Nickel-Based Alloy Overlay On A 2.25Cr1Mo PDF

Saghafifar, Hassan (2011) Microstructural stability of a
nickel-based alloy overlay on a 2.25Cr1Mo steel

substrate. PhD thesis, University of Nottingham.
Access from the University of Nottingham repository:

http://eprints.nottingham.ac.uk/12393/1/hassan.pdf
Copyright and reuse:
The Nottingham ePrints service makes this work by researchers of the University of
Nottingham available open access under the following conditions.
This article is made available under the University of Nottingham End User licence and may
be reused according to the conditions of the licence. For more details see:
http://eprints.nottingham.ac.uk/end_user_agreement.pdf
For more information, please contact [email protected]

Microstructural stability of a nickel-based alloy
overlay on a 2.25Cr1Mo steel substrate
Hassan Saghafifar, M.Sc.
Thesis submitted for the degree of

Doctor of Philosophy
Department of Mechanical, Materials, and Manufacturing
Engineering
The University of Nottingham
July 2011
Abstract
Ni-based superalloy weld overlays are widely used in electricity generating

plants to significantly reduce high temperature corrosion problems of ferritic
steel components under service conditions. Welding a nickel alloy similar to
IN625 onto the outer bore of a 2.25Cr1Mo steel tube enhances its service life
as a superheater tube in the highly corrosive environment of a Waste-to-
Energy boiler. For the purposes of studying the effects of high temperature
service on the microstructure of this laid tube with a weld overly, a series of
thermal exposure tests at 650˚C was performed for different times from 1 day
up to 128 days. The microstructural evolution was studied using a
combination of analytical techniques along with changes in hardness profile
across the interface. Changes in the microstructure were examined using OM,
SEM, EPMA, EBSD and FIB-STEM.
Hardness survey results indicate significant changes in the interfacial region

during ageing. Formation of a soft zone ~300 μm wide and its subsequent re-
hardening was observed in the steel side while the hardness of the bulk steel
remained unchanged. Development of a hard band ~30 μm wide adjacent to
the interface in the weld overlay region and hardening of the bulk overlay
material occurs in the early stages of ageing and remained unchanged with
ageing time. Thermodynamic calculations were performed using Thermo-Calc
software and TCFE6 and TTNI7 databases to aid interpretation of
experimental data.
Microstructural evolution in the steel region is related to the carbide

transformation process and carbon migration. In the bulk of the steel tube, the
bimodal microstructure is stable and in the tempered martensite/bainite areas,
the initial M3C transform to M23C6 through a series of metastable carbides
while in the ferrite grains, M2C carbide precipitates and transforms directly to
the equilibrium carbide. The main process for supply of carbon atoms is
transformation of carbides and the rates of transformation are related to the as-
welded microstructure. On the steel side of interfacial region where bainite
was formed after welding, the stability of metastable carbides is related to the
carbon content. Long term ageing causes Mo replenishment in the coexisting
ferrite and fine grains have formed in this region following long term ageing.
Experimental observations confirmed that a network of alloyed M23C6 carbide

precipitates was formed at the interface in the steel side which are believed to
interrupt the carbon migration across the interface. Moreover, there is a
carbide precipitation region within ~100 μm from the interface in the weld
overlay. Beyond this region intermetallic phases like Mo-rich μ and Nb-rich δ
were formed in the interdendritic regions and along the grain boundaries. Hard
band formation is related to precipitation of the σ phase.
Microstructural stability of a nickel-based alloy on a 2.25Cr1Mo steel substrate...i

Table of Contents
Contents
Abstract …………………………………………………………………………………………………… i
Table of Contents .…………………………………………………………………………………… ii
Acknowledgements ……………………………………………………………………………… viii
List of Abbreviations ……………………………………………………………………………… ix
1- Introduction ....................................................................................... 1
2- Literature review ................................................................................ 5
2-1- Introduction ........................................................................................................... 5
2-2- General Background ............................................................................................... 6
2-2-1- Energy challenges and WtE plants .......................................................................... 6
2-2-2- Materials requirements for boiler in WtE plants .................................................... 7
2-2-3- Surface engineering for boiler components and weld overlay ............................... 8
2-2-4- Alloy 625 weld overlay and welding processes .................................................... 10
2-3- 2.25Cr1Mo steel ................................................................................................... 12
2-3-1- Physical metallurgy of 2.25Cr1Mo steel ............................................................... 13
2-3-2- Carbide transformation in 2.25Cr1Mo microstructure at high temperature ....... 14
2-3-3- Effects of welding on the 2.25Cr1Mo microstructure .......................................... 17
2-4- Alloy 625, a nickel based super alloy .................................................................... 19
2-4-1- Solidification of Alloy 625 and weld deposit of Alloy 625 ..................................... 19
2-4-1-1- Solidification of Alloy 625 ............................................................................ 20
2-4-1-2- Solidification of weld deposit of Alloy 625 .................................................. 23
2-4-1-3- Effect of heat input during welding and cooling rate .................................. 24
2-4-1-4- Effect of dilution .......................................................................................... 25
2-4-2- Phase precipitation during high temperature exposure in Alloy 625 ................... 26
2-4-2-1- Precipitation Hardening Phases .................................................................. 27
Microstructural stability of a nickel-based alloy on a 2.25Cr1Mo steel substrate...ii

Table of Contents
2-4-2-2- Carbides and carbo-nitrides ........................................................................ 29
2-4-2-3- Topologically Closed Packed (TCP) phases .................................................. 32
2-4-3- TTT diagrams for wrought Alloy 625 ..................................................................... 34
2-4-4- Precipitation behaviour of weld deposit .............................................................. 36
2-5- Cladding and dissimilar metal welding ................................................................. 37
2-5-1- Interface and its different regions ........................................................................ 38
2-5-2- Interface and Diffusion ......................................................................................... 40
2-6- Microstructural evolution in dissimilar weld joints ............................................... 41
2-6-1- Different Cr-content steels and formation of soft and hard zones ...................... 41
2-6-2- Steel/nickel alloys and formation of carbide arrays ............................................. 42
2-6-3- Phase transformation at 2.25Cr1Mo steel and Alloy 625 weld overlay ............... 43
2-7- Evaluation of microstructural evolution using thermodynamic calculation .......... 46
2-7-1- Thermo-Calc software .......................................................................................... 47
2-7-2- Modelling of microstructure evolution in Ni-based alloys ................................... 48
2-7-3- Prediction of microstructural states in Cr-Mo steels ............................................ 49
2-7-4- Thermodynamic aspects of carbon migration in dissimilar weld joints ............... 50
2-8- Summary .............................................................................................................. 50
3- Experimental procedure ................................................................... 52
3-1- Introduction ......................................................................................................... 52
3-2- Materials .............................................................................................................. 53
3-3- High temperature exposure tests and thermal exposed samples ......................... 54
3-4- Sample Preparation .............................................................................................. 58
Microstructural stability of a nickel-based alloy on a 2.25Cr1Mo steel substrate...iii

Table of Contents
3-4-1- Composition analysis using spark emission spectroscopy .................................... 58
3-4-2- Sample preparation for optical and SE microscopy .............................................. 59
3-4-3- FIB-TEM sample preparation ................................................................................ 60
3-5- Principles of characterization methods ................................................................. 63
3-5-1- Optical emission spectroscopy ............................................................................. 63
3-5-2- Optical Microscopy ............................................................................................... 63
3-5-3- Scanning Electron Microscopy (SEM) ................................................................... 64
3-5-4- Transmission Electron Microscopy (TEM)............................................................. 65
3-5-5- Ion beam imaging ................................................................................................. 66
3-5-6- Energy Dispersive X-ray Analysis (EDX) ................................................................. 66
3-5-7- Electron Probe Micro Analysis (EPMA) ................................................................. 67
3-5-8- EBSD ...................................................................................................................... 69
3-6- Hardness measurements ...................................................................................... 70
3-6-1- Microhardness method ........................................................................................ 70
3-6-2- Nanohardness method ......................................................................................... 72
4- Characterization of the as-welded condition ..................................... 73
4-1- Introduction ......................................................................................................... 73
4-2- Macrostructural features and chemical composition ............................................ 74
4-3- Microstructural characterization of 2.25Cr1Mo steel substrate ............................ 77
4-3-1- Inner bore microstructure .................................................................................... 80
4-3-2- Central zone microstructure ................................................................................. 80
4-3-3- Transition zone microstructure ............................................................................ 82
Microstructural stability of a nickel-based alloy on a 2.25Cr1Mo steel substrate...iv

Table of Contents
4-3-4- Far-HAZ zone microstructure................................................................................ 84
4-3-5- Near-HAZ zone ...................................................................................................... 87
4-4- Interface characterisation ..................................................................................... 89
4-5- Microstructural characterization of the weld overlay ........................................... 91
4-5-1- Chemical composition profile through the weld overlay region .......................... 95
4-5-1-1- Mixing zone ................................................................................................. 95
4-5-1-2- Bulk zone ..................................................................................................... 97
4-5-2- EPMA results ......................................................................................................... 99
5- Characterization following high temperature exposure ................... 102
5-1- Introduction ....................................................................................................... 102
5-2- Overview ............................................................................................................ 103
5-3- Hardness survey ................................................................................................. 109
5-4- Microstructural characterization ........................................................................ 117
5-4-1- Microstructural evolution in weld overlay region .............................................. 117
5-4-1-1- Microstructural evolution in the early stages of thermal exposure .......... 118
5-4-1-2- Microstructural evolution in the mid-term thermally exposed samples .. 126
5-4-1-3- Microstructural evolution after long term thermal exposure ................... 134
5-4-2- Microstructural evolution in the steel region ..................................................... 144
5-4-2-1- Tempering of the near-HAZ region ........................................................... 145
5-4-2-2- Formation of soft zone .............................................................................. 152
5-4-2-3- Re-hardening and fine grain formation in the near-HAZ region ............... 163
5-4-2-4- Formation of carbides at interface ............................................................ 179
Microstructural stability of a nickel-based alloy on a 2.25Cr1Mo steel substrate...v

Table of Contents
6- Calculation of phase equilibria in Ni-based alloys and 2.25Cr1Mo steel

using Thermo-Calc software ................................................................... 185
6-1- Introduction ....................................................................................................... 185
6-2- CALPHAD principles ............................................................................................ 187
6-3- Thermo-Calc software ........................................................................................ 191
6-3-1- Methodology ...................................................................................................... 191
6-3-2- Multicomponent Databases ............................................................................... 193
6-4- Calculation of phase equilibria in wrought IN625 ............................................... 194
6-5- Calculation of phase equilibria in the weld overlay material .............................. 203
6-5-1- Stable phases in the Bulk zone ........................................................................... 204
6-5-1-1- Effect of carbon on phase equilibria ......................................................... 207
6-5-2- Stable phases in the mixing zone ........................................................................ 208
6-6- Non-equilibrium solidification of weld overlay material ..................................... 213
6-6-1- Bulk zone............................................................................................................. 213
6-6-2- Mixing zone......................................................................................................... 215
6-7- Calculation of stable and metastable phase equilibria in 2.25Cr1Mo steel ......... 218
6-7-1- Effect of temperature on the stable phases ....................................................... 219
6-7-2- Metastable phase equilibria ............................................................................... 220
6-7-3- Effect of carbon content on the stable phase equilibria .................................... 222
7- Discussion and conclusions ............................................................. 225
7-1- Introduction ....................................................................................................... 225
7-2- Microstructural evolution in the steel substrate ................................................. 226
Microstructural stability of a nickel-based alloy on a 2.25Cr1Mo steel substrate...vi

Table of Contents
7-2-1- Carbide transformation in the central zone ....................................................... 227
7-2-2- Carbide transformation in the HAZ region ......................................................... 231
7-2-3- Carbon migration ................................................................................................ 236
7-2-3-1- Thermodynamic prediction ....................................................................... 237
7-2-3-2- Carbon supply model ................................................................................ 239
7-2-4- Formation of fine ferrite grains in the near-HAZ region of the D128 sample..... 242
7-3- Microstructural evolution in the weld overlay .................................................... 243
7-3-1- Microstructure in the as-welded sample ............................................................ 244
7-3-2- Microstrcutral evolution following high temperature exposure ........................ 246
7-3-2-1- Bulk zone hardening .................................................................................. 246
7-3-2-2- Mixing zone hardening .............................................................................. 247
7-3-2-3- Phase precipitation in the interfacial region ............................................. 248
7-4- Summary ............................................................................................................ 250
7-5- Future work ........................................................................................................ 252
8- References ..................................................................................... 253

9- Appendix No.1: Optical emission spectroscopy ............................... 271
Microstructural stability of a nickel-based alloy on a 2.25Cr1Mo steel substrate...vii

Acknowledgments
Acknowledgements
I would like to express my deepest gratitude to Professor Graham McCartney
and Professor Phillip Shipway for their continued support and guidance during
my PhD studies. Their attitude, enthusiasm, patience and encouragement
regarding my research are greatly appreciated.
I would like to thank the technical staff in Advanced Materials Group. Special
thanks to Mr Keith Dinsdale and Dr Nigel Neate for their constant help and
expertise using SEM and TEM. I would like to thank Mr Thomas Buss and Mr
Martin Roe for their assistance using different characterization techniques. I
would also like to thank Mr Graham Malkinson for help in samples
preparation and thermal exposure. Finally, I am very grateful for the help from
Dr Deen Zhang.
I would like to take this opportunity to thank the University of Nottingham
International Office for the funding my Ph.D. programme.
Very special thanks to Hoda, the one and only, who I love deeply, for taking
very good care of me during this period of time, for her unconditional love,
support, understanding, patience, and for encouraging me to follow my Ph.D.
program.
Microstructural stability of a nickel-based alloy on a 2.25Cr1Mo steel substrate...viii

List of abbreviations
List of abbreviations
CCT Continues - Cooling - Transformation
EDS Energy Dispersive X-Ray Spectrometry
G. B. Grain Boundary
IPF Inverse Pole Figure
ISE Ion beam induced Secondary Electron
S.D. Standard Deviation
SIMS Secondary Ion Mass Spectrometry
TCP Topological - Close - Pack
TIG Tungsten – Inert - Gas
TM/B Tempered Martnesite and or Bainite
TTT Time – Temperature – Transformation
Tγ Transformed Austenite (γ)
Microstructural stability of a nickel-based alloy on a 2.25Cr1Mo steel substrate...ix

Chapter1 … Introduction
Chapter 1
1- Introduction
In the 21st century, the world faces an important challenge of providing
plentiful, cheap energy to meet the needs of growing global demands. It is
believed that a strong set of electricity generation technologies and fuels
including fossil fuels and renewable sources should be developed to assure
adequate supplies. One of the renewable energy sources is waste [1] and
conversion of all kinds of it, including municipal solid waste (MSW) and
biomass to energy in waste-to-energy (WtE) plants is recognized globally as
one of the means for electricity supply [2]. In order to improve the thermal
efficiency of the WtE plants, it is essential to raise the operation temperature
of the boiler [3].
Conventionally, the chromium-molybdenum series of ferritic steels such as
2.25Cr1Mo steel is used widely in the power generation and process industries
because of their high strength at elevated service temperature, combined with
Microstructural stability of a nickel-based alloy on a 2.25Cr1Mo steel substrate...1

microstructural stability [4, 5]. However, ferritic steels do not provide
adequate resistance in high temperature corrosive environments such as WtE
boilers. It should be noted that the typical life time of components in a power
station is about 200,000 hours [6] and thus the cost and performance of
materials for this kind of application should be optimised [7].
Optimisation of cost and performance suggests a surface engineering approach
using a Ni-based alloy with high Cr content as a coating on the candidate
ferritic steel. Coatings can be produced by several methods such as co-
extrusion, diffusion treatment, thermal spraying, laser cladding and weld
overlaying [8]. Good bond integrity, heat transfer and corrosion resistance are
the essential characteristics for a coating. Among various methods of cladding,
there is a growing interest in weld overlays because of the intermixing and
metallurgical bonding of the weld metal to the substrate [9].
Alloy 625 is a nickel based alloy with alloying elements such as Cr, Mo and
Nb and is one such material which is commonly used for cladding processes
[10]. There is large body of work on the microstructural evolution of Alloy
625 with different microstructures (wrought and as-cast) at elevated
temperatures and it has concluded that chemical composition and initial
microstructure are critical parameters.
In addition, the microstructural changes in 2.25Cr1Mo steel after long term
service have been extensively reported in the literature (see section 2-3-2).
Moreover, there are some papers regarding effects of welding on the as-
welded microstructure and its evolution at high temperature (see section 2-3-
3). However, little detailed work has been undertaken on the microstructural

evolution of Alloy 625 weld overlay on a 2.25Cr1Mo substrate and their
reciprocal effects.
Nevertheless, there are a number of studies which are devoted to
microstructural analysis of dissimilar material weld joints at high
temperatures. It has been concluded that alloying elements such as carbon
diffuse across the interface according to their chemical potential gradient. The
carbon migration process is a well known process in the many dissimilar weld
joints and it occurs from the low alloy side of interface to the high alloy side.
There are many important factors which affect the carbon diffusion process
such as time and temperature and also the type of stable phases in each
material.
The stable phases in multicomponent materials such as steels and nickel-based
alloys can be predicted by using thermodynamic calculation software such as
Thermo-Calc. Knowing the stable phases in each system helps the systematic
analysis of the effects of chemical composition changes on the microstructural
evolution in the complex systems.
Therefore, the aim of this investigation is to understand the effects of service
conditions at temperatures above 600˚C on the microstructure of the
2.25Cr1Mo steel/nickel-based alloy weld overlay system. The following
objectives are set up to reach this aim:
1) To clearly characterize the microstructure and chemical composition of
weld overlay and 2.25Cr1Mo steel and the compositional variation at
the weld interface in the as-welded condition.

2) To evaluate microstructural changes in both materials following
elevated temperature exposure of samples at 650°C and investigate the
effects of this treatment on the hardness of both materials.
3) To better understand carbon diffusion during high temperature
exposure and its effects on the microstructure and determine the
important parameters affecting carbon diffusion in both materials and
develop a model for the microstructural changes.
4) To analyse the stable phases in both materials by performing
thermodynamic calculations using Thermo-Calc software and to
understand the effects of intermixing during welding and carbon
migration during thermal exposure on the stable phases which are
related to microstructural evolution.

