H. A. H.

SAADAWI

A STUDY ON THE WIRE ARC ADDITIVE MANUFACTURING OF DUPLEX

STAINLESS STEEL PLATES: OPTIMIZATION OF PROCESS PARAMETERS
AND INVESTIGATION OF MECHANICAL PROPERTIES

THE GRADUATE SCHOOL OF NATURAL AND APPLIED SCIENCES

OF
ATILIM UNIVERSITY

HASSAN A. H. SAADAWI
ATILIM UNIVERSITY

A DOCTOR OF PHILOSOPHY THESIS

IN
MODELING AND DESIGN OF ENGINEERING SYSTEMS
(MAIN FIELD OF STUDY: METALLURGICAL AND MATERIALS
ENGINEERING)
2022

JANUARY 2022
A STUDY ON THE WIRE ARC ADDITIVE MANUFACTURING OF DUPLEX
STAINLESS STEEL PLATES: OPTIMIZATION OF PROCESS PARAMETERS
AND INVESTIGATION OF MECHANICAL PROPERTIES

A THESIS SUBMITTED TO
THE GRADUATE SCHOOL OF NATURAL AND APPLIED SCIENCES
OF
ATILIM UNIVERSITY

BY

HASSAN A. H. SAADAWI

IN PARTIAL FULFILLMENT OF THE REQUIREMENTS

FOR
THE DEGREE OF DOCTOR OF PHILOSOPHY
IN
MODELING AND DESIGN OF ENGINEERING SYSTEMS

(MAIN FIELD OF STUDY: METALLURGICAL AND MATERIALS

ENGINEERING)

JANUARY 2022
Approval of the Graduate School of Natural and Applied Sciences, Atılım University.

Prof. Dr. Ender Keskinkılıç

Director

I certify that this thesis satisfies all the requirements as a thesis for the degree of Doctor
of Philosophy in Modeling and Design of Engineering Systems (Main Field of
Study: Metallurgical and Materials Engineering), Atılım University.

Prof. Dr. Ender Keskinkılıç

Program Chair

This is to certify that we have read the thesis A STUDY ON THE WIRE ARC
ADDITIVE MANUFACTURING OF DUPLEX STAINLESS STEEL PLATES:
OPTIMIZATION OF PROCESS PARAMETERS AND INVESTIGATION OF
MECHANICAL PROPERTIES submitted by HASSAN A. H. SAADAWI and that in
our opinion it is fully adequate, in scope and quality, as a thesis for the degree of Doctor
of Philosophy.

Asst. Prof. Dr. Kâzım Tur Assoc. Prof. Dr. Erkan Konca
Co-Supervisor Supervisor

Examining Committee Members:

Assoc. Prof. Dr. Erkan Konca

Met. & Materials Eng. Dept., Atılım University

Prof. Dr. Bilgehan Ögel

Met. & Materials Eng. Dept., METU

Prof. Dr. C.Hakan Gür

Met. & Materials Eng. Dept., METU

Asst.Prof. Dr. Hakan Kalkan

Manufacturing Eng. Dept., Atılım University

Asst.Prof. Dr. Ozan Özkan

Met. & Materials Eng. Dept., Atılım University

Date: January 20, 2022

I declare and guarantee that all data, knowledge and information in this document
has been obtained, processed and presented in accordance with academic rules and
ethical conduct. Based on these rules and conduct, I have fully cited and referenced
all material and results that are not original to this work.

Name, Last Name : HASSAN SAADAWI

Signature :
 a ABSTRACT

A STUDY ON THE WIRE ARC ADDITIVE MANUFACTURING OF

DUPLEX STAINLESS STEEL PLATES: OPTIMIZATION OF PROCESS
PARAMETERS AND INVESTIGATION OF MECHANICAL PROPERTIES

SAADAWI, HASSAN
PhD in Modeling and Design of Engineering Systems (MODES)
Supervisor: Assoc. Prof. Dr. Erkan Konca
Co-Supervisor: Asst.Prof.Dr. Kâzım Tur
January 2022, 140 pages

In recent years, additive manufacturing (AM) for metals, which is used to create
products using layer-by-layer deposition of materials, has been a subject of interest as
a means of enhancing the efficiency of manufacturing processes in various industries.
In this thesis, super duplex stainless steel parts have been wire arc additively
manufactured (WAAM) using the cold metal transfer welding technique. To obtain the
optimum combination of additive manufacturing parameters (voltage, current and
speed) that yields the desired bead quality, response surface method has been used to
build statistical models based on numerical simulation of additive welding process.
COMSOL Multiphysics 5.5 software has been used to perform the experimental
instead of real process and Expert-Design software was used to obtain the statistical
models. The optimal values of the controlled variables were found as 200 amperes and
15 volts for current and voltage, and 10 mm/s for welding speed. Then, the fracture
toughness tests using the single edge notched tension (SENT) specimens were
implemented at a sub-zero temperature. Fracture toughness data has been used to
establish crack-growth resistance curves for both, as-received rolled plate and
additively manufactured parts and compare the results based on J-integral (energy
release rate). The J-integral value at the initiation of stable crack growth (Ji) of SDSS
Grade 2507 as-received plate specimens is approximately 17% higher than the (Ji)
value of WAAM’d DSS Grade 2209 and 31% higher than the (Ji) value of WAAM’d
SDSS Grade 2509.
Keywords: Additive Manufacturing, Super Duplex Stainless Steel, Cold Metal
Transfer Welding, Response Surface Method.

iii
 b ÖZ

TEL ARK EKLEMELİ İMALAT YÖNTEMİYLE DUBLEKS PASLANMAZ

ÇELİK PLAKA ÜRETİMİNDE PROSES PARAMETRELERİNİN
OPTİMİZASYONU VE PLAKA MEKANİK ÖZELLİKLERİNİN
İNCELENMESİ

SAADAWI, HASSAN

Doktora, Mühendislik Sistemlerinin Modellenmesi ve Tasarımı Ana Bilim Dalı

Tez Yöneticisi: Doç.Dr. Erkan Konca
Ortak Tez Yöneticisi: Dr. Öğr.Üyesi Kâzım Tur
Ocak 2022, 140 sayfa
Son yıllarda, malzemelerin katman katman biriktirilmesi yoluyla ürünler oluşturmak
için kullanılan eklemeli imalat (Eİ), çeşitli endüstrilerde üretim süreçlerinin
verimliliğini artırmanın bir yolu olarak ilgi konusu olmuştur. Bu tezde, süper dubleks
paslanmaz çelik plaka parçalar soğuk metal transferi kaynak tekniği kullanılarak tel
ark eklemeli imalat (WAAM) ile elde edilmiştir. İstenen kaynak kalitesini veren
eklemeli üretim parametrelerinin (voltaj, akım ve hız) optimum kombinasyonunu elde
etmek için, eklemeli kaynak işleminin nümerik simülasyonuna dayalı istatistiksel
modeller oluşturmak üzere yanıt yüzeyi yöntemi kullanılmıştır. Deneysel süreci
çalışmak üzere COMSOL Multiphysics 5.5 yazılımı ve istatistiksel modelleri elde
etmek için Expert-Design yazılımı kullanılmıştır. Kontrol edilen değişkenlerden akım
ve gerilim için optimum değerler sırasıyla 200 amper ve 15 volt, kaynak hızı için ise
10 mm/sn olarak bulunmuştur. Daha sonra, sıfır altı sıcaklıklarda tek kenar çentikli
çekme (SENT) numuneleri kullanılarak kırılma tokluğu testleri gerçekleştirilmiştir.
Kırılma tokluğu verileri, hem haddelenmiş plaka hem de eklemeli imalat ile üretilmiş
plaka parçalar için çatlak ilerleme direnci eğrileri oluşturmak ve J-integraline (enerji
salım hızı) dayalı sonuçları karşılaştırmak için kullanılmıştır. SDSS Grade 2507
haddelenmiş plaka numuneleri için kararlı çatlak büyümesi başlangıcındaki J-integral
değeri (Ji), WAAM DSS Grade 2209'un (Ji) değerinden yaklaşık %17 ve WAAM
SDSS Grade 2509'un (Ji) değerinden ise %31 daha yüksek olarak belirlenmiştir.
Anahtar Kelimeler: Eklemeli İmalat, Süper Dubleks Paslanmaz Çelik, Soğuk Metal
Transfer Kaynağı, Tepki Yüzey Yöntemi.

iv
 c DEDICATION

I would like to dedicate this small piece of work to

My mother soul, My father soul and My wife and children

v
 d ACKNOWLEDGMENTS

Thanks to Allah almighty who enabled me to finish my thesis successfully. It is

a privilege for me to express my sincere gratitude to my thesis supervisors
Assoc.Prof.Dr. Erkan Konca and Asst.Prof. Dr. Kâzım Tur for their supervision,
guidance, training, help and continuous encourage during my work. My gratitude is
also extended to my committee members: Prof. Dr. Bilgehan Ögel and Asst.Prof.
Dr. Ozan Özkan for their comments and suggestions during all phases of thesis. I am
grateful to all my thesis committee members for their contributions. I also would like
to express my appreciation and gratefulness to Prof. Dr. C. Hakan Gür, the chair of
the Department of Metallurgical and Materials Engineering, and the director of
Welding Technology and Non-Destructive Testing Research/ Application Center
(KTTMM) at Middle East Technical University, Dr. Koray Yurtışık the head of
Welding R&D unit at KTTMM, engineers and technicians of KTTMM who made all
possible faithfully to help me especially, Eng. Mine Kalkancı and Eng. Batuhan
Ersan. Finally, my appreciation goes to my parents, my wife, my children and my
brothers who encouraged and supported me throughout the time of this work.

vi
 e TABLE OF CONTENTS

ABSTRACT ................................................................................................................ iii

ÖZ ............................................................................................................................... iv

DEDICATION ............................................................................................................. v

ACKNOWLEDGMENTS .......................................................................................... vi

TABLE OF CONTENTS ........................................................................................... vii

LIST OF FIGURES ................................................................................................... xii

LIST OF TABLES .................................................................................................... xvi

LIST OF SYMBOLS AND ABBREVIATIONS .................................................... xvii

CHAPTER 1 ................................................................................................................ 1

INTRODUCTION .................................................................................................... 1

1.1 Background ..................................................................................................... 1

1.2 Relevant Studies ............................................................................................. 3

1.3 Knowledge Gap .............................................................................................. 4

1.4 Objectives ....................................................................................................... 4

1.5 Details of Thesis Format................................................................................. 4

CHAPTER 2 ................................................................................................................ 6

THEORY .................................................................................................................. 6

2.1 General............................................................................................................ 6

2.2 Additive Manufacturing Technologies ........................................................... 6

2.3 Benefits and Constraints ................................................................................. 7

2.3.1 Benefits of AM Technologies .................................................................. 7

2.3.2 Limitations of AM Technologies ............................................................. 8

2.4 Classification of Additive Manufacturing Technologies ................................ 8

2.4.1 Liquid Based Technologies...................................................................... 8

vii
 2.4.2 Powder Based Technologies .................................................................... 9

2.4.3 Wire Based Technologies ........................................................................ 9

2.5 Wire Arc Additive Manufacturing (WAAM) ................................................. 9

2.5.1 GTAW Based WAAM System .............................................................. 11

2.5.2 PAW Based WAAM System ................................................................. 11

2.5.3 GMAW Based WAAM System ............................................................. 12

2.6 Cold Metal Transfer ...................................................................................... 12

2.6.1 Cold Metal Transfer Welding Cycle ...................................................... 14

2.6.2 Cold Metal Transfer Welding Parameters ............................................. 16

2.7 The Material; Super Duplex Stainless Steel ................................................. 17

2.7.1 General ................................................................................................... 17

2.7.2 Brief History .......................................................................................... 18

2.7.3 Duplex Stainless Steels Grades; Past and Present ................................. 19

2.7.4 Super Duplex Stainless Steel Grades ..................................................... 21

2.8 Physical Metallurgy and Microstructure of SDSS ....................................... 22

2.8.1 Structure Stability and Solidification Mode .......................................... 22

2.8.2 The Role of Alloying Elements in Duplex Stainless Steels ................... 24

2.8.3 Desired Microstructure .......................................................................... 24

2.8.4 Precipitation of Secondary Phases ......................................................... 24

2.9 Metallurgy of Additively Manufactured SDSS ............................................ 26

2.10 The Property: Fracture Toughness ............................................................. 27

2.11 Fracture Mechanics ..................................................................................... 27

2.12 Linear Elastic Fracture Mechanics ............................................................. 28

2.13 Elastic Plastic Fracture Mechanics ............................................................. 30

2.13.1 Crack Tip Opening Displacement (CTOD) ......................................... 30

2.13.2 J- Integral ............................................................................................. 31

viii
 2.14 Crack-Growth Resistance Curves ............................................................... 32

2.15 Methods of Measuring and Types of Specimen ......................................... 32

2.16 Fracture of Duplex Stainless Steels ............................................................ 34

CHAPTER 3 .............................................................................................................. 35

NUMERICAL SIMULATION STUDY ................................................................ 35

3.1 General.......................................................................................................... 35

3.2 Wire Arc Additive Manufacturing Process Variables .................................. 36

3.3 Response Surface Design Matrix ................................................................. 37

3.4 Numerical Modelling and Simulation of WAAM Process ........................... 38

3.5 Numerical Model Description ...................................................................... 39

3.6 Numerical Model Assumptions .................................................................... 39

3.7 Governing Equations and Boundary Conditions .......................................... 40

3.8 Model Validation .......................................................................................... 40

3.9 Simulation Results ........................................................................................ 44

3.10 Creation of Statistical Models .................................................................... 61

3.11 Checking the Adequacy of the Models ....................................................... 68

3.12 Optimization ............................................................................................... 68

3.13 Discussion ................................................................................................... 69

3.13.1 Interaction effects of welding current and voltage on distortion .......... 69

3.13.2 Interaction effects of welding current and welding speed on distortion

......................................................................................................................... 69

3.13.3 Interaction effects of welding voltage and welding speed on distortion

......................................................................................................................... 70

CHAPTER 4 .............................................................................................................. 71

EXPERIMENTAL STUDIES ................................................................................ 71

4.1 General.......................................................................................................... 71

ix
 4.2 Materials ....................................................................................................... 71

4.3 Wire Arc Additive Manufacturing Procedure .............................................. 72

4.3.1 Equipment .............................................................................................. 72

4.3.2 Sample Build-up .................................................................................... 72

4.4 Integrity Assessment of the Additively Manufactured Beads ...................... 75

4.5 Mechanical Characterization ........................................................................ 76

4.5.1 Tensile Test ............................................................................................ 76

4.5.2 Hardness Test ......................................................................................... 77

4.6 Microstructural Examination ........................................................................ 77

4.7 Fracture Toughness Testing using SENT According to BS 8571 ................ 77

4.7.1 Post-Test Measurements ........................................................................ 81

4.7.2 Calculating J........................................................................................... 82

4.7.3 Definition of the R-curve ....................................................................... 83

CHAPTER 5 .............................................................................................................. 85

RESULTS AND DISCUSSION ............................................................................ 85

5.1 Background ................................................................................................... 85

5.2 Integrity Assessment of the Additively Manufactured Beads ...................... 85

5.2.1 Appearance and parts characterization .................................................. 85

5.2.2 Macroscopic characterization ................................................................ 86

5.2.3 Radiographic Examination..................................................................... 86

5.3 Microstructure Characterization ................................................................... 87

5.3.1 As-Received Product Forms .................................................................. 87

5.3.2 WAAM deposited beads ........................................................................ 88

5.4 Mechanical Characterization ........................................................................ 95

5.4.1 Tensile Test ............................................................................................ 95

5.4.2 Hardness Test ......................................................................................... 97

x
 5.5 Fracture Toughness Testing.......................................................................... 98

5.5.1 Fracture Toughness of As-received SDSS Rolled Plate Grade 2507 .. 100

5.5.2 Fracture Toughness of WAAM’d DSS Grade 2209 Plate ................... 103

5.5.3 Fracture Toughness of WAAM’d SDSS Grade 2509 Plate ................. 106

CHAPTER 6 ............................................................................................................ 110

CONCLUSIONS .................................................................................................. 110

6.1 CONCLUSIONS ........................................................................................ 110

6.2 SUGGESTIONS FOR FUTURE WORK .................................................. 111

REFERENCES......................................................................................................... 112

APPENDIX A .......................................................................................................... 119

The Developed Matlab Codes to calculate the crack length and J value ............. 119

APPENDIX B .......................................................................................................... 123

Fracture Surface and Load-CMOD Curve of Rolled Plate and Additively

Manufactured Plate .................................................................................................. 123

APPENDIX C .......................................................................................................... 135

Crack Length Measurements of Rolled Plate and Additively Manufactured Plate . 135

xi
 f LIST OF FIGURES

Figure 2.1 Examples of AM application ...................................................................... 7

Figure 2.2 Classification of AM processes based on the state of the raw material...... 9

Figure 2.3 Electron beam and Laser-beam AM ........................................................ 10

Figure 2.4 Gas tungsten arc welding .......................................................................... 11

Figure 2.5 Plasma arc welding .................................................................................. 12

Figure 2.6 GMAW welding transfer modes. ............................................................. 13

Figure 2.7 Gas metal arc welding .............................................................................. 13

Figure 2.8 Fronius CMT welding system ................................................................. 15

Figure 2.9 CMT welding cycle ................................................................................. 16

Figure 2.10 Microstructure of duplex stainless steels ................................................ 17

Figure 2.11 Schaeffler diagram ................................................................................. 18

Figure 2.12 Some applications where DSSs are employed........................................ 19

Figure 2.13 The solubility limit of nitrogen in DSS ................................................. 22

Figure 2.14 Pseudo-phase binary diagram of duplex grades (2205/2507) ................ 23

Figure 2.15 Time-temperature precipitation diagram of duplex stainless steels ....... 25

Figure 2.16 Conditions of flaw propagation .............................................................. 28

Figure 2.17 Plate with the initial crack ...................................................................... 28

Figure 2.18 The three modes of loading .................................................................... 29

Figure 2.19 Displacement at the original crack tip .................................................... 31

Figure 2.20 Crack-growth resistance, R- curve ......................................................... 32

Figure 2.21 Standardized fracture mechanics test specimens .................................... 33

Figure 3.1. The plan of study ..................................................................................... 35

Figure 3.2 Response surface design for k=3 variables............................................... 36

xii
Figure 3.3 Schematic diagram and geometrical size of the computational domain... 39

Figure 3.4 Cooling curves recorded in the middle of weld pools .............................. 42

Figure 3.5 Heat distribution of case 1 ........................................................................ 42

Figure 3.6 Heat distribution of case 2 ........................................................................ 43

Figure 3.7 Heat distribution of case 3 ........................................................................ 43

Figure 3.8 Simulation result of experiment (1). ......................................................... 45

Figure 3.9 Simulation result of experiment (2). ......................................................... 46

Figure 3.10 Simulation result of experiment (3). ....................................................... 47

Figure 3.11 Simulation result of experiment (4). ....................................................... 48

Figure 3.12 Simulation result of experiment (5). ....................................................... 49

Figure 3.13 Simulation result of experiment (6). ....................................................... 50

Figure 3.14 Simulation result of experiment (7). ....................................................... 51

Figure 3.15 Simulation result of experiment (8). ....................................................... 52

Figure 3.16 Simulation result of experiment (9). ....................................................... 53

Figure 3.17 Simulation result of experiment (10). ..................................................... 54

Figure 3.18 Simulation result of experiment (11). ..................................................... 55

Figure 3.19 Simulation result of experiment (12). ..................................................... 56

Figure 3.20 Simulation result of experiment (13). ..................................................... 57

Figure 3.21 Simulation result of experiment (14). ..................................................... 58

Figure 3.22 Simulation result of experiment (15). ..................................................... 59

Figure 3.23 (a) Response surface and (b) contour surface of the interaction effect of
current and voltage on longitudinal distortion ........................................................... 62

Figure 3.24 (a) Response surface and (b) contour surface of the interaction effect of
current and speed on longitudinal distortion .............................................................. 63

xiii
Figure 3.25 (a) Response surface and (b) contour surface of the interaction effect of
voltage and speed on longitudinal distortion ............................................................. 64

Figure 3.26 (a) Response surface and (b) contour surface of the interaction effect of
current and voltage on transversal distortion ............................................................. 65

Figure 3.27 (a) Response surface and (b) contour surface of the interaction effect of
current and speed on transversal distortion ................................................................ 66

Figure 3.28 (a) Response surface and (b) contour surface of the interaction effect of
voltage and speed on transversal distortion ............................................................... 67

Figure 3.29 Scatter plot of the actual and predicted values distortion ....................... 68

Figure 4.1 Fronius TPS 4000 CMT power source ..................................................... 73

Figure 4.2 Schematic of WAAM setup ...................................................................... 74

Figure 4.3 Real WAAM’d sample for Grade 2509 .................................................... 75

