1

HAZ Properties in High Performance Steel

Solid-State Welds

Adonyi, Y. PhD, PE
Omer Blodgett Chair of Welding and Materials Joining Engineering
LeTourneau University, Longview, TX
yoniadonyi@letu.edu

Abstract

The AASHTO/AWS A 709 Quenched and Tempered High Performance Steels (HPS) with up to
100 ksi yield strength, good weldability and weathering characteristics, as well as the A1010
ferritic stainless steels have already been used in hundreds of bridges across North America.
They could be used in arctic environments as well because of their excellent toughness,
although they all require heat input limitations when welded using traditional arc welding.

This study was intended for creating a solid-state welding process database and in hope of
improving low temperature HAZ toughness, when compared with fusion welds. Actual- and
simulated High Frequency- (HF), Friction Stir- (FSW) and HF/FSW Hybrid welding processes
were compared with similar heat input SMAW, SAW and GMAW/FCAW fusion welds.

It was found that all solid-state welds had superior HAZ toughness when compared to fusion
weld HAZs. However, each solid-state welding process required Post Weld Heat Treatment and
presented low productivity and high capital cost issues. Nevertheless, it was concluded that
solid-state welding remains a strong alternative when joining dissimilar combinations required
for prefabricated transition pieces.

1. Introduction.

This section includes an overview of the two types of structural steels involved in the study,
together with a brief description of the solid-state welding processes used.
2

1.1 High Performance Steel (HPS) Background.

The AASHTO/AWS A709 type grade 70 or HPS 70W steel, which has been used in the
construction of over 400 new bridges across the United States, is a 70 ksi minimum yield
strength weathering steel typically welded using fusion welding processes. These processes
include Shielded Metal Arc Welding (SMAW), Gas Metal Arc Welding (GMAW), Flux Cored Arc
Welding (FCAW) and Submerged Arc Welding (SAW). The high heat inputs of these processes
cause grain growth in the heat affected zone (HAZ), reducing the strength and toughness. The
fusion zone experiences a reduction in toughness, but the strength is regained through alloying.
Typical defects within the fusion zone can include: hydrogen induced cracking, porosity, andr
slag inclusions. Micro alloying and inclusion control, such as low sulfur levels were the keys to
improved toughness and improved weldability (no preheat for up to 2.0 inch thick plates). A
typical chemical composition and mechanical properties are shown in Tables I and II in
Appendix I

1.2 Ferritic/Martensitic Structural Stainless Steel Background.

The ASTM A1010 Grade 50 duplex ferrite/martensite stainless steel is a newer material used for
bridge fabrication. The use of fusion welding processes causes significant grain growth in the
HAZ and significantly increases the hardness, which reduces the HAZ toughness. Defects in the
fusion zone can include: solidification cracking, porosity and slag inclusions. Made under the
Duracorr@ trade name, it is a dual-phase stainless steel product with a 12% chromium content
produced by Arcelor Mittal USA [7]. This structural steel has a wide variety of applications that
require good corrosion resistance, good edge retention, high strength, high hardness and wear
resistance [5]. The mechanical properties and chemical composition of ASTM A1010 can be
seen in Tables III and IV in Appendix 1. The specified mechanical properties of Duracorr are to
meet the standard of ASTM A1010. Various testing methods used to measure the mechanical
and corrosion properties of Duracorr were carried out [8].

1.3 Solid State Welding Process Overview.

Three different solid state welding processes will be discussed in this section: 1) High
Frequency Welding (HFW), 2) Friction Stir Welding (FSW), and a 3) HF-enhanced or hybrid
HFI/FSW. In all solid-state welding process, the base metal is not melted, producing good
toughness in the weld. Peak temperatures are much lower compared to those of fusion welding
and, as a result, there are finer grain sizes in the HAZ and there is no chance of hot cracking.
3

1.3.1. High Frequency Welding.

This technology is about six decades old and a well-known solid-state welding process that
utilizes Electric Resistance (Joule) heating of the adjoining surfaces typically using a high-
frequency alternating current at 100-300 kHz for welding of steels. Depending on the electrical
resistivity of the metals to be welded, such as with aluminum or copper, frequencies of up to 1
MHz must be used. The eddy currents responsible for heating can be induced using a coil (no
contact) or by using sliding contacts adjacent to the weld, Figure 1.