Chapter2 … Literature review
Chapter 2
2- Literature review
2-1- Introduction
This chapter provides the context for the themes explored within this thesis.
The underlying aspects are materials for boiler applications and their
properties are presented and in particular, the use of weld overlay as a
corrosion barrier is discussed. Emphasis is placed upon microstrucutral
features of 2.25Cr1Mo steel and Alloy 625 and their stabilities in high
temperature service. In particular, focus is given to carbide transformation in
the 2.25Cr1Mo steel and solidification behaviour of an Alloy 625 weldment
and its secondary phase formation at high temperatures.
Work on different types of dissimilar weld joints is presented and their
microstructural evolution discussed. The most important issue in these
systems is carbon diffusion between dissimilar materials. The reactions and

phase transformation at the interface are presented. Finally, a thermodynamic
calculation method and its application in prediction of stable phases in the
multi component systems is presented in the final section.
2-2- General Background
2-2-1- Energy challenges and WtE plants
Energy may be the most important factor that will influence the shape of
society in the 21st century. The cost and availability of energy significantly
impacts our quality of life, the health of national economies, the relationships
between nations, and the stability of our environment. Biomass and waste are
renewable energy sources in the modern world [1].
Today‟s Waste-to-Energy (WtE) plants are power plants that utilize municipal
solid waste as their fuel rather than coal, oil or natural gas. WtE plants recover
the thermal energy contained in the waste in boilers that generate steam that
can then be sold directly to industrial customers, or used on-site to drive
turbines for electricity production [3].
The use of fossil fuels for electricity generation poses a unique set of
challenges. On the one hand, this kind of fuel is plentiful and available at low
cost in much the world. On the other hand, traditional methods of fuel
combustion emit pollutants and CO2 at high levels relative to other energy-
generation options [11]. Environmental concerns are a great challenge in
modern power generation and it is claimed that in addition to reduction of
waste materials, WtE plants produce electricity from waste with less
environmental impact than landfill process of waste[12].

The principle of WtE technology is feeding the waste material to an
incinerator and producing thermal energy by combusting it. In order to
improve the thermal efficiency of these plants, there is a need to raise the
steam temperature and pressure. The main technology is the development of
high temperature materials with sufficient resisitance to the aggressive
environment. Extensive research and development programmes have resulted
in numerous high strength alloys with sufficient corrosion resistance for the
boiler tubing [2, 3, 13-15].
2-2-2- Materials requirements for boiler in

WtE plants
Low alloy ferritic steels are used extensively in modern power generation for
components which operate at temperature less than 550˚C and steam pressure
less than 35 MPa where creep deformation limits the service life which is
about 200,000 h [16-18]. Creep is an important design consideration for steam
tubes operating at temperatures in excess of 400˚C. The chromium–
molybdenum series of steels are used widely in both power plant and process
industries because they offer high strength at ambient and elevated
temperature, combined with adequate toughness. The availability of these low
alloy steels makes them economically attractive [5].
These steels contain chromium from 0.5 wt.% to 12 wt.% and some
molybdenum, usually up to a maximum of 1 wt.%. Other strong carbide
formers such as vanadium and niobium may also be added. The 2.25Cr1Mo
steel is one of the most important steels in this group [4]. Metallurgical
changes occur when creep-resisting steels are used at elevated temperature

because their properties largely depend on the precipitation and stability of
alloy carbides. The initial microstructure of the steel is also important and
some factors such as grain size, solid solution hardening, type of structure,
precipitation or dispersion hardening and segregation at structural
discontinuities such as grain boundaries are important too.
As waste has a variable composition, the combustion gas from waste
incineration furnaces fluctuates both in gas temperature and composition.
Therefore, corrosion in WtE plants is a more challenging issue than it is in
conventional power plants. Corrosive gases such as hydrogen chloride (HCl)
combined with fly ash (dust) contained in the combustion gas result in a
severe high temperature corrosive environment [8, 19-21]. Low alloy steels
suffer from severe corrosion in chlorine-containing environments; however,
nickel-based alloys are significantly more resistant than steels, as the steels
mass change in a corrosive environment is about 6.5 times compared to that of
the Inconel 625 in the similar environment [1]. The presence of Cr and Mo in
the nickel alloys are very important in corrosion resistance [22].
2-2-3- Surface engineering for boiler

components and weld overlay
There has been an increasing interest in the need to optimise the cost and
performance of materials with appropriate corrosion resistance and
mechanical properties for boiler components. Materials having both attributes
are very scarce and therefore in many applications, bimetallic systems are
necessary.

Materials with satisfactory corrosion resistance must be highly alloyed in
nickel, chromium and molybdenum; thus they are expensive. Moreover there
are many difficulties in utilizing these materials since the mechanical
properties of these alloys may not meet the requirements of structural
components. There is thus a need for surface engineering, using a protective
surface layer to protect the structural components against corrosion and/or
erosion. This allows the structural components to meet the codes and
standards in terms of strength while relying on the surface protective layer to
protect against corrosion. This protective layer can be produced by several
methods such as: weld overlaying; co-extrusion [7]; diffusion bonding [9];
thermal spraying [23]; and laser cladding [24]. The protective layer needs to
exhibit good bond integrity, high heat transfer, good corrosion resistance, and
low porosity, since these are the essential characteristics that provide optimum
life of boiler components.
Among all the cladding processes, there is a growing interest in weld
overlaying in light of the metallurgical bonding between the weld metal and
substrate, providing strong adhesion and reducing the possibility of crack
propagation along the interface. Welding processes can be automated to yield
large output at a reduced cost. The durability of weld overlay cladding is
higher than that of a protective layer produced by other methods, since a dense
coating layer that is chemically bonded with the base metal and as thick as
several millimetres can be produced by this method [25].
Commercially available nickel-based superalloys have been used for weld
overlays [26, 27]. In order to achieve the maximum high-temperature

corrosion protection with these alloys, there is a need to maintain the lowest
possible dilution from the underlying steel in the overlay. Excessive dilution
in welding that may result up to 20 wt% iron content in the nickel-based
superalloy weld overlay is known to reduce the resistance to high temperature
chloride attack to half compared to un-diluted alloy [28].
It is also important that overlay welding does not result in embrittlement or
degradation in the mechanical properties of the overlaid component. Nickel
alloy weld overlays on steel have been found to exhibit similar mechanical
properties to ferritic steel [10]. Furthermore, tests on the components after
long term service have indicated that nickel alloys increased resistance of the
overlaid boiler components to overheating and to falling clinker [10].
2-2-4- Alloy 625 weld overlay and welding

processes
Alloy 625 is a nickel-based superalloy with corrosion resistance and high
strength at temperature above 500˚C [29]. It contains about 22 wt% Cr, about
9 wt% Mo, less than 5 wt% Fe and about 3.5 wt% Nb. Welding and
overlaying with Alloy 625 filler metal is common in many industrial plants in
the oil and gas industries [30]. These welding products closely match the
coefficient of expansion of many ferritic steels, accept iron dilution without
cracking, weld readily with dissimilar metals and possess the corrosion
properties inherent of bulk Alloy 625 [31]. Since this nickel-based alloy has a
long established reputation for its resistance to highly corrosive environments,
it has been used in WtE environments [32, 33]. The bimetallic Alloy 625/steel
system is also a good candidate for many other applications such as offshore

and onshore components [30] and black liquor recovery boilers, and for some
waste heat recovery systems in steel making, such as basic oxygen furnace
(BOF) hoods and electric arc furnace (EAF) hoods [10].
The welding process allows successful use of a number of specific deposition
techniques which can be chosen according to application. For each
application, there is a need to compromise between deposit thicknesses,
surface chemical composition, deposition rate, and welding parameters. Some
argue that one of the most successful processes for weld overlaying Alloy 625
is the submerged- arc-welding (SAW) process [34]. However, there are others
applications where the automatic gas-metal-arc-welding (GMAW) is used as a
standard process for applying Alloy 625, such as production of boiler water-
walls. The pulsed-gas-metal-arc-welding (PGMAW) process has a wide range
of use in overlay applications too, since this process has the lowest dilution
level which is related to lower heat input compared to other processes [30].
Figure 2-1 shows a schematic diagram of weld overlay application on a water
wall of the WtE boiler.
Figure ‎2-1. Schematic diagram of weld overlay beads to cover the tube-membrane
water-wall in the WtE boilers [10].

There are some parts in the boiler such as the superheater which is constructed
from individual tubes and therefore the weld overlay should be applied by a
spiral mode. The spiral weld overlay can be applied with a low energy input
which will only influence the microstructure of the substrate within a limited
depth [35], so that its strength will not change significantly. A superheater
spiral weld overlaid tube is shown schematically in Figure 2-2.
Figure ‎2-2. Schematic diagram of spiral weld overlaid tube [36].
2-3- 2.25Cr1Mo steel
A Cr-Mo low alloy ferritic steel used extensively in modern power generation
has a nominal composition of 0.1 wt.% carbon, 1.9 to 2.6 wt.% chromium and
0.8 to 1.13 wt.% molybdenum. This steel is known as 2.25Cr1Mo steel or T22
[37] and 10CrMo910 [38] in the literature. Its popularity is due to its excellent
mechanical and creep properties as well as good resistance to oxidation and
hydrogen embrittlement at elevated temperatures. In spite of these excellent
properties, a number of problems such as temper embrittlement and creep
embrittlement have often been associated with it use [39].

2-3-1- Physical metallurgy of 2.25Cr1Mo steel
2.25Cr1Mo steel is a low carbon ferritic steel which has been successfully
used in non-nuclear elevated temperature applications for a number of years.
The amount of chromium in this steel confers increased oxidation resistance,
so allowing it to be used at 550 to 600°C. The molybdenum content of this
steel is known to be about the optimum level to give the required creep
resistance. The presence of chromium in the range of 2.25 wt.% counteracts
the embrittling tendency caused by presence of molybdenum [40]. This steel is
normally regarded as the best low alloy creep resisting steel with an optimum
combination of properties with widespread application [5].
The mechanical and corrosion resistance of Cr-Mo steels depend upon their
microstructure [41] which can be predicted using knowledge of their TTT and
CCT diagrams. Figure 2-3 presents these diagrams for 2.25Cr1Mo steel. Large
amounts of 2.25Cr1Mo steel enter service in either the quenched and tempered
condition, or the normalized condition.
Figure ‎2-3. Corresponding (a) TTT and (b) CCT diagrams for a 2.25Cr1Mo steel. The
CCT diagram shows the terminology used in describing air-cooling form the
austeniteisation (i.e. normalizing) and furnace cooling (i.e. annealing) [42].

Some believe that in former case, the microstructure usually consists of
tempered martensite and/or bainite, while in the later case it contains a mixed
microstructure of ferrite and very fine pearlite [43]. However, others argue
that in the air cooled and tempered condition, the microstructure is largely
bainitic with varying amounts of ferrite depending on the cooling rate of
product after tempering [5].
For power plant applications, this steel is usually air cooled after
austenitisation [42]. The initial microstructure has a great effect on the service
life of this steel. By increasing the bainite content to about 50-60%, the creep
resistance of this steel can be maximized [44]. The tempering temperature for
this steel depends on the desired mechanical properties at elevated
temperatures, and it is in the range of 650-750°C.
The major metallurgical advantage of this steel is that it will transform to a
fully bainitic microstructure over a wide range of temperatures. Thus the
tempered microstructure contains many carbide precipitates which act as
dispersion hardening particles. The precipitates present in this state are
spheroidsed carbides originally present in the bainite, predominantly M3C
(M=Cr, Fe, Mo) and the smaller incoherent Mo2C needles [42].
2-3-2- Carbide transformation in 2.25Cr1Mo

microstructure at high temperature
During tempering or service at high temperature, the morphology, distribution
and type of carbides in the 2.25Cr1Mo steel will change depending on the
time and temperature [45]. This transformation depends on the initial
microstructure of the steel and its grain size as well as the applied stress and

temperature of service. Moreover, the carbon content has a critical influence
on the ageing behaviour of this steel [46]. There is a large body of research
which analysed the carbides in this steel after long term service [39, 43, 44,
47, 48]. However, some of the assessments of carbide transformation in this
steel are based on the extrapolation of accelerated thermal ageing tests [49] or
the creep rupture data obtained from laboratory tests [50-52]. The validity of
such an approach is thought to be justified since the accelerated ageing tests
have shown to produce carbides whose chemical compositions are comparable
to those of the carbides which precipitate under actual service exposure
conditions [53].
The initial microstructure of 2.25Cr1Mo steel contains numerous, finely
dispersed carbides embedded in the matrix which undergo changes in
composition and shape during service. Therefore, these changes can be used to
form a time-temperature recorder which can be used in remanent life
assessments [54]. The carbide transformation sequences occurring during the
tempering of this steel, proposed by different researchers are very different.
The first and widely referenced work is that by Baker and Nutting [55]. They
and some other researchers [5] argue that the equilibrium carbide in this steel
is M6C and that there are different sequences in either bainite or ferrite for
carbide transformation as follows:
In bainite:
In ferrite:

It should be noted that the ageing temperature and time has a great influence
on these sequences. For example, it is noted that thermal ageing at 850 K to
900 K for durations less than 10,000 hours could not lead the transformation
of metastable carbides to the equilibrium carbide [56].
Further experiments showed that M6C and M23C6 are stable carbides after long
term thermal ageing at elevated temperatures [57]. The sequences of carbide
transformation are shown in Figure 2-4. This diagram shows that increasing
exposure time at temperature above 650˚C results in M23C6 growth at the
expense of cementite and Mo2C in addition to Cr7C3 precipitates. Therefore it
causes a decrease in the number of particles in the microstructure.
The final carbides are shown to be M23C6 and M6C; however, some
researchers have argued that two carbides cannot coexist at equilibrium in the
2.25Cr1Mo steel. Therefore, it is suggested that the equilibrium carbide
should be one of the Mo-rich M6C and the Cr-rich M23C6.
Figure ‎2-4. Summary of carbide transformation sequence [58].

It is found that the nucleation and growth of carbides is sensitive to the
presence of minor alloying additions [59]. Finally, there is some recent
researches which shows that in silicon-containing steels, the transformation
sequence in the bainitic regions is given as [58]:
While in ferrite grains the sequence is given as [43]:
Moreover, carbide transformation sequences have different rates in the various
microstructures. These changes occur at a slower rate in a mixed
microstructure of bainite and allotrimorphic ferrite when compared with a
fully bainitic microstructure [60, 61].
2-3-3- Effects of welding on the 2.25Cr1Mo

microstructure
The low alloy ferritic steels, (especially 2.25Cr1Mo steel) are used in as-
manufactured and as-welded conditions in service. Welding processes alter the
microstructure in the near-weld region and produce a complex microstructure
leading to sharp variations in service behaviour [62]. The weldability of this
steel is good, although reheat cracking and cold cracking in the heat affected
zone (HAZ) and in the fusion zone are sometimes observed [63]. The
difficulties related to welding can be minimized by preheating or by using low
heat inputs, such as multi-pass welding, in order to reduce austenite grain
growth and to increase the amount of fine grained bainitic microstructure [64].

The microstructure of the weld metal and HAZ is dependent on the heat input,
the welding procedure and the post weld heat treatment. It is claimed [64] that
the microstructure of the 2.25Cr1Mo steel weld metal is largely bainitic,
consisting of packets of elongated and outlined platelets. These are interpreted
as individual bainitic ferrite platelets outlined by retained carbon-enriched
austenite. Carbides may possibly also be present between platelets.
Metallographic analysis of the HAZ adjacent to the fusion boundary [63, 65]
reveals a bainitic structure consisting of laths of ferrite with aligned carbide
precipitates between the laths. There is limited information about the types of
carbide in the HAZ of 2.25Cr1Mo steel and it is reported that Mo2C and M6C
carbides are not seen in this region [66]. As the heat input increases, this
microstructure becomes coarser. Pro-eutectoid ferrite was never observed in
this zone. It is reported that the austenite grain size differs across the HAZ;
adjacent to the fusion boundary there is a prior coarse grain region which
changes to a fine grain region as the distance from the fusion boundary
increases [66, 67] .
The number of carbides in the fine-grained zone is higher than in other regions
and these carbides tend to be coarser [63]. Therefore, the microstructural
evolution in these two regions is different during post weld heat treatment
(PWHT) or during service life. This is considered to be consequence of
differences in the austenite grain size, bainite morphology and carbide size
distribution governed by cooling rate and chemical composition. These
differences all affect carbide transformation during PWHT or service life [68].

2-4- Alloy 625, a nickel based super alloy
For evaluation of the weld overlay characteristics, it is necessary to have a
good knowledge of the standard Alloy 625. The solidification behaviour of
Alloy 625 is presented since this is relevant to work addressing a weld deposit
of this material. Moreover, phases which precipitate during high temperature
service are identified and described. This information aids understanding of
the relationship between the phases and their stabilities in the microstructure
during high temperature exposure.
The development of Alloy 625 was started in the 1950s to meet the then-
perceived demand for a high-strength main steam-line piping material [69].
Alloy 625 has been among the most successful nickel based alloys for long
term engineering applications as it is a solid-solution strengthened alloy.
Although the alloy was originally designed as a solid solution strengthened
alloy, it has been shown that some secondary phases form during solidification
[70] or precipitate during ageing treatment or service at temperatures in the
range of 600 to 750˚C [71].
2-4-1- Solidification of Alloy 625 and weld

deposit of Alloy 625
To be termed Alloy 625, the composition has to fall within certain limits
which are mentioned in 2-2-4. However, variations within this range can
affect the microstructure and thus properties. In cases such as hard-facing [72]
and weld cladding [73] where Alloy 625 is used as weld deposit, the
composition will be affected by substrate dilution. It is suggested that
solidification reactions in such diluted compositions will be different to those

of pure Alloy 625 and consequently phase formation processes will differ.
Moreover, solidification during welding is an inherently non-equilibrium
process and the microstructures generated during welding are often not those
predicted by applying equilibrium considerations to existing phase diagrams
[74].
Phase formation during solidification of nickel alloys has been widely
presented in the literature, and provides a foundation for the discussion of
phase formation process during solidification in a weld overlaid Alloy 625 on
the steel substrate. The effects of the following issues on the formation of
these solidified microstructures are then presented.
1) Increase in the cooling rate which cause changes to the equilibrium
condition
2) Changes in the composition due to mixing with steel
2-4-1-1- Solidification of Alloy 625
In order to investigate phase formation during solidification, the sequence of
solidification reactions is of prime importance. Alloy 625 is considered as a
modified version of alloy 718 [71]. This assumption is helpful for
understanding the solidification behaviour of Alloy 625. The dominating
solidification reaction in these nickel alloys is the enrichment of the remaining
interdendritic liquid in niobium as the γ dendrites form from the liquid. This
causes formation of niobium-rich Laves and/or niobium carbide in those
regions during the final stages of solidification [71]. Another way of
illustrating the solidification is a pseudo ternary equilibrium diagram which is

shown in Figure 2-5. This schematic presentation is based on the alloy 718
data [75] and is modified for Alloy 625 [71]. As shown in Figure 2-5, the
C/Nb ratio dictates the solidification path from three available paths and
affects the resultant microstructure. Path 1 is related to the high C/Nb ratios
and leads to the formation of γ+NbC with no Laves phase. Path 2 is for
intermediate C/Nb ratios and firstly forms γ+NbC, followed by Laves phase
formation at the end of solidification. At low C/Nb ratios (path 3) γ+Laves
phase forms with no NbC. Examination of microstructures in typical heats of
Alloy 625 reveal the Path 1 (γ + NbC) or Path 2 (γ + NbC + Laves)
microstructures [71].
Figure ‎2-5. Schematic view of solidification paths in Alloy 625 [71]
The F.C.C. structure of the nickel matrix has a high solubility for many
substitutional alloying elements, but commercial nickel alloys often contain a
large number of minor phase particles such as carbides, nitrides, borides and
intermetallic compounds in their as-solidified condition [76]. This reflects the
fact that the solidification rate and other alloying elements such as N and B
also affect microstructure.

