensitic steels 
have relatively low n values. Work-hardening values typically decrease with increas-
ing strength of the AHSS.
2.5 Current and future trends in AHSS
In recent years there has been significant research into developing the third generation of 
AHSSs. While second-generation AHSSs such as TWIP steels show superior strength–
ductility combinations, they are often considered incompatible with current automo-
tive stamping processes because of the exceedingly high deformation levels required to 
obtain maximum strength, and the significantly higher alloying contents make cost and 
welding significant issues. A number of concepts are currently being used to develop 
third-generation AHSSs that show improved combinations of strength and ductility 
compared with the first generation, without the alloying cost of the second generation.
26 Welding and Joining of AHSS
One example is lowering the manganese content from the typical TWIP con-
centrations to between 4 and 6 wt%. These steels exhibit a combination of TRIP 
and TWIP behaviours. However, this manganese concentration can still lead to pro-
cessing difficulties and associated costs. Other attempts have included combining 
ultrafine ferrite grain sizes in a TRIP steel by controlled thermomechanical process-
ing. Again, laboratory results show promise, but implementation in a production 
0.200
0.180
0.160
0.140
0.120
0.100
0.080
0.060
0.040
0.020
0.000
0.001 0.01 0.1
Strain rate (1/s)
(T
S+
YS
)*
U
E/
2 
(J
/m
m
3 )
1 10 100 1000
DP600-Gl
DP600-HR
DP800-GA
TRIP590-EG
TRIP600-CR
TRIP780-CR
TRIP980-CR
(a)
0.1
0.09
0.08
0.07
0.06
0.05
0.04
0.03
0.02
0.01
0
0.001 0.01 0.1 1 10 100 1000
Strain rate (1/s)
E 
10
%
 (J
/m
m
3 ) DP600-GlDP600-HR
DP800-GA
TRIP590-EG
TRIP600-CR
TRIP780-CR
TRIP980-CR
(b)
Figure 2.13 Effect of strain rate on energy absorption before necking (a) and at 10% strain 
(b) for different transformation-induced plasticity (TRIP) and dual-phase (DP) steels.
AISI (2003b), with kind permission from the Steel Market Development Institute.
27Properties and automotive applications of AHSS
environment seems difficult. Overall, this is one of the major concerns with any new 
steel grade. The cost pressures are such that these steels really do need to be able to 
be processed along with other grades without major processing changes. They also 
need to demonstrate high levels of product uniformity and repeatability within and 
between batches. This is a real challenge for these more complex strip compositions 
and processing routes.
One option that has attracted considerable research interest is the quenching and 
partitioning process; this involves compositions that are similar to current grades and 
a heat treatment process that is commercially feasible, although to date the work is 
restricted to the laboratory. Quenching and partitioning steels have a higher range of 
tensile strength than TRIP and DP steels (1300–1800 MPa), with elongation in the 
range of 16–18% (Cao, Wang, Shi, & Dong, 2010).
References
AISI. (1998). UltraLight steel auto body final report. American Iron and Steel Institute.
AISI. (2003a). Formability characterization of a new generation of high-strength steels. American 
Iron and Steel Institute/U.S. Department of Energy Technology Roadmap Program. 
Report No. TRP 0012.
AISI. (2003b). Characterization of fatigue and crash performance of new generation high-strength 
steels for automotive applications. American Iron and Steel Institute/U.S. Department 
of Energy Technology Roadmap Program. Report No. TRP 0038.
Bleck, W., Larour, P., & Baumer, A. (2004). High strain rate tensile testing of modern car 
body steels. In J. F. Nie, & M. Barnett (Eds.), Paper presented at the 3rd International 
 Conference on Advanced Materials Processing (pp. 21–29). Australia: Institute of Materials 
Engineering.
Cao, W., Wang, C., Shi, J., & Dong, H. (2010). Application of quenching and partitioning to improve 
ductility of ultrahigh strength low alloy steel. Materials Science Forum, 654–656, 29–32.
Chiang, J., Lawrence, B., Boyd, J. D., & Pilkey, A. K. (2011). Effect of microstructure on 
retained austenite stability and work hardening of TRIP steels. Materials Science and 
Engineering A, 528, 4516–4521.
Choi, I. D., Bruce, D. M., Kim, S. J., Lee, C. G., Park, S. H., Matlock, D. K., et al. (2002). 
Deformation behavior of low carbon TRIP sheet steels at high strain rates. ISIJ Interna-
tional, 42(12), 1483–1489.
Davies, R. G. (1978). Influence of martensite composition and content on the properties of dual-
phase steels. Metallurgical Transactions A, 9A, 671–679.
Davies, R. G. (1984). Side-wall curl in high strength steels. Journal of Applied Metalworking, 
3(2), 120–126.
Davies, G. (2012). Materials for automobile bodies. Boston, MA: Butterworth-Heinemann.
DeCooman, B. C., Chin, K., & Kim, J. (2011). High Mn TWIP steels for automotive applications. 
In M. Chiaberge (Ed.), New trends and developments in automotive system engineering. 
InTech, ISBN: 978-953-307-517-4. http://dx.doi.org/10.5772/14086. Available from 
http://www.intechopen.com/books/new-trends-and-developments-in-automotive-system- 
engineering/high-mn-twip-steels-for-automotive-applications.
Dicello, J. A. G., & George, R. A. (1974). Design criteria for the dent resistance of auto body 
panels. SAE Technical Paper No. 740081. Detroit, MI: SAE World Congress.
http://dx.doi.org/10.5772/14086
http://www.intechopen.com/books/new-trends-and-developments-in-automotive-system-engineering/high-mn-twip-steels-for-automotive-applications
http://www.intechopen.com/books/new-trends-and-developments-in-automotive-system-engineering/high-mn-twip-steels-for-automotive-applications
28 Welding and Joining of AHSS
Fredriksson, K., Melander, A., & Hedman, M. (1989). Influence of prestraining and ageing 
on the fatigue properties of a dual-phase sheet steel with tensile strength of 410 MPa. 
Scandinavian Journal of Metallurgy, 18, 155–165.
Frommeyer, G., Brux, U., & Neumann, P. (2003). Supra-ductile and high-strength manganese 
TRIP/TWIP steels for high energy absorption purposes. ISIJ International, 43(3), 438–446.
Horvath, C. D., & Fekete, J. R. (2004). Opportunities and challenges for increased usage of 
advanced high strength steels in automotive applications. Paper presented at the Interna-
tional Conference on Advanced High Strength Sheet Steels for Automotive Applications, 
Winter Park, CO.
Kang, Y., Han, Q., Zhao, X., & Cai, M. (2013). Influence of nanoparticle reinforcements on the 
strengthening mechanisms of an ultrafine-grained dual phase steel containing titanium. 
Materials and Design, 44, 331–339.
Keeler, S. P., & Brazier, W. G. (1975). Relationship between laboratory material characteriza-
tions and press shop formability. Microalloying, 75, 517–530.
Llewellyn, D. T., & Hudd, R. C. (1998). Steels – metallurgy and applications (3rd ed.). Oxford: 
Butterworth Heinemann.
Malden, D. E. (2011). Fundamentals of automobile body structure design. SAE International.
Nikhare, C., Hodgson, P., & Weiss, M. (2011). Necking and fracture of advanced high strength 
steels. Materials Science and Engineering A, 528, 3010–3013.
Prakash, A., Hochrainer, T., Reisacher, E., & Reidel, H. (2008). Twinning models in self-consistent 
texture simulations of TWIP steels. Steel Research International, 79(8), 645.
Ramos, L. F. V. (1979). A study of strengthening mechanisms of the Fe-C-Mn dual-phase steels. 
M.S. Thesis No. T-2189. Golden, CO: Colorado School of Mines.
Sakuma, Y. (2004). Recent achievements in manufacturing and application of high-strength 
steel sheets for automotive body structure. In Paper presented at the International 
Conference on Advanced High Strength Sheet Steels for Automotive Applications, Winter 
Park, CO.
Streicher, A. M., Speer, J. G., & Matlock, D. K. (2002). Forming response of retained austenite 
in a C-Si-Mn high strength TRIP sheet steel. In Paper presentedat the International 
Conference on TRIP Steels, Ghent, Belgium.
Sugimoto, K., Kobayashi, M., Nagasaka, A., & Hashimoto, S. (1995). Warm stretch-formability 
of TRIP-aided dual phase sheet steels. ISIJ International, 35(11), 1407–1414.
ULSAB. (2001). ULSAB-AVC body structure materials. Technical Transfer Dispatch #6. 
Available from http://www.autosteel.org/∼/media/Files/Autosteel/Programs/ULSAB-AVC
/avc_ttd6.pdf. Accessed 12.11.12.
Vaissiere, L., Laurent, J. P., & Reinhardt, A. (2002). Development of pre-coated boron steel 
for applications on PSA Peugeot Citroen and RENAULT bodies in White. SAE Paper No 
2002-01-2048. Detroit, MI: SAE World Congress.
Wang, W., Li, M., He, C., Wei, X., Wang, D., & Du, H. (2013). Experimental study on high 
strain rate behavior of high strength 600–1000 MPa dual phase steels and 1200 MPa fully 
martensitic steels. Materials and Design, 47, 510–521.
WorldAutoSteel. (2011). Future steel vehicle overview report. Available from http://c315221. 
r21.cf1.rackcdn.com/FSV_OverviewReport_Phase2_FINAL_20110430.pdf. Accessed 
12.11.12.
http://www.autosteel.org/%7E/media/Files/Autosteel/Programs/ULSAB-AVC/avc_ttd6.pdf
http://www.autosteel.org/%7E/media/Files/Autosteel/Programs/ULSAB-AVC/avc_ttd6.pdf
http://c315221.r21.cf1.rackcdn.com/FSV_OverviewReport_Phase2_FINAL_20110430.pdf
http://c315221.r21.cf1.rackcdn.com/FSV_OverviewReport_Phase2_FINAL_20110430.pdf
Welding and Joining of Advanced High-Strength Steels (AHSS). http://dx.doi.org/10.1016/B978-0-85709-436-0.00003-5
Copyright © 2015 Elsevier Ltd. All rights reserved.
Manufacturing of advanced 
high-strength steels (AHSS)
M.-C. Theyssier
ArcelorMittal R & D Center, Maizières les Metz, France
3.1 Introduction
For any advanced high-strength steels (AHSS) family,1 the automotive customer is 
awaiting a differentiation in a product’s properties. This can be summarized through 
the following simplified requirements:
 • Higher yield strength (YS) and tensile strength (TS) associated with a minimum required 
elongation
 • Adapted range of the YS-to-TS ratio
 • Better bending properties and/or an improved hole expansion ratio
 • Good weldability in homogeneous and heterogeneous configurations
These requirements are achieved thanks to a combination of phase constituents 
with an appropriate mix of soft and hard phases. This necessitates higher alloying 
contents compared with reference grades (such as classical interstitial-free [IF] steels 
or high-strength, low alloyed [HSLA] grades).
To successfully achieve the final required properties, the best or optimal combina-
tion of metallurgical phase constituents in the final product necessitates the optimal 
adaptation of alloying content and process settings: the so-called ‘metallurgical route’.
For steels with high strength levels (>780 MPa) in particular, this implies the appro-
priate adjustments of carbon (C) and manganese (Mn) and additions of all types of 
alloying and micro-alloying elements, including silicon (Si), chromium (Cr), molybde-
num (Mo), aluminium (Al), boron (B), vanadium (V), titanium (Ti) and niobium (Nb). 
The metallurgical concepts are obtained through the optimal balance of the different 
phases (ferrite, austenite, bainite, martensite) with suitable mixing and structures.
This higher alloying content not only has a positive impact on the final properties, 
as required by the automotive customer in the context of making vehicles lighter (car-
bon dioxide emissions control), it also has an impact on the production of the grades 
and generates technical challenges all along the processing route.
This chapter gives an overview of key challenges faced when producing AHSS 
grades during steelmaking, from liquid steel to coated strips. Internal soundness and 
1 The final delivery state of the considered AHSS product – hot rolled, cold rolled and annealed, cold rolled 
and annealed and electrogalavanized or cold rolled and galvanized states. A general remark about this 
chapter: even if several of the comments could apply to them, the chapter is not specifically adapted to very 
high amounts of manganese (e.g. twinning-induced plasticity steels) nor very high contents of aluminium 
(‘low density’ steels), which are at the first steps of their industrial history.
3
30 Welding and Joining of AHSS
quality, the possible occurrence of surface defects, phase transformation along the 
route and the hardness value of semi-finished products are described and linked with 
the consequences on production capability.
Globally, these challenges have been addressed by steelmaking plants for the past 
5–10 years but recently with higher intensity. Indeed, to enable customers to propose 
concepts for lighter vehicles without sacrificing passengers’ safety, the automotive 
industry promotes rapid development of ever-higher-strength steel. The return on 
experience regarding the associated production challenges is indeed being consoli-
dated by a few available publications.
A particular example is the dedicated first conferences on advanced steels (Yuqing, 
Dong, & Gan, 2011, Advanced Steels – Proceedings of the First International Confer-
ence on Advanced Steels; International symposium on new developments in advanced 
high strength sheet steels, AIST, Colorado, USA, June 23–27, 2013)
3.2 Key challenges faced in producing AHSS grades
3.2.1 Steelmaking
3.2.1.1 Liquid steel refining/analysis
In liquid steel production, the first related challenge is linked to the analysis of achieve-
ments to ensure the appropriate standard deviations of alloying element contents. 
Those standard deviations must be defined to allow a good range of mechanical prop-
erties as required by consumers of the final product (e.g. minimum and maximum YS).
For higher alloying contents (as is the case for AHSS grades), the choice of raw 
materials may be adapted and tighter control over this might be required. This can 
also imply an improvement of chemical mastery through improvements in analytical 
measurement sensors and the definition of reference samples.
Depending on the AHSS product being considered, the level of control over residual 
elements may have a significant impact on the optimization of the final product’s prop-
erties. This necessitates defining the suitable maximum limits of phosphorous, sulphur 
and nitrogen contents (ppm); sulphur and nitrogen contents are mainly linked to better 
control of inclusions such as manganese sulphide (MnS) and aluminium nitride (AlN).
Slag/metal reactions, as well as oxide and non-oxide inclusions or precipitates, can 
be mastered thanks to the well-known thermodynamic laws and conveniently using 
integrated software as the one developed by ArcelorMittal Global R&D: Ceqcsi.2 This 
software takes into account:
 • all phases, such as the slag, liquid and solid, of iron-based alloys covering most composi-
tions from cast iron, low-carbon steels, alloys and stainless steels;
 • oxide, sulphur, carbide and nitride phases (either stoechiometric or in a solid solution state, 
such as spinels and carbonitrides);
 • and gaseous phase.
2 Calculs d’EQuilibre Chimique pour la Sidérurgie (Chemical EQuilibrium Calculations for the Steel 
Industry).
31Manufacturing of AHSS
The history and latest developments of these models and software have been 
described by Lehmann et al. (2009).
Thus evaluating the nature and quantity of oxides and inclusions (in parts per mil-
lion) that are created during the elaboration step and the casting step is possible. As an 
illustration, the results of oxide calculations are shown in Figure 3.1.
In this particular example, if it seems that formation of calcium sulphide too early could 
be detrimental to the hole expansion property of the final product, the formation would 
need to be delayed. The only way to do this is to decrease the calcium addition and keep 
the same level of cleanliness (i.e. the small concentrations of total oxygen and sulphur).
Other examples provided in Figures3.2–3.4 show the large difference in the inclu-
sions content between a low-carbon steel and a higher alloyed steel (only the non- 
oxide inclusions are plotted here; precipitation is not computed until the product exits 
the casting machine). As shown in these figures, the nature and quantity of metallic 
inclusions may vary greatly with the AHSS grade.
Depending on the harmfulness of the inclusions along the production route and in 
the final product, the inclusions population, quantity and size may have to be opti-
mized. In AHSS grades the properties gap between soft and hard phases is a critical 
location for damage; this can be greatly enhanced by the occurrence of defects such as 
inclusions.
The hardness, morphology and position of the inclusions vary with each product 
application and may thus be detrimental to in-use properties by creating privileged 
locations for cracks (see the example in Figure 3.5).
Figure 3.1 Result of Ceqcsi software at the high temperature level, calculated precipitation 
of calcium sulphide and reaction with the oxides during metal cooling. CA2 = CaO–2Al2O3; 
CA = CaO–Al2O3; LO = liquid oxides.
Figure courtesy of Lehmann.
32 Welding and Joining of AHSS
Temperature (°C)
15
50 80
0
0.0
ALN
MNS
FCC
BCC
LIQUID
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.01200
1000
800
600
400
200
0
85
0
90
0
95
0
10
00
10
50
11
00
11
50
1500 1400 1300 1200 1100 1000 900 800
12
00
12
50
13
00
13
50
14
00
14
50
15
00
M
et
al
lic
 fr
ac
tio
n
C
on
te
nt
s 
(p
pm
)
Figure 3.2 Results of Ceqcsi software in the case of a low C steel grade containing 0.09% Al 
and 300 ppm S.
Figure courtesy of Lehmann.
Temperature (°C)1
55
0
80
0
0.0
BN
M2B_TETR
FCC
NBTICN
MNS
BCC
LIQUID
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.01200
1000
800
600
400
200
0
85
0
90
0
95
0
10
00
10
50
11
00
11
50
1500 1400 1300 1200 1100 1000 900 800
12
00
12
50
13
00
13
50
14
00
14
50
15
00
M
et
al
lic
 fr
ac
tio
n
C
on
te
nt
s 
(p
pm
)
Figure 3.3 Results of Ceqcsi software in the case of a HSLA-Nb steel grade containing 20 
ppm Nb, controlled S.
Figure courtesy of Lehmann.
Damage can be controlled, case by case, through steel chemical analysis adjustments, 
calcium treatment and/or electromagnetic swirl stirring.
As an example of chemical analysis adjustment, a ‘classical’ case is verifying the harm-
fulness of manganese sulphide inclusions when forming automotive parts. In the laboratory 
33Manufacturing of AHSS
15
50 80
0
0.0
NBTICN
TI4C2S2
M2B_TETR
FCC
ALN
MNS
BCC
LIQUID
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.01200
1000
800
600
400
200
0
85
0
90
0
95
0
10
00
10
50
11
00
11
50
1500 1400 1300 1200 1100 1000 900 800
12
00
12
50
13
00
13
50
14
00
14
50
15
00
M
et
al
lic
 fr
ac
tio
n
C
on
te
nt
s 
(p
pm
)
Temperature (°C)
Figure 3.4 Results of Ceqcsi software in the case of a TRIP-type steel grade with higher C, 
Mn, Si, Al contents; controlled S.
Figure courtesy of Lehmann.
Figure 3.5 Example of damage initiation (here beside titanium nitride (TiN) inclusion).
Image courtesy of A. Perlade.
this can be evaluated through tension, bending or hole expansion tests, with analysis of the 
cause of fracture. Such analysis may lead to the definition of maximum suitable sulphur 
content. This type of limitation is then discussed with the production units and can thus be 
introduced in the plant’s production control plans for the grades concerned. In particular, 
refer to Nadif et al. (2009) for details on desulphurization industrial practices.
34 Welding and Joining of AHSS
Calcium treatment can be applied during the secondary metallurgy step. It has 
three main objectives:
 • To improve the steel casting ability by avoiding nozzle clogging
 • To minimize slab surface defects associated with the presence of inclusions
 • To modify the sulphide morphology, as shown in Figure 3.6. This third objective has differ-
ent beneficial effects, among which is included better compatibility between the matrix and 
the inclusion during hot deformation.
Electromagnetic swirl stirring necessitates that an appropriate device be installed at 
the top level of the mould. The functioning principle of the device consists of pushing 
the inclusions away from the surface through electromagnetic agitation of melted steel. 
When parts are being formed, for example, by bending, the high strain near the surface 
of certain products can be facilitated by the absence of inclusions as potential loca-
tions of damage nucleation.
In the same way optimizing dephosphorization at high temperatures adapts to 
the detrimental consequences encountered along the production route and at the 
level of a product’s properties, which occur to a larger extent in higher alloyed 
grades. As an illustration, Maier and Faulkner (2003) studied the influence of 
Mn and phosphorous (phosphorus being ‘of great interest because it causes 
Al2O3
Al2O3
CaO
Al2O3
+ MnO
MnS
(Mn Ca) S
+ SiO2
Silico aluminates
Figure 3.6 Effect of calcium treatment on the control of inclusion morphology.
35Manufacturing of AHSS
embrittlement’) on the arc weld of C–Mn materials used in pressure vessels. They 
detailed the state of Mn and phosphorus segregations at the grain boundaries as 
dependent on the size of the microstructure. Solutions can be found by the cus-
tomer in part during the welding process. On the other hand, controlling phospho-
rus during the upstream production step through imposing limited phosphorous 
content (depending on the grade) is also a valuable technical option. This control 
necessitates dephosphorization pretreatments to pig iron and/or mastering slag 
metal engineering.
3.2.1.2 Continuous casting and slab yard
Internal soundness
During solidification along the casting process, inclusions and precipitates are formed 
(see Chapter 2), but micro- and macrosegregations patterns also appear. The composi-
tion of the slab is not homogeneous throughout its thickness.
The microsegregation pattern results from solute redistribution (C, Mn, Si, 
phosphorus, sulphur) during solidification at the level of the dendrites. This leads 
to a variation in the solute concentration between the centre and the outside of a 
dendrite arm. The distribution of the solute in the dendrites can be derived using 
mass balance equations taking into account diffusion of the solute in the liquid and 
solid phase. During solidification, significant concentration gradients occur in the 
solid at the solid–liquid interface for substitution elements (e.g. Mn, Si, Cr, nickel, 
Mo), whereas no concentration gradient is associated to interstitial elements (e.g. 
C, hydrogen, nitrogen), for which the diffusion coefficient is 10–10,000 times 
higher.
The Fourier number (αi) characterizes the diffusion length of the solute (i) in 
the solid during solidification compared with the microstructure scale. This classi-
cal parameter can be used to qualify the intensity of microsegregation (Bobadilla 
& Lesoult, 1997; Brody & Flemings, 1966; Clyne & Kurz, 1981):
 αi = (Dsi ts) /(λs/2)
2
 
where Dsi is the diffusion coefficient in the solid for the solute i; ts is the solidification 
time corresponding to the solidification range divided by the cooling rate; and λs is the 
spacing of the secondary dendrite arm.
Figure 3.7 summarizes the fact that the level of microsegregation for the solute (i) 
increases as both the partition coefficient of the solute (ratio of the solid and liquid 
compositions at the solid–liquid interface) and the diffusion coefficient in the solid 
decrease. It illustrates in particular the fact that the intensity of microsegregation is 
higher for substitutional solute elements than for interstitial solute elements. Moreover, 
because the diffusion coefficient of solute elements in the γ-phase (austenite) is lower 
than in the δ-phase (ferrite), alloys undergoing the solidification sequence (l → l + δ) 
exhibit less microsegregationthan alloys with γ occurring during the solidification 
phase.
36 Welding and Joining of AHSS
Depending on the nominal composition of the alloy, this might result in a more or 
less critical scheme of microsegregation pattern. As an example, Figure 3.8 shows that 
for a pseudo-binary iron–C–Mn diagram at 1.6% Mn, the higher the C content, the 
higher the γ occurring during solidification, and so the higher the microsegregation 
level.
Solidification is initiated from the skin in contact with the cold wall of the 
mould to the midthickness of the slab. At midthickness, the remaining liquid steel 
in the last moments of solidification is the richer alloyed part of the ‘in-formation’ 
slab.
The following are the main causes of solute-enriched liquid displacements in the 
mushy zone during continuous casting (Myazawa & Schwerdtfeger, 1981):
 • Deformations of the solid shell caused by compressive cooling or expansive reheating and 
by the discontinuity and imperfection of the mechanical support of the caster (bulging, roll 
position defect). This induces alternative variations of the mushy volume and sucking or 
ejection flows of the interdendritic segregated liquid.
 • Deformations of the solid skeleton in the mushy zone caused by solidification shrinkage 
during cooling. This contraction of the solid leaves free volume, which attracts some 
solute-enriched liquid.
Once the solidification of the whole thickness is completed, the scale of the chem-
ical heterogeneity at the slab’s ‘midplane’ is 10 or 100 times higher than that of the 
dendritic structure. This chemical heterogeneity is called ‘macrosegregation’.
Throughout the process after casting, segregations are at the origin of the alternate 
layers of hard and soft phase through the product’s thickness, the so-called banded 
structure. The morphology of the bands, present at the hot coil (after hot rolling) and 
cold coil levels (after hot rolling, cold rolling and annealing or coating), may have an 
influence on the final bending or hole expansion, that is, on the forming properties of 
the hot-rolled or cold-rolled products.
Figure 3.7 Information about the microsegregation behaviour of various solute elements 
depending on the coefficient partition and solid diffusion.
Figure courtesy of Bobadilla (1999).
37Manufacturing of AHSS
δ δ δ δ
δ δ
γ γ γγ γ γ γ γ
γ γ
1540
1520
1480
T 
(°
C
)
1440
1400
0.1
Soft, extra soft
Semi-hard
Medium carbon
0.2 0.3 0.4
Carbon (%)
(a)
(a)
(b)
(b)
(c)
(c)
L
L+δ
γ
L+δ+γ
L→L+δ
L→L+δ→L+δ+γ
L→L+δ →L+δ+γ
→δ→δ+γ
→ δ+γ→γ
→ L+γ→ γ L→L+γ →γ
→γ
δ+γ
δ L+γ
(d)
(d)
Hard
Figure 3.8 Pseudo-binary iron–carbon–manganese diagram at 1.6% Mn. The solidification 
sequence depends on the carbon content in the alloy as is illustrated in the different schemes 
for soft and extra soft grades (a), medium carbon (b), semi-hard steel (c) and hard steel (d). 
(δ indicates “ferrite phase” and γ indicates “austenite phase”).
Figure courtesy of Bobadilla (1999).
38 Welding and Joining of AHSS
Macrosegregation and microsegregation patterns strongly depend on the chemical 
composition of the alloy. But some process specificities and parameters also have an 
impact:
 • Soft reduction devices (when available at the casting machine level) allow controlled com-
pression of solid skins to be applied by roll pinching. This results in a beneficial ejection 
of the interdendritic segregated liquid, which has a direct impact on the macrosegregation 
pattern, lowering it (Figure 3.9).
 • The type of caster (especially when considering a classical >200-mm-thick slab or a thinner 
slab caster for <140-mm-thick slabs) also has an impact.
 • The main casting parameters (mainly influencing microsegregation) are all the process 
parameters that can have an impact on heat extraction at the skin level and, consequently, on 
the solidification rate and thermal gradient; this is the case for the casting speed, the super 
heat (difference of temperature between the liquid steel in the mould and the liquidus tem-
perature) and the secondary cooling (Segunpta et al., 2011).
The resulting banded structure (alternated bands of different phases) is also affected 
by thermal control at the hot rolling stage. The most influential parameter is the coiling 
temperature, which has a direct impact on the Carbone diffusion and thus, in the end, 
on the repartition of the phase in the thickness of the coil.
After cold rolling, during the annealing step, well-known grain nucleation and growth 
mechanisms produce a topology with more or less band heredity. This is strongly depen-
dent on the heating speed and on the soaking temperature, whether it is in the austen-
ite–ferrite or fully austenitic domain. This is, of course, not without consequences on 
Figure 3.9 Fraction areas of Mn and P segregations, according to the type of the support 
(bar block, one-piece roll or divided roll) and soft reduction (SR) optimization: if the block 
bars (walking bars which carry out a soft reduction in a quasi continuous way) give a result 
close to the perfection, the soft reduction with the divided rollers represents a very significant 
improvement compared to the quality obtained without any soft reduction actuator.
Image courtesy of J.M. Jolivet.
39Manufacturing of AHSS
the final product properties. As an example, if the annealing conditions of a particular 
grade are such that they promote the heredity of the banded structure through the whole 
thickness of the strip, the risk of damage being initiated during product bending may be 
enhanced at the level of the hard bands that are close to the surface.
Slab surface defects
To guarantee good surface quality of continuous cast steels, avoiding carbon concen-
trations that are in a critical peritectic range, at which the risk of longitudinal cracks 
increases, is desirable. Two models are available to predict the critical peritectic range 
of a steel grade:
 • The carbon equivalent (Ceq) criterion is the classical approach: the effect of the alloying 
elements is taken into account to calculate the effective carbon composition for a particular 
grade. Various values of coefficients for the different elements are found in the literature 
(Wolf, 1991). These coefficients are obtained from thermal analysis, thermodynamics calcu-
lations or plant observations. If the calculated carbon equivalent is in the critical peritectic 
range (0.08–0.15 wt%), the grade is assumed to be sensitive to the formation of longitudinal 
cracks.
 • The peritectic predictor equations developed by Blazek, Lanzi, Gano, and Kellogg (2007) 
were obtained by calculations using Thermocalc (version M with the TCFE3 database) and 
were tuned to match experimental data. The investigated compositions span a large range of 
alloying elements, including high manganese, aluminium and silicon contents. Non-linear 
regression equations for the lower and upper limits of the critical peritectic range have been 
proposed. Based on these equations, if the carbon content of the grade is in the critical range, 
the grade is assumed to be peritectic.
Both approaches have been compared and similar results obtained for the grade 
Fe–0.9wt%Mn–0.3wt%Si. Nevertheless, the carbon equivalent formula was not 
adapted for high-alloyed steels and some improvements were needed to take into 
account the effect of aluminium.
The carbon equivalent formula has been updated to span a larger range of alloying 
elements (Bobadilla team); see the Ceq formulae (*) below. The available thermo-
dynamic databases were critically analysed to select the most coherent database to 
determine the coefficient. The predictions made with the peritectic predictor equations 
with the Ceq criteria or the new formulation are similar.
Ceq defined as (*) Ceq = [C] + 0.0146 [Mn]– 0.0027 [Si] – 0.0385 [Al]2 
 – 0.0568 [Al] – 0.064 [Mo] + 0.0021 [Cr] + 0.02 [Ni] – 0.006 [W] – 0.012 [V] 
 + 0.8297 [S] + 0.0136 [Mn][Si] – 0.0104 [Si][Al] + 0.0026 [Si][Al]2+ 0.0134 [Mn][Al] + 0.0031[Mn][Al]2.
Courtesy of M. Bobadilla
When 0.07 < Ceq < 0.15, the risk of longitudinal cracks occurrence is enhanced.
This work is a good illustration of the necessary continuous adaptation of knowledge 
to newly developed products for which the previously established formulae sometimes 
reach their limits of validity.
40 Welding and Joining of AHSS
The toughness property of solid matter at the slab level might also be insufficient to 
ensure a bending–unbending step, free of transverse cracks. Indeed, for high alloying 
contents, the ductility might be too low to ensure a deformation free of defects all 
along the casting (temperature and deformation speed) range.
As shown by Tuling et al. (2011), the level of ductility depends on several factors:
 • Solidified austenitic grain size (bigger grains are more prone to damage).
 • Phase transformation: among the different products, the peritectic range is still a risky zone. 
But equally, for a given grade, the ferrite phase quantity and structure may also play a detri-
mental role. Indeed, at the slab bending–unbending stage, the quantity of ferrite that already 
precipitated – and, more precisely, its structure – are possible weak areas during slab defor-
mation where cracks could initiate. This is particularly the case if, at that level of the caster, 
the ferrite phase forms a film all along the grain boundaries.
 • An AlN inclusion or other precipitates characteristics.
In general, phase type transformation at cooling (with ferrite growth at the grain 
boundaries): type, location and size of precipitates and austenitic grain size play a 
crucial role on the risk of transverse cracks occurring.
For a given grade on a given casting machine, adjusting the secondary cooling 
helps avoid defects; it consists in adapting the cooling flow of the slab along the cast-
ing machine to obtain the suitable temperature at the bending–unbending level for the 
given grade. The principle is to avoid the most fragile structure at that stage of the 
caster. In severe cases, slab surface defects require a repairing step through grinding 
or scarfing.
After continuous casting, all the different production steps, including the slab cool-
ing to atmosphere temperature, storage and transportation in the slab yard and reheat-
ing in the furnace (which is located at the hot rolling entry) are made difficult in the 
case of fragile slabs.
As far as AHSS products are concerned, preexisting defects, grain size and tough-
ness may not be favourable for those intermediate steps. When the risk of slab breaks 
is evaluated to be high, some cautious production practices are used, for example, 
slab cooling in stacks and hot charging. The principle of those practices consists of 
avoiding transporting the slab at low temperatures (<250 °C) between exit from the 
continuous casting machine and entry into hot rolling. Indeed, the toughness of the 
matter is lower at those temperatures and the slabs are thus more fragile.
3.2.2 Hot rolling
3.2.2.1 Reheating furnace
Slab reheating remains of primary importance for controlling the amount of microal-
loying elements taken into solution and for controlling starting grain size.
Titanium nitride (and, to a lesser extent, niobium carbonitrides and AlN precipi-
tates) is the most stable compound with poor dissolution in the usual range of reheating 
temperatures. Remaining precipitates (including fine carbides) participate to the control 
of grain size in the subsequent stages of production.
41Manufacturing of AHSS
In the reheating furnace the scale layer, which is formed along the temperature–
time path, is not composed of a simple single layer. Once again, depending on the 
grade composition and alloying content, a wide range of oxide chemistries and asso-
ciated morphologies can be formed – simple wüstite, magnetite, hematite, fayalite, 
silice – in local or continuous layer configurations (Alaoui Mouayd et al., 2011) (see 
the example in Figure 3.10).
Some elements that have a lower melting temperature, such as copper, are prone 
to melting along the grain boundaries at the metal–scale interface, possibly inducing 
specific defects at the coil surface in the subsequent hot rolling process.
When oxides remain at grain boundaries, or depending on the scale mechanical 
properties, different surface defects may form, from different types of slivers to sur-
face diverse heterogeneous morphologies. Those heterogeneities of surface morphol-
ogy can induce dispersions of the product emissivity, which may result in a poor 
thermal control in the last stages of the hot rolling process (during cooling on the run 
out table cooling and during coiling). As a consequence, to avoid defects on the hot 
coil surface, the thermal path along the hot strip mill as well as the primary and the 
secondary descalings must be optimized.
In the roughing and finishing mills stands’ roll gaps, the scale intervenes as an 
intermediate body between the rolls and the metal being hot rolled. The chemical 
nature of the scale and the thickness of the layer can lead to different friction coef-
ficients and possibly localized defects on the roll’s surface. Once again, this directly 
impacts product and process achievements.
NAS: NAS+1.6%Si:
2 sub-layers:
-Internal (2): Thin (#14 µm),
Si 6%At, O 47%At => Fe–Si oxide
-External (3) : Thin (#13 µm),
Si 0.9%At, O 46%At => FeO
=> Similar to SAIS!
Mono-layer(1): Thin (#7 µm)
O 50%At => FeO (wustite)
=> Representative of NAIS
..
Figure 3.10 Scanning electron microscopic scale images and their chemical composition 
through X-ray diffraction showing oxidation of pilot samples. NAS, nonalloyed steel.
Figure courtesy of Alaoui Mouayd.
42 Welding and Joining of AHSS
3.2.2.2 Hot rolling
Depending on the grade composition (alloying and microalloying contents) and on the 
hot rolling schedule, the strain hardening at each pass of the roughing and finishing 
mills and the dynamic recrystallization ability between the passes vary.
The temperature of no-recrystallization (Tnr) can differ from grade to grade 
and depends on the hot rolling schedule. This Tnr value determines the state of 
recrystallization of the hot coil matter when it exits the hot rolling mill. The mate-
rial at this stage is partially or fully recrystallized, depending on whether the 
temperature when exiting the finishing mill is higher or lower compared with the 
Tnr value.
All this explains the interest of optimizing the rolling schedule for the given grade. 
Of course, this must be done in coherence with all other production constraints.
In its optimized version, the microstructural design of the hot-rolled product, 
including austenite grain morphology and ferrite precipitation control when exiting 
the mill, is better known as thermomechanical processing (TMP). It consists of 
modifying the thermomechanical treatment for a given composition of steel to achieve 
the required grain size and phase distribution, both of which control the properties of 
the steel. In the case of AHSS grades, TMP in the reheating furnace to finishing mill 
area is fairly advanced but still under development (Perlade et al., 2008).
As far as hot strain hardening is concerned, the tendency is that the higher the 
alloying content (up to approximately 5–7% total in AHSS grades), the higher the hot 
deformation resistance through the influence of the Si, carbon, manganese, Nb, Ti. 