Figure 4.4 Tensile test samples .................................................................................. 76

Figure 4.5 SENT specimens ....................................................................................... 78

Figure 4.6 Orientations of the SENT samples ........................................................... 78

Figure 4.7 SENT specimens place and orientation in HAZ for Grade 2509 ............. 79

Figure 4.8 Fatigue pre-crack ...................................................................................... 80

Figure 4.9 Fracture toughness test setup .................................................................... 80

Figure 4.10 Measurement of crack length ................................................................. 81

Figure 4.11 Load versus CMOD curve ...................................................................... 82

Figure 4.12 Fitting the R-curve to the SENT crack extension data ........................... 84

Figure 5.1 Additively manufactured plates of Grade 2509 ........................................ 85

Figure 5.2 Macro-image of the cross-section of Grade 2509..................................... 86

Figure 5.3 X-ray films ................................................................................................ 87

Figure 5.4 Microstructure of As-received SDSS Grade 2507 (100X) ....................... 87

Figure 5.5 Microstructural examination points .......................................................... 88

xiv
Figure 5.6 Microstructures of fusion line 1 ................................................................ 89

Figure 5.7 Microstructures of fusion line 2 ................................................................ 90

Figure 5.8 Microstructures of (OHAZ) lower fusion line .......................................... 91

Figure 5.9 Microstructures of weld metal upper fusion line ...................................... 92

Figure 5.10 Ferrite/austenite ratio at fusion line 1 ..................................................... 93

Figure 5.11 Ferrite/austenite ratio at fusion line 2 ..................................................... 94

Figure 5.12 Ferrite/austenite at OHAZ. ..................................................................... 94

Figure 5.13 Ferrite/austenite weld metal. ................................................................... 95

Figure 5.14 Stress-Strain curves of Grade 2507 and Grade 2509 SDSS samples ..... 96

Figure 5.15 Specimens (a) before and (b) after the tensile tests. ............................... 97

Figure 5.16 Hardness investigation area, WAAM’d Grade 2509 .............................. 97

Figure 5.17 Micro-hardness map. .............................................................................. 98

Figure 5.18 Fracture toughness samples (a) before and (b) after the tests. ................ 98

Figure 5.19 Measurements points .............................................................................. 99

Figure 5.20 Typical J-R curve .................................................................................. 100

Figure 5.21. (a) Fracture surface, (b) Load-CMOD of Grade 2507 sample 1 ......... 101

Figure 5.22 J-integral vs crack extension curve for as-received SDSS plate........... 102

Figure 5.23. (a) Fracture surface, (b) Load-CMOD of Grade 2209 sample 1 ......... 104

Figure 5.24 J-integral vs crack extension curve for WAAM’d Grade 2209 plate ... 105

Figure 5.25 J R-curve lower-bound Fitting Parameters ........................................... 106

Figure 5.26 Load-CMOD of WAAM’d Grade 2509 plate at the weld metal. ......... 107

Figure 5.27 Load-CMOD of WAAM’d SDSS Grade 2509 plate at OHAZ. ........... 107

Figure 5.28 J-R curve of different DSS Grades ....................................................... 108

xv
 g LIST OF TABLES

Table 2.1 Chemical compositions of the modern duplex stainless steels .................. 21

Table 3.1 Control variables ........................................................................................ 36

Table 3.2 WAAM process variables with their limits................................................ 37

Table 3.3 Established design matrix .......................................................................... 38

Table 3.4 Super duplex stainless steel properties....................................................... 39

Table 3.5 Actual and simulation results of temperature ............................................ 44

Table 3.6 Response values ......................................................................................... 60

Table 3.7 Coefficient values of models ..................................................................... 61

Table 3.8 Criterion of numerical optimization........................................................... 69

Table 3.9 Desirable solution selected......................................................................... 69

Table 4.1 Chemical compositions of the steels used (wt%)....................................... 72

Table 4.2 Constant process parameters. ..................................................................... 74

Table 5.1 Mechanical properties of Grade 2507 and Grade 2509 super duplex stainless
steel. ........................................................................................................................... 96

Table 5.2 Crack length of as-received SDSS plate Grade 2507 sample 1 ............... 101

Table 5.3 J-integral values of the as-received SDSS plate Grade 2507 ................... 102

Table 5.4 Crack length of WAAM’d DSS Grade 2209 plate sample 1 ................... 103

Table 5.5 J-integral values of the WAAM’d DSS Grade 2209 plate ....................... 104

Table 5.6 Fitting parameters of J integral ................................................................ 105

Table 5.7 J-integral values of WAAM’d SDSS Grade 2509 plate. ......................... 108

Table 5.8 J-integral values of different DSS Grades................................................ 109

xvi
 h LIST OF SYMBOLS AND ABBREVIATIONS

AM : Additive Manufacturing
PBF : Powder Bed Fusion
DED : Directed Energy Deposition
WAAM : Wire Arc Additive Manufacturing
SDSS : Super Duplex Stainless Steels
CMT-W : Cold Metal Transfer Welding
PREN : Pitting Resistance Equivalent Number
HAZ : Heat Affected Zone
SENT : Single Edge Notched Tension
q : Heat Flux (W/m2)
Q : Power of Heat Source (W)
CMOD : Crack Mouth Opening Displacement (mm)
CTOD : Crack Tip Opening Displacement (mm)
J : J-integral Fracture Toughness (N/mm)
K : Stress intensity factor (N/mm1.5)
a0 : Initial crack length (mm)
af : Final crack length (mm)
E : Modulus of elasticity (GPa)
Jel : Elastic component of J (N/mm)
Jep : Plastic component of J (N/mm)
Up : Area under the plastic part of the load - CMOD curve (N·mm)
np : Dimensionless function of geometry

xvii
 i CHAPTER 1

1. INTRODUCTION

1.1 Background

Competition in the manufacturing industry is very harsh and in order to obtain

competitive advantages the companies must continuously work on product
development. Meanwhile, the companies should also follow the developments in the
manufacturing technology. This is because there is a symbiosis between technology
and product development, and it is only if both are concurrently enhanced then
successful products can be accomplished with optimal lead time [1].
In recent years, additive manufacturing (AM) for metals has been a subject of great
interest as a means of enhancing the efficiency of manufacturing processes in various
industries, such as aircraft, military, automotive, oil, medical [2, 3]. AM is the
formalized term for what was originally called as rapid prototyping or 3D printing, and
today it is commonly used to create products using layer-by-layer deposition of
materials. Metal AM process can be broadly classified into two groups, Powder Bed
Fusion Additive Manufacturing (PBF) and Directed Energy Deposition Additive
Manufacturing (DED). In PBF AM, material in powder shape is fused in layers by a
heat source. In the DED AM, the products are produced by adding material in the shape
of powder, wire or sheets in layers and fused with a heat source [2, 4].
Among various AM processes, wire arc additive manufacturing (WAAM) deposits a
metal wire using an arc heat source. In the WAAM process, a traditional welding
machine may be used as a heat source that decreases the initial investment costs
compared to the laser and electron beam heat sources. In addition, the cost of raw
materials of the wire type is roughly 10 per cent of the cost of raw materials of the
powder type. The WAAM process therefore produces a high deposition rate at a low
cost, making it a highly favored process for the manufacturing of large components
using expensive materials [3, 5].
Today, WAAM has become a popular manufacturing tool for various engineering
materials such as titanium, aluminum, nickel alloys and steels. Compared to traditional

1
subtractive manufacturing, WAAM technology can reduce fabrication time and post-
machining time depending on the component size [1, 6].
More recently, there has been an increasing interest in applying AM to super duplex
stainless steels (SDSS) due to the greater demand for SDSSs in the offshore
applications. Since SDSSs have excellent mechanical properties and corrosion
resistance, they are suitable for use in harsh corrosive environments containing
chlorides such as seawater [7, 8]. It is well known that an adequate percentage of
austenite (γ) and ferrite (α) phases in the resulting duplex stainless steels
microstructure plays a key role in the final properties needed. The ferrite phase offers
strength and corrosion resistance while the austenite phase increases the ductility.
However, when SDSS parts are made using additive manufacturing technology the
subsequent pass reheats the metal deposited by the preceding pass. As result, undesired
ratio of α/γ phases is obtained since austenite fraction is increased due to precipitation
of secondary austenite [8, 9].

As mentioned earlier, SDSSs are used in marine structures where low temperatures are
not uncommon. At sub-zero temperatures, the fracture mechanism for ferrite goes from
ductile to brittle with low fracture deformation. The move from one mechanism to
another is called as the ductile-to-brittle transition [10]. Therefore, precise prediction
of fracture toughness plays a critical role in any structural integrity assessment. These
structural integrity assessments rely on the assumption that the fracture resistance of
the product is equal to the resistance obtained from the toughness test specimen [11].

The focus of the present work has been the study of the fracture toughness of wire arc
additively manufactured super duplex stainless steels at sub-zero temperatures using
single edge notched tension (SENT) specimens. To obtain the optimum combination
of additive manufacturing parameters that yield the desired bead quality, response
surface method has been used to build statistical models. Numerical simulation was
used to perform the experimental instead of real welding process. Experimental work
of this thesis study has been carried out at the Welding Technology and Non-
Destructive Testing Research/Application Center at Middle East Technical University
(KTTMM-METU) in Ankara, Turkey.

2
1.2 Relevant Studies

Few studies have investigated additive manufacturing of duplex stainless steels.

Davidson et al. [12] fabricated super duplex stainless steels (type-2507) using selective
laser melting (SLM) and found that the as-built piece contained 93% ferrite with some
Cr2N particles. Heat treatment at 1040°C reduced the ferrite fraction to 55%.
Hengsbach et al. [13] used SLM technology to produce a type-2205 DSS piece, where
99% ferrite was obtained, while heat treatment at 1000°C reduced the ferrite fraction
to 66%. Posch et al. [14] investigated WAAM of DSS using 2209 grade filler metal
and cold metal transfer (CMT) welding process. The microstructure obtained had a
ferrite number of 30 FN (standardized value which is related to the ferrite content in
austenitic and duplex stainless steels weld metal) and mechanical properties were
comparable with the filler metal. Eriksson et al. [15] also studied the CMT process for
the additive manufacturing of SDSS, where a 20% ferrite fraction and acceptable
mechanical properties were achieved. Papula et al. [16] produced a high-density
duplex stainless steel 2205 specimens (<0.03% porosity) by selective laser melting
(SLM) and good mechanical properties were achieved. The microstructure was almost
fully ferritic, but the desired microstructure was obtained by post-processing heat
treatment. Hejripour et al. [17] studied the thermal behavior of two different parts
(Wall and Tube) made of 2209 duplex stainless steel (DSS) using GMAW process.
They used the thermal cycle data to study the evolution of microstructure and found
that slow cooling significantly promoted austenite formation in a ferrite matrix.
Hosseini et al. [18] studied the microstructures of 2209 duplex stainless steel additively
manufactured by GMAW technique with high and low heat input and they found that
the austenite fraction increased up to 10% in the reheated beads rather than first bead
that increased up to 8% for high and low heat input. Stützer et al. [19] used GMAW-
cold wire AM technique to improve alloy compositions of DSS by mixing filler wires
with different nickel contents. They melted approximately 23% cold wire with the wire
electrode in the AM experiments and made it possible to achieve 39–72% ferrite in the
specimens. Lervåg et al. [20] fabricated additively manufactured super duplex stainless
steel parts using CMT process with three different heat inputs and studied the effect of
heat input on the mechanical properties. They found that a slight reduction in
toughness occurs when the heat input was increased, also yield strength, tensile
strength and hardness were decreased with increasing heat input.
3
1.3 Knowledge Gap

Duplex stainless steels are increasingly used for marine applications which are
exposed to low service temperatures. Cold metal transfer wire arc additive
manufacturing has a very good potential to produce DSS components. However, there
is a significant lack of information/knowledge on different aspects of this
manufacturing technique that must be clarified/explored before it can be proposed to
be used in practice. For example, the effects of wire arc welding process parameters
on the distortion of SDSS parts produced by WAAM have not been studied yet. In
addition, the knowledge on the mechanical properties of SDSS components produced
by WAAM, especially their fracture behavior at sub-zero temperatures, is very limited.

1.4 Objectives

The objectives of this research are:

 To perform heating-cooling/thermal cycle simulations in order to find the
optimum combination of wire arc welding process parameters that minimize
the distortion of DSS parts produced by WAAM.
 To conduct WAAM of DSS Grade 2209 and Grade 2509 using CMT welding
technology with optimized process parameters.
 To examine the microstructural evolution and to determine the strength of DSS
parts produced by CMT WAAM.
 To perform and evaluate fracture toughness testing of additively manufactured
DSS at sub-zero temperatures using the single edge notched tension (SENT)
specimens and compare the results with those of as received DSS Grade 2507
rolled plate..

1.5 Details of Thesis Format

The thesis starts with the background literature relevant to this research project in
Chapter 2. It includes a history of AM and a few of the technologies available today.
This chapter also briefly describes the properties of duplex stainless steel, which lead
into the specific super duplex stainless steel. The fracture mechanics analysis approach
employed in this project is also detailed. The simulation works are presented in

4
Chapter 3. The experimental procedures of the project, as well as the equipment used
to achieve results, are described in Chapter 4. Chapter 5 discusses the results of all
experiments and the information that can be derived from them. In Chapter 6, the thesis
is ultimately concluded together with suggestions for future work.

5
 j CHAPTER 2

2. THEORY
2.1 General

This chapter provides a theoretical background for,

 Additive manufacturing technologies, concepts and classifications including

cold metal transfer, as the wire arc additive manufacturing process.

 Super duplex stainless steel, as the material

 Fracture toughness, as the property

2.2 Additive Manufacturing Technologies

Additive Manufacturing (AM) has been in existence for three decades and was initially
used for model making and prototyping. AM technology has now evolved into a rapid
manufacturing process [2]. AM refers to a collection of techniques used to create three-
dimensional parts by depositing and melting successive layers of base material using
an additive technique. AM is seen as opposed to conventional subtractive processes
using machining operations (such as turning, milling, grinding) or molding methods
(such as pressing, casting, injection molding). AM processes are often described by
terms such as 3D printing, rapid prototyping, direct digital manufacturing, fast
manufacturing, and solid freeform manufacturing. Although AM does not replace any
of the conventional manufacturing processes, it represents a key enabler for
development and innovation, and it also adds capacity to complement traditional
processes as well. Therefore, AM is receiving more and more attention and effort
around the world because there is a tremendous worldwide interest in evaluating AM's
potential as a useful and possibly disruptive technology. AM technology is rapidly
developing and getting more and more integrated into manufacturing industry and our
daily lives [21]. AM can handle different kinds of materials such as polymers,
ceramics, and metals [2]. Indeed, AM's excitement encompasses many fields, from
computer science and product design to new materials and lean engineering [22].
Figure 2.1 shows some applications of AM.

6
 (a) (b)
Figure 2.1 Examples of AM application: (a) steel propeller, (b) titanium structure
element of aircraft wing [22]

However, public attention has been focused on AM only in recent years [22]. AM
innovations have already developed themselves at the level of actual production in
some sectors [22].

One of the current problems for AM is that many traditional manufacturing sectors are
not fully aware of, or do not fully understand, how AM can be used. AM will replace
some of the production methods, but not all of them, while it has the added potential
to complement many of those it cannot replace [22]. AM is much more than a process
used to make customized novel items or prototypes. With new developments in AM,
we are living in an age at the cusp of industrialized rapid manufacturing taking over as
a process for producing many products as well and making designing and creating new
ones possible [21].

2.3 Benefits and Constraints

2.3.1 Benefits of AM Technologies

Nowadays AM of metal parts is well known as an important alternative to traditional

processes because of its ability to minimize component costs by minimizing material
wastage and time to market. In addition, AM can also enable greater design freedom,
which potentially leading to weight savings as well as facilitating the manufacture of
complex assemblies that used to consist of many subcomponents [22, 23].

7
2.3.2 Limitations of AM Technologies

In order to take full advantage of AM technology, it is important to be aware of certain

limitations of this technology, such as:
- Part size: Part size limited to system size in some AM technologies.
- Processability of different materials: Although many alloys are available, specific
approaches are required for difficult-to-weld alloys.
- Post-processing: In some cases, machining is required to achieve better surface
finishing or dimensional precision.
- Residual stresses and distortions are major concerns for AM. Part tolerances could
be affected. Premature failure has also been an issue [22].

2.4 Classification of Additive Manufacturing Technologies

The AM area has been developed over the past two decades as a result of various
technologies being researched and developed [2]. In the frame of AM technology,
various processes can be described, where a range of materials (including polymers,
metals, ceramics and composites), different techniques of deposition and fusion are
applied [22]. Because of its intrinsic characteristics, each method is naturally suited to
some applications [23]. American Society of Testing Materials (ASTM) Committee
F42 has categorized AM processes to seven categories as Vat polymerization, Material
jetting, Material extrusion, Sheet lamination, Binder jetting, Powder bed fusion and
Directed energy deposition [22].

The uses of metal have been gaining popularity in AM technology. For the AM of
metallic parts, AM processes could be classified based on the state of the raw material
as liquid, powder and wire based processes as shown in Figure 2.2 [22, 24].

2.4.1 Liquid Based Technologies

The application of liquid-based material is relatively new in metal AM. This

technology melts the raw material and propels the fused metal droplets to build up a
metal component using an electromagnetic field. This speeds up the manufacturing of
components with high-intensity [24].

8
 Additive Manufacturing
Technologies

Liquid based Powder based Wire based

Liquid Metal 3D Printing Selective Laser Melting Wire Laser AM

Solid Freeform Fabrication Selective Laser Sintering Wire Electron Beam AM

Wire Arc AM

Figure 2.2 Classification of AM processes based on the state of the raw material

2.4.2 Powder Based Technologies

Powder bed fusion (PBF) is an additive manufacturing process based on powder as

raw material where a source of heat selectively fuses a region of a powder bed. Powder
bed fusion is the most commonly used method for metal build up with AM. Although
PBF process is suitable for the fabrication of small components because of better
geometrical and dimensional accuracy, its deposition rate and the fabricated build
volume is relatively low. In order to melt powder bed regions either a laser beam or an
electron beam is used [22, 24].

2.4.3 Wire Based Technologies

Due to their high potential for use in metal AM, wire based additive manufacturing
has been extensively studied recently. A simple system of wire based AM consists of
a motion system, heat source and feedstock [23, 24]. Wire based AM includes three
different techniques: laser, electron beam and arc welding. Laser and electron beam
based techniques are typically the same [24] as schematically represented in Figure 2.3
[22].

2.5 Wire Arc Additive Manufacturing (WAAM)

WAAM is a near-net shape AM system that allows complex parts to be built with
minimum finishing. The system builds up the components with layer-by-layer
deposition of material. Complex parts can be built up with enhanced material

9
properties, and hybrid components can be created. The savings in time and material
are expected to make WAAM an attractive option for large part manufacturing [25].

Electron LASER
beam gun beam gun

Wire feeder Electron LASER Wire feeder

beam beam
Deposit layers Substrate Deposit layers

a b

Figure 2.3 Electron beam and Laser-beam AM [22]

WAAM technologies have been investigated to be used as an AM method since 1990s.
Recent improvements in welding have made it possible to use a multi-choice welding
source instead of a laser or an electron beam source [26]. WAAM uses welding
equipment: welding power source, torches and wire feeding systems. Motion can be
provided by robotic systems or computer numerical controlled table [22, 24].
High deposition rates, low cost of materials and equipment, and good structural
integrity make WAAM a suitable candidate for replacing the current manufacturing
process, especially with respect to parts of low and medium complexity. For this AM
process, a wide variety of materials are available including steels, stainless steels,
nickel and titanium alloys [22, 23, 27].

Residual stresses and distortions are of great concern for the WAAM of large scale
components. Distortions result in poor tolerances while residual stresses affect the
component's mechanical behavior. In addition, post-processing operations are required
due to the poor finish of the surface obtained with WAAM [22].

In the WAAM process, 3D metallic components are built by depositing beads of weld
metal layer by layer, using welding processes such as gas metal arc welding (GMAW),

10
gas tungsten arc welding (GTAW), and plasma arc welding (PAW) combined with a
motion system [24, 26].

2.5.1 GTAW Based WAAM System

GTAW uses a non-consumable tungsten electrode combined with a separately fed wire
to produce the weld deposit, as shown in Figure 2.4. During the deposition process,
wire feed orientation affects material transfer and the deposit consistency. Back
feeding, side feeding, and front feeding can be used. A GTAW torch with a gas lens is
used to generate laminar flow of shielding gas to reduce oxidation. Gas lens is one of
modern GTAW torch components, composed of a copper body with layered mesh
screens of steel or stainless steel [27].