There are a few publications describing phase formation during solidification
and important factors which affect formation of Nb-rich phases in Alloy 625
[70, 71]. Differential thermal analysis (DTA) was performed to establish
approximate equilibrium solidus and liquidus temperatures of Alloy 625 with
different Nb, Si and C contents. These DTA results reveal that the presence of
these elements decreases liquidus temperature from 1406˚C to 1352˚C and
solidus temperature from 1380˚C to 1275˚C which causes an increase of the
melting range (ΔT) and formation of minor phases during the final stages of
solidification.
Microstructural examination of as-solidified Alloy 625 revealed that there are
two different types of elemental segregation during solidification of Alloy
625. The continuous segregation pattern within the primary austenite (γ)
dendrites causes the liquid solubility to be exceeded during the terminal stages
of solidification, promoting another segregation between austenite (γ) and the
various minor phases [70]. Therefore, it is found that solidification of Alloy
625 terminates by a eutectic-like reaction of enriched remaining liquid phase
forming a solid solution alloy (γ) and various niobium-rich phases such as
NbC, M6C and Laves phase, the details of which depend on the chemical
composition. The solidification reaction in Alloy 625 with a levels of Nb=3.53
wt%, C= 0.035 wt%, Si=0.46 wt% is as follows [70]:
The amount of eutectic-type constituents can be calculated, at least semi-
quantitatively, from DTA thermo-grams. Using this technique, it has been
found that the minor phases to matrix ratio is about 3% to 5% and depends on

the carbon and silicon contents [70]. The important factors affecting the final
microstructure are carbon to niobium ratio (C/Nb) as well as iron and silicon
contents. The higher C/Nb ratio results in more carbide while the higher iron
and silicon contents promote more Laves phase in the microstructure.
2-4-1-2- Solidification of weld deposit of Alloy 625
Work by Cieslak [77] describes the solidification of gas tungsten arc (GTA)
weld deposit of Alloy 625 as a function of composition and determined the
fundamental melting/solidification parameters such as liquidus and solidus
temperatures. In his research, he correlated the solidification microstructure
and weldability of Alloy 625 as a function of chemical composition, along
with sequences of solidification reactions leading to the formation of the as-
welded microstructure.
The DTA results obtained from re-melting weld deposits showed that the
important alloying elements (i.e. Nb, C and Si) decreased the liquidus from
1354˚C to 1333˚C and the solidus temperature from 1250˚C to 1158˚C which
caused an increase in the melting range (ΔT). The melting temperature range
was related to the presence of a eutectic-type constituent which increased hot
cracking susceptibility. These possible eutectic phases are identical to those of
the as- solidified Alloy 625 [77], the dominant phase depending on the weld
wire characteristics [78] .
In research relating to Alloy 625 overlay deposits on a nickel-based material
produced by plasma transferred arc welding (PTAW), it is argued that the
C/Nb ratio is the primary factor controlling the secondary phase formation

[79]. As the C/Nb ratio decreased from 0.024 to 0.016, the content of γ/NbC
eutectic constituent also decreased from 1.25 to 0.77 vol%, but the content of
γ/Laves eutectic constituent increased from 1 to 2.5 vol%, resulting in a total
content of eutectic constituents being increased from 2.25 to 3.27 vol% [79].
Formation of the Laves phase increases the tendency for hot cracking during
welding [77] and reduces the ductility in as cast material [71]. Moreover, the
corrosion resistance decreases due to spatial inhomogeneity in the
composition [80].
2-4-1-3- Effect of heat input during welding and
cooling rate
In the industrial application of Alloy 625 as a weld metal, other phases such as
sigma (σ) and P phases appear as a result of non-equilibrium solidification
[81]. The appearance of these phases in the microstructure can be related to
local segregation. During non-equilibrium solidification, the thermodynamic
driving force for formation of these phases is achieved by a sufficient local
segregation.
In a weld overlay structure, heat input during the welding is an important
parameter which can affect the resultant microstructure, and metallographic
examination has indicated that the formation of secondary phases in Alloy 625
weld overlays depends on the heat input [33]. It is claimed that the lower heat
input would result in more rapid cooling and less time being available for
segregation. Rapid cooling would promote a lower degree of segregation of
the alloy constituents, a finer dendrite spacing and finer dispersion of the
terminal phase.

Moreover, it is shown that the phase fraction of secondary phases found in
Alloy 625 weldment decreased from about 3 % to about 1% as the cooling rate
increased from 0.1 to 104 K/s. This can be explained as a consequence of
interaction between solute atoms and the trapped vacancies in the solid [82].
It should be noted that increasing the cooling rate has no effect on the type of
carbides formed, but it is observed that the distribution of carbide particles
changes from relatively few large particles to many fine precipitates as the
cooling rate is increased [83].
2-4-1-4- Effect of dilution
Dilution of the overlaid material with substrate material occurs during welding
processes. This dilution results in a variation of chemical composition in an
overlay and thus there is a variation in the solidification behaviour and
microstructure since these are sensitive to chemical composition. In general,
during welding of an Alloy 625 on a steel substrate, the content of Fe, C and
Si increases in the Alloy 625 overlay composition. It has been shown that
addition of Fe up to 6 wt.% to Alloy 625 did not change the solidification
behaviour. Moreover, the C-enriched alloy solidification behaviour is identical
to pure Alloy 625 [84]. In both of them, molybdenum and niobium have been
shown to exhibit strong tendencies to segregate to interdendritic regions and
their segregation was balanced by inverse segregation of nickel and iron to
dendrites cores [84].
It is believed that the Laves phase fraction increases with an increase in the
content of Fe and Si alloying elements in Alloy 625. The formation of Laves
phase and NbC particles in the interdendritic regions decreases the toughness

of weld metal. The impact energy at room temperature of the weld metal made
with a standard (Fe = 3.19 wt%, Si = 0.22 wt%) filler wire is about 42.3 J.
This value increased to about 152.7 J in a weld made with the low Fe and Si
wire (Fe = 0.80 wt.%, Si = 0.08 wt.%) [71]. In addition, it has been shown that
addition of iron caused alteration in the Laves phase composition. The more
iron in the Alloy 625 (increasing from 2 wt% to 6 wt%), the larger the iron
concentration in the Laves phase (increasing form ~1 wt% to 4 wt% ) in the
as-solidified sample [84].
2-4-2- Phase precipitation during high

temperature exposure in Alloy 625
The main aim of this section is to describe phase changes that occur at high
temperature in Alloy 625. It is clear that solid-state transformation or phase
precipitation during high temperature exposure relies on the initial
microstructure. Therefore, in order to understand the precipitation of
secondary phases in a weld microstructure of Alloy 625 during high
temperature exposure, the available information concerning precipitation
processes in the wrought Alloy 625 can be used as a guideline.
Alloy 625 was not originally developed as a precipitation hardened alloy [85]
but in the 1970‟s, a long time ageing study showed that there are various
phase transformations that can take place in this material as a function of time
and temperature [86]. It should be noted that secondary phases which are
precipitated during production processes can significantly influence the
microstructure of the wrought Alloy 625 products [87]. Different carbides and
inter-metallic compounds can precipitate in Alloy 625 after thermal exposures

at temperatures in the range of 600 - 1000˚C. Due to its high content of
molybdenum, chromium and niobium, Alloy 625 exhibits complex
precipitation behaviour [88]. The precipitation behaviour in the wrought alloy
is also dependent upon the carbon content and final annealing treatment [89].
A large amount of research has been carried out to correlate the properties and
precipitation behaviour of wrought Alloy 625 [71, 88-93]. In order to simplify
this, the following categories can be considered:
1) Precipitation of hardening phases
2) Carbides and carbo-nitrides
3) Topologically closed-packed (TCP) phases
2-4-2-1- Precipitation Hardening Phases
The efforts for identification of these phases in Alloy 625 go back to early
stages of its application. Comprelli and Wolff [94] noted that in Alloy 625
material and other researchers found in weld metal of Alloy 625 [95] , there is
an increase in hardness (from ~270 to ~370 Vickers) after ageing at 700˚C.
Moreover, tensile strength increases (form ~880 to ~1000 N.mm-2) and a
significant decrease of elongation (form 32 % to 20 %) occurred after ageing
at 700˚C.It was assumed that this was due to the precipitation of a coherent
phase (γ‟) with formula [Ni3(Nb, Al, Ti)] as seen in other nickel alloys [96].
However, research work of Bohm [97] showed that the increase in the
hardness at temperatures above 750˚C is caused by the precipitation of a stable
incoherent orthorhombic phase with formula [Ni3(Nb,Mo)].

The long term ageing study was carried out and the mechanical properties of
Alloy 625 were determined for samples exposed 8,000 hours at 649˚C, 760˚C
and 871˚C [86]. It was found that an increase in tensile strength was observed
after exposure at 649˚C (812.2 Mpa compared to 456.4 Mpa in mill annealed
condition) but that tensile strength dropped significantly to about 671.6 Mpa
in samples aged at 760˚C and further decreasing happen to about 439.2 Mpa at
871˚C. The reported X-ray diffraction results of this study showed that two
M6C phases and a Ni3Nb phase were present in the material aged at 871˚C, but
the increased strength at 649˚C was not explained by the X-ray data [87].
Further investigations in this regard showed that strengthening at intermediate
temperatures (between 600 to 700˚C) in Alloy 625 occurred by precipitation
of the metastable γ” phase. This phase has a body centred tetragonal D022
structure while the equilibrium intermetallic phase is the δ phase with D0a
structure. It is believed that the metastable γ” phase [Ni3(Nb,X)] forms on
ageing at temperatures above 600˚C and transforms to the orthorhombic δ
phase [Ni3(Nb,Mo)] upon prolonged ageing[98].
Figure ‎2-6. Time-Temperature-Precipitation (TTP) diagram corresponding to γ”‎and‎δ‎

precipitation in Alloy 625 [98].

Moreover, it was shown that δ phase [Ni3(Nb,Mo)] formation is possible
directly from the supersaturated solid solution at higher temperatures [99]. It is
claimed that the equilibrium phase (δ) often nucleates at grain boundaries
during thermal exposure while dislocations act as effective nucleation sites for
precipitation of the γ‟‟ phase [98, 100]. Time-temperature-precipitation (TTP)
curves corresponding to the formation of these phases are presented in Figure
2-6, while their micrographs are shown in Figure 2-7.
Figure ‎2-7. Different precipitates found in Alloy 625; (a)‎ δ‎ precipitates‎ formed‎ after‎
ageing at‎820˚C‎for‎200‎hours, (b) plates of δ‎phase growing from a NbC precipitate, and
(c)‎γ’’‎precipitates‎formed‎after‎ageing‎at‎675˚C‎for‎100‎hours [93].
2-4-2-2- Carbides and carbo-nitrides
NbC carbides with B1 structure [82] form in this alloy during the final stages
of solidification; these are generally termed primary NbC particles [71]. In the
solution annealed condition (above 1100˚C), all the phases, except primary
NbC, have been put into solid solution and further carbide precipitation in
Alloy 625 during ageing has been observed from 650˚C onwards [101]. It has
been reported that three different carbides i.e. MC, M6C and M23C6 are present
in the temperature range from as low as ~600˚C to as high as ~1050˚C[102].

Others have argued that the temperature range is more limited being in the
range of 760-980˚C [103].
The type of carbide formed during heat treatment depends upon the
temperature. At higher temperatures, roughly from 870˚ to 1040˚C, the
carbides are both MC (NbC) as thin grain boundaries films, and M6C, where
M is principally Ni, Cr, and Mo. At temperatures in the 700˚C - 870˚C range,
the grain boundary carbides are primarily M23C6 where M is almost entirely
Cr. After intermediate temperature heat treatments, all three carbides can
usually be found [103, 104].
However, Muzyka [105] has reported that the primary MC carbides
decompose during annealing on prolonged exposure at elevated temperatures
into M6C and M23C6 [106]. Many researchers report precipitation of all
carbides on grain boundaries [71]. It should be noted that the chemical
composition of MC with intergranular precipitation is different from
intragranular primary MC precipitates with irregular shape [107, 108]. The
lattice parameter of the primary MC carbide was found to lie between the
reported lattice parameter of NbC and TiC phases [101]. The thin film on the
grain boundaries is reported to be mainly NbC [71]. Two different
morphologies for intergranular MC have been reported, namely the diamond
shape and dendritic form [93].
The M6C and M23C6 carbides commonly have blocky, irregular shapes and
form as series of separate, discrete grain boundary particles. Figure 2-8 shows
TEM micrographs of these types of carbides. The MC carbide formed in the
early stages of heat treatment (after 1 hour) at 900˚C while the M23C6 carbide

was formed after a prolonged heat treatment (100 hours) at lower temperature
(800˚C). The M6C carbide precipitates after 1 hour at higher temperature
(1025˚C). The stability of different carbides in solutionized Alloy 625 forged
bar has been reported by Vernotloier and Cortial [109, 110] who mentioned
that fine M23C6 carbides precipitated along austenite grain boundaries in a
sample heat treated for 8 h at 700˚C, whereas a sample heat treated for 8 h at
800˚C contained M6C carbide, and the sample heat treated at 1050˚C
contained only MC carbide. The precipitation of the M6C carbide at 800˚C
caused degradation of mechanical properties (reduction of yield stress ~150
N.mm-2) [109] and corrosion resistance (3 times) [110] of Alloy 625.
Figure ‎2-8. Images of different intergaranular carbides;‎(a)‎dendritic‎MC‎(900˚C,‎1‎h),‎

(b)‎ diamond‎ shaped‎ MC‎ (900˚C,‎ 1‎ h),‎ (c)‎M 23C6 blocky‎ shape‎ (800˚C,‎ 100 h), (d) M6C
needle‎shape‎(1050˚C,‎1h)‎[93].

Moreover, slight changes in the chemical composition resulted in different
type of carbides [111]. Table 1 summarizes the crystal structures and typical
compositions of the carbides types which found to be present in the
microstructure of wrought Alloy 625 [71].
Table 1.Structures and typical compositions of the carbide precipitates [71].
Phase Structure Location Typical Composition
(Ti0.07 Cr0.04 Fe0.02 Ni0.09 Nb0.75 Mo0.03 )C

Cubic , Fm3m
Matrix
MC a0= variable
(Ti0.53 Cr0.03 Ni0.04 Nb0.39 Mo0.01 )C
Grain
a0=0.43nm (Ti0.15 Cr0.04 Fe0.01 Ni0.12 Nb0.67 Mo0.01 )C
Boundaries
Cubic , Fd3m Grain
M 6 C (Cr0.21 Fe0.02 Ni0.37 Nb0.08 Mo0.24 Si0.08 )6 C
a0=1.13nm Boundaries
Cubic , Fm3m Grain
M 23 C6 (Cr0.85 Fe0.01 Ni0.07 Mo0.07 )23 C6
a0=1.08nm Boundaries
2-4-2-3- Topologically Closed Packed (TCP)
phases
The last group of precipitates are intermetallic compounds commonly known
as topologically-close-packed (TCP) phases because of their characteristic
crystal structures [91, 92]. These can be listed as follows:
1) Laves Phase
2) Sigma (σ) Phase
3) Mu (μ) Phase
4) Ni2M Phase
Laves phase: Laves phase is a hexagonal A2B compound that can be found in
number of binary alloys [112]. In more complex alloys, Laves phase can
contain significant amounts of other alloying or impurity elements, and depart
considerably in composition from a simple A2B type chemistry. In Alloy 625,

the Laves phase has noticeable compositional variations that reflect the
differences in processing history of the alloy in which the particles were
formed [71]. Irrespective of processing, Laves phase is generally found to be
significantly enriched in Nb (12 to 27 wt%), Mo (12 to 23 wt%), and Si (2 to 9
wt%) while the Fe, Cr levels are 3 to 5 wt% and 13 to 22 wt% respectively
which are not noticeably different compared to the nominal Alloy 625
composition [33, 77].
Sigma‎ (σ)‎ phase: this phase typically appears in Fe-Cr alloys that contain
more than 20 wt% of Cr, and in these alloys and it has the approximate
composition of FeCr. Sigma (σ) phase has a complex, tetragonal layered
structure and tends to nucleate from M23C6 carbides to which it is structurally
and compositionally related. Although sigma (σ) phase formation is not
common in Ni-Cr systems, it can form in nickel based alloys that contain
about 10 wt% Mo and more than 15 wt% Fe and other alloying elements that
tend to stabilize sigma (σ) phase including Si and W [113].
Mu‎ (μ)‎ phase: Some researchers [92] believe that the Mu (μ) phase has a
A6B7 type composition where A is Mo and/or W and B is Fe and/or Co. It has
a hexagonal close-packed structure and D85 crystal system. Others claim that
Mu (μ) phase has a complex, rhombohedral structure, which occurs in Cr
containing alloys with large amounts of Mo and W and stabilized by Fe [113].
Mu (μ) phase particles tend to nucleate on M6C carbides which are structurally
and compositionally (Mo-rich) similar. For a certain Ni-Cr-Mo-Fe alloy, the
Mu (μ) phase is reported to have a composition as follows [92]:
(Ni0.36Cr0.16Fe0.04Co0.02) (Mo0.39W0.03)

Ni2M: This intermetallic compound with Pt2Mo-type structure is reported to
precipitate in Alloy 625 after prolonged service at temperatures close to but
less than 600˚C [114, 115]. This phase is known as Ni2M or Ni2 (Cr, Mo) and
has six variants and precipitates in the austenite grains. These particles exhibit
a snowflake-like morphology and are uniformly distributed in the matrix
[106]. There are other minor phases in the Alloy 625 matrix [116]; however,
due to their negligible fractions, further discussion is not necessary for the
purpose of this work.
2-4-3- TTT diagrams for wrought Alloy 625
It has been shown that different phases precipitate in Alloy 625 depending
upon temperature and duration of ageing [117] and TTT diagram is a helpful
tool in analysing of the precipitation behaviour in this alloy. Several TTT
diagrams have been presented for Alloy 625 which differ in certain aspects
from each other. There are several possible reasons for these differences; the
first is variation in composition within the ranges given for Alloy 625. The
second source of variability is related to the techniques which are used to
identify the phase transformations.
With these facts in mind, a TTT diagram for commercial Alloy 625 was first
published by E. Schanbel and co-workers [88] and is presented in Figure2-9.
It is clear that heat treatment of the material during its processing can
significantly change the TTT diagram; for example by solution annealing, the
precipitation behaviour of Alloy 625 is changed. Crum and co-workers [88]
have published a revised TTT diagram which is shown in Figure2-10.

Figure‎2-9. Time-Temperature-Transformation diagram for Alloy 625 [88].
Figure‎2-10. TTT diagram for solution treated Alloy 625 [88].
Figure ‎2-11. Approximate TTT diagrams for phase formation at high temperatures in
nominal Alloy 625 [71].