The consequence is higher rolling forces for the same hot rolling reduction rate com-
pared with that of HSLA grades.
Figure 3.11 summarizes the maximum stress as measured with the hot axial com-
pression test for different temperatures and strain rates and several different AHSS 
grades.
In the context of reducing carbon dioxide emissions, the usage of AHSS grades in 
car bodies allows vehicles to be made lighter. This can be obtained by decreasing the 
strips’ thickness, while keeping tight control of the strips’ width and flatness.To achieve dimensional requirements at hot and cold coil levels, the hot rolling 
schedule of AHSS grades must be adapted to the higher hardness of the product, 
taking into account the specific expectations for delivering low thicknesses and the 
technical characteristics of the mills. Optimizing the hot rolling schedule concerns the 
main parameters of the schedule setting: temperatures, thickness reductions per pass, 
rolling speed and stand roll profiles. In this task one also has to consider the friction 
coefficient, which may vary with the product (see Section 3.2.2.1) and also depends 
on the mill settings (e.g. lubrication of the rolls).
Optimizing the grade chemistry to lower the product hardness at hot rolling is a 
possible way to help achieve the coils’ dimensional targets. An example of such a tun-
ing attempt is described by Mostert et al. (2013), who showed that a DP600 grade with 
reduced Si content allows the dimensional feasibility to be enlarged. This is explained 
in that paper by the composition dependence of the mean flow stress as obtained from 
industrial trials. The contribution from Si is only approximately 16% less than the Mn 
contribution.
43Manufacturing of AHSS
When exiting the hot strip mill, control of the product properties through TMP in 
the run out table to coiling area is already fairly advanced thanks to the development 
of several thermodynamic and physics/metallurgy-based models, like those described 
by Perlade et al. (2005) and Pethe et al. (2011).
Those models’ hearth calculates the repartition of the phase and the precipitation 
strengthening of the steel at the hot-rolled coil level (case of carbon–Mn and carbon–
Mn–Nb grades). In particular, the heat evolution linked to the phase transformation 
during cooling is taken into account.
3.2.2.3 Coiling and coil yard
Linked to phase transformation when exiting the hot rolling mill along the run out 
table, the coiling step and at the coil yard level, the choice of finishing mill tem-
perature, coiling temperature and temperature–time paths have a major influence on 
achieving product property targets and the homogeneity of the product properties 
throughout the whole coil body. This is, of course, particularly true in the case of hot-
rolled AHSS products for which the final microstructure is directly produced in the 
hot rolling mill.
For each of the metallurgical concepts, the typical phase transformation curves 
(continuous cooling transformation curves) are specific; ferrite, perlite, bainite and 
martensite domains are more or less extended and translated along the temperature 
and transformation axis. As examples, elements like carbon, Mn, Cr, Mo and boron 
delay the ferrite and perlite transformations; carbon, Mn and Cr delay the bainite 
transformation. On the other hand, Si and Al expand the ferrite domain to the left and 
increase the Ac1 and Ac3 temperatures.
400
350
300
250
200
150
100
50
0
800 850 900 950 1000 1050 1100 1150 1200 1250 1300
0.1 s–1
1 s–1
10 s–1
50 s–1
Higher alloying content within:
C < 0.25%
Mn < 2.5%
Si < 1.8%
Cr < 0.4%
Al < 1.4%
Others: B, Nb, Ti...
St
re
ss
 (M
Pa
)
Temperature (°C)
Figure 3.11 Maximum stress (MPa) obtained after hot compression tests for different 
advanced high-strength steel grades.
44 Welding and Joining of AHSS
In general, thermal heterogeneities encountered in the run out table, during coiling 
and at the coil yard level, may be at the origin of possible phase transformations in 
localized areas and, as a consequence, of property and hardness heterogeneities. These 
heterogeneities necessitate particular attention to avoid ‘out-of-tolerance’ dimensions 
in some parts of the coils, especially out-of-tolerance thickness variations at the cold 
coil level.
Poliak et al. (2009) illustrate an example of this production issue that is a direct 
consequence of the head and tail overhardness at the AHSS hot coils extremities – the 
so-called gauge hash effect. During coil cooling, some parts of the coils are cooled at 
higher rates than others, especially in the outer rings. This creates a pseudo-periodic 
overhardness localized at the coil extremities, which finally results in thickness irregu-
larities after cold rolling. In severe cases this necessitates adaptations to the hot rolling 
schedule.
3.2.3 Pickling
Most pickling lines are continuous, meaning that coils are welded head to tail, 
one to another to allow a continuous production flow. Flash butt welding at entry 
to the pickling line is a solid-state welding process that is followed by a weld 
grinding step to remove the flash-formed weld. AHSS welding through flash butt 
requires adaptations to welding process parameters (welding speed, power) to 
ensure a robust and safe enough weld that is able to pass through the entire line, 
whether the production line is limited to the pickling baths or comprises a contin-
uous pickling and cold rolling process. Laser welding is also sometimes used and 
may offer higher production flexibility (Wallmeyer, 2013). In particular, welding 
with various heterogeneous configurations (implying an AHSS grade on one side 
and a different accompanying steel grade on the other side of the weld joint) can 
be facilitated.
Ichiyama and Kodama (2007) presented an innovative proposition involving high 
currents during the flash butt welding process. This specific practice goal is to extrude 
internal oxides and inclusions, since those defects are the main possible causes of 
embrittlement of flash butt welds.
Because of the wider variety of scales coming from the hot rolling process (see 
Section 3.2.2), management of pickling baths sometimes needs to be adapted. When 
necessary, this means changing the acid concentration, changing the pickling inhibi-
tor, tuning the temperature of the baths and adapting the line speed. In any case, the 
solutions implemented to improve the product quality are the result of the best tech-
nical compromise between the production flow constraints and the options for better 
management of the acid baths.
In difficult cases, for example, when oxides penetrate the grain boundaries 
or when the intermediate layer oxides change the pickling kinetics, acids stron-
ger than hydrogen chloride or sulphuric acid or a mix of specific acids might be 
considered.
45Manufacturing of AHSS
3.2.4 Cold rolling
The main challenge of cold rolling AHSS steels consists of achieving the required 
dimensions (width × thickness × flatness) for steels that exhibit higher cold strain 
hardening and, in some cases, higher risks of breakage at higher reduction rates. 
As for hot rolling (Section 3.2.2), market evolution makes the rolling of harder grades 
necessary, with the goal of achieving thinner strips.
In Figure 3.12 one can see the evolution of stress up to a 90% reduction rate in a cold 
compression test machine (plane strain deformation configuration) in a laboratory.
The increase in the hardness of higher alloyed grades during cold rolling requires 
the optimization of cold rolling schedules and especially of the reduction rates at each 
stand, using the full benefits of the mill’s total capacity.
Another difficulty is linked to the possible occurrence of defects, either at the prod-
uct surface, bulk or strip edges (heterogeneities are inherited from upstream produc-
tion steps or result from difficult edge trimming). This increases the risk of strip breaks 
during cold rolling.
Indeed, after high cold reduction rates, the high internal stresses resulting from 
cold strengthening (Figure 3.12) are accompanied by a reduced total elongation of 
the material being cold rolled. This favours the initiation of cracks at critical-sized 
defects. In those conditions a situation of high tension at the strip edges during the 
cold rolling process enhances the risk of cracks opening in the transverse direction.
Figure 3.12 Stress on different types of AHSS grades after cold plane strain compression 
tests in alaboratory. The equivalent strain of 0.8 is obtained for a reduction rate of 50%, and 
the equivalent strain of 1.6 is obtained for a reduction rate of 75%.
46 Welding and Joining of AHSS
3.2.5 Annealing and coating
3.2.5.1 On-line welding
Seam welding is the most expanded technique for on-line welding as a continu-
ous process to ensure coil-to-coil assembly at the entry of continuous annealing 
lines and galvanizing lines. This is a resistance welding technique that requires an 
overlap of the two extremities of the coils so one can be joined to another. Seams 
are copper-based, round electrodes that ensure current conduction and the Joule 
heating effect.
The general trend is that when spot welding is difficult, the same is true for 
on-line seam welding. The processing of hard AHSS often requires adapting param-
eters, such as the level of pressure that is applied between the seams, the welding 
speed, the temperature, the use of a device to post heat the weld and a better and 
tighter in-line control. As in the case of flash butt welding (Section 3.2.3) and for the 
same reasons, laser welding at this stage of production is considered as a relevant 
technical alternative.
3.2.5.2 Thermal issues and challenges
As for any other steel grade, controlling the temperature of the product during anneal-
ing is of prime importance to achieve the targeted properties because they result 
from the appropriate volume fraction of the phase and recrystallization kinetics. 
This implies the ability to properly measure the strip temperature along its path in the 
continuous annealing line in the different furnace zones: heating, soaking and cooling.
In the case of AHSS grades, as explained in previous chapters, there may 
remain some surface heterogeneity linked to specific roughness and surface topog-
raphy after the oxide layer is removed through the pickling process. This may 
be at the origin of dispersions of the product’s surface emissivity, which itself 
results in heating heterogeneities. That is why appropriate sensors for measuring 
temperature are necessary: the use of wedge pyrometers that take advantage of 
the dark body principle, among others, can be of great help in overcoming this 
difficulty.
The particular case of water quenching, which makes specific flatness and surface 
difficulties arise (as in full martensitic grades (Hurtig et al., 2008)), necessitates adapted 
control and modelling. In an analogous manner, ‘new’ developments in cooling paths 
implying quench and partitioning (Li et al., 2013) open the door for necessarily 
overcoming various product and process challenges.
3.2.5.3 Oxidation and coating issues
The higher the alloying content of the annealed grade (especially regarding Mn, Si and 
Al contents), the greater the risk of external selective oxidation at continuous anneal-
ing lines or galvanizing lines.
At continuous annealing lines, such a surface defect can result in remaining oxides 
or colouration issues. This high sensitivity of the surface to the furnace atmosphere 
47Manufacturing of AHSS
and cooling fluid (e.g. water for quenched products) makes adapting the surface 
engineering of those products necessary. Particular care when in contact with gas in 
the furnace and with liquids (degreasing baths, pickling baths, temper mill lubricants 
and the rinsing section) is required.
At galvanizing lines, a report by Staudte (Staudte, Mataigne, Loison, & Del Frate, 
2011) summarizes the main difficulties associated with the surface oxidation prob-
lem that affects product coatability, namely, coverage and adherence of the zinc 
coating. This article recommends furnace atmosphere dew point control to improve 
wettability and adherence in the case of transformation-induced plasticity grades 
(Mn–Al or Mn–Si). The coating adheres well at dew points in the range of −20 °C 
to +10 °C. Dew points of −20 °C and higher significantly improve coatability by 
ensuring Al and Si diffusion into the bulk of the material. Because the manganese 
oxide at the surface is not a continuous layer, the enrichment of Mn at the surface 
does not seem to be detrimental for hot-dip coatability in case of this Mn–Al 
grade.
Even higher dew points must be applied for Mn–Si grades compared with Mn–Al 
grades, allowing the transition from external to internal selective oxidation. See also 
Mataigne (2001) on that particular topic. With high dew point annealing, excellent 
hot-dip coatability can be obtained. The benefit of this technology has even been con-
firmed industrially.
Apparently, coating quality of the DP grade with higher Mn (2–3% Mn, low Al and Si) 
could not be improved by increasing the oxidation potential via increasing the dew point 
during annealing. Any increase in the oxidation power leads to thickening of the exter-
nal Mn-rich oxide (iron participation in surface oxidation). This results in poor wettability. 
In general, the galvanizing difficulties associated with high Mn, Si and Al contents are 
being studied.
Another example (Blumenau et al., 2012) shows that preoxidation can ensure coat-
ing success when using direct flame furnace galvanizing lines. The conditions of this 
preoxidation are strongly dependent on the alloying content and strip dimensions. 
Preoxidation method is even claimed to be useful in the case of very high Mn contents 
(products containing 20–25% Mn, such as Twin Induced Plasticity (TWIP) steels).
Norden (2013) also described the interest of reducing preoxidation of radiant tube 
furnace galvanizing lines in case of a iron–Mn–Al transformation-induced plasticity 
steel.
3.2.6 Robustness along the route
As for any other product, the inheritance of defects along the industrial route is a 
possible issue whether at the coil surface, at the coil edges or in the material bulk. 
In high-strength steels, however, this ‘whole route’ question is more acute in the 
sense that higher risks of defects and heterogeneities occur from the very first steps 
of production:
 • surface or internal defects coming from brittleness or poor internal soundness of the slab;
 • specific microstructures such as banded structures;
48 Welding and Joining of AHSS
 • high macrosegregation in the midthickness; and
 • imperfections of the trimmed edge.
In some cases this makes not only study of the origin of the defect but also its evolution 
along the route necessary to be able to find the appropriate preventive or curative solu-
tion. This comment is particularly true for the question of varying mechanical and in-use 
properties along coils or between coils. This is an important issue because it directly 
relates to customers’ final requirements and satisfaction. As an example, overhardness 
that is sometimes obtained at the head and tail of the coils and at the coil edges may be 
related to an uncontrolled cooling rate of the outer parts of the coils along the hot process.
Another example is the achievement the flatness targets of the delivered coils, 
which results in mastering this parameter along the route, through the rolling mills 
and the annealing and cooling steps. Hardness of the product during rolling, phase 
heterogeneities along the route and buckling risk at the quench step of the continuous 
annealing line make optimizing the usage of available actuators and adapting repairs 
at the temper mill and leveller tools necessary.
As far as process risks along the route are concerned, one important topic is the risk 
of hydrogen (H) pickup, which can result in delayed fracture in the final stamped part. 
This so-called delayed fracture risk is an unexpected cold cracking that occurs on a 
stamped part after several hours or days. The condition necessary for the delayed frac-
ture phenomenon to occur include a critical combination of applied or residual stresses, 
metallurgical factors such as crystallographic structures and the presence of diffusible H.
High-strength steels are all the more sensitive to delayed fracture whenthe strength 
is high; indeed, high strengths often are associated with high applied residual stresses. 
Combined with this factor, the higher the diffusible H content (resulting from a 
balance between H pickups along the industrial route and H diffusion in the steel 
bulk), the higher the risk.
For AHSS grades that combine a large variety of phases – ferrite, bainite, martensite – 
allowing rather good diffusion of H in the bulk of the material, the major process risks 
associated with H are located at the end-of-route coating lines and, more specifically, 
at electrogalavanizing lines.
Dedicated laboratory trials making possible a comparison of the behaviour 
of different high-strength grades have been developed (e.g. Lovicu, Bottazzi 
et al., 2012). Those testing procedures allow a controlled strain or stress to be 
applied to samples that may have been previously electrochemically hydrogenated. 
In parallel, other laboratory efforts concentrate on the development of accurate mea-
surements of diffusible H, such as thermal desorption analysis (Georges, 2009). Those 
studies allow the risk for each of the considered high-strength steels to be evaluated and 
the diffusible H content under which this risk is eliminated to be defined. Achieving the 
required H content in the final product was also a main topic of those research efforts.
3.2.7 Elements of manufacturing issues from the customer’s 
perspective
Karbasian and Tekkayya (2010) give an overview of the technical challenge that hot 
stamping represents, especially in controlling phase transformation at the cooling 
49Manufacturing of AHSS
stage, the details of which are not described in this chapter. As far as the cold stamp-
ing process is concerned, some particular difficulties arise in the case of high-strength 
steels.
Because of their low anisotropy coefficient (r value around 1), high-strength steels 
are not particularly good material in ‘shrink drawing’ deformation mode, which is a 
quite common deformation mode during stamping. Deformation of the corners is not 
easy and may result in rupture or risk of wrinkling. In the same way, depending on the 
strain hardening exponent ‘n’, the ‘stretching’ as deformation mode might be more or 
less difficult: the lower the n coefficient (for high Ys/Ts values); the more difficult the 
deformation in stretching mode is.
For most AHSS grades, optimum bending properties and hole expansion minimum 
values are requested by the customer. These specifications depend on the grade and the 
automotive parts that are targeted.
One of the most severe difficulties encountered during AHSS forming remains 
the spring-back effect, which is clearly higher for higher stress levels (for equivalent 
Young modulus values). In this context hot stamping presents the largest factor of 
interest, among others, to more or less annihilate this drawback.
The behaviour of high-strength steels during forming is still not fully understood. 
Structural parameters explaining the bending ability of high grades are being studied 
(Sadagopan and Urban, 2003).
Another large domain of investigations regarding AHSS formability is again linked 
to the risk of delayed fracture associated with H embrittlement. In general, the higher 
the mechanical resistance of the product, the higher the risk. This can be particularly 
enhanced at locations where wrinkles appear. For example, Carlsson (2005) recom-
mends forming process parameters to avoid wrinkling as much as possible and, in this 
way, lower the delayed fracture risk.
Higher strengths of stamped products result in rapid wear of the press tools (espe-
cially at the level of the small radii). This can lead to the necessity of applying 
hardening techniques either through local thermal treatment or specific additions 
at the surface involving hard coatings such as Cr or Ti carbides. Moreover, with 
harder grades to be formed, the total required power for the press is obviously higher 
(if cold stamping). This partly explains the reasons for the current success of hot 
stamping.
3.2.8 Elements of costs and economics
The high volatility of raw materials is a current economic trend that cannot but 
be taken into account in AHSS product development globally. As an illustration, 
Table 3.1 gives an idea of the evolution of the cost of some of the main ferro alloys 
used in AHSS compositions.
This makes it necessary, in a context of high worldwide industrial competition, the 
special attention to the most appropriate alloy mix to obtain an optimal set of product 
properties. In parallel, it is required to optimize the choice of ferro alloys in order 
to obtain the best possible compromise between cost and required analytical/product 
quality.
50
W
elding and Joining of A
H
SS
Table 3.1 Evolution of ferro alloy costs, showing rapid change over time
Alloy
Cost ($/T ferro) by year
Source2000 2002 2004 2006 2008 2010 2011 2012
HC FeMn 423 483 1274 737 2662 1449 1379 1164 Metal Bulletin
Mn metal — — 1617 1389 3730 2942 — — AM purchasing
Sdt SiMn 469 492 1272 749 2222 1445 1313 1217 Metal Bulletin
Sdt FeSi75 535 557 921 870 2006 1761 1846 1454 Metal Bulletin
HC FeCr 875 688 1586 1376 5082 2717 2708 2387 Metal Bulletin
MC FeCr 1388 1350 2161 2350 9361 4483 5002 4649 Metal Bulletin
FeMo 7021 9920 44,269 59,054 69,370 40,139 38,322 31,414 Metal Bulletin
FeV 9790 7721 27,206 38,454 61,182 30,062 28,742 24,976 Metal Bulletin
FeTi70 3675 3962 10,548 16,371 7577 6763 8349 7395 Metal Bulletin
FeNb 9208 9220 8750 9459 23,249 22,767 26,377 24,309 COMEXT
Ni cash 8638 6772 13,823 24,244 21,104 21,804 22,890 17,533 LME
Cu cash 1813 1559 2865 6721 6955 7534 8821 7949 LME
51Manufacturing of AHSS
3.3 Future trends
Requirements for ever safer and lighter vehicles will guide the trends of the steel 
market for the design of future vehicles. That is the reason why new generations of 
AHSS grades are already being studied and even starting development in the market. 
Those new generation steels will include higher alloying contents (from 8 to 10%, 
up to 20–30%) and result in interesting properties such as lower density and a better 
combination of strength and formability, with improved performance regarding crashes 
and the ability to make vehicles lighter. Manufacturing challenges will, of course, be 
specific to those new grades, whatever their final process: hot or cold stamping or roll 
forming.
Even though we know the automotive sector strongly promotes competition 
between materials (e.g. Al, plastic, carbon fibres, reinforced composites), the vari-
ety of production tools and the possibility for adaptations are high in the steel 
industry; there is thus great scope for the provision of new recyclable materials for 
the automotive industry. In parallel and in addition, press tools used by customers 
follow the general trends for high productivity (wide progressive press; develop-
ments in hot press stamping). This will also provide greater potential for new steel 
product offers.
References
Alaoui Mouayd, A., Sutter, E., Tribollet, B., & Koltsov, A. (2011). Pickling and over-pickling 
mechanisms of high alloyed steel grades. Eurocorr 2011, Stockholm.
Blazek, K. E., Lanzi, O., Gano, P. L., & Kellogg, D. L. (2007). Calculation of the peritectic 
range for steel alloys. In AIST 2007.
Blumenau, M., Gusek, Ch.O., Norden, M., & Schönenberg, R. (2012). Industrial use of pre- 
oxidation during continuous hot dip coating of high alloyed steels.
Bobadilla, M., & Lesoult, G. (1997). Chapitre 4: La coulée et la solidification des aciers. Les 
aciers Spéciaux. Technique & Documentation. ISBN: 2-7430-0222-0.
Brody, H. D., & Flemings, M. C. (1966). Solute redistribution in dendritic solidification. Trans 
AIME, 236, 615.
Carlsson, B. (2005). Effect of wrinkling on delayed fracture in deep drawing. In Proceedings of 
the 24th international deep-drawing research group congress IDDRG’2005, June 20–22, 
2005. France: Besançon.
Clyne, T. W., & Kurz, W. (1981). Solute redistributionduring solidification with rapid solute 
diffusion. Metallurgical Transactions A, 12 A, 965.
Georges, C. (2009). Determination of diffusible hydrogen content in coated high strength steels. 
In B. Somerday, P. Sofronis, & R. Jones (Eds.), Effects of hydrogen on materials, pro-
ceedings of the 2008 international hydrogen conference (pp. 493–500). Copyright © 2009 
ASM International®.
Hurtig, P., Nilsoson, T., & Larsson, J. (2008). Development of cold rolled martensitic electro-
galvanized steels. In International conference on new developments in advanced high-
strength sheet steels, June 15–18, 2008, Orlando, Florida, USA (pp. 137–146).
Ichiyama, Y., & Kodama, S. (2007). Flash butt welding of high strength steels. In ‘Nippon steel 
technical report’, No. 95 (pp. 81–87).
52 Welding and Joining of AHSS
Karbasian, H., & Tekkayya, A. E. (2010). A review on hot stamping. Journal of Materials pro-
cessing technology, 210, 2103–2118.
Lehman, J., Bontems, N., Simonnet, M., & Gardin, P. (2009). New developments on slag model-
ling at arcelormittal Maizières. In Proceedings of the VIII international conference on molten 
slags, fluxes and salts.
Li, W., Yong, Z., Weijun, F., Xinyang, J., & Speer, J. (2013). Industrial application of Q&P 
sheet steels. In Proceedings of the international symposium on new developments in 
advanced high strength sheet steels, June 23–27, 2013. USA Colorado: AIST.
Lovicu, G., Bottazzi, M., D’Aiuto, F., De Sanctis, M., Dimatteo, A., Santus, C., et al. (November 
2012). Hydrogen embrittlement of automotive advanced high-strength steels. The Minerals, 
Metals & Materials Society and ASM International, 43A, 4075–4087.
Maier, P., & Faulkner, R. G. (2003). Effects of thermal history and microstructure on 
 phosphorus and manganese segregation at grain boundaries in C–Mn welds. Materials 
Characterization, 51, 49–62.
Mataigne, J. M. (2001). In Proc. 8th Proc. Int. Conf. On Zn and Zn Alloy coated sheet steels, 
Galvatech’11, Genova, AIM, June 21–24.
Mostert, R., Ennis, B. L., & Hanlon, D. N. (2013). Adapting AHSS concepts to industrial prac-
tice. In Proceedings of the international symposium on new developments in advanced 
high strength sheet steels, June 23–27, 2013. USA Colorado: AIST.
Myazawa, K., & Schwerdtfeger, K. (1981). Macrosegregation in continuously cast steel slab: 
preliminary theoretical investigation on the effect of steady state bulging. Arch Eisenhüt-
tenwes, 52, 415.
Nadif, M., Suero, J., Rodhesly, C., Salvadori, D., Schadow, F., Schutz, R., et al. (Juillet/Aout 
2009). Desulphurization practices in ArcelorMittal flat carbon Western Europe. La Revue 
de Metallurgie – CIT, 270–279.
Norden, M. (2013). Recent trends in hot dip galvanizing of advanced high strength steels at 
ThyssenKrupp steel Europe. Iron & Steel Technology, 67–73.
Perlade, A., Grandemange, D., & Iung, T. (2005). Application of microstructural modelling for 
quality control and process improvement in hot rolled steels. Ironmaking and Steelmaking, 
32(4), 299–302.
Perlade, A., Grandemange, D., Huin, D., Couturier, A., & Oostsuka, K. (September 2008). A 
model to predict the austenite evolution during hot strip rolling of conventional and Nb 
micro alloyed steels. La Revue de Métallurgie – CIT, 443–451.
Pethe, N., Zheng, K., Huin, D., Moretto, C., & Poliak, E. (2011). Dynamic run-out-table cool-
ing simulator and temperature controllers. In Sensors, sampling and Simulation for pro-
cess control. Hoboken, New Jersey: John Wiley and Sons, Inc. TMS, The Minerals, Metals 
and Materials Society.
Poliak, E., Pottore, N. S, Skolly, R. M., Umlauf, W. P., & Brannbacka, J. C. (2009). Thermo-
mechanical processing of advanced high strength steels in production hot strip rolling. la 
metallurgia italiana, 1–8.
Sadagopan, S., & Urban, D. (2003). Formability characterization of a new generation of high 
strength steels. Work Performed under Cooperative Agreement No. DE-FC07–97ID13554. 
Prepared for U.S. Department of Energy, Prepared by American Iron and Steel Institute, 
Technology Roadmap Program Office, For availability of this report, contact: Office of 
Scientific and Technical Information, P.O. Box 62, Oak Ridge, TN 37831, (615) 576–8401.
Sengupta, J., Casey, J., Crosbie, D., Nelson, B., Kladnik, G., Gao, N. (2011). Qualitative and 
quantitative techniques for evaluating manganese segregation in advanced high strength 
steels at ArcelorMittal Dofasco’s No. 1 continuous caster. In AIESTech 2011 proceedings 
(Vol. II(52)) (pp. 731–740).
53Manufacturing of AHSS
Staudte, J., Mataigne, J. M., Loison, D., & Del Frate, F. (2011). Galavanizability of high Mn 
grade versus mixed Mn-Al and Mn-Si grades. In 8th International conference on Zinc and 
Zinc Alloy coated steel sheet, Genova, AIM; Galvatech (Italy), June 21–24.
Tuling, A., Banarjee, J. R., & Mintz, B. (2011). Influence of peritectic phase transformation on 
hot ductility of high aluminium TRIP steels containing Nb. Material Science and Technol-
ogy, 27(11), 1724–1731.
Wallmeyer, R. (2013). Coil-to-coil joining with laser welding in particular for ‘sophisticated 
steel grades’. In Proceedings of the international symposium on new developments in 
advanced high strength sheet steels, June 23–27, 2013. USA Colorado: AIST.
Wolf, M. M. (1991). Estimation method of crack susceptibility for new steel grades. In ECC 
Florence, 1st European conference on continuous casting, Florence, Italy, September 
23–25.
Yuqing, W., Dong, H., & Gan, Y. (2011). Advanced steels – proceedings of the first international 
conference on advanced steels. In Proceedings of the international symposium on new 
developments in advanced high strength sheet steels, June 23–27, 2013. USA Colorado: 
AIST.
 
This page intentionally left blank
Welding and Joining of Advanced High Strength Steels (AHSS). http://dx.doi.org/10.1016/B978-0-85709-436-0.00004-7
Copyright © 2015 Elsevier Ltd. All rights reserved.
Resistance spot welding techniques 
for advanced high-strength steels* 
(AHSS)
M. Tumuluru
Research and Technology Center, United States Steel Corporation, Pittsburgh, PA, USA
4
4.1 Introduction
Based on the expressed objectives of various automotive companies to build more 
fuel-efficient and safer cars, as well as the trends in steel usage to meet increasingly 
stringent government requirements across the globe, it is anticipated that advanced 
high-strength steel (AHSS) usage in automotive bodies will climb to 50% by 2015 
(Horvath, 2004; Pfestorf, 2006). For example, over 50% of the body in white (BIW) 
of the 2014 MDX vehicle from Honda was made from AHSS grades with strength 
levels in excess of 590 MPa (Keller, 2014). In addition, AHSS in the Hyundai Sonata 
accounted for 21% of parts in the 2010 model year, whereas in the 2014 model, AHSS 
accounted for nearly 53% of the parts in BIW applications (Chang, 2015).
Resistance spot welding is the main method of joining used in the automotive indus-
try, with each vehicle containing several thousand welds (Tumuluru, 2006a). Further-
more, resistance spot welding alone or in combination with other joining processes, 
such as adhesive joining or laser welding, accounts for more than 70% of all welding 
done in BIW applications (Tumuluru, 2013). Recent investment and equipment upgrade 
trends in automotive assembly plants across the globe suggest that the resistance spot 
welding process will continue to be the most dominant process for automotive joining 
for several years to come. Therefore, weldability assessment of AHSS grades using the 
resistance spot welding process is a critical step in the implementation of these steels 
in automotive applications. To be able to successfully use these steels, characterizing 
and understanding the resistance spot welding behavior of AHSS grades is important.
To provide corrosion protection for BIW applications, most steels are used with 
either galvannealed (zinc–iron alloy) or galvanized (pure zinc) coatings. These coatings 
are typically applied eitherby dipping incoming steel coils in a molten pool of zinc or 
by electrolytically plating zinc onto steel surfaces. The typical coating weights for steel 
in automotive applications range from 40 to 60 g/m2 for galvannealed applications and 
* Disclaimer: The material in this chapter is intended for general information only. Any use of this material 
in relation to any specific application should be based on independent examination and verification of its 
unrestricted availability for such use and a determination of suitability for the application by professionally 
qualified personnel. No license under any patents or other proprietary interest is implied by the publication 
of this chapter. Those making use of or relying on the material assume all risks and liability arising from 
such use or reliance.
56 Welding and Joining of AHSS
50–70 g/m2 per side for galvanized applications. Therefore, understanding the effect of 
these coatings on the resistance spot welding behavior of these steel grades is important.
The two most commonly used grades of AHSS in automotive BIW welded 
applications are dual-phase and transformation-induced plasticity (TRIP) steels 
(Ducker Worldwide report 2007). Although other AHSS grades such as complex- 
phase and twinning-induced plasticity steels are commercially available, they have 
limited applicability in the automotive industry because of their high cost and limited 
global availability. According to Ducker Worldwide report 2007, by 2020 dual-phase 
steels will account for nearly 280 lb, and TRIP steel, along with complex-phase steel, 
will account for approximately 55 lb, of a typical passenger vehicle. Given the pro-
jected extensive use of these two steel grades, this chapter focuses on the weld charac-
terization and the welding behavior of these two commonly used AHSS grades.
4.2 Characterizing welding behavior
Several tests are generally used to characterize the resistance spot welding behavior 
of AHSS. These include the welding current range, metallographic characterization of 
the microstructures in the weld and the heat-affected zone, microhardness, and weld 
tensile tests (American Welding Society D8.9M-2012; Tumuluru, 2006b).
4.2.1 Welding current range
The useful current range is the difference between the welding current required to pro-
duce a minimum weld size (Imin) and the current that causes expulsion of weld metal 
(Imax). The minimum weld size is typically defined as 4√t, where t is the nominal sheet 
thickness. This definition is generally used in the automotive and steel industries. The 
procedure to determine the current range is described in detail in a specification from the 
American Welding Society D8.9M-2012. Peel test coupons measuring 140 × 50 mm2 are 
generally used in determining the current range (Figure 4.1). The coupons overlap by 
Anchor weld Test weld
Overlap
W
50 mm
50 mm
140 mm
L
Figure 4.1 Schematic of a peel test coupon (Tumuluru, 2006b).
Source: Tumuluru, M. (2006b). A comparative examination of the resistance spot welding 
behavior of two advanced high strength steels. In: SAE technical paper No. 2006-01-1214, 
presented at the SAE congress, Detroit, MI. Copyright © SAE International. Reprinted with 
permission from 206-01-1214.
57Resistance spot welding techniques for AHSS
25 mm, and a shunt or anchor weld is made on one side of each coupon pair. On the other 
side, test welds are made 35 mm from the edge. The test welds are peeled open and the 
weld sizes are measured using calipers. The current range is useful because it provides 
a range of welding currents over which welds of an acceptable size can be produced.
Before determining the current range, the electrode tips are generally conditioned 
by making about 100 welds. Current ranges are identified by first determining the 
lowest welding current that produces the minimum acceptable weld size. Then, the 
current is gradually increased until weld metal expulsion results. The range of current 
between Imin and Imax is regarded as the welding current range.
4.2.2 Weld lobes
Another way to characterize the suitability of a given grade of AHSS is to determine the 
weld lobes for the steel. Weld lobes are graphical representations of the useful current 
ranges that provide acceptable welds without expulsion and button sizes that are above 
the minimum required sizes. In other words, weld lobes are similar to the welding 
current ranges. Weld lobes are typically determined using three different weld times. 
Before determining weld lobes, generally 100 conditioning welds are made to condi-
tion the electrode tips. To determine weld lobes, the welding currents that produced 
minimum weld sizes are determined at each of the weld times chosen. Weld times (the 
duration of the passage of welding current through the electrodes) are chosen based 
on a suggested nominal welding time for a given thickness of steel. These weld times 
vary depending on the specification to which a given grade of steel is being tested. 
The welding current then is increased until expulsion occurs. The expulsion current 
is determined for three specified weld times chosen. When the nominal welding time 
is known, it is typical to use ±10% of this time to select the other two weld times.
4.2.3 Weld shear tension and cross-tension tests
Weld shear tension strength and cross-tension strength (CTS) are determined to assess 
the load-bearing ability of welds (Tumuluru, 2006a) For determining the shear tension 
strength, 140- × 60-mm samples are sheared and a single spot weld is made at the 
center of an overlapped area measuring 45 mm (Figure 4.2). For cross-tension tests, 
the test coupons used are 150 mm long and 50 mm wide (Figure 4.3). Two coupons are 
placed at 90° to each other and a spot weld is made at the center of the overlapped area. 
140 mm
45 mm
60 mm
Figure 4.2 Shear tension test coupon dimensions (top) and layout (bottom) (Tumuluru, 2006a).
58 Welding and Joining of AHSS
Before making the weld test samples, the electrode tips are conditioned by making 
100 welds on flat panels. Per the American Welding Society D8.9M-2012, all shear 
and cross-tension test samples are generally prepared with a specified weld size. This 
is normally the electrode face diameter weld size, which is slightly bigger than 90% 
of the electrode face diameter. Additional details on the testing methodology are avail-
able from the American Welding Society D8.9M-2012.
4.2.4 Weld fracture appearance
Appearance of fractures in the welds is generally determined on all weld tensile test 
samples after the tests. Weld fractures are typically classified as full button pull-out, an 
interfacial fracture, or a partial interfacial fracture. In the full button pull-out fracture 
mode the entire weld nugget pulls out from the sheets because of a fracture occur-
ring outside of the weld area. In an interfacial fracture, the entire weld fails through 
the plane of the weld. In a partial interfacial fracture, part of the weld nugget fails 
through the plane of the weld and some portion of the weld pulls out as a partial button 
(Figure 4.4).
It is also possible to have a combination of two failure modes in which a portion 
of the nugget is pulled out of one of the sheets and the rest of the nugget shears at the 
Figure 4.3 Cross-tension test specimen dimensions 
and layout (Tumuluru, 2006a).
Figure 4.4 Weld fracture types showing interfacial (top) and button pull mode (bottom).
59Resistance spot welding techniques for AHSS
interface. A detailed description of various fracture morphologies that are possible 
in resistance spot welds is provided by the American Welding Society D8.1M-2007.
4.2.5 Weld microhardness
Microstructures of the weld and heat-affected zone are generally examined to check 
for any imperfections, such as voids and cracks, and to provide an understanding of 
the tensile properties of the weld. Weld microhardness profiles are determined by 
measuringhardness at 0.4-mm intervals along a diagonal in a weld cross section 
(American Welding Society D8.9M-2012). If more information about softening in the 
heat-affected zone is required, indentations can be spaced more closely, at 0.2-mm 
intervals. In such a case, however, the indentations may need to be staggered to main-
tain a sufficient distance between successive indentations.