Tungsten electrode

Contact tube
Shielding gas
Shielding
gas nozzle Power
supply
Gas lens

Filler metal Arc

Figure 2.4 Gas tungsten arc welding [27]

2.5.2 PAW Based WAAM System

PAW has also been widely studied as a process for the AM processing of metallic
materials. In plasma welding, arc energy density can reach three times that of GTAW,
resulting in fewer weld distortions and smaller welds with higher welding speeds.
Figure 2.5 shows WAAM system based on PAW [27].
11
 Tungsten electrode
Plasma gas
Contact tube nozzle
Shielding gas
Plasma gas
Shielding
gas nozzle

Power
supply

Filler metal Arc

Figure 2.5 Plasma arc welding [27].

2.5.3 GMAW Based WAAM System

GMAW is a welding process in which an electric arc is formed between a consumable

wire electrode and the substrate metal. The wire is usually perpendicular to the
workpiece [27]. GMAW is the most common heat source choice for AM due to its
welding wire direct-feeding spool, which is coaxial to the welding torch [24]. As
depicted in Figure 2.6, there are four welding modes in GMAW: the globular mode,
short-circuiting mode, spray mode, and pulsed-spray mode. Cold metal transfer (CMT)
was later introduced as modified GMAW [28]. CMT performs better than GMAW
because of its capacity to produce a high deposition rate with a lower heat input [23,
24, 27]. Figure 2.7 shows a GMAW based WAAM system.

2.6 Cold Metal Transfer

Cold Metal Transfer welding is a modified GMAW process which was developed by
Fronius of Austria in 2004 [29]. An illustration of Fronius CMT welding set-up is
shown in Figure 2.8 [30]. CMT can be regarded as a short circuit GMAW method in
its most elementary form. GMAW short circuit comprises the lowest range of welding
currents and electrode diameters.
12
 50

Globular
mode
40
Spray mode
Voltage

30 Pulsed mode

Short-circuit
mode
20

0 150 300 450

Current

Figure 2.6 GMAW welding transfer modes. [28]

Wire feed
rolls

Wire electrode

Contact tube

Shielding gas Power

supply
Shielding
gas nozzle
Arc

Figure 2.7 Gas metal arc welding [27]

This kind of transfer provides a tiny, quick-freezing deposited metal. During the arcing
process, metal is transferred from the wire to the substrate only in a short period of

13
time, when the electrode is in contact with the weld pool; hence the term "short circuit
transfer" which means no metal is transferred across the arc gap [31]. It is critical that
the current density must be sufficiently high to heat the electrode and facilitate metal
transfer, but low enough to reduce spatter caused by violent separation of the metal
drop [31].

The CMT unit consists of two separate wire-drives (front and rear), separated by the
wire buffer. The front drive (servomotor), located in the torch, (see Figure 2.8-a),
drives the wire forth and back at a pace of up to 90 times per second in a dabbing
motion. At the same time, the rear drive drives the wire directly from the filler spool
within the wire drive. It is important to note that Process Control digitally controls
both drives. A wire buffer (Figure 2.8-b) is interposed between the two drives to
decouple them from one another to ensure a constant wire feed [31].

Unlike the GMAW, CMT incorporates a digital process-control that detects the short
circuit at the workpiece, and then retracts the wire mechanically to help detach the
molten droplet. The retraction of the wire significantly reduces the spatter which is
usually associated with traditional short circuit GMAW. The second most noticeable
difference is a reduction in the thermal input, since the droplet detachment is
practically current-free, hence the word “Cold Metal Transfer.” [31].

2.6.1 Cold Metal Transfer Welding Cycle

A typical electrical cycle of CMT process can be defined as the time required to deposit
a droplet of molten metal on the substrate. Current and voltage waveform analysis is
important for the study of the energy distribution of different phases in droplet transfer
process. As shown on Figure 2.9, the process is divided into three phases:
(i) The peak current phase: High current with constant voltage causes the easy
ignition of the arc and then heats the wire to form a droplet.
(ii) The background current phase: The current is dropped to prevent the globular
transfer of the droplet produced on the tip of the wire. This phase continues till short
circuiting takes place.
(iii) The short-circuiting phase: The arc voltage is brought to zero in this phase.
At the same time, the wire feeder is given a return signal which gives the wire a back-
14
drawing force. This phase helps detachment of the liquid and its transfer onto the
substrate [29].

7
5 6

2
B
A

3 8

1. Power source
2. Remote-control unit
3. Cooling unit
4. Robot interface A
5. Wire feeder
6. Robacta Drive CMT
7. Wire buffer
8. Wire supply B

Figure 2.8 Fronius CMT welding system [30]

15
 Figure 2.9 CMT welding cycle [29, 31]

2.6.2 Cold Metal Transfer Welding Parameters

The most important variables of the CMT process which affect the AM quality are
welding current (wire feed speed), arc voltage (arc length), travel speed, electrode
extension, torch angle, and electrode diameter. Knowledge and control of these
variables are necessary to consistently produce appropriate quality product.
Nonetheless, modifying one variable usually involves adjusting additional parameters
to maintain an appropriate quality since the variables are not entirely independent of
each other [31].

2.6.2.1 Arc Length

With the CMT, the arc length is mechanically acquired and adjusted. That means the
arc remains stable irrespective of the surface condition of the workpiece itself. The
CMT uses a self-correction mechanism via a constant-potential power source in
addition to a mechanical response [31].

2.6.2.2 Wire Feed Speed

With the CMT, current and voltage are linked together via a linear mathematical
relationship (proprietary to Fronius International LLC) that integrates voltage and
16
amperage process controls into the feed speed of the wire and depends on the thermal
and electrical resistivity properties of the material substrate and filler used [31].

2.6.2.3 Welding Speed

Welding speed in the CMT process affects the weld quality deeply. The deposition
rate of filler metal per unit length increases when the welding speed is reduced. As the
welding speed increases, the thermal energy per unit length of weld transmitted from
the arc to base metal is decreased [31].

2.7 The Material; Super Duplex Stainless Steel

2.7.1 General

The microstructure of Duplex Stainless Steels (DSS) consists of ferrite and austenite
phases as shown in the Figure 2.10. DSSs are considered to be the preferred material
for many applications such as marine equipment, chemical, oil and gas industries
because they offer a unique combination of corrosion resistance properties and high
mechanical properties to be considered when selecting materials [32, 33, 34].

(a) (b)
Figure 2.10 Microstructure of duplex stainless steels: (a) wrought base material and
(b) weld metal [35]

A stainless steel's resistance to localized corrosion and mechanical properties is closely

related to its alloy content. Chromium (Cr), molybdenum (Mo), and nitrogen (N) are
the primary elements which contribute to the pitting corrosion resistance, strength, and
toughness. An empirical relationship called the Pitting Resistance Equivalent Number
(PREN) was developed to relate the composition of a stainless steel to its relative
17
pitting resistance in solutions containing chloride [36]. For austenitic and duplex
stainless steels, the PREN is calculated as follows:
PREN = Cr + 3.3(Mo + 0.5W) + 16N (2.1)
Where alloying elements are in weight percentages.

2.7.2 Brief History

Development of stainless steels in UK and Germany dates back to the first decades of
the 20th century. First grades of stainless steel have only been rich in Cr. After the
inventions of martensitic and ferritic grades, austenitic grades were developed by Ni
alloying and became the most popular stainless steel grades [37].
In stainless steels, there are three main types of microstructures, i.e., ferritic, austenitic,
and martensitic. These microstructures could be obtained by adjusting the steel
chemistry appropriately. Stainless steels can be classified into several different groups
of these three major microstructures which are ferritic stainless steels, austenitic
stainless steels, martensitic stainless steels, duplex stainless steels, and precipitation
hardening stainless steels [38]. Figure 2.11 includes the most popular diagram
developed experimentally by Schaeffler for weld metal, which allows the
identification of different categories of stainless steels on the basis of chemical
composition [38].
Nickel equivalent = Ni % + 30 C % + 0.5 Mn %

32

28
AUSTENITE
24
DUPLEX
20
A+M
16
2507
12 MARTENSIT 2205
A+M+F 2304

8 F+M FERRITE
4
4 8 12 16 20 24 28 32
36 Chromium equivalent = Cr % + Mo % + 1.5 Si % + 0.5 Nb %

Figure 2.11 Schaeffler diagram [39]

18
2.7.3 Duplex Stainless Steels Grades; Past and Present

Duplex stainless steel is a general term for stainless steels with ferrite and austenite
phases having a ratio of approximately 1:1. Various improvements have been made to
duplex stainless steel in the course of its history, and it has reached the present day [7].
The material can be made available in as-cast and wrought forms. Both grades are
increasingly employed for a wide range of applications, such as heat exchangers,
pressure vessels, storage tanks, desalination systems, piping assemblies in off-shore
platforms, and reaction vessels as shown in Figure 2.12 [37].

(a) (b)

(c)

Figure 2.12 Some applications where DSSs are employed; (a) shell-tube-heat-
exchanger, (b) pipe line, and (c) food processing.

19
The development of DSSs could be divided to four generations:

2.7.3.1 First Generation from Dawn -Development of Duplex Stainless Steels

Duplex stainless steel was initially used as a casting in the 1930s. At the beginning, it
was difficult to process sheet and tube products. These first-generation duplex stainless
steels have sufficient mechanical properties and corrosion resistance for practical use.
However, these steels were not sufficient for use in welded structures. The reason for
this is that the welded portion has the disadvantage that the mechanical properties and
corrosion resistance are greatly reduced as compared to the base metal because of
excessively big coarsened ferritic grains in the heat affected zone (HAZ) [7, 34, 38].

2.7.3.2 Second Generation-Ensuring Weldability

With the 70’s, new refining processes with VOD (Vacuum Oxygen Decarburization)
and AOD (Argon Oxygen Decarburization) in addition to continuous casting allowed
the production of much cleaner steels with a very low carbon content (0.03% or less)
and a high nitrogen content (0.1% or more) [34, 38]. These second-generation duplex
stainless steels like UNS S32205 have been improved in weldability. The above-
mentioned problems with HAZ were greatly improved by controlling of N and C
contents [7].

2.7.3.3 Third Generation-High Corrosion Resistance

In the 1990s, the need for even more corrosion-resistant duplex stainless steels
increased, and third-generation super duplex stainless steels were developed with a
high value of pitting corrosion index PREN and are being used in marine oil well-
related facilities. By convention, stainless steels with a PREN value of more than 40
are considered super stainless steels, so duplexes in this category are called super
duplex stainless steels [7].

2.7.3.4 At Present-Resource Saving and Ultra High Corrosion Resistance

In recent years, the direction of development of duplex stainless steel has largely been
focused on higher corrosion resistance. With a pitting resistance of 49, the SAF 2707
hyper-duplex stainless steel has significantly improved corrosion resistance in chloride
solutions compared to super duplex stainless steels [7, 40]. Table 2.1 lists the chemical
compositions and PREN values of the modern wrought duplex stainless steels. [36]

20
 Table 2.1 Chemical compositions and PREN values of the modern duplex stainless
steels [36]

USN Grade C Cr Ni Mo N Si Mn PREN

Early grade
S32900 329 0.08 23-28 2-5 1-2 -- 0.7 1 32
Lean duplex
S32304 2304 0.03 21 – 24 3 – 5.5 0.05-0.6 0.05-0.2 1 2.5 28
Standard duplex
S31803 2205 0.03 21-23 4.5-6.5 2.5-3.5 0.08-0.2 1 2 34
Super duplex
S32750 2507 0.03 24-26 6-8 3-5 0.2-0.3 0.8 1.2 42.5
Hyper duplex
S32707 2707 0.03 26-29 5.5-9.5 4-5 0.3-0.5 0.5 1.5 49.2

2.7.4 Super Duplex Stainless Steel Grades

In the 70s, a famous super duplex grade, which is known as Ferralium 255 (UNS
S32550) with 25 % Cr and high Mo additions, was developed by Langley alloy. The
grade had nitrogen additions but well below today's levels; it remained difficult to
weld. The new generation of super duplex grades was produced in the 90s, with 25-26
% Cr and increased Mo and N contents. These grades include 2507 and Zeron 100.
Additions of nitrogen in duplex stainless steels seemed extremely significant. It
contributes to many properties particularly in the welded areas, including the high
temperature stability of the two-phase microstructure [34]. In addition, nitrogen is also
considered to strengthen the localized corrosion resistance (pitting and crevice)
properties of the stainless steels. Such two main contributions of nitrogen explain why
nitrogen content in duplex grades has been gradually increased to the levels close to
the solubility limit value as shown in Figure 2.13 [41]. The solubility limit of nitrogen
steadily increases with increase of chromium and molybdenum contents and hence
improve the corrosion behavior of the SDSS [34].

21
 N, wt.%
0.40 Nitrogen
solubility limit Hyper DSS

Super DSS
0.25
Standard DSS
0.20 Lean DSS

Stability of austenite Stability of ferrite

not guaranteed difficult to achieve

20 35 Cr + Mo, wt. %

Figure 2.13 The solubility limit of nitrogen in DSS [41]

2.8 Physical Metallurgy and Microstructure of SDSS

2.8.1 Structure Stability and Solidification Mode

The stainless steel solidification mode is very important as it affects the corrosion and
mechanical properties of welded and cast components. The volume fraction of
austenite and ferrite are influenced by the solidification mode [42]. The ternary phase
diagram of iron-chromium-nickel system is a roadmap of the duplex stainless steels’
metallurgical behaviour. A 68% iron section through the ternary diagram (Figure 2.14)
shows that these alloys solidify as ferrite (α), which then partially transforms into
austenite (γ) as temperature drops, depending on the composition of the alloy.
Austenite / ferrite volume fractions are specifically related to composition and
temperature. As a result, the properties and microstructure at room temperature are
closely related to the solution annealing temperature i.e. higher solution annealing
temperatures produces higher ferrite contents [34]. A microstructure of approximately
50:50 percent of ferrite and austenite can be achieved at room temperature by water
quenching from the appropriate solution annealing temperature [36]. Nitrogen is
typically added as an alloying element to stabilize and accelerate the formation of the
austenite phase [43, 44].

22
 1800
α - Solidification
Possible primary
1600 carbides, nitrides

1400

transformation
1200 α  γ1
Temperature (°C)

1000

800
Carbides, nitrides
α  σ+γ3
α  χ  σ+γ3
600

α'
400 (475Co embrittlement)
ε- phase

200
γ  martensite
0
30 20 10 Cr %
0 10 20 Ni%

Figure 2.14 Pseudo-phase binary diagram of duplex grades (2205/2507) [44]

23
2.8.2 The Role of Alloying Elements in Duplex Stainless Steels

The interactions between the main alloying elements, especially chromium,

molybdenum, nitrogen, and nickel, are very complex. To achieve a stable duplex
structure that can be well fabricated, care must be taken to ensure that each of these
elements is at the correct level. In addition to the phase balance, a second major
concern is the formation of harmful intermetallic phases [36].
Chromium and molybdenum are ferrite stabilizers as they promote the body centered
cubic (BCC) crystal structure of iron and they are used to increase the strength,
corrosion resistance, hardenability and wear resistance of DSS. Whereas nickel and
nitrogen are austenite stabilizers as they promote the transformation of crystal structure
from BCC to face centered cubic (FCC) structure which increases the precipitation
mechanism of austenite phases and these two alloying elements increase the pitting
corrosion resistance, impact toughness and the tensile properties of DSS. To develop
the duplex structure, the ferrite formers, chromium and molybdenum should be
balanced by the austenite formers, nickel and nitrogen [45].

2.8.3 Desired Microstructure

The microstructure of super duplex stainless steels is a ferritic-austenitic structure, and

hot-rolled material has elongated grains in the rolling direction (Figure 2.10). The
ferrite / austenite ratio will impact the DSS properties. Report 938-C of the American
Petroleum Industry (API) states that the ferrite content in DSS is to be within the range
of 30-65% . However, the best properties are obtained when the ferrite content is about
50% [35, 46].

2.8.4 Precipitation of Secondary Phases

Slow cooling in the critical temperature ranges can result in the formation of various
types of intermetallic phases that are detrimental to the mechanical and corrosion
properties of SDSS [47]. The amount and the type of alloying elements have a direct
influence on the precipitation actions of different phases, as shown in Figure 2.15 [36].
By increasing the Cr and Mo concentrations, the Time-Temperature Transformation
(TTT) curves of the precipitation reactions can be shortened, thus increasing the
stability area of the sigma phase [47]. Different types of precipitates, which can form
24
in SDSS after thermal aging, are introduced taking into account two temperature
ranges:
1- High Temperature Range (550-1000°C) Precipitations
2- Low Temperature Range (250-550°C) Precipitations

Mo W Si
1000°C
Carbides
Cr Nitrides
Mo σ- phase
W χ- phase
γ2- phase
Si
R- phase

Cr π- phase
Mo ε- phase
W α- phase
Cu G- phase
300°C
Cr Mo W Cu

Time

Figure 2.15 Time-temperature precipitation diagram of duplex stainless steels [35]

2.8.4.1 High Temperature Range (550-1000°C) Precipitations

Sigma (σ) phase, which is a Fe-Cr-Mo rich phase with a tetragonal structure is the
most well-known deleterious phase precipitating in DSS from the ferrite at
temperatures below austenite formation between (600-1000°C) [35]. In addition,
because of the eutectoidal formation of the sigma phase, secondary austenite (γ2) is
present at the ferrite-austenite boundaries. Sigma phase has a powerful embrittlement
effect that decreases toughness. Secondary austenite is normally chromium-depleted
compared to the primary form, resulting in a localized reduction in resistance to the
pitting corrosion [10].

SDSS can precipitate chi (χ) phase, R phase, and chromium nitrides, in addition to
sigma phase and secondary austenite. The R phase is an intermetallic compound rich
in molybdenum and usually forms between 550-700°C but its practical significance is

25
lower since its nucleation takes longer [47]. Chi (χ) is a cubic intermetallic phase rich
of Mo, precipitating between 700°C and 900°C. It precipitates in smaller amounts
compared to σ. Chi (χ) nucleation is rapid due to its cubic structure, whereas its growth
rate is slow. Chi (χ) is usually considered to be an intermediate phase, replaced at
longer aging periods by the sigma phase [35].

Chromium nitride is formed in the temperature range of 700 - 900°C followed by rapid
cooling. At this temperature range, nitrogen does not have enough time to diffuse in
austenite and has limited amount of austenite formation. As a result, when ferrite
formations are supersaturated with nitrogen, chromium nitrides form at the grain
boundaries [47]. The chromium nitride precipitation reduces the pitting corrosion
resistance due to loss of chromium and nitrogen [38].

2.8.4.2 Low Temperature Range (250-550°C) Precipitations

The most popular transformation that occurs below 500°C is the spinodal
decomposition of the ferrite into chromium-rich ferrite (α’) and iron-rich ferrite (α) on
a very small scale, which also known as the 475°C embrittlement. A subsequent
hardening and embrittlement of the ferrite is observed. This explains why most
applications are confined to temperatures below 250°C. SDSS grades with 25% Cr are
very sensitive to this phenomenon [34, 35].

2.8.4.3 Microstructure and Mechanical Properties of DSS

Mechanical properties of SDSSs are described as a mixture of austenite and ferrite

properties. Combination of the toughness of austenite and strength of ferrite make DSS
a desirable material, which can be used in a wide variety of applications. Interestingly,
due to their finer microstructure, their strength is higher than both austenitic and ferritic
stainless steels. The excellent mechanical properties of SDSSs are restricted to
temperatures lower than 250°C due to the danger of 475°C embrittlement as discussed
previously [34, 35, 36].

2.9 Metallurgy of Additively Manufactured SDSS

The superior mechanical properties and pitting corrosion resistance of SDSS are due
to the balanced microstructure consisting of austenite and ferrite phases. This

26
microstructure is significantly influenced and altered by heating, and it has been
noticed to suffer the loss of corrosion resistance and reduced impact toughness [48].
WAAM is a layer-by-layer additive manufacturing process based on welding
technology. In this process the fresh runs reheat the underlying deposited beads.
Reheated beads have a distinct microstructure compared to single pass welds.
Common observations are the possible precipitation of secondary austenite and
deleterious secondary phases [35]. Maintaining a proper austenite / ferrite phase
balance while avoiding brittle phase formation is necessary to maintain acceptable
toughness and corrosion resistance of the beads [48].

2.10 The Property: Fracture Toughness

Although metallic materials are characterized, with some exceptions, by a significant

ductility, metallic structures and components may show, under particular conditions,
sudden unexpected failures, where plastic strains are limited or absent at all. This
phenomenon, which may occur even if stresses acting on structural elements are well
below material’s yield strength, and is generally correlated to the presence of planar
flaws and called brittle failure. Brittle failure phenomena may occur in structures
and/or components where safety and cost aspects are of particular significance. Repair
operations may be critical or impossible. These considerations underline the
importance of defining assessment procedures, in order to check if materials are fit for
their intended use in design conditions or a detected flaw in service can be considered
acceptable or not [49].