Finally, Floreen [71] presented an approximate TTT diagram for nominal (as
defined by Floreen) heats of wrought Alloy 625 (see Figure 2-11). Carbo-
nitrides are not indicated on the Floreen diagram due to wide difference in the
composition of nominal heats; however, the main intermetallic phase i.e.
Laves phase is identified in this diagram. Although the precipitation of δ and
γ‟‟ are similar in all of these diagrams, carbide precipitation curves are
different in terms of temperature and time. Additionally, the TTT diagram for
the specific solution treated alloy (see Figure2-10) contains different variants
of M6C carbide while there is only one variant in the diagram shown in Figure
2-11; these variations are related to differences in the chemical composition.
In addition to these differences, the Ni2M phase, which was experimentally
detected in Alloy 625, was not present in the TTT diagrams referred to in the
literature [118]. Additionally, Figure2-9 and 2-10 have features inconsistent
with the basic principles of TTT diagrams.
2-4-4- Precipitation behaviour of weld deposit
Thermal exposure also can markedly change the microstructure of a weld
deposit. However, there are issues in the application of TTT diagrams for
prediction of precipitation behaviour of Alloy 625 with an as-welded
microstructure. For example, even if care is taken to avoid Laves phase in the
as-welded structure, the enrichment of Nb in the interdendritic regions will
lead to accelerated precipitation reactions in the ageing experiments.
Figure 2-12 shows an example of a TTT diagram for Delta (δ) precipitation in
GTA deposited Alloy 625 weld metal compared to wrought Alloy 625 [71].
As seen, the time for Delta (δ) formation was reduced by a factor of 50

compared to that of the wrought material. It is reported by Cortial [95, 119]
that in the rough-welded condition with TIG, the minor phases are present in
the interdendritic regions, which is normal for any metal in the as-solidified
state. Different interdendritic phases have been identified which are Laves
phases, MC and M6C carbides. This indicates that the M6C carbide was
formed during the non-equilibrium conditions associated with welding. It is
reported that the corrosion resistance of the as-solidified IN625 is reduced due
to several microstructural changes which occur during rapid solidification.
Phase formation during the re-solidification of laser melted surface and/or
solidification of the weld overlay has been shown to be a source of high
corrosion rate under severe conditions [26, 80].
Figure ‎2-12. TTT diagram showing thermal exposure conditions for formation of Delta
in weld metal compared to wrought Alloy 625 [71].
2-5- Cladding and dissimilar metal welding
There is a large body of research addressing dissimilar welding and their
results may be relevant to this work which is concerning microstructural
evolution in a weld overlay/substrate system. The effects of the welding
process and inter-diffusion of elements across the interface during post weld

heat treatment (PWHT) and service can be used as a guide for understanding
of the weld overlay/steel substrate system.
Dissimilar metal welds are widely used in high temperature applications in
industrial plants and are generally fusion welded. There are different types of
dissimilar weld joints such as joints between steels of different phase type
(austenitic-ferritic joints) and/or steels of differing substitutional alloy content
(primarily ferritic-ferritic joints). Moreover, there are weld joints between
different types of steel and nickel alloys. The weldability, compatibility, and
long term microstructural and mechanical stability of the welded joints, are of
the highest importance in the analysis of dissimilar weld joints.
2-5-1- Interface and its different regions
When a dissimilar metals joint enters service at high temperatures,
microstructural changes and elemental redistribution may occur with these
changes being most marked close to the fusion boundary. The weld interface
of the weldment is classically defined as the interface between the liquid metal
and the heat affected zone.
However, microstructural examination on a finer scale reveals a more
complicated metallurgical condition at the fusion boundary. Savage [120] has
presented a schematic illustration of the various regions of the weld which is
shown in Figure 2-13. As presented, there are two additional zones between
the melted weld metal and heat affected zone. These are the unmixed zone and
the partially melted zone, with the apparent interface being between these two
regions.

Figure ‎2-13. Schematic illustration showing different regions of a heterogeneous weld

and the proposed terminology of these regions [120].
Usually, all the material in the so called HAZ is regarded as remaining solid
right up to the boundary of the fully molten region. However, metallographic
examination at high magnifications reveals significant localized melting
within this zone near to the fusion boundary. It has become apparent,
therefore, that the definition of the fusion boundary is dependent upon both
the magnification and metallographic technique used to examine this region.
Figure ‎2-14. Showing photomicrographs of the welding interface indicates identification

of unmixed zone with elemental profile [121].

In dissimilar weld joints, an important issue is the difference in chemical
composition of the materials which are being welded together. The interface
can be identified by the point where the elemental profiles start to change such
as shown in Figure 2-14. As seen, a clear layer of unmixed zone exists
between the weld metal and substrate. However, there are different materials
in this region and therefore this zone is actually the inter-mixing zone.
2-5-2- Interface and Diffusion
Dissimilar joints are prone to migration of alloying elements across the
interface during their service at elevated temperature. This occurs owing to
wide difference between chemical compositions of materials. The main factors
that influence the stability of the welded joint are the phase nucleation,
element diffusion, and the rate of phase transformation. These factors depend
significantly on temperature and the diffusion path which is related to the
structure of the interface; secondary phases have been found to nucleate in an
area near to the interface.
The microstructural transformation in a treated weld joint is usually a
complicated issue when diffusion between two different materials occurs at
the interface. In such cases, several mutually influencing phase
transformations can often be observed. The rate of phase transformation is
dependent on the temperature and chemical composition of the materials.
Changes in the chemical potentials of the elements can cause phase
precipitation, phase dissolution, and/or interface replacement in the diffusion
zone along the welded interface [122, 123].

2-6- Microstructural evolution in dissimilar weld

joints
The mechanical performance of dissimilar weld joints is dependent on the
microstructural changes taking place in the interface region at high
temperature. Owing to differences between the two materials, transport of
elements across the interface occurs, which could potentially result in either
dissolution or precipitation of phases such as intermetallics and carbides. The
formation of phases is dependent on the heat treatments which are applied
during welding (including post weld heat treatment) as well as the service
condition.
2-6-1- Different Cr-content steels and
formation of soft and hard zones
In all dissimilar steel joints, it is known that carbon migration across the weld
interface occurs from the low-alloy steel to the more highly alloyed austenitic
[123-128] or ferritic steels [67, 129-135]. This happens generally during post
weld heat treatment (PWHT) or subsequent service at elevated temperature.
As a result of this migration, a carbon depleted zone (soft zone) is formed in
the heat affected zone (HAZ) of the low alloy steel and a carbon enriched zone
(hard zone) is formed in the highly alloyed steel [128, 133, 136]. Since the
solubility of carbon in ferrite is extremely low, the partitioning of carbon
involves the dissolution of carbides on the low alloy side of the weld, and the
precipitation of carbides on the high alloy side of the weld. The precipitation
of carbides at temperatures where the joints enter service begins with the

formation of cementite, followed by a number of alloy carbides as the system
tends towards equilibrium [131, 137-139].
2-6-2- Steel/nickel alloys and formation of

carbide arrays
There is a wide body of research regarding the development of the interfacial
structure in steel/nickel alloy weld joints [6, 140-147]. The low Cr ferritic
steels include 2.25Cr1Mo, AISI 4130 and T11 while nickel alloys include
Inconel 625, Inconel 182 (Ni-15% Cr-10% Fe-8% Mn-2% Nb), Hastelloy G
(Ni-22% Cr-20% Fe-7% Mo-2% Co-2% Cu-2% W) and Alloy 800 (Fe-30%
Ni-20% Cr-1% Mn-0.5% Ti).
Moreover, other different types of steels including stainless steels such as
SS310 [141, 145, 147] and high Cr ferritic steels such as 9Cr1Mo [146, 148]
are welded to nickel alloys. These dissimilar weld joints are used very
frequently in the welding of power plant components, since it is claimed that
failures which occur by low ductility circumferential cracking in Cr steel weld
joints can be resolved by using nickel alloys as the weld metal [143].
However, the joint made with a nickel-based weld is not immune to failure. It
is claimed that nickel-based joints after failure or long-term service exposure
contain arrays of particles, mainly M23C6 and M6C in the ferritic steel, parallel
and very close (~1µm) to the weld interface. The precipitates show two
general morphologies; the most frequent is a single plane of precipitates with
mean size >0.5 μm, and the other is a wide band of finer precipitates (>100
nm). These morphologies are classified as types I and II respectively [6, 140].

It is reported that during welding of 2.25Cr1Mo steel with different nickel
alloy fillers, a band of martensite ~0.5 to 50 μm wide forms between the steel
and the weld metal due to dilution of steel into the weld metal [6, 143]. The
interface between the steel and the martensite band acts as a site for
precipitation of coarse carbides during ageing, and a line of these particles
grow on the narrow martensite band. It is believed that microscopic cracking
is generally associated with these precipitates [6].
These precipitates develop during long-term high temperature service
exposure and in the particular system (2.25Cr1Mo steel/ Inconel 182) it is
reported that interfacial precipitation can be accelerated by laboratory ageing
at 630-700˚C and precipitates and hardness profile in a sample aged at 630˚C
for 6000 h are similar to those after long term service [6]. However, above
657˚C, extended laboratory ageing on multi-bead weld samples can cause
interfacial precipitate dissolution and gross decarburization of some regions in
the weld metal, which are not typical features in the ex-service joints [140].
2-6-3- Phase transformation at 2.25Cr1Mo

steel and Alloy 625 weld overlay
There are two publications which address the 2.25Cr1Mo steel substrate and
Alloy 625 weld overlay system [149, 150]. One of them addresses phase
transformation at the interfacial region and evaluation of microstructural
characteristics which could affect the mechanical properties and corrosion
resistance of weld overlay. In this research, the interface microstructure of
different cladding methods were evaluated [149]. The other paper relates to
dilution of steel in the weld overlay and its subsequent solidification

behaviour [150]. There are some differences in the nature of the systems
examined, including the iron content of the weld overlay and the interdiffusion
structure and width. Moreover, different secondary phases in the weld overlay
were reported.
Ayer [149] reported that SIMS images of the interface between 4130 steel and
Alloy 625 for chromium and nickel exhibited a sharp change at the interface.
However, the iron image indicated a somewhat greater movement of iron into
the Alloy 625 cladding which is about 20 μm. He found that these images are
similar for 2.25Cr1Mo steel/IN625. The overall composition measured by
EDS in a TEM with a 40 nm microprobe demonstrated that transition occurred
over approximately a 5 μm distance. The composition of the ferritic steel was
modified only by moving ~1 μm away from the interface while the
composition in the Alloy 625 side was modified to less than 10 wt.% Fe over a
4 μm distance.
Nevertheless, DuPont [150] has found that the process of weld cladding can
be visualized to consist of the melting of a small region of the steel surface by
the molten Alloy 625 weld pool. The elements from this melted region of the
steel are homogenized in the liquid phase which provides a greater
intermixing and minimizes the gradient of compositional profile. The
transition region was seen to be ≈20 μm and beyond this transition region
there is no systematic variation in composition with distance. Thus, the weld
metal composition was shown to be homogeneous on a macroscopic scale
with 27.2 ± 5.8 wt.% Fe content.

These researchers have reported different secondary phases in the weld
overlay material which may be associated with the differences in compostion
and welding process. Ayer [149] reported that grain boundary precipitates
were either M6C or NbC carbides. The stoichiometry of M6C carbides was
determined to be as follows:
(Mo0.4 Ni0.36 Cr0.16 Fe0.08)6 C
In contrast, DuPont [150] has reported that secondary phase formation and
microsegregation behaviour of the main alloying elements (Fe, Ni, Cr, Mo and
Nb) are very similar to those for Alloy 718. The Laves phase formed in a
eutectic-type morphology with γ at final stages of solidification as follows:
Ayer [149] has reported that, as in conventional welding, HAZ was observed
in the steel and the grain size of the steel had been modified over a major
portion of the HAZ. Some interesting microstructural features were also
observed at the interfacial region. At the interface between the FCC austenite
and BCC ferrite regions, a thin zone of un-tempered martensite was observed.
The thickness of the martensite region varied in the range of 0.2 to 0.5 μm. In
addition to martensite, precipitation of particles was also observed at the
boundary between the un-tempered martensite and ferrite. The particles were
identified to be M23C6 carbides.
Formation of these carbides is dependent upon the carbon content of this
region. The ferrite is changed to austenite during the welding and the carbon
solubility is enhanced. Upon cooling the cladding to room temperature, a

small region of the steel in the interface region is transformed to austenite or
austenite + M23C6 carbide region. This reversion is the result of the presence
of alloying elements in the area near to interface (such as nickel) which
stabilise austenite to lower temperatures.
2-7- Evaluation of microstructural evolution

using thermodynamic calculation
The understanding of microstructural evolution in a dissimilar weld joint
between nickel alloy and steel system is based on the depth of knowledge of
the phase equilibria of both materials in this system. An accurate knowledge
of phase fields and transformation temperatures in the steel and the nickel
alloy is therefore vital for successful studies of microstructural evolution
during thermal exposure. As both of the materials are considered to be
multicomponent, so the thermodynamic calculations in each of them would be
an important approach in understanding the changes in the microstructure of
the system.
The CALPHAD (CALculation PHAse Diagram) method has proved to be an
efficient tool for solving various problems in the field of materials research
and thermochemistry [151]. In recent years, significant progress has been
made in the application of thermodynamic methods to the calculation of
multicomponent phase diagrams in steels [152-154], nickel based superalloys
[155-158] and other Ni-Cr-Mo alloys [159]. An important reason for
employing thermodynamic methods is that they offer a means of systematising
and modelling the behaviour of one interesting phase such as a carbide in
complex systems like nickel superalloys. This allows the phase relations of

multi-component, multiphase systems to be modelled from those of their
subsystems.
Various computer-based thermodynamic models such as Thermo-Calc [160]
(used during this research) and MTDATA [58, 161] are commercially
available for the calculation of multiphase, multicomponent equilibria as a
function of pressure and temperature in conjunction with appropriate
thermodynamic property databases. The thermodynamic databases contain
assessed Gibbs energy parameters for the thermodynamic models for
individual phases. The databases are then linked to software for the calculation
of phase equilibria for the practical applications of interest. The success of the
CALPHAD technique depends upon the reliability of these databases.
2-7-1- Thermo-Calc software
Thermo-Calc is a software and database package for all kinds of phase
equilibrium, phase diagram and phase transformation calculations and
thermodynamic assessments. Moreover, many types of process simulations
can be performed using this programme. It has been developed for complex
heterogeneous interaction systems with strongly non-ideal solution phases and
can be applied to any thermodynamic system in different fields like
metallurgy, alloy development, semiconductors etc. depending on the kind of
database it is connected to [160].
The underlying methodology of this programme is usually a Gibbs free energy
minimization process and Thermo-Calc makes use of a very general algorithm
to find the equilibrium state of a system. The development of Thermo-Calc
started at the Royal Institute of Technology in Stockholm [162] . This

technique allows much more flexibility for the user setting the external
conditions for the equilibrium state than any other thermodynamic software.
For example, Thermo-Calc is the only software that allows explicit conditions
on individual phase compositions or configuration whereas most software can
handle conditions on the overall composition only. For example, activities and
chemical potentials of the components, volumes, enthalpies, entropies etc. can
also be set as conditions. Many quantities that must be calculated by so called
“target calculations”, i.e. a double minimization procedure, with other
software can be calculated directly and much faster with Thermo-Calc.
The flexible way to set conditions is particularly useful when Thermo-Calc is
used as a subroutine package in application programs, for example in
microstructure evolution or process simulations. For phase diagram
calculations, the facility that each condition is set separately and that any
condition can be used as axis variable means that Thermo-Calc can calculate
innumerable types of diagrams. Thermo-Calc was designed originally for
multi-component systems and all its facilities are available for systems with
up to 40 components [163].
2-7-2- Modelling of microstructure evolution
in Ni-based alloys
The design of new or modified nickel based superalloys for high temperature
application is an extremely expensive and time consuming process. Therefore
thermodynamic calculations are being introduced in attempt to reduce these
costs [164]. Thermodynamic simulation using a CALPHAD software like

Thermo-Calc can bring some useful information to aid understanding of the
phase transformations in complex multicomponent systems of the nickel based
alloys [165]. However, little experimental data are available on the
effectiveness of these predictive tools in these systems.
In the case of Ni-based superalloys, interesting data are the liquidus, the
solidus, the hardening phase (γ‟ and/or γ‟‟) solvus i.e. temperature above
which no hardening phase exists at equilibrium, and the temperature range for
the existence of minor phases like carbides, carbo-nitrides, borides and TCP
phases. Amounts and compositions of such phases are also required to be
computed. Some of these data which are obtained from the equilibrium phase
diagram for Inconel 625 are presented elsewhere [142].
2-7-3- Prediction of microstructural states in
Cr-Mo steels
The microstructural changes in the Cr-Mo steels, especially in the dissimilar
weld joints, can affect the service life of the component or the properties of
the joint. Thus it is of great practical importance to be able to predict structural
changes that occur during the thermal histories associated with production,
post weld heat treatment and service. This problem has been tackled both
experimentally and by application of numerical calculation. Extensive
literature is available for the microstructure – property correlations and
identification of carbides in a variety of Cr-Mo steels. However, there are a
few works which are related to thermodynamic properties of these Cr-Mo
steels [166, 167] and 2.25Cr1Mo steel [168].

There are some models developed by Bhadeshia for carbide transformation
sequences in power plants steels [169, 170] which are dependent upon the
accurate thermodynamic data of the stable carbides in these steels. Moreover,
there is a paper which uses phase diagrams to predict the stable phases in the
microstructure of these steels [153].
2-7-4- Thermodynamic aspects of carbon
migration in dissimilar weld joints
An interesting feature in dissimilar weld joints is the carbon migration. It is
important to predict the microstructural changes in these joints during thermal
histories in production and service. Computer simulation of diffusion in the
multiphase systems has been thoroughly investigated [152, 171, 172]. In order
to perform accurate simulations, one needs an accurate thermodynamic
description of different materials in the dissimilar weld joint system which is
available within the thermodynamic softwares such as Thermo-Calc system.
2-8- Summary
There is a large body of work on the microstructural evolution in the
2.25Cr1Mo steel and Alloy 625 at service temperatures. Moreover, the
microstructural evolution in dissimilar weld joints between steels and nickel-
based alloys are available. In the current work, the microstructure of the
various regions of a Ni-based weld overlay on a 2.25Cr1Mo steel substrate in
as-welded condition will be characterised.

High temperature exposure tests will be used to produce a set of simulated
service microstructures to allow characterization of the microstructural
changes in the different regions with ageing time. This work will examine the
microstructural changes that occur in the 2.25Cr1Mo steel and the weld
overlay material system and the reciprocal effects on each material due to
carbon migration.
Moreover, thermodynamic modelling work has been carried out for the
simulation of phase formation during welding and at certain temperature with
different carbon contents in each material of this system. This information can
then be used as a guide to the long term service performance of weld overlaid
components.

Chapter3 … Experimental procedure
Chapter 3
3- Experimental procedure
3-1- Introduction
The experimental procedure used for the performance of high temperature
exposure tests and the microstructural characterization of different samples
presented in this thesis will be discussed in this chapter. The first sections
concern the starting materials and experimental procedure for thermal
exposure tests. The next sections cover the sample preparation for the various
chemical and surface techniques used to examine the changes in the
microstructure before and after thermal exposure tests. Methods of
microstructural examination and characterization are then briefly reviewed in
the following sections.