4.3 General considerations in resistance spot 
welding of AHSS
In resistance welding the materials to be joined are heated through I 2 Rt, where I is the 
current used for welding, R is the resistance offered to the passage of current, and t is 
the duration of the current’s passage. Therefore, the resistance offered by the steel is an 
important factor that controls weld nugget development. The term R here is a sum of all 
resistances, including the resistivities of the two steel sheets being welded, the interfa-
cial resistance at the sheet-to-sheet interface, as well as the two interfacial resistances 
at the sheet-to-electrode interfaces. The interfacial resistances between the sheets and 
the bulk resistivities are critical to heat development because cooling the electrodes in 
water removes the heat at the sheet-to-electrode interfaces. However, the steel resistiv-
ity must be controlled to prevent the generation of excessive heat. Excessive or uncon-
trolled heat generation can lead to weld metal expulsion, which is undesirable. Because 
of the high alloy content of AHSS compared with low-strength steel, AHSS has high 
resistivity and is therefore likely to heat rapidly at the sheet-to-sheet interface. If heat 
generation is not controlled properly, weld metal expulsion will result.
One way to control the effect of the higher resistivity of AHSS is to use higher 
electrode force compared with that used for welding low-strength steels, such as inter-
stitial-free steels. The beneficial effect of using a higher electrode force can be seen in 
Figure 4.5 (Tumuluru, 2008a). As the electrode force increased from 2.9 kN (650 lbf) to 
5.3 kN (1200 lbf), the welding current increased from 0.6 to 1.4 kA. A current of 1.0 kA 
is generally regarded as acceptable for production use. It is clear from Figure 4.5 that a 
welding force of 3.6 kN (800 lbf) is required to obtain the 1-kA current. It is cautioned 
that with the use of high force, the indentation of the electrode into the base mate-
rial should be monitored because higher electrode force causes deeper indentation. 
In general, the indentation at a given force is greater in a low-strength steel than in a 
high-strength one. As a general guideline, the electrode indentation should be less than 
25% of the base material thickness to avoid the possibility of creating a stress raiser.
60 Welding and Joining of AHSS
4.3.1 Welding dual-phase steel
The resistance spot welding behavior of coated dual-phase steel has been the focus of 
previous research (Tumuluru, 2006a, 2006b). This understanding led to the successful 
implementation of dual-phase steel, ranging in nominal strength from 590 to 980 MPa, 
into automotive production. The microstructure of dual-phase steel contains a fine dis-
tribution of a hard martensite phase in a soft and ductile ferrite phase. The amount of 
martensite depends on the strength of the steel; 780-MPa steel typically contains about 
25% martensite. This unique combination of martensite and ferrite gives dual-phase 
steel high strength and high ductility. The welding currents obtained for various dual-
phase steels are shown in Figure 4.6 (Tumuluru, 2006a). This indicates that production 
welding of dual-phase steels can be accomplished with relative ease using a variety 
of suitable welding conditions and that welds of acceptable quality can be achieved. 
Because dual-phase steel has higher alloy content than low-strength steel, it has high 
resistivity and generally requires much lower welding currents to weld.
AHSS can be welded using a wide variety of electrode tip shapes. The two 
most commonly used tip shapes are a truncated cone and a dome (also known as 
a ball nose). A description of these and other electrode tip designs is provided by 
ISO 5821:2009. Figure 4.7 shows the effect of the electrode tip shape on welding 
current ranges. It is clear that dome-shaped tips provide more consistent welds 
than a truncated cone-shaped tip. The dome-shaped (also referred to as ball-nosed) 
tips produced a broader current range and more welds that meet a certain mini-
mum strength requirement for 1.6-mm steel, which is typically around 8800 N. 
The dome-shaped tips also provide a larger contact area than the cone-shaped 
Figure 4.5 A plot showing welding current range as a function of electrode force for 1-mm, 
780-MPa, dual-phase steel (Tumuluru, 2008a).
61Resistance spot welding techniques for AHSS
electrodes and thereby reduce the current density (Chan et al., 2006). As a result, 
dome-shaped electrodes increase the current required to produce a weld. This, in 
turn, increases the welding current range.
From Figure 4.7 it is also apparent that the use of pulsed currents did not widen the 
welding current range for the 1.6-mm 980 dual-phase steel. In general, the use of pulsed 
currents is useful for thinner-gauge steel to better control nugget growth and avoid 
expulsion. It is also apparent from Figure 4.7 that, in the case of 980 dual-phase steel, 
the use of the dome-shaped electrode significantly increased the welding current range.
4.3.2 Welding TRIP steel
TRIP steel contains austenite and bainite in a matrix of ferrite. When subjected to plastic 
deformation, the austenite transforms into martensite. This strain-induced transformation 
of austenite to martensite gives added ductility to the steel. As a result, TRIP steel pos-
sesses better formability than dual-phase steel. The properties and the physical metallurgy 
of dual-phase and TRIP steel have been extensively studied and reported in the literature 
(Baik, Kim, Jin, & Kwon, 2000; Takahashi, Uenshi, & Kuriyama, 1997). While both types 
of steel can contain up to 0.15 weight-percent carbon, TRIP steel generally contains addi-
tional alloying elements to avoid the formation of cementite so that the austenite phase is 
enriched in carbon. These alloying additions can cause differences in the weld hardness.
A comparative study completed to examine the welding behavior of 780-MPa 
dual-phase and TRIP steels reported that both types of steel exhibited similar weld 
lobes (Figure 4.8). This suggests that the welding behavior of the two is similar 
(Tumuluru, 2006b). Except for the current required to produce the minimum weld size 
Figure 4.6 A plot showing the welding current ranges obtained for 1.6-mm dual HDGA steel 
at various strengths (Tumuluru, 2006a).
62 Welding and Joining of AHSS
for the dual-phase steel, the welding currents required to produce welds of a minimum 
diameter and those that resulted in the first instance of expulsion were almost iden-
tical. Even the welding current required to obtain a weld of minimum diameter was 
only 200 A lower for dual-phase steel when compared with TRIP steel. At the 18-cycle 
weld time that was used to prepare the tensile test samples in the study reported by 
Biro, Mingsheng, Zhiling and Zhou (2008), the weld lobe for the dual-phase steel was 
only 100 A lower than that of the TRIP steel. This small difference could be from the 
inherent variation present in the determination of the lobes. The practical implications 
Figure 4.7 The effect of electrode tip shape on the welding current range for 780-MPa (a) 
and 980-MPa (b) dual-phase steels (1.6-mm). The steels contained hot-dipped galvannealed 
coating. Panel (b) also shows the effect of the use of pulsed currents on current ranges.
63Resistance spot welding techniques for AHSS
of these observations reported by Tumuluru (2006b) are that 780-MPa TRIP steel can 
be welded with similar welding parameters as those required to weld 780-MPa dual-
phase steel.The second implication from this study was that acceptable welds with no 
imperfections can be obtained, even with the use of simple-to-use and easily adopt-
able welding parameters. It should, however, be noted that in auto body fabrication 
under shop-floor conditions, fit-up of parts generally dictates the welding parameters 
Figure 4.8 Weld lobes obtained for 780-MPa dual-phase (a) and transformation-induced 
plasticity (TRIP) (b) steel (1.6-mm) (Tumuluru, 2006b).
Copyright © SAE International. Reprinted with permission from 206-01-1214.
64 Welding and Joining of AHSS
required to obtain acceptable welds, and these parameters may differ from those that 
produce acceptable welds under laboratory conditions.
4.4 Coating effects
Two types of coatings are generally applied to steel sheets used in the automotive 
industry, namely, galvanized and galvannealed coatings. Galvanized coatings con-
tain essentially pure zinc with about 0.3–0.6 weight-percent aluminum. The term 
galvanize comes from the galvanic protection that zinc provides to a steel substrate 
when exposed to a corroding medium. A galvannealed coating is obtained by heating 
the zinc-coated steel at 450–590 °C immediately after the steel exits the zinc bath. This 
additional heating allows iron from the substrate to diffuse into the coating. Because 
of the diffusion of iron and its alloying with zinc, the final coating contains around 
90% zinc and 10% iron. There is no free zinc present in the galvannealed coating 
because of this alloying. Dipping coils of steel into a molten bath of zinc is known 
as the hot-dip process. HDGA refers to hot-dipped galvannealed products, whereas 
HDGI refers to hot-dipped galvanized products. HDGA coatings contain less alumi-
num (about 0.15–0.4 weight-percent) than HDGI coatings. Another way of applying 
the coatings is through an electrolytic plating process. Scanning electron micrographs 
showing cross-sectional views of HDGI and HDGA coatings are shown in Figure 4.9.
In an investigation undertaken to examine whether differences in the resistance 
spot welding behavior of 780-MPa dual-phase steel with an HDGA coating exist 
compared with steel with an HDGI coating, the welding current ranges, weld shear 
tension strength and CTS, and weld microhardness profiles across the welds were 
examined (Tumuluru, 2008b). The results indicated that 780-MPa dual-phase steel 
showed similar overall welding behavior with HDGA and HDGI coatings. This work 
also showed that the weld shear tension strength and CTS were independent of the type 
of coating. HDGA-coated steel was able to be welded at a slightly lower current than 
Research and technology center
2 µm Mag = 2.50 K X
WD = 6 mm
EHT = 15.00 kV
Signal A = SE2
(a) (b)
Figure 4.9 Scanning electron microscope views of cross sections of hot-dipped galvanized (a) 
and hot-dipped galvannealed (b) coatings on 780-MPa dual-phase steel (Tumuluru, 2008b).
65Resistance spot welding techniques for AHSS
the HDGI-coated steel: the reason for this behavior was attributed to the differences in 
surface resistitivity between the coatings (Figure 4.10).
4.5 Microstructural evolution in welds
Because electrodes are cooled in water in resistance spot welding, the weld cool-
ing rates are extremely rapid. Spot welds with a thickness up to 2 mm typically 
solidify in less than three or four cycles. It has been shown through modeling 
that even at 500 °C the cooling rates in spot welding were in excess of 1000 °C/s 
(Li, Dong, & Kimchi, 1998). For steel, the critical cooling rate (ν) required to 
achieve martensite in the microstructure is determined using the following equation 
(Easterling, 1993):
 log v = 7.42 − 3.13 C − 0.71 Mn − 0.37 Ni − 0.34 Cr − 0.45 Mo 
For 780-MPa dual-phase steel, the critical rate is about 240 °C/s. As a result, 
a martensitic structure is typically present in both the weld and the heat-affected 
zone. Even in the near heat-affected zone, martensite is the predominant constit-
uent (Figure 4.11). However, in the far heat-affected zone (the part of the heat-af-
fected zone that is closer to the unaffected base material), some of the martensite 
is tempered and a decrease in the hardness can be noted (Figure 4.12). As can be 
seen from Figure 4.12, this drop in the far heat-affected zone increases from the 
780–980-MPa strength level because of the higher alloying content in the 980-
MPa steel. This phenomenon of heat-affected zone softening in welds was studied 
Figure 4.10 Plot of welding current ranges for hot-dipped galvannealed (HDGA)- and 
hot-dipped galvanized (HDGI)-coated 780 MPa dual phase steels. Notice the higher current 
required to obtain the minimum weld size for the HDGI-coated steel (Tumuluru, 2008b).
66 Welding and Joining of AHSS
Research and technology center
1µm
1µm 1µm
Mag = 5.00 K X
Mag = 5.00 K X Mag = 5.00 K X
WD = 13 mm
WD = 19 mmWD = 18 mm
EHT = 15.00 kV
EHT = 15.00 kVEHT = 15.00 kV
Signal A = SE2
Signal A = SE2Signal A = SE2
(a) (b)
(c)
Figure 4.11 Scanning electron micrographs showing the microstructure of welds (a), the near 
heat-affected zone (b), and the far heat-affected zone (c).
100
150
200
250
300
350
400
450
500
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32
H
ar
dn
es
s 
(V
H
N
)
Indentations
780 DP
980 DP
Figure 4.12 A plot showing weld microhardness traverses for 780- and 980-MPa dual-phase 
steel. The heat-affected zone showed a decrease in hardness of about 30 VNH for the 
980-MPa steel.
67Resistance spot welding techniques for AHSS
by Biro et al. (2008), who found evidence of martensite tempering in dual-phase 
steel welds. Other researchers have confirmed these findings, as well (Okita, 
Baltazar Hernandez, Nayak, & Zhou, 2010).
4.6 Weld shear tension strength and cross-tension 
strength (CTS)
4.6.1 Shear tension strength
Results of actual testing and finite element modeling (FEM) show that the shear ten-
sion strength in AHSS grades increases as the base material strength increases. The 
fracture mode in shear tension testing of AHSS grades up to 980 MPa shows that there 
are essentially two types of fracture modes, namely, the full button pull-out mode 
(also known as button pull or plug-type fractures) and the interfacial fracture mode. 
For pull-out failure, the results of the finite element simulations by Radakovic and 
Tumuluru (2008) showed that there was a strong correlation between failure load and 
the material strength, sheet thickness, and weld diameter. The load required to cause 
interfacial failure was more strongly dependent on the weld diameter and less depen-
dent on the sheet thickness. The predicted failure loads were found to adhere to the 
following correlations:
 FPO = kPO · σUT · d · t (4.1)
 FIF = kIF · σUT · d2 (4.2)
where FPO is the failure load for pull-out failure, FIF is the failure load for an interfa-
cial fracture, σUT is the tensile strength of the material, d is the weld diameter, and t 
is the sheet thickness. These equations were derived based on the fact that the force 
required to cause failure is equal to the product of the strength of the material and the 
cross section of the failed area. In this analysis the material was assumed to be homo-
geneous. Therefore the strengths of the weld and the base metal are both equal to σUT. 
In Eqns (4.1) and (4.2), kPO and kIF were constants determined from the modeling.
Radakovic and Tumuluru (2008) also determined that there is a critical sheet thick-
ness above which the expected failure mode could move from pull-out to interfacial 
fracture. Further, they found that as the strength of the sheet increases, the fracture 
toughness of the weld required to avoid interfacial fractures must also increase. In 
higher-strength, less ductile steel this is not likely to occur, and interfacial fracture 
could become the expected failure mode. The load-carrying capacity of the samples 
that failed via interfacial fracture was more than 90% of the maximum loadassociated 
with the full button pull-out. This indicates that the load-bearing capacity of these 
welds is not significantly affected by the fracture mode. The mode of failure should 
therefore not be the only criteria used to judge the results of the shear tension test. The 
load-carrying capacity of the weld should be considered the most important parameter 
when evaluating the shear tension test results in AHSS.
68 Welding and Joining of AHSS
4.6.2 Cross-tension strength (CTS)
The results of cross-tension testing for both 780 and 980 steel grades are shown in 
Figure 4.13 (Radakovic & Tumuluru, 2012; Tumuluru & Radakovic, 2010). In Figure 
4.13 the weld CTS was plotted as a function of weld size. As the plot shows, the CTS 
increased with weld size for both grades. In this test the full button pull-out fracture 
mode occurred in both grades and at all weld sizes. As can be seen in Figure 4.13, CTS 
for the 980-MPa steel was slightly lower than that for the 780-MPa steel; this differ-
ence was more noticeable at weld sizes larger than 4√t. Tumuluru and Kashima (2009) 
reported that CTS decreased as the carbon content of the base material increased from 
0.05% to 0.2% in dual-phase steel. This research also showed that as the carbon con-
tent increased, the tensile strength of the base material also increased and the ductility 
decreased (Tumuluru & Kashima, 2009). This indicates an inverse relationship between 
the tensile strength of the base material and the weld CTS. A similar trend in CTS with 
increasing base material strength was observed by Sakuma and Oikawa (2003).
The modeling performed by Tumuluru and Radakovic (2010) indicated that the 
failure load in the cross-tension test is related to the sheet thickness and the strength 
and ductility of the heat-affected zone. Actual cross-tension test results supported the 
model predictions that there was a correlation between failure load and weld size. In 
the cross-tension testing, at all weld sizes tested, full button pull-out fractures occurred 
in the two steel grades tested. This result agreed with the model results, which indi-
cated that the weld would not overload until the button size became much smaller 
than those achieved in the test samples. In the cross-tension test, the constraint of the 
sample grips prevents lateral movement of the sheet as the sample is pulled vertically. 
Modeling also showed that this constraint causes high tensile force to develop in the 
sample perpendicular to the direction of the applied load. For this reason, pull-out 
failures are the preferred failure mode in this test, even for very small weld sizes. 
2
4
6
8
10
12
2 3 4 5 6 7 8 9
780 DP
980 DP
 C
TS
 (k
N
)
 Weld size (mm)
Figure 4.13 Cross-tension test results for 1.2-mm, 780- and 980-MPa dual-phase steels. The 
cross-tension strength (CTS) is shown as a function of weld size. Full button pull-out failures 
occurred at all weld sizes in both the grades (Radakovic and Tumuluru, 2008).
69Resistance spot welding techniques for AHSS
Tumuluru and Radakovic examined an actual crash-tested vehicle and found that the 
deformation of the sheets around the welds in this vehicle was similar in appearance 
to that which occurred in the cross-tension test, with buckling of the joined sheets 
between spot welds. However, this type of buckling or crumpling in the crash-tested 
vehicle was the result of compressive loading and not tensile loading as predicted 
by the finite element method for the cross-tension test. Based on this work, it was 
concluded that the cross-tension test is neither a discriminating test for assessing the 
weldability of high-strength steel nor does it represent the type of loading that the spot 
welds in a vehicle undergo in a real crash event.
4.7 Summary
Resistance spot welding is the predominant method of joining used in automotive 
assembly plants for welding a wide variety of parts, a trend that is likely to continue for 
the foreseeable future. Several new grades of AHSS, including dual-phase and TRIP 
steel, have been commercialized between 2000 and 2010. These steel grades are being 
increasingly used, especially in BIW applications, to meet the ever-increasing demands 
across the globe for improved fuel efficiency and occupant protection. Research com-
pleted to date indicates that these AHSS grades are easily weldable. In general, AHSS 
grades require the use of higher electrode forces and larger tip sizes to achieve accept-
able welding current ranges. AHSSs with a tensile strength of 980 MPa or higher show 
softening in the far heat-affected zone, which can affect their cross-tension test behav-
ior. Research has also demonstrated that the fracture morphology from weld tension 
testing should not be used as the sole criterion to assess the results of shear tension 
tests. Load to failure is an important attribute of the test that should be emphasized.
References
American Welding Society Specification D8.1M-2007. (2004). Specification for automotive 
quality – Resistance spot welding of steel. Miami, FL: American Welding Society.
American Welding Society Specification D8.9M-2012. (2012). Recommended practices for 
test methods for evaluating the resistance spot welding behavior of automotive sheet steel 
materials. Miami, FL: American Welding Society.
Baik, S. C., Kim, S., Jin, Y. S., & Kwon, O. (2000). Effects of alloying elements on mechanical 
properties and phase transformation of cold rolled steel sheets. In SAE technical paper 
2000-01-2699, Detroit, MI.
Biro, E., Mingsheng, X., Zhiling, T., & Zhou, N. (2008). Effects of heat input and martensite on 
HAZ softening in laser welding of dual phase steels. ISIJ International, 48(6), 809–814.
Chang, I. (2015). Recent trend of welding and joining application in automotive industry. In 
Paper presented at the international congress of the International Institute of Welding, 
Seoul, South Korea.
Chan, K., Scotchmer, N., Bohr, J. C., Khan, I., Kuntz, M., & Zhou, N. (2006). In Effect of elec-
trode geometry on resistance spot welding of AHSS, 4th international seminar on advances 
in resistances in resistance welding, November 14–16, 2006, Wels, Austria.
Ducker Worldwide report 2007, www.duckerworldwide.com.
http://www.duckerworldwide.com
70 Welding and Joining of AHSS
Easterling, K. E. (1993). Modeling the weld thermal cycle and transformation behavior in the 
heat affected zone. In H. Cerjak, & K. E. Easterling (Eds.), Mathematical modeling of weld 
phenomena. The Institute of Materials.
Horvath, C. D. (2004). The future revolution in automotive high strength steel usage. In Paper 
presented at great designs in steel, American Iron and Steel Institute, Southfield, Mich.
ISO 5821:2009: Resistance welding – spot welding electrode caps, www.iso.org.
Keller, J. (2014). Weight down….Value up … and the new MDX. www.cargroup.org.
Li, M. V., Dong, D., & Kimchi, M. (1998). Modeling and analysis of microstructure devel-
opment in resistance spot welds of high strength steels. SAE Technical Paper 982278. 
Warrendale, PA: SAE International.
Okita, Y., Baltazar Hernandez, B. H., Nayak, S. S., & Zhou, N. (2010). Effect of HAZ-softening 
on the failure mode of resistance spot-welded dual-phase steel. In Paper presented at the 
sheet metal welding conference XIV, May 11–14, Livonia, MI.
Pfestorf, M. (2006). BMW – functional properties of the advanced high strength steels in the 
body-in-white. In Paper presented at great designs in steel, American Iron and Steel Insti-
tute, Southfield, Mich.
Radakovic, D. J., & Tumuluru, M. (April 2008). Predicting resistance spot weld failure modes 
in shear tension tests of advanced high-strength automotive steels. Welding Journal, 
American Welding Society, Miami, FL, 96S–105S.
Radakovic, D. J., & Tumuluru, M. (2012). An evaluation of the cross-tension test of resistance 
spot welds in high strength dual phase steels. Welding Journal, 91, 8S–15S.
Sakuma, Y., & Oikawa, H. (2003). Factors to determinestatic strength of spot-weld for high 
strength steel sheets and development of high-strength steel sheets with strong and stable 
welding characteristics. Nippon Steel Corporation, Japan, Nippon Steel Technical Report 
No. 88.
Takahashi, M., Uenshi, A., & Kuriyama, Y. (1997). Properties of high strength TRIP steel 
sheets. Automotive Body Materials. IBEC.
Tumuluru, M. (2006a). An overview of the resistance spot welding of coated high strength dual 
phase steel. Welding Journal, 85(8), 31–37s.
Tumuluru, M. (2006b). A comparative examination of the resistance spot welding behavior of 
two advanced high strength steels. In SAE technical paper No. 2006-01-1214, presented at 
the SAE congress, Detroit, MI.
Tumuluru, M. (2008a). Some considerations in the resistance spot welding of dual phase steels. In 
Paper presented at the 5th international seminar on advances in resistance welding, September 
24–26, 2008, Toronto, Canada, organized by Huys Industries, Weston, Ontario, Canada.
Tumuluru, M. (June 2008b). The effects of coatings on the resistance spot welding behavior 
of 780 MPa dual phase steel. Welding Journal, American Welding Society, Miami, FL, 
161S–169S.
Tumuluru, M. (2013). Evolution of steel grades, joining trends and challenges in the automotive 
industry. In Invited keynote presentation, American Welding Society FABTECH welding 
show and conference, Chicago, IL.
Tumuluru, M., & Kashima, T. (2009). Effect of alloying elements on resistance spot weld per-
formance in dual phase steels. In Paper presented at AWS FABTECH conference, Chicago, 
November 17, 2009. Miami, FL: American Welding Society.
Tumuluru, M., & Radakovic, D. J. (2010). Modeling of cross-tension behavior of dual phase and 
TRIP steels. In Sheet metal welding conference XIV. Livonia, MI: American Welding Society.
http://www.iso.org
http://www.cargroup.org
Welding and Joining of Advanced High Strength Steels (AHSS). http://dx.doi.org/10.1016/B978-0-85709-436-0.00005-9
Copyright © 2015 Elsevier Ltd. All rights reserved.
Laser welding of advanced 
high-strength steels (AHSS)
S.S. Nayak1, E. Biro2, Y. Zhou1
1University of Waterloo, Waterloo, ON, Canada; 2ArcelorMittal Global Research, 
Hamilton, ON, Canada
5
5.1 Introduction
In recent years the use of advanced high-strength steel (AHSS) has increased in popu-
larity in the automotive industry because of its excellent combination of high strength 
and ductility, which allows the thickness of steel used to make auto body parts to be 
decreased, which in turn facilitates a reduction in vehicle weight while improving 
safety (Blank, 1997; Gan, Babu, Kapustka, & Wagoner, 2006). Typical AHSS families 
used in automotive construction, for example, dual-phase (DP) steel, transformation- 
induced plasticity (TRIP) steel, complex phase steel, martensitic steel, and twinning- 
induced plasticity (TWIP) steel, are characterized by yield strengths and ultimate tensile 
strengths higher than 300 and 600 MPa, respectively (Bhadeshia & Honeycombe, 2006; 
World Auto Steel, 2009). Understanding the individual microstructures of these steels 
is critical to understand the changes that occur during laser welding. TRIP steel has 
a microstructure consisting of retained austenite and martensite islands dispersed in a 
ferrite matrix. TWIP steel contains a fully austenitic microstructure because of its high 
manganese content, that is, 17–24 wt% (De Cooman, Kwon, & Chin, 2012). Finally, 
martensitic steel contains a fully martensite microstructure, as is suggested by its name. 
Of all AHSS, only DP steel and TRIP steel have been identified as potential candidates 
for car body fabrication because of their better formability compared with martensitic 
steel and their low manufacturing cost compared with TWIP steel. Because of this, the 
research on AHSS has mainly focused on DP and TRIP steels; therefore, this chapter 
outlines a review of the work on these two steels carried out so far.
A design technique that is unique to the laser welding process is the production of 
laser-welded blanks (LWBs), which are also known as tailor-welded blanks. LWBs are 
composed of two or more sheets of similar or dissimilar materials, thicknesses and/or 
coating types welded together, which are formed to fabricate three-dimensional auto-
motive body parts (Auto/Steel Partnership, 1995). The advantages of LWBs include 
weight reduction, cost minimization, material usage and scrap reduction with improved 
part integration. In general, LWBs are made with mild or interstitial free steels; how-
ever, in recent years the ability of LWBs to reduce weight has been increased by using 
high-strength steels such as high-strength low-alloy (HSLA) steel and AHSS, which 
also improves the crash performance of the blanks (Kusuda, Takasago, & Natsumi, 
1997; Shi, Thomas, Chen, & Fekte, 2002; Uchihara & Fukui, 2006). LWBs are almost 
72 Welding and Joining of AHSS
exclusively joined in a butt joint configuration. Hence this chapter likewise focuses on 
butt joints in various AHSSs, mainly DP steels and TRIP steels. Before discussing the 
microstructure and properties in laser welding of AHSS, however, a brief background 
of the laser welding process is provided in Section 5.2.
5.2 Background
The beam intensity, average power and flexibility of beam delivery at different loca-
tions make laser welding a popular welding process. Traditional carbon dioxide lasers 
have been predominately used in industry because of their plug efficiency when 
compared with solid-state lasers such as neodymium:yttrium–aluminium–garnet 
(Nd:YAG). However, ytterbium fibre lasers have recently gained industrial acceptance 
because of their high plug efficiencies, lower capital costs and the flexibility of deliv-
ering the laser with a flexible fibre instead of fixed optics, as is needed for carbon 
dioxide lasers. Further information about the physics and operation of lasers can be 
found in laser-welding textbooks (Dawes, 1992; Duley, 1999).
The automotive industry uses a keyhole mode for laser welding, which has a high 
power density, resulting in deep penetration and narrower welds. Therefore laser welding 
does not require any special joint preparation or addition of filler materials. High welding 
speeds are achieved in laser welding because of the high power density, which significantly 
increases the production rate. The combination of high power density and speed leads to 
lower heat input in laser welding, which minimizes metallurgical heterogeneities across 
the weldments, for example, extension of the fusion zone and heat-affected zone (HAZ). 
Lower heat input also reduces the thermal distortion of the workpieces, minimizing the 
machining requirements after welding. Laser welding allows the joining of sheets where 
access to only one side is possible, allowing greater flexibility of the joint design. Based on 
the above-mentioned advantages, mentioning that laser welding is a potential process for 
welding AHSS, especially for body-in-white applications, is worthwhile.
5.3 Laser welding of AHSS
Automakers can conveniently tailor the design of automotive parts through the combi-
nation of strength, formability and crashworthiness by incorporating AHSS in LWBs. 
Before the analysis of forming and crashworthiness of LWBs, however, various weld-
ability issues must be understood to design and produce high-quality parts with rea-
sonable production and tooling costs. In industry LWBs are manufactured in a keyhole 
laser welding mode; however, laboratory-based research also has examined welds 
made in a conduction mode to understand the metallurgical and mechanical behaviour 
of laser-welded AHSS (Biro, McDermid, Embury, & Zhou, 2010; Panda, Hernandez, 
Kuntz, & Zhou, 2009; Sreenivasan, Xia, Lawson, & Zhou, 2008; Xia, Tian, Zhao, & 
Zhou, 2008a; Xia, Tian, Zhao, & Zhou, 2008b). Therefore this chapter reports results 
of AHSS welded with both these laser welding modes.
73Laser welding of AHSS
5.3.1 Key issues
Laserwelding of AHSS involves many challenges, and the most important ones are 
described in this section. AHSS is welded in a butt weld configuration when manu-
facturing LWBs to eliminate the issues involved in forming and die design. In butt 
welding the zinc coating present on the AHSS would not be harmful, but if lap weld-
ing is used it would be harmful (Li, Lawson, Zhou, & Goodwin, 2007). However, 
butt welding requires precise fit-up and alignment of the workpieces because of the 
narrow beam size compared with other fusion welding processes. Any gap between 
the steel sheets may result in significant weld concavity and undercut, which degrades 
weld performance. In addition, misalignment of the workpieces produces a notch that 
reduces the fatigue life of the LWBs. Most industrial laser welding processes use 
keyhole mode welding because it gives deep penetration and fast welding speeds. The 
rapid fluid velocity, as high as 3000 mm/s (Zhao, White, & DebRoy, 1999), associated 
with keyhole mode laser welding and the instability of the keyhole often lead to rough 
and ropy bead surfaces. This can be an issue when considering the auto body parts 
where weld surface appearance is important, for example, in automotive applications 
where the welds are visible. However, a smooth and acceptable bead surface can be 
obtained by controlling the welding parameters.
In addition to these challenges, metallurgical issues such as softening (i.e. a 
decrease in the hardness relative to the base metal) have consistently occurred at the 
tempered region or subcritical HAZ of DP steel welds (Biro et al., 2010; Li et al., 
2013; Panda et al., 2009; Sreenivasan et al., 2008; Xia et al., 2008a; Xia et al., 2008b; 
Xu et al., 2012). The HAZ softening phenomenon occurs because the martensite 
phase in DP and martensitic steels is tempered; the extent of softening increases 
with the martensite content. Thus more HAZ softening generally occurs in DP steels 
with higher martensite volume fractions. The details of these steels are discussed in 
Sections 5.4.2 and 5.5.1.
5.4 Microstructure of laser-welded AHSS
As in all other welding processes, the microstructure of laser welds in AHSS are 
broadly made of three parts, namely, a base metal, a HAZ and a fusion zone. A typical 
example of laser-welded DP steel is illustrated in Figure 5.1(a). The microstructure of 
a DP steel base metal consists of ferrite grains and martensite islands elongated in the 
rolling direction (Figure 5.1(b)). The HAZ, in turn, is composed of three subparts based 
on the peak temperature experienced by the workpiece during welding and the critical 
transformation temperature of the AHSS. The three subparts of HAZ are a tempered 
region, or subcritical HAZ, where the peak temperature experienced during welding 
is below the Ac1 temperature of the steel (Figure 5.1(c)); the intercritical HAZ, where 
the peak temperature during welding is between the Ac1 and Ac3 temperatures of the 
steel (Figure 5.1(d)); and the supercritical HAZ, where the steel is heated above its 
Ac3 temperature (Figure 5.1(e)). In the fusion zone the peak temperature exceeds the 
melting point of the steel. The microstructure of the HAZ depends only on solid-state 
74 Welding and Joining of AHSS
transformations, whereas the microstructure of the fusion zone is based on both the 
solidification behaviour of the steel and the solid-state transformations as the fusion 
zone cools from melting temperature to room temperature.
The geometry of the fusion zone and the HAZ is dependent on the welding param-
eters and the type of laser welding. Table 5.1 compares the different laser welds based 
on the geometry of the fusion zone and the HAZ in DP980 steel (Xu et al., 2013). It 
should be noted that a 4-kW diode laser weld in a 1.2-mm-thick DP980 steel sheet 
has a very wide HAZ and fusion zone compared with other laser welding processes. 
Figure 5.1 Microstructure developed in fibre laser welding (6-kW power and welding at 
16 m/min) of DP980 (1.2-mm-thick) steel: weld profile (a), base metal (b), subcritical heat-affected 
zone (HAZ) (c), intercritical HAZ (d), supercritical HAZ (e), and fusion zone (f). F, ferrite; 
M, martensite; RD, rolling direction; TM, tempered martensite (Westerbaan et al., 2012).
75
L
aser w
elding of A
H
SS
Table 5.1 Comparison of different laser welding processes based on the size of the heat-affected zone 
(HAZ) and fusion zone formed in DP980 steel (Xu et al., 2013)
Welding 
type
Power 
(kW)
Welding 
speed (m/min)
Spot size 
(mm)
Thickness of 
the workpiece 
(mm)
Average 
width of the 
HAZ (µm)
Average 
width of the 
fusion zone 
(µm) Reference
Diode 4 1.6 12 × 0.5 1.2 4000 3000 Xia, Sreenivasan, 
Lawson, Zhou, 
and Tian (2007)
Nd:YAG 3 3 0.6 1.17 1000 750 Sreenivasan 
et al. (2008)
CO2 6 6 – 1.8 1000 1000 Kim, Choi, Kang, 
and Park 
(2010)
Fibre 6 16 0.6 1.2 250 450 Xu et al. (2013)
76 Welding and Joining of AHSS
This is because of the considerably larger laser beam spot size in diode laser weld-
ing, which produces a lower energy density, leading to conduction mode welding. By 
contrast, in fibre laser welding, the higher energy density causes keyhole mode weld-
ing, which is more efficient when compared with the conduction mode in diode laser 
welding; in this case the fibre laser welding could occur at higher speeds, resulting in 
a narrower weld.
5.4.1 Fusion zone
As mentioned previously, the fusion zone melts and solidifies during laser weld-
ing. Because of the high welding speeds and narrow weld widths, the cooling rates 
experienced in laser welding are very high (Gould, Khurana, & Li, 2006). Thus, the 
microstructure within the fusion zone consists of the constituents formed via nonequi-
librium solidification of the liquid base metal. Epitaxial solidification occurs in the 
fusion zone, starting at the fusion boundary (i.e. the area of contact between the weld 
pool and the unmelted substrate of the workpiece) and gradually growing towards the 
weld centre line to meet growing columnar grains from the opposite fusion boundary 
(Figure 5.1(a)). Epitaxial solidification involves the growth of the solid in the direc-
tion in which the grains at the fusion boundary are oriented. In AHSS the fusion zone 
microstructure and that of its constituents depend on two important factors: cooling 
rate and chemistry. The following sections discuss the influence of cooling rate and 
chemistry in detail.
The cooling rate in laser welding is directly related to the welding speed, that is, 
the higher the welding speed, the higher the cooling rate. However, the cooling rate 
in the laser welding process is commonly higher than the critical cooling rate to form 
the martensite phase in most AHSS, which has high hardenability because of its high 
alloying content. Thus martensite is commonly observed in the fusion zone of almost 
all AHSS laser welds. For example, Gu, Yu, Han, Li, and Xu (2012) reported that 
Nd:YAG laser welding (3 kW) of hot-stamped (martensitic) steel forms a fusion zone 
with only a martensite phase in the entire range of welding speeds (3.6–7.8 m/min). In 
addition, Xia et al. (2008b) also concluded that no significant variation in the fusion 
zone microstructure is expected when the welding speed increases from 1.2 to 2.2 m/
min in diode laser welding of TRIP steel. This was based on the observation that there 
was an insignificant increase in the fusion zone hardness with welding speed.
The fusion zone microstructure in laser-welded AHSS is strongly dependent on 
the carbon content. For example, it was recently reported that martensite content in 
the fusion zone microstructure in diode laser welding of AHSS decreased with car-
bon and alloying additions (Santillan Esquivel, Nayak, Xia, & Zhou, 2012). A mixed 
microstructure containing martensite, ferrite and bainite phases is formed in the fusion 
zone, containing less than 0.12 wt% carbon, which is also confirmed by the hardness 
values,which measured lower-than-predicted martensite hardness with similar carbon 
content (Santillan Esquivel et al., 2012). Carbon increases the hardenability of AHSS 
and shifts the continuous cooling transformation curve toward the right. In addi-
tion to carbon, other alloying additions in AHSS, for example, manganese, silicon, 
aluminium, chromium and molybdenum, also enhance the formation of martensite by 
77Laser welding of AHSS
retarding the kinetics of ferrite and bainite formation and increasing the hardenability 
of the steel. Therefore, because of the carbon and other alloying elements used in 
AHSS and the high cooling rates associated with laser welding, the fusion zone is 
usually highly martensitic.