2.11 Fracture Mechanics

The aim of fracture mechanics is to establish the limit conditions beyond which the
propagation of an unstable flaw occurs (i.e. crack driving forces determining the flaw
propagation overtake material’s resistance to fracture, see Figure 2.16) [49].
According to fracture behavior of materials, two main branches of fracture mechanics
are taken into consideration: Linear Elastic Fracture Mechanics (LEFM) for brittle
materials and Elastic-Plastic Fracture Mechanics (EPFM) for ductile materials [50,
51].

27
 Applied
Materials
stresses

Environmental Flaw
Crack
conditions: Material dimension
driving s
-Temperature resistance to
force
-Hydrogen fracture Component
geometry

Applied strain
rate

Figure 2.16 Conditions of flaw propagation [49]

2.12 Linear Elastic Fracture Mechanics

The basic work for linear elastic fracture mechanics is by Griffith, where he defined a
critical stress for failure. As seen in Figure 2.17 a plate with an initial crack length (2a)
subjected to tensile stress σ, was used to describe the flaw propagation. At the tip of
the crack, the stress is greater than the average stress of σ. Because of this high stress,
the material near the tip of the crack will experience large strains and ultimately
collapse, allowing the crack to spread ahead [50, 52, 53].

Figure 2.17 Plate with the initial crack [49]

28
The critical stress σc required for crack propagation in a brittle material is described
by the expression:
2𝛾𝑒 𝐸
𝜎𝑐 = √ (2.2)
𝜋𝑎

Where:
E is modulus of elasticity
γe is specific surface energy
a is one half of the length of an internal crack [54]

The modern theory of fracture mechanics started with the analyses carried out by
Irwin, where the stress and strain distributions occurring at the tip of a two-dimensional
crack can be defined by means of Stress Intensity Factor “K”.

As shown in Figure 2.18, crack propagation may occur according to three different
modes, namely the opening mode (mode I), the in-plane shearing mode (mode II), and
the out-of-plane shearing mode (mode III) [54]. From an engineering point of view,
Mode I is the most important mode as the propagation of crack-like flaws tends to
occur perpendicular to the maximum tensile stress direction.

Mode I Mode II Mode III

Opening mode Sliding mode Tearing mode
(in-plane shear) (out-of-plane shear)

Figure 2.18 The three modes of loading [54]

29
The mode I stress intensity factor (KI) can be represented by the following general
equation:
𝐾𝐼 = 𝑌𝜎√𝜋𝑎 (2.3)

The parameter Y depends on:

-Structural component geometry,
-Flaw geometry,
-Type of stress distribution
σ represents the nominal value characterising the stress distribution;
a represents the characteristic flaw dimension [49].

Under an increasing load crack propagation/initiation may initiate when K I reaches a

special critical value KIC.
KI = KIC
KIC is a real material characteristic called fracture toughness [55].

2.13 Elastic Plastic Fracture Mechanics

Linear elastic fracture mechanics (LEFM) lose its legitimacy for ductile materials
because the fracture processes at the crack tip are followed by severe plastic
deformation in these materials [54]. Therefore, the characterization of the fracture
behavior of ductile materials requires Elastic Plastic Fracture Mechanics (EPFM) [51].
The most important concepts in EPFM are crack tip opening displacement (CTOD)
and J-integral proposed by Wells (1961) and Rice (1968), respectively [54]. The
mechanical behavior near the crack-tip of ductile materials can be defined by both
concepts (J and CTOD). Each of them can serve as a measure of the toughness of the
material and can therefore be used to determine criteria for fracture [50, 51].

2.13.1 Crack Tip Opening Displacement (CTOD)

For elastic-plastic materials (ductile materials), the crack tip opening displacement
(CTOD) is a fracture mechanics parameter describing crack-tip conditions. It is
observed that for materials with high toughness, the high degree of plastic deformation
give rise to blunt the crack tip before subsequent ductile tearing as shown in
Figure 2.19. The CTOD (δ) value can be connected to KI as follows:

30
 𝐾𝐼2
𝛿= (2.4)
𝑚𝜎𝑌𝑆 𝐸 ′

where
m is a dimensionless constant that is approximately 1.0 for plane stress and 2.0 for
plane strain.
E′ is the effective Young’s modulus, defined as:
E′=E for plane stress and E′=E/(1-v2) for plane strain
σys is the 0.2% offset yield stress [51, 54].

Deformed crack flank

Ductile tearing

CTOD (δ )

Blunt crack tip

Original crack tip

Figure 2.19 Displacement at the original crack tip [54]

2.13.2 J- Integral

Plastic deformation takes place in front of the crack tip for common structural steels
including DSS with a relatively high toughness. The J-integral can be determined for
these materials, which describes the rate of energy release of non-linear elastic
materials. This toughness determination method is based on the amount of energy
required to initiate and propagate a crack [56]. J reflects the energy "spent" for crack
propagation per unit fracture surface under the most general conditions, both when the
material behavior is primarily linear elastic and when the size of the plastic zone is
large. The following equation is valid for materials with elastic behavior:

31
 𝐾2
𝐽=𝐺= (2.5)
𝐸′

G=Crack extension force

E′=E for plane stress
E′=E/(1-v2) for plane strain [49]

2.14 Crack-Growth Resistance Curves

Many high toughness materials do not show catastrophic failure at a particular value
of J or CTOD. Instead, these materials display an increasing R curve, where J and
CTOD rise with the progress of cracks. An increasing of R curve is typically correlated
with micro voids coalescence. Figure 2.20 demonstrates a standard J resistance curve
for a small amount of visible crack growth due to blunting. When J increases, the
material at the crack tip fails locally and the crack progresses further. Due to the rising
of R curve, the initial crack growth is usually stable but instability may later be
encountered [54].

JR

Stable tearing
JC Fracture initiation

Blunting crack

Crack Extension (Δa)

Figure 2.20 Crack-growth resistance, R- curve [54]

2.15 Methods of Measuring and Types of Specimen

For fracture toughness testing of a material a lot of specimens are used to analyse
characteristic values [55]. Single edge notched bend (SENB) (Figure 2.21-a) and
compact tension (CT) (Figure 2.21-b) specimens with a high crack depth are the most
32
frequently used types of specimens [57], and their use is defined in international
standards such as ASTM E1820 and BS-7448 [58]. These specimens are subjected to
a bending load that results in a high crack-tip constraint condition, which is
unrepresentative of constraint at the crack tip in conditions like a surface crack in a
pipe [58]. In recent years, the clamped single edge notched tension (SENT) specimens
(Figure 2.21-c) have been commonly used [57]. Generally, the lower constraint of the
SENT specimen, compared to the SENB specimen, results in higher values of fracture
toughness [59, 60].

F
F

B
W
B

4W

(a) H
a
B F

W
1.2W

1.25W
F
F
(b) (c)
Figure 2.21 Standardized fracture mechanics test specimens: (a) single edge-notched
bend, (b) compact specimen, (c) single edge notched tension. [61, 62]
33
Fracture toughness testing with SENT specimens is now standardized in BS 8571 and
SENT specimens can be used for higher and lower temperature testing, different
material thicknesses and wide range of engineering applications. SENT test results can
be considered for use with fitness-for-service evaluation provided that the structural
specification to be tested has lower constraints than the specimen [59].

2.16 Fracture of Duplex Stainless Steels

The room temperature fracture behaviour of duplex stainless steels is well established.
Compared to other stainless steels the fracture toughness of DSS is very high and this
high toughness could be ascribed to the presence of austenite [10, 63]. The ductile
austenite retards the cleavage fracturing of ferrite [63, 64]. The grain size together with
crystallographic texture of SDSS can also influence the behavior of the fracture. It has
been shown that there is a strong rolling texture in specimens which results in highly
anisotropic resistance properties [10]. The fracture toughness values are higher in the
transverse than in the longitudinal direction with respect to the rolling direction [63].
When a DSS part is additively manufactured using arc welding, the fracture toughness
of welded layers and their heat affected zones are generally much lower compared to
that of the rolled plate.. Deposited metals have a fine grain size because of their rapid
cooling rate. In addition, there can be partial dissolution of austenite grains in the heat
affected zone, as well as general matrix ageing. Ductility is therefore limited, and the
work hardening rate is much higher than that of the rolled plate.. During additive
manufacturing, reheating of welded layers can also nucleate brittle intermetallic phases
within the matrix that further promote cleavage fracture. The fracture behavior of
welded layers is thus often the critical factor in a product's performance [10].
At sub-zero temperatures, the fracture mechanism for ferrite shifts from ductile to
brittle fracture. This phenomenon called the ductile-to-brittle transition. The cause for
this change in fracture behavior for the ferrite is the restricted slip and dislocation/twin
interactions. The ductile-to-brittle transition temperature (DBTT) of SDSS depends on
the chemical content, microstructure, existence of detrimental phases and
precipitations [10, 64, 65]. In conclusion, although the fracture toughness of duplex
stainless steels is generally well studied, fracture behavior of additive manufactured
duplex stainless steels (especially at low temperatures) needs further studies to be
conducted [10].

34
 k CHAPTER 3

3. NUMERICAL SIMULATION STUDY

3.1 General

Wire arc additive manufacturing is a welding based process and can be abbreviated it
as (WAAM). The process is commonly used in industrial applications due to its high
efficiency, durability and low cost compared to some other processes. One of the
challenges faced by engineers and technicians in this process is the selection of the
best welding parameters to achieve high quality at low cost. By strategic planning of
experiments design, effective statistical models can be developed to solve these
significant problems. In this work, response surface method has been used to build
statistical models to study the effect of WAAM process variables (welding voltage,
current and welding speed) on longitudinal and transversal distortion of product. The
statistical models would be helpful in setting process parameters at optimum values to
achieve a low displacement resulted from longitudinal and transversal distortion. The
plan of study is shown in Figure 3.1. Instead of doing real WAAM process, which
leads to additional costs caused by losing expensive material, numerical simulation
was used to simulate the WAAM process and predict the longitudinal and transversal
distortion. Design-Expert version 7.1.3 software [66] was used to obtain the statistical
models and optimization. For the simulation, a commercial software package,
COMSOL Multiphysics 5.5 [67] has been used to perform the thermal and mechanical
tests.

Figure 3.1. The plan of study

35
3.2 Wire Arc Additive Manufacturing Process Variables

According to the previous studies that covered CMT welding based AM of SDSS, the
upper and lower values of welding parameters are decided as given in Table 3.1. [15,
17, 18, 19]

Table 3.1 Control variables

Lower Upper
Current I (A) 140 220
Voltage U (V) 15 25
Speed S (mm/s) 4 10

The response surface design constructed from three sets of points, first set based on
factorial design which consist of 2k experiments and coded as (±1, ±1, ±1). Second
set, center points, coded as (0, 0, 0). Third set, star points, consist of 2k experiments
and located at a specified distance α from the center point and coded as (±α,0, 0), (0,
±α, , 0 ), (0, 0, ±α). The value of α can be calculated by the equation (3.1). [68]
α = 2k/4 (3.1)
k = Number of process parameters.
For k=3  α = ± 1.682
All sets are shown diagrammatically in Figure 3.2 for k = 3

Figure 3.2 Response surface design for k=3 variables

36
Five levels for each WAAM process variables have been defined based on the response
surface method. The five levels were coded as (-α , -1 , 0 , 1 , α). The original values
(X) of the variable at α can be calculated by the equation (3.2). [68]
X = (((±α+1)/2) × (XU – XL)) + XL (3.2)
Where:
X : the original value of the variable at α.
XU : the upper variable value.
XL : the lower variable value.
The selected WAAM process variables with their limits and codes are given in
Table 3.2

Table 3.2 WAAM process variables with their limits

Parameter Units Notation Factor levels

Coded value -1.682 -1 0 1 1.682

Current (A) I 112.72 140 180 220 247.28

voltage (V) U 11.6 15 20 25 28.41

Speed (mm/s) S 2 4 7 10 12

3.3 Response Surface Design Matrix

In factorial design, experiments are carried out on all possible combinations of

parameter levels, and these combinations form a design matrix when written in the
form of a table. The total number of the experimental runs (N), based on full factorial
central rotatable design, is given by the equation (3.3): [69]

N=2k+2k+no (3.3)

Where no refers to the number of the center point.

In the present work, k = 3 which leads to α = 1.682 (formula 3.1). There are 8 corner
points, 6 star points and 6 runs for the center point, yielding a total number of twenty
experimental runs, N = 20 (formula 3.3). The selected design matrix is shown in
Table 3.3 .
37
 Table 3.3 Established design matrix

Experiment Current Voltage Speed

No I (A) U (V) S (mm/s)
1 140 15 4
2 220 15 4
3 140 25 4
4 220 25 4
5 140 15 10
6 220 15 10
7 140 25 10
8 220 25 10
9 112.72 20 7
10 247.28 20 7
11 180 11.6 7
12 180 28.41 7
13 180 20 2
14 180 20 12
15 180 20 7
16 180 20 7
17 180 20 7
18 180 20 7
19 180 20 7
20 180 20 7

3.4 Numerical Modelling and Simulation of WAAM Process

In particular, the selection of the optimum process parameters in this study is based on
the evolution of the longitudinal and transversal distortion simulating the welding of
super duplex stainless steel Grade (2507), the process simulations are designed in the
form of single layer and 3 runs using COMSOL Multiphysics software. The process

38
parameters are selected according design matrix given in Table 3.3 and material
properties are given in Table 3.4. [36]

Table 3.4 Super duplex stainless steel properties [36]

Property Value Unit

Thermal conductivity 20 W/(m.K)
Heat capacity at constant pressure 487 J/(kg.K)
Poisson's ratio 0.3 --
Young's modulus 200 ×109 Pa
Density 7750 kg/m3
Coefficient of thermal expansion 14 ×10-6 1/K

3.5 Numerical Model Description

Figure 3.3 shows the schematic diagram and geometrical size of the computational
domain considered. Points T1, T2 and T3 were selected to estimate the temperature.
The distance between each point was 2 mm. Points D1 and D2, with a distance of 50
mm in between, were selected to estimate the longitudinal and transversal distortion.

D1

D2

10 mm
T1
T3
T2

Figure 3.3 Schematic diagram and geometrical size of the computational domain

3.6 Numerical Model Assumptions

 Heat flux is deemed at top surface.

 The arc heat flux is defined by the Gaussian distribution.
 The established model does not consider the molten pool's flow but takes into
account the simple transfer of heat.
39
3.7 Governing Equations and Boundary Conditions

For the accuracy of the finite element results, the accurate modeling of the heat source
is of most significance since it determines the heat input from the welding process and
the resultant distortions. A Gaussian distribution can approximate the heat source
energy, and the heat flux (q) on the top surface is given by the following expression
[2]:

2𝑄 2𝑅 2
𝑞 = 𝜋𝑅2 𝑒𝑥𝑝 (− ) (3.4)
𝑠 𝑅𝑠2

𝑄 = 𝜂𝑉𝐼 (3.5)

where Q is the arc power of the heat source, Rs is radius of heat source, R is distance
from local point and η = 0.8.
Equations (3.6) and (3.7) [2], were used to calculate the boundary conditions of cooling
between the workpiece and the environment through convection and radiation,
respectively.

qc = h(T −T0 ) (3.6)

qr =ε σ (T4 −T04) (3.7)

where ε = 0.6 is the emissivity of SDSS and h (10 W/m2.K ) is the natural convective
heat transfer coefficient. [70, 71]

3.8 Model Validation

To demonstrate the efficacy of the numerical model, an experimental validation is

provided by comparing the actual temperature of a WAAM with the simulated ones.
The model was validated using data obtained from two published work and data
obtained from laboratory works as described below:

40
Case 1
Vahid et al. [18] studied the wire arc additive manufacturing of a duplex stainless steel
in terms of thermal cycle analysis and microstructure characterization. A voltage of
24.5 V, a current of 147 A, and a welding speed of 6.5 mm/s were used. The
temperature was 1043oC at 3.75 mm away from the fusion line of the first pass.

Case 2
Carlos et al. [72] studied the welding effects on the duplex stainless steel
microstructure by carried bead-on-plate welding. A voltage of 25 V, a current of 180
A, and a welding speed of 6 mm/s were used. The temperature was 1250oC at 4 mm
away from the fusion line of the first pass.

Case 3
Laboratory work has been done by carried bead-on-plate using a voltage of 20 V, a
current of 250 A, and a welding speed of 7 mm/s. An indication of the temperature of
the deposited metal was obtained by measuring the temperature of the center of the
weld pool by inserting a thermocouple into the weld pool behind the arc during
depositing. Type K-chromel thermocouples, which are the most common sensor
calibration type, are Ni based and exhibit good corrosion resistance. Their melting
point is 1370oC and they are reliable and accurate at temperatures up to 1260oC. During
the measurements, the thermocouple material was protected with a ceramic protection
tube. Cooling curves obtained by thermocouple reading and temperature data
acquisition during the welding processes are represented in Figure 3.4.

41
 1400

1200

1000

Temprature (oC)
800

600

400

200

0
0 5 10 15 20 25 30
Time (sec)

Figure 3.4 Cooling curves recorded in the middle of weld pools

The simulation was executed using the process parameters of case 1, case 2 and case
3 at point T3, 4 mm from the fusion line.
Figure 3.5, Figure 3.6 and Figure 3.7 illustrate the result of simulation for case 1, case
2 and case 3 respectively.

Figure 3.5 Heat distribution of case 1

42
Figure 3.6 Heat distribution of case 2

Figure 3.7 Heat distribution of case 3

43
The results obtained from simulation were recorded and compared with the actual
results of experiments as shown in Table 3.5.

Table 3.5 Actual and simulation results of temperature

Temperature (oC)
Case No. Error
Experiment Simulation
Case 1 1043 1128 8.2 %
Case 2 1250 1385 10.8 %
Case 3 1188 1277 7.5 %

When the temperatures obtained from the simulations are compared to the results of
actual experimental temperature measurements; the relative percentage errors are up
to 8.2 % for case1, and 10.8 % for case 2, and finally 7.5% for case 3. Considering the
heat transfer assumptions in the model, and considering the accuracy of thermocouple
readings and the delay in thermocouple response times, these differences are small and
expected so that it is concluded that the simulation of the model using COMSOL
Multiphysics holds.

3.9 Simulation Results

The results obtained from simulation were recorded and surmised in figures from
Figure 3.8 to Figure 3.22. Since the values of distortion and temperature were equal
for the last 6 experiments only one of them is presented. The heat distributions were
presented in part (a) of the figures while the displacements resulting from longitudinal
and transversal distortion were shown in part (b) of the figures for the 15 experiments
in design matrix, recorded in Table 3.3. The results displayed in (3D) for experiments
were shown in part (c) of the figures.

44
 (a) (b)

(c)
Figure 3.8 Simulation result of experiment (1). a) Thermal cycle, b) Displacements,
c) Total displacement results displayed by FEM simulations

45
 (a) (b)

(c)
Figure 3.9 Simulation result of experiment (2). (a) Thermal cycle (b) Displacements
(c) Total displacement results displayed by FEM simulations

46
 (a) (b)

(c)
Figure 3.10 Simulation result of experiment (3). (a) Thermal cycle (b) Displacements
(c) Total displacement results displayed by FEM simulations

47
 (a) (b)

(c)
Figure 3.11 Simulation result of experiment (4). (a) Thermal cycle (b) Displacements
(c) Total displacement results displayed by FEM simulations

48
 (a) (b)

(c)
Figure 3.12 Simulation result of experiment (5). (a) Thermal cycle (b) Displacements
(c) Total displacement results displayed by FEM simulations

49
 (a) (b)

(c)
Figure 3.13 Simulation result of experiment (6). (a) Thermal cycle (b) Displacements
(c) Total displacement results displayed by FEM simulations

50
 (a) (b)

(c)
Figure 3.14 Simulation result of experiment (7). (a) Thermal cycle (b) Displacements
(c) Total displacement results displayed by FEM simulations

51
 (a) (b)

(c)
Figure 3.15 Simulation result of experiment (8). (a) Thermal cycle (b) Displacements
(c) Total displacement results displayed by FEM simulations

52
 (a) (b)

(c)
Figure 3.16 Simulation result of experiment (9). (a) Thermal cycle (b) Displacements
(c) Total displacement results displayed by FEM simulations
53
 (a) (b)

(c)
Figure 3.17 Simulation result of experiment (10). (a) Thermal cycle (b)
Displacements (c) Total displacement results displayed by FEM simulations
54
 (a) (b)

(c)
Figure 3.18 Simulation result of experiment (11). (a) Thermal cycle (b)
Displacements (c) Total displacement results displayed by FEM simulations
55
 (a) (b)

(c)
Figure 3.19 Simulation result of experiment (12). (a) Thermal cycle (b)
Displacements (c) Total displacement results displayed by FEM simulations

56
 (a) (b)

(c)
Figure 3.20 Simulation result of experiment (13). (a) Thermal cycle (b)
Displacements (c) Total displacement results displayed by FEM simulations

57
 (a) (b)

(c)
Figure 3.21 Simulation result of experiment (14). (a) Thermal cycle (b)
Displacements (c) Total displacement results displayed by FEM simulations

58
 (a) (b)

(c)
Figure 3.22 Simulation result of experiment (15). (a) Thermal cycle (b)
Displacements (c) Total displacement results displayed by FEM simulations

59
The values of temperature (T) and displacements resulting from longitudinal and
transversal distortions (D1 and D2) for each experiment in design matrix were listed
in Table 3.6.