3-2- Materials
The starting material in this work was a weld overlaid tube with 25 mm
outside diameter as shown in Figure 3-1(a). The tube consists of a weld
overlay layer with approximately 1.7 to 2 mm thickness which is applied by a
proprietary spiral welding method on the outer bore of a steel tube with 15
mm internal diameter and about 3.5 mm wall thickness. This tube is designed
to be used in a boiler super heater tube in a WtE plant. A wrought IN625 plate
was used to compare the characteristics of the weld overlay with Alloy 625.
Figure ‎3-1. Showing (a) the weld overlaid tube and weld bead specifications, (b) the
whole cross sectional sample of the tube.

The tube was nominally a 2.25Cr1Mo steel and the weld overlay was a nickel-
based alloy similar to IN625. The typical composition ranges of these
materials are given in Table 3-1.
Table ‎3-1.Nominal chemical composition of 2.25Cr1Mo steel and IN 625 alloy (wt.%)
Element Steel 2.25Cr1Mo IN625
C <0.1 <0.1
Al - <0.4
Si <0.5 <0.5
Ti - <0.4
Cr 2-2.5 20-23
Mn <0.5 <0.5
Fe Balance <5
Ni - Balance
Nb - 3.15-4.15
Mo 0.9-1.2 8-10
3-3- High temperature exposure tests and
thermal exposed samples
In order to simulate the effects of high temperature service conditions on this
overlaid tube, individual samples were produced by cutting sections of tube.
The sections were cut with about 7 to 10 mm thickness and each section was
subsequently cut into quarters, yielding 4 specimens. Figure 3-2(a) shows a
typical cross sectional samples used in this work. The specimen dimensions
are suitable for mounting and preparation processes.
The aim of this research is to analyse the evolution of microstructure during
high temperature exposure. Referring to previous research, the microstructure
of 2.25Cr1Mo ferritic steel undergoes phase changes above 675˚C which are

undesirable for service condition [6, 140]. Therefore, specimens were kept
below this temperature. They were heated in a furnace at 650°C for high
temperature exposure experiments for various times, from 1 day to 128 days
to ensure maximum rates of phase transformations which are applicable for
service conditions.
The sample identification represents its exposure time in days and therefore
includes “D” e.g. D128 represents a sample which was at elevated temperature
for 128 days. All specimens were air cooled after their exposure time. Due to
the oxidation of the steel occurring during the high temperature exposure tests
in the air, for long term experiments i.e. 64 and 128 day tests, these samples
were sealed in evacuated quartz tubes at a residual pressure of 10 -5 torr before
being placed in the furnace and were subsequently air cooled to room
temperature. As the main aim was to analyse the phase transformation and
precipitation behaviour at this temperature, the use of a vacuum will not affect
on the behaviour.
It should be mentioned that for some conditions such as the as-welded and the
128 day conditions, the longitudinally sectioned samples (see Figure 3-2(b))
were examined. These samples were used for analysis of the microstructral
features along the length of tube. Moreover, in order to investigate how the
nickel alloy microstructure behaves in the absence of the steel substrate, one
tube section with about 10 mm length was cut and its steel substrate was
removed by machining which is shown in Figure 3-2(c). The thermal effect of
machining on the weld overlay material will be negligible after 128 days
exposure.

Figure ‎3-2. Showing (a) the typical cross sectional sample used for high temperature
exposure tests, (b) the longitudinal section of the tube used for the analysis of the
interface in the as-welded condition and after 128 days, and (c) the free standing of weld
overlay (WD128) sample after high temperature exposure and mounting.

Moreover, hardening due to machining should be relieved in the first hours of
thermal exposure. It should be noted that three different sets of high
temperature exposure tests were carried out in this work which are reported in
Table 3-2. In the first set, the exposure time started at 1 day giving the D1
sample. Afterwards, the exposure time was increased by a factor of 2 for the
next exposure test, up to a maximum of 128 days.
Based on the results of the first set and in order to obtain a more detailed
understanding of microstructural changes during high temperature exposure, a
second set of tests were undertaken. In this set, the first sample was exposed
for 12 days and the exposure time was increased by four days for the next test;
the final sample in the second set has been exposed for 32 days. Because of
performing two sets of exposure tests, two samples i.e. D16 and D32 were
produced twice, providing a measure of the reproducibility of the tests.
Table ‎3-2: Sample identification including sample name, its quantity and the exposure
time‎in‎hours‎at‎650˚C
Sample Name Number of samples in each set Time(hours)

set1 set2 Set3
As-welded 1 0 0 0
D1 1 0 0 24
D2 1 0 0 48
D4 1 0 0 96
D8 1 0 0 192
D12 0 1 0 288
D16 1 1 0 384
D20 0 1 0 480
D24 0 1 0 576
D28 0 1 0 672
D32 1 1 0 768
D64 1 0 0 1536
D128 1 0 0 3072
WD128 0 0 1 3072

In the third set of tests, only the free standing weld overlay sample was
exposed. In order to compare its precipitation behaviour with the D128
sample, this sample was exposed for 128 days at 650˚C in the air, as there is
no concern about oxidation and is termed WD128 sample. Moreover, the
furnace ambient has no effect on the precipitation of the weld overlay
material. Comparison of WD128 microstructure with D128 reveals the effect
of steel on the weld overlay micorstructural evolution.
3-4- Sample Preparation
3-4-1- Composition analysis using spark

emission spectroscopy
This technique was employed to determine chemical composition of both the
overlay and tube materials using a FOUNDARY-MASTER instrument. The
instrument requires a flat surface of at least 15 X 15 mm and Figure 3-3 shows
a mounted cross section of a sample prepared for this experiment.
Figure ‎3-3. Showing a cross section of the flat sample using for chemical composition
analysis after mounting

The steel composition was determined from analysis of the surface shown in
Figure 3-3. The other side of sample was used for the analysis of the overlay
which was performed on two different flat planes. The position of these planes
was referenced to the distance from the steel/nickel alloy interface at the
midpoint of the sample. The distances were 700 μm and 500 μm respectively.
3-4-2- Sample preparation for optical and SE

microscopy
Specimens for the microstructural characterization studies were prepared by
conventional mounting, grinding and polishing methods. Because of oxidation
and decarburization of steel in the samples during high temperature exposure
tests, and to avoid its effects, approximately 1 mm of specimen surface were
removed by grinding. The cross sectional samples were firstly hot mounted
using different Metprep mounts (conductive and non-conductive) and a
Buehler Metaserv automatic mounting press. The mounted samples were
ground mechanically to 1200 grit with SiC papers, then polished with 6
microns and 1 micron diamond paste respectively to obtain mirror surface. For
some characterization processes such as EBSD, the roughness of surface
finished with 1 micron diamond paste was not suitable; in these cases the final
polishing was conducted with colloidal silica suspension (Si2O, 0.1 μm, pH:
9.0).
As a result of the widely dissimilar compositions of weld overlay material and
steel substrate in the samples, it was necessary to etch samples sequentially to
reveal the microstructure of each part. 2% NITAL solution was used for the
steel part and the etching time has a great effect on the microstructure and

after running many tests, it was found that by 30 seconds etching time the best
result can be achieved.
The weld overlay region required highly corrosive etch and there are many
processes for Alloy 625 etching which is reported in hand books and former
researches. In the present investigation, two different methods were used for
etching the weld overlay as well as wrought IN625, namely electrolytic
etching and etching with AQUA REGIA.
In former method, a solution of 10 wt.% oxalic acid in water, was used with
electrode potentials of 2-3 V. The cathode of the device was connected to a
stainless steel plate which was suspended in the solution. The anode was
connected to the sample with a wire. The desirable etching time in this process
depends on electrode potentials and decreases by increasing voltage. The
electrolytic etching process required a non-conductive mount which causes
limitations for using EDX of the samples in the electron microscopy.
The latter etching method used AQUA REGIA solution which consists of a
25% by volume of HNO3 and 75% by volume of HCl. It was found that this
solution must be heated up to 50°C prior to etching. This reduces the etching
time and causes more corrosion rate during etching and protrudes the
dendrites after etching. As the AQUA REGIA etchant solution is very
corrosive, the etching time is less than 10 seconds.
3-4-3- FIB-TEM sample preparation
The TEM analysis required a thin foil which is electron transparent and the
main issue is to prepare a sample and choose a proper thinning process among

different thinning processes. The Focused ion beam (FIB) instrument is a
powerful tool in material studies as micromachining for transmission
microscope specimen preparation is possible with it. Moreover, it can be used
for high resolution imaging process. In this section, the FIB-TEM sample
preparation will be outlined.
Because of sputtering action of the ion beam, the FIB can be used to locally
remove or mill away material. Therefore, this process can be used for
production of the sample from the bulk material and also for its thinning
process. The microstructure of the D64 and the D128 samples were studied
using the FIB-TEM method in the Department of Materials, Loughborough
University. Three specimens were prepared from these samples; the first one
was taken from the weld interface and two other samples were taken from
~100 μm and ~500 μm distant from the interface in the steel region.
Figure 3-4(a) shows the locations of the TEM specimens in the D64 sample.
TEM specimens were cut form similar locations in the D128 sample. Figure
3-4(b) shows a micro-volume sample for production of a TEM sample using
the FIB method. It is clear that material around the micro-volume sample has
been removed. Finally, the micro-volume sample has been cut from the bulk
with ion beam again.
An Omniprobe micro-manipulator was used to lift out the micro-volume
sample. This sample was thinned using FIB to produce a TEM specimen.
Finally, a Pt gas injector system is used for local Pt thin film deposition. The
TEM specimen with a deposited film of Pt and mounted in the sample holder
is shown in Figure 3-4(c).

Figure ‎3-4. Showing (a) TEM specimens locations in the cross sectional sample with
different distances from the interface, (b) the micro-volume sample which is used for the
TEM thin foil preparation, and (c) the TEM electron transparent specimen in the
sample holder which is covered by deposition of Pt films.

3-5- Principles of characterization methods
3-5-1- Optical emission spectroscopy
The chemical composition analysis of the steel substrate and the weld overlay
material as well as wrought IN625 were performed using an optical emission
spectrometer. The specimen preparation procedure is detailed in section 3-4-1.
The optical emission spectrometer determines element concentration via
a quantitative measurement of the optical emission from excited atoms.
This high-temperature discharge provides sufficient energy to promote the
atoms into high energy levels. The atoms decay back to lower levels by
emitting light. Since the transitions are between distinct atomic energy levels,
the emission lines in the spectra are narrow. The spectra of samples containing
many elements can be very congested, and spectral separation of nearby
atomic transitions requires a high-resolution spectrometer. Since all atoms in a
sample are excited simultaneously, they can be detected simultaneously using
a polychromator with multiple detectors. This ability to simultaneously
measure multiple elements is a major advantage of optical emission
spectroscopy. In this research, a FOUNDRY-MASTER instrument was used
which works with electron discharge in argon atmosphere. Different software
was used for analysis of the steel and the nickel alloy. The results of this
analysis are presented in appendix No.1.
3-5-2- Optical Microscopy
Optical microscopy is a conventional process for characterisation of the
microstructure. In this work, there are some features in the microstructure of

some samples which are only visible with an optical microscope. In other
words, the optical microscopy images show distinctive regions compared to
the other microscopy methods. In this work, primary microstructural analysis
was carried out on the samples using a Nikon model OPTIPHOT microscope.
This microscope was equipped with a Nikon digital camera DMX1200F for
imaging. Three different objective magnifications were used in this method.
For overall images it was 5X while for detailed images it was 40X and finally
for high magnification images, it was 100X.
3-5-3- Scanning Electron Microscopy (SEM)
SEM provides valuable information regarding the microstructural features in
the cross sectional samples. The SEM characterization was performed on a
FEI microscope. This FEI XL30 scanning electron microscope was fitted with
a tungsten filament which operates in high vacuum mode. The microscope
was primarily operated at an accelerating voltage of 20 kV, spot size of 4-6
and a working distance of 10 mm in both secondary electron (SE) and back
scattered electron (BSE) imaging modes.
In a typical SEM system, a stream of electrons generated by the filament of
the electron gun is accelerated toward the specimen using a positive electrical
potential. This stream is confined and focused using the metal apertures and
magnetic lenses into a thin, focused, monochromatic beam. This beam is
focused onto the sample using a magnetic lens. This electron beam irradiates
the surface of the specimen which in turn produces a signal in the form of
either x-ray fluorescence, secondary or back scattered electrons. The signal
produced by the secondary electrons is detected and sent to a cathode ray tube

(CRT) image. By changing the width (W) of the electron beam scan, the
magnification (M=D/W) can be changed where D is the width of the CRT.
Since D is constant, the magnification can be increased by decreasing the
width (W).
The secondary electron imaging method was mainly used for analysing the
microstructure of samples in the etched condition. However, backscattered
electron imaging was used for the analysis of different phases available in
samples in the un-etched condition, since the BSE imaging mode relies on the
differences in the mean atomic number of the phases. The mean atomic
numbers of all the phases which are expected to occur in the 2.25Cr1Mo steel/
nickel alloy system are listed in appendix No.2.
3-5-4- Transmission Electron Microscopy

(TEM)
Transmission electron microscopy is powerful imaging tool for detailed
crystal structure and observing the fine particles in the matrix as well as other
features such as grain boundaries. As mentioned in section 3-4-3, FIB-TEM
specimens were produced from different regions of the D64 and the D128
samples. A Philips Tecani F20 FEG-(S)TEM equipped with an Oxford ISIS
EDS system and a Fischione high annular dark field detector (HAADAF) was
operated at 200 kV in STEM mode for high resolution imaging and analytical
studies.
In a STEM, a narrow spot of the focused electron beam scanned over the
sample in a raster. The rastering of the beam across the sample makes this
mode of TEM suitable for analysis techniques such as mapping with EDS and

annular dark field imaging (ADF). These signals can be obtained
simultaneously, allowing direct correlation of image and quantitative data. By
using a STEM and a high-angle detector, it is possible to form atomic
resolution images.
3-5-5- Ion beam imaging
In the same manner that images are generated in an SEM, the ion beam can
rastered over a sample surface and the emitted electrons, particles (atoms and
ions), and electromagnetic radiation can be detected. In the FIB instrument,
the imaging is based on detecting the low-energy electrons referred to as ion-
induced secondary ions (ISEs).
As an alternative to chemical etching, by controlling the FIB, the contrast
mechanism for ISE can be changed from channelling which reveals grain
microstructure, to a contrast that reveals the phase distribution. In this
technique, only surface atoms layers are removed so actual surfaces are
remained for the quantification. The ion beam imaging analysis was
performed on an FEI Nova 600 Nanolab dual beam system which combines
both a FIB and a FEG-SEM in a single instrument.
3-5-6- Energy Dispersive X-ray Analysis

(EDX)
Energy dispersive X-ray (EDX) analysis was used together with SEM for
local elemental identification. When the incident electron beam interacts with
the specimen the bombardment of high energy electrons on the orbiting inner
most electron shell of an atom will lead to the escape of high energy electrons

and leave a vacancy. An electron from the outer orbiting shell jumps into the
empty electron vacant shell and emits energy. The emitted energy, in the form
of X-ray photons is specific for each element in the periodic table. The
induced X-ray fluorescence in the sample is energy-analyzed using a cooled
semiconductor detector. The element-specific spectral lines are identified to
give the elemental composition for the incident area.
This technique employed with an Oxford Instruments Si detector in
combination with the SEM in the present work. Data were converted using a
standardless ZAF correction procedure. The process is suitable for
determining elements with atomic number higher than 12; therefore light
elements such as B, C, etc. cannot be quantified.
3-5-7- Electron Probe Micro Analysis

(EPMA)
An electron probe micro analyzer (EPMA) is an analytical tool used for
determination of chemical composition of small volumes of solid materials. It
works similarly to a SEM-EDX but emitted X-rays are wavelength-analysed.
The basic measurement in this method is the intensity ratios of X-rays from a
sample and a standard which is most conveniently taken to be the pure
element [173].
The measured intensities require certain instrumental correction, including
subtraction of back-ground, the chief source of which is the continuous X-ray
spectrum. The composition at the analysed point is calculated from the
corrected intensities by applying „matrix corrections‟ which take account of
the various factors governing the relationship between intensity and

composition. These are commonly applied in the form of „ZAF correction‟
with separate correction factors dependent on atomic number, absorption and
fluorescence [174].
A JEOL JXA-8200 electron microprobe in Department of Archaeology at
University of Nottingham (with thermionic emission from an electron gun
fitted with a tungsten „hairpin‟ filament which operates at high accelerating
voltage under high vacuum mode) was employed to examine the light
elements such as carbon distribution in the microstructure of the samples. The
micro probe analyser was primarily operated at an accelerating voltage of 20
kV and 10-4 Torr or better vacuum. Magnetic lenses are used to focus the
beam into a fine probe incident on the surface of the specimen and a probe
diameter of 0.2-1 μm is typical. The electron beam current used in this
research was about 10 nA which is in the range of typical current (1-100 nA)
and working distance of this microprobe analyser is about 11 mm. In
microprobe analysis, the X-ray spectrum is recorded with either with 4
wavelength dispersive spectrometers (WDS) or one energy dispersive
spectrometer (EDS).
In the current research, the WDS method was used to obtain quantitative
information about the composition of different phases in each sample. Prior to
quantitative analysis, it was necessary to calibrate each spectrometer using
pure metal standards from Micro-Analysis Consultants Ltd. Moreover, the
φρΖ correction method supplied by JEOL was used to obtain quantitative
results.

3-5-8- EBSD
Electron backscattered diffraction (EBSD) is a technique which allows
crystallographic information to be obtained from the samples in the scanning
electron microscope (SEM). In EBSD, a stationary electron beam strikes a
tilted crystalline sample and the diffracted electrons form a pattern on a
fluorescence screen. This pattern is characteristic of the crystal structure and
orientation of the sample region from which it was generated.
The diffraction pattern can be used to measure the crystal orientation, measure
grain boundary misorientations, discriminate between different materials, and
provide information about local crystalline perfection. When the beam is
scanned in a grid across a polycrystalline sample and the crystal orientation
measured at each point, the resulting map will reveal the constituent grain
morphology, orientations, and boundaries. These data can also be used to
show the preferred crystal orientations (texture) present in the material. A
complete and quantitative representation of the sample microstructure can be
established with EBSD.
An Oxford Instruments HKL Technology Nordlys II EBSD detector fitted to
FEI XL-30 and controlled with CHANNEL5 software was used to obtain
preliminary EBSD results for the D128 sample. In order to determine the
chemical composition and crystallographic nature of observed phases in the
D32 and the D64 samples, the combined EBSD and EDS analyses were
carried out in Oxford Instruments laboratory. A Nordlys S detector and X-
MAX 80 ESD detector from Oxford Instruments installed on a FEG SEM
were used for this study. The Flamenco software was used to acquire both

EBSD and EDS data. Tango and Salsa software modules in CHANNEL5 were
used for post processing the EBSD data acquired. The EBSD data was
obtained with a probe current of 10 nA and an acceleration voltage of 20 kV.
3-6- Hardness measurements
3-6-1- Microhardness method
The Vickers hardness tests were conducted with a LECO-M400 machine and
were measured using a 200 gf load on specimens at various locations of
samples. The measurements were carried out on the steel and weld overlay
regions systematically at various distances from the weld interface. Four sets
of measurement were carried out on different locations of a sample which are
shown Figure 3-5(a).
In each set, there are different intervals between two successive indents. In the
preliminary measurements, the distance is more than ~200 μm, while in the
final measurements it was reduced to the standard minimum distance. In the
measurements of low hardness regions, and in order to keep the standard
minimum distance between two successive indents, the hardness
measurements were carried out on an inclined line. The distance from the
interface increased by ~100 μm and a ~100 μm horizontal distance is
considered between them. Figure 3-5(b) shows a set of these hardness
measurements which was carried out on the steel substrate of the D32 sample.
On the basis of the preliminary results, the detailed measurements of hardness
in the harder regions were carried out by decreasing the intervals between
indents to ~20 μm.