TRIP steel is unique because it is generally alloyed with high amounts of either silicon 
or aluminium to delay carbide precipitation (De Cooman, 2004). Interestingly, microstruc-
ture in the weld pool of TRIP steel laser welds is reported to be strongly influenced by the 
alloying addition (Xia et al., 2008a). For example, in diode laser welding silicon-alloyed 
TRIP formed a fully martensite structure (Figure 5.2(a)) for the reasons discussed above. 
However, since aluminium is a ferrite stabilizer, it transformed the TRIP steel fusion zone 
into a mixed microstructure consisting of high-temperature ferrite (δ-ferrite) as a primary 
Si-alloyed
Al-alloyed
(a)
(b)
Figure 5.2 Fusion zone microstructure of silicon-alloyed (a) and aluminium-alloyed (b) 
transformation-induced plasticity steel in diode laser welding. F, ferrite (Xia et al., 2008b).
78 Welding and Joining of AHSS
phase, which subsequently transformed into side-plate ferrite (Figure 5.2(b)), martensite 
and bainite (Xia et al., 2008a). The solidification sequence in aluminium-alloyed TRIP is 
as follows. In the first stage δ-ferrite solidifies as the primary-phase dendrites and grows 
into the liquid. These dendrites are rich in aluminium, so the remaining liquid, which is 
depleted in TRIP with a lower aluminium content, solidifies to form stable austenite via 
a peritectic reaction. This sequence leads to a unique microstructure of δ-ferrite dendrites 
with interdendritic austenite, which upon further cooling decomposes to either marten-
site or bainite, depending on the welding speed, which in the range of 1.6–2.2 m/min 
leads to a cooling rate of 50–100 K/s (Xia et al., 2008a).
5.4.2 Heat-affected zone
In the HAZ of AHSS laser welds the temperature rises rapidly to a peak and then quickly 
cools again. The heating and cooling rates depend on the welding parameters and the 
distance from the fusion boundary. Based on the peak temperature, different transfor-
mations occur within the HAZ. In the supercritical HAZ the base metal microstructure 
changes to austenite during heating. Depending on how high above the Ac3 temperature 
of the steel the peak temperature is, grain growth may also occur. When the region of the 
HAZ with a temperature above the steel’s Ac3 temperature cools, it typically transforms 
into martensite because of the high hardenability of AHSS (Figure 5.1(e)). In the inter-
critical HAZ the base metal microstructure starts forming austenite, which nucleates at 
the grain boundaries. Again, upon cooling the austenite typically transforms into mar-
tensite and the ferrite remains unchanged. The volume fraction of martensite in this area 
of the HAZ increases as the peak temperature rises from the Ac1 to the Ac3 temperature. 
For example, the intercritical HAZ microstructure in DP steel contains fine martensite 
grains in a ferrite matrix (Figure 5.1(d)), which is very different from the base metal 
(Figure 5.1(b)). It should be noted that the intercritical HAZ microstructure in DP steel 
may resemble that of the base metal (Biro & Lee, 2004). The subcritical HAZ experi-
ences a peak temperature below the Ac1 line of the steel, wherein the martensite phase 
in the base metal is tempered (Figure 5.1(c)), causing the HAZ to soften or the hardness 
to drop below that of the base metal. Because of a decrease in tempering temperature, 
the severity of HAZ softening decreases with increasing distance from the Ac1 iso-
therm. Martensitic and DP steels are more prone to HAZ softening. For example, many 
researchers have reported the severity of HAZ softening of DP steels increases with 
increasing heat input and the steel grade (Biro et al., 2010; Xia, Biro, Tian, & Zhou, 
2008). HAZ softening has been reported as detrimental to the performance of laser 
welds, the details of which are discussed in Section 5.6.
5.5 Hardness
Hardness across the LWBs is determined by the corresponding microstructure, which 
has been discussed as being dependent on the welding parameters, steel chemistry and 
initial microstructure. The effects of these on the performance are discussed separately.
Welding parameters such as welding speed, power and laser spot size affect heat 
input, which has a large effect on the properties after welding. Figure 5.3 compares 
79Laser welding of AHSS
microhardness profiles across the DP980 LWBs made using different laser-welding param-
eters. The readers may note that HAZ softening (marked as ‘softening’ in Figure 5.3(c)) 
may occur in the outer part of the HAZ, irrespective of the laser welding process used. 
ν
ν
µ
ν
Figure 5.3 Hardness profile across the DP980 steel welds made using a diode laser at 
1.0 m/min (a), a neodymium:yttrium–aluminium–garnet laser at 3.0 m/min (Sreenivasan 
et al., 2008) (b), and a fibre laser at 16 m/min (Westerbaan et al., 2012) (c). BM, base metal; 
HAZ, heat-affected zone.
80 Welding and Joining of AHSS
However, the width of the soft zone decreases with increasing welding speed and 
decreasing beam width. As mentioned in earlier sections, HAZ softening has been asso-
ciated with tempering of the martensite phase of the base metal (Baltazar Hernandez, 
Panda, Kuntz, & Zhou, 2010; Biro et al., 2010; Xia et al., 2008). The hardness of the 
HAZ increases between the edge of the tempered region and the fusion boundary. This 
is related to an increase in the martensite volume fraction in the supercritical region of 
the HAZ (Figure 5.1(c) and (d)). The hardness of the HAZ smoothly merges with the 
fusion zone hardness, which shows a maximum value because the fusion zone experi-
ences the highest cooling rate in the weldment. The microstructure of the fusion zone 
is typically fully martensitic because of the high cooling rates experienced and high 
hardenability of AHSS (Figure 5.1(f)). It should be noted that minimal HAZ softening 
has been reported in TRIP steel welds, in which decomposition of retained austenite in 
the base metal occurs in the temperature below the Ac1 line, resulting in an increase in 
hardness (Xia et al., 2008a, 2008b).
The hardness of the fusion zone of AHSS strongly depends on its carbon con-
tent. Santillan Esquivel et al. (2012) carried out diode laser welding of several AHSSs 
(DP600, DP780, TRIP780) with similar and dissimilar combinations. Three different 
regions were identified when fusion zone hardness was plotted versus carbon con-
tent (weight percent) and compared with the theoretical martensite hardness (calcu-
lated from the carbon content), as depicted in Figure 5.4. Region I comprised the 
fusion zone, which has high carbon content, resulting in a completely martensitic 
structure and hardness similar or close to that of theoretical martensite hardness. A 
mixed microstructure of martensite and bainite was seen in region II, which resulted 
in slight decrease in hardness from the theoretical martensite hardness. In region III 
a significant diversion from the martensite hardness was noticed because AHSS with 
low carbon content (<0.1 wt%) formed a ferritic microstructure with little bainite 
and/or martensite (Santillan Esquivel et al., 2012).
Figure 5.4 Fusion zone hardness 
versus carbon content in laser 
welding of various advanced 
high-strength steels in similar and 
dissimilar combinations (SantillanEsquivel et al., 2012).
ν
81Laser welding of AHSS
5.5.1 Factors affecting HAZ softening
Laser welding heat input and the amount of alloying in the steel are two factors that 
influence HAZ softening and its characteristics in DP steels. With decreasing heat 
input there is less time available to complete the martensite tempering reaction in 
the base metal. It should be noted that the HAZ of laser welds may not fully temper. 
Therefore the time that the HAZ of laser welds are at the elevated temperatures do not 
allow for as much diffusion as processes using higher heat. However, higher heat input 
increases the time that the subcritical area of the HAZ is at elevated temperatures and 
hence results in more severe softening (Biro et al., 2010; Xia et al., 2008). A typical 
example of difference in martensite morphology in HAZ of one DP780 steel welded 
with low and high heat input is shown in Figure 5.5. Higher heat input has resulted in 
severe decomposition of the martensite grain shown, whereas at lower heat input there 
still remains a large fraction of untempered martensite in the subcritical HAZ.
Another recent study (Nayak, Hernandez, & Zhou, 2011) of the effects of chemistry 
on HAZ softening in DP steels concluded that when the same welding parameters are 
used the degree of softening is more for lean composition (DPL) steel when compared 
with moderate (DPM) and rich (DPR) steels. Note that DPL represents the steel with 
lower content of alloying elements (e.g. manganese, chromium and silicon), the con-
tents of which are higher in the DPR steel; for the DPM steel, their concentrations are 
between that of lean and rich steel. This was attributed to the severity of martensite 
decomposition (Figure 5.6(a)–(c)), which was suggested by the size of the precipi-
tated cementite particles (Figure 5.6(d)–(f)) in the tempered region of the welds. In 
the tempered region of HAZ, DPL steel (Figure 5.6(d)) formed coarser cementite par-
ticle and DPR steel (Figure 5.6(f)) formed a finer one with DPM steel (Figure 5.6(e)), 
forming cementite particles with a size between those formed in the DPL and DPR 
steels. It has been reported that the tempering characteristics in DP steel strongly 
depend on the martensite morphology in the base metal (Baltazar Hernandez, Nayak, 
& Zhou, 2011; Nayak et al., 2011). Therefore DPR steel, which contained a twinned 
structure of martensite because of higher martensite carbon content (0.36 wt%) in the 
Figure 5.5 Effect of heat input (low (a) and high (b)) on martensite tempering in DP780 steel 
(Biro et al., 2010).
82 Welding and Joining of AHSS
(a) (d)
(e)
(f)
(b)
(c)
Figure 5.6 Effect of chemistry on the severity of martensite tempering in DP980 steels. 
Microstructures showing tempered martensite in the subcritical heat-affected zone of lean 
composition dual-phase (DPL) (a), moderate composition dual-phase (DPM) (b), and rich 
composition dual-phase (DPR) steel (c) welds. The representative bright field images of the 
extracted cementite and corresponding selected area diffraction patterns in the inset images 
showing [010] zone axis of cementite from DPL (d), DPM (e), and DPR steels (f) (Nayak 
et al., 2011).
83Laser welding of AHSS
base metal, formed finer cementite, whereas other lower martensite carbon contents in 
DPL (0.273 wt%) and DPM steels (0.269 wt%) resulted in a martensitic lath structure 
(Nayak et al., 2011). It should be noted that although DPM steel contains similar car-
bon content as DPL steel, it has a higher amount of alloying elements, which retard the 
coarsening kinetics of the cementite (Chance & Ridley, 1981; Miyamoto, Oh, Hono, 
Furuhara, & Maki, 2007) formed during the martensite tempering reaction, which 
resulted in finer precipitates compared with those formed in DPL steel.
The effects of heat input during welding on the kinetics of the martensite tempering 
or HAZ softening has been studied recently (Biro et al., 2010; Xia et al., 2008). In their 
(Xia et al., 2008; Biro et al., 2010) study they compared the martensite decomposition 
with the heat input of the welding process. They have provided a modified formula 
for calculating heat input at the subcritical HAZ, that is, at the Ac1 temperature, which 
was based on Rosenthal’s solution for a moving line power source in a thin plate. The 
calculated heat input then was used to determine the time constant, that is, the time 
required to heat the material from an ambient to the Ac1 temperature (Eqn (5.1)).
 
τ =
1
4πeλρc
[Qnet/ (vd)]
2
(TAc1 − T0)
2
 (5.1)
In Eqn (5.1), Qnet is the laser power (watts), v is welding speed (millimetres/second), 
d is sheet thickness (millimetres), λ is the thermal conductivity (30 W/m/K), ρ is 
the steel density (7860 kg/m3), c is the specific heat capacity of steel (680 J/kg/K), 
TAc1 is the Ac1 temperature (Kelvin) and T0 is the ambient temperature (298 K). The 
term Qnet/(vd) is the heat input (more precisely, the thickness-normalized net energy 
absorbed per unit weld length), which was calculated from the difference in distance 
between the weld centre line to the Ac1 isotherm and the weld centre line to the fusion 
boundary (Xia et al., 2008). The plots of the HAZ softening kinetics measured from 
laser welds in various grades of DP steel are presented in Figure 5.7, which show the 
change in hardness increases with DP steel strength (martensite volume fraction), and 
softening kinetics increases with increasing martensite carbon content and decreases 
with increasing alloying with carbide-forming elements. For example, the change in 
hardness is much greater for DP780 than either DP600 or DP450 (Figure 5.7(a)). The 
rate of martensite decomposition increased with increasing carbon content and when 
similar amounts of carbide-forming alloying additions were used. However, the steel 
with less alloying showed a faster rate of decomposition for its respective martensite 
carbon content (Figure 5.7(b)). Also, one should note the degree of softening increased 
with heat input, that is, diode laser welding resulted in higher softening. Figure 5.7(c) 
shows an example indicating the effect of chemistry on HAZ softening in DP980 
steel. One can see that steel with rich composition (DPR) has higher resistance to 
softening compared with lean (DPL) and moderate (DPM) composition steels, which 
is attributed to the smaller degree of decomposition of martensite (Figure 5.6(c)) 
and finer cementite (Figure 5.6(f)) observed in the tempered region of the welds 
(Nayak et al., 2011). Therefore process parameters should be tailored to material 
chemistry and microstructure to make DP steel LWBs with minimum softening.
84 Welding and Joining of AHSS
5.6 Performance of laser-welded AHSS
5.6.1 Strength and durability
Hardening in the fusion zone and supercritical HAZ increases the strength of the weld-
ment, which in turn decreases the ductility when strained in a direction parallel to 
Figure 5.7 Comparison of the 
softening kinetics in dual-phase 
(DP) steels: the effect of grade 
(Xia et al., 2008) (a), the effects 
of DP steel chemistry and grade 
(Biro et al., 2010) (b), and the 
effect of chemistry (c) on the 
softening of DP980 steel welds 
(Nayak et al., 2011). C, carbon; 
Cr, chromium; DPL, lean com-
position dual-phase steel; DPM, 
moderate composition dual-phase 
steel; DPR, rich composition dual-
phase steel; Mo, molybdenum; 
Nd:YAG, neodymium:yttrium–
aluminium–garnet.
τ
φ
 τ
ν
85Laser welding of AHSS
the weld line. Conversely, HAZ softening reduces the local strength when the load-
ing direction is perpendicular to the weld line, which resulted in strain localization, 
leading to premature failure at the tempered region at low loads and elongations 
(Panda, Sreenivasan, Kuntz, & Zhou, 2008; Westerbaan et al., 2012; Xu et al., 2012). 
Figure 5.8(a) shows engineering stress–strain curves from tensile tests of the base metal 
and transverse to the laser weld in samples of DP980 steel (Sreenivasan etal., 2008). 
The yield strength and ultimate tensile strength (UTS) of the welded samples were 
lower than the base metal values with reduction in the overall specimen elongation. 
This was due to necking in the subcritical HAZ in all the welded specimens (Panda 
et al., 2008; Sreenivasan et al., 2008). The Nd:YAG laser welds had higher strength 
and ductility (elongation) compared with the diode laser welds because of a narrower 
tempered region and less severe HAZ softening associated with the Nd:YAG laser 
welding. Thus it is concluded that the transverse strength and ductility of laser-welded 
Figure 5.8 Comparison of (a) the tensile test plots (Sreenivasan et al., 2008) and (b) the 
S (Stress)–N (number of cycles) curves (Xu et al., 2012) of DP980 laser-welded blanks 
fabricated using different lasers. BM, base metal; DLW, diode laser welding; FLW, 
fibre laser welding; S, single linear weld; M, multiple linear welds; Nd:YAG, 
neodymium:yttrium–aluminium–garnet.
86 Welding and Joining of AHSS
DP980 is dependent on the properties of the tempered HAZ. For example, high weld-
ing speed achieved using fibre laser resulted in a narrower tempered zone (Xu et al., 
2012), which in turn gave rise to a 96% joint efficiency, that is, the ratio of the UTS 
of the laser welds and the base metal, whereas diode laser welding resulted in a lower 
joint efficiency because of the wider and softer tempered HAZ. The LWBs prepared 
by fibre laser welding also lead to improved fatigue life when compared with diode 
laser welding (Figure 5.8(b)), even with multiple linear welds present in the gauge 
length of the fatigue test coupons (Xu et al., 2012). For example, the fatigue strength 
at 1 × 107 cycles (sometimes called the conditional fatigue limit) was about 100 MPa 
lower for diode laser welds than the base metal, whereas it was even lower (∼150 MPa) 
in the low cycle fatigue region at 2 × 103 cycles (Figure 5.8(b)). On the other hand, 
fibre laser welds had a fatigue life close to that of the base metal at stress amplitudes 
above 300 MPa. However, the fatigue strength became lower and more scattered at 
stress amplitudes below 300 MPa. This suggested that the narrower tempered zone in 
fibre laser welding did not affect the tensile properties of the weldment, but fatigue 
resistance is susceptible to HAZ softening, irrespective of the size of the tempered 
zones. The fatigue data for the multiple linear welds (Figure 5.8(b)) exhibited a larger 
scatter and lower fatigue strength, indicating that the probability of dynamic fatigue 
failure at lower stress amplitudes increased with an increasing number of tempered 
zones. It should be noted that the failure location in both tensile and fatigue testing of 
the DP980 steel was at the tempered region of the HAZ, irrespective of the laser weld-
ing type (Sreenivasan et al., 2008; Westerbaan et al., 2012; Xu et al., 2012).
5.6.2 Formability
The formability of AHSS is reduced significantly after welding, which makes forming 
LWBs a challenging task. The forming behaviour of an LWB can be predicted by 
understanding a few important points: (1) material property changes occur in the HAZ 
of the weld; (2) non-uniform deformation occurs because of differences in thickness, 
properties and/or surface characteristics; (3) the effects of the welded zone on the 
strain distribution, failure site and crack propagation; and (4) how the weld line moves 
during the forming process. Among these complexities, the welding process plays the 
major role. Factors arising in laser welding processes that influence formability can 
be classified into four categories, namely, the type of laser, welding parameters, prop-
erties of the base materials, and changes in material properties of the weld and HAZ). 
This section discusses the effects of these factors on the formability of AHSS LWBs.
Formability properties, such as hardness, tensile strength and fatigue strength depend 
on heterogeneity in the microstructure across the LWBs and is represented by the lim-
ited dome height (LDH) obtained in formability tests. In general, formability can be 
related to the hardness and strength of a weld. For instance, Sreenivasan et al. (2008) 
observed a significant decrease in the LDH of welds due to HAZ softening in the 
subcritical HAZ; more severe HAZ softening led to larger reductions in the LDH. 
Furthermore, when the location of the failure in LDH samples was analysed, that fail-
ure always occurred in the tempered HAZ. Figure 5.9 shows the relation between the 
LDH and softening in DP980 laser welds. Larger reductions in hardness led to lower 
87Laser welding of AHSS
formability of the welded blanks. The formability of the diode welds was less than that 
of the Nd:YAG welds because the tempered zone in the diode welds is wider than in 
the Nd:YAG welds (Figure 5.9). With an increase in the welding speed, the formability 
of the welded DP steel samples approached that of the base metal. These results indi-
cate that when using higher power densities and higher speed keyhole welding mode, 
there is narrower, less softened, tempered HAZ, which leads to better formability. 
Therefore, for DP steel, it is better to weld with an Nd:YAG laser, and probably a fibre 
laser, in keyhole mode and at the maximum achievable welding speed.
No significant difference in the formability of the DP980 steel was observed with 
respect to welding orientation relative to the rolling direction or how the weld was 
positioned relative to the punch (on the face or root side of the weld), as HAZ soften-
ing dominated the formability behaviour (Sreenivasan et al., 2008). Another study of 
DP800 showed that the formability of equal-thickness LWBs was 20% lower than the 
base metal because of microstructural changes and an increase in microhardness of 
the fusion zone and HAZ (Wu, Gong, Chen, & Xu, 2008). It also was noted that when 
samples were stretched parallel to the weld, cracks nucleated and propagated normal to 
the weld line (Saunders & Wagoner, 1996), whereas when samples were stretched per-
pendicular to the weld, failure occurred in the weaker materials (Panda, Li, Hernandez, 
Zhou, & Goodwin, 2010; Sreenivasan et al., 2008).
The forming behaviour changes when different material combinations and weld 
line positions are used in LWB fabrication. Figure 5.10 shows the change in the LDH 
with weld line positions for DP600 (1.2 mm)–HSLA (1.14 mm) and DP980 (1.2 mm)–
HSLA (1.14 mm) LWBs. The welded samples always had a lower LDH compared 
with the parent metals, which is attributed to the presence of the tempered HAZ and 
the difference in material properties within the blank, which induced nonuniform 
Figure 5.9 Formability of the laser-welded blanks versus the reduction in weld metal 
hardness in DP980 steel (Sreenivasan et al., 2008). LDH, limited dome height; Nd:YAG, 
neodymium:yttrium–aluminium–garnet.
88 Welding and Joining of AHSS
deformation. Interestingly, with larger differences in parent material properties, lower 
formability was observed because of higher nonuniformity in the deformation during 
stretch forming (Panda et al., 2010). For instance, the LDH of DP980–HSLA was 
lower than that of the DP600–HSLA combination. The lower LDH values were mea-
sured for DP600–HSLA LWBs when welds were positioned at −15, 0 and +15 mm 
from the punch pole, whereas the LDH increased (27–33%) when the weld was placed 
−30 mm away from the punch pole. Similarly, for DP980 (1.2 mm)–HSLA (1.14 mm) 
LWBs, the LDH was lower when the weld line was positioned +15 mm from the punch 
pole, and it increased by (∼150%) when the weld was placed −30 mm away from the 
punch pole. This suggested that the position of the weld line has a strong influence on 
the forming behaviour of dissimilar LWBs, and the formability can be increased by 
keeping the weld line away from the punch. However, the level of increase in form-
ability with respect to weld line position (either toward the positive side or negativeside) depends on the material combination. It was also noted that the load progression 
curve for an LWB is between that of the parent metals, and the slope of the curve 
depends on the amount of each parent metal in the LWB. Interestingly, the percentage 
of elongation from a uniaxial tensile test does not reflect the same trend as that of LDH 
for LWBs (Panda et al., 2010). Uniaxial tensile tests could not predict the actual press 
performance of LWBs. The strain distribution profile across the LWBs generally cor-
relates well with the LDH and the failure location (Panda et al., 2010). It is interesting 
to note that the failure location in the LDH tests of the dissimilar LWBs were mostly 
in the HSLA base metal because its hardness was even lower than that of the tempered 
HAZ of the DP600 and DP980.
Weld line geometry also plays an important role in the forming behaviour of 
LWBs. A recent study by Li et al. (2013) of the effects of weld line geometry and posi-
tion on DP (1.2-mm-thick) steel and HSLA (1.14-mm-thick) steel LWBs reported 
that the hardness across the welds is correlated to predict the failure location and the 
Figure 5.10 Comparison of formability (limited dome height) of laser-welded blanks with 
different weld line positions (Panda et al., 2010). DP, dual phase; HSLA, high strength low 
alloy; TWB, tailor welded blank.
89Laser welding of AHSS
LDH values. The formability is dependent on the weld line position and increases 
when the weld is located farther from the blank centre because of the develop-
ment of more uniform strain during LDH tests (Li et al., 2013; Panda et al., 2010). 
The curvilinear welds form an inconsistent extension of the HAZ on either side of 
the weld line; more severe HAZ softening, the difference in the base metal hardness 
and minimum hardness at the HAZ, is observed at the inner region of the curvilin-
ear welded blanks (Li et al., 2013). The effect of weld line geometry on formability 
is insignificant for DP980 and HSLA steels because the effect of HAZ softening 
dominated in DP980 steel and laser welding does not alter the HSLA steel. A 
strong correlation between HAZ softening and failure location in DP980 steel is 
observed; fracture consistently occurs in the soft zone located 3–5 mm away from 
the weld centre line (Figure 5.11). In general, predicting the failure location in cur-
vilinear welded DP steels is easier compared with HSLA because fracture always 
occurred in the soft zone at the inner region of the curve (Li et al., 2013). Weld line 
geometry typically has a stronger influence on the formability of lower-grade DP 
steel. For instance, the strain distribution profiles indicate that formability of only 
DP600 LWBs is affected significantly by weld line geometry, whereas both HSLA 
and DP980 showed comparable strain profiles in the linear and curvilinear welds 
(Li et al., 2013).
Figure 5.11 Failure locations in formed DP980 laser-welded blanks with different weld 
locations of linear welds: 0 mm (a), 15 mm (b) and 30 mm from centre (c); and curvilinear 
welds: 0 mm (d), 15 mm (e) and 30 mm from centre (f) (Li et al., 2013).
90 Welding and Joining of AHSS
5.7 Future trends
Although a considerable amount of work on the laser welding of AHSS has been 
reported so far, there is a lack of study on the entire AHSS family (namely TWIP, CP 
and martensitic steels) in relation to the effect of laser welding on microstructure, ten-
sile properties, fatigue life and formability. Furthermore, with the increased inclusion 
of fully martensitic press-hardened steels in applications such as B pillars and rails, 
open literature on the impact performance of these joints is critical. However, these 
applications have further challenges because the part welds will be hot formed; there-
fore the welds will undergo further transformations. What effect heat treating and die 
cooling have on the laser-welded microstructure (which will be very different from the 
base material) or the mechanical properties of the final part are unknown. In the area 
of welding process more work must be done to develop techniques to minimize weld 
concavity observed in AHSS because severe concavity reduces the strength (espe-
cially the fatigue resistance) of LWBs. Finally, research to understand how HAZ soft-
ening affects the properties of weldments after welding, and not just welded coupons, 
needs to be carried out. This is especially true because there are more applications 
where laser welding is being used for assembly applications; in these cases the weld 
zone is a part of the load path in a potential crash situation.
References
Auto/Steel Partnership. (1995). Tailor welded blank design and manufacturing manual: Techni-
cal report. Available from http://www.a-sp.org/publication.htm. Accessed 11.10.12.
Baltazar Hernandez, V. H., Nayak, S. S., & Zhou, Y. (2011). Tempering of martensite in dual-
phase steels and its effects on softening behavior. Metallurgical and Materials Transactions 
A, 42A, 3115–3129.
Baltazar Hernandez, V. H., Panda, S. K., Kuntz, M. L., & Zhou, Y. (2010). Nanoindentation 
and microstructure analysis of resistance spot welded dual phase steel. Materials Letters, 
64(2), 207–210.
Bhadeshia, H., & Honeycombe, R. (2006). Steels: Microstructure and properties. Oxford: 
Cambridge.
Biro, E., & Lee, A. (2004). Welded properties of various DP600 chemistries. In: Proceedings of 
Sheet Metal Welding Conference XI, May 11–14, Sterling Heights, MI, paper 5–2.
Biro, E., McDermid, J. R., Embury, J. D., & Zhou, Y. (2010). Softening kinetics in the subcriti-
cal heat-affected zone of dual-phase steel welds. Metallurgical and Materials Transactions 
A, 41A, 2348–2356.
Blank, W. (1997). Cold-rolled, high-strength sheet steels for auto applications. JOM, 48, 26–30.
Chance, J., & Ridley, N. (1981). Chromium partitioning in isothermal transformation of a eutec-
toid steel. Metallurgical Transactions A, 12A, 1205–1213.
Dawes, C. (1992). Laser welding. Toronto: McGraw-Hill.
De Cooman, B. D. (2004). Structure–properties relationship in TRIP steels containing carbide-free 
bainite. Current Opinion in Solid State and Materials Science, 8(3–4), 285–303.
De Cooman, B. C., Kwon, O., & Chin, K.-G. (2012). State-of-the-knowledge on TWIP 
steel. Materials Science and Technology, 28(5), 513–527.
http://www.a-sp.org/publication.htm
91Laser welding of AHSS
Duley, W. W. (1999). Laser welding. New York: John Wiley and Sons.
Gan, W., Babu, S., Kapustka, N., & Wagoner, R. (2006). Microstructural effects on the spring-
back of advanced high-strength steel. Metallurgical and Materials Transactions A, 
37A(11), 3221–3231.
Gould, J. D., Khurana, S. P., & Li, T. (2006). Predictions of microstructures when welding auto-
motive advanced high-strength steels. Welding Journal, 85, 111s–116s.
Gu, Z., Yu, S., Han, L., Li, X., & Xu, H. (2012). Influence of welding speed on microstructures 
and properties of ultra-high strength steel sheets in laser welding. ISIJ International, 52(3), 
483–487.
Kim, C. H., Choi, J. K., Kang, M. J., & Park, Y. D. (2010). A study on the CO2 laser welding 
characteristics of high strength steel up to 1500 MPa for automotive applications. Journal 
of Achievement in Materials Manufacturing Engineering, 39, 79–86.
Kusuda, H., Takasago, T., & Natsumi, F. (1997). Formability of tailored blanks. Journal of 
Materials Processing and Technology, 71, 134–140.
Li, X., Lawson, S., Zhou, Y., & Goodwin, F. (2007). Novel technique for laser lap welding of 
zinc coated sheet steels. Journal of Laser Applications, 19(4), 259–264.
Li, J., Nayak, S. S., Biro, E., Panda, S. K., Goodwin, F., & Zhou, Y. (2013). Effect of weld line 
position and geometry on the formability of laser welded high strength low alloy and dual-
phase steel blanks. Materials and Design, 52, 757–766.
Miyamoto, G., Oh, J. C., Hono, K., Furuhara, T., & Maki, T. (2007). Effect of partitioning of 
Mn and Si on the growth kinetics of cementite in tempered Fe-0.6 mass% C martensite.Acta Materialia, 55, 5027–5038.
Nayak, S. S., Baltazar Hernandez, V. H., & Zhou, Y. (2011). Effect of chemistry on nonisothermal 
tempering and softening of dual-phase steels. Metallurgical and Materials Transactions 
A, 42A, 3242–3248.
Panda, S. K., Baltazar Hernandez, V. H., Kuntz, M. L., & Zhou, Y. (2009). Formability analysis 
of diode-laser-welded tailored blanks of advanced high-strength steels. Metallurgical and 
Materials Transactions A, 40A, 1955–1967.
Panda, S. K., Li, J., Hernandez Bultazar, V. H., Zhou, Y., & Goodwin, F. (2010). Effect of weld 
location, orientation, and strain path on forming behavior of AHSS tailor welded blanks. 
Journal of Engineering Materials and Technology, 132, 0410031–0410111.
Panda, S. K., Sreenivasan, N., Kuntz, M. L., & Zhou, Y. (2008). Numerical simulations and 
experimental results of tensile test behavior of laser butt welded DP980 steels. Journal of 
Engineering Materials and Technology, 130, 0410031–0410039.
Santillan Esquivel, A., Nayak, S. S., Xia, M. S., & Zhou, Y. (2012). Microstructure, hardness 
and tensile properties of fusion zone in laser welding of advanced high strength steels. 
Canadian Metallurgical Quarterly, 51(3), 328–335.
Saunders, F. I., & Wagoner, R. H. (1996). Forming of tailored-welded blanks. Metallurgical and 
Materials Transactions A, 27A(9), 2605–2616.
Shi, M. F., Thomas, G. H., Chen, X. M., & Fekte, J. R. (2002). Formability performance 
comparison between dual phase and HSLA steels. L&SM, 27–32.
Sreenivasan, N., Xia, M., Lawson, S., & Zhou, Y. (2008). Effect of laser welding on formability 
of DP980 steel. Journal of Engineering Materials and Technology, 130, 0410041–0410049.
Uchihara, M., & Fukui, K. (2006). Formability of tailor welded blanks fabricated by differ-
ent welding processes: study of tailor welded blanks using automotive high-strength steel 
sheets (1st report). Welding International, 20, 612–621.
Westerbaan, D., Nayak, S. S., Parkes, D., Xu, W., Chen, D. L., Bhole, S. D., et al. (2012). 
Microstructure and mechanical properties of fiber laser welded DP980 and HSLA steels. 
Livonia: American Welding Society.
92 Welding and Joining of AHSS
World Auto Steel. (2009). Advanced high strength steels application guidelines; version 4.1. 
Available from http://www.worldautosteel.org. Accessed 28.11.12.
Wu, Q., Gong, J., Chen, G., & Xu, L. (2008). Research on laser welding of vehicle body. Optics 
and Laser Technology, 40(2), 420–426.
Xia, M., Biro, E., Tian, Z., & Zhou, Y. N. (2008). Effects of heat input and martensite on HAZ 
softening in laser welding of dual phase steels. ISIJ International, 48, 809–814.
Xia, M., Sreenivasan, N., Lawson, S., Zhou, Y., & Tian, Z. (2007). A comparative study of form-
ability of diode laser welds in DP980 and HSLA steels. Journal of Engineering Materials 
and Technology, 129, 446–452.
Xia, M., Tian, Z., Zhao, L., & Zhou, Y. N. (2008a). Fusion zone microstructure evolution of 
Al-alloyed TRIP steel in diode laser welding. Materials Transactions, 49(4), 746–753.
Xia, M., Tian, Z., Zhao, L., & Zhou, Y. N. (2008b). Metallurgical and mechanical properties of 
fusion zones of TRIP steels in laser welding. ISIJ International, 48(4), 483–488.
Xu, W., Westerbaan, D., Nayak, S. S., Chen, D. L., Goodwin, F., Biro, E., et al. (2012). Micro-
structure and fatigue performance of single and multiple linear fiber laser welded DP980 
dual-phase steel. Materials Science and Engineering A, 553, 51–58.
Xu, W., Westerbaan, D., Nayak, S. S., Chen, D. L., Goodwin, F., & Zhou, Y. (2013). Tensile and 
fatigue properties of fiber laser welded high strength low alloy and DP980 dual-phase steel 
joints. Materials and Design, 43, 373–383.
Zhao, H., White, D. R., & DebRoy, T. (1999). Current issues and problems in laser welding of 
automotive aluminium alloys. International Materials Reviews, 44(6), 238–266.
http://www.worldautosteel.org
Welding and Joining of Advanced High-Strength Steels (AHSS). http://dx.doi.org/10.1016/B978-0-85709-436-0.00006-0
Copyright © 2015 Elsevier Ltd. All rights reserved.
High-power beam welding of 
advanced high-strength steels 
(AHSS)
L. Cretteur
ArcelorMittal R & D, Automotive Application Research Center, Montataire, France
6
6.1 Introduction
The automotive industry is facing the complex challenge of improving passen-
ger safety while reducing fuel consumption and maintaining acceptable cost and 
comfort for the user. Many different solutions to address these issues are being 
developed. Among them, the use of high-strength steel (HSS) allows mass to be 
reduced (and therefore the fuel consumption to be reduced) by decreasing the 
sheet thickness applied while keeping the same mechanical behavior. The assem-
bly of different parts into a global car body, however, becomes one key issue. 
In mild steel automotive structures, the weld is usually stronger than the base 
metal and is not considered to be a factor limiting part integrity. In HSS structures 
the weld may become the weak point of the assembly. Increasing the alloying 
content of HSS leads to the formation of very hard metallurgical microstructures 
in the welds, which may limit weld ductility and toughness. As the base metal 
becomes stronger, the welds can carry higher loads. Therefore validation of the car 
body design must integrate specific weld properties, not only base metal proper-
ties. Furthermore, weld behavior must be improved to reach the target properties 
of the welded component.
Laser welding applications in the automotive industry has increased tremen-
dously since the early 1990s. During the 1990s, laser welding was mainly used 
to produce laser-welded blanks (LWBs), which were made using carbon dioxide 
(CO2) lasers. In spite of their high cost and limited flexibility, CO2 lasers were par-
ticularly adapted to the production of LWBs for the automotive industry. Indeed, 
the large volumes of parts requested by the automotive industry could make the 
laser welding cost-efficient because of its high welding speed and quality, result-
ing in a productive process with low scrap rates. Conversely, laser welding on 
the complete body in white (BIW) was rare during this decade, mainly because 
of the low flexibility of CO2 lasers in three-dimensional applications. Once high-
power lasers with increased flexibility and optic fibers came to the market, new 
automotive applications became possible. Laser welding and brazing then became 
major processes, competing with the traditional resistance spot welding process 
used by many car makers. Indeed, laser welding has unique qualities that make it 
94 Welding and Joining of AHSS
particularly adapted to the requirements of the automotive industry. Some of these 
characteristics include the following:
 • High welding speed: a high production rate is a requirement in the automotive industry. 
Lasers may reach welding speeds of several meters per minute, allowing a large number of 
welds in a short time.