Table 3.6 Response values

Exp. Current Voltage Speed D1 D2 T

No (A) (V) (mm/s) (mm) (mm) (°C)
1 140 15 4 1.41 0.88 2086
2 220 15 4 1.67 1 2975
3 140 25 4 1.7 1.02 3365
4 220 25 4 1.91 1.11 4100
5 140 15 10 0.95 0.56 1171
6 220 15 10 1.27 0.74 1821
7 140 25 10 1.32 0.77 1922
8 220 25 10 1.65 0.93 2897
9 112.72 20 7 1.08 0.71 1389
10 247.28 20 7 1.56 0.97 2766
11 180 11.6 7 1.03 0.68 1289
12 180 28.41 7 1.58 0.98 2854
13 180 20 2 1.74 1.06 3786
14 180 20 12 1.23 0.71 1265
15 180 20 7 1.37 0.87 2085
16 180 20 7 1.37 0.87 2085
17 180 20 7 1.37 0.87 2085
18 180 20 7 1.37 0.87 2085
19 180 20 7 1.37 0.87 2085
20 180 20 7 1.37 0.87 2085

60
3.10 Creation of Statistical Models

The second-degree response surface function representing the longitudinal and

transversal distortion (D1 and D2) dimensions and temperature can be expressed as
follows [68]:
Y = b0 + b1I + b2U + b3S+ b11I2 + b22U2 + b33S2+ b12IU + b13IS+ b23US (3.8)
The coefficients values of statistical models were calculated by using Design-Expert
software [66] and the results are presented in Table 3.7.
Table 3.7 Coefficient values of models

Parameter Coefficient D1 D2 T
constant b0 1.37 0.87 2075.03
I b1 0.14 0.072 407.48
U b2 0.16 0.085 502.53
S b3 -0.17 -0.12 -655.7
2
I b11 0.007 -0.005 62.51
2
U b22 0.001 -0.009 60.39
2
S b33 0.065 0.011 220.91
IU b12 -0.005 -0.006 21.37
IS b13 .023 0.016 0.12
US b23 .028 0.019 -72.13

The statistical models developed are given below. The WAAM process control
variables are in their coded form:
D1 = 1.37 + 0.14I + 0.16U -0.17S+ 0.007I2 + 0.001U2 + 0.065S2-0.005IU +
0.023IS+ 0.028US (3.9)
D2 = 0.87 + 0.072I + 0.085U -0.12S-0.005I2 -0.009U2 + 0.011S2 -0.006IU +
0.016IS+ 0.019US (3.10)
T = 2075.03 + 407.48I + 502.53U-655.7S+ 62.51I2 + 60.39U2 + 220.91S2+
21.37IU + 0.12IS-72.13US (3.11)

The statistical models furnished above can be employed to predict the longitudinal and
transversal distortions and temperature for the range of parameters used in the
investigation by substituting their respective values in coded form. Based on these
models, the interaction effects of the process parameters on the distortion were

61
computed and plotted as response surface and contour surface and depicted in figures
from Figure 3.23 to Figure 3.28.

D1: Longitudinal distortion (mm)

(a)

D1: Longitudinal distortion (mm)

U: Voltage (V)

I: Current (A)

(b)
Figure 3.23 (a) Response surface and (b) contour surface of the interaction effect of
current and voltage on longitudinal distortion

62
 D1: Longitudinal distortion (mm)

(a)

D1: Longitudinal distortion (mm)

S: Speed (mm/s)

I: Current (A)

(b)
Figure 3.24 (a) Response surface and (b) contour surface of the interaction effect of
current and speed on longitudinal distortion

63
D1: Longitudinal distortion (mm)

(a)

D1: Longitudinal distortion (mm)

S: Speed (mm/s)

U: Voltage (V)

(b)
Figure 3.25 (a) Response surface and (b) contour surface of the interaction effect of
voltage and speed on longitudinal distortion

64
 D2: Transversal distortion (mm)

(a)

D2: Transversal distortion (mm)

U: Voltage (V)

I: Current (A)

(b)
Figure 3.26 (a) Response surface and (b) contour surface of the interaction effect of
current and voltage on transversal distortion

65
 D2: Transversal distortion (mm)

(a)

D2: Transversal distortion (mm)

S: Speed (mm/s)

I: Current (A)

(b)
Figure 3.27 (a) Response surface and (b) contour surface of the interaction effect of
current and speed on transversal distortion

66
 D2: Transversal distortion (mm)

(a)

D2: Transversal distortion (mm)

S: Speed (mm/s)

U: Voltage (V)

(b)
Figure 3.28 (a) Response surface and (b) contour surface of the interaction effect of
voltage and speed on transversal distortion

67
3.11 Checking the Adequacy of the Models

The validity of the models can be judged from the scatter plots shown in Figure 3.29,
which indicate that the predicted values of the longitudinal and transversal distortion
(D1 and D2) are in good agreement with the results obtained from the simulation.

Predicted
Predicted

Actual Actual

(a) (b)

Figure 3.29 Scatter plot of the actual and predicted values for (a) longitudinal and (b)
transversal distortion

3.12 Optimization

Optimization is searching for combination of factor levels that fulfill the requirements
imposed on each of the responses and factors simultaneously. The simultaneous
optimization of multiple responses in this work can be performed with help of Design-
Expert software by selecting the desired target for each process variable and response.
The possible targets are maximization or minimization of the responses. The
"importance" of each target can be changed in relation to the other targets from 1 to 5.
In this study, importance was chosen to be constant. Table 3.8 shows the suggested
criterion for numerical optimization and Table 3.9 shows the most desirable solution
selected by the Design-Expert software.

68
 Table 3.8 Criterion of numerical optimization

Name Goal Lower limit Upper limit Importance

Current (I) A In range 140 220 3
Voltage (U) Volt In range 15 25 3
Speed (S) mm/s In range 4 10 3
Temperature (T) °C In range 1500 2500 3
Longitudinal
Minimize 0.95 1.91 3
distortion (D1) mm
Transversal
Minimize 0.56 1.11 3
distortion (D2) mm

Table 3.9 Desirable solution selected

Name Desirable value

Current (I) A 200

Voltage (U) volt 15

Speed (S) mm/s 10

Temperature (T) °C 1500

Longitudinal distortion (D1) mm 1.17

Transversal distortion (D2) mm 0.70

3.13 Discussion

3.13.1 Interaction effects of welding current and voltage on distortion

As can be seen in Figure 3.23 and Figure 3.26, the longitudinal and transversal
distortions increase with increasing voltage and current because both have a positive
relationship with distortion. The longitudinal and transversal distortions are at their
maximum values of 1.68 mm and 1.01 mm, respectively when the welding current and
voltage are at their maximum values.

3.13.2 Interaction effects of welding current and welding speed on distortion

The interaction effect of current and welding speed on distortion is shown in
Figure 3.24 and Figure 3.27, the current has a positive relationship with distortion,
while speed has a negative relationship. As result, the value of distortion increases with

69
the increase in current for all values of speed. However, the rate of increase for
distortion gradually decreases as speed increases from its lower limit to upper limit.
The longitudinal and transversal distortions are at their maximum values of 1.74 mm
and 1.06 mm, respectively when the welding current is at its maximum value and speed
is at its minimum value.

3.13.3 Interaction effects of welding voltage and welding speed on distortion

The distortion increases as voltage increases for all values of speed. However, the rate
of increase for distortion gradually decreases as the speed increases as shown in
Figure 3.25 and Figure 3.28. The longitudinal and transversal distortions are at their
maximum values of 1.74 mm and 1.06 mm, respectively when the welding voltage is
at its maximum value and speed is at its minimum value.

70
 l CHAPTER 4

4. EXPERIMENTAL STUDIES

4.1 General

This chapter is devoted to the experimental studies and divided into six sections. In
section 4.2, the delivery conditions and chemical compositions of substrate and
additive filler metals are introduced. Section 4.3 deals with build-up of the wire arc
additive manufactured layers. Inspection procedures for integrity assessment of
additively manufactured beads were addressed in section 4.4. This section is followed
by sections 4.5 and 4.6, which give the descriptions of mechanical and
metallurgical characterization techniques, respectively. The last section 4.7 forms the
main focal issue of the thesis, as it concerns fracture toughness testing using single
edge notched tension (SENT) specimens.

4.2 Materials

Three different steels have been employed in this experimental work:

1- Super duplex stainless steel rolled plates grade 2507 with a thickness of 12 mm
were subjected to the investigation and conformed to ASTM A240 S32750 and
EN 10088-2 X2CrNiMoN 25-7-4 (1.4410) [73, 74].

2- Austenitic stainless steel grade ASTM A240 316L plates with the dimensions
of 400 mm × 200 mm × 10 mm were used as substrates to carry out the
experiments. This work was focused only on the AM layers and did not include
the interface layer; hence it was not mandatory to use SDSS plates as substrate.

3- Grade 2209 duplex stainless steel filler wire which was standardized as per
SFA/AWS 5.9 ER2209 and ISO 14343-A G 22 9 3 N L was used as the
feedstock material in the first experiments [75].

4- Grade 2509 super duplex stainless steel filler wire which was standardized as
per SFA/AWS 5.9 ER2594 and ISO 14343-A G 25 9 4 N L was used as the
feedstock material in the second experiments [75].
71
The chemical compositions of steels used are shown in Table 4.1.

Table 4.1 Chemical compositions of the steels used (wt%). [73, 75]

Material C Cr Ni Mo Mn Si N
Grade 2507 as received
0.02 24.7 6.6 3.6 0.8 0.32 0.28
Plate
Grade 2209 filler wire
0.01 22.7 8.5 3.2 1.5 0.5 0.17
for WAAM
Grade 2509 filler wire
0.01 25.2 9.4 3.9 0.4 0.4 0.24
for WAAM
Grade 316L Plate
0.03 16.8 10.7 2.1 2.0 0.75 0.1
( substrate for WAAM )

4.3 Wire Arc Additive Manufacturing Procedure

4.3.1 Equipment

The system used to perform additive manufacturing deposition is composed by the

integration of two different systems:
 Welding system. In this work, the semi-automatic Fronius TPS 4000 CMT
power source was used as the WAAM process. This WAAM method provides
relatively low-heat input with high deposition rate.
 Three-axis movement system, the motion of the system should be easily
controlled.
The power source used for the described experimental work in this section is shown in
the Figure 4.1.

4.3.2 Sample Build-up

Cold metal transfer welding technique was used to deposit all beads. A reversed
polarity (electrode positive) direct current (DCEP) between the wire electrode and the
substrate was used in all runs. Two blocks with the size of 400 mm × 200 mm × 30
72
mm were deposited using Grade 2209 duplex stainless steel filler metal and Grade
2509 super duplex stainless steel filler metal severally. The substrate was Grade 316L
austenitic stainless steel plate. The plate was cleaned to remove dirt, located at flat
position, aligned on the platform manually and clamped by the help of pneumatic
cylinders. After clamping, the beads are built in a zigzag direction until 13 layers of
material have been deposited using mechanized torch. Schematic WAAM setup could
be seen in Figure 4.2. The employed welding parameters which were obtained from
simulation studies in Chapter 3, were kept constant for all deposited layers as given in
the Table 4.2. To protect the weld pool from atmospheric contamination a mixture of
Ar + 2% O2 was used as shielding gas. The real sample which was welded for
investigation is shown in Figure 4.3.

Figure 4.1 Fronius TPS 4000 CMT power source

73
 CMT torch
Deposition direction

Deposit layers of Grade Grade 316L

2207 or Grade 2509 Substrate
filler wire

Figure 4.2 Schematic of WAAM setup

Table 4.2 Constant process parameters.

Parameter Setting

Welding process CMT

Welding position PA (1G)
Average current 200 A
Average voltage 15 V
Welding speed 10 mm/s
Shielding gas flow rate 18 L/min
Contact tube distance 12 mm

74
 30 mm

400 mm

Figure 4.3 Real WAAM’d sample for Grade 2509

4.4 Integrity Assessment of the Additively Manufactured Beads

Deposited beads performed by WAAM were visually inspected to check surface

imperfections. To examine possible volumetric discontinuities, X-ray radiographic
examination was applied according to ASTM E1742 [76] and films were evaluated as
per the requirements and definitions in ASME BPV Code Section IX [77]. The quality
level employed for radiography was 2 % (2-2T using hole type IQI).

75
4.5 Mechanical Characterization

4.5.1 Tensile Test

In order to determine mechanical properties (strength and ductility) of the super duplex
stainless steel, tensile tests were carried out with a universal tensile testing machine
(BESMAK BMT-E series) using round shape tensile samples which were prepared
according to ASTM E8 standard [78]. Figure 4.4-a shows the dimensions of the
samples. Five samples were cut from the received plate of SDSS Grade 2507 as shown
in Figure 4.4-b. Moreover, as depicted in Figure 4.4-c, two samples were sectioned
from the additively manufactured plate of SDSS Grade 2509.

R= 6 mm

D= 6 mm

10 mm

(a)

(b) (c)
Figure 4.4 Tensile test samples

76
4.5.2 Hardness Test

In order to investigate the effects of the thermal cycles on the hardness, Vickers micro
hardness measurements were taken along the deposited material of Grade 2509. The
indentations were made under the load of 0.5 kg over 10 s with a distance of 0.3 mm
between the indentations. Sections perpendicular to the deposition direction were cut,
prepared and examined in the etched condition. The specimens were prepared and
tested according to ISO 6507 standard [79].

4.6 Microstructural Examination

Because of critical effect of microstructure upon the fracture behavior of a material,

thorough analysis of the microstructure was required for the rolled plate grade 2507
and additively manufactured layers of Grade 2509 SDSS . Samples were extracted
from rolled plate as well as additively manufactured layers and subjected to
metallographic examination. Samples were hot-mounted in conductive bakelite and
grinded for macro-etch inspection with different grades of SiC papers which is helpful
in making sure whether imperfections are present or not. For microstructural
examinations, specimens were further grinded and polished using 6, 1 and ¼ μm
colloidal alumina suspension. The samples were etched using the etchant that contains
20 ml HCl+20 ml ethyl alcohol+1 g CuCl2. The samples were examined using optical
microscope (OM) to measure the size of grains as per ASTM E112 [80] and volume
fraction of phases by manual point count as per ASTM E562 [81] and by an image
analysis software as per ASTM E1245 [82].

4.7 Fracture Toughness Testing using SENT According to BS 8571

As specified in the BS 8571 standard SENT specimens were machined with a square
cross-section [62]. The specimen dimensions, including initial crack length, are shown
in Figure 4.5. For the base material characterization, the six specimens were taken out
from the SDSS as-received plate, for the WAAM characterization, the other six
specimens were taken out from the deposited SDSS.

77
 W= 12
mm
Figure 4.5 SENT specimens
All the specimens were notched using electro-discharge machining, with a Ø300μm
wire to reach an initial crack length of 4 mm. As depicted in Figure 4.6, the notch
orientation of the rolled plate specimens was parallel to the rolling direction and for
the WAAM’d plate specimens the notch orientation was perpendicular to the
deposition direction.

(a)

(b)

Figure 4.6 Orientations of the SENT samples: (a) As-received plate, (b) Additively
manufactured sample, Grade 2207
78
For the deposited 2509 Grade SDSS characterization, the other six specimens were
taken out from the additively manufactured part, here the study was focused on the
overlapped heat affected zone (OHAZ) as depicted in Figure 4.7. The (OHAZ) phrase
here is defined as the critical zone that is affected by heat more than once. So that, the
location of pre-crack was chosen carefully.

Fusion lines Weld metal

Overlapped
HAZ

Figure 4.7 SENT specimens place and orientation in HAZ for Grade 2509

The notched specimens were fatigue pre-cracked in three-point bending (see

Figure 4.8) at room temperature within the limits of ISO 12135 standard [61]. The pre-
crack length was chosen to be 1.5 mm. Universal tensile testing unit was used to
conduct the fracture toughness tests at a temperature of -10°C, which is the expected
service temperature for the offshore equipment to be manufactured out of SDSS.
Localized cooling was applied to cool down the specimens using a flow of liquid
nitrogen vapor as illustrated in Figure 4.9.

Tests were performed at a range of crack extensions between 0.45 mm and 0.90 mm
for the specimens from as-received rolled plate and between 0.3 mm and 1.08 mm for
the WAAM specimens which satisfied standard criteria (0.2 mm to 20% of the
specimen ligament (W-a0)) [62]. The results for each specimen (load vs. CMOD) were
used to provide one point on the J R-curve.

79
 Figure 4.8 Fatigue pre-crack

Figure 4.9 Fracture toughness test setup

80
4.7.1 Post-Test Measurements

After executing the SENT tests, measurements of the samples were taken, and data
analysis is carried out to determine a value for the fracture toughness. First, heat tinting
was performed on the specimens by heating them up to 300°C. Then they were
quenched in a bath of liquid nitrogen (-196°C) before they were broken. Due to the
brittle behavior of the quenched samples they were broken without any further plastic
fracture behavior.
High resolution images of broken samples were taken. As prescribed by the ISO 12135
standard, nine reference measurements points were plotted on each image as indicated
in Figure 4.10 [61]. The outer two positions of measurement are taken at 0.01 times
the total sample width, and the remaining 7 measurements are spaced evenly with each
other. Image analysis software ImageJ [83] was used to perform the measurements.
The value of ao is obtained by equation 4.1.

a Measure initial and final

crack lengths at positions 1 to 9.
b Not to scale.
c Reference lines.
d Crack plane.
e Machined notch.
f Fatigue precrack.
g Initial crack front.
h Stretch zone.
i Crack extension.
j Final crack front.
k Side groove.

Figure 4.10 Measurement of crack length [59].

1 𝑎1 + 𝑎9 𝑗=8
𝑎0 = [( ) + ∑𝑗=2 𝑎𝑗 ] (4.1)
8 2

81
4.7.2 Calculating J

J shall be calculated at the assessment point using equation (4.2):

𝐾2 𝜂𝑝 𝑈𝑝
𝐽 = 𝐽𝑒𝑙 + 𝐽𝑝𝑙 = + (4.2)
𝐸′ 𝐵𝑏0

𝐸 ′ = 𝐸 ⁄(1 − 𝜐 2 ) (4.3)
where
Jel = elastic part of the J-integral which directly linked to the Stress Intensity Factor K.
Jpl = plastic part of the J-integral.
E’ = E for plane stress (E is Young’s modulus).
ν is Poisson’s ratio.
b0 = (W-a0) is the remaining ligament.
a0 is the initial crack length.
ηp is a dimensionless function of the geometry.
Up is the area under the plastic part of the load versus crack mouth opening
displacement (CMOD) curve, in (N·mm) (see Figure 4.11).

Key
X Crack mouth opening displacement
Y Applied force
1 Line parallel to the elastic part of the load-displacement trace (line 0A), offset
to intersect at the point of assessment
Figure 4.11 Load versus CMOD curve

82
Stress Intensity Factor K for clamped specimens shall be calculated using equation 4.4
𝑃 √𝜋𝑎0 𝑎
𝐾 = [(𝐵.𝐵 0.5
] 𝐺 ( 𝑊0 ) (4.4)
𝑁) 𝑊

where:
𝑎 𝑎 𝑖−1
𝐺 ( 𝑊0 ) = ∑12 0
𝑖=1 𝑡𝑖 ( 𝑊 ) (4.5)

where ti for H/W = 10 are as follows and valid for 0.05 ≤ a0/W ≤ 0.95:

i ti i ti
1 1.197 7 −36.137
2 −2.133 8 51.215
3 23.886 9 −6.607
4 −69.051 10 −52.322
5 100.462 11 18.574
6 −41.397 12 19.465

J is calculated using equation 4.2 and equation 4.6:

𝑎 𝑖
𝜂𝑝 = ∑10 0
𝑖=0 𝜑𝑖 ( 𝑊 ) (4.6)

Where φi for H/W = 10 are as follows and are valid for 0.05 ≤ a0/W ≤ 0.7:
i φi i φi
0 1.000 6 −77.984
1 −1.089 7 38.487
2 9.519 8 101.401
3 −48.572 9 43.306
4 109.225 10 −110.770
5 −73.116

4.7.3 Definition of the R-curve

The data were plotted as fracture toughness J against the stable crack extension, Δa.
The J–R curve line was fitted to the test data as a lower-bound to crack extensions

83
between 0.2 mm and the maximum crack extension from the test data, as shown in
Figure 4.12.