Figure 1. Schematic illustration of contact high frequency welding, Ref. 3.

Advantages of HFW include: excellent reproducibility, high productivity, relative simplicity and
low heat input. Disadvantages of HFW relative to fusion welding are: difficult to apply to low
volume production, less flexibility in changing over one product to another, very high welding
speeds that requires precise preparation and base metal feeding and cutting devices. High
Frequency Induction Welding is widely used to make pipe from steel and other materials.

1.3.2. Friction Stir Welding.

This is a relatively new solid state welding process that revolutionized the welding industry since
its introduction in 1991 to join previously unweldable aluminum alloys such as the Al-Cu and Al-
Li systems (Ref. 4). The process builds on the positive experience gained in many decades of
Friction Welding of small parts that had radial symmetry which could be spun at high speeds
such as axles, valve stems, etc. The novelty of using a spinning tool to generate frictional heat
made possible the joining of abutted edges, see Figure 2.
4

Figure 2. Schematic representation of the Friction Stir Welding Process (Ref. 3)

Advantages of FSW include welding in solid state of materials susceptible to solidification

cracking and porosity formation, joining of dissimilar materials, less mismatch between the weld
and HAZ properties, low distortion. Disadvantages have mostly been related to tool wear, low
welding speeds, the need for strong backing and extensive fixturing, and difficulties with
nondestructive evaluation. Yet, joint strength has been shown to increase by as much as 30
percent over fusion welding and the joints also have much higher elongation and decreased
shrinkage. One of the biggest advantages of FSW is the ability to join materials that were
‘unweldable’ by fusion processes such as the 7xxx and 2xxx series aluminum alloys as well as
dissimilar materials.

1.3.3. High Frequency/Friction Stir Welding Combination.

The Friction Stir/High Frequency Hybrid process is expected to combine the benefits of both
high frequency induction welding and friction stir welding into one process. The set-up is shown
in Figure 3.
5

Figure 3. Friction Stir/High Frequency hybrid weld setup at LeTourneau University. Welding
direction is from left to right, with the coil (arrow) leading.

2. Objectives

The primary objective of this study was to use new solid-state welding processes to weld and
study HAZ phenomena in HPS 70W and A1010 steels, using High Frequency Welding , Friction
Stir Welding and an HF-enhanced FSW. Baseline fusion welds and Gleeble simulations were
to be used for comparison, all focused on HAZ properties.

3. Methodology

Overall, the problem of defining equivalent HAZ thermal cycles and heat inputs between the
solid-state and fusion welds was to be solved first for similar and dissimilar combinations for
0.25 inch thick plate. Subsequently, Gleeble HAZ simulations were to be used for a one-on-one
comparison for HAZ toughness and actual weld HAZ property predictions were to be made.

3.1 Baseline Fusion Welds.

In this study, only HPS 70W and A1010 plates of 0.25” thickness were used because it could be
welded in full-penetration mode using all processes involved. Each plate was beveled at a
30°angle in order to make a single pass weld. Four processes were used: the Shielded Metal
Arc Welding , Gas Metal Arc Welding , Flux Cored Arc Welding and Submerged Arc Welding
(SAW). The actual welding parameters and heat inputs can be seen in Tables V through VIII in
Appendix II.
3.2 Friction Stir Welds.

Thermocouples were used to take temperature measurements in the heat affected zone of all
welds for later use in the Gleeble simulations, while COMSOL numerical simulations and
6

infrared camera recordings of the process were made. Figure 4 shows the comparison between
the measured temperature profile and the predicted profile.

Figure 4. Predicted vs actual temperature profiles.

The IR thermal camera recording of the FSW temperatures matched the thermal couple
readings therefore validating our thermal cycle. Downforce measurements were performed via
two load cells placed underneath the table. Position sensors allowed the monitoring of the
horizontal and vertical displacement via a PLC controller but the ‘constant force’ control mode
was used for all experiments.

Metallography was performed on cross-sections from the welds. Hardness measurements were
made using a Vickers Micro Hardness testing machine. Visual inspection under a microscope
and Eddy Current testing were also performed on the welds. A 10% oxalic acid and Kallings
No.2 etchants were used to reveal the microstructures.