 • High quality of the welded components: because of the highly concentrated energy of the 
laser beam, very limited distortions of the final parts occur. The welded components do not 
need to be reworked after welding and are “ready to use.”
 • Versatility in terms of weld geometry and accessibility: laser welding needs access to only a 
single side, and many different weld shapes can be made.
 • Adaptation to mass production: laser welding is highly repeatable and, with proper auto-
mation, may be easily automated. This is in line with the production needs of automotive 
manufacturing.
During the same period of time, automotive designs also included massive use of 
advanced high-strength steel (AHSS). Various AHSS families were introduced and 
used successfully to meet passive safety and lightweight design requirements. While 
until the 1990s car body design was mainly based on low-carbon and high-strength, 
low-alloy (HSLA) grades, AHSSs typically have increased carbon and alloy content, 
which needs to be accounted forwhen designing weldments and welding procedures. 
Currently, AHSS with tensile strengths in the range of 600–1500 MPa, with various 
chemistries and microstructures, are commonly used. In terms of weldability, how-
ever, different issues remain.
 • The chemical composition of automotive AHSS is characterized by increased alloying com-
pared with the traditional deep drawing or HSLA grades. In particular, carbon and man-
ganese are frequently added to increase the strength of the base metal and promote the 
formation of strengthening phases. Considering the high production rates imposed by the 
automotive industry, and particularly the high welding speed of laser welding, these chem-
istries have a tendency to form hard welds with martensitic and bainitic microstructures 
because of their highly hardenable chemistries.
 • Many AHSS have multiphase microstructures and have a high martensite content (nearly 
100% for the highest steel grades). Martensite within these structures tempers in the tem-
pered area of the heat-affected zone (HAZ). This results in local areas of the microstructure 
that are softer than the surrounding base material.
 • The influence of weldments comprising various areas containing hard and soft microstruc-
tures on the final in-use properties of the joint needs to be clarified.
The aim of this chapter is to give an overview of the different issues occurring 
during laser welding of AHSS in BIW applications. The influence of the laser weld-
ing process on the local metallurgy in the fusion zone and in the HAZ is presented. 
Finally, because welds are designed to transfer loads within a structure, the mechan-
ical properties of AHSS laser welds are highlighted. Most of the discussion on weld 
strength is concerned with single welds; however, some results of how welds function 
globally within a structure also are presented.
95High-power beam welding of AHSS
6.2 Back to basics: fundamentals of high-power beam 
welding
6.2.1 Principles of the keyhole welding process
The aim of this section is not to detail the physics of the laser welding process but to 
recall basics to facilitate the understanding of the following sections. The interested 
reader can find more details about the fundamentals of laser welding in the literature 
(AWS welding handbook, 2007; Mazumder, 1993a,b).
6.2.1.1 Keyhole formation
High-power beam welding is mainly applied by two different energy sources: high-
power lasers and electron beams. In both cases the interaction between the beam and 
the steel sheet is relatively similar. The energy from the beam is concentrated on a 
small surface (the so-called spot) to obtain a very high energy density. The key factor is 
to obtain, by focusing the beam, an energy density in excess of 106 W/cm2. This is the 
equivalent of a 2-kW beam focused to a 0.5-mm diameter spot, which may easily be 
achieved by various high-power laser systems. When energy densities above 106 W/cm2 
are achieved, the metal at the impingement point is heated well above its melting 
temperature and vaporizes to form a capillary filled with metallic vapors (so-called 
deep-penetration mode or keyhole welding mode). The keyhole allows the beam to 
deeply penetrate inside the material, resulting in a deep and narrow weld (typically 
1–2 mm wide). For comparison, in gas tungsten arc welding or gas–metal arc welding 
the energy densities at the workpiece are on the order of 104 W/cm2. With this lower 
energy density, the heat transferred from the source to the sheet is capable of melting 
only the sheet surface. Heat is transferred farther into the material by conduction. 
As a result, the weld tends to be relatively wide (centimeter scale) and have a lower 
penetration.
A deep-penetration welding mode may be applied by a variety of laser 
sources delivering both continuous and pulsed beams. Most laser applications 
in the automotive industry today use continuous-wave lasers. The laser source 
may be a CO2 type or based on a solid-state active medium such as neodym-
ium:yttrium–aluminum–garnet (Nd:YAG) crystals, fiber, or diodes. While the 
laser source significantly influences production issues by determining the wave-
length (affecting the choice of optics, shielding gases, potential use of optic 
fiber, and health and safety issues), the basic thermal mechanisms (formation 
of the molten pool) and consecutive metallurgical transformations are not sig-
nificantly affected by the type of laser as long as the keyhole welding mode is 
used. The metallurgical phenomena described hereafter concern welding with 
all laser sources and, by extension, electron beam welding because the electron 
beam welding process is also characterized by a transfer of the beam energy to 
the metal through a keyhole.
96 Welding and Joining of AHSS
6.2.1.2 Welding in keyhole mode
Welding in keyhole mode occurs by translating the keyhole, which results from the 
laser–metal interaction, at a high speed along the metal surface. As a result, a fusion 
line is obtained. Depending on the design needs, both linear and curved weld lines 
can be easily obtained thanks to the flexibility of modern laser cells. The solid metal 
in front of the beam is melted by the beam, flows around the keyhole wall, and solid-
ifies at the rear of the keyhole (Figure 6.1). The final fusion zone is a mixture of both 
materials being welded.
Once the laser beam source has been selected (CO2, Nd:YAG, fiber, diode), the 
following main factors influence the weld quality:
 • Heat input can be expressed as the ratio between the beam power and the welding speed. 
Heat input strongly influences weld depth and width. In most automotive laser welding 
applications the maximum power capacity of the laser (typically 3–8 kW) is used to weld, 
and the speed is adjusted to obtain full penetration. Depending on the sheet thickness and 
available power, typical welding speed ranges from 1 to 10 m/min. In fully penetrated welds 
the keyhole is open at the bottom of the weld. Because of travel speed, it was also observed 
that the keyhole is not vertical but is instead elongated and inclined (Fabbro, 2002; Fabbro, 
Slimani, Coste, & Briand, 2005; Pan & Richardson, 2011).
 • Spot size mainly influences the weld width. The spot size depends on both intrinsic beam 
properties and the choice of optics used in the process and represents the sharpness of the 
welding tool. The highly concentrated energy from a small spot increases the energy density, 
increasing the ability to create the keyhole and to penetrate deeper into the sheet. A small 
spot (diameter <0.4 mm), however, also represents a very sharp and sensitive tool, which 
may be destabilized by misalignments or gaps between the sheets to be welded. As a result, 
most automotive applications use spots diameters ranging from 0.4 to 0.7 mm (Brockmann, 
2010; Kielwasser, 2009; Larsson, 2007).
 • Joint geometry. Because laser welding is a contactless process, it is a very versatile tool 
that may be applied to various joint configurations. The automotive industry typically 
welds in the butt, lap, and overlap joint configurations; however, other joint designs may 
also be used.
Figure 6.1 Principle of fluid flow in high-power beam welding.
97High-power beam welding of AHSS
6.2.1.3 Influence of a metal’s physical properties on weldability
The formation of the weld is the result of the interaction of the beam with the metal; 
therefore it is clear that the physical properties of the base metal significantly influence 
the laser welding process. Keyhole formation results from the local melting and vapor-
ization of the metal caused by a highly concentrated energy. It is therefore obvious 
that the melting and vaporization temperatures of the workpiece are very important 
properties during laser welding. The reflectivity of the metal dictates the amount of 
energy that is reflected away from the material surface during laser impingement. This 
affects the amount of remainingenergy that is absorbed by the workpiece. It must be 
noted that reflectivity is a function of the beam wavelength and material temperatures 
(reflectivity decreases with temperature). Finally, heat conductivity determines how 
quickly heat flows from the weld. More heat is needed to weld materials with high 
thermal conductivity to compensate for the heat lost to the surrounding material.
Considering the current range of AHSS (complex-phase, dual-phase, transformation- 
induced plasticity, and press-hardened steels), the physical properties do not vary 
significantly from one product to the other (Table 6.1). The slight differences in physi-
cal properties of AHSS products do not significantly influence keyhole formation and the 
process efficiency. The variations of physical properties influence the process only when 
different classes of materials are compared (e.g., comparing AHSS with stainless steel).
6.2.2 Thermal cycle in laser welding
The keyhole deep-penetration welding process is characterized by a low heat 
input and a high welding speed. As a result, the weld cools very quickly. 
Measuring the cooling rate of laser welds is relatively complex because the weld 
and its corresponding HAZ are very narrow. However, the use of thin thermocouples 
allows the cooling rate to be evaluated at different positions in the HAZ. An 
example of an experimental measurement of cooling rate is given in Figure 6.2. 
Thermocouples were positioned on the surface of the sheet at different distances from 
the sheet edges before welding in a butt joint configuration. The sheets used were 
1.5-mm thick, and the welding was done using a 4-kW Nd:YAG laser at 5 m/min. 
Table 6.1 Typical properties of various materials
DP600 DP980 CP800
Aluminum 
alloy (6061)
Stainless 
steel 
(austenitic)
Thermal 
conductivity 
(W/m °K)
37 36 42 160 15–17
Melting 
temperature 
(°C)
1483 1473 1454 585 1500
Thermal conductivity and melting temperature data are from internal ArcelorMittal projects.
98 Welding and Joining of AHSS
Using this experimental setup, the HAZ was exposed to very rapid heating followed by 
rapid cooling, without a hold time at a high temperature. In this case the cooling 
rate (Δ700–300) was approximately 250 °C/s. The cooling rate in any case depends 
on the specific welding parameters, but this example shows the order of magnitude 
of typical cooling rates in automotive laser welding applications. It was also noted 
that similar cooling rates were observed at different locations away from the fusion 
line. This means that although various positions in the HAZ differ in terms of the 
maximum temperature reached during welding, cooling rate and hold time did not 
change as a function of distance from the weld. From a metallurgical point of view, 
the laser welding process can be compared to ultrafast heating instantaneously fol-
lowed by a rapid quench. The metallurgical modifications that are induced by the 
laser welding process are driven mainly by the maximum temperature reached at 
each point within a very limited time. Because of this, diffusion-based metallurgical 
phenomena rarely occur or do so only at very short distances.
6.2.3 Material flow in the weld
As presented in Figure 6.1, a laser weld line is obtained by moving the keyhole across 
the sheet surface. The metal is melted in front of the laser beam and flows around the 
keyhole. A turbulent weld pool collects behind the keyhole, leading to a mixing of 
both constitutive materials A and B in the weld before solidification occurs. Because 
of the fast cooling rate and corresponding fast solidification, the weld zone is not 
homogeneously mixed in spite of the turbulent pool. This is observed very clearly on 
overlap-welded joints. Figure 6.3 illustrates overlap laser welds made of mild steel 
(1.5 mm) and DP1180. It is clear from the micrograph that the microstructures of 
the weld metal in the top and bottom sheets are not identical. The difference in these 
microstructures is due to local differences in chemical composition. Indeed, during 
Figure 6.2 Thermal gradient during laser welding.
ArcelorMittal internal report.
99High-power beam welding of AHSS
keyhole welding, the base metal melted in front of the keyhole tends to flow mainly 
along a horizontal plane toward the rear of the weld, where it solidifies, allowing only 
limited time for mixing.
In butt welds, the mixture of both materials A and B being welded is much better; 
however, even butt welds do not have homogenous weld zones. The weld illustrated 
in Figure 6.4 is made of two different steel grades with significantly different chem-
ical compositions. A TRIP780 grade (1.5-mm thick) with a high manganese content 
(1.6%) was welded to a deep drawing mild steel (1.8-mm thick) containing only 0.1% 
manganese. Welding was performed with a 4-kW Nd:YAG laser at a welding speed 
of 3.5 m/min. When the weld was etched using nital, the fusion zone appeared to 
be relatively homogeneous. However, the use of other etchants (picric acid solution; 
Figure 6.4b) reveals heterogeneities. Further chemical analysis then was done using a 
microprobe on a scanning electron microscope. From the microprobe analysis it is clear 
that the manganese content in the weld is not homogeneous. A higher concentration of 
manganese is observed on the TRIP780 side of the fusion zone. Although segregation 
was observed, the influence of mixing was also observed as a gradient of composition 
across the weld. The lack of mixing is due to the rapid cooling during laser welding. 
In spite of a turbulent flow within the weld pool, the fusion zone solidified before 
homogeneous mixing occurred. The areas with various compositions illustrate the 
fluid flow occurring in the weld pool.
Figure 6.3 Segregation in overlap welds of mild steel (top) with DP1180 (bottom).
ArcelorMittal internal report.
100 Welding and Joining of AHSS
The average weld composition can be considered as the average composition of 
both constitutive materials, but this assumption is not true locally. Very strong gradi-
ents can be observed at a microscopic scale. Local chemical composition represents a 
mixture of both constitutive materials in various proportions. Similar phenomena were 
reported by Mujica, Webera, Pinto, Thomy, and Vollersten (2010) on laser welds made 
with iron–manganese alloys with high manganese content.
The lack of homogeneity of the solidified laser weld pool is not a large problem. 
However, it explains why microhardness measurements across welds may exhibit 
some scatter, as areas with various compositions may be indented. Also, guidelines or 
conclusions drawn about the average composition of the weld must be considered with 
care because the local composition in the weld may differ from the nominal average 
composition.
6.3 Metallurgical phenomena in laser welding of AHSS
From a microstructural point of view, a laser weld can be divided into two different 
areas: the fusion zone, where the metal melted during the welding process undergoes 
solidification, and the HAZ, where peak temperature during welding was less than the 
melting temperature but still high enough for the material to undergo solid-state metal-
lurgical transformations. Both areas, as shown in Chapter 2.2, are characterized by the 
maximum temperature reached locally, followed by rapid cooling. It should be noted 
that the cooling rate in laser welding is so rapid that the as-welded microstructures 
cannot be predicted by an equilibrium phase diagram.
Typical hardness profiles across the weld are depicted in Figure 6.5. The weld cen-
ter systematically has a very high hardness compared with the unaffected base mate-
rial. The HAZ, adjacent to the fusion zone, can either harden or soften. TRIP780 and 
lower-strength dual-phase (DP) steels such as DP450 exhibit only hardening. Some 
softening can occur in higher-strength DP steels with an ultimate tensile strength 
Figure 6.4 Cross section of laser-weldedTRIP780 (right) and DC04 (left) with different 
etching/analysis: nital etching (a), picric acid solution (b), and microprobe analysis of 
manganese concentration (c).
ArcelorMittal internal report.
101High-power beam welding of AHSS
(UTS) higher than 600 MPa; however, this is most prevalent in DP980 and stronger 
DP steels. Finally, as would be expected, the base metal hardness may be correlated 
to its strength.
6.3.1 Weld microstructure
The microstructure of the fusion zone results from the fast solidification and cooling 
of the molten pool. From a metallurgical point of view, the cooling of the weld can 
be considered as a fast quench. As a result, considering the chemical composition of 
most AHSSs (with carbon varying from 0.06% to 0.22% and manganese varying from 
1% to 2.5%), the fusion zone has a predominantly martensitic structure, as illustrated 
in Figure 6.6. Depending on the chemical composition of the steel, some bainite may 
also be found in the fusion zone.
Because the fusion zone is mainly martensitic, its hardness can be easily predicted. 
The hardness of a martensitic structure is a function of its carbon content (Grange, 
Hribal, & Porter, 1977). In the current range of AHSS chemical contents, the weld 
hardness linearly depends on the carbon content of the base material (Figure 6.7).
Many different carbon equivalent (Ceq) approaches have been developed to evaluate 
the weldability of various steel grades (ASM, 1997, chapter 13). The main objectives of 
0
50
100
150
200
250
300
350
400
450
500
550
600
–3.00 –2.00 –1.00 0.00 1.00 2.00 3.00
Position (mm)
H
ar
dn
es
s 
(H
V0
.5
)
DP980
TRIP800
DP780
DP590
DP450
Base metal HAZWeldHAZ Base metal
Figure 6.5 Hardness profiles of laser welds on various advanced high-strength steels. HAZ, 
heat-affected zone.
ArcelorMittal internal report.
102 Welding and Joining of AHSS
Ceq approaches are to evaluate the risk of forming martensite and related cold cracking 
issues in arc welds. It should be noted that a Ceq formula defined for arc welding pro-
cesses must be carefully applied to laser welds because the welding process dynamics 
are fundamentally different. The cooling rate of laser welds is much faster, and the 
residual stresses and distortions occurring around the weld are very low because of 
limited heat loss around the weld during the process. Finally, laser welding typically 
Figure 6.6 Martensitic microstructure in a DP980 laser weld.
ArcelorMittal internal report.
Figure 6.7 Influence of carbon content on the average hardness of laser butt welds (2-mm 
sheets).
ArcelorMittal internal study.
103High-power beam welding of AHSS
is done without adding any filler wire. These three factors are fundamentally different 
from gas metal arc welding. Because these factors differ from arc welding, the resulting 
weld metallurgy in laser welding is also completely different from that of arc welding. 
In laser welds the Ito–Bessyo definition of Ceq (Eqn (6.1)) can be applied to predict the 
final microstructure. Based on experiments, CeqIto >0.2 systematically leads to a fully 
martensitic microstructure.
 CeqIto = C + Si/30 + (Mn + Cu + Cr) /20 + Ni/60 + Mo/15 V/10 + 5B (6.1)
Considering the typical chemical composition of AHSSs, particularly with regard 
to carbon and manganese, most AHSSs exhibit a CeqIto >0.2; therefore, most laser 
welds are expected to have a martensitic microstructure. However, even though laser-
welded AHSSs often have a martensitic microstructure because of the rapid cooling, 
the small distortions and low residual stresses that occur around a laser weld mean that 
there is also not a high risk of cracking upon cooling.
6.3.2 HAZ softening
As described in Chapter 2.2, the HAZ is characterized by a temperature peak followed 
by fast cooling, where the peak temperature is high enough to induce metallurgical trans-
formations but low enough that melting does not occur. Even if the duration at the peak 
temperature is short, significant metallurgical modifications may occur. Many AHSS fam-
ilies have partially or fully martensitic microstructures. The martensite volume fraction 
of the base metal could be relatively low, as in DP600, but may also represent a unique 
microstructure, as in press-hardened steels, for which the final microstructure is obtained 
by hot stamping and die quenching. Martensite is obtained when the face-centred cubic 
(FCC) crystal structure of austenite (stable at high temperatures) does not have sufficient 
time to rearrange into the body-centred cubic (BCC) structure of ferrite (stable at low 
temperatures) during cooling. When this occurs, the FCC lattice shears, forming a body 
centered tetragonal (BCT) structure, entrapping the carbon atoms that were dissolved in 
the austenite structure. This microstructure is metastable. Upon reheating, carbon tends 
to migrate out of the martensite structure, forming a ferritic matrix and carbides, which 
reduce the distortions in the BCT matrix and relieve internal stressors in a process known 
as tempering. This transformation can be measured by a local reduction in hardness. This 
process is thermally activated (Badeshia, 2006). Martensite tempering can be observed 
when holding for a long time at a temperature of 150–200 °C, but it occurs particularly 
quickly at temperatures higher than 500 °C.
Figure 6.8 illustrates the hardness profile across a laser butt weld in DP1180. The 
hardness profile can be decomposed into the following areas:
 • The fusion zone, typically exhibiting a martensitic microstructure, is the result of the fast 
solidification and cooling of the molten pool, as mentioned in Chapter 3.1.
 • The HAZ, which can be divided into two parts:
 – The supercritical HAZ, which is closest to the fusion zone. Like the fusion zone, this area 
of the HAZ is also martensitic. During welding, the peak temperature experienced in this 
area exceeds the Ac3 temperature and then cools rapidly.
104 Welding and Joining of AHSS
 – The subcritical HAZ, which is adjacent to the base material. In this area, the peak tem-
peratures during welding remain below the Ac1 temperature. Martensite initially present 
in the base metal is partially or completely tempered. Because of the short duration at a 
high temperature, softening the material to levels achievable using furnace heat treatment 
is impossible. In the HAZ of a laser weld (Figure 6.8), the resulting microstructure is 
then a mix of the initial martensite, tempered martensite, and ferrite. Depending on the 
distance to the weld and the corresponding peak temperature reached locally, the propor-
tion of the different phases varies. This results in a continuous variation in hardness from 
the minimal hardness (to the location of the Ac1 isotherm, where the strongest tempering 
occurs) to the base metal hardness.
 • The base metal is not sufficiently heated to undergo any transformations. The typical micro-
structure of a DP1180 steel grade is shown in Figure 6.8. The base metal microstructure is 
mainly martensitic, with some islands of ferrite.
In AHSS containing martensite, the HAZ softening phenomena can be observed for 
all welding processes and all welding configurations because this is a metallurgical phe-
nomenon that is not related to the welding process. However, the width of the HAZ 
depends on the welding process (and the related time/temperature of each point of the 
HAZ). Figure 6.9 depicts the influence of the welding process conditions for a given 
material. Thin sheets (1 mm) of electrogalvanized DP1180 have been welded in a butt 
joint configuration under various welding conditions. A narrow weld was obtained using 
a 6-kW CO2 laser at a welding speed of 8 m/min, whereas the wide weld was made at 
3 m/min. It can be clearly seen that the profiles are relatively similar, showing severe 
Figure 6.8 Hardness profile and typical microstructure of a DP1180 laser weld (1-mm thick-
ness) using a 4-kW neodymium:yttrium–aluminum–garnetlaser at 3.5 m/min. HAZ, heat- 
affected zone.
ArcelorMittal internal report.
105High-power beam welding of AHSS
tempering of the HAZ. At low speed (high heat input), the HAZ is clearly wider; more 
heat is transferred to the work piece, so heat is able to be conducted farther from the 
weld zone. This results in a wider area in which the material was exposed to tempera-
tures sufficient to induce tempering (>300 °C). Finally, the minimum hardness measured 
in the HAZ also depends on the heat input. At higher heat input, the time spent above the 
tempering temperature by each point of the HAZ is longer than in the weld made at low 
heat input, which led to more pronounced tempering. A similar tempering phenomenon 
in fully martensitic press-hardened steel welded in overlap configuration was described 
by Gu et al. (2011).
From a macroscopic point of view, HAZ softening tends to decrease the local mechan-
ical properties of the weldment. The affected local mechanical properties may influence 
the global mechanical behavior of the complete part. The consequences of HAZ soft-
ening depend both on the width of the softened area and on the load case applied to the 
weld. Transverse tensile load has been applied to the wide and narrow welds mentioned 
above. In the case of the narrow weld, the strength reduction due to the presence of the 
softened HAZ was only 6% of the initial sheet strength and remained above the minimal 
strength required for a DP1180 grade (Figure 6.10). Even in the case of a large weld with 
more significant softening, the strength reduction was only 13% compared with the base 
metal. Because of the narrow dimensions of laser welds, the properties of the neighbor-
ing material restrict the thinning of the softened material, which reduces the detrimental 
effects of HAZ softening.
Figure 6.9 Heat-affected zone softening in a DP1180 laser weld under various welding 
conditions.
ArcelorMittal internal report.
106 Welding and Joining of AHSS
Vickers hardness usually is considered to be linearly correlated to the UTS of a 
material (ISO 18265). This simple rule cannot, however, be used in the case of HAZ 
softening. The above-mentioned results clearly show that, in spite of the decrease in 
hardness corresponding to 25% of the base metal hardness for a narrow weld (and 
35% for large weld), the ultimate strength of the weld was only reduced by 6% (and 
13%, respectively). The linear correlation between hardness and UTS is to be used 
only in the case of a homogeneous material, not for materials exhibiting small-scale 
changes in mechanical properties, since the properties of the surrounding areas restrict 
the deformation of the soft zone The same trend was observed by Xia et al. (2007), 
Panda et al. (2009), and Panda, Sreenivasan, Kuntz, and Zhou (2008): When the weld 
and the width of the corresponding HAZ are further increased in very large welds 
made by direct diode welding on DP980, the weld strength in the transverse direction 
tends to decrease. It may therefore be concluded that both HAZ softening and the 
width of the softened area must be accounted for when estimating the mechanical 
properties of welding exhibiting HAZ softening.
6.4 Laser-welded blanks (LWBs): issues related to the 
use of AHSS
6.4.1 Principle and typical applications of LWBs
LWBs, also called tailor-welded blanks, are composite blanks that are made of two 
or more sheets that are laser welded before stamping. The sheets making up an LWB 
Figure 6.10 Influence of the width of the heat-affected zone (HAZ) on tensile properties.
ArcelorMittal internal report.
107High-power beam welding of AHSS
can be of different grades or thicknesses or even the same material, depending on the 
application. LWBs allow blanks to be designed so that the final part uses the most 
appropriate material at the right place of the component (ULSAB, 1998) or, in the 
case of LWBs made of the same material, decreased material scrap. By combining 
different steel grades and thicknesses in a part, it is possible to simultaneously opti-
mize the weight and the properties of the final component. Figure 6.11 illustrates 
the crash behavior of a front rail made of two different materials. The front section 
of the part is made of a 600-MPa steel grade with good ductility, while the rear 
section has a 1500-MPa strength. During a crash, the front section deforms and 
absorbs energy, while the rear section remains stable and ensures the safety of the 
car’s occupants.
The following are typical applications of tailored blanks (FSV, 2011; ULSAB, 
1998):
 • Front and rear rails, which optimize the management of crash energy by absorbing energy at 
the vehicle’s extremities
 • B-pillars, where a softer material is used in the bottom area to obtain localized energy 
absorption and deformation in the seat area, while the upper area remains stable to protect 
the passenger in the case of a side impact
 • Door inners, where the door panel is reinforced at the hinges to support the weight of the 
door, while a thin gauge is used for the majority of the part to minimize its weight in the 
section that does not carry a load.
6.4.2 Issues in AHSS applications
The process of producing LWBs can be briefly divided into two steps.
 1. Laser butt welding of the sub-blanks in flat conditions. The key issue in this phase is ensur-
ing a good joint fit-up with a very narrow gap.
 2. Stamping the welded blank as a single component. Stamping is characterized by significant 
deformation of the sheets. The key issue for the weld is being able to support the same level 
of deformation as the surrounding material.
Figure 6.11 Principle of crash management through the application of a laser-welded blank.
ArcelorMittal internal study.
108 Welding and Joining of AHSS
The main concern regarding the introduction of AHSS in LWBs is the formability 
of the weld and the surrounding HAZ. As described in Section 6.3, because of the 
combination of the chemical composition and the cooling speed of the weld, welding 
AHSS usually leads to a highly martensitic microstructure. It is well known that the 
formability of martensite is very low. However, the formability of the weld does not 
solely depend on the formability of martensite. An LWB can be briefly described as 
the combination of three elements: each of the adjacent base materials and the weld 
joining them. Each element has its own mechanical properties. The formability of 
an LWB cannot be defined as the formability of each independent element (Gaied, 
Pinard, Schmit, & Roelandt, 2007). This is particularly true for a weld because it rep-
resents only a narrow portion of the component. The formability of the weld is highly 
influenced by the surrounding materials.
The following results (Figure 6.12) illustrate the capacity of laser welds made 
of different materials to be deformed plastically. Tensile tests have been carried 
out using a standard geometry, with the weld located in the middle of the sample, 
orientated parallel to the tensile direction. In spite of its high hardness and its 
martensitic microstructure (Figure 6.6), the welded coupon can have a uni-
form elongation of up to 12%, which is much higher than would be expected of 
martensite.
Because of the limited width of a laser weld, its behavior is highly influenced by 
the surrounding materials. As a result, significant formability of the weld is obtained 
in spite of its high hardness. Its high formability allows the LWB to be successfully 
formed (Gaied, Cretteur, & Schmit, 2013).
Figure 6.12 Weld elongation in longitudinal tensile tests.
ArcelorMittal internal report.
109High-power beam welding of AHSS
6.5 Body-in-white joining applications
6.5.1 Why use lasers for body-in-white joining?
The development of solid-state laser sources in the past decade has offered new weld-
ing solutions for BIW assembly. In particularly improved electrical efficiency (cost 
reductions) and the capability to combineoptic fibers and long working distances (tool 
flexibility) offers new possibilities (Brockmann, 2010; Kessler, 2010). As a result, 
between 2000 and 2010, laser welding has been widely introduced to the assembly 
line, replacing the resistance spot welding process for various applications (Radscheit 
& Löffler, 2004). The reasons to use lasers can be summarized in three categories:
 • Cost reduction: Compared with the standard spot welding process, laser welding allows much 
higher productivity. This is because of the laser’s ability to move almost instantly between welds 
by directing the weld to the workpiece using rapid tilting mirrors, without the need for moving 
heavy mechanical parts. As a result, the production time is mainly used for welding and is not 
taken up by robot movement. In spite of the higher investment required, the improved productiv-
ity leads to a global cost reduction (Forrest, Reed, & Kizymsa, 2007; Kielwasser, 2009).
 • Flexibility in design: Laser welding requires joint accessibility from only one side, whereas 
resistance spot welding usually requires both sides to be accessible. One-sided access allows 
sheets to be joined to tubes or profiles, as well as joining parts with tight spaces that will 
not fit a spot welding gun. Because of the narrow width of the laser line, laser welding also 
allows flange widths to be reduced, allowing, for example, for wider windows with better 
visibility for the driver (Larsson, 2007, 2009, pp. 6–11).
 • Weldment performance: One of the main advantages of the laser welding process is the 
flexibility in the design of the joints. Long welds can be made to improve the stiffness of 
the assembly. Various weld geometries (e.g., C-shape, S-shape) can be made to improve the 
static or dynamic behavior. This topic is detailed in Section 5.2.
Laser welding is being used industrially at two stages of the BIW assembly:
 • For sub-assembly of components: The main reason to use lasers is to increase the produc-
tivity. Short stitches are done to replace resistance spot welds. Welds can either be linear or 
have more complex shapes to optimize their behavior (Figure 6.13).
Figure 6.13 C-shaped welds on a mass-produced part.
ArcelorMittal internal study.
110 Welding and Joining of AHSS
 • To weld the entire BIW during vehicle assembly, lasers are used to make preferentially long 
welds on large panels. The main advantage of using long welds is to increase the stiffness 
of the BIW. Figure 6.14 shows an industrial application at Volvo Car Corporation, where the 
roof is laser welded to the car body.
6.5.2 Properties of AHSS laser welds and laser-welded 
components
While the formability of the laser weld is a key issue for LWB applications, static 
and dynamic strength are key properties when laser welding is used for BIW 
assembly.
6.5.2.1 Static strength
The main advantage of laser welding over resistance spot welding is the ability to 
adjust the weld length to tailor the strength of the joint to the mechanical require-
ments of the welded component. Although in spot welding there is some flexibility 
to change weld size, the diameter is limited by the diameter of the welding elec-
trodes (it is difficult to make a spot weld much larger than the electrode diameter 
without risking expulsion). However, it is easy to program a longer laser stitch to 
increase the weld strength. As a base approximation, one may assume that, when 
loaded in tensile shear, the weld strength is proportional to the weld length. A 
refined analysis shows that this linear trend is valid as long as the failure mode 
remains the same (Figure 6.15). Short welds are more prone to interfacial failure, 
whereas long welds lead to failure of the HAZ or the base metal. Similar behavior 
can be observed in resistance spot welds when loaded similarly; small-diameter 
welds tend to fail at the interface, whereas larger welds exhibit button pull-out 
(van der Aa, 2013).
Figure 6.14 Laser welding on an XC90 car body.
Courtesy of Volvo Car Corporation, personal communication from Mr Larsson.
111High-power beam welding of AHSS
The failure mode can be understood as the result of a failure along the weakest 
load-bearing area of the assembly. The weakest area can be either:
 • Interfacially through the weld, at the faying surface. This failure mode depends on the fusion 
zone properties (martensite under shear loading) and weld width.
 • Near the weld, in the HAZ, where softening may reduce the local properties.
 • Far from the weld, in the base metal, where failure strength depends on the base metal 
properties (under axial loading) and thickness. Thin sheets and low-strength steel grades are 
more prone to base metal failure.
It is obvious that the failure mode depends on the local properties of the area of 
concern; however, it is also strongly dependent on the local load-bearing section. As a 
consequence, for a given steel grade, the ratio w:t, where w is the weld width and t is the 
sheet thickness, is a major factor driving the failure mode. Gu et al. (2011) also reported 
interfacial failure occurring in press-hardened steel welds in spite of significant HAZ 
softening. This highlights the influence of the weld geometry on the weld properties.
Because of the keyhole energy transfer mode, laser welds are very narrow. The 
weld width depends primarily on the beam diameter and, to a lesser degree, on the 
total heat input. For a given laser configuration (power and optical conditions), the w:t 
ratio decreases rapidly with increasing sheet thickness, leading to interfacial failure in 
thick sheets. The failure mode is therefore not intrinsic to the base metal weldability 
but is mainly driven by sample geometry. This is demonstrated in Figure 6.16, where 
the percentage of welds made in a wide range of steels is graphed against steel thick-
ness. It clearly shows that interfacial failure becomes the predominant failure mode in 
sheets thicker than 1.5 mm, irrespective of metallurgy. It should be noted that a similar 
trend occurs in spot welding of AHSS (Radakovic & Tumurulu, 2008).
Figure 6.15 Tensile shear strength on laser weld stitches of different lengths.
ArcelorMittal internal study.
112 Welding and Joining of AHSS
The comparison of spot and laser weld strengths cannot be restricted to the basic tensile 
shear test. Other loading modes must also be taken into account. The following results 
were extracted from a test campaign performed to evaluate the weld strength of various 
AHSS combinations in both quasi-static and dynamic conditions. The trials were per-
formed on a high-speed testing machine at 5 mm/min for quasi-static tests and 0.5 m/s for 
dynamic tests in pure shear, pure tear (cross-tension), or mixed loading modes (Figure 
6.17). From those trials, the strength at failure and the energy absorbed during the trial 
were measured.
Figure 6.16 Influence of sheet thickness on the failure mode of overlap welds for a 
wide range of products, including DP450-600-780-980; TRIP700-800; FB450-600; and 
M800-1200.
ArcelorMittal internal report.
Figure 6.17 Sample geometry for 
quasi-static and dynamic tests.
Cretteur, Bailly, Pic, Tchorbadjiysky, and 
Cotinaut, 2010.
113High-power beam welding of AHSS
It must be noted that the energy absorbed depends on the deformation of the sample 
and is not just due to the mechanical properties of the weld. However, because all the 
trials used the same sample geometry, this comparison is relevant.
Laser stitches were made using a length of 27 mm. C- and S-shaped welds with 
the same overall weld length were created. This leads to various apparent lengths and 
widths of the welds. A shape factor, expressed as the width-to-length ratio of the weld, 
can be defined according to Table 6.2.
The weld strength at failure can be easily described with an elliptic representation, 
with major axes representing pure shear and normal loading (Figure 6.18). The different 
laser weld geometries arecompared with an 8-mm-diameter resistance spot weld.
Based on Figure 6.18, the following can be observed:
 • In quasi-static conditions the resistance spot weld and the various laser welds have equiva-
lent failure strengths. Replacing a spot weld with 25–30 mm of laser weld has been shown 
to be valid in other work when involving ultra-high-strength steel with thicknesses between 
1.5 and 2 mm (Pic, Tchorbadjisky, & Faisst, 2010).
Table 6.2 Definition of the shape factor
Length of 
fused zone 
(mm)
Shape length 
l (mm)
Shape width 
w (mm)
Shape factor 
(w/l)
Linear stitch 27 27 1 0.04
C-shape 27 16 5 0.31
S-shape 27 14.7 5 0.34
Figure 6.18 Quasi-static (a) and dynamic strength (b) of DP600 2-mm and 1.5-mm welds.
Cretteur et al. (2010).
114 Welding and Joining of AHSS
 • Laser welds have higher failure strengths than spot welds when dynamically loaded. Laser 
weld geometry did not affect dynamic weld strength.
The energy absorption characteristics of these welds also were analyzed. Figure 6.19 
illustrates the failure strength and energy absorption for welds joining press-hardened 
steel to DP steel:
 • In tearing conditions the failure strength is lower for resistance spot welds than for the var-
ious laser-welding geometries. The energy absorbed by the spot-welded sample was also 
significantly lower than that measured for the laser-welded samples.
 • In shear the failure strength is equivalent for all the welding processes. However, energy 
absorption is slightly better for the resistance spot welds because of the different failure 
modes of the welds. The laser welds failed interfacially under shear loading. This failure 
mode led to lower total energy absorption (Figure 6.20).