The R-curve equation form was:

𝐽 = 𝑚 + 𝑙(Δ𝑎)𝑋 (4.7)
where m, l and x are constants.

Key
_____ Fitted R-curve, extrapolated to 0.2 mm crack extension to determine J0.2 or
δ 0.2
1 Initiation value of fracture toughness, J0.2 or δ 0.2
 Data with Δa < 0.2 mm or Δa > 0.2 × (W - a0) mm: excluded from curve fit
but plotted
 Data used in fitting the R-curve
X Δa (mm)
Y CTOD, J
Figure 4.12 Fitting the R-curve to the SENT crack extension data

84
 m CHAPTER 5

5. RESULTS AND DISCUSSION

5.1 Background

This chapter presents the experimental results, including mechanical characterization

and microstructural analysis of specimens. Fracture toughness results of as-received
rolled plate and additively manufactured plates are also presented.

5.2 Integrity Assessment of the Additively Manufactured Beads

5.2.1 Appearance and parts characterization

Top, side, and front views of the wire arc additively manufactured sample with plates
of Grade 2509 SDSS is shown in Figure 5.1, displaying the added beads direction and
the final height which was approximately 30 mm.

Figure 5.1 Additively manufactured plates of Grade 2509

85
5.2.2 Macroscopic characterization

A sample from transverse side of WAAM’d plate Grade 2509 , as shown in Figure 5.2,
was taken for macroscopic examination. The macro-image of the sample shows the
layers of manufactured part. The layer bands are in a concave shape and the thickness
of the layers varies between 2 mm to 3 mm.

Fusion lines

Figure 5.2 Macro-image of the cross-section of the additively manufactured sample

of Grade 2509

5.2.3 Radiographic Examination

After visual inspection, the top and side surface of additively manufactured part was
machined as shown in Figure 5.3-a, and X-ray radiographic examination was done
aaccording to ASTM E1742 [76], and radiographs of Figure 5.3-b were evaluated as
per the requirements and definitions in ASME BPV Code Section IX [77]. The quality
level employed for radiography was 2 % (2-2T using hole type IQI).
The additively manufactured part was free of cracks and no lack of fusion or
volumetric discontinuities were detected.

86
 (a) (b)

Figure 5.3 X-ray films

5.3 Microstructure Characterization

5.3.1 As-Received Product Forms

As-received (Grade 2507) super duplex stainless steel rolled plate has elongated ferrite
and austenite grains in the rolling direction. An optical micrograph of this material is
shown in Figure 5.4. The dark phase is ferrite and the bright phase is austenite in the
wrought microstructure.

Ferrite

Austenite

Rolling direction

Figure 5.4 Microstructure of As-received SDSS Grade 2507 (100X)

87
5.3.2 WAAM deposited beads

For microstructural examination, four samples were taken from the side surface of
deposited beads of Grade 2509 SDSS plates, two from fusion line between beads, one
above the fusion line (weld metal) and one below the fusion line (overlapped heat
affected zone (OHAZ)) as shown in Figure 5.5.
The revealed microstructures of various positions in the additively manufactured beads
are shown in Figure 5.6 to Figure 5.9. The fusion line1 and 2 microstructures are
illustrated in Figure 5.6 and Figure 5.7 respectively. The microstructure contains a
mixture of primary and secondary austenite.
Figure 5.8 shows the microstructure of the lower fusion line where Widmanstätten
austenite formed because of the reheating of the underlying layer. The secondary
austenite (γ2) is formed from the metastable ferrite at a lower temperature.
Widmanstätten and intergranular austenite become finer at the top layer, upper fusion
line, due to a higher cooling rate, as shown in Figure 5.9. The high cooling rate of the
top layer is caused by convection and radiation without any further heat accumulation.

Upper
Fusion line 2

Fusion line 1

Lower

Figure 5.5 Microstructural examination points

88
 Austenite
Austenite

Ferrite Austenite

Ferrite
Ferrite

(a)

Intergranular austenite

Widmanstätten austenite

(b)
Figure 5.6 Microstructures of fusion line 1, (a) 290X magnification and (b) 1160X
magnification

89
 (a)

Widmanstätten austenite

Secondary austenite

(b)
Figure 5.7 Microstructures of fusion line 2, (a) 290X magnification and (b) 1160X
magnification

90
 (a)

Intergranular austenite

Secondary austenite

Widmanstätten austenite

Austenite

(b)
Figure 5.8 Microstructures of (OHAZ) lower fusion line, (a) 290X magnification and
(b) 1160X magnification

91
 (a)

Widmanstätten austenite

Intergranular austenite

(b)
Figure 5.9 Microstructures of weld metal upper fusion line, (a) 290X magnification
and (b) 1160X magnification

92
The ferrite/austenite ratio is an important microstructural parameter affecting the
mechanical properties and corrosion resistance of SDSS. While the SDSS raw material
consists of approximately equal proportions of ferrite and austenite, the additively
manufactured SDSS can show a significantly wider range. When using wire arc
additive manufacturing, the solidification is normally fully ferritic and austenite forms
during cooling [84]. Unbalanced ferrite/austenite ratio influences mechanical and
corrosion properties, and therefore, investigations are required. Different industry
standards set the limit for the ferrite fraction in base metal, heat affected zone, and
weld metal of DSS welds. For instance, API TR 938-C stated that ferrite content should
be 30–65% for base metal, 40–65% for HAZ, and 25–60% for the weld zone [46],
whereas NACE MR0175/ISO 15156-1 states that ferrite content in the weld metal
should be in the range of 30% to 70% volume fraction [85]. Different approaches such
as point counting (PC) and image analysis (IA) can be used to measure ferrite/austenite
ratio. IA is a quick method but demands high quality microstructural images and
computer software packages [84, 86, 87]. The microstructural images of the samples,
which were taken from places shown in Figure 5.5, were chosen to determine the
ferrite/austenite ratio of the WAAM using image analysis approach. The
ferrite/austenite ratio was calculated using ImageJ 1.5j8 image analysis software [83]
and the results are shown in Figure 5.10, Figure 5.11, Figure 5.12 and Figure 5.13.

Austenite Austenite Austenite

Measurement Ferrite
290X 1160X Average
Area % 40.71 % 33.46 % 37.09 % 62.91 %

Figure 5.10 Ferrite/austenite ratio at fusion line 1

93
 Austenite Austenite Austenite
Measurement Ferrite
290X 1160X Average
Area % 43.19 % 29.68 % 36.44 % 63.56 %

Figure 5.11 Ferrite/austenite ratio at fusion line 2

Austenite Austenite Austenite

Measurement Ferrite
290X 1160X Average
Area % 46.05 % 34.69 % 40.37 % 59.63 %

Figure 5.12 Ferrite/austenite at OHAZ.

94
 Austenite Austenite Austenite
Measurement Ferrite
290X 1160X Average
Area % 37.49 % 25.54 % 31.52 % 68.48 %

Figure 5.13 Ferrite/austenite weld metal.

As stated previously, during additive manufacturing, the increase in temperature
causes a significant change in the microstructure of the duplex stainless steel in terms
of the volume fraction and the morphology of the ferrite and austenite phases. These
microstructural changes, expectedly, changes the mechanical properties. As shown in
Figure 5.10 and Figure 5.11, the ferrite fraction in two different fusion lines was
62.91% and 63.56% with an average of 63.2%, whereas Figure 5.12 and Figure 5.13
show that the ferrite percentage in the upper and lower fusion lines were 68.48% and
59.63%. The average ferrite percentage around the fusion line is 63.65%. Although the
average result is acceptable according to NACE MR0175/ISO 15156-1 [85], the ferrite
percentage still remains high and will probably affect the mechanical properties
negatively. Maintaining a proper ferrite / austenite phase balance is necessary to
maintain the acceptable toughness and corrosion resistance of the beads.

5.4 Mechanical Characterization

5.4.1 Tensile Test

Five specimens were extracted from the received plate of SDSS grade 2507 and one
sample was sectioned from the additively manufactured block of SDSS grade 2509.
Table 5.1 contains the tensile properties of examined samples and tensile properties

95
from 2507 SDSS data sheet [88]. Although the present microstructure of WAAM’d
specimens is different from the wrought material, the mechanical properties such as
UTS and elongations are comparable, and similar to the results of others in the
literature [75, 88]. Figure 5.14 shows the stress-strain curves of as-received super
duplex stainless steel plate Grade 2507 and WAAM’d plate made with Grade 2509
super duplex stainless steel filler wire. Figure 5.15 shows fractured samples after
tension tests.

Table 5.1 Mechanical properties of Grade 2507 and Grade 2509 super duplex
stainless steel.
Ultimate Yield Modulus of Elongation
Tensile Strength Elasticity (%)
Sample
Strength (MPa) (GPa)
(MPa)
Grade 2507
as received rolled 836 ± 7 570 ± 10 193 ± 5 39.16 ± 3
Plate
WAAM’d plate made
845 565 196 45.36
with Grade 2509
Grade 2507 plate
800-1000 >550
Data sheet [88]
Grade 2509 filler wire
832 659 30.00
Data sheet [75]

Figure 5.14 Stress-Strain curves of Grade 2507 and Grade 2509 SDSS samples
96
 (a) (b)

Figure 5.15 Specimens (a) before and (b) after the tensile tests.

5.4.2 Hardness Test

Vickers hardness indentations were made around the fusion line as shown in
Figure 5.16. Maximum hardness value was 325 HV for the WAAM part of Grade 2509
at the fusion lines. A microhardness map is also shown in Figure 5.17. According to
SANDVIK data sheet, the hardness of SDSS Grade 2507 is less than 300 HV [88].
The hardness of additively manufactured duplex stainless steel is often higher than
base material due to the strain induced by the thermal cycle. The increase in hardness
is clear in fusion line between beads.

Figure 5.16 Hardness investigation area, WAAM’d Grade 2509

97
 Figure 5.17 Micro-hardness map.

5.5 Fracture Toughness Testing

The fracture toughness tests were conducted on eighteen specimens, six specimens for
the as-received SDSS plate Grade 2507 characterization and six specimens for the
WAAM’d DSS Grade 2209 characterization and six specimens for the WAAM’d
SDSS Grade 2509 characterization. The tests were conducted at a temperature of -
10°C. Figure 5.18 shows the images of some samples before and after test. The
specimens were fatigue pre-cracked at room temperature within the limits of ISO
12135 standard [61]. The pre-crack length was chosen to be 1.5 mm.

(a) (b)

Figure 5.18 Fracture toughness samples (a) before and (b) after the tests.
98
For identification and measurement of crack length, heat tinting was performed on the
specimens by heating them up to 300°C. Then they were quenched in a bath of liquid
nitrogen (-196°C) before they were broken. Due to the brittle behavior of the quenched
samples they were broken without any further plastic fracture behavior. After that,
high-resolution images of each sample were taken. Then, nine reference measurements
points were plotted on each image as indicated in Figure 5.19.

Crack Fatigue
extension pre-crack

Figure 5.19 Measurements points

Fracture toughness determination using J-integral vs. crack growth resistance (J-R)
curve is a useful tool for the evaluation of the structural integrity of a material when
defects are present [54]. A typical J-R curve is shown in Figure 5.20. From it, the
material fracture toughness at the initiation of stable crack growth (Ji) and at maximum
crack extension (Jc) can be derived. According to the BS 8571 standard [62], the data
located in the red zone in Figure 5.20 is the qualified data. The dashed lines parallel to
the Y-axis (J-integral) are the boundaries of the qualified data region which intersects
the x-axis (Δa) at 0.2 and 0.2 (W-a0) mm, respectively.

99
 0.2 0.2 (W-a0)

Jc

Ji

Figure 5.20 Typical J-R curve

To perform the analysis of all samples a Matlab© [89] script is written to automatically
determine the relevant values from the graph and calculate the crack extension and J
value. The Matlab© script was included in Appendice A.

5.5.1 Fracture Toughness of As-received SDSS Rolled Plate Grade 2507

Typical load-displacement curves and fracture surfaces as obtained from fracture test
of as-received SDSS plate Grade 2507 can be seen in Figure 5.21. All fracture surfaces
suggest that a stable crack occurred without prior pop-ins. Table 5.2 summarize the
crack length measurements (a0, af and Δa). The load-displacement curves, fracture
surface and crack length measurements of other samples were included in Appendices
B and C.

100
 Table 5.2 Crack length of as-received SDSS plate Grade 2507 sample 1

Specimen SDSS Grade 2507 (1)

Width (mm) 12.05 (mm)
Thickness (mm) 12.06 (mm)
position a0 af
0.813 5.531 6.077
2.124 5.760 6.535
3.394 5.883 6.694
4.664 5.901 6.861
5.934 5.927 6.974
7.225 5.830 6.782
8.515 5.751 6.658
9.765 5.663 6.685
11.055 5.399 5.945
a0 (mm) 5.773
af (mm) 6.650
∆a (mm) 0.878

BM1
8
Position (mm)

0
0 1 2 3 4 5 6 7 8 9 10 11 12
Crack length (mm)
(a)
Force (mm)

CMOD (mm)
(b)

Figure 5.21. (a) Fracture surface, (b) Load-CMOD of Grade 2507 sample 1
101
The J-integral values against the crack extension, Δa for as-received SDSS plate Grade
2507 are summarized in Table 5.3 and shown in Figure 5.22

Table 5.3 J-integral values of the as-received SDSS plate Grade 2507

Specimen No. J (N/mm) Δa (mm)

Specimen 1 1558.00 0.8775
Specimen 2 1271.00 0.5631
Specimen 3 1190.60 0.5064
Specimen 4 1114.60 0.5023
Specimen 5 1058.30 0.4577
Specimen 6 940.03 0.4463

As-received SDSS plate Grade 2507

1800
1600
1400
1200
J (N/mm)

1000
800
600
400
200
0
0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1
Δa (mm)

Figure 5.22 J-integral vs crack extension curve for as-received SDSS plate
Grade 2507

102
5.5.2 Fracture Toughness of WAAM’d DSS Grade 2209 Plate

Typical load-displacement curves and fracture surfaces as obtained from fracture test
of WAAM’d DSS Grade 2209 plate can be seen in Figure 5.23. All fracture surfaces
suggest that a stable crack occurred without prior pop-ins. Table 5.4 summarize the
crack length measurements (a0, af and Δa). The load-displacement curves, fracture
surface and crack length measurements of other samples were included in Appendices
B and C.

Table 5.4 Crack length of WAAM’d DSS Grade 2209 plate sample 1

Specimen DSS Grade 2209 (1)

Width (mm) 12.03 (mm)
Thickness (mm) 12.08 (mm)
position a0 af
0.697 5.600 6.346
2.006 5.875 6.993
3.296 6.032 7.307
4.606 6.149 7.444
5.935 6.208 7.483
7.224 6.169 7.326
8.553 6.110 7.228
9.843 6.032 6.875
11.133 5.777 6.542
a0 (mm) 6.033
af (mm) 7.138
∆a (mm) 1.105

103
 WM1
8

6
Position (mm)

0
0 1 2 3 4 5 6 7 8 9 10 11 12

Crack length (mm)

(a)
Force (mm)

(b)
CMOD (mm)
(b)

Figure 5.23. (a) Fracture surface, (b) Load-CMOD of Grade 2209 sample 1

The J-integral values against the crack extension, Δa for WAAM’d DSS Grade 2209
plate are summarized in Table 5.5Table 5.3 and shown in Figure 5.24.

Table 5.5 J-integral values of the WAAM’d DSS Grade 2209 plate

Specimen No. J (N/mm) Δa (mm)

Specimen 1 1091.00 1.1046
Specimen 2 879.63 0.5917
Specimen 3 803.45 0.5879
Specimen 4 694.44 0.4619
Specimen 5 681.75 0.4318
Specimen 6 593.39 0.2565

104
 WAAM’d DSS Grade 2209 Plate
1200

1000

800
J (N/mm)

600

400

200

0
0 0,2 0,4 0,6 0,8 1 1,2

Δa (mm)

Figure 5.24 J-integral vs crack extension curve for WAAM’d DSS Grade 2209 plate

The J R-curves have been obtained by curve fitting the test data. Normally, a lower-
bound curve was used and the R-curve equation is:

J = I (Δa)X (5.1)

The fitting parameters I and X for the as-received SDSS plate Grade 2507and
WAAM’d DSS Grade 2209 plate are given in Table 5.6. The J value against the crack
extension based on fitting parameters were shown in Figure 5.25.

Table 5.6 Fitting parameters of J integral

J = I (Δa)X
Material
I X

As-received SDSS plate Grade 1749 0.65

2507
WAAM’d DSS Grade 2209 1039 0.45
plate

105
 As-received plate Grade 2507
2500,00 Base Metal
WAAM’d
AM Metal Grade 2209 plate

2000,00
J (N/mm)

1500,00

1000,00

500,00

0,00
0 0,2 0,4 0,6 0,8 1 1,2 1,4 1,6
Δa (mm)
Figure 5.25 J R-curve lower-bound Fitting Parameters

5.5.3 Fracture Toughness of WAAM’d SDSS Grade 2509 Plate

For the deposited 2509 Grade SDSS characterization, as mentioned in section 4.6, the
study was focused on the overlapping heat affected zones (OHAZ) as depicted in
Figure 4.7. The tests were conducted but the results were very low so that a resistance
curve for specimens taken out from additively manufactured part of Grade 2509 SDSS
could not be generated. Alternatively, their fracture toughness were calculated roughly
using load vs. CMOD curves at weld metal (see Figure 5.26) and OHAZ (see
Figure 5.27). The technical crack initiation value a0 was determined using a stable
crack propagation of Δa = 0.2 mm. The results are summarized in Table 5.7.

106
 Force (N)

CMOD (mm)

Figure 5.26 Load-CMOD of WAAM’d SDSS Grade 2509 plate at the weld metal.
Force (N)

CMOD (mm)

Figure 5.27 Load-CMOD of WAAM’d SDSS Grade 2509 plate at OHAZ.

107
 Table 5.7 J-integral values of WAAM’d SDSS Grade 2509 plate.

Test location P Up a0 J
Grade 2509 (N) (N.mm) (mm) (N/mm)
Weld metal 54166 35735 5.7 421.14

OHAZ 50833 18907 5.7 240.81

The fracture toughness results have demonstrated that there is a disparity between J
integral values of As-received plate Grade 2507 and WAAM’d DSS Grade 2209 plate.
Based on BS 8571 standard, the initiation fracture toughness value is the value of J at
0.2 mm of crack extension and the maximum value at 0.2(W-a0) from the R-curve fit.
Figure 5.28 shows the J-R curves obtained for the investigated As-received plate Grade
2507and WAAM’d SDSS Grade 2209 plate and present the values of Ji and Jc which
are also summarized in Table 5.8 .

As-received plate Grade 2507

WAAM’d Grade 2209 plate
WAAM’d Grade 2509 plate at weld metal
WAAM’d Grade 2509 plate at OHAZ
OHAZ

1970.29 N/mm

612.38
N/mm 1126.33 N/mm

507.1 N/mm
421.14 N/mm
240.81 N/mm

Figure 5.28 J-R curve of different DSS Grades

108
 Table 5.8 J-integral values of different DSS Grades

Material Ji (N/mm) Jc (N/mm)

As-received rolled plate Grade 2507 612.38 1970.29

WAAM’d DSS Grade 2209 plate 507.17 1126.33

WAAM’d SDSS Grade 2509 plate at
421.14 ---
weld metal
WAAM’d SDSS Grade 2509 plate at
240.81 ---
OHAZ

The Ji values of additively manufactured SDSS Grade 2509 plate at the weld metal
and OHAZ also presented as two points. Two results of fracture toughness can be
distinguished. On the one hand, 2507 As-received plate and WAAM’d 2209 Grades
plate show two J integral values, at crack initiation (Ji = 612.38 N/mm and 507.17
N/mm respectively) and at maximum crack extension (Jc = 1970.29 N/mm and
1126.33 N/mm for 2507 and 2209 Grades respectively). Both DSS grades are
characterized by noticeable increase of J values during crack propagation. On the other
hand, for WAAM’d SDSS Grade 2509 plate the Ji values at the weld metal and OHAZ
were 421.14 N/mm and 240.81 N/mm, respectively.

As-received SDSS plate Grade 2507 specimens have an approximately 17% higher (Ji)
value than additively manufactured DSS Grade 2209 plate. For the additively
manufactured SDSS Grade 2509 plate, the value of (Ji) at weld metal was
approximately 31% lower than SDSS Grade 2507 and 16 % lower than DSS Grade
2209. As well, the value of (Ji) at the OHAZ was approximately 60% lower than SDSS
Grade 2507 and 52% lower than DSS Grade 2209. The discrepancy in fracture
toughness value could be explained by the fact that both samples had significant
anisotropy in terms of distribution and morphologies of the phases, ferrite and
austenite. Currently, the reason for a difference in fracture toughness value is an
assumption, and further analysis, such as fractography, is required to better understand
this behavior.