Weld Tempering: The PWHT simulation was performed in a furnace at 1200°F for 10 minutes
followed by air cooling to room temperature. Similarly, for the solid-state welding part,
completed welds were tempered in a furnace at 1200oF for 10 minutes to validate the need for
the trailing coil concept.

Gleeble Simulations: Oversized Charpy-type specimens (11 mm x 11 mm cross section) were

used to simulate HAZ thermal cycles using the single-pass peak temperatures and cooling rates
measured using the IR camera and thermocouples in certain locations in the HAZ.
7

High Frequency Welding Experiments: As the brief overview on the topic showed, the standard
configuration of HF welding consists of a stationary coil and a moving part. Making special
individual “batch” HF welding equipment with the inductor coil on the bottom and impeder on the
top of the abutting edges would be a major change to the process. Thermatool Inc already has
such a small scale device in their labs that works on 10” long specimens, Figure 5.

Figure 5. HFW welding fixture used at Thermatool, with the HF coil removed from the top of the
two samples pressed against one another, arrow (zero gap) during welding.
Parameters for the 0.25 in. thick samples were: 70% power, 10 kHz, 3-3.5 sec weld
time, 10 ksi forging pressure.

4. Results and Discussion

4.1 HPS-70W Fusion Weld Baseline

The welding parameters used and mechanical properties of the fusion welds are listed in
Appendix II. One SAW coarse grained HAZ or CGHAZ was picked to illustrate the need to limit
heat input (and grain coarsening) in the HAZ, Figure 6.
8

Figure 6. Optical HAZ of the SAW weld made at 30 kJ/in. heat input, HPS side, 100X. Notice
the hard and brittle CGHAZ (arrow), that measured over 357 HV in average hardness

4.2 A1010 (Duracorr) Similar and HPS 70W+ A 1010 Dissimilar Fusion Baseline Welds.

The similar A 1010 welds exhibited predictable properties based on welding recommendation
from the manufacturer. As no matching strength filler material was used, all welds broke in the
weaker but ductile austenitic weld metal, as expected. The additional color mismatch between
the black A 1010 ferritic/martensitic base metal and the shiny E 304 Cr/Ni composition filler
metal was pointed out to be ‘bothersome’ by some as bridge manufacturers. In other words, the
HAZ toughness loss issue was not considered, because the austenitic weld metal would likely
absorb impact energy, but fail in a plastic collapse mode.

The dissimilar welds had HAZ properties identical to those in the similar A1010 and HPS-70W
welds discussed above for each material. THE A1010 RESULTS HAVE NOT BEEN
DISCUSSEDThe area with the lowest toughness in these welds will probably be the GCHAZ on
the A1010 side of the welds. The same parameters and austenitic filler metal were used for
these welds The fusion zone in each weld tended to have similar properties to each of the
A1010 welds for the same process, but the CGHAZ in the A 1010 side had the highest hardness
and presumably the lowest toughness, Figure 7.
9

400
SMAW - Hardness, A1010+HPS70
350
300
250
200
150
100
50
0
-16 -14 -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 14 16
Distance from Center (mm)

Figure 7. Typical hardness distribution across a dissimilar weld, Ref. 1. Note the difference in
HAZ hardenability between the two steels, with the A1010 HAZ having almost twice
the hardness of the base metal and weld.

These welds showed the same heat input sensitivity as the A1010 welds, and at higher heat
inputs, such as in the SAW weld, the weld metal was able to transform to martensite, giving the
fusion zone a similar hardness to the GCHAZ on the A1010 welds. These welds allow the HPS-
70W to showcase its excellent weldability, as the hardness is much lower in the GCHAZ on the
HPS-70W side. The typical concern with meeting a minimum heat input, as with the HPS-70W
welds, is waived because the chromium in the A1010 increases the solubility of hydrogen and
eliminates the possibility for hydrogen induced cracking. The area with the lowest toughness in
these welds will probably be the GCHAZ on the A1010 side of the welds and they will likely fail
in the austenite/ferrite mix fusion zone or the GCHAZ on the A1010 side of the weld.