The failure mode can be significantly influenced by the outer dimensions of the 
weld. Figure 6.21 shows how the weld shape factor influences the probability of 
Figure 6.19 Strength at failure (left) and energy absorption (right) of Usibor1500P 1.8-mm 
and DP600 1.5-mm samples for various welding conditions.
Cretteur et al. (2010).
Figure 6.20 “Plug out” failure mode in an S-shaped (a) and C-shaped weld (b).
ArcelorMittal internal report project.
115High-power beam welding of AHSS
interfacial failure during the mechanical tests of various weld shapes and material 
combinations. A shape factor of 1 (corresponding to any geometry with equal width 
and length, for example, a circle or any shape that could be inscribed within a square) 
is more favorable to “plug out” failure modes. This is related to the different stress 
concentrations around the weld for different shape factors.
The influence of various weld shape factors on the failure mode was also observed 
to scale up in larger components. Crash tests were performed on hat channel-shaped 
structures with various welding conditions: resistance spot welding, laser stitch, and 
S-shaped laser welds, as well as weld bonding (combination of spot welding with Beta-
mate 1496 high-strength adhesive). These crash tests clearly show that laser stitches 
have the highest rate of failure during crash testing (33%), as may be seen in Figure 
6.22. Resistance spot welding also showed some weld failures. By shaping the laser 
welds, however, no more weld failures occurred during testing, even in cases where 
the parts were severely deformed during the test. In terms of energy absorption the best 
laser-welded solution (S-shaped welds of 21 mm × 12 mm) was able to absorb 10% more 
energy than the spot weld reference (Cretteur, Bailly, Pic, Tchorbadjiysky, & Cotinaut, 
2010).
6.5.2.2 Stiffness
The reduction in sheet thickness allowed by the use of ultra-high-strength steel has a 
direct and negative impact on the component stiffness. However, the choice of the join-
ing technique can compensate for the loss of stiffness (Audi, 2007; Daimler, 2009; Pic 
et al., 2010). In particular, laser welding offers the possibility of producing continuous 
joints, increasing the component’s stiffness. Figure 6.23 shows the torsional stiffness 
=
Figure 6.21 The occurrence of interfacial failure depends on the weld shape factor.
Pic, Tchorbadjiysky, Faisst and BeaLaser (2010).
116 Welding and Joining of AHSS
of DP600 clam shell beams (made of two hat sections). The hat sections were welded 
using various laser weld geometries. Then the beams were torsionally loaded, and the 
corresponding stiffness was calculated. It should be noted that the stiffness is highly 
dependent on the initial geometry (Cretteur et al., 2010) of the component; therefore, 
× × ×
Figure 6.22 The influence of the welding process on weld integrity during frontal crash tests 
of TRIP800 1.5-mm-thick steel.
ArcelorMittal internal report project.
Figure 6.23 Evolution of torsional stiffness of DP600 1.2-mm beams with the joining process.
ArcelorMittal internal report project.
117High-power beam welding of AHSS
values presented in Figure 6.23 must not be considered absolute and only illustrate 
a trend. When component stiffness was measured, a continuous weld line increased 
stiffness by 15%. Weld geometry also plays a role in the final stiffness. An interrupted 
line of 50-mm stitches led to an intermediate improvement in stiffness. This linearly 
corresponded to the proportion of the welded length of the beam flanges. Using curved 
welds instead of linear stitches tends to lower the beneficial effect of laser welding 
on stiffness. With C-shaped welds, the key factor is not the effective weld length but 
the apparent weld length l, as defined in Table 6.2. This shows that while C-shaped 
welds are preferred for crash conditions, linear welds are promoted for stiffness-driven 
components. The versatility of laser welding allows the weld design to be adapted, 
depending on the main function of the AHSS component.
6.6 Conclusions
The rapid evolution of laser technologies between 1990 and 2010 has led to a wide 
range of laser-welding applications in the automotive industry, both for the manufac-
ture of components and the assembly of the BIW.
AHSSs also penetrated the automotive market during the same period. The rapid 
heating and cooling cycles experienced in laser welding, combined with the complex 
microstructure and relatively high alloying content of AHSS, lead to significant mod-
ifications of the base metal in the vicinity of the weld. Specifically, the fusion zone is 
prone to being highly martensitic. However, it was proven that the fusion zone’s high 
hardness does drastically not limit the strength and formability of the weld. Also, HAZ 
softening frequently occurs in AHSS welds, particularly on steels containing marten-
site (such as DP or press-hardened steels). Even if the decrease in the hardness of the 
HAZ can be considered as a local reduction of the metal’s mechanical properties, it 
was proven that the properties of the welded component are only slightly affected by 
the presence of this locally weakened area. Because the HAZ in laser welds is typi-
cally very narrow, the effect of HAZ softening on the overall mechanical properties is 
also limited because the mechanical properties of the neighboring base material and 
fusion zone dominate as the HAZ narrows. Although weld microstructure affects local 
properties, weld geometry is of primary importance when determining the overall 
strength and formability of the weld.
Adapting the weld geometry to the targeted application is also a major advantage of 
the laser welding process. While many other welding processes are limited in terms of 
weld geometry, laser welding offers an infinite variety of weld shapes (lines, curves, 
C- or S-shapes of various dimensions). Modifying the weld definition on a given com-
ponent could lead to improved stiffness and crash behavior.
The use of AHSS in the automotive industry aims to optimize simultaneously car 
body weight and mechanical properties. Joining processes must also be considered 
when optimizing the auto body design. Laser welding offers the largest versatilityand potential solutions for improving properties. Laser welding offers solutions to 
optimize the behavior of an AHSS component.
118 Welding and Joining of AHSS
Acknowledgments
The author thanks J. Larsson from Volvo Car Corporation for providing images presented in this 
chapter. The author also thanks Ms Tainturier, Ms Laveau, Ms Bayart, Ms Gayet and Mr Luquet, Mr 
Gaied, Mr Marakchi, Mr Lucas, Mr Bobadilla, Mr Bailly, Mr Pic, and Mr Yin from ArcelorMittal 
R&D for their participation in the chapter, and particularly Mr Biro for reviewing the chapter.
References
van der Aa, E. (2013). Welding of pre-deformed AHSS. Results from RFCS project REFORM, 
Conf Joining in Car Body Engineering 2013, Bad Nauheim, Germany.
ASM handbook weld integrity and performance. (1997). Chapter 13 ASM.
Audi. (2007). Die Karosserie des neuen Audi A5 coupé. In Euro car body 2007 conference, 
BadNauheim/Frankfurt.
American Welding Society. (2007). AWS welding handbook, Welding processes (9th ed.) 
(Vol. 3). Part 2, Chapter 14.
Badeshia, H. (2006). Steels. In S. Honeycombe (Ed.), Chapter 9: The tempering of martensite.
Brockmann, R. (2010). Joining process for profile intensive car body structures. In Vincentz 
Network GmbH (Ed.), Fugen in Karrosseriebau, Bad-Nauheim (D).
Cretteur, L., Bailly, N., Pic, A., Tchorbadjiysky, A., & Cotinaut, L. (2010). Properties optimi-
zation on car body modules through adequate selection of the joining processes. Munich 
2010: International Auto Body Congress.
Daimler, A. G. (2009). The new E-Class. In Euro carbody 2009 conference, BadNauheim/
Frankfurt.
Fabbro, R. (2002). Basic processes in deep penetration laser welding. In ICALEO conference 
2002, Scottsdale, USA October 14–17.
Fabbro, R., Slimani, S., Coste, F., & Briand, F. (2005). Study of keyhole behaviour for full pene-
tration Nd–Yag CW laser welding. Journal of Physics D: Applied Physics, 38, 1881–1887.
Forrest, M., Reed, D., & Kizyma, A. (2007). Business case for laser welding in body shops: 
Challenges and opportunities. Berlin: IABC.
FSV. (2011). Future steel vehicle consortium. overview report. WorldAutoSteel consortium. 
www.worldautosteel.org/projects/future-steel-vehicle/phase-2-results/.
Gaied, S., Cretteur, L., & Schmit, F. (2013). Modeling of laser welded blanks formability, revue 
de metallurgie, 110, 199–203.
Gaied, S., Pinard, F., Schmit, F., & Roelandt, J. M. (2007). Formability assessment of laser 
welded blanks by numerical and analytical approaches. In Sheet metal welding conference, 
Detroit, 2007.
Grange, R., Hribal, C., & Porter, L. (November 1977). Hardness of tempered martensite in car-
bon and low-alloy steels. Metallurgical Transactions A, 8(11), 1775–1785.
Gu, Z., Yu, S., Han, L., Meng, J., Xu, H., & Zhang, Z. (2011). Microstructures and properties of 
ultra-high strength steel by laser welding. ISIJ International, 51(7), 1126–1131.
ISO 18265. Metallic materials—Conversion of hardness values.
Kessler, B. (2010). Fiber laser spot welding guns versus remote laser and resistance spot- 
welding processes. conf. Fügen in Karrosseriebau 2010, Bad-Nauheim.
Kielwasser, M. (2009). First production experience acquired with laser scanner welding at PSA 
Peugeot Citroen. Bad-Nauheim: European Automotive Laser Applications.
http://www.worldautosteel.org/projects/future-steel-vehicle/phase-2-results/
119High-power beam welding of AHSS
Larsson, J. (2007). The new Volvo V70 and XC70 car body. In Euro car body conference 2007, 
BadNauheim.
Larsson, J. (May 2009). Designed for laser welding: The Volvo XC60, industrial laser solutions 
for manufacturing.
Mazumder, J. (1993a). Laser-beam welding, ASM handbook, Welding brazing and soldering 
(Vol. 6) (pp. 262–269).
Mazumder, J. (1993b). Procedure development and practice consideration for laser beam weld-
ing, ASM handbook, Welding brazing and Soldering (Vol. 6) (pp. 874–880).
Mujica, L., Webera, S., Pinto, H., Thomy, C., & Vollertsen, F. (2010). Microstructure and 
mechanical properties of laser-welded joints of TWIP and TRIP steels. Materials Science 
and Engineering A, 527, 2071–2078.
Panda, S. K., Baltazar Hernandez, V. H., Kuntz, M. L., & Zhou, Y. (August 2009). Formability 
analysis of diode-laser-welded tailored blanks of advanced high-strength steel sheets. 
Metallurgical and Materials Transactions A, 40A.
Panda, S. K., Sreenivasan, N., Kuntz, M. L., & Zhou, Y. (August 26, 2008). Numerical simula-
tions and experimental results of tensile test behavior of laser butt welded DP980 steels. 
Journal of Engineering Materials and Technology, 130(4), 041003.
Pan, Y., & Richardson, I. M. (2011). Keyhole behaviour during laser welding of zinc coated 
steel. Journal of Physics D: Applied Physics, 44(4), 45502.
Pic, A., Tchorbadjisky, A., Faisst, F., & BeLaser, M. (2010). Laser welding optimization for steel-
based body-in-white structures. In European laser automotive application conference, Bad-
Nauheim 2010.
Radakovic, D. J., & Tumuluru, M. (April 2008). Predicting resistance spot weld failure modes 
in shear tension tests of advanced high-strength automotive steels. Welding Journal, 87.
Radscheit, C., & Löffler, K. (2004). Laser applications on GolfV: concept and implemen-
tation into production. In International conference “advanced metallic materials and 
their joining”, 2004 Bratislava, Slovakia.
ULSAB. (1998). Ultra light steel auto body consortium, program report/materials and 
processes. WorldAutoSteel Consortium. www.worldautosteel.org/projects/ulsab/ 
ultralight-steel-auto-body-ulsab-programme/.
Xia, M., Sreenivasan, N., Lawson, S., Zhou, Y., & Tian, Z. (January 2007). A comparative 
study of formability of diode laser welds in DP980 and HSLA steels. Journal of Engi-
neering Materials and Technology, 129(3), 446–452.
http://www.worldautosteel.org/projects/ulsab/ultralight-steel-auto-body-ulsab-programme/
http://www.worldautosteel.org/projects/ulsab/ultralight-steel-auto-body-ulsab-programme/
 
This page intentionally left blank
Welding and Joining of Advanced High Strength Steels (AHSS). http://dx.doi.org/10.1016/B978-0-85709-436-0.00007-2
Copyright © 2015 Elsevier Ltd. All rights reserved.
Hybrid welding processes in 
advanced high-strength steels 
(AHSS)
S. Chatterjee, T. van der Veldt
Tata Steel Research and Development, Joining and Performance Technology, 
Wenckebachstraat, The Netherlands
7
7.1 Introduction
The laser beam welding process can be combined, in principle, with the arc welding 
process. When these two welding techniques are coupled as one process, the laser 
beam and arc (Gas Tungsten arc welding, plasma or gas–metal arc welding (GMAW)) 
interact at the same time in one zone (plasma and weld pool) and mutually influence 
and assist one another. Such process coupling is referred to by the term hybrid welding 
process (DVS, 2005).
One of the remarkable characteristics of laser welding is the narrow and deep config-
uration of the weld. This narrow weld is the result of the high energy concentration of the 
process and the high welding speed, which result in a low heat input into the workpiece. 
Automotive joining lines use the narrow weld characteristic and cost advantage of the high 
processing speed, but its narrow weld leads to some difficult metallurgical problems. Also, 
laser welding requires special attention because the narrow laser beam (a few microns) 
has intolerant to poor fit between the parts. The arc welding process, on the other hand, is 
more tolerant to the poor fit between the parts because of its relatively wider arc spot (a 
few millimetres). Also GMAW has the possibility to add filler metal to better bridge gaps 
and control the metallurgical influence on the weld microstructure. In GMAW, however, 
the speed of welding is relatively much lower than that of laser welding.
In laser–arc hybrid welding, a laser beam is used in combination with an arc pro-
cess to produce a weld seam. The laser ensures deep penetration at a higher welding 
speed, whereas the arc produces arelatively wider and smooth weld face and alleviates 
problems caused by misalignment or gaps in the joint configurations (Ataufer, 2005; 
Duley, 1999; Ishide, Tsubota, Watanabe, & Ueshiro, 2003; Kutsuna & Chen, 2002; 
Petring, Fuhrmann, Wolf, & Poprawe, 2003; Schubert, Wedel, & Kohler, 2002; Steen, 
2003; Steen & Eboo, 1979; Steen et al., 1978; Tsuek & Suban, 1999).
In the late 1970s at Imperial College London, a group of scientists lead by 
Professor William M. Steen performed the first attempts at combining a laser and an 
arc (tungsten inert gas) welding process (Steen et al., 1978; Steen & Eboo, 1979). 
This early investigation showed that the combination of a laser beam and arc within a 
common process zone is more than a simple combination of two heat sources. These 
experiments demonstrated that the laser radiation had an essential impact on the 
arc behaviour, leading to a stabilized arc column and a contraction of the arc spot. 
122 Welding and Joining of AHSS
This new hybrid laser–arc process can penetrate deeper with a narrow weld bead com-
pared with both the solo laser and the solo arc welding techniques. A significantly higher 
welding speed can also be obtained with this hybrid technique. However, this inno-
vation did not immediately find practical applications because the laser process itself 
was not viable on an industrial scale (Bagger, Flemming, & Olsen, 2005; Seyffarth & 
Krivtsun, 2002). In the early 1990s, when the multi-kilowatt laser systems become 
available, the challenges in developing welding applications changed from the perfor-
mance of the beam sources (e.g. with respect to penetration or welding speed) to the 
requirements considering fit-up and part tolerances. Possibilities and limitations of the 
hybrid laser–arc welding technique were investigated all over the world (in the United 
States, Europe and Japan) (Beyer, Imholff, Neuenhahn, & Behler, 1994; Dilthey & 
Wieschemann, 1999; Ishide, Tsubota, & Watanabe, 2002; Magee, Merchant, & Hyatt, 
1991). After renewed research efforts, promising success in industrial applications 
(e.g. in the shipyard industry) was accomplished quickly (Merchant, 2003; Denney, 
2002; Dilthey, Wieschemann, & Keller, 2001). Hybrid laser–arc welding became one 
of the hot topics in laser processing. New industries such as the pipeline and the auto-
motive industries, became interested. The continued development of high-power, sol-
id-state lasers with smaller form factors, greater efficiency and lower cost has greatly 
influenced the usability of this hybrid welding in industries. New technologies, in 
particular the fibre delivery system, helped integrate the process into conventional 
motion systems such as robots, gantries and automation, which increased the accep-
tance of the hybrid system in automotive industries. The recent development of laser 
sources with even more high-power density, such as disc lasers and fibre lasers, com-
bine higher laser power in a higher beam quality. In thick-plate welding this enables 
deeply penetrating welds with high aspect ratios, but at the same time it increases 
even further demands on edge quality. A hybrid welding solution could reduce these 
fit-up problems. However, for very thin sheets, hybrid welding is not really suitable; 
the thinnest application is around 1 mm (Hansen, 2012). In that respect laser–GMAW 
hybrid welding has a limitation for many parts of the auto body. This chapter describes 
the laser–GMAW hybrid welding process from an automotive perspective.
7.2 Laser–arc hybrid process description
The laser–arc hybrid welding process is schematically shown in Figure 7.1, along with 
a photograph of a laser–GMAW hybrid welding head. The arc, in addition to the laser 
beam, supplies heat to the weld metal in the upper weld region, giving the weld seam 
its ‘wine glass shape’ (a wider weld face and a narrower weld root). Filler metal is 
supplied to the weld pool by the electrode wire. The mutual influences exerted by the 
laser and arc during the process can differ in intensity as a function of the arc or laser 
process used and the process parameters.
Understanding physical phenomena in hybrid welding is important because all 
states of solid, liquid, vapour and plasma exist in a small space. A keyhole is generally 
formed in the molten pool with the laser beam with high-power density and, simulta-
neously, a plume (laser-induced metal vapour), and spatters are formed in the space. 
123Hybrid welding processes in AHSS
GMA plasma and droplets from the filler wire also exist above the molten pool. 
An increase in arc voltage is noted during hybrid welding, and the amount of this 
increase is in direct proportion to the power of the yttrium–aluminium–garnet laser. 
The plume, which evolves towards the incident laser beam, affects the phenomenon 
such that the arc column becomes brighter and longer, leading to an increase in the 
arc voltage in this hybrid welding (Naito, Mizutani, & Katayama, 2006). In the case 
of a hybrid carbon dioxide (CO2) laser and pulsed metal active gas welding, the arc 
approaches a laser-induced plume at low voltages but covers the molten pool just 
below the wire at higher voltages (Sugino, Tsukamoto, Nakamura, & Arakane, 2005). 
A laser-induced plume often acts as an arc current path between the electrode or wire 
and the plate when the laser beam and the heat source are close together. The inter-
action between the arc and the laser-induced plume generally depends on the type 
of laser and shielding gas, the arc current, the distance between an electrode and the 
plate, the distance between a laser irradiation spot and an electrode target on the plate 
and the inclination of the electrode.
Laser beam Electrode
Arc
Fusion zone
Weld direction
Workpiece
Vapour cavity
Plasma
Laser-induced metal 
vapour
(a)
(b)
Figure 7.1 (a) Principles of the laser–arc hybrid welding process (H. Staufer, M. Rührnößl 
and G. Miessbache, January 02, 2013, Hybrid welding for the automotive industry.) and (b) a 
laser–GMAW hybrid welding head.
124 Welding and Joining of AHSS
7.3 Laser–arc hybrid process parameters for welding 
automotive AHSS
The hybrid laser–arc welding technique has proved its suitability in many indus-
trial applications. However, a large number of parameters have to be set correctly to 
achieve proper joint quality. Some of these parameters are described below.
7.3.1 Energy input
The heat input to which the weldment is exposed as a result of the hybrid pro-
cess can be kept low compared with the standalone arc processes. In general, an 
increase in laser power increases the weld penetration. In the case of hybrid laser–
arc welding (as opposed to the laser-only process) this increase in penetration is 
accentuated because the reflectivity of the workpiece metal is reduced as the metal 
is heated by the arc. The laser’s or the arc’s character may predominate, however, 
depending on the selected power input ratio. At Tata Steel, a series of experimental 
welding has been done by varying laser power and GMAW power; for a 2.5-kW 
laser power when the arc power exceeds 10% of the laser power, the weld penetra-
tion decreases (Chatterjee, Mulder, and van der Veldt, 2013).
Laser power is the dominant factor influencing weld penetration. The welding 
voltage has been shown not to influence the weld penetration depth by a great deal, but 
the weld bead gets wider if the welding voltage increases, giving a lower depth-to-width 
ratio for a same laser power. The arc voltage (and wire feed rate) therefore need to 
be increased for wider fit-up gaps to avoid any lack of fusion. The welding current 
generally is matched to the filler wire diameter (a higher welding current for a larger 
wire diameter). Considering a given wire diameter and voltage settings, an increase 
in welding current gives a deeper weld with a higher depth-to-width ratio. Nilsson, 
Heimbs, Engström, and Alexander (2003) studied the effect of Metal inertgas weld-
ing (MIG) power on the weld geometry of hybrid welds and reported the width of 
the heat-affected zone increases with increasing MIG power. Also, the depth of the 
undercut increases with increasing MIG power (Figure 7.2).
7.3.2 Welding speed
Ability to achieve a faster welding speed is one of the great advantages of the hybrid 
welding process. Because of the inherent high energy density, a laser beam can be 
moved very quickly during laser welding, but maintaining a stable arc at a high speed 
is a difficult task. However, Ono, Shinbo, Yoshitake, and Ohmura (2002) studied this 
and reported that the arc remains steady in hybrid welding, even at high speeds. They 
showed the welding speed limit for hybrid welding is at least seven times higher than 
that for arc welding (as shown in Figure 7.3). In arc welding the arc is actually main-
tained by thermionic emission from the sheet. When the welding speed is high, the 
heating becomes insufficient and the arc becomes unstable. By contrast, during hybrid 
welding, the electron density in a keyhole formed by laser radiation reaches 1017–1020/
cm3 (Ono et al., 2002) Moreover, the surrounding area is in a molten state, so that 
125Hybrid welding processes in AHSS
Figure 7.2 (a) Width of the heat-affected zone (HAZ) and the weld seam, depending on the 
MIG power. (b) Depth of the undercut as a function of the MIG power.
Nilsson K, Heimbs S, Engström H, Alexander FH. 2003. Parameter influence in CO2-laser/
MIG hybrid welding. IIW Doc. IV-843-03. © IIW.
Figure 7.3 The welding speed limit 
for forming a uniform weld using 
hybrid and arc welding.
Ono M, Shinbo Y, Yoshitake A, and 
Ohmura M. 2002. Development of 
laser-arc hybrid welding. NKK 
Technical Review, 86.
126 Welding and Joining of AHSS
thermionic emission takes place very easily. In fact, the plume generated from the 
laser interaction with the material feeds the arc process. When arc welding is com-
bined with laser welding in this region, a stable arc is maintained, even when the 
welding speed is high.
The weld penetration increases when the welding speed is decreased because the 
heat input per unit length of the weld is higher. Also, the capability of the filler wire 
to fill the gap is improved at lower welding speeds (at a constant rate of filler wire 
feeding). The ratio between welding speed and filler wire feeding is important to the 
stability of the keyhole and thus to the stability of the process itself.
7.3.3 Relative arrangement of the laser and the MIG torch
To get the maximum weld penetration, the laser is positioned perpendicular to the 
weld seam, and the arc torch is kept at an angle to the laser beam. For a fillet type 
of joint configuration, keeping the arc torch with an angle to the joint line can be 
beneficial. The leading or trailing position of the arc torch is a determining factor of 
the weld characteristics. The arc-leading configuration helps to obtain an increase in 
penetration.
The distance between the laser and the filler wire tip is one of the most important 
parameters to control in hybrid laser–arc welding. A short distance, typically 1.5 mm, 
between the laser spot and the filler wire tip has been shown to be favourable for a steady 
keyhole. Nilsson et al. (2003) studied this by keeping the laser beam at the joint centre 
and laterally displacing the MIG torch up to 2 mm from the joint centre (Figure 7.4). 
With increasing lateral displacement of the MIG torch, the weld became more asym-
metrical. When a lateral displacement is 2 mm, the distance between the laser beam 
and the arc becomes so large that the two processes started acting separately without 
any synergy. The arc is still attracted by the laser beam up to a 1.5 mm gap, so that the 
typical hybrid weld forms.
7.3.4 Focal point position
The maximum weld penetration for the hybrid laser–arc process is generally obtained 
when the laser beam is focused below the surface of the top sheet. Depending on 
Figure 7.4 Micrographs of the weld seams with a lateral displacement B of the arc V-joint.
Nilsson K, Heimbs S, Engström H, Alexander FH. 2003. Parameter influence in CO2-laser/
MIG hybrid welding. IIW Doc. IV-843-03. © IIW.
127Hybrid welding processes in AHSS
the requirement of the weld shape, position of the focal point needs to be selected. 
Campana, Fortunato, Ascari, Tani, and Tomesani (2007) carried out hybrid welding 
experiments using 8-mm-thick plates and reported that the focal position of the laser 
beam must be kept below the surface of the upper substrate to achieve best result 
(Figure 7.5). The distance between the substrate surface and the laser beam focus 
depends on the GMAW metal transfer mode: it should be less for a short arc and more 
when using a pulsed/spray mode.
7.3.5 Angle of electrode
The penetration of the weld increases with the angle of the electrode to the workpiece 
surface up to 50°. The gas flow along the welding direction provided by the arc torch 
deflects the plasma induced by the laser, and this plasma reduces the absorption of the 
laser beam when CO2 lasers are used. Therefore the angle of the electrode to the top 
surface of the workpiece is often set at around 40–50°.
7.3.6 Joint gap
The laser–arc hybrid process is well known for its tolerance to inaccurate joint prepa-
ration and joint fit-up. The capability of hybrid lap welding to bridge gaps is consid-
erably greater than that of laser welding because the filler wire used in hybrid welding 
supplies enough weld metal to fill gaps. By contrast, when a gap is present in laser 
welding with no filler metal, the amount of molten metal tends to be insufficient to fill 
the gap, resulting in weld defects such as underfill or burn-through. Ono et al. (2002) 
used lap welding on sheets with different thicknesses and varied gaps to investigate 
the gap tolerance of hybrid welding vis-à-vis laser welding. Their results are shown 
in Figure 7.6(a) and (b), which illustrates that the gap tolerance for hybrid welding is 
much higher than that of laser welding alone.
Nilsson et al. (2003) studied the gap tolerance of a hybrid process on butt joint 
configurations. They also found that the joint without a gap shows no undercut, but 
Figure 7.5 Influence of the laser beam’s focal position on hybrid weld penetration.
Campana G, Fortunato A, Ascari A, Tani G, and Tomesani L. 2007. The influence of arc transfer 
mode in hybrid laser-mig welding. Journal of Materials Processing Technology 191, 111–113.
128 Welding and Joining of AHSS
joints with a gap show varying levels of undercuts. This can be solved by increasing 
the speed of the wire feed. Nilsson et al. found a relationship between necessary wire 
feed speed and gap width (shown in Figure 7.7).
The welding speed cannot be increased linearly. As the weld travel speed 
increases, increasingly larger amounts of filler wire have to be melted to fill the 
gap. This takes more time and requires lower process rates, and the speed has to be 
reduced for wider gaps.
Figure 7.6 (a) Gap tolerance in 
laser lap welding. YAG, yttrium–
aluminium–garnet. (b) Gap toler-
ance in hybrid lap welding.
Ono M, Shinbo Y, Yoshitake A, and 
Ohmura M. 2002. Development of 
laser-arc hybrid welding. NKK Tech-
nical Review, 86.
129Hybrid welding processes in AHSS
7.3.7 Optimization of welding parameters: torch angle, stick-out 
and beam-to-wire distance
Tata Steel performed welding experiments with different hybrid welding parameters 
using lap shear specimens. Parameters (Table 7.1) such as laser power, wire feed 
speed, welding travel speed and shielding gas composition were kept fixed during the 
experiments to observe the effects of torch angle, wire stick out (for arc voltage) and 
the gap between the laser beam and the arc during weld penetration. The experiments 
were designed as per the Box–Behnken method (Table 7.2).
The results of these tests, shown in Figure 7.8(a)–(c), indicate how the weld pen-
etration changes withtorch angle and at different stick outs.
0
0 0.5 1
Gap width (mm)
1.5 2
0
5
10
15
20
25
0.2
0.4
0.6
0.8
W
el
di
ng
 tr
av
el
 s
pe
ed
 (m
/m
in
)
W
ire
 fe
ed
 ra
te
 (m
/m
in
)
1
1.2
1.4
1.6
1.8
2
Figure 7.7 Welding and wire speed as a function of gap width.
Nilsson K, Heimbs S, Engström H, Alexander FH. 2003. Parameter influence in CO2-laser/MIG 
hybrid welding. IIW Doc. IV-843-03. © IIW.
Table 7.1 Welding parameters
Fixed parameters
Laser power 2.5 kW
Wire feed speed 7 m/min
Welding travel speed 1 m/min
Shielding gas 92% Ar + 8% CO2
Varied parameters
Torch angle 5–65°
Stick out 13–21 mm
Distance between the laser beam and MIG wire 0.5–2.5 mm
130 Welding and Joining of AHSS
Figure 7.8 illustrates that when a small stick out is used, the penetration is highest 
when the distance from the laser beam to the arc is larger. Similarly, for a large stick 
out, the laser-to-arc distance should be smaller to obtain higher penetration. Changing 
the torch angle does not change this trend but shifts the optimum.
7.3.8 Shield gas composition
The predominant constituent of the shielding gas is generally an inert gas such as helium 
or argon (Ar). A shielding gas that provides a higher ionization potential is required 
because the plasma can deflect or absorb a portion of the laser energy when CO2 lasers 
are used. Helium is, therefore, often preferred over Ar for laser welding, but its lightness 
is a disadvantage; it is often combined with Ar, which is heavier, without substantially 
altering the weld penetration depth. The addition of reactive gases such as oxygen (O2) 
and CO2 has an influence on the weld pool wetting characteristics and bead smoothness.
The weld penetration of hybrid welding varies with the plasma shape, which is 
determined by shielding gas parameters, especially the plasma height interacting with 
the incident laser. The higher the plasma height interacting with incident laser, the 
shallower the weld penetration. The effect of shielding gas parameters on plasma 
shape is achieved in two ways: laser–arc plasma interaction and the direction and 
velocity of gas flow. Figure 7.9 (Gao, Zeng, & Hu, 2007) shows the effect of the heli-
um-to-Ar ratio on the plasma shape during the hybrid welding process.
Dilthey et al. (2001) investigated the effect of shielding gas on the porosity of the 
bead surface and spattering in hybrid welding of galvannealed steel sheets in an over-
lapping fillet joint configuration (without a gap in the flat position). They reported the 
Table 7.2 Experimental design as per the Box–Behnken method
Experiment no.
Torch angle 
(degrees) Stick out (mm)
Distance between the laser 
beam and MIG wire (mm)
1 5 13 1.5
2 5 21 1.5
3 65 13 1.5
4 65 21 1.5
5 5 17 0.5
6 5 17 2.5
7 65 17 0.5
8 65 17 2.5
9 35 13 0.5
10 35 13 2.5
11 35 21 0.5
12 35 21 2.5
13 35 17 1.5
14 35 17 1.5
15 35 17 1.5
131Hybrid welding processes in AHSS
Figure 7.8 Effects of stick out (mm) and laser-to-wire distance on weld penetration when 
torch angle is (a) 5°, (b) 35° and (c) 65°.
132 Welding and Joining of AHSS
CO2 in Ar + CO2 shielding gas is effective in decreasing the pits, whereas O2 decreases 
the size but increases the number of pits. Increasing the CO2 and O2 gas ratio in 
Ar + CO2 and Ar + O2 shielding gas deepens the penetration and widens the weld bead, 
respectively. Both CO2 and O2, however, increase spattering. A mixture of Ar, CO2 
and O2 as a shielding gas is most suitable for welding of galvannealed steel sheets 
because it produce less pitting, porosity and spatter.
7.4 Applications in the automotive industry
Welding in the automotive industry predominantly involves joining sheet metals. 
Therefore the energy input to the substrate has to be very low compared with that 
when welding heavier-gauge steel and needs to be controlled precisely to avoid any 
distortion. The process has to be extremely fast to cope with the productivity of auto-
motive production lines. Ease of automation or robotization is another criterion for 
such welding processes, as is robustness. Currently, resistance spot welding and laser 
welding are the two major welding processes that are considered most suitable for 
automotive manufacturing lines. With the increased use of advanced high-strength 
steels (AHSS) in recent years, however, welding in automotive manufacturing has 
become more challenging. As mentioned in previous chapters, AHSS grades differ 
significantly from conventional formable low-alloy automotive steels in terms of total 
alloy content, microstructures and different thermophysical properties, which require 
different welding practices for AHSS grades. There are some limitations for applying 
laser–GMAW hybrid welding to manufacturing of many auto body parts. It is not 
suitable for very thin sheets. Accessibility is also an issue because the welding heads 
are quite large. Still, this hybrid welding process is chosen by the automotive industry 
as an enabling technology to reduce both distortion and mass without compromising 
structural crashworthiness.
Depending on the materials and joint configurations, each car contains some hybrid 
welds. The main application of hybrid welding in automobiles is in the chassis and 
suspension. Volkswagen and Audi in particular are two examples of companies con-
vinced by the benefits of hybrid laser–MIG welding (Brettschneider, 2003; Beyer, 
Brenner, & Poprawe, 1996; Graf & Staufer, 2003; Staufer, 2003). For manufacturing 
Figure 7.9 Plume formation for different shielding gas mixtures.
Gao M, Zeng X and Hu Q. 2007. Effects of gas shielding parameters on weld penetration of 
CO2 laser-tungsten inert gas hybrid welding. Journal of Materials Processing Technology 
184, 177–183.
133Hybrid welding processes in AHSS
the doors of the Volkswagen Phaeton, hybrid welding is applied, in addition to MIG 
and laser welding. One door includes 48 hybrid-welded seams with a total length of 
3570 mm. The seams are mainly fillet seams on the lap joints and some butt joints. 
To meet rigidity requirements for the doors and to save weight at the same time, 
having a tailor-made combination of sheet, casting and extruded materials would be 
necessary. At some spots these parts can be jointed only by hybrid welding because 
of the required speeds and given tolerance. Without the hybrid process, Volkswagen 
would have had to use heavy casting material.
The new Audi A8 also uses hybrid welding. Each vehicle comprises a total of 4.5 m 
of weld seam. Hybrid welding is used in the lateral roof frame, which is equipped with 
various functional sheets (Figure 7.10). Daimler also uses hybrid welding to produce 
the axle components of their C-class vehicles (Staufer, 2009).
7.5 Costs and economics
Some important superior features of hybrid welding compared with pure laser or arc 
welding are listed below.
 • Higher welding speed. Productivity is improved through increased welding speed. For sheet 
material it is possible to enhance speed by 30% compared with conventional laser welding, 
without the addition of the arc power.
 • When using a hybrid combination, the investment cost for the power source is significantly 
less and the electrical efficiency is much higher. A 1-kW reduction in the neodymium: 
yttrium–aluminium–garnet laser beam power leads to a reduction of approximately 35 kVA 
in the electric power consumed. Thus, using a 2-kW laser instead of a 4-kW laser is possible 
in the hybrid process, resulting in savings in initial investment outlays.
 • Larger tolerance of the joint configuration due to gap bridging with the added GMAW wire. 
Therefore the cost of edge preparation and of poor quality due to improper joint fit-up is 
negligible, which improves the overall economics of the hybrid process.
Figure 7.10 Hybrid welding in the Audi A8 roof area (green parts are made of sheet metal, 
red parts are casted structures and blue indicates extruded parts).
Helten (2003).
134 Welding and Joining ofAHSS
 • Good weld quality, with low and predictable distortion, is obtained, which implies a 
reduction in the need for rework. This potentially reduces the labour costs incurred from 
rectification work.
 • Introducing AHSS materials that are non-weldable by autogenous laser welding is possible 
using hybrid processes. This, in turn, helps reduce the auto body weight, making it more fuel 
efficient, greener and a safer vehicle at an affordable cost.
In conclusion, the hybrid laser–GMAW process combines the advantages of both 
arc and laser processes, resulting in high joint completion rates with increased tol-
erance to fit-up and without compromising joint quality and distortion control. The 
benefits to the industry include increased productivity, simplified setup procedures and 
reduced reworking costs after welding. Table 7.3 summarizes the economic advan-
tages of a welded component for the automotive industry (Staufer, 2009).