109
 a CHAPTER 6

6. CONCLUSIONS

6.1 CONCLUSIONS

In this thesis, cold metal transfer wire arc additive manufacturing was studied to
produce rectangular plates out of Grade 2509 and Grade 2209 duplex stainless steel
wires. The employed WAAM process parameters were obtained from simulation
studies which aimed parameter optimization to minimize the distortion of the plates
deposited. Microstructural characterization and conventional tensile and hardness
testing of the produced DSS plates were performed. Finally, fracture toughness tests
of additively manufactured SDSS were performed at a sub-zero temperature using the
single edge notched tension (SENT) specimens. The main conclusions that can be
drawn from this study are as follows:

1. It is found that the response surface method, combined with process simulation,
can be used to build statistical models to predict the distortion of WAAM products
within the working range of process variables and obtain optimum process
parameters.

2. The longitudinal and transversal distortions increase with the increase in arc
voltage and welding current for any given value of the welding speed. However,
this increase in distortion gradually decreases as the welding speed increases.

3. Specific to the CMT WAAM of DSS plates produced in this study, the desirable
longitudinal (D1) and transversal (D2) distortions have been found to be 1.17 mm
and 0.70 mm corresponding to the input WAAM process variables of 200 A of
weld current, 15 V of weld voltage and 10 mm/s of weld speed.

4. The strength of the WAAM parts were comparable to the wrought material.
Ultimate tensile strength of samples made out of Grade 2507 and Grade 2509 DSS
were determined as 836 MPa and 845 MPa, respectively.
110
5. Although the ferrite/austenite ratio of the deposited samples varied along the cross
section of the deposited samples, the values remained within the range desired
from duplex stainless steels. Specifically, the ferrite/austenite ratios determined at
fusion line1, fusion line2, upper fusion line and lower fusion line were
(62.91/37.09), (63.56/36.44), (59.63/40.37) and (68.48/31.52), respectively.

6. The initiation J (J0.2) values for the as-received rolled plate and WAAM’d plate
are 612.38 N/mm and 507.17 N/mm, respectively. For the Grade 2509, the Ji
values at the weld metal (deposited during WAAM) were 421.14 N/mm and
240.81 N/mm at the overlapped heat affected zone (OHAZ). Lower fracture
toughness is a common weakness of nearly all additive manufacturing processes
using raw materials in powder or wire form. It is speculated that the significantly
lower fracture toughness values of the WAAM samples should be related with their
anisotropy.

6.2 SUGGESTIONS FOR FUTURE WORK

Suggestions for future work can be classified into two areas:

1. Modeling and simulation aspect.

 Extend the simulation work to study the effect of WAAM parameters
on the fracture toughness of SDSS.
 It could be interesting to estimate the J-R curve of additively
manufactured SDSS based on simulation.
2. Experimental aspect
 Detailed fractographic examination of the SENT samples.
 Inherent aspects of additive manufacturing processes resulting in lower
fracture toughness for engineering parts especially for structural
applications needs much more attention.
 Investigation of the impact of post-weld heat treatment on the
anisotropic behavior, corrosion and fracture toughness of additively
manufactured SDSS.
 Investigation of the fatigue crack propagation behavior of additively
manufactured SDSS.

111
 b REFERENCES

[1] A. Staffanson and P. Ragnartz, “Improving the product development process

with additive manufacturing,” , MSc thesis, Mälardalen University, Sweden,
2018.

[2] A. Yadav, A. Ghosh, and A. Kumar, “Thermal transport phenomena in multi-

layer deposition using arc welding process,” 3D printing and additive
manufacturing, pp. 15-27, 2019.

[3] S.H. Lee, “Optimization of cold metal transfer-based wire arc additive
manufacturing processes using gaussian process regression,” Metals, vol. 10, no.
4, p. 461, 2020.

[4] M. Liberini, A. Astarita, G. Campatelli, A. Scippa, F. Montevecchi, G.

Venturini, M. Durante, L. Boccarusso, F. Memola, C. Minutolo and A. Squillace,
“Selection of optimal process parameters for wire arc additive manufacturing,”
Procedia CIRP, vol. 62, pp. 470-474, 2017.

[5] C.R. Cunningham, J.M. Flynn, A. Shokrani, V. Dhokia and S.T. Newman,
“Invited review article: Strategies and processes for high quality wire arc
additive manufacturing,” Additive Manufacturing, vol. 22, pp. 672-686, 2018.

[6] B. Wu, Z. Pan, D. Ding, D. Cuiuri, H. Li, J. Xu, and J. Norrish, “A review of the
wire arc additive manufacturing of metals: properties, defects and quality
improvement,” Journal of Manufacturing Processes, vol. 35, pp. 127-139, 2018.

[7] K. Ogawa , “Development and Recent Trends of Duplex Stainless Steel,” The
Japan Welding Engineering Society, 2015.

[8] B. Jang, I. Moon, M. Lim, S. Kim, S. Kim, J. Lee, H. Park and J. Koh, “Heat
Treatment Effect on Super Duplex Stainless Steel UNS S32750 FCA Multipass
Welds,” Journal of Welding and Joining, vol. 32, no. 2, pp. 48-53, 2014.

[9] X. Zhang, K. Wang, Q. Zhou, J. Ding, S. Ganguly, G. Marzio, D. Yang, X. Xu,

P. Dirisu, and S.W. Williams, “Microstructure and mechanical properties of
TOP-TIG-wire and arc additive manufactured super duplex stainless steel
(ER2594),” Materials Science and Engineering: A, vol. 762, p.138097, 2019.

[10] A.O. Humphreys, “The Low Temperature Fracture Behaviour of the Super
Duplex Stainless Steel Zeron 100” PhD thesis, University of Birmingham, UK,
1997.

[11] C. Bayley, “Evaluation of the single edge notch tension specimen for quantifying
fracture toughness,” Defence Research and Development, Canada, 2015.

112
[12] K. Davidson and S. Singamneni, “Selective Laser Melting of Duplex Stainless
Steel Powders: An Investigation,” Materials and Manufacturing Processes , vol.
31, no. 12, pp. 1543-1555, 2016.

[13] F. Hengsbach, P. Koppa, K. Duschik, M.J. Holzweissig, M. Burns, J. Nellesen,

W. Tillmann, T. Tröster, K. Hoyer, and M. Schaper, “Duplex stainless steel
fabricated by selective laser melting – Microstructural and mechanical
properties,” Materials & Design , vol. 133, pp. 136-142, 2017.

[14] G. Posch, K. Chladil and H. Chladil, “Material properties of CMT- metal

additive manufactured duplex stainless steel blade-like geometries,” Welding in
the World, vol. 61, no. 5, pp. 873-882, 2017.

[15] M. Eriksson, C. Fredrik, M. Lervåg, C. Sørensen, A. Robertstad, B.M. Brønstad,

B. Nyhus, R. Aune, X. Ren and O.M. Akselsen, “Additive manufacture of
superduplex stainless steel using WAAM,” in 5th International Conference of
Engineering Against Failure (ICEAF-V 2018), 2018.

[16] S. Papula, M. Song, A. Pateras, X. Chen, M. Brandt, M. Easton, Y.

Yagodzinskyy, I. Virkkunen, and H. Hänninen, “Selective laser melting of
duplex stainless steel 2205: effect of post-processing heat treatment on
microstructure, mechanical properties, and corrosion resistance,” Materials , vol.
12, no. 15, p. 2468, 2019.

[17] F. Hejripour, F. Binesh, M. Hebel, and D.K. Aidun, “Thermal modeling and
characterization of wire arc additive manufactured duplex stainless steel,”
Journal of Materials Processing Technology, vol. 272, pp. 58-71, 2019.

[18] V.A. Hosseini, M. Högström, K. Hurtig, M. Asuncion,V. Bermejo, L. Stridh,

and L. Karlsson, “Wire-arc additive manufacturing of a duplex stainless steel:
Thermal cycle analysis and microstructure characterization,” Welding in the
World, vol. 63, no. 4, pp. 975-987, 2019.

[19] J. Stützer, T. Totzauer, B. Wittig, M. Zinke, and S. Jüttner, “GMAW cold wire
technology for adjusting the ferrite–austenite ratio of wire and arc additive
manufactured duplex stainless steel components,” Metals, vol. 9, no. 5, p. 564,
2019.

[20] M. Lervåg, C. Sørensen, A. Robertstad, B.M. Brønstad, B. Nyhus, M. Eriksson,

R. Aune, X. Ren, O.M. Akselsen, and I. Bunaziv, “Additive manufacturing with
superduplex stainless steel wire by CMT process,” Metals, vol. 10, no. 2, p. 272,
2020.

[21] A. Bandyopadhyay, T.P. Gualtieri and S. Bose, “Global Engineering and

Additive Manufacturing,” in Additive Manufacturing, Taylor & Francis Group,
2016, pp. 1-18.

[22] ADDISPACE team, “State of the Art Additive Manufacturing Technologies:

Technologie, Trends, Opportunities, Challenges and Applications to the
Aerospatial Sector,” European Regional Development Fund, 2019.

113
[23] S.W. Williams, F. Martina, A.C. Addison, J. Ding, G. Pardal, and P. Colegrove,
“Wire+ arc additive manufacturing,” Materials Science and Technology, vol. 32,
no. 7, pp. 641-647, 2016.

[24] J.Z. Li, M.R. Alkahari, N.A. Rosli, R. Hasan, M. N. Sudin, and F.R. Ramli,
“Review of Wire Arc Additive Manufacturing for 3D Metal Printing,”
International Journal of Automation, vol. 13, no. 3, pp. 346-353, 2019.

[25] A. Anca, V.D. Fachinotti, G. Escobar Palafox, and A. Cardona, “Computational

modelling of shaped metal deposition,” International journal for numerical
methods in engineering, vol. 85, no. 1, pp. 84-106, 2011.

[26] J. González, I. Rodríguez, J. L. Prado-Cerqueira, J. L. Diéguez, and A. Pereira,

“Additive manufacturing with GMAW welding and CMT technology,”
Procedia Manufacturing , vol. 13, pp. 840-847, 2017.

[27] Z. Pan, D. Ding, B. Wu, D. Cuiuri, H. Li, and J. Norrish, “Arc welding processes
for additive manufacturing: a review,” Transactions on intelligent welding
manufacturing , pp. 3-24, 2018.

[28] P. Kah, H. Latifi, R. Suoranta, J. Martikainen, and M. Pirinen, “Usability of arc

types in industrial welding,” International Journal of Mechanical and Materials
Engineering, vol. 9, no. 1, pp. 1-12, 2014.

[29] S. Selvi, A. Vishvaksenan, and E. Rajasekar, “Cold metal transfer (CMT)

technology-An overview,” Defence technology, vol. 14, no. 1, pp. 28-44, 2018.

[30] FRONIUS Team, “CMT: Cold Metal Transfer,” FRONIUS

INTERNATIONAL. https://1library.net/document/y4kxvm9q-cmt-cold-metal-
transfer-mig-mag-transfer-process.html, [May. 12, 2020].

[31] T. P. Hasselberg, “A Feasibility Study of “Cold Metal Transfer” – Gas Metal

Arc Welding (CMT-GMAW) Nickel Base Superalloy Inconel 718™,”, MSc
thesis, Rensselaer Polytechnic Institute, USA, 2009.

[32] N. R. Tailleart, F. J. Martin, R. J. Rayne, R. A. Bayles, A. Rubinoff, L.

Levenberry, and P. M. Natishan, “Communication—Restoring the Pitting
Resistance of Sensitized Duplex Stainless Steel Using Interstitial Hardening,”
Journal of The Electrochemical Society, vol. 163, no. 8, p. C423, 2016.

[33] X. Xu, M. Zhao, Y. Feng, F. Li, and X. Zhang, “A Comparative Study of Critical
Pitting Temperature (CPT) of Super Duplex Stainless Steel S32707 in NaCl
Solution,” International Journal of Electrochemcal Science, vol. 13, pp. 4298-
4308, 2018.

[34] J. Charles and P. Chemelle, “The history of duplex developments, nowadays

DSS properties and duplex market future trends,” World Iron & Steel, vol. 1, pp.
46-57, 2012.

114
[35] V. Hosseini, “Super Duplex Stainless Steels – Microstructure and Properties of
Physically Simulated Base and Weld Metal,” , PhD thesis, University West,
Sweden, 2018.

[36] TMR Stainless Steel, “Practical guidelines for the fabrication of Duplex
Stainless Steel,” International molybdenum, London, UK, 2014.

[37] K. Yurtisik, “Kinetice and Microstructure Analysis of Fatigue Fracture Progress

in Weld Joints of Duples Stainless Steel Grade 2205,” , PhD thesis, Middle East
Technical Universty, Turkey, 2013.

[38] R. Cervo, “Experimental and Numerical Analysis of Welding and Heat

Treatment of Duplex Stainless Steels,” , PhD thesis, Università degli Studi di
Padova, Italy, 2010.

[39] J. Charles, “Duplex Stainless Steels a Review after DSS ‘07 held in Grado,” Steel
Research International , vol. 79, no. 6, pp. 455-465, 2008.

[40] K. Göransson, M-L. Nyman, M. Holmquist, and E. Gomes, “Sandvik SAF 2707
HD®(UNS S32707): a hyper-duplex stainless steel for severe chloride
containing environments,” Revue de Métallurgie , vol. 104, no. 9, pp. 411-417,
2007.

[41] M Knyazeva and M Pohl, “Duplex steels. Part II: carbides and nitrides,”
Metallography, microstructure, and analysis, vol. 2, no. 5, pp. 343-351, 2013.

[42] S. Sharafi, “Microstructuer of Super- Duples Stainless Steel,” PhD thesis,

University of Cambridge, UK, 1993.

[43] J. C. Lippold, D. J. Kotecki, Welding Metallurgy and Weldability of Stainless

Steel, New Jersey: Jone Wiley & Sons, Inc., 2005.

[44] M. Knyazeva and M. Pohl, “Duplex steels: Part I: Genesis, formation, structure,”
Metallography, Microstructure, and Analysis , vol. 2, no. 2, pp. 113-121, 2013.

[45] A. Vinoth Jebaraj, L. Ajaykumar, C. R. Deepak, and K. V. V. Aditya,

“Weldability, machinability and surfacing of commercial duplex stainless steel
AISI2205 for marine applications–A recent review,” Journal of Advanced
Research, vol. 8, no. 3, pp. 183-199, 2017.

[46] API TECHNICAL REPORT 938-C, “Use of Duplex Stainless Steels in the Oil
Refining Industry,” American Petroleum Institute, USA, 2011.

[47] S. Naskar, “IMPROVED CORROSION PERFORMANCE IN SUPER-

DUPLEX WELDS,” sandvik research and development centre, Sweden , 2015.

[48] T. Pickle, N. Henry, P. Morriss, L. Tennis, D. Wagner, and R. E. Baumer, “Root

pass microstructure in super duplex stainless steel multipass welds,” Welding
Journal , vol. 98, no. 5, pp. 123s-134s, 2019.

[49] IIS Team, “Assessment of Planar Flaws Fracture Mechanics Basic Principles,”
Istituto Italiano della Saldatura, Italy, 2004.
115
[50] G. Cornacchia, “Numerical Models and Experimental Simulation of Irradiation
Hardening and Damage Effects on the Fracture Toughness of 316L Stainless
Steel,” , PhD thesis, The University of Manchester, UK, 2012.

[51] Y. Huang, “Numerical Studies for Improving Fracture Toughness Resistance

Curve Testing Using Single-Edge (Notched) Tension Specimens,” , PhD thesis,
Western University, Canada, 2016.

[52] N. M. Dhansay, “Fracture Mechanics Based Fatigue and Fracture Toughness

Evaluation of SLM Ti-6Al-4V,” , MSc thesis, University of Cape Town, South
Africa, 2015.

[53] W.D. Callister, Fundamentals of Materials Science and Engineering. Fifth

Edition, Ed., USA: John Wiley & Sons, Inc., 2001.

[54] T.L. Anderson, Fracture Mechanics Fundamentals and Applications. Third

Edition ed., UK: Taylor & Francis Group, 2005.

[55] SLV Team, “The Welding Engineer´s Current Knowle,” German welding
center, Germany, 2000.

[56] H. Sieurin, “Fracture Toughness Properties of Duplex Stainless Steels,” Royal

Institute of Technology, Sweden, 2006.

[57] Z. Liu, X. Chen, and X. Wang, “Analysis of three-dimensional clamped single

edge notched tension (SENT) specimens,” in Pressure Vessels and Piping
Conference, 2018.

[58] C. Soret, Y. Madi, J. Besson and V. Gaffard, “Use of the Sent Specimen in
Pipeline Design,” in 20th JTM- EPRG European pipeline research group, Paris,
France, 2015.

[59] P. L. Moore and A. M. Crintea, “Single Edge Notched Tension (SENT) Testing
At Low Temperatures,” in 11th International Pipeline Conference IPC2016, ,
UK, 2016.

[60] P. Moore, “Optimisation of SENT test specimen design,” in Eurojoin 9

conference, Norway, 2015.

[61] The International Organization for Standardization, “Metallic materials —

Unified method of test for the determination of quasistatic fracture toughness,”
ISO 12135, Second edition, 2016.

[62] British Standards Institution, “Method of Test for Determination of Fracture

Toughness in Metallic Materials Using Single Edge Notched Tension (SENT)
Specimens,” BS 8571, 2018.

[63] J. Nilsson, “Super duplex stainless steels,” Materials science and technology ,
vol. 8, no. 8, pp. 685-700, 1992.

[64] T.J. Marrow, “Brittle Fracture of Duplex Stainless Steel,” in 12th European
Conference on Fracture (ECF12), UK, 1998.
116
[65] J. Pilhagen, “The Fracture Mechanisms in Duplex Stainless Steels at Sub-zero
Temperatures,” Royal Institute of Technology, Sweden, 2013.

[66] Stat-Ease, Inc. 2007. Design-Expert® Software, V7, USA.

[67] COMSOL, Multiphysics®, “V. 5.6. www.comsol.com. COMSOL AB,

Stockholm, Sweden.”.

[68] V. Gunaraj and N. Murugan, “Prediction and optimization of weld bead volume
for the submerged arc process—part 1,” Welding journal, vol. 79, no. 10, pp.
286s-294s, 2000.

[69] D. C. Montgomery, Design and Analysis of Experiments. John Wiley &Sons,

2001.

[70] V. A. Kostin and G. M. Grygorenko, “Simulation of the Additive Process of

Forming 3D Products from HSLA Steel 09G2S,” in Proceedings of the
COMSOL Conference, Rotterdam, Holland, 2017.

[71] M. T. Santos, P. V. Muterlle and G. C. Carvalho, “Emissivity Caracterization in

Stainless Steels Alloys for Application in Hidroeletric Turbines,” Applied
Mechanics and Materials, vol. 719, pp. 3-12, 2015.

[72] C. R. Xavier, H. G. Delgado, and J. A. de Castro, “An experimental and

numerical approach for the welding effects on the duplex stainless steel
microstructure,” Materials Research, vol. 18, no. 3, pp. 489-502, 2015.

[73] ASTM Subcommitte A01.17, ASTM A240, “Standard Specification for

Chromium and Chromium-Nickel Stainless Steel Plate, Sheet, and Strip for
Pressure Vessels and for General Applications,” American Society for Testing
and Materials, USA, 2005.

[74] European Standard, DIN EN 10088-2, “Stainless Steels - Part 2: Technical

Delivery Conditions For Sheet/Plate And Strip Of Corrosion Resisting Steels For
General Purposes,” Deutsches Institut fur Normung E.V. (DIN), Germany, 2000.

[75] ESAB Team, “Welding Filler Metal Data book,” ESAB, USA, 2016.

[76] ASTM E1742, “Standard Practice for Radiographic Examination,” American

Society for Testing and Material, USA, 2012.

[77] ASME Committe, “ASME Boiler and Pressure Vessel Code, New York:,” The
American Society of Mechanical Engineers, USA, 2011.

[78] ASTM Subcommitte E28.04, ASTM E8, “Standard Practice for Radiographic
Examination,” American Society for Testing and Materials, USA, 2013.

[79] The International Organization for Standardization, “Metallic materials –

Vickers hardness test - Part 1: Test method,” ISO 6507-1, 2018.

117
[80] ASTM Subcommitte E04.08, ASTM E112, “Standard Test Methods for
Determining Average Grain Size,” American Society for Testing and Materials,
USA, 2012.

[81] ASTM Subcommitte E04.14, ASTM E562, “Standard Test Method for
Determining Volume Fraction by Systematic Manual Point Count,” American
Society for Testing and Materials, USA, 2011.

[82] ASTM Subcommitte E04.14, ASTM E1245, “Practice for Determining the
Inclusion or Second Phase Constituent Content of Metals by Automatic Image
Analysis,” American Society fot Testing and Materials, USA, 2011.