On the high heat input end when multi-wire SAW processes are used, the as-solidified cast weld
metal properties typically undermatch those of the base metal (Ref. 2). Additionally, all these
welds can be potentially susceptible to solidification cracking, see Figure 8. If the weld does not
crack during solidification, the resultant ferrite/martensite mixture will remain susceptible to
hydrogen-induced cracking(in case not an austenitic filler metal is used like in the recommended
practices).
10

Figure 8. Optical micrograph, dissimilar fusion weld. Notice the high delta ferrite content,
possible hot cracking and cold cracking tendencies.

4.3 Similar and Dissimilar Solid-State Welds (HF, HF/FSW and FSW)

The HF welds were made at Thermatool Inc. All similar combinations were of excellent quality,
while the most promising proved to be the dissimilar combination, Figure 9.

Figure 9. Dissimilar HF weld between HPS and A 1010 (right) at low and high magnifications.

The peak hardness in these welds, 329 HV, was located at the bond line. Also The GCHAZ
contained more martensite, and had a higher hardness, 299 HV, in the GCHAZ than any of the
fusion welds had in their fusion zones. The HAZ measured to be 4.45 mm from the bond line to
the outer edge of the GGHAZ.
11

While the FSW had a high heat input, it experienced the lowest HAZ peak temperatures of all of
the welds. The result was the widest HAZ measuring 7.74 mm from the edge of the stir zone to
the outer edge of the CGHAZ. The lower peak temperature practically eliminated the GCHAZ
noted in the fusion welds.

Figure 10. Thermal Mechanically Affected Zone (TMAZ) in a Friction Stir weld on the A 1010
side, 100 X magnification

Average
Grain Size:

Hybrid A1010 Stir

Average
Hybrid A1010 Base Grain Size:
Metal, HAZ, Stir Zone

Hybrid A1010

Figure 11. Details of the FSW microstructure, A 1010.

The role of HF preheating (HF/FSW combo) was significant, making welding easier by reducing
the steel flow stress and improving tool wear. There was less buildup of the W-Re tool and the
FSW welding speed was increased by 62% as a result of using the HF preheat, Figure 12.
12

Figure 12. Top view of the A 1010 FSW showing the effect of the HF preheat on the weld
quality (right of the separating line, arrow).

The hardness profiles of a FSW and FSW/HF Hybrid weld can be seen in Figure 13. The graph
shows a decrease in hardness across the weld was achieved in addition to the increase in
welding speed.

A1010 - A1010 Hardness Profile

400
350
Vickers Hardness

300
250
200
FSW - A1010
150
100 FSW/HFI A1010
Weld
50 centerline
0
0 5 10 15 20 25 30
Distance Across the Weld [mm]

Figure 13. Hardness profiles comparing use of HF preheating in FSW of A1010.

When welding the dissimilar HPS 70W + A1010 combination, the HPS 70W was placed on the
Advancing side of the tool, while the A1010 with a wormhole seen in Figure 14, was placed on
the Retreating side of the tool.
13

Figure 14. Dissimilar A1010 (left) + HPS 70W (right) macro section, assembled using electronic
montage, as the two different etchants had to be used to reveal the microstructures.

Note the wormhole on the root side of the weld, indicating that more parameter optimization is
required to complete better quality dissimilar welds. The microstructure images, Figure 15, show
these areas. Visual Inspection revealed kissing bonds and tungsten inclusions. Tungsten
inclusions were verified using Electron Dispersive Spectroscopy (EDS) on a SEM. Images of
these defects can be found below as well.

Kissing Bond Tungsten Inclusion in Stir

Zone of A1010 weld

Figure 15. Optic and electron micrographs in different areas of the FSW weld. Note the W
inclusion left from the W-Re tool .

Eddy Current Testing was then performed to detect any other flaws and to analyze the change
in impedance across the face of the polished weld sample. While wormholes were found near
14

the root of the weld, the impedance measured was quite different for the fusion- vs solid-state
welds, as shown in Figure 16.

Conductivity in A1010, SMAW vs. FSW

16
14
Impedance [Ohms]

12
10
8
FSW
6
4 SMAW
2
0
0 5 10 15 20 25 30 35
Distance [mm]

Figure 16. The more even electricalonductivity measurements across the FSW vs. when an
austenitic filler metal is used in SMAW of A 1010 (Eddy Current testing) shows the
superiority of the solid state welding process.