References
Ataufer, H. (2005). Laser hybrid welding and laser brazing: state of the art in technology and 
practice by the examples of Audi A8 and VW-phaeton. In: Proceedings of 3rd Interna-
tional WLT Conference on laser in manufacturing, 2005, Munchen (pp. 203–208).
Bagger, C., Flemming, O., & Olsen (February, 2005). Review of laser hybrid welding. Journal 
of Laser Applications, 17(1).
Beyer, E., Brenner, B., & Poprawe, R. (1996). Hybrid laser welding techniques for enhanced 
welding efficiency. In: ICALEO proceedings, section D, Detroit, USA.
Beyer, E., Imholff, R., Neuenhahn, J., & Behler, K. (1994). New aspects in laser welding with 
an increased efficiency. In: Proceedings of the laser materials processing – ICALEO ’94. 
Orlando, FL, USA: LIA (Laser Institute of America).
Brettschneider, C. (September, 2003). A8 meets DY. Eurolaser (Zeitschrift für die Industrielle 
Laseranwendung).
Campana, G., Fortunato, A., Ascari, A., Tani, G., & Tomesani, L. (2007). The influence of arc 
transfer mode in hybrid laser-mig welding. Journal of Materials Processing Technology, 
191, 111–113.
Table 7.3 Economic advantages of a welded component for the 
automotive industry
Parameters Benefits
Welding speed +30%
Shop floor space −50%
Wire consumption −80%
Shop floor staff −30%
Reduction of cost of materials Up to €7
Full penetration Fewer variants
Need for quality control Absolutely stable process
Staufer H. Industrial robotic application of laser-hybrid welding. Olsen, F.O. (Ed.): Hybrid Laser-arc Welding. 
Cambridge, Woodhead, 2009, 197–199.
135Hybrid welding processes in AHSS
Chatterjee, S., Mulder, R., van der Veldt, T. (2013). Effect of Laser-MIG hybrid welding param-
eters on properties of welded HSLA sheets for automotive applications. XII-2152–13.
Denney, P. (September, 2002). Hybrid laser welding for fabrication of ship structural compo-
nents. Welding Journal, 81(9), 58.
Dilthey, U., & Wieschemann, A. (1999). Prospects by combining and coupling laser beams and 
arc welding processes. International Institute of Welding. IIW Doc. XII-1565–99.
Dilthey, U., Wieschemann, A., & Keller, H. (2001). CO2-laser GMA hybrid and hydra welding: 
innovative joining methods for shipbuilding. LaserOpto, 33, 307.
Duley, W. W. (1999). Laser welding. New York: Wiley.
DVS – Deutscher Verband für Schweißen und verwandte Verfahren (Hrsg.): DVSMerk-
blatt 3216 – Laserstrahl-lichtbogen-hybridschweißverfahren. Düsseldorf, DVSVerlag 
(2005).
Gao, M., Zeng, X., & Hu, Q. (2007). Effects of gas shielding parameters on weld penetra-
tion of CO2 laser-TIG hybrid welding. Journal of Materials Processing Technology, 184, 
177–183.
Graf, T., & Staufer, H. (2003). Laser-hybrid welding drives VW improvements. Welding 
Journal, 82(1), 42.
Hansen (Ed.). (2012). ‘Join this’ laser community – The laser magazine from Trumpf (02) 
(p. 17).
Helten, St., (2003). Applikation des Laser-MIG-Hybrid- schweissverfahrens im Aluminium- 
karosseriebau des Audi A8. European Automotive Laser Applications conference, Bad 
Nauheim.
Ishide, T., Tsubota, S., & Watanabe, M. (2002). Latest MIG, TIG, arc-YAG laser hybrid welding 
systems for various welding products. In: First international symposium on high-power 
laser macroprocessing. SPIE.
Ishide, T., Tsubota, S., Watanabe, M., & Ueshiro, K. (2003). Latest MIG, TIG arc YAG laser 
hybrid welding systems. Journal of Japan Welding Society, 72(1), 22–26.
Kutsuna, M., & Chen, L. (2002). Interaction of both plasma in CO2 laser-MAG hybrid welding 
of carbon steel. IIW Doc.XII-1708–02.
Magee, K. H., Merchant, V. E., & Hyatt, C. V. (1991). Laser assisted gas metal arc weld charac-
teristics. In: Proceedings of the laser materials processing – ICALEO ‘90, Nov 4–9 1990 
(vol. 71). Boston, MA, USA: LIA (Laser Institute of America).
Merchant, V. (August, 2003). Shipshape laser applications. Status report. Industrial laser 
solutions.
Naito, Y., Mizutani, M., & Katayama, S. (2006). Electrical measurement of arc during hybrid 
welding – welding phenomena in hybrid welding using YAG laser and TIG arc. Quarterly 
Journal of Japan Welding Society (JWS), 24(1), 45–51.
Nilsson, K., Heimbs, S., Engström, H., & Alexander, F. H. (2003). Parameter influence in 
CO2-laser/MIG hybrid welding. IIW Doc. IV-843–03.
Ono, M., Shinbo, Y., Yoshitake, A., & Ohmura, M. (2002). Development of laser-arc hybrid 
welding. NKK Technical Review, 86.
Petring, D., Fuhrmann, C., Wolf, N., & Poprawe, R. (2003). Investigation and applications of 
laser arc hybrid welding from thin sheets up to heavy section components. In: Proceedings 
of the 22nd ICALEO’03 (pp. 1–10). Jacksonville: LIA. Section A.
Schubert, E., Wedel, B., & Kohler, G. (2002). Influence of the process parameters on the welding 
results of laser-GMA welding. In: Proceedings of ICALEO’02. Scottsdale: LIA (Session A – 
Welding).
Seyffarth, P., & Krivtsun, I. V. (2002). Laser-arc processes and their applications in welding 
and material treatment. London: Taylor & Francis.
136 Welding and Joining of AHSS
Staufer, H. (2003). Laser hybrid welding and laser brazing at Audi and VW. IIW, Doc. 
IV-847–03.
Staufer, H. (2009). Industrial robotic application of laser-hybrid welding. In F. O. Olsen (Ed.), 
Hybrid laser-arc welding (pp. 197–199). Oxford: Woodhead.
Steen, W. M. (2003). Laser material processing (3rd ed.). London, New York: Springer.
Steen, V. M., & Eboo, M. (1979). Arc augmented laser beam welding. Metal Construction, 97, 
332–335.
Steen., et al. (1978). Arc augmented laser beam welding. In 4th International Conference on 
advances in welding processes (pp. 257–265). Paper no. 17.
Sugino, T., Tsukamoto, S., Nakamura, T., & Arakane, G. (2005). Fundamental study on 
welding phenomena in pulsed laser-GMA hybrid welding. In: Proceedings of the 24th 
ICALEO (98) (pp. 108–116). Miami: LIA.
Tsuek, J., & Suban, M. (1999). Hybrid welding with arc and laser beam. Science and Technology 
of Welding and Joining, 4(5), 308–311.
Welding and Joining of Advanced High Strength Steels (AHSS). http://dx.doi.org/10.1016/B978-0-85709-436-0.00008-4
Copyright © 2015 Elsevier Ltd. All rights reserved.
Metal inert gas (MIG) brazing 
and friction stir spot welding of 
advanced high-strength steels 
(AHSS)
M. Shome
Research & Development, Tata Steel, Jamshedpur, India
8
8.1 Introduction
To meet weight reduction targets of automobiles, several new grades of steel have been 
developed over the past two decades under the broad category of advanced high-strength 
steels (AHSSs). These new steels include dual-phase (DP), transformation-induced plas-
ticity, ferrite–bainite, complex-phase and twinning-induced plasticity steels. AHSSs have 
tailored microstructures consisting of ferrite, bainite, martensite and retained austenite 
that provide an excellent combination of high strength and ductility (Kuziak, Kawalla, 
& Waengler, 2008; Zrnik, Mamuzic, & Dobatkin, 2006). To meet corrosion resistance 
requirements, a large percentage of these steels are produced in the form of galvanized 
and galvannealed products. Despite possessingexcellent mechanical properties, these 
steels require special precautions during forming and welding.
Welding is an integral part of automotive manufacturing, and therefore the weld-
ability of steels is an essential requirement for their application. The carbon equiv-
alent, microstructure, mechanical properties, type of zinc coating and thickness are 
key factors that require attention while selecting the joining process and establishing 
the joining conditions for AHSSs. Under the influence of the welding thermal cycle, 
the microstructures of the base metal are altered in the heat-affected zone (HAZ). 
Consequently, the mechanical properties are impaired locally, resulting in poor joint 
performance. For example, in DP steels the HAZ tends to soften with tempering 
of martensite. In transformation-induced plasticity steels the hardness of the weld 
increases significantly, and therefore the joint is likely to suffer brittle fracture under 
loading and would also exhibit poor fatigue properties.
Zinc is vaporized during laser welding or gas–metal arc welding (GMAW) of coated 
steels, causing porosities and exposing the surface to the atmosphere without sacri-
ficial protection. Interfacial fracture, electrode life and zinc loss are issues that arise 
during resistance spot welding (RSW) of bare and coated AHSSs. The performance 
and durability of an automotive structure largely depends on the design and quality of 
the joints (Davies, 2012, p. 248). For increased usage of AHSS, addressing weldability 
requirements and exploring alternative joining processes is therefore important. In that 
direction, this chapter focuses on the welding of DP steels using two entirely different 
processes, namely, metal inert gas (MIG) brazing and friction stir spot welding (FSSW).
138 Welding and Joining of AHSS
8.2 MIG brazing
Coated steel sheets encounter a number of difficulties during GMAW, also known as 
MIG/metal active gas welding. Excessive zinc evaporation leads to spatter, porosity for-
mation, non-uniform bead geometry and, most important, loss of protection against corro-
sion. All these problems lead to poor weld quality, increased cleaning costs after welding, 
more rework and lower productivity (Guimaraes, Mendes, Costa, Machado, & Kuromoto, 
2007; Holliday, Parkar, & Williams, 1995, 1996; Howe & Kelly, 1988; Parker, Williams, 
& Holiday, 1988). To alleviate these issues a gas metal arc brazing (or MIG brazing) 
process is recommended. This process combines the advantages of both GMAW (a high 
deposition rate) and conventional brazing processes (cold joining). Because consumables 
with low melting points are used, the high welding speed and the low thermal input ensure 
stable operation and good joint properties in terms of appearance, HAZ microstructure, 
strength and resistance against corrosion (Quintino, Pimenta, Iordachescu, Miranda, & 
Pépe, 2006). MIG-brazed joints demonstrate good fatigue properties that ensure reliable 
performance during a vehicle’s life cycle (Lepisto & Marquis, 2004). During brazing, the 
torch is usually maintained at a 70° travel angle and a 20° working angle and is traversed 
along the edge of the upper sheet using both push and pull modes (Figure 8.1). Argon 
gas shielding is preferred. While this process has been successfully applied to galvanized 
low-carbon steels, its applicability to AHSSs is yet to be fully explored. In that direction, 
investigations assessing the performance of coated DP steel when MIG brazed were 
carried out.
8.2.1 Experimental methodology
The composition and mechanical properties of galvanized DP590 steel and MIG braz-
ing filler wire (CuAl8) are shown in Tables 8.1 and 8.2, respectively. The thickness of 
Direction of
MIG brazing 70°
Base metal - 
Base metal - 
Direction of
MIG brazing
Push mode
Direction of
MIG brazing
Pull mode
200 mm
120 mm
100 mm
Filler
wire
20°
Base metal
Base metal
BrM
Figure 8.1 Metal inert gas (MIG) brazing setup and procedure.
139Metal inert gas (MIG) brazing and friction stir spot welding of AHSS
the steel was 1.4 mm. As shown in Figure 8.1, MIG brazing was done in a lap joint 
configuration over a length of 300 mm using pulse-synergic conditions, wherein the 
current and the welding speed varied. The parameters for joining the steel are given in 
Table 8.3.
Samples for metallographic, microhardness, shear tensile and high-cycle fatigue 
testing were sliced from the lap joint assembly. For microscopy the specimens were 
polished and etched using 2% nital. The specimens were examined under an optical 
microscope and a field emission scanning electron microscope equipped with energy 
dispersive X-ray spectroscopy. Micro-hardness was traversed from the braze metal to 
the base metal under a load of 100 gf using a standard Vickers micro-hardness testing 
Table 8.1 Chemical composition and mechanical properties of 
galvanized dual-phase steel
Chemical composition (wt%) Mechanical properties
Coating 
thickness 
(μm)Carbon Manganese Silicon Titanium
Yield 
strength 
(MPa)
Ultimate 
tensile 
strength 
(MPa)
Braking 
load 
(kN) %El
0.097 1.631 0.24 0.002 386 626 17.53 24.5 19
UTS, Ultimate tensile strength; %El, %Elongation.
Table 8.2 Chemical composition and mechanical properties of the 
filler wire
Wire 
type
Wire 
diameter 
(mm)
Chemical composition (wt%) Ultimate 
tensile 
strength 
(MPa)
Hardness 
(Hv)Copper Silicon Manganese Aluminium
CuAl8 1.0 Bal. 0.113 0.318 8.01 450 100
Bal., Balance.
Table 8.3 Metal inert gas brazing parameters
Specimen
Current 
(A) Voltage (V)
Wire feed 
speed 
(m/min)
Welding speed 
(mm/min)
Heat input 
(J/mm)
Welding 
mode
DP1 108 18.0 5.0 600 136 Push
DP3 108 18.0 5.0 400 204 Push
DP3 128 19.0 6.0 600 170 Push
DP4 108 18.0 5.0 400 204 Pull
140 Welding and Joining of AHSS
machine. The sliced lap joint samples were machined to prepare standard shear tensile 
test specimens according to the DIN EN 10002-1 standard (Figure 8.2). The tensile shear 
tests were carried out in a universal testing machine with a 100-kN capacity at a cross-
head speed of 0.5 mm/min. Three samples were tested for each heat input to evaluate 
the joint strength. High-cycle fatigue testing was performed on the tensile shear samples 
using a 50-kN resonant testing machine. For the tension–tension tests, the maximum 
load was equivalent to 80% of the ultimate quasi-static load under a load ratio of 0.1. 
The maximum load, minimum load, maximum and minimum crosshead displacement 
and the frequency were monitored during fatigue testing using data acquisition software.
8.2.2 Bead geometry and microstructure
A schematic representation of the weld joint is shown in Figure 8.3. As demonstrated in 
Figure 8.4, the actual dimensions of the weld bead depend on the welding parameters. Fig-
ure 8.5 shows that the width (W), leg length (L) and cross-sectional area (A) increase with 
increasing current. At the same time, the bead height (H) and wetting angle (θ) decrease. 
This is expected because the flowability of the melt increases with increasing temperature, 
causing a wider bead. The bead geometry varies with welding direction; in push mode 
(DP2) the bead is wider and flatter, whereas in pull mode (DP4) the bead is narrow and 
raised for the same heat input (204 J/mm). Lower capillary pressure associated with the 
push mode increases the wetting angle and produces a wider bead. The extent of zinc loss 
at the back side depends on the heat intensity. The zinc loss is least in DP1. Despite DP2 
having a higher heat input than DP3, the zinc loss is less at the back of the sheet because 
the current is of a low order. Therefore the brazing heat intensity, which is directly related 
to the welding current, is critical in determining the extent of zinc evaporation.
The microstructure at different locations of the weld joint (Figure 8.6) depends on 
the peak temperature attained and the subsequent cooling rate. While DP steel has 
dispersed martensitephases in a predominantly ferrite matrix, the HAZ contains lath 
martensite or bainite. The higher cooling rate (40–120 °C/s) produces these phases 
during phase transformation from austenite (Gould, Khurana, & Li, 2006). Hardness 
greater than 300 Hv confirms the presence of hard phases in the coarse grain HAZ 
(CGHAZ) compared with a base metal hardness of ∼180 Hv. While higher peak tem-
perature (Tp) is responsible for coarsening of austenite grains, the ensuing rapid coo-
ing condition enforces displacive transformation to form the hard phases.
100
120
35 130 35
20
20
30
1.4
Figure 8.2 Schematic diagram of a shear tensile specimen. All dimensions are in millimetres.
141Metal inert gas (MIG) brazing and friction stir spot welding of AHSS
(a)
(b)
C L
W
A
WM BM
d
θ
BH
Dilution
of BM
Weld metal
Base metal HAZ
HAZ Upper sheet
Lower sheet
D0.5 mm
a
CGHAZ
cb
FGHAZ
Figure 8.3 Schematic diagram of a cross-section of a metal inert gas-brazed bead (a) and 
weld metal and the heat-affected zone (HAZ) (b). BM, base metal; CGHAZ, coarse grain 
heat-affected zone; FGHAZ, fine grain heat-affected zone; WM, weld metal.
Figure 8.4 Effect of heat input on the bead appearance of metal inert gas-brazed joints.
Distinct dendrites are observed in weld metals with copper matrix (Figure 8.6). 
These dendrites predominantly are supersaturated solid solutions of copper in iron, 
which are formed by localized melting of the base metal and mixing with the mol-
ten copper, and remain scattered because of rapid cooling during solidification. The 
142 Welding and Joining of AHSS
dendritic constituents in welds made with higher heat input have a larger size and 
higher density because of the more dissolution of iron from the base metal by the 
Marangoni effect. The iron in the dendrites is responsible for the high hardness of the 
weld metal.
The energy-dispersive spectroscopic analysis shown in Figure 8.7 reveals that the weld 
metal matrix essentially consists of a copper–aluminium alloy where the copper content 
θ
Figure 8.5 Bead profile of metal inert gas-brazed joints at different heat inputs. H, height; 
L, length; W, width.
Figure 8.6 Field emission scanning electron micrographs of metal inert gas-brazed joints of 
different heat inputs. CGHAZ, coarse grain heat-affected zone; Fe, iron; FGHAZ, fine grain 
heat-affected zone; WM, weld metal.
143Metal inert gas (MIG) brazing and friction stir spot welding of AHSS
is about 85% and aluminium content is about 6%. The iron content in the iron-rich den-
drites is about 80%. Some copper and aluminium also were found in the dendrites that are 
retained during the solidification process. The interface zone between the weld and the 
HAZ consists of mainly iron with some aluminium and copper. The interface thickness 
increased from 5.22 μm in DP1 to 6.07 μm in DP2 with an increase in heat input. Again, for 
the same heat input (e.g. 204 J/mm), the interface thickness was higher for pull mode than 
push mode. This indicates that more heat used in pull mode resulting in a thicker interface. 
The dendrite volume fraction, iron content in the dendrite and interface thickness are 
significant because they influence the strength of the weld metal and the overall perfor-
mance of the weld joint.
In the MIG welding process the weld metal attains high hardness values because 
steel consumables are used. In the MIG brazing process the copper-based consum-
ables are of low strength; hence the hardness of the weld metal is of the order of that 
of the base metal (Figure 8.8). That also occurs because of the iron-rich dendrites in 
the weld metal. Compared with the weld metal and the base metal, the HAZ has the 
highest hardness because of its high martensite content.
8.2.3 Mechanical properties
Shear tensile properties: Tensile properties corresponding to different MIG brazing 
parameters are listed in Table 8.4. When the heat input is increased from 136 to 204 J/
mm by decreasing the welding speed from 600 to 400 mm/min, the joint strength 
increases. When the heat input is increased from 136 to 170 J/mm by increasing the 
current from 108 to 128 A, the joint strength is significantly reduced. To explain this 
discrepancy, the load-bearing capacity of the joint needs to be considered from the 
perspective of bead geometry. In particular, the height of the bead H plays a major 
Figure 8.7 Energy-dispersive spectroscopic analysis of metal inert gas-brazed joints corre-
sponding to different parameters. Al, aluminium; Cu, copper; Fe, iron; HAZ, heat-affected 
zone; WM, weld metal.
144 Welding and Joining of AHSS
role in determining the strength of the joint. A bead with a smaller H/W value is 
likely to fail in the weld, and a larger value could lead to interfacial fracture. In DP1, 
that is, in the joint with the lowest heat input, the leg length L is small and hence 
there is insufficient bonding between the weld metal and the parent metal; therefore 
failure occurs through the interface under lower tensile loads (Figure 8.9). The best 
performance is provided by DP2 because the H and L values are large and the H/W 
ratio of 0.6 is favourable. In this case the joint efficiency is as high as 98% and fail-
ure occurs in the HAZ. In DP3, however, failure takes place at the weld because of 
the smaller H value. For the same heat input (204 J/mm), push mode (DP2) shows 
greater strength than pull mode (DP4). Push mode has a lower wetting angle than 
pull mode (DP4) and results in a longer L and a shorter H (Figure 8.5).
Fatigue properties: High-cycle fatigue results represented by the load (S) amplitude 
versus the number of cycles (N) to failure curve in Figure 8.10 indicate that the endurance 
350
300
250
200
150
100
0 2000 4000 6000 8000 10,000
Distance (microns)
DP1 DP2 DP3 DP4
WM HAZ BM
H
ar
dn
es
s 
(H
ν)
Figure 8.8 Micro-hardness profile of a metal inert gas-brazed joint (top sheet). BM, base 
metal; HAZ, heat-affected zone; WM, weld metal.
Table 8.4 Shear tensile test data for metal inert gas-brazed joints
Specimen
Heat input 
(J/mm)
Tensile load 
(kN)
Location of 
failure
Joint 
efficiency (%)
Welding 
process
DP1 136 15.88 Interface 91 Push mode
DP2 204 17.21 Heat- 
affected 
zone
98 Push mode
DP3 170 14.84 Weld metal 85 Push mode
DP4 204 16.84 Interface 96 Pull mode
145Metal inert gas (MIG) brazing and friction stir spot welding of AHSS
limit of 2 × 106 cycles were attained at 10% of the tensile load. This is irrespective of the 
bead’s geometry. However, joints receiving greater heat input were able to withstand 
more cycles. At 60–80% tensile loading, fatigue failure occurred along the (1) interface 
in joints made with high heat input (e.g. DP1), (2) weld metal for intermediate heat input 
(e.g. DP2, DP3) and (3) in the HAZ for lower heat input (e.g. DP4). At a lower load, 
irrespective of the heat input, all joints failed in the HAZ.
During fatigue testing, failure can occur at any one of the following three locations: 
the interface, the weld metal or the HAZ. As mentioned earlier, the weld geometry 
predominantly determines the type of failure. The weld root between two overlapping 
sheets acts as the default notch with stress concentration. In case of interfacial failure 
cracks are initiated at the weld root and propagate through the interface towards the 
weld toe. Again, in case of weld metal failure, the crack initiates at the weld root and 
propagates through the weld metal in a direction perpendicular to the applied load. For 
HAZ failure, however, the crack initiates from the weld toe and propagates through the 
fine grain HAZ (FGHAZ) across the sheet thickness (Figure 8.11). Small cracks may 
Figure 8.9 Cross-sectional view indicating failure location under quasi-static loading.
146 Welding and Joining of AHSS
originate from several spots along the weld toe or weld root. They subsequently grow 
and coalesce to become larger cracks (Lassen & Recho, 2006). The remaining ligamentof the sheet or weld section eventually becomes too small to bear the load and failure 
takes place. This notch effect is more pronounced at higher loads as the gap Crack 
opening displacement between the two sheets increases during cyclic loading.
8.3 Friction stir spot welding (FSSW)
FSSW is a relatively new process that recently received considerable attention from 
the automotive industry. FSSW has proven to be a cost-effective and productive means 
for joining light materials such as aluminium (Gerlich, Su, & North, 2005). This 
Figure 8.10 High-cycle fatigue plots of metal inert gas-brazed joints.
Figure 8.11 Fracture location after high-cycle fatigue test. BM, base metal; COD, crack 
opening displacement; HAZ, heat-affected zone; WM, weld metal.
147Metal inert gas (MIG) brazing and friction stir spot welding of AHSS
process avoids the severe heating and cooling rates experienced during RSW. It is an 
attractive technology for spot welding of high-strength and AHSSs. FSSW of steel is 
usually carried out using a cylindrical polycrystalline boron nitride (PCBN) tool with 
a convex, scrolled shoulder and a protruding pin, as shown in Figure 8.12. The tool 
is plunged into two overlapping sheets at a specific rate to a predetermined depth. 
The frictional heat generated by the interaction between the tool and material softens 
the metal, and the rotating pin causes material to flow in both circumferential and 
axial directions. The scrolls on the shoulder are such that when the tool is rotated in 
a counterclockwise direction, the scrolls assist in moving the material from the outer 
periphery of the shoulder towards the central pin. The tool is then retracted rapidly 
either immediately or after a dwell time. The rotational speed of the tool, plunge rate, 
plunge depth and dwell time are the four principal parameters in FSSW. The pressure 
applied by the tool shoulder enhances the stirring effect and produces an annular sol-
id-state bond around the pin.
Between 2000 and 2010 there have been successful attempts to join AHSSs 
by the friction stir process; however, the tool life and weld quality are still being 
assessed for widespread commercial purposes. Feng et al. (2005) reported that 
solid-state joints were produced in 1.6-mm-thick DP600 steel applying 1500 rpm 
of tool speed with a weld time varying between 1.6 and 3.2 s just by changing the 
plunge rate. The bond strength increased with increasing weld time as the width 
Figure 8.12 (a) Top view of a polycrystalline boron nitride tool. (b) Schematic detailed view 
of the tool tip. CCW, counter clockwise.
148 Welding and Joining of AHSS
of the bonding ligament became larger. Interestingly, the thermo-mechanically 
affected zone (TMAZ) exhibited a microstructure and hardness similar to that of 
the base metal. Hovanski et al. (2007) successfully lap-joined hot-stamped boron 
steel by applying a rotational velocity of 800–2000 rpm and a weld cycle time of 
1.9–10.5 s. Longer dwell time resulted in a direct increase in lap shear strength 
of 40–90% for all plunge rates. The effect of rotational speed on weld strength 
was dependent on the plunging conditions. The original microstructure containing 
martensite was mostly retained, except for a thin region of ferrite that formed at 
the interface region within the bond area. Cracks around the nugget propagated 
through this softer region.
A comparative study between RSW and FSSW of 1.2-mm-thick zinc-coated DP600 
steel revealed that the microstructure of the HAZ is similar in both cases. Martensite is 
observed in the fusion zone of RSW and stirred zone (SZ) of FSSW, but with different 
morphologies (Khan et al., 2007). However, the TMAZ contains a mixture of lath mar-
tensite, bainite and ferrite. Furthermore, in both processes failure load increases with 
an increase in nugget size or bond area, which in turn depends on the energy input. 
Aota & Ikeuchi (2009) observed that failure load in thin, low-carbon sheets increased 
with plunge depth, and failure mode changed from interface rupture to plug rupture 
at plunge depths greater than 0.16 mm. The failure load corresponding to plug rupture 
conditions increased with dwell time and was almost completely unchanged over 0.4 s 
at a plunge depth of 0.14 mm.
The body of information available, however, does not mention processing param-
eters that can produce commercially feasible bond sizes equivalent to nugget sizes of 
RSW. The work mentioned in the subsequent section attempts to address this issue. 
FSSW of 1.6-mm-thick DP590 steels was carried out with the intent to produce a 
small bond with adequate mechanical properties. Efforts to evaluate and fine-tune 
parameters based on real-time thermo-physical response of the material during weld-
ing and to investigate microstructural characteristics and mechanical performance of 
the joints also were made.
8.3.1 Welding of DP590 steel
The composition and mechanical properties of DP590 steel are listed in Table 8.5. 
The PCBN tool used for welding had a shoulder diameter of 25 mm and a pin height 
of 1 mm with a base diameter of 3–4 mm. Lap shear tensile specimens of 175 × 45 mm 
with an overlap of 35 mm, as shown in Figure 8.13, were considered for testing and is 
shown in Figure 8.13. Two spacers 40 mm in length were attached to both ends of the 
specimen to induce pure shear and to avoid initial realignment during testing (Figure 
8.12).
8.3.2 Processing parameters and mechanical response
The parameters considered for lap welding are listed in Table 8.6. The welding 
cycle in FSSW begins as soon as the tool makes contact with the steel surface. 
As the pin enters the first sheet, the material gets work-hardened, and thereby 
149Metal inert gas (MIG) brazing and friction stir spot welding of AHSS
the force required to stir the material increases. The interaction between the tool 
and the material involves energy, which is calculated using the following formula 
(Khan et al., 2007):
 
QFSW =
N∑
n = 1
F (n) · [x (n) − x (n − 1)] +
N∑
n = 1
T (n) ·ω (n) ·Δ t
 
where F is the experimentally measured normal force, x is the displacement, T is the 
axial torque and ω is the angular velocity (2π*RPM/60). The values for the parameters 
given in Table 8.6 are plotted in Figure 8.14.
As shown in Table 8.6, the bond (nugget) diameters obtained are acceptable by 
RSW standards. From the various parameter combinations attempted, it can be said 
that a high rotational speed of 1600 rpm applied for a longer time of 72 s (i.e. a 
feed rate of 2 mm/min) leads to large nugget diameters (>11 mm). For a feed rate 
of 10 mm/min, the nugget diameter reduces to 4.7 mm, but the welding time is 20 s. 
By judiciously adopting higher feed rates (228 or 300 mm/min) along with a higher 
rotational speed (2400 rpm), however, obtaining nuggets that are of appropriate size 
within a short time of ∼4 s is possible (Sarkar, Pal, & Shome, 2014). Since the feed 
rate is extremely high, a dwell time of 1 s at the end of the plunging stage ensured 
effective joining.
The forces acting on the tool are the x-, y- and z-forces; however, the z-force is 
most critical because the tool penetrates along that direction. During this process, 
Table 8.5 Composition and mechanical properties of DP590 steel
Composition (wt%) Mechanical properties
Carbon Manganese Silicon
Ultimate 
tensile 
strength 
(MPa)
Yield 
strength 
(MPa)
Elongation 
(%)
0.009 0.98 0.31 617 365 29
Figure 8.13 Dimensions of the lap shear tensile test specimen for friction stir spot welding.
150 Welding and Joining of AHSS
heat is generated by the friction between the rotating pin and the workpiece, as well 
as by adiabatic heat during plastic deformation of the material, which is reflected in 
the decrease in the z-force (Figure 8.15). The softened material is displaced and the 
progressing pin encounters a fresh layer of material. Fourment & Guerdoux (2008) 
showed, through numerical simulation, that the maximum temperatureof the work-
piece is located at the bottom of the pin. This causes the material under the pin to 
soften, facilitating tool progression (Khan et al., 2007). With increasing rotations per 
minute, the deformation as well as heat generation increase. As a result, the material 
is thermally softened more quickly with increasing rotations per minute. This effect 
can be seen in the z-force curves, where the first peak occurs sooner and at a lower 
load with increasing rotations per minute. The ensuing thermo-mechanical condition 
enables solid-state diffusion between the discretely mixed solid entities within the 
SZ. Because of the prevailing high strain and strain rate, the dynamic recrystallization 
process also becomes active.
As the pin comes in contact with the second sheet, the z-force starts ris-
ing again for the reasons stated above. This is the cause of the second peak in 
Figure 8.15(a).The rise and fall of the z-force is more pronounced in case of a 
350
300
250
200
150
100
50
350
(b)(a)
300
250
200
150
100
50
2 4 6 8 10400
To
ta
l e
ne
rg
y 
(k
J)
To
ta
l e
ne
rg
y 
(k
J)
600 800 1000 1200 1400
Rotational speed (rpm) Feed rate (mm/min)
1600
Figure 8.14 Change of energy with rotational speed (rotations per minute [rpm]) (a) and feed 
rate (b).
Table 8.6 Correlation between welding parameters and bond 
diameter in lap welding
Depth of 
penetration 
(mm)
Rotations 
per minute
Feed rate 
(mm/min)
Dwell 
time (s)
Weld 
time (s)
Bond 
diameter 
(mm)
2.2 400 2 0 72 4.3
800 7.8
1200 11.7
1600 11.5
2.2 1600 10 0 20 4.7
2.4 2400 228 1 4 5.1
151Metal inert gas (MIG) brazing and friction stir spot welding of AHSS
400-rpm weld. At higher rotations per minute, the bottom sheet is further soft-
ened and hence there is less variation in z-force. Material softening and pin 
immersion causes upward displacement of the extruded material. The axial 
force increases when the tool shoulder contacts the extruded material. The 
Figure 8.15 Force (a, c) and torque (b) plots for an identical feed rate, but different rotational 
speeds (rpm).
152 Welding and Joining of AHSS
thermal expansion associated with heating of the metal adds to this effect. The 
peak force is caused by the extruded material squeezing between the tool shoul-
der and the workpiece. Therefore, the z-force peaks at the last stage of plung-
ing when the rotating tool shoulder is in firm contact with the workpieces. 
A relatively smooth z-force plot suggests that the FSSW process is more stable at 
higher rotations per minute.
The spikes in x-force and y-force plots are caused by tool vibrations while pene-
trating through the metal. Such vibrations reflect slow and delayed heating and cau-
tions for adjustment of welding parameters. With increasing rotations per minute, 
sufficient heating followed by softening takes place early and is sustained through-
out. Consequently, the spikes are reduced as the process stabilizes, for example, in 
the case of 1600 rpm. Again, when the shoulder comes in contact with the extruded 
material, the x and y directional forces encounter some oscillations. These are prob-
ably caused by transversal load variations on the tool due to inadequate contact 
between the extruded material and the shoulder (Davies, 2012, p. 248). It has been 
observed that the z-force and spindle torque for the 400-rpm weld is significantly 
different from that obtained with higher rotations per minute but an identical feed 
rate (Zimmer, Langlois, Laye, & Bigot, 2010). This can be explained by the frictional 
heat input at the beginning of the plunge, at 400 rpm, being insufficient to cause 
proper stirring. This occurs because the tool experiences more resistance from the 
material at such low rotational speeds.
Increasing rotations per minute marginally reduces the torque (Figure 8.15(b)) 
and gradually attains a steady state of operation. At higher feed rates, the ini-
tial work-hardening rate is high because the workpiece is at ambient temperatures 
(Figure 8.16). However, fewer vibrations are created by the tool–metal interaction 
at higher feed rates (Figure 8.16(c)) because of better process stability. Of the two, 
rotational speed has a greater influence on process stability than feed rate. The 
high force and torque values observed at numerous rotations per minute (2400) 
and a high feed rate (228–300 mm/min) suggest that, with a very short welding 
time (∼4 s), the material offers substantial resistance to stirring because it does not 
soften to the extent reported earlier for lower parameters (Figure 8.17). Because 
the depth of penetration is greater, the tool shoulder meets the material early and 
contributes to maximum heat generation. This results in the stirring of a larger 
volume of material, and hence much resistance is encountered, which is reflected 
by the higher torque and z-force values. It may be noted, however, that assuming 
the spot size requirement of friction stir spot welds to be 3.5–5√t, as in the case 
of spot welding, the high parameter conditions are more favourable. They produce 
appropriately sized spots within a time cycle that is productive and close to the 
RSW nugget size.
8.3.3 Structure–property correlation
The cross-section of an FSSW joint is shown in Figure 8.18, wherein the following 
four zones are observed: (1) the SZ, (2) the TMAZ, (3) the HAZ and (4) the base metal. 
It may be noted that the dimension of each of the zones increases with increasing heat 
153Metal inert gas (MIG) brazing and friction stir spot welding of AHSS
input. While the rotations per minute increase the heat input, the feed rate decreases 
the heat input as well as the weld time.
The microstructure of the DP590 base metal is shown in Figure 8.19; it has a DP 
microstructure containing islands of hard martensite embedded in a softer ferritic matrix. 
Figure 8.16 Force (a, c) and torque (b) plots for identical rotational speeds but different feed 
rates.
154 Welding and Joining of AHSS
Figure 8.17 Force (a, c) and torque (b) plots for different rotational speeds and feed rates. dt, 
dwell time.
155Metal inert gas (MIG) brazing and friction stir spot welding of AHSS
The microstructures of the various zones shown in Figure 8.18 are given in Figures 
8.19–8.21. The microstructure in the SZ consists of fine grains of ferrite (Figure 8.20(a)). 
Traversing farther inside, from the SZ through the TMAZ, the grain size progressively 
becomes larger. This microstructural variation is consistent with the strain and tempera-
ture gradient that develops along the thickness of the sheet from the surface as an effect 
of stirring (Zimmer et al., 2010). With increasing rotations per minute, the grain size of 
the SZ increases. The TMAZ microstructure consists of blocky ferrite at fewer rotations 
per minute (Figure 8.19 (b)). The microstructure produced at higher rotations per minute, 
Figure 8.18 Schematic profile superimposed on macrograph of FSS weld – cross-section 
view. Base metal (BM), thermo-mechanically affected zone (TMAZ), heat-affected zone 
(HAZ), and stirred zone (SZ).