Figure ‎3-5. Showing (a) four preliminary hardness measurement sets of different regions
in a sample, (b) indicates a set of measurement in the steel region of D32 sample.
In order to characterize the boundary of different hardness regions, the
hardness measurement has been repeated on the certain areas of the sample
between two indents. This process was repeated for 4 different regions of the
sample and each time 5 tests were carried out. The hardness values versus the
distance from the interface were used to establish the hardness profile of the
sample in one location. Based on these results, the comparison between
hardness profiles of different samples can be made.
This procedure was applied on the as-welded sample as well as thermally
exposed samples for identification of hardness profile. The hardness profile

changes are crucial in this investigation, as they are related to the
microstructural evolution during high temperature exposure experiments.
3-6-2- Nanohardness method
A MML Nanotest system manufactured by Micro Materials Ltd was used to
carry out the nano-indentation for hardness measurement of the WD128
sample. The indentation experiments were performed using a standard
diamond Berkovich indenter. A 5 mN load was used with a 3X10 grid in
which indents were approximately 5 μm apart. Due to limitations of
microhardness method in the edges, this type of indentation was carried out on
two different regions i.e. bulk and edges of polished WD 128 sample in order
to reveal the hardness changes in these regions.

Chapter4 … Characterization of the as-welded condition
Chapter 4
4- Characterization of the as-welded condition
4-1- Introduction
The material in this project is a weld overlaid tube which has a potential
application as a superheater tube in a WtE boiler. The purpose of this chapter
is to make an extensive characterization for different regions of this tube in the
as-welded (i.e. as-deposited) condition. In the first section, the macrostructure
of the tube is reviewed and the chemical composition of different regions is
presented.
In the second section, the microstructure of the steel substrate is reported and
its different zones are introduced. It should be noted that the substrate is a
ferritic steel and the analyses were carried out to evaluate the effects of the

welding on its microstructure. A brief discussion about the interface region is
also presented in a following section.
The final section considers the weld overlay of similar composition to IN625
and in this section, characteristics of different zones in this material will be
reported. Moreover, secondary phases in this material will be analysed.
4-2- Macrostructural features and chemical

composition
The experimental work described here was conducted on the cross section of
weld overlaid tube to gain an understanding of the macrostructure. There are
two main regions, namely the steel substrate tube and the weld overlay as
shown in Figure 4-1. As shown, the overlay region thickness is between 1.7
and 2.0 mm and the steel substrate thickness is about 3.0 to 3.5 mm. The
variations in the thicknesses are because of surface roughness of the weld
overlay and the wavy interface between the two.
The welding process for cladding on the outer bore of steel tube causes a wavy
interface between the weld overlay and substrate. The amplitude of the
interface undulations depends on the location. There are large undulations
with a few millimetres wavelength and several hundred microns amplitude in
some points while there are some parts with smaller wavelength and few
microns amplitude.
The results of the chemical composition analysis of the weld overlay are
presented in the Table 4-1. The results are from two longitudinal planes
parallel to the tube axis but at different distances from the interface (700 μm

and 500 μm). The details of the sample preparation were presented in section
3-4-1.
Figure ‎4-1. Showing typical sample in this project with the thickness of the weld overlay
and the steel substrate; the interface undulation is related to welding and alters form
point to point.
The chemical composition of a wrought IN625 plate is presented in Table 4-2
for comparison. It is clear from Table 4-1 that the chemical composition of
weld overlay is similar to that of alloy IN625. By comparing these tables, it is
clear that the weld overlay has significantly more iron and carbon and lower
molybdenum and niobium contents. However, the chromium content is about
the same in the both materials.
Moreover, it can be seen that some minor elements such as Ti in both
materials are in the same range while other minor elements i.e. Cu, Co and Al
contents are negligible in the weld overlay material. The higher Fe and C
levels in the weld overlay material come from either dilution during cladding
or the use of a consumable with higher iron and carbon compared to the
standard IN625 or a combination of both.

Table ‎4-1.Chemical composition of the weld overlay from different planes with 700 μm‎
and 500 μm‎distance‎from the interface
Composition (wt.%)
Element Average (wt.%)
at 700μm at 500μm
C 0.02 to 0.03 0.01 to 0.02 0.02
Ti 0.17 to 0.18 0.18 to 0.19 0.18
Cr 20.6 to 20.7 21.0 to 21.1 20.9
Fe 8.3 to 8.7 7.3 to 7.7 8.0
Ni Balance Balance Balance
Nb 3.1 to 3.2 3.1 to 3.3 3.2
Mo 8.1 to 8.2 8.1 to 8.3 8.2
Table ‎4-2.Measured chemical composition of the wrought IN625, wt.%
Element wt.% Average (wt.%)

C 0.004 to 0.006 0.005
Al 0.1 to 0.11 0.1
Ti 0.20 to 0.22 0.21
Cr 21.8 22.0 21.9
Fe 3.09 to 3.28 3.18
Co 0.09 to 0.1 0.1
Ni Balance Balance
Cu 0.17 to 0.174 0.17
Nb 3.4 to 3.6 3.5
Mo 9.1 to 9.2 9.2
Table ‎4-3.Measured chemical composition of the steel substrate, wt.%
Element wt.% Average (wt.%)

C 0.09 to 0.11 0.10
Si 0.22 to 0.25 0.24
Cr 2.31 to 2.34 2.33
Mn 0.42 to .44 0.43
Fe Balance Balance
Cu 0.10 to 0.11 0.11
Ni 0.9 to 0.1 0.1
Mo 0.91 to 0.92 0.92

The chemical analysis of the steel substrate is presented in Table 4-3. The
chromium content is approximately 2.25 wt.%, the molybdenum content is ~ 1
wt.% and the carbon content is ~ 0.1 wt.%. Therefore, it can be assumed that
the steel substrate in the overlaid tube is the well known 2.25Cr1Mo steel
which is also known by the code 10CrMo910 and T22. The results of the
optical emission spectroscopy are presented in the appendix No.1.
4-3- Microstructural characterization of

2.25Cr1Mo steel substrate
Figure 4-2(a) is a low magnification image of the steel region of the tube
cross-section. Two different zones in the 2.25Cr1Mo steel microstructure are
clearly visible after etching with 2% Nital and are labelled in this image.
1) A HAZ zone close to the interface, about 1 mm wide and with a
uniform microstructure.
2) A central zone of the steel region, ~ 2 to 2.5 mm wide and with a
bimodal microstructure.
The optical microscopic images of different zones of the steel substrate in the
as-welded condition are shown in Figure 4-2(b) and (c). The HAZ can be
divided into two regions as follows (see Figure 4-2(b)):
1) A Near-HAZ ~ 300 μm wide with a uniform microstructure
2) A Far-HAZ ~ 700 μm wide with a non-uniform microstructure.

Figure ‎4-2. Showing different zones in the cross section of the steel substrate in the
etched condition; (a) low magnification image indicates the HAZ and the central zones,
(b) higher magnification optical microscope image of two zones in the HAZ and the
transition zone and (c) higher magnification optical microscope image of the central and
inner bore zones.

It is clear that by moving away from the interface, the uniformity of the HAZ
zone decreased which is related to the maximum temperature and rate of
heating/cooling of the thermal cycle during the welding process. Moreover,
there is a zone ~ 200μm wide between the far-HAZ and central zone. The
microstructure of the steel has changed gradually from a bimodal structure to a
more uniform microstructure in this zone which is termed the transition zone
in this thesis.
Figure 4-2(c) shows the inner bore of the tube; further investigation revealed
that the steel microstructure close to the inner bore area is slightly different
from that in the central zone. This area ~200 μm wide has a higher fraction of
ferrite grains compared to the central zone. There are thus five main different
zones in the 2.25Cr1Mo steel substrate of the weld overlaid tube with different
microstructures, and from the tube‟s inner bore towards welding interface they
are referred as following:
1) An inner bore zone ~150 to 200 μm wide with a higher proportion
ferrite grains in the microstructure.
2) A central zone ~ 2000 to 2500 μm wide with approximately equal
portions of ferrite and carbide containing regions.
3) A transition zone ~ 200 μm wide which has a transformed two phase
microstructure.
4) A far-HAZ zone ~ 700 μm wide with a one phase microstructure;
however, the former bimodal structure is still visible.
5) A near-HAZ zone ~ 300 μm wide with a uniform microstructure.

4-3-1- Inner bore microstructure
It is probable that the central and the inner bore zones of the steel were not
affected by the thermal cycle of the welding process and therefore they retain
their initial microstructures. Figure 4-3 shows different images of these
regions. The optical microscope image of the inner bore zone is shown in
Figure 4-3(a1) and reveals predominantly ferrite (α-Fe) grains and some small
dark etching areas.
The width of this zone varies from point to point and is ~150 to 200 μm wide
as indicated in the lower magnification SE micrograph of this region (see
Figure 4-3(b1)). The higher magnification SE micrograph of the inner bore
zone microstructure is presented in Figure 4-3(c1) and reveals that ferrite (α-
Fe) grains have a featureless appearance after etching while the dark optical
microscope areas surface have a mottled contrast. It is suggested that bright
particles are carbides in a tempered martensite or bainite structure. These areas
will be termed TM/B (tempered martensite/bainite) in the rest of the thesis.
4-3-2- Central zone microstructure
The bimodal microstructure of the central zone is seen in the majority of
2.25Cr1Mo steel region and consists of approximately equal proportions of
ferrite (α-Fe) grains and TM/B areas. This type of microstructure is seen in the
central part of steel tube between the inner bore and the transition zones. The
optical microscope image of this zone is presented in Figure 4-3(a2) and
shows its main features which are large areas of TM/B and ferrite (α-Fe)
grains of different size.

Figure ‎4-3. Showing microstructure of the inner bore (1) and the central (2) zones in the
etched condition; (a1,a2) optical microscope images, (b1,b2) lower magnification SE
micrographs and (c1,c2) higher magnification SE micrographs.

As seen in Figure 4-3(b2), there are some locations in the central zone where
the TM/B areas can be considered as matrix and small grains of ferrite (α-Fe)
are formed in these areas. Nevertheless, there are other locations where a
group of ferrite (α-Fe) grains are formed and their grain boundaries are clearly
visible. There are some bright particles on the grain boundaries of ferrite
grains as labelled in this image. The bright particles are seen more clearly in
the higher magnification SE micrograph (see Figure 4-3(c2)) on the
boundaries of ferrite grains and in the TM/B areas. Referring to section 2-3, it
is believed that these particles are likely to be carbides as a common feature in
a tempered martensitic or bainitic structure. It must be noted that the etching
process causes relief of carbides on the surface and so carbides are seen as
bright in the SE micrographs.
4-3-3- Transition zone microstructure
The transition zone is between the far-HAZ and the central zone. In this zone,
the bimodal microstructure including (α-Fe) and TM/B has partially
transformed to austenite (γ-Fe) during the welding process. The austenite
subsequently transformed to martensite or bainite during the rapid cooling of
the weld thermal cycle. The resultant regions after cooling are termed Tγ
(transformed γ) in this thesis.
Figure 4-4(a) shows the etched microstructure of the transition zone. In the
higher peak temperature parts of this zone (upper parts of Figure 4-4(a)) the
microstructure was completely transformed to γ. However, some
untransformed ferrite grains are present in the central part and some TM/B
areas appear in the lower parts of this image.

Figure ‎4-4. Showing microstructure of the transition zone in the etched condition; (a)
optical microscope image indicates the‎ untransformed‎ (α-Fe) and the untransformed
TM/B‎ as‎ well‎ as‎ the‎ Tγ,‎ (b) lower magnification SE micrograph of the untransformed
(α-Fe) and Tγ areas in the transition zone and (c) higher magnification SE micrograph
indicates‎the‎Tγ structure in the TM/B areas and corners of a ferrite‎(α-Fe) grain.

The untransformed (α-Fe) grains and TM/B are related to the lower peak
temperature during the welding process. In the lower parts of transition zone,
only some parts of TM/B areas and the corners of (α-Fe) grains were
transformed to the γ during heating up period. Figure 4-4(b) shows a low
magnification SE micrograph of the transition zone which is similar to the
optical microscope image. The untransformed (α-Fe) grains and TM/B areas
are clearly labelled in this image.
The higher magnification SE micrograph of this zone is presented in Figure
4-4(c) and reveals that the Tγ structure is different from the initial
microstructure. The edges and corners of ferrite grains convert to Tγ after the
welding process while the centre of the grains remains untransformed. In the
transition zone, the TM/B areas and the ferrite (α-Fe) grains were transformed
to a uniform structure which is the main feature of the HAZ. The
transformation process from the initial structure to austenite (γ) at high
temperature and then to the Tγ structure is critically dependent on the heating
and cooling rates of the thermal cycle of the welding and the peak temperature
of the region. Moreover, it is suggested that the TM/B areas were transformed
to austenite in a shorter time compared to the ferrite grains.
4-3-4- Far-HAZ zone microstructure
In this work, the HAZ region is divided into a near-HAZ and a far-HAZ. It is
suggested that, during welding, the temperature of the steel surface must reach
at least to that of IN625 melting temperature. Referring to former research
[150], there is a range for IN625 melting temperature which is between
approximately 1250 and 1420˚C. Therefore, the regions of the steel adjacent

to the weld overlay were heated above the austenitisation temperature and
rapidly cooled afterwards. The region where austenitisation and air cooling led
to a uniform lath-like structure is termed near-HAZ.
The uniformity of the austenite phase depends on the carbide dissolution and
austenitisation temperature which is a function of distance from the interface.
The region where a completely uniform lath-like structure was not seen is
termed the far-HAZ. Thus, the area closer to the interface is heated up to a
higher temperature, so the austenite in this area is more uniform compared to
other areas which are far from the interface. It is suggested that in the far-HAZ
zone, the carbide dissolution in the austenite during welding may not have
been completed.
Figure 4-5(a) shows an optical microscope image from the area about 500 μm
distant from the interface. It is shown in Figure 4-2(b) that the width of the
far-HAZ zone is ~700 μm and therefore the image with ~500 μm distance
from the interface is showing the central part of the far-HAZ zone. Traces of
the initial bimodal microstructure of the central zone are still distinguishable
in this image.
A low magnification SE micrograph of this zone, taken from the same region
(~500 μm distant from the interface) is presented in Figure 4-5(b). There are
some regions with more bright contrast particles compared to the other parts.
Therefore, this type of microstructure is called non-uniform Tγ in this thesis.
Figure 4-5(c), shows a higher magnification image of this zone. The former
TM/B areas and ferrite (α-Fe) grains are labelled in this image.

Figure ‎4-5. Showing microstructure of the area ~500 μm‎distant from the interface (far-
HAZ) zone in the etched condition; (a) optical microscope image indicates former (α-Fe)
and TM/B, (b) lower magnification SE micrograph of the non-uniform far-HAZ zone
and (c) higher magnification SE micrograph contains blocky carbide particles in the
former TM/B area.

The main feature of the far-HAZ zone is the presence of the un-dissolved
carbide particles in the former TM/B areas which is related to incomplete
austenitisation during heating up period. The un-dissolved carbide particles in
the far-HAZ zone (see Figure 4-5(c)) are blocky which are different compared
to the initial TM/B area (see Figure 4-3(c2)).
4-3-5- Near-HAZ zone
Figure 4-6(a) shows an optical microscope image of the interfacial region
(close to the overlay) in the etched condition. The microstructure of this
region (defined as the near-HAZ zone) is uniform without any un-dissolved
blocky carbides. A high magnification SE micrograph of the interfacial region
is shown in Figure 4-6(b). The lath-like structure is visible in the
microstructure close to interface which is the main characteristic of rapidly
cooled austenite.
Figure 4-6(c) shows the microstructure of the area about 300 μm distant from
the interface. It can be seen that the nature of the uniform Tγ microstructure is
also visible in this area. Therefore it is suggested that the width of this zone is
~300 μm which is different from point to point. It can be said that the
appearance of the un-dissolved carbides could be considered as the border
between the near-HAZ and the far-HAZ zones.
Comparison of Figure 4-6(c) and Figure 4-5(c) reveals the differences
between the uniform Tγ and the non-uniform Tγ microstructure.

Figure ‎4-6. Showing microstructure of the near-HAZ zone in the etched condition; (a)
optical microscope image shows the microstructure of the uniform Tγ‎close to interface,
(b) higher magnification SE micrograph of the interfacial region illustrates lath-like
structure and (c) higher magnification SE micrograph of the area ~300 μm‎distant from
the interface shows uniform‎Tγ.

It is suggested that at a smaller distance from the interface, un-dissolved
blocky carbide particles disappeared from the microstructure and replaced
with the lath-like structure. The length of the laths of the Tγ microstructure
increased from less than 10 μm in the area away from the interface (see Figure
4-6(c)) to about 30 μm adjacent to interface (see Figure 4-6(b)).
4-4- Interface characterisation
In order to understand the intermixing process of the fusion weld overlay and
the steel substrate, a comprehensive investigation on the interface region was
carried out. Figure 4-7(a) and (b) show the typical interface in the cross
sectional samples in the etched condition. Nital did not affect the weld overlay
region; therefore the interface can be clearly identified. EDX line scans across
the interface showed the compositional variation in the interfacial region. The
composition changes from a Fe-rich to a Ni- rich alloy in less than 5 μm (see
Figure 4-7(c)). This profile for different shapes of interface was similar.
Further investigation was carried out on longitudinal sections. Figure 4-8(a)
shows a BSE micrograph of the interfacial region in these samples and
indicates the semi-melted interface areas. It is clear that during the welding
process, some areas of the steel did not fully melt. The semi-melted areas in
the interfacial region proved that the interface is not a planar feature between
the steel and weld overlay. Figure 4-8(b) shows an un-melted steel part in the
weld overlay region. The presence of the semi-melted regions and or un-
melted parts in the interfacial region cause an undulating interface and
therefore it can be considered as a three dimensional feature.

Figure ‎4-7. Showing the nature of the interface in the weld overlay/substrate system; (a,
b) SE micrographs from different locations of interface etched in Nital indicate different
interface shapes, (c) compositional profile at the interface shown in figure(a).