Finally, Gleeble HAZ simulations were performed on A1010 samples for SMAW, SAW, FSW,
and FSW/HF thermal cycles, because that CGHAZ was deemed to be the most susceptible to
failure in welded structures. All results were compared by performing Charpy V-Notch testing
and then analyzing the specimens by executing fractography, hardness testing, and
metallography. Charpy V-Notch testing was performed at 0 °F, -20 °F, and -30 °F.
Figure 17 shows the as-welded simulated CGHAZ toughness tested at 0 °F. The base metal
toughness value used was the nominal value published by the manufacturer.
15

Toughness vs. Simulated Welding Process

200
Impact Toughness [ft-lbs.]

150

100

50

0
Base Metal HF+FS FSW SAW SMAW
W
Figure 17. Relative simulated CGHAZ toughness, 0°F for the A 1010 base metal only.

These results showed a significant improvement in HAZ toughness from fusion welding with
FSW alone. Implementation of the Hybrid System with a trailing HFI Coil further improved the
HAZ toughness from FSW and increased the average by 34%.

5. Conclusions

While it is difficult to outperform traditional fusion welding processes on HPS-70W and A1010
steels in terms of cost, flexibility and productivity, there is much promise for welding them using
solid state processes. High Frequency and HF-enhanced FSW weld performed well, but both
type of welds needed post weld heat treatment. Especially promising were the HAZ toughness
results in solid-state weld Gleeble simulations, as well as dissimilar joining between the two
types of steels examined in this study.

6. References

[1] Guide Specification for Highway Bridge Fabrication with HPS 70W (HPS 485W) Steel, 2nd
ed., June 2003
[2] Wilson, Alexander D. “Production of High Performance Steels for U.S.A. Bridges.”
[3] Wilson, Alexander D. “Production Experience with new Heavy Plate Grades for Bridges and
Shipbuilding Using Microalloying.”
[4] Wilson, Alexander D. “Clean Steel Technology – Fundamental to the Development of High
Performance Steels.”
16

[5] Krauss, George. Steels: Heat Treatment and Processing Principles. Materials Park, OH:
ASM International, 1989. Print.
[6] ASTM Standard A 709/A 709M – 01a, “Standard Specification for Carbon and High-Strength
Low-Alloy Structural Steel Shapes, Plates, and Bars and Quenched-and-Tempered Alloy
Structural Steel Plates for Bridges,” ASTM International, West Conshohocken, PA,
www.astm.org.
[7] ArcelorMittal USA, Duracorr: Life-Cycle Cost-Effective 12% Chromium Stainless Steel.
November 2008, www.arceelormittal.com.
[8] F.B. Fletcher, A.D. Wilson, “Stainless Steel for Accelerated Bridge Construction.”
[9] R.W.K. Honeycombe and H.K.D.H. Bhadeshia. Steels microstuctures and properties.
Edward Arnold, London, 1995.
[10] ASTM Standard A 1010/A 1010M – (01) 2011, “Standard Specification for Higher Strength
Martensitic Stainless Steel Plate, Sheet, and Strip,” ASTM Internation, West Conshohocken,
PA, www.astm.org.
[11] Carrouge, D., Woollin P, October 2002, Microstructural change in high temperature heat-
affected zone of low carbon weldable 13 %Cr martensitic stainless steel. Stainless Steel World:
16 – 22.
[12] AWS Welding Handbook: Welding Processes, Vol. 2, 8th ed. American Welding Society.
[13] Subarc Welding: Advantages of Submerged Arc Welding. ESAB Welding & Cutting USA,
www.esabna.com.
[14] Wei, Lingyun and Nelson, Tracy W. “Characterization of Microstructures and
Transformation Behavior in Friction Stir Welded HSLA-65.” Trends in Welding Research,
Proceedings of the 8th International Conference. ASM International, 2009.
[15] Konkol, P.J. and Mruczek, M.F. July 2007. “Comparison of Friction Stir Weldments and
Submerged Arc Weldments in HSLA-65 Steel.” Welding Journal. American Welding Society and
the Welding Research Council.
[16] Ding, Jeffrey R., and Schneider Judy. “Advances in Solid State Joining of Haynes 230 High
Temperature Alloy.”
[17] Messler, Robert W. “Principles of Welding: Processes, Physics, Chemistry, and Metallurgy”.
1999, John Wiley & Sons, Inc.
[18] Hamby, Gary. “High Performance Steel Designers’ Guide”. 2nd ed, April 2002. U.S.
Department of Transportation, Federal Highway Administration.
[19] AWS Welding Handbook: Materials and Applications – Part 2, Vol. 4, 8th ed. American
Welding Society.
17

[20] Adonyi, Y. and Swenson, J. “Deformation Effects during Variable High Frequency Welding
of Modern Steels”. Trends in Welding Research, Proceedings of the 8th International
Conference. 2009,

Acknowledgments: This work was sponsored by the Federal Highway Administration R&D
Labs, McLean, VA. PROJECT NO. DTFH61-10-D-00017.