Figure 8.19 Microstructure of the base metal (BM), thermo-mechanically affected zone 
(TMAZ), heat-affected zone (HAZ) and stirred zone (SZ) (800 rpm, 2 mm/min feed rate).
156 Welding and Joining of AHSS
however, shows an increasing amount of bainite/acicular ferrite structure (Figure 8.20(b)). 
With increasing rotations per minute the grain size of the TMAZ increases. For 1200–
1600 rpm, as one travels from the TMAZ to the HAZ, the microstructure shows an 
increasing amount of bainite/acicular ferrite and a decreasing amount of ferrite.
The microstructure in the subregions of the HAZ tend to develop in relation to the local 
thermal cycle experienced during welding. The HAZ exhibits a CGHAZ surrounding the 
TMAZ, an FGHAZ encompassing the CGHAZ and an inter-critical HAZ encompassing 
the FGHAZ. The HAZ in general has a finer structure than the base metal, consisting 
primarilyof polygonal ferrite and pearlite (Figures 8.19 and 8.20). With increasing rota-
tions per minute, the grain size of the HAZ increases. The microstructures of the TMAZ 
and HAZ welded at higher feed rates are considerably finer than the ones welded with 
the same rotations per minute but a lower feed rate (Figures 8.20 and 8.21). The TMAZ 
shows a volume fraction of acicular ferrite and bainitic sheaves that are oriented ran-
domly with respect to one another. There are also traces of martensite in the TMAZ of 
DP specimens welded with the maximum parameters.
Metallographic evidence also suggests that at maximum parameters a thin region 
of ferrite is formed, originating from the interface of the two-sheet stack to a location 
near the pinhole at the centre of the nugget. This distinctive band of ferrite remains 
Figure 8.20 Microstructure of the stirred zone (a), thermo-mechanically affected zone (b) and 
heat-affected zone (1600 rpm, 2 mm/min feed rate) (c).
157Metal inert gas (MIG) brazing and friction stir spot welding of AHSS
along both sheet surfaces, as well as throughout the weld nugget. The soft band of 
ferrite is shown in Figure 8.22. It may be noted that as we move towards the central 
pin hole, the coarse ferritic grains are gradually replaced by fine ferrite grains. The 
micro-hardness of the TMAZ and HAZ increases with increasing rotations per minute 
(up to 1200 rpm); this can be observed in Figure 8.23. The TMAZ region of the weld 
has a hardness above 210 Hv. The hardness of the HAZ also increases with an increas-
ing feed rate. When compared with fewer rotations per minute and a lower feed rate, 
DP590 shows considerably high hardness values (302 Hv) under 2400 rpm and 228 mm/
min feed rate (Figure 8.23(c)). Interestingly, the maximum parameters cause softening 
in the HAZ.
It has been reported that more rotations per minute result in thermal cycles with a higher 
peak temperature (Tp) (Cui, Fujii, Tsuji, & Nogi, 2007; Fourment & Guerdoux, 2008; 
Khan et al., 2007). Again, with increasing feed rate the heat input and Tp decrease. 
This is responsible for the coarser microstructure in the different zones that is obtained 
at higher rotations per minute, whereas a finer structure is observed with higher feed 
Figure 8.21 Microstructure of the thermo-mechanically affected zone at 1600 rpm, 10 mm/min 
feed rate (a) and 2400 rpm, 228 mm/min feed rate, 1 s dwell time (b) and the heat-affected zone 
(2400 rpm, 228 mm/min feed rate, 1 s dwell time) (c).
158 Welding and Joining of AHSS
Figure 8.22 (a) Macrostructure of friction stir spot weld in DP590 steel showing the interfacial 
ferritic band. (b) Magnified view of region A in panel (a).
Figure 8.23 Micro-hardness plots for different rotations per minute (rpm) (a), different feed 
rates (b) and maximum parameters (c).
159Metal inert gas (MIG) brazing and friction stir spot welding of AHSS
rates. The heavy local deformation in the SZ is associated with a temperature increase 
up to 1100–1200 °C (Hovanski et al., 2007; Lienert, Stellwag, Grimmett, & Warke, 
2003; Zimmer et al., 2010). This triggers dynamic strain-induced recrystallization 
followed by rapid cooling with the withdrawal of the tool. Consequently, a fine fer-
rite microstructure is produced upon transformation from austenite. The strain and 
peak temperatures of the thermal cycles decrease with the depth of the sheet, and 
so does the phase transformation conditions. The TMAZ region experiences high 
Tp (<1100 °C) and longer time spent at a high temperature. This results in varying 
degrees of austenite grain coarsening. Microstructural evidence indicates that the Tp 
in the TMAZ attains the temperature of austenite, effecting appreciable grain growth. 
The final microstructure of the TMAZ depends on the effects of strain, strain rate, 
temperature and cooling rate.
Mechanical stabilization of austenite causes the amount of bainite to vary from 
top to bottom in the TMAZ (Larn & Yang, 2000; Lee, Bhadesia, & Lee, 2003). 
When an externally applied stress exceeds the yield strength of austenite, it is pos-
sible that the transformation of austenite into bainite is retarded. This is because 
displacive transformations occur by the advance of glissile interfaces, which can 
be hindered or rendered sessile upon encountering defects such as dislocations or 
grain boundaries. Such defects act as obstacles to the migration of the interface into 
austenite, similar to the effects that lead to work hardening when the passage of slip 
dislocation is obstructed. If the density of the dislocations is increased by deforming 
the austenite plastically, then these dislocations would limit the growth of the bainite 
plates. The final fraction of bainite may then become smaller in the deformed aus-
tenite. If phase transformation immediately follows deformation, ferrite is nucleated 
intragranularly at the places with highest dislocation density, thus resulting in grain 
refinement, as seen in the SZ (Hickson, Hurley, Gibbs, Kelly, & Hodgson, 2002). 
Under these circumstances, ferrite nucleates on the unrecovered dislocation sub-
structures in the austenite grains. If there is a delay between deformation and phase 
transformation, however, recovery alleviates the substructure from becoming the 
preferred site for ferrite nucleation.
In the upper region of the TMAZ, the strain generated by stirring is con-
siderably higher than in the lower region, resulting in a structure consist-
ing primarily of ferrite with some amount of bainite. The lower part of the 
weld experiences lower strains and also greater undercooling, as the bot-
tom surface dissipates heat to the backing plate, which acts as a heat sink. 
This leads to a larger amount of bainite. Sluggish transformation kinetics pre-
vail in this region. The microhardness of the TMAZ represents the bainitic– 
acicular ferritic structures found. At 1600 rpm, the microhardness decreases 
because of excessive grain coarsening. The higher hardness values at the border 
of the TMAZ are caused by rapid cooling. This is probably because of the geom-
etry of the FSSW process, wherein the temperature of the entire TMAZ region 
is expected to be high; therefore, the highest cooling rates are at the edges of the 
TMAZ, corresponding to higher hardness (Reynolds, Tang, Posada, & Deloach, 
2003).
The HAZ experiences considerably lower Tp than the TMAZ and hence shows a finer 
grain size. Material in the CGHAZ experiences the highest temperature in the HAZ. 
160 Welding and Joining of AHSS
The microstructure suggests that Tp was well above the effective A3 temperature, thus 
allowing some austenite grain growth. Farther away, the temperature experienced by the 
FGHAZ is less than the A3 temperature. The decomposition of austenite to ferrite and 
pearlite upon cooling promotes fine grains in this region. The inter-critical HAZ was 
characterized by a bimodal distribution of fine ferrite grain sizes surrounded by coarse 
grains as it was exposed to temperatures in the two-phase inter-critical region. In the case 
of samples welded at a higher feed rate (Figure 8.21), a characteristic lower Tp (because 
of less available time and low heat input), followed by a high cooling rate, result in a rel-
atively finer structure in all the zones. The rapid cooling associated with low heat input 
conditions enable the formation of fine bainite, acicular ferrite and traces of lath marten-
site in the TMAZ of DP steel. The high micro-hardness values are due to the presence of 
fine low-temperature transformation products, including martensite (>300 Hv).
8.3.4 Mechanical properties
Lap shear specimens of different FSSW joints, shown in Figure 8.13, were first tested 
under quasi-static loading conditions. The tensile test results of weld joints are summa-
rized in Table 8.7, which shows that the nugget diameter and maximum load increase with 
increasing rotational speed (rotations per minute) of the tool and decreases with an increas-
ingfeed rate. Compared with the minimum prescribed breaking load of spot welds (Ref: 
BS1140:1993), the friction stir spot welds exhibited considerably higher breaking loads.
Some of the tensile tests were interrupted to investigate crack propagation during 
shear tensile loading. A transverse cross-section of a partially failed weld specimen 
is demonstrated in Figure 8.24, showing crack propagation from the sheet interface 
along the thin ferrite region within the weld nugget. Examination of partially failed 
DP590 FSSW specimens show consistent failure along this softened region of fer-
rite within the weld nugget. This ferrite band provides an easy route for failure with 
reduced strength. Failure ultimately takes places when the crack traverses the entire 
weld and reaches the central depression of the pin hole. The path of the final fracture 
is marked with dashed lines in Figure 8.24(a). Higher rotations per minute ensures 
Table 8.7 Shear tensile test results of friction stir spot welds of 
DP590 steel
Welding parameters
Dwell 
time (s)
Nugget 
diameter 
(mm)
Minimum 
acceptable 
breaking 
load (kN)
Breaking 
load (kN)
Penetration 
depth (mm)
Rotations 
per minute
Feed rate 
(mm/min)
2.2 400 2 0 4.30 11.65 18.6
1200 11.76 23.0
1600 11.51 24.3
1600 10 4.75 21.8
2.4 2400 228 1 5.14 12.61 23.7
161Metal inert gas (MIG) brazing and friction stir spot welding of AHSS
favourable temperatures and greater intermixing and hence results in higher failure 
loads. Increasing the feed rate decreases the weld time and also the peak tempera-
ture, resulting in decreasing failure loads. For optimized parameters, the tensile crack 
initiates at the original notch tip and propagates along the sheet interface into the 
weld along the softer ferrite band; it finally reaches the central depression of the pin 
hole, causing shear failure. The crack propagates along the ferrite band because of the 
favourable stress concentration and the presence of a region with less hardness.
Based on the average failure loads under quasi-static loading, welded samples of 
DP590 steel were subjected to cyclic loading conditions with a load ratio of 0.1. The 
fatigue performance was evaluated to determine the number of cycles to failure as 
a function of load amplitude and load ratio. As shown in Figure 8.25, the lap shear 
fatigue test results indicate that the endurance limit of 2 × 106 cycles is obtained in 
DP590 steel at loads of 3.08 kN. Failure occurring in the lap joints under a load ratio 
of 0.2–0.6 is shown in Figure 8.26. Under cyclic loading conditions, friction stir 
spot welds fail from kinked cracks originating from the original notch tip and then 
propagating through the upper and lower sheet thickness either along the boundary 
between the TMAZ and the HAZ or along the outer fringes of the weld nugget, 
taking the shortest and weakest route. The favourable microstructure of the welds 
offers considerable resistance to the propagation of fatigue cracks, even when tested 
under high loads.
Figure 8.24 (a) Macrostructure of an interrupted shear tensile test specimen. (b) Magnified 
view of region I.
162 Welding and Joining of AHSS
The cracks subsequently open up at the surface of the sheets along the circumfer-
ence of the bond diameter. Figure 8.27 shows the cross-section of specimens failed 
under a load ratio of 0.6. Under high-cycle loading conditions, fatigue cracks I and II 
appear to emanate from the original crack tips of the weld at A and B, respectively, 
and propagate through the upper and lower sheet thicknesses, respectively. A shear 
failure, marked by F, occurs at the end of fatigue cracks I and II. These two cracks 
finally cause the failure of the specimen. During high-cycle fatigue testing under lower 
loads (load ratio ≤0.6), cracks propagate through the upper and lower sheet thickness. 
Cracks I and II both become transverse through cracks that propagate across the width 
Figure 8.25 Fatigue plots: maximum 
load vs number of cycles (a) and load 
ratio vs number of cycles (b).
Figure 8.26 Appearance of fatigue failure on the surface of a lap joint at a load ratio of 
0.2 (a), 0.4 (b) and 0.6 (c).
163Metal inert gas (MIG) brazing and friction stir spot welding of AHSS
of the specimen (Figure 8.27). These cracks finally cause the failure of the specimen. 
At higher load values (load ratio >0.6), after propagating through the upper and lower 
sheet thicknesses, fatigue cracks I and II become circumferential cracks that propagate 
along the nugget’s circumference.
8.4 Conclusions
8.4.1 MIG brazing
A galvanized DP steel sheet could be successfully joined by MIG brazing using 
copper–aluminium base (CuAl8) filler wire. Proper selection of parameters could lead 
to an efficiency of more than 90%. The dispersed iron-rich phases in the copper matrix 
enhance the strength of the weld metal and are at par with DP590 steel hardness. The 
volume fraction of dendrites containing iron increases as the MIG brazing heat input 
is increased and therefore is parameter dependent.
High shear tensile strength properties are associated with large bead sizes which 
results in failure in the HAZ. The push mode provides the most adequate dimensions 
with respect to weld height, leg length and wetting angle for superior weld perfor-
mance. Fatigue endurance limit of 2 × 106 cycles is usually attained at 10% of the 
tensile load. However, larger bead geometries give better results.
8.4.2 Friction stir spot welding
Two overlapping DP590 steel sheets, each of 1.6 mm thickness, were successfully spot 
welded with different rotational speeds and feed rates using the PCBN tool. With a rota-
tional speed of 2400 rpm, a feed rate of 228 or 300 mm/min and a dwell time of 1 s, it is 
possible to achieve suitable nugget diameters between 5 and 6 mm, comparable with spot 
weld nuggets. With these parameters completing the entire welding cycle in 4 s is possi-
ble, which is close to RSW practice. At higher parameter values the process in terms of 
z-force and torque is lower and more stable, and therefore longer tool life is expected.
Figure 8.27 (a) Failure of lap joints under a load ratio of 0.6. (b) Schematic representation of 
crack propagation.
164 Welding and Joining of AHSS
The frictional and adiabatic heat, tool pressure and stirring causes local mixing 
of discrete, plasticized entities, followed by solid-state diffusion between them for 
joining to happen. The microstructure of the SZ and TMAZ containing polygonal 
ferrite, acicular ferrite and bainite phases are refined. Dynamic recrystallization of the 
strained austenite, a low peak temperature and rapid cooling at the end of welding all 
combine to refine the microstructures in the different zones. A combination of a high 
rotational speed and a high feed rate produces the best microstructure, apart from 
meeting productivity requirements.
The tensile loads of FSSW joints are much higher than the minimum acceptable 
loads for resistance spot welds. An out-of-plane notch tip between the two sheets 
seems to restrict crack initiation and propagation during lap shear tensile testing. This 
situation is advantageous. The amenable microstructures in all the zones of the joint 
produced with higher parameters ensure better ductility. Under quasi-static loading, 
the crack originates at the notch tip and propagates along the sheet interface through 
a softer ferritic band before culminating at the central pin hole, causing shear failure. 
Under cyclic loading conditions, the fatigue crack originates at the original notch tip 
and propagates through the sheet thickness along the TMAZ–HAZ boundary or the 
outer fringes of the weld circumference. Fatigue failures occur well away from the 
central pin hole under all load ranges.
Acknowledgements
The contents of this chapter were extracted from completion reports of two Tata Steel- 
sponsored projects carried out at the Welding Technology Centre of Jadavpur University, 
Kolkata. Theauthor sincerely thanks Prof. T.K. Pal of Jadavpur University for his sup-
port during the course of these projects. Special thanks are extended to Sushovan Basak 
and Rajarshri Sarkar, Research Scholars at Jadavpur University, for generating some use-
ful information on MIG brazing and friction stir spot welding, respectively. The author is 
indebted to the management of Tata Steel India for permitting this paper to be published 
and be part of this book.
References
Aota, K., & Ikeuchi, K. (2009). Development of friction stir spot welding using rotating tool 
without probe and its application to low-carbon steel plates. Welding International, 23, 572.
Cui, L., Fujii, H., Tsuji, N., & Nogi, K. (2007). Friction stir welding of a high carbon steel. 
Scripta Materialia, 56, 637.
Davies, G. (2012). Materials for automobile bodies (2nd ed.). BH, Kidlington, Oxford. p. 248.
Feng, Z., Santella, M. L., David, S. A., Steel, R. J., Packer, S. M., Pan, T., et al. (2005). Friction 
stir spot welding of advanced high-strength steels – a feasibility study. SAE International.
Fourment, L., & Guerdoux, S. (2008). 3D numerical simulation of the three stages of friction 
stir welding based on friction parameters calibration. International Journal of Material, 
1, 1287.
165Metal inert gas (MIG) brazing and friction stir spot welding of AHSS
Gerlich, A., Su, P., & North, T. H. (2005). Tool penetration during friction stir spot welding of 
Al and Mg alloys. Journal of Materials Science, 40, 6473.
Gould, J. E., Khurana, S. P., & Li, T. (2006). Predictions of microstructures when welding auto-
motive advanced high-strength steels. Welding Journal, 85, 111s.
Guimaraes, A. S., Mendes, M. T., Costa, H. R. M., Machado, J. D. S., & Kuromoto, N. K. 
(2007). An evaluation of the behaviour of a zinc layer on a galvanised sheet joined by MIG 
brazing. Welding International, 21, 271.
Hickson, M. R., Hurley, P. J., Gibbs, R. K., Kelly, G. L., & Hodgson, P. D. (2002). The produc-
tion of ultrafine ferrite in low carbon steel by strain induced transformation. Metallurgical 
and Materials Transactions A, 33A, 1019.
Holliday, R., Parkar, J. D., & Williams, N. T. (1995). Electrode deformation when spotwelding 
coated steels. Welding World, 3, 160.
Holliday, R., Parkar, J. D., & Williams, N. T. (1996). Relative contribution of electrode tip 
growth mechanism in spot welding zinc coated steels. Welding World, 4, 186.
Hovanski, Y., Santella, M. L., & Grant, G. J. (2007). Friction stir spot welding of hot-stamped 
boron steel. Scripta Materialia, 57, 873.
Howe, P., & Kelly, S. C. (1988). A comparison of the resistance spot weldability of bare, hot-
dipped, galvannealed, and electrogalvanized DQSK sheet steels. In International Congress 
and Exposition, Detroit, Michigan (p. 325).
Khan, M. I., Kuntz, M. L., Su, P., Gerlich, A., North, T., & Zhou, Y. (2007). Resistance and fric-
tion stir spot welding of DP600: a comparative study. Science and Technology of Welding 
and Joining, 12(2), 175.
Kuziak, R., Kawalla, R., & Waengler, S. (2008). Advanced high strength steels for the automo-
tive industry. Archives of Civil and Mechanical Engineering, VIII, 103–118.
Larn, R. H., & Yang, J. R. (2000). The effect of compressive deformation of austenite on bainitic 
ferrite transformation in Fe-Mn-Si-C steels. Materials Science and Engineering, A278, 278.
Lassen, T., & Recho, N. (2006). Fatigue life analyses of welded structures. London: ISTE.
Lee, C. H., Bhadesia, H. K. D.H., & Lee, H. C. (2003). Effect of plastic deformation on the 
formation of acicular ferrite. Materials Science and Engineering: A, A360, 249.
Lepisto, J. S., & Marquis, G. B. (2004). MIG brazing as a means of fatigue life improvement. 
Welding in the World, 48, 28.
Lienert, T. J., Stellwag, W. L., Jr., Grimmett, B. B., & Warke, R. W. (2003). Friction stir welding 
studies on mild steel. Welding Journal, 1S–9S.
Parker, J. D., Williams, N. T., & Holiday, R. J. (1988). Mechanism of electrode degradation 
when spotwelding coated steels. Science and Technology of Welding and Joining, 3, 65.
Quintino, L., Pimenta, G., Iordachescu, D., Miranda, R. M., & Pépe, N. V. (2006). MIG brazing 
of galvanized thin sheet joints for automotive industry. Metal Manufacturing Processes, 
21, 63.
Reynolds, A. P., Tang, W., Posada, M., & Deloach, J. (2003). Friction stir welding of DH36 
steel. Science and Technology of Welding and Joining, 8(6), 456.
Sarkar, R., Pal, T. K., & Shome, M. (2014). Microstructures and properties of friction stir spot 
welded DP590 dual phase steel sheets. Science and Technology of Welding and Joining, 
19, 436.
Zimmer, S., Langlois, L., Laye, J., & Bigot, R. (2010). Experimental investigation of the influ-
ence of the FSW plunge processing parameters on the maximum generated force and 
torque. International Journal of Advanced Manufacturing Technology, 47, 201.
Zrnik, J., Mamuzic, I., & Dobatkin, S. V. (2006). Recent progress in high strength low carbon 
steels. Metalurgija, 45, 323.
 
This page intentionally left blank
Welding and Joining of Advanced High Strength Steels (AHSS). http://dx.doi.org/10.1016/B978-0-85709-436-0.00009-6
Copyright © 2015 Elsevier Ltd. All rights reserved.
Adhesive bonding techniques 
for advanced high-strength 
steels (AHSS)
K. Dilger, S. Kreling
TU Braunschweig, Institute of Joining and Welding, Braunschweig, Germany
9
9.1 Introduction: the exigency of adhesive bonding 
of high-strength steels
The motivations for building lighter car bodies are varied. First, the reduction of car-
bon dioxide emissions and the accordant statutory laws represent a challenge to the 
automotive industry; furthermore, the reduced weight of the car body also allows an 
improvement in driving characteristics and the integration of comfort features while 
conserving the weight of the complete automobile. To achieve lighter car bodies, 
several methods can be applied, from new structural concepts to the application of 
advanced materials and adequate joining methods. These methods can be classified 
as construction lightweight design or material lightweight design, although both are 
always interconnected because design has to be chosen according to the material and 
vice versa. Fiber-reinforced plastics, especially carbon fiber-reinforced plastics, have 
been in the public focus lately for use as lightweight materials in various applica-
tions including car body engineering. Even though these materials have excellent 
strength, stiffness-to-weight ratios, and several other advantages, their applicability— 
especially in large-volume vehicles—is limited by their high cost as well as by the 
lack of manufacturing techniques for producing high-quality parts in large batches. 
Light metals, such as aluminum or magnesium, generally have mechanical properties 
inferior to those of steel and also require complex pretreatment for aging-resistant 
adhesive bonding. Furthermore, weldbonding is not easily transferable to aluminum 
joints. For these reasons steel, especially advanced high-strength steels (AHSS), is a 
material class with a very high potential for building structural lightweight parts for 
medium- and large-volume vehicles.
The key to lighter structural parts is reducing the sheet thickness while increasing 
the yield strength and thus keeping the part strength constant. Further improvement 
can be achieved by the use of cleverly bonded joint design concepts. Compared with 
spot welds only, improvements in relative stiffness of up to 40% can be realized by 
combining classical spot welds with adhesive bonding. This allows a further reduction 
of sheet thickness and thus a reduction of weight; the strength remains unchanged.
A major challenge in the use of AHSS is to find joining methods that allow the full 
utilization of the material properties and thus the full lightweight potential. A main 
168 Welding and Joining of AHSS
problem is that welding always produces a heat-affected zone (HAZ) in the material 
andleads to a local change of microstructure, which is critical because the strength of 
AHSS is achieved by the carefully controlled microstructure. The material characteris-
tics change because of recrystallization, grain growth, or precipitations. Depending on 
the material, its thermomechanical history, and the quantity of heat introduced into the 
welding process, either local softening or hard spots can occur. A typical failure mode 
of spot-welded joints, especially under crash loads, is unbuttoning—this failure mode 
is particularly critical because very low amounts of crash energy are absorbed. An 
example of changes in the mechanical properties is shown by the hardness profile of a 
spot weld in Figure 9.1. Figure 9.2 shows a specimen with typical unbuttoning failure.
500
450
400
350
300
250
7 6 5 4 3 2 1 0
Distance from center [mm]
H
V
0.
5
(a) (b)
Figure 9.1 Vickers hardness distribution over a spot weld with a diameter of 6 mm (a). 
(b) Material of the upper adherend is boron–manganese steel (t = 1.5 mm), and that of 
the lower adherend is H300 (t = 1 mm).
Figure 9.2 Plug failure of a spot weld.
169Adhesive bonding techniques for AHSS
The impact of the warm joining techniques described above on material properties 
illustrates the interest in cold joining methods that do not lead to decreased strength or 
energy absorption. For this reason, adhesive bonding is especially interesting for joints 
of structural parts that are subjected to crash loads. The development of advanced 
toughened adhesives that achieve high strength as well as high fracture toughness and 
energy absorption has further increased the attractiveness of adhesive bonding as a 
joining method for automotive parts in recent years. Another advantage of adhesive 
bonding is its ability to deal with problems that often occur when diverging materials 
such as aluminum, magnesium, or carbon fiber-reinforced plastic are joined. These 
problems include differences in the thermal expansion coefficient or the risk of gal-
vanic corrosion. Furthermore, bonded overlap joints allow the transmission of a lam-
inar force, which leads to the reduction or elimination of local stress peaks caused by 
spot welds or flow drill screws. (Lutz & Symietz, 2009)
Nevertheless, adhesive bonding itself also provides some technological challenges, 
and there are also disadvantages regarding the achievable joint strength, the depen-
dence of adhesive properties on temperature, and the stability of the joints under aging 
conditions. Another point that is especially relevant for AHSS is the influence of sur-
face layers such as zinc coatings or cinder on the strength of the whole part. Most of 
these challenges, however, can be met by choosing the right joint geometries, adhe-
sives, and surface pretreatment methods. The following sections discuss the influences 
of joint geometry, several materials, and surface layers on the behavior of adhesive 
joints of AHSS.
9.2 Challenges in adhesive bonding of AHSS
9.2.1 Influence of joint geometry on the strength of adhesive 
bonds
Most adhesive joints in real parts are subjected to nearly pure shear loads because this 
is the preferred loading condition for these joints, which often allows the highest bond 
strength. This joint geometry is well represented by the single-lap shear specimen, as 
described in DIN EN 1465 and shown in Figure 9.3. In addition to the adhesive and 
the overlap geometry (width and length), the joint strength of this simple geometry 
also depends on the thickness and the material of the adherends (Tong & Luo, 2011). 
To take a look at these correlations, the different tensions in a single-overlap joint are 
discussed here (Habenicht, 2009; Tong & Luo, 2011).
First, shear tension occurs inside the adhesive layer; this is represented by the 
simple term τν = F/A. Second, the adherends are elastically or—with higher forces— 
plastically deformed; the parts of the adherend that are far from the end of the overlap 
carry higher loads and thus are more deformed. After the overlap ends the stress or 
the deformation is constant over the length of the joining partner. This is illustrated 
in Figure 9.3. In addition to the uniform shear tension, further tension caused by this 
deformation of the adherends is superposed and leads to stress peaks at both ends of 
the overlap. These stress peaks usually cause the first defects in the adhesive layer and 
then lead to the failure of the joint.
170 Welding and Joining of AHSS
The influence of the overlap width on the joint strength is linear because stress dis-
tribution over the width of the adherend is constant, so increasing the width is a simple 
approach for increasing the transferable load of a joint. Nevertheless, in most con-
structions the ability to realize this is limited by geometric or design reasons. Unlike 
the width, an increase in the overlap length, which is easier to realize by design, does 
not linearly increase the transferable load. This is caused by stress peaks, as explained 
above. Figure 9.4 shows the correlation between overlap length, transferable load, and 
the yield point of the adherend material.
At a small overlap (lo1) it is not the stress peaks, but the tensions caused by the 
displacement of the adherends, that are dominant. The transferable load is marginal 
τε
τν
σ
σ
Figure 9.3 Stress distribution in a single-overlap adhesive bond. 1, adherend 1; 2, adherend 
2; 3, adhesive layer; t, adherend thickness; w, overlap with adherend; lo, overlap length; σ1, 
stress distribution in adherend 1; σ2, stress distribution in adherend 2; τν, uniform shear tension 
in the adhesive layer caused by displacement; τε, shear tension in the adhesive layer caused by 
deformation of the adherends.
σ
τ
Figure 9.4 Formation of stress depending on the overlap length.
171Adhesive bonding techniques for AHSS
because of the small bonded area and is considerably lower than the strength of the 
adherends; in this case the material strength Rp0,2 cannot be used. At a medium overlap 
(lo2) the bonded area is large enough so the adherend reaches the yield point Rp0,2. For 
this overlap geometry, an optimized usage of the adherend material is achieved because 
the strength of the adhesive joint is of the same magnitude as the yield point of the 
material. Further increase of the overlap length (lo3) leads to plastic deformation of the 
adherend, which cannot be sustained by the adhesive layer and therefore causes failure 
of the joint. This shows that increasing the overlap length does not further increase 
the joint strength if tensions in the magnitude of the material yield point are obtained. 
Because of the formation of stress peaks the central area of the joint bears only a small 
fraction of the load. Of course, in this context the transferred load has to be aligned 
to the thickness of the adherend. This illustrates the lightweight potential of adhesive 
bonds of AHSS. Because of the high yield point of these materials, either a decrease in 
the sheet gauge or an effective increase in the overlap, and hence of the transferable load 
at a constant sheet gauge, is possible. To use the maximum lightweight potential of the 
materials for each joint, the optimum sheet gauge and overlap according to the occur-
ring loads have to be found, given that reducing the thickness or overlap also allows 
weight to be reduced. As also described by Adonvi (2005), however, the strength of 
adhesively bonded joints between AHSS parts is always limited to the cohesive strength 
of the adhesive itself, thus raising the question of whether the strength of commercially 
available structural adhesives is sufficient to use the full potential of advanced steel.
Decreasing the sheet thickness to achieve a higher lightweight potential nonetheless 
also reduces the bending stiffness of the part, which again, depending on part design 
and load, results in the threat of peel loads occurring in the adhesive layer, significantlylowering the sustainable maximum strength. Considering AHSS with a very high yield 
point, this is especially critical because the strength is proportional to the thickness, 
while the bending stiffness decreases in relation to the thickness cubed, which means 
that the use of low gauges contains the threat of peel loads occurring.
Considerations of the geometric aspects of the bonding area and adherends show 
that, because of their high yield strength, AHSS further enhances the strength of adhe-
sively bonded joints. Nevertheless, the joint design always has to be taken into account 
to prevent peel loads and enable sufficient bond areas.
9.2.2 Crash behavior of adhesively bonded AHSS
In addition to the static maximum and cyclic loads that occur during a vehicle’s life-
time, the adhesively bonded joints also need to perform well during a crash. For con-
ventional steel parts this means that the adhesive has to bear the highest possible loads 
during the crash so the metal parts can deform plastically and thus absorb most of 
the energy. If parts manufactured from AHSS are used the difference is that, because 
of the high yield point and low possibility of plastic deformations, far less energy is 
absorbed by the part itself. In this case it is even more important that the adhesive layer 
does not fail because of brittleness but absorbs as much energy as possible. Modern 
structural adhesive bonds, however, often show layer thicknesses of the magnitude far 
less than 1 mm; such thin layers are not able to absorb large amounts of crash energy. 
172 Welding and Joining of AHSS
Hence considering the function of the part, or at the part itself, is important. The 
B-pillar is, for example, a typical automotive structural part that is manufactured from 
AHSS. In a side crash it is important that the pillar does not deform massively; this 
would cause severe injuries to the car’s occupants. Thus the function of the adhesive 
layer bonding the inner and outer sheets of the B-pillar is not to absorb large amounts 
of energy, for example, by plastic deformations, but to ensure that the parts are kept 
together. Because of the low plastic deformation resulting from their high yield points, 
the function of most parts manufactured from AHSS is not to absorb energy by plastic 
deformation but to maintain the structure of the car and save the occupants. Hence the 
main function of the adhesive is to keep the joined parts together, not to fail because 
of brittleness, and to show tough behavior, even at low temperatures.
9.2.3 Bondability of different kinds of AHSS
The adhesive bondability of high-strength steel strongly depends on the alloying ele-
ments in the steel. The characteristics of the joint under mechanical loads and aging 
conditions also are influenced by surface layers or coatings that are applied during the 
manufacturing process for several reasons. In general, AHSS for automotive applica-
tions can be categorized as coated and uncoated materials; among the coated materi-
als, zinc coatings or coatings to prevent cindering during heat treatments are widely 
used. For this reason first uncoated and then zinc-coated and press-hardened steel with 
coatings to prevent cindering are described in the following sections.
9.2.4 Uncoated AHSS
The application of uncoated AHSS is, as for other kinds of steel, quite limited because of 
their poor resistance against corrosion. These materials are commonly coated with oil to 
prevent corrosion during manufacturing processes or storage. Most structural hot-curing 
adhesives that are used in the automotive industry show a good tolerance toward oil 
contamination and are able to absorb the oil from the surface and build a durable bond. 
A typical threshold value for the amount of oil that can be absorbed by an adhesive is 
about 3 g/m². Thus when uncoated materials that have a high degree of oil contamina-
tion are adhesively bonded, a prior cleaning step should always be considered.
Another aspect to consider when adhesively bonding uncoated AHSS is the forma-
tion of different oxides of the alloying elements on the bonding surface. This is espe-
cially meaningful for AHSS because more alloying elements with a higher tendency 
to build oxides are used. Depending on the oxides that are built, local areas with either 
poor resistance against corrosion or poor adhesion to the adhesive can be formed. 
Hence when bonding uncoated AHSS taking a close look at the alloying elements and 
considering the resistance to corrosion of the oxides that can be built are essential.
9.2.5 Zinc coatings
Conventional high-strength steels as well as AHSSs are often coated with zinc to 
inhibit corrosion. For this reason, special adhesives that show very good adhesion on 
173Adhesive bonding techniques for AHSS
the zinc layer as well as on certain amounts (up to several grams per square meter) 
of oil residues that can occur on the surfaces, for example, after the deep drawing 
process, have been developed. These adhesives are well established and state of the 
art for structural adhesive bonds; hence adhesion between the adhesive and zinc layer 
occurs in most cases.
A common process for coating high-strength steel with zinc is galvannealing, 
which maintains the parts at elevated temperatures after the hot-dip coating process. 
In detail, first, the uncoated material is coated in a bath of liquid zinc and then heated 
at temperatures around 550°C. During the time spent at the elevated temperature, the 
zinc coating alloys with the iron by diffusing between the molten zinc and iron of 
the base material. As a consequence, the final coating contains about 90% zinc and 
10% iron, which strongly depends on the diffusion temperature and time. Because 
of the diffusion process, this coating has a very strong bond to the base material. 
Another advantage of the galvannealing process is that the coating does not contain 
aluminum, as galvanized coatings do. The aluminum is added in the galvanizing pro-
cess to improve adhesion between the coating and the base material. When adhesively 
bonding to the surface, there are also areas containing aluminum oxide, which usually 
have poor properties of creep corrosion and long-term stability. Thus, when adhesively 
bonding zinc-coated steel, it is always important to mind the type of the coating and 
possible influences, especially on the long-term durability of the joints.
Compared with high-strength steel, AHSS grades use higher amounts of alloys 
such as manganese, silicium, molybdenum, or carbon, which have a higher affinity to 
oxygen than iron itself. This can lead to minor adhesion or defects inside the zinc layer 
(Li, 2011) because of the difficulties in reducing their more stable oxides. However, 
this problem is in the focus of steel manufacturers, and coatings with good adhesion 
to AHSS are available on the market. Good adhesion between the base material the 
and coating is especially important because, as discussed above, the transferable loads 
of adhesively bonded AHSS are usually significantly higher than those of conven-
tional high-strength steel, and compared with mechanical joints or welds the load is 
transferred entirely through the interface between the steel and the coating. Bandekar 
(2009) reported that when coating delamination occurs, joint strength and impact load 
are decreased, and X-ray photoelectron spectroscopic analysis showed that the coating 
was removed mostly in extra deep drawing, interstitial-free steel samples, and it was 
at the gamma phase of the galvannealed layer. Furthermore, this reference indicates 
that if the joints failed cohesively, the joint strength was not sensitive to steel grades, 
which is plausibly explained above.
9.2.6 Press-hardened steels
In the press-hardening process boron–manganese steel is heated to about 800°C 
and then plastically deformed. Heating often occurs inside an oven in an inert gas 
atmosphere; afterward the parts are transferred into