[83] W. Rasband, ImageJ, National Institutes of Health, Bethesda, Maryland, USA,

1997-2018..

[84] A. Putz , M. Althuber, A. Zelić, E. M. Westin, T. Willidal, and N. Enzinger,

“Methods for the measurement of ferrite content in multipass duplex stainless
steel welds,” Welding in the World 63, no. 4 (2019): , vol. 63, no. 4, pp. 1075-
1086, 2019.

[85] The International Organization for Standardization, “Petroleum, petrochemical,

and natural gas industries — Materials for use in H2S-containing environments
in oil and gas production,” ANSI/NACE MR0175/ISO 15156-1, 2015.

[86] R. Mohammed, G. M. Reddy, and K. S. Rao, “Effect of Welding Process on

Microstructure, Mechanical and Pitting Corrosion Behaviour of 2205 Duplex
Stainless Steel Welds,” in IOP Conference Series: Materials Science and
Engineering, IOP Publishing, 2018.

[87] V. Hosseini, K. Hurtig, D. Eyzop, A. Östberg, P. Janiak and L. Karlsson, “Ferrite

content measurement in super duplex stainless steel welds,” Welding in the
World 63, no. 2 (2019), vol. 63, no. 2, pp. 551-563, 2019.

[88] Sandvik Team, “Sandvik SAF 2507,” Sandvik Research and Development
Centre, Sweden, 2020.

[89] MATLAB and statistics Toolbox R2013a, The MathWorks, Inc., Natick,
Massachusetts, United States.

118
 c APPENDIX A

The Developed Matlab Codes to calculate the crack length and J value

%Calculate the crack length

clc
filename = 'AM.xlsx'; % Excel file name
B = xlsread(filename,1); % Sheet 1
W = xlsread(filename,2); % Sheet 2
P1=B(:,1);
ao1=B(:,2);
af1=B(:,3);
P2=B(:,4);
ao2=B(:,5);
af2=B(:,6);
P3=B(:,7);
ao3=B(:,8);
af3=B(:,9);
P4=B(:,10);
ao4=B(:,11);
af4=B(:,12);
P5=B(:,13);
ao5=B(:,14);
af5=B(:,15);
P6=B(:,16);
ao6=B(:,17);
af6=B(:,18);
a11=(((ao1(1)+ao1(9))/2)+ao1(2)+ao1(3)+ao1(4)+ao1(5)+ao1(6)+ao1(7)+ao1(8))/8;
a21=(((af1(1)+af1(9))/2)+af1(2)+af1(3)+af1(4)+af1(5)+af1(6)+af1(7)+af1(8))/8;
a12=(((ao2(1)+ao2(9))/2)+ao2(2)+ao2(3)+ao2(4)+ao2(5)+ao2(6)+ao2(7)+ao2(8))/8;
a22=(((af2(1)+af2(9))/2)+af2(2)+af2(3)+af2(4)+af2(5)+af2(6)+af2(7)+af2(8))/8;
a13=(((ao3(1)+ao3(9))/2)+ao3(2)+ao3(3)+ao3(4)+ao3(5)+ao3(6)+ao3(7)+ao3(8))/8;
a23=(((af3(1)+af3(9))/2)+af3(2)+af3(3)+af3(4)+af3(5)+af3(6)+af3(7)+af3(8))/8;
a14=(((ao4(1)+ao4(9))/2)+ao4(2)+ao4(3)+ao4(4)+ao4(5)+ao4(6)+ao4(7)+ao4(8))/8;
a24=(((af4(1)+af4(9))/2)+af4(2)+af4(3)+af4(4)+af4(5)+af4(6)+af4(7)+af4(8))/8;
a15=(((ao5(1)+ao5(9))/2)+ao5(2)+ao5(3)+ao5(4)+ao5(5)+ao5(6)+ao5(7)+ao5(8))/8;
a25=(((af5(1)+af5(9))/2)+af5(2)+af5(3)+af5(4)+af5(5)+af5(6)+af5(7)+af5(8))/8;
a16=(((ao6(1)+ao6(9))/2)+ao6(2)+ao6(3)+ao6(4)+ao6(5)+ao6(6)+ao6(7)+ao6(8))/8;
a26=(((af6(1)+af6(9))/2)+af6(2)+af6(3)+af6(4)+af6(5)+af6(6)+af6(7)+af6(8))/8;
del1=a21-a11;
del2=a22-a12;
del3=a23-a13;
del4=a24-a14;
del5=a25-a15;
del6=a26-a16;
PW1=W(:,1);
119
wo1=W(:,2);
wf1=W(:,3);
PW2=W(:,4);

wo2=W(:,5);
wf2=W(:,6);
PW3=W(:,7);
wo3=W(:,8);
wf3=W(:,9);
PW4=W(:,10);
wo4=W(:,11);
wf4=W(:,12);
PW5=W(:,13);
wo5=W(:,14);
wf5=W(:,15);
PW6=W(:,16);
wo6=W(:,17);
wf6=W(:,18);
w11=(((wo1(1)+wo1(9))/2)+wo1(2)+wo1(3)+wo1(4)+wo1(5)+wo1(6)+wo1(7)+wo1
(8))/8;
w21=(((wf1(1)+wf1(9))/2)+wf1(2)+wf1(3)+wf1(4)+wf1(5)+wf1(6)+wf1(7)+wf1(8))
/8;
w12=(((wo2(1)+wo2(9))/2)+wo2(2)+wo2(3)+wo2(4)+wo2(5)+wo2(6)+wo2(7)+wo2
(8))/8;
w22=(((wf2(1)+wf2(9))/2)+wf2(2)+wf2(3)+wf2(4)+wf2(5)+wf2(6)+wf2(7)+wf2(8))
/8;
w13=(((wo3(1)+wo3(9))/2)+wo3(2)+wo3(3)+wo3(4)+wo3(5)+wo3(6)+wo3(7)+wo3
(8))/8;
w23=(((wf3(1)+wf3(9))/2)+wf3(2)+wf3(3)+wf3(4)+wf3(5)+wf3(6)+wf3(7)+wf3(8))
/8;
w14=(((wo4(1)+wo4(9))/2)+wo4(2)+wo4(3)+wo4(4)+wo4(5)+wo4(6)+wo4(7)+wo4
(8))/8;
w24=(((wf4(1)+wf4(9))/2)+wf4(2)+wf4(3)+wf4(4)+wf4(5)+wf4(6)+wf4(7)+wf4(8))
/8;
w15=(((wo5(1)+wo5(9))/2)+wo5(2)+wo5(3)+wo5(4)+wo5(5)+wo5(6)+wo5(7)+wo5
(8))/8;
w25=(((wf5(1)+wf5(9))/2)+wf5(2)+wf5(3)+wf5(4)+wf5(5)+wf5(6)+wf5(7)+wf5(8))
/8;
w16=(((wo6(1)+wo6(9))/2)+wo6(2)+wo6(3)+wo6(4)+wo6(5)+wo6(6)+wo6(7)+wo6
(8))/8;
w26=(((wf6(1)+wf6(9))/2)+wf6(2)+wf6(3)+wf6(4)+wf6(5)+wf6(6)+wf6(7)+wf6(8))
/8;
del11=w21-w11;
del22=w22-w12;
del33=w23-w13;
del44=w24-w14;
del55=w25-w15;

120
del66=w26-w16;
aoB=[a11 a12 a13 a14 a15 a16];
afB=[a21 a22 a23 a24 a25 a26];
aoW=[w11 w12 w13 w14 w15 w16];
afW=[w21 w22 w23 w24 w25 w26];
DB=[del1 del2 del3 del4 del5 del6];
DW=[del11 del22 del33 del44 del55 del66];
Pb=[P1 P2 P3 P4 P5 P6];
Pw=[PW1 PW2 PW3 PW4 PW5 PW6];
%===========================================================
===================
% Rolled Plate files
%filename1 = 'b1.xls';
%filename1 = 'b2.xls';
%filename1 = 'b3.xls';
%filename1 = 'b4.xls';
%filename1 = 'b5.xls';
%filename1 = 'b6.xls';
%====================
%%WAAM’d plate files
%filename1 = 'w1.xls';
%filename1 = 'w2.xls';
%filename1 = 'w3.xls';
filename1 = 'w4.xls'; %RRR
%filename1 = 'w5.xls';
%filename1 = 'w6.xls';
%====================
AA = xlsread(filename1,1);
H=AA(:,4);
V=AA(:,3);
plot(H,V)
hold on
[x,y]=ginput(2);
s=(y(2)-y(1))/(x(2)-x(1));
H2=H(size(H));
V2=V(size(V));
n=H2(1)-(V2(1)/s);
Ph=[n,H2(1)];
Pv=[0,V2(1)];
plot(Ph,Pv,'k')
%plot(x,y,'r')
n1=x(1)+((V2(1)-y(1))/s);
ph1=[x(1),n1];
pv1=[y(1),V2(1)];
plot(ph1,pv1,'r');
AA1=trapz(H,V);
AA2=trapz(Ph,Pv);

121
AA3=(H2(1)-n)*(V2(1))/2;
xlabel('CMOD (mm) ')
ylabel('Force (N) ')
hold off
B1=AA(1,6);
W1=AA(1,5);
Up1=AA1-AA2;
Pf1=V2(1);
%=================================================
%a0=aoB(1);
%a0=aoB(2);
%a0=aoB(3);
%a0=aoB(4);
%a0=aoB(5);
%a0=aoB(6);
%a0=aoW(1);
%a0=aoW(2);
%a0=aoW(3);
a0=aoW(4);
%a0=aoW(5);
%a0=aoW(6);
t=[1.197 -2.133 23.886 -69.051 100.462 -41.397 -36.137 51.215 -6.607 -52.322
18.574 19.465];
Q=[1 -1.089 9.519 -48.572 109.225 -73.116 -77.984 38.487 101.401 43.306 -
110.770];
E=195000;
V=0.3;
%%%%
Aw1=a0/W1;
for i=1:12;
G(i)=t(i)*((Aw1)^(i-1));
end;
for k=1:11;
n(k)=Q(k)*((Aw1)^(k-1));
end;
G1=sum(G);
n1=sum(n);
K1=Pf1*sqrt(pi*a0)*G1/(W1*sqrt(B1*B1));
E1=E/(1-(V^2));
bo=W1-a0;
Je1=((K1^2)/E1);
Jp1=((n1*Up1)/(B1*bo));
J1=Je1+Jp1;
a0
Up1
J1

122
 d APPENDIX B

e Fracture Surface and Load-CMOD Curve of Rolled Plate and Additively

Manufactured Plate

8 BM1
Position (mm)

0
0 1 2 3 4 5 6 7 8 9 10 11 12

Crack length (mm)

(a)
Force (mm)

CMOD (mm)
(b)

Figure B.1 a. a) Fracture surface, (b) Load-CMOD of Grade 2507 sample 1

123
 BM2
Position (mm) 8

0
0 1 2 3 4 5 6 7 8 9 10 11 12
Crack length (mm)

(a)
Force (mm)

CMOD (mm)
(b)

Figure B.2 a. a) Fracture surface, (b) Load-CMOD of Grade 2507 sample 2

124
 BM3
8
Position (mm)

0
0 1 2 3 4 5 6 7 8 9 10 11 12
Crack length (mm)

(a)
Force (mm)

CMOD (mm)
(b)

Figure B.3 a. a) Fracture surface, (b) Load-CMOD of Grade 2507 sample 3

125
 BM4
Position (mm) 8

0
0 1 2 3 4 5 6 7 8 9 10 11 12
Crack length (mm)

(a)
Force (mm)

CMOD (mm)
(b)

Figure B.4 a. a) Fracture surface, (b) Load-CMOD of Grade 2507 sample 4

126
 BM5
Position (mm) 8

0
0 1 2 3 4 5 6 7 8 9 10 11 12
Crack length (mm)

(a)
Force (mm)

CMOD (mm)
(b)

Figure B.5 a. a) Fracture surface, (b) Load-CMOD of Grade 2507 sample 5

127
 Position (mm)
8
BM6
6

0
0 1 2 3 4 5 6 7 8 9 10 11 12
Crack length (mm)

(a)
Force (mm)

CMOD (mm)
(b)

Figure B.6 a. a) Fracture surface, (b) Load-CMOD of Grade 2507 sample 6

128
 WM1
Position (mm) 8

0
0 1 2 3 4 5 6 7 8 9 10 11 12

Crack length (mm)

(a)
Force (mm)

CMOD (mm)
(b)

Figure B.7 a. Fracture surface, (b) Load-CMOD of Grade 2209 sample 1

129
 WM2
Position (mm) 8

0
0 1 2 3 4 5 6 7 8 9 10 11 12 13
Crack length (mm)

(a)
Force (mm)

CMOD (mm)
(b)

Figure B.8 a. Fracture surface, (b) Load-CMOD of Grade 2209 sample 2

130
 WM3
Position (mm) 8

0
0 1 2 3 4 5 6 7 8 9 10 11 12 13
Crack length (mm)

(a)
Force (mm)

CMOD (mm)
(b)

Figure B.9 a. Fracture surface, (b) Load-CMOD of Grade 2209 sample 3

131
 WM4
Position (mm) 8

0
0 1 2 3 4 5 6 7 8 9 10 11 12 13
Crack length (mm)

(a)
Force (mm)

CMOD (mm)
(b)

Figure B.10 a. Fracture surface, (b) Load-CMOD of Grade 2209 sample 4

132
 WM5
Position (mm) 8

0
0 1 2 3 4 5 6 7 8 9 10 11 12 13
Crack length (mm)

(a)
Force (mm)

CMOD (mm)
(b)

Figure B.11 a. Fracture surface, (b) Load-CMOD of Grade 2209 sample 5

133
 WM6
Position (mm) 8

0
0 1 2 3 4 5 6 7 8 9 10 11 12 13
Crack length (mm)

(a)
Force (mm)

CMOD (mm)
(b)

Figure B.12 a. Fracture surface, (b) Load-CMOD of Grade 2209 sample 6

.

134
 f APPENDIX C
g Crack Length Measurements of Rolled Plate and Additively Manufactured
Plate
Table C.1 Crack length of as-received SDSS plate Grade 2507 sample 1

Specimen SDSS Grade 2509 (1)

Width (mm) 12.05 (mm)
Thickness (mm) 12.06 (mm)
position a0 af
0.813 5.531 6.077
2.124 5.76 6.535
3.394 5.883 6.694
4.664 5.901 6.861
5.934 5.927 6.974
7.225 5.83 6.782
8.515 5.751 6.658
9.765 5.663 6.685
11.055 5.399 5.945
a0 (mm) 5.773
af (mm) 6.650
∆a (mm) 0.878
Table C.2 Crack length of as-received SDSS plate Grade 2507 sample 2

Specimen SDSS Grade 2509 (2)

Width (mm) 12.08 (mm)
Thickness (mm) 12.07 (mm)
position a0 af
0.544 5.790 6.043
1.911 6.110 6.710
3.236 6.307 6.819
4.54 6.449 7.049
5.907 6.519 7.225
7.211 6.413 7.102
8.557 6.325 6.890
9.925 6.148 6.678
11.229 5.831 6.184
a0 (mm) 6.260
af (mm) 6.823
∆a (mm) 0.563

135
Table C.3 Crack length of as-received SDSS plate Grade 2507 sample 3

Specimen SDSS Grade 2509 (3)

Width (mm) 12.03 (mm)
Thickness (mm) 12.08 (mm)
position a0 af
0.642 5.811 6.184
1.949 6.122 6.557
3.338 6.288 6.889
4.624 6.371 6.972
5.952 6.433 7.055
7.259 6.371 6.806
8.628 6.309 6.765
9.955 6.060 6.578
11.262 5.728 6.122
a0 (mm) 6.215
af (mm) 6.722
∆a (mm) 0.506

Table C.4 Crack length of as-received SDSS plate Grade 2507 sample 4

Specimen SDSS Grade 2509 (4)

Width (mm) 12.05 (mm)
Thickness (mm) 12.07 (mm)
position a0 af
0.444 5.686 5.982
1.799 6.045 6.551
3.153 6.256 6.826
4.508 6.362 6.868
5.884 6.425 6.952
7.217 6.319 6.826
8.55 6.193 6.699
9.905 5.982 6.530
11.28 5.644 6.045
a0 (mm) 6.156
af (mm) 6.658
∆a (mm) 0.502

136
Table C.5 Crack length of as-received SDSS plate Grade 2507 sample 5

Specimen SDSS Grade 2509 (5)

Width (mm) 12.06 (mm)
Thickness (mm) 12.05 (mm)
position a0 af
0.525 5.639 6.035
1.881 5.954 6.369
3.237 6.140 6.556
4.594 6.285 6.744
5.971 6.327 6.890
7.327 6.223 6.751
8.662 6.140 6.577
9.998 5.922 6.390
11.333 5.639 5.994
a0 (mm) 6.079
af (mm) 4.053
∆a (mm) 0.458

Table C.6 Crack length of as-received SDSS plate Grade 2507 sample 6

Specimen SDSS Grade 2509 (6)

Width (mm) 12 (mm)
Thickness (mm) 12.06 (mm)
position a0 af
0.577 5.733 5.942
1.938 5.942 6.458
3.299 6.169 6.602
4.639 6.231 6.767
6 6.231 6.808
7.34 6.220 6.664
8.701 6.128 6.540
10.041 5.942 6.313
11.402 5.733 6.087
a0 (mm) 6.075
af (mm) 6.521
∆a (mm) 0.446

137
Table C.7 Crack length of WAAM’d DSS Grade 2209 sample 1

Specimen DSS Grade 2209 (1)

Width (mm) 12.03 (mm)
Thickness (mm) 12.08 (mm)
position a0 af
0.697 5.600 6.346
2.006 5.875 6.993
3.296 6.032 7.307
4.606 6.149 7.444
5.935 6.208 7.483
7.224 6.169 7.326
8.553 6.110 7.228
9.843 6.032 6.875
11.133 5.777 6.542
a0 (mm) 6.033
af (mm) 7.138
∆a (mm) 1.105

Table C.8 Crack length of WAAM’d DSS Grade 2209 sample 2

Specimen DSS Grade 2209 (2)

Width (mm) 12 (mm)
Thickness (mm) 12.09 (mm)
position a0 af
0.393 5.853 6.200
1.789 6.135 6.591
3.207 6.244 6.830
4.625 6.309 7.004
6 6.374 7.112
7.418 6.352 7.047
8.836 6.265 6.960
10.233 6.135 6.656
11.673 5.831 6.179
a0 (mm) 6.207
af (mm) 6.799
∆a (mm) 0.592

138
Table C.9 Crack length of WAAM’d DSS Grade 2209 sample 3

Specimen DSS Grade 2209 (3)

Width (mm) 12.03 (mm)
Thickness (mm) 12.02 (mm)
position a0 af
0.444 5.735 6.067
1.797 6.045 6.421
3.216 6.133 6.642
4.591 6.177 6.819
5.989 6.133 6.908
7.364 6.023 6.731
8.739 5.978 6.598
10.115 5.624 6.288
11.49 5.359 5.846
a0 (mm) 5.958
af (mm) 6.545
∆a (mm) 0.588

Table C.10 Crack length of WAAM’d DSS Grade 2209 sample 4

Specimen DSS Grade 2209 (4)

Width (mm) 12.05 (mm)
Thickness (mm) 12.04 (mm)
position a0 af
0.455 5.627 5.954
1.813 5.867 6.304
3.192 6.107 6.566
4.572 6.151 6.653
5.952 6.129 6.700
7.353 6.063 6.566
8.711 5.954 6.435
10.09 5.867 6.282
11.514 5.496 5.823
a0 (mm) 5.962
af (mm) 6.424
∆a (mm) 0.462

139
Table C.11 Crack length of WAAM’d DSS Grade 2209 sample 5

Specimen DSS Grade 2209 (5)

Width (mm) 12.01 (mm)
Thickness (mm) 12.02 (mm)
position a0 af
0.382 5.438 5.713
1.781 5.819 6.136
3.18 5.967 6.368
4.58 5.988 6.558
5.958 5.988 6.516
7.378 6.051 6.474
8.777 5.882 6.389
10.177 5.713 6.115
11.555 5.417 5.755
a0 (mm) 5.854
af (mm) 6.286
∆a (mm) 0.432

Table C.12 Crack length of WAAM’d DSS Grade 2209 sample 6

Specimen DSS Grade 2209 (6)

Width (mm) 12.01 (mm)
Thickness (mm) 12.05 (mm)
position a0 af
0.463 5.658 5.899
1.875 6.097 6.404
3.287 6.316 6.580
4.654 6.514 6.711
6.066 6.536 6.799
7.5 6.426 6.689
8.912 6.326 6.678
10.301 6.260 6.458
11.691 5.690 5.865
a0 (mm) 6.269
af (mm) 6.525
∆a (mm) 0.257

140