Figure ‎4-8. Showing the weld overlay/substrate interfacial region, (a) low magnification
BSE micrograph from the interfacial region in the longitudinally sectioned sample
indicates the semi melted areas, (b) higher magnification SE micrograph form the un-
melted steel part in the weld overlay region.
4-5- Microstructural characterization of the weld

overlay
Etching samples with Aqua-Regia reagent revealed the weld overlay
microstructure. Figure 4-9(a) shows an optical microscope image from the
whole region of the weld overlay. One edge of this region is the interface with
the steel and the other one is the tube outer surface. The weld overlay
microstructure has a dendritic structure which is a characteristic of an as-

solidified material. The columnar dendrites exist through the whole region and
show directional solidification aligned with the radial direction.
Figure 4-9(b) shows a higher magnification optical microscope image of the
weld overlay close to interface with the steel. There is a region in this image
where the columnar dendrites are not visible. Due to partial melting of the
steel surface and intermixing with the weld pool during the welding process,
the microstructure of this region is different from the other parts. This region
of less than 30 μm wide is called the mixing zone in the rest of this thesis.
However, other parts of the weld overlay region have a similar structure.
Therefore the region beyond the mixing zone is considered as the bulk zone of
the weld overlay.
The most important feature in the dendritic microstructure of the bulk weld
overlay is inhomogeneity of the chemical composition between the dendrite
cores and interdendritic regions. The segregation phenomenon in this region
can be seen in the BSE micrograph (BSE signal is sensitive to the mean
atomic number) and is shown in Figure 4-9(c).
As seen in this image, there are some bright features which are labelled as
interdendritic regions of higher mean atomic number. Moreover, the darker
areas can be considered as the cores of dendrites. It means that during
solidification, the segregation of higher atomic number elements such as Mo
and Nb cause the interdendritic regions to become brighter in contrast
compared to that of the dendrite cores.

Figure ‎4-9. Showing microstructure of the weld overlay region; (a) low magnification
optical microscope image indicates the edges of the weld overlay region, (b) high
magnification optical microscope image of the mixing zone close to interface and (c) BSE
micrograph of the weld overlay region indicates inhomogeneity in the microstructure of
bulk zone.

EDX analyses, with line scans, were carried out on the different locations of
the weld overly region in order to reveal the chemical composition differences
between the interdendritic and dendrite core regions. Figure 4-10(a) shows an
EDX line scan results from a location in the bulk zone at ~500 μm from the
interface. The corresponding location of this analysis is presented in Figure
4-10(b). It can be seen the bright contrast regions match with the higher
niobium and molybdenum content regions.
Figure ‎4-10. Showing inhomogeneity in the bulk zone of the weld overlay (a) EDX
analysis results showing concentration profiles of Mo and Nb, (b) location of the line
scan indicates the bright regions correspond to the higher Mo and Nb regions.

4-5-1- Chemical composition profile through

the weld overlay region
Based on the previously mentioned results, it is suggested that there are two
main zones in the weld overlay region as follows:
1) A mixing zone ~ 30 μm wide
2) A bulk zone with a dendritic structure
EDX area analyses were carried out on different locations of these two zones
to analyse the differences in their chemical composition. In order to compare
the results of different parts, all the EDX analyses were taken in the form of a
box 10 μm wide and 100 μm long. EDX analyses were repeated on five or six
different locations with the same distance from the interface.
4-5-1-1- Mixing zone
The chemical composition of the mixing zone was determined with two sets of
area analyses. The first set was carried out on many different locations at ~1
μm distant from the interface. The schematic diagram of this set of analyses is
shown in Figure 4-11 and the results are presented in Table 4-4. This table
presents the results from 6 different locations and reveals that the chemical
composition in the first 10 μm of weld ovverlay varies significantly.
The second set of analyses was carried out on different locations at ~10 μm
distant from the interface (D=10 μm). Table 4-5 shows the results of 5 points
of this set of analyses and it seen that by increasing the distance from the
interface, the average of iron content decreases while other elements increase.

Figure ‎4-11. Showing schematic of EDX analysis on different locations of the mixing
zone‎with‎1‎μm‎distance‎from‎the‎interface
Table ‎4-4. Chemical composition of the mixing zone in the weld overlay region at
different areas 1 μm distant from the interface. (D=1 μm)
Element (wt.%) A(1) A(2) A(3) A(4) A(5) A(6) Mean S.D.
Cr 10.5 13.1 15.4 15.3 15.1 11.3 13.5 2.2
Fe 57.3 44.7 33.4 33.7 36.9 54.5 43.4 10.6
Ni 26.8 35.3 43.2 43.2 40.5 28.5 36.2 7.3
Nb 1.2 1.5 1.9 1.8 1.7 1.5 1.6 0.3
Mo 3.9 5.1 5.8 5.9 5.6 4.1 5.0 0.9
Table ‎4-5. Chemical composition of the mixing zone in the weld overlay region at
different areas 10 μm distant from the interface. (D=10 μm)
Element (wt.%) A(1) A(2) A(3) A(4) A(5) Mean S.D.
Cr 16.2 19.6 18.8 14.8 17.8 17.4 1.9
Fe 30.7 16.1 17.6 36.7 23.1 24.8 8.7
Ni 43.0 53.1 51.5 40.2 48.7 47.3 5.5
Nb 3.2 3.2 3.9 2.2 3.0 3.1 0.6
Mo 6.6 7.7 7.7 5.9 7.3 7.0 0.8
A comparison of the mean values of Fe and Ni in these tables reveals the
intermixing of the weld overlay with the steel substrate. The variation in the

chemical composition at a fixed distance from the interface is probably related
to the interface undulations. The presence of the semi melted and/or un-melted
parts of the steel substrate in the area close to interface causes higher iron
content in some locations of the mixing zone.
4-5-1-2- Bulk zone
In order to have a chemical composition profile in the bulk zone, the EDX
area analyses were carried out on locations with different distances from the
interface. The EDX results for one set of these measurements are shown in
Figure 4-12. As variations of high atomic number elements i.e. Mo and Nb are
smaller than other elements, their profiles are presented in a separate graph. It
is seen that after a distance of ~30 μm from the interface, the composition of
the bulk zone is almost uniform through whole region. The border between the
mixing and the bulk zones can be defined by the dashed line shown in Figure
4-12.
This type of measurement was repeated on different locations of weld overlay
region and the results were very similar. The resultant profiles exhibit a
plateau between 20 to 40 μm from the interface. Comparison between the
compositions of the bulk zone in different profiles reveals that nickel and iron
content vary within a range of 3 wt.%. However, the concentration variation
range of other elements in different profiles is less than 1 wt.%. Finally, the
composition in the bulk zone is close to the analysis results from the spark
emission spectroscopy (see Table 4-1).

60
Cr
50
Fe
Element wt.%
40 Ni
30
20
10 Mixing Bulk
0
1 10 100 1000
Distance‎from‎interface(μm)‎
10
9 Nb
8
7
Element wt.%
Mo
6
5
4
3
2
1 Mixing Bulk
0
1 10 100 1000
Distance from interface(μm)
Figure ‎4-12. Showing typical EDX analysis results for the weld overlay region on
logarithmic distance scale; (above) indicates variation of the matrix elements i.e. Cr, Fe
and Ni as a function of distance from the interface, (below) shows variation of minor
alloying elements i.e. Mo and Nb in the same way.

4-5-2- EPMA results
EPMA analysis using WDX spectrometers was carried out on the interfacial
region of a cross sectional sample. The methodology employed is described in
section 3-5-7. Figure 4-13 shows elemental colour maps for Cr, C, Nb, Ti, Mo
and Fe as well as a backscattered electron image from the interfacial region
(with CP in the data bar) between the steel (lower part) and weld overlay
(upper part). The concentrations represented by the different colours are
shown by the scales to the right of the image.
The mixing zone, previously identified by SEM/EDX, is clearly visible as a
~30 μm wide zone of high Fe and low Ni, Cr, Mo and Nb compared to the
bulk of the weld overlay. The levels of these elements away from the mixing
zone are also in agreement with the SEM/EDX results. Moreover, EPMA
results give more distinct information about distribution of the niobium and
molybdenum in the weld overlay region.
As seen, the concentrations of the both of these elements are increased in the
mixing zone with increasing distance from the interface (Mo concentration at
the interface is ~1 at.% and increases to about 6 at.% while Nb concentration
at interface is almost zero and increases to ~2 at.%). However, the Mo
distribution in the bulk zone is somewhat different from the Nb distribution.
The Mo map appears more uniform whereas the Nb is concentrated in some
regions in the bulk zone which are red and visible in Figure 4-13.
It should be noted that most of the Nb-rich points correlate with Ti-rich points.
This fact can be related to formation of secondary phases at final stages of

solidification. Presence of Ti and Nb possibly is related to (MC) carbide.
Moreover, the carbon concentration range is not valid which is related to its
low concentration.
Figure ‎4-13. Showing different element maps in the interfacial region with the steel at
the lower part and weld overlay at the upper part; the variation in chemical composition
is visible in the mixing zone. The boundary between the mixing and the bulk zones is
clearly visible and some (Nb, Ti)-rich points are present in the bulk zone. The carbon
concentration range is not valid and its map did not show anything.

In summary, as a result of the welding of the nickel alloy on the outer bore of
the steel tube, following regions are present in the weld overlaid steel tube in
the as-welded condition:
1) 2.25Cr1Mo steel substrate
a) A inner bore region which contains ferrite grains
b) A central zone region with bimodal microstructure which
contains approximately equal portions of ferrite grains and
TM/B areas as the main zone
c) A transition region which shows the extent of austenitisation
during the welding
d) A far-HAZ region ~700 μm wide with a non-uniform
microstructure including un-dissolved carbides
e) A near-HAZ zone region ~300 μm wide with a uniform lath-
like microstructure
2) Nickel alloy weld overlay region
a) A mixing zone ~30 μm wide with high Fe content
b) A bulk zone with ~9 wt.% Fe content and 0.02 wt.% C which
are higher than the standard wrought IN625 which is
mentioned in Table 4-2.

Chapter5 … Characterization following high temperature exposure
Chapter 5
5- Characterization following high

temperature exposure
5-1- Introduction
In order to prevent failures in service, it is necessary to evaluate the effects of
service conditions on the microstructure and mechanical properties of the steel
tube and the weld overlay, particularly in the interfacial region. In this project,
the main focus is on the effect of high temperature exposure on the hardness
and microstructure of different regions of the weld overlay and the steel
substrate. The weld overlay is a nickel based alloy similar in composition to
IN625 but with a higher iron and carbon contents and the substrate is
2.25Cr1Mo ferritic steel.

Due to the differences in the chemical composition of these materials,
transport of various elements across the interface is likely to occur during the
high temperature exposure. As the results reveal a complex set of interactions,
a brief overview will be presented in section 5-2. The aim of this is to provide
the reader with a succinct description in which the primary processes in the
microstructural evolution are defined. Moreover, schematic diagrams
displaying the main features in the microstructure of both materials are
outlined.
In the following sections of this chapter, the results of hardness tests and
microstructural analysis of the thermally exposed samples are presented. In
these sections, evidences about the microstructure evolution in the weld
overlay region and different zones of the steel are described in detail. The
results are classified by region in terms of the microstructural evolution
processes.
5-2- Overview
The hardness and microstructure of samples were analysed in detail (see
section 5-3 and 5-4) and changes in the microstructure are classified in the
present section. The changes in the microstructure were compared with the as-
welded condition. The microstructures of the steel substrate as well as the
weld overlay in the as-welded condition are reported in chapter 4. In the as-
welded condition, there is a significant variation in the microstructure of the
steel from the near-HAZ region to the tube inner bore due to the different
thermal cycles experienced.

Figure ‎5-1. Showing (a) schematic diagram of the as-welded sample microstructure,
indicates different region in the steel, (b) schematic diagram of the D1 sample
microstructure, indicates tempering in the steel and the fine veins formation in the weld
overlay, (c) indicate the hardness values of both materials; in the steel due to tempering
it is decreased while it is increased for the weld overlay region due to hardening.

Figure 5-1(a) shows the schematic diagram of the as-welded sample. In this
steel, the initial microstructure contains ferrite (α-Fe) grains and the TM/B
areas. This bimodal structure converts to austenite during welding and then by
rapid cooling lath-like structure was formed and termed (Tγ). Due to the
gradient in peak temperature during the welding, the uniformity of HAZ
microstructure decreases with increasing the distance from the interface
(compare the near-HAZ with far-HAZ).
Thermal exposure at 650˚C for 1 day causes tempering of the lath-like Tγ
structure in the HAZ region to give a tempered structure with small carbide
precipitates. Figure 5-1(b) shows schematically this issue. The hardness drops
in the HAZ region as seen in Figure 5-1(c). Thermal exposure causes
hardening of the weld overlay, and hardness of the weld overlay in the D1
sample increases compared to the as-welded condition (see Figure 5-1(c)).
Moreover, due to carbon migration across the weld interface at high
temperature and presence of carbide former elements, carbide particles are
likely to precipitate along the grain boundaries in the weld overlay. Therefore,
some features, termed veins, formed in the weld overlay side of interfacial
region. These veins ~50 μm long are shown in Figure 5-1(b).
After 8 days exposure (D8 sample), tempering of the steel and a decrease in its
hardness continues. This is accompanied by increasing the length of the veins
in the weld overlay to about 100 μm. Moreover, the hardness increases up to a
depth of 20 μm into the weld overlay. This region is termed a “hard band”
which is emphasised as a dark band in Figure 5-2(a). The maximum hardness
of this region rises to about 500 kgf mm-2.

Figure ‎5-2. Showing (a) schematic diagram of the D8 sample, indicates fully tempered
structure in the steel and the vein growth and hard band formation in the weld overlay
(b) schematic diagram of the D16 sample, indicates soft zone in the former near-HAZ
region and further growth in the veins and wider hard band in the weld overlay and (c)
indicates the hardness variation in different parts of two samples.

Further increase in exposure time causes apparently further carbon migration
from the steel to the weld overlay and a soft zone forms on the steel side of the
interface. This soft zone is first apparent after 12 days exposure and remains
even after 32 days of exposure. The microstructure of the D16 sample is
shown schematically in Figure 5-2(b). The main characteristics of the soft
zone are a large ferrite grain structure and low hardness (see Figure 5-2(c)).
The width of the soft zone is ~300 μm which is equivalent to the near-HAZ
region. The soft zone is largely free from precipitates.
In the weld overlay region of the D16 sample, the width of the hard band
increased to ~50 μm and its maximum hardness increased to ~600 kgf mm-2.
The hard band width is correlated to the mixing zone (see section 4-5-1-1)
which is a nickel rich region with ~30 at.% iron content material. Moreover,
comparison of the D8 and D16 samples indicates that veins become thicker
with increasing exposure time at 650˚C.
After 64 days exposure, see Figure 5-3(a), two important changes took place
in the near-HAZ region. First, fine particles were precipitated within the large
ferrite grains and secondly a network of coarse carbide particles was formed at
the interface between the weld overlay and the steel (in the steel side). This
has led to a re-hardening of the near-HAZ region (see Figure 5-3(c) compared
to Figure 5-2(c)).
Within the weld overlay region, the features termed veins did not extend
further into the sample but appeared more distinct, suggesting a coarsening of
the precipitates within them (see Figure 5-3(a) compared to Figure 5-2(b)).

Figure ‎5-3. Showing (a) schematic diagram of the D64 sample, indicates precipitates
within the large ferrite grains and a coarse precipitate network at the interface region,
(b) schematic diagram of the D128 sample indicate fine grains of ferrite and TM/B
formation in the former near-HAZ region; and (c) indicates the hardness variation of
the different regions of two samples.

Finally, after 128 days, the hardness values of the near-HAZ region and the
hard band in the weld overlay underwent little further change. However, the
grain size of the ferrite in the near-HAZ region decreased very significantly
from ~150 μm to less than 10 μm. Also in the bulk zone of the weld overlay
region the TCP precipitates become visible (see Figure 5-3(b)).
This overview provides a framework for presenting the detailed experimental
results in the following sections. The hardness results are presented in section
5-3 followed by the microstructral observations after different ageing times at
650 ˚C in section 5-4.
5-3- Hardness survey
The microhardness measurements for both materials, the weld overlay (nickel
alloy) and steel tube (2.25Cr1Mo) were carried out in the as-welded condition
and after each thermal exposure experiment. In this section, the results of
hardness tests for different samples are presented. The hardness measurement
procedure was described in section 3-6.
It should be noted that several hardness measurements in a sample show
similar profiles and in a particular zone, there is ~30 kgf mm-2 difference
between the maximum and the minimum values. Therefore, for displaying the
changes in the hardness profile of different samples, one set of measurements
is presented in the diagrams for each sample. The typical ~30 kgf mm-2 error
bar should be considered in each diagram which represents the difference
between values of a certain zone in the sample.

The hardness values of the weld overlay region in the as-welded condition are
presented in Figure 5-4(a), and show a range of 220 - 250 kgf mm-2 for both
zones. The hardness values for the steel region in the as-welded condition are
presented in Figure 5-4(b) and show the maximum hardness values are
corresponding to the near-HAZ region (~300 μm wide) and are between 430
and 460 kgf mm-2.
The hardness values in the far-HAZ region are in the range of ~200 to ~430
kgf mm-2 and decrease with increasing distance from the interface. The
hardness values in the central zone are in the range of ~150 to ~180 kgf mm-2.
It can be suggested that the hardness profile in the as-welded condition is
correlated to the peak temperature during welding.
High temperature exposure has a significant effect on the hardness of both
regions (steel and weld overlay). The hardness values for the weld overlay and
the steel regions of the D1 sample are presented in Figure 5-4(a) and (b)
respectively. As seen, in the weld overlay region, the hardness values increase
to the range of ~280 to ~310 kgf mm-2. Moreover, the hardness values in both
regions of the HAZ in the D1 sample are decreased compared to those of the
as-welded sample. The hardness values in the HAZ region varied in the range
of ~220 to ~250 kgf mm-2, which are still higher compared to those of the
central zone.
The central zone hardness values range is between ~150 and ~180 kgf mm-2,
in the D1 sample, which is similar to that of the as-welded sample. This means
that high temperature exposure for 1 day did not affect the hardness values of
this zone in the steel region.

a) 500
D1
450
As-welded
Hardness (kgfmm-2)
400
350
300
250
200
-2000 -1500 -1000 -500 0
Distance from interface into weld overlay (μm)
b) 500
D1
400 As-welded
Hardness (kgfmm-2)
300
200
Near Far
HAZ HAZ Central zone
100
0 500 1000 1500 2000 2500 3000 3500
Distance from interface into steel (μm)
Figure ‎5-4. Showing (a) hardness profile in the weld overlay region of the as-welded and
the D1 samples, (b) hardness profile in the steel region of the as-welded and the D1
samples.
The hardness values of the D4 sample in different regions are shown in Figure
5-5. As seen, increasing the exposure time to four days did not change the
hardness profiles significantly in the steel region. However, hardness values in

the weld overly region increased to the range of ~320 to ~350 kgf mm-2,
which is higher compared to that of the D1 sample (~280 to ~310 kgf mm-2).
a) 500
Bulk
Mixing zone
450 D8 zone
Hardness (kgfmm-2)
D4
400
350
300
250
200
-2000 -1500 -1000 -500 0
b ) 500
D8
D4
400
Hardness (kgfmm-2)
300
200
Near Far
HAZ HAZ Central zone
100
0 500 1000 1500 2000 2500 3000 3500
Figure ‎5-5. Showing (a) hardness profile in the weld overlay of the D4 and the D8
samples indicates hard band formation, (b) hardness profile in the steel region of the D4
and the D8 samples, indicates hardness decreasing in the HAZ region.