The author would like to thank the many LeTourneau University students who worked on
different stages of this project: Hayden Adams, Nathan Dix, Devin Peluso, Puhao Wang, and
others.
18

Appendix I

Table I. HPS 70W Chemical Requirements, [6]

Element Composition
(%)
Carbon 0.11 max
Manganese 1.10-1.35
Phosphorus 0.020 max
Sulfur* 0.006 max
Silicon 0.3-0.5
Copper 0.25-0.40
Nickel 0.25-0.40
Chromium 0.45-0.70
Molybdenum 0.02-0.08
Vanadium 0.04-0.08
Aluminum 0.010-0.040
Nitrogen 0.015 max
* The steel shall be calcium treated for sulfide shape control.

Table II,- Mechanical Properties, [6]

Plate Thickness Yield Strength Tensile Minimum
(inches [mm]) (ksi [MPa]) Strength Elongation
ksi [MPa] (% in 2 in.)
HPS 70W To 4 [100] 70 [485] 85-110 [585- 19
760]

Table III. Typical ASTM A1010 Chemistry Requirements, [10]

Element Composition
(%)
Carbon 0.03
Manganese 1.50
19

Phosphorus 0.040
Sulfur 0.030
Silicon 1.00
Chromium 10.5-12.5
Nickel 1.50
Nitrogen 0.030
Molybdenum -

Table IV. Typical Mechanical Properties, ASTM A1010 [10]

Ultimate Yield Strength Minimum Brinell
Tensile (ksi [MPa]) Elongation Hardness
Strength (%) (maximum)
(ksi [MPa])
ASTM A1010 70 [485] 50 [345] 18 223*
Grade 50
*Converts to 20 HRC

Appendix II - Fusion Welding - References

Table V. Consumables Used, Lincoln Electric
HPS 70W – HPS 70W A1010 – A1010 HPS 70W – A1010
SMAW Excalibur 9018M MR Excalibur 309/309L-16 Excalibur 309/309L-16
(5/32”) (5/32”) (5/32”)
GMAW Super Arc LA 100 (0.045”) Blue Max MIG 309LSi Blue Max MIG 309LSi
(0.045”) (0.045”)
FCAW Ultra Core 81Ni1C-H N/A N/A
(1/16”)
SAW Lincoln Weld LA-85 Wire Lincoln Weld 309L Wire Lincoln Weld 309L Wire
(1/8”) (3/32”) (3/32”)
Lincoln Weld MIL800-HPNi Lincoln Weld P2007 Flux Lincoln Weld P2007 Flux
Flux

Table VI. HPS 70W + HPS 70W Welding Data.

20

SMAW GMAW FCAW SAW

Travel Speed (in./min) 5.5 13.25 16.1 17.25
Voltage (Volts) 24 27 27 30.2
Current (Amps) 160 275 365 400
Transfer Efficiency 0.75 0.86 0.86 1
Heat Input (kJ/in.) 31.4 28.9 31.5 42.0

Table VII. A1010 + A1010 Welding Data.

SMAW GMAW SAW
Travel Speed (in./min) 5.5 12.25 15
Voltage (Volts) 22.7 25.4 32.5
Current (Amps) 135 215 290
Transfer Efficiency 0.75 0.86 1
Heat Input (kJ/in.) 25.0 23.0 37.7

Table VIII. HPS 70W + A1010 Welding Data.

SMAW GMAW SAW
Travel Speed (in./min) 5.5 12.25 15
Voltage (Volts) 22.5 25.4 32.5
Current (Amps) 145 225 260
Transfer Efficiency 0.75 0.86 1
Heat Input (kJ/in.) 26.6 24.0 33.8