Figure 5-5(a) shows the hardness profile in the weld overlay region of the D8
sample. As seen, the hardness of the mixing zone increased significantly to the
range of 470 - 500 kg fmm-2. Meanwhile, the hardness values of the bulk zone
underwent little changes compared to the D4 sample. In the steel region, as
shown in Figure 5-5(b), the hardness values of the HAZ region in the D8
sample decreased to the range of 170 - 200 kgf mm-2; which is very close to
the hardness values of the central zone.
A comparison of the hardness profiles in the as-welded sample (see Figure
5-4) and the D8 sample (see Figure 5-5) shows that the hardness profile in the
steel region becomes uniform through whole the steel region. Moreover, after
8 days of exposure, the hardness values increased in the mixing zone of the
weld overlay region which is called hard band.
Increasing the exposure time to 16 days causes further changes in the hardness
profiles of both regions. The weld overlay hardness profile in the D16 sample
is presented in Figure 5-6(a) and it is obvious that the maximum hardness in
the hard band increases to a range of 590 - 620 kgf mm-2. Moreover, the hard
band width increases up to ~50 μm. The hardness values of the bulk zone
show little changes compared to those of the D8 sample.
The hardness profile in the steel region of the D16 sample is presented in
Figure 5-6(b). The hardness values decrease to a range of 110 - 130 kgf mm-2,
in the near-HAZ region. A soft zone (low hardness), ~300 μm wide, formed in
all mid-term exposed samples (from 12 to 32 days). However, the hardness
values in other regions of the steel show small changes (less than 10 kgf mm-2)
compared to those of the D8 sample.

a) 700 Bulk zone

D32
Hard band
600
Hardness (kgf mm-2)

D16
500
400
300
-2000 -1500 -1000 -500 0
b) 300 Near Far HAZ Central zone D32

HAZ
D16
Hardness (kgf mm-2)
250
200
150
100
0 500 1000 1500 2000 2500 3000
samples indicates formation of hard band, (b) hardness profile in the steel region of the
D16 and the D32 samples, indicates formation of soft zone.
Exposure at 650˚C for 32 days did not affect the hardness profiles compared
to those of the D16 sample. As shown in Figure 5-6(a), the maximum
hardness of the hard band increases to the range of 620 - 650 kgf mm-2, in the
D32 sample, while its width is similar to that observed in the D16 sample.

a) 700 Bulk zone

D128
Hard band
600
Hardness (kgf mm-2)

D64
500
400
300
-2000 -1500 -1000 -500 0
b) 300 Near Far HAZ Central zone D128

HAZ
D64
Hardness (kgf mm-2)
250
200
150
100
0 500 1000 1500 2000 2500 3000
samples, (b) hardness profile in the steel region of the D64 and the D128 samples,
indicates re-hardening of the near-HAZ region.
Moreover, the hardness values of the bulk zone in the D32 sample are almost
similar to those of the D16 sample. The hardness profile in the steel region of
the D32 sample shown in Figure 5-6(b) indicates that the soft zone hardness
values and width are similar to those of the D16 sample. The hardness values

of the far-HAZ region in the D32 sample are almost the same as those of the
central zone. Further investigations about the hardness profiles in samples of
the second set of tests, including D12, D16, D20, D24, D28 and D32 samples,
did not reveal any significant changes compared to those presented in Figure
5-6. Therefore, it is suggested that the hardness profiles in this figure are valid
for all mid-term exposed samples
The hardness profile of the weld overlay region in the D64 sample, presented
in Figure 5-7(a), indicates that increasing the exposure time did not affect the
hardness values in this region. However, the hardness profile of the steel
region in the D64 sample shows a significant change (about 50 kgf mm-2) in
the Near-HAZ region compared to that of the D32 sample. The hardness
values of the soft zone have increased to the range of 160 - 180 kgf mm-2. This
range is almost the same as that of the central zone (see Figure 5-7(b)).
In other words, increasing the exposure time to 64 days caused re-hardening
of the soft zone. Examination of hardness values in the D128 sample did not
reveal significant changes in the hardness profiles of both regions compared to
those of the D64 sample. These hardness profiles for both regions are
presented in Figure 5-7(a) and (b).
WD128 sample hardness survey
The WD128 sample was aged after having had the steel substrate machined
away and hardness measurement was carried out on the bulk zone of the
WD128 sample using microhardness method and same procedure. The
preliminary results did not reveal any significant change in the hardness of the
bulk zone in the WD128 sample compared to those of the D128 sample. In

order to investigate the formation of a hard band in this sample, the hardness
measurement in the edges of the sample was required.
The refined hardness measurement was done using a nano-indentation method
as described in section 3-6-2 at two different locations of the WD128 sample.
The hardness values at the edges of sample are reported as 7.7±0.6 GPa.
However, the hardness of the bulk zone is reported as 5.4±0.6 GPa. Therefore,
it is suggested that a hard band formed even in the absence of solid state
diffusion from the steel substrate.
5-4- Microstructural characterization
The microstructure evolution in the weld overlay and the steel regions are
linked together but in the following sections, the microstructural changes in
the different regions are represented separately. In the next section, the
microstructural evolution in the weld overlay region is reported and the
microstructural changes in the different zones of the steel region are described
in the latter section.
5-4-1- Microstructural evolution in weld

overlay region
Microstructural changes occur during the high temperature exposure. For a
better understanding of these changes, the thermally exposed samples should
be divided into following three categories:
1) Early stages of thermal exposure including samples D1 to D8

2) Mid-term thermally exposed samples including D16, D32 and D64
3) Long term thermally exposed samples including D128 and WD128
5-4-1-1- Microstructural evolution in the early
stages of thermal exposure
Examination of the D1 sample reveals that some features named as veins
formed in the mixing zone of the weld overlay.
Figure ‎5-8. Showing overall view of the weld overlay region in the D1 sample; (a) optical
microscope image in the etched condition indicates veins ~50 μm‎ long, (b) SE
micrograph in the etched condition indicates formation of the veins along the dendrites.

An optical microscope image from the D1 sample in the etched condition
shows these veins close to interface (see Figure 5-8(a)). These veins ~50 μm
long are the main features in the microstructure of the D1 sample. SE
micrograph of the interfacial region in the etched conditions presented in
Figure 5-8(b) shows that veins are formed along the dendrites. Moreover, the
length and population of these veins vary from point to point. Further
investigation confirmed that there is no significant change in the bulk zone of
the weld overlay region in the D1 sample compared to that of the as-welded
sample.
Figure 5-9(a) shows a BSE micrograph of the interfacial region in the polished
condition and indicates a vein ~30 μm long with darker contrast compared to
the matrix. It is suggested that veins contain some precipitates with lower
mean atomic number compared to that of the matrix. Figure 5-9(b) shows a
higher magnification optical microscope image of the veins in the etched
condition, indicating formation along the dendrites. Figure 5-9(c) shows a SE
micrograph in the etched condition indicating formation of veins with
different sizes at the interdendritic regions. Moreover, some fine precipitates
are visible in the large veins.
Examination of the D2 sample indicates the growth of veins. An optical
microscope image of the D2 sample presented in Figure 5-10(a) shows the
veins are ~100 μm long. Further investigations reveal that these veins outline
the grain boundaries of weld overlay region. Figure 5-10(b) presents a SE
micrograph of the D2 sample in the etched condition and the grain structure of
the weld overlay region is clearly visible in this image.

Figure ‎5-9. Showing different aspects of the veins in the D1 sample; (a) BSE micrograph
in the un-etched condition indicates a vein ~30 μm‎long with darker contrast compared
to the matrix, (b) optical microscope image in the etched condition indicates veins
locations and (c) SE micrograph in the etched condition indicates different sizes of the
veins and precipitates in the larger one.

An EDX line scan across the wide veins (see Figure 5-10(c)) reveals the Cr
and Mo concentrations are about 50% higher in veins compared to those of the
matrix. Considering higher Cr and Mo content in the veins and their dark
contrast in the BSE micrograph (see Figure 5-9(a)), it is suggested that (Cr,
Mo)-rich carbide are likely to precipitate along the grain boundaries of the
weld overlay region.
The analysis of the D4 sample reveals that growth of the veins in the
interfacial region has continued. The optical microscope image of the
interfacial region in the D4 sample in the etched condition is shown in Figure
5-10(d). In addition to the veins, some grain boundaries are present in the area
~300 μm away from the interface.
Figure 5-10(e) shows a SE micrograph of the interfacial region in the D4
sample in the etched condition. The veins have formed mostly in the
interdendritic regions. These interdendritic regions are protruding from the
matrix in the etched condition.
Further investigation with the SEM reveals that fine precipitates are present on
the grain boundaries in the area ~300 μm away from the interface. These
precipitates on the grain boundaries are shown in Figure 5-10(f). It should be
noted that the grain boundaries are present in some parts of the microstructure
where the precipitates are large enough to be observed. It is clear that the grain
boundaries in the area ~300 μm away from the interface become visible by
precipitation of fine particles along them. EDX spot analysis revealed that
some of these fine precipitates which are large enough to be analysed, are
enriched in Nb, Mo compared to the matrix.

Figure ‎5-10. Showing microstructure of the D2 and the D4 sample in the etched
condition; (a) optical microscope image of the D2 sample, (b) SE micrograph of the veins
in the D2 sample and (c) EDX line scan results of a vein in the D2 sample; (d) optical
microscope image of the D4 sample, (e) SE micrograph of the veins in the D4 sample and
(f) SE micrograph of the grain boundaries in the area‎~300‎μm‎from‎the‎interface in the
D4 sample

Microstructural analysis of the D8 sample in the etched condition, revealed a
dark etch band of ~20 μm wide, adjacent to the interface which is shown in
Figure 5-11(a). As was indicated in section 5-3, a hard band formed in the
interfacial region of the D8 sample which was ~20 μm wide. It is suggested
that formation of the dark band is related to hardening of this region. Figure
5-11(b) shows a BSE micrograph of the D8 sample and indicates more distinct
darker contrast features compared to the D1 sample (see Figure 5-9(a)). As
seen, these features are outlined the grain boundaries.
Figure ‎5-11. Showing the interfacial region of the D8 sample in different conditions; (a)
optical microscope image in the etched condition indicates the dark band ~20 μm‎wide,
b) BSE micrograph of interfacial region in the un-etched condition indicates darker
contrast grain boundaries in the area close to interface.

Figure 5-12(a1) shows a lower magnification SE micrograph of the interfacial
region in the D8 sample in the etched condition while Figure 5-12(a2) shows
an area ~300 μm distant from the interface in the same condition. The grain
boundary features are present in both of these images. Higher magnification
SE micrographs of these features in both regions are shown in Figure 5-12(b1)
and (b2) respectively. These images indicate fine precipitates along the grain
boundary features.
BSE micrographs of two regions in this sample at different distances from the
interface (i.e. ~50 μm and ~300 μm) in the un-etched condition are presented
in Figure 5-12(c1) and (c2) respectively. The grain boundary features in the
area close to the interface have darker contrast compared to the matrix.
Further investigation indicates that dark grain boundaries are present within
~100 μm distance from the interface. However, grain boundaries in the area
far from the interface contain bright contrast precipitates. It is suggested that
darker precipitates are likely to correspond carbide precipitates in the
interfacial region while the bright precipitates in the bulk zone are Mo, Nb
enriched precipitates. The changes in the microstructure of the weld overlay
region in the early stage of thermal exposure is summarised in the following
points:
1) (Cr, Mo)-rich carbide precipitation along the grain boundaries within
~100 μm distance from the interface.
2) (Nb, Mo)-rich precipitates formation along the grain boundaries in the
bulk zone.
3) Formation of a dark etch band, ~20 μm in width, adjacent to interface.

Figure ‎5-12. Showing micrographs of the interfacial region (1) and the bulk zone (2) in
the D8 sample; (a1, a2) lower magnification SE micrographs in the etched condition
indicate grain boundary features, (b1, b2) higher magnification SE micrographs indicate
fine precipitates and (c1, c2) BSE micrographs in the un-etched condition indicate the
dark precipitates in the interfacial region and bright precipitates in the bulk zone.

5-4-1-2- Microstructural evolution in the mid-term
thermally exposed samples
In this section, the microstructural evolution in three samples i.e. D16, D32
and D64 are reported. Detailed investigation on the weld overlay
microstructure in the etched condition using optical microscope revealed that
the dark etch band width was increased to ~50 μm in the D16 sample as
indicated in Figure 5-13(a).
However, as seen in this image, the length of the veins in the D16 sample did
not change significantly compared to that of the D8 sample and remained at
~100 μm. The appearance of the dark etch band is related to the etching of the
weld overlay in Aqua-Regia. For example, electro-etching of this sample
results in a different appearance to the interfacial region. A higher
magnification image of the interfacial region of the D16 sample in the electro-
etched condition is presented in Figure 5-13(b). As seen, the dark etch band
region does not appear in this condition. This image indicates that the hard
band in the D16 sample has granular structure which is different from the
dendritic structure of the bulk zone.
A higher magnification image of the bulk zone (~300 μm distance from the
interface) in the D16 sample etched in Aqua-Regia is presented in Figure
5-13(c) and shows the sub grains. It should be noted that an identical
microstructure was observed for all of the mid-term thermally exposed
samples in the etched condition, for both etching processes. Hence, the
corresponding images of the D32 and the D64 samples are not presented here
as they are very similar to those of the D16 sample.

Figure ‎5-13. Showing optical microscope images of the D16 sample weld overlay region
in the etched condition; (a) lower magnification image indicates the dark band and the
grain boundaries, (b) higher magnification image of the electro-etched sample indicates
grains in the mixing zone and (c) higher magnification image of the bulk zone indicates
formation of sub grains.

Further examinations were carried out on the microstructure of the D16
sample in the etched condition using SEM. A SE micrograph of the interfacial
region of this sample, presented in Figure 5-14(a1), shows the dark band
region and grain boundaries. The EDX area analyses results from the D16
sample have shown that composition of the interfacial region in this sample
generally is similar to the as-welded condition (see Figure 5-14(a2)).
Therefore, the formation of the hard band is not related to the compositional
changes.
A higher magnification SE micrograph of the grain boundaries in the
interfacial region is presented in Figure 5-14(b1). The EDX spot analysis of
the large particles present along the grain boundaries in this region is shown in
Figure 5-14(b2) and indicates higher Cr and Mo compared to the matrix.
As seen in Figure 5-14(c1), fine precipitates in the bulk zone are individual
precipitates which are formed in the interdendritic regions. As seen, there are
different types of precipitates in this region. There are some larger blocky
particles dispersed in the interdendritic regions. EDX spot analysis on these
particles indicates higher Nb and Ti content compared to the matrix.
Moreover, there are some irregular shape precipitates, which EDX spot
analysis on them shows higher Nb and Mo content compared to those of the
matrix (see Figure 5-14(c2)).
The microstructure of the weld overlay region in the etched condition in the
D32 and the D64 samples are similar to those of the D16 sample. However,
there are some minor differences in the interfacial region of the mid-term
thermally exposed samples which can be revealed by BSE micrographs.

Figure ‎5-14. Showing SE micrographs (1) and EDX results (2) of the weld overlay region
in the etched condition in the D16 sample; (a1) lower magnification image of the
interfacial region, (a2) EDX results for the hard band; (b1) higher magnification image
of the grain boundaries close to interface, (b2) EDX results for a particle in this region;
(c1) higher magnification image from the bulk zone indicates two different particles, (c2)
EDX results for a Mo, Nb- rich particle in this region.

Increasing exposure time causes coarsening of the (Cr, Mo)-rich carbide
precipitates along the grain boundaries within ~100 μm of the interface.
Figure 5-15(a1) shows a higher magnification BSE micrograph from the
interfacial region in the D16 sample and dark grain boundaries are hardly
visible in this image. In the same region of D64 sample the dark grain
boundaries became more distinct at the same magnification (see Figure
5-15(b1)). Figure 5-15(c1) shows the EDX line scan results across these dark
grain boundaries in the D64 sample and indicates higher Cr and Mo contents
in them.
Figure 5-15(a2) and (b2) show the region at a ~100 μm distance from the
interface in the D16 and D64 sample respectively. The dendritic
microstructure is visible in D16 sample while in the D64 sample there are
some bright precipitates at the end of dark grain boundaries. Figure 5-15(c2)
shows the EDX line scan results across the large bright particles in this region
of the D64 sample and shows a higher Mo and Nb content in them compared
to those of the matrix.
EPMA analysis was carried out on a region at a distance of ~50 μm from the
interface in the D64 sample and its results are shown in Figure 5-16. This
image shows the elemental colour maps for all elements in the weld overlay
region as well as a BSE micrograph of the analysed area (indicated in the
image). The dark grain boundary is correlated to the Cr-rich regions in the Cr
map. Moreover, the Mo map also shows a band with slightly higher Mo
concentration in the regions which are correlated to the grain boundary.

Figure ‎5-15. Showing microstructural evolution in the interfacial region (1) and the area
at‎a‎~100‎μm‎distance‎from‎the interface (2) in the mid-term samples; (a1) the interfacial
region of the D16 sample (b1) the same region in the D64 sample and (c1) the EDX
results of the dark‎grain‎boundaries‎in‎this‎region.‎(a2)‎the‎area‎at‎a‎~100‎μm‎distance‎
from the interface in the D16 sample, (b2) the same region in the D64 sample and (c2)
the EDX results of bright precipitates in this region.

Therefore, it is suggested that (Cr, Mo)-rich carbide are likely to precipitate
along the grain boundaries in this region. Moreover, there are some bright
contrast precipitates in Figure 5-16 which are correlated to higher
concentrations of Nb and Mo. It is suggested that these precipitates are related
to formation of intermetallic phases. However, it should be noted that some of
the Nb-rich points are related to the Ti-rich points too. The average
concentrations in (at.%) for the elements except carbon in this region are in
the same range compared to those of the D16 sample (see Figure 5-14(a2)).
Figure ‎5-16. Showing EPMA analysis results on the weld overlay region of the D64
sample in the un-etched condition indicates dark and bright precipitates.

EDX line scan results from the interfacial region in the polished condition in
D32 and the D64 samples are presented in Figure 5-17(a1) and (a2)
respectively and show that the concentration profiles for different elements in
this region are very similar to those of the as-welded sample (see Figure 4-12).
Further analyses were carried out on the D32 and the D64 samples in order to
identify the differences in precipitate population by Oxford-instruments
Company. BSE micrographs of an area at ~50 μm distance from the interface
in the D32 and the D64 samples are presented in Figure 5-17(b1) and (b2).
Figure ‎5-17. Showing different images for midterm exposed samples; (a1) EDX line scan
for the interfacial region of the D32 sample, (a2) EDX line scan in the D64 sample; (b1)
BSE micrograph of precipitates along the grain boundaries in the D32 sample and (b2)
BSE micrograph of precipitates around the grain boundaries in the D64 sample.

A comparison of these images shows that there are more dark precipitates
around the grain boundaries in the D64 sample compared to those of the D32
sample. These images indicate that although the length of carbide precipitation
region close to interface is constant, the population of these carbides was
increased with increasing exposure time.
Evolution processes in the weld overlay region of the mid-term thermally
exposed samples, are summarized as following points:
1) Coarsening of the carbide precipitates along the grain boundaries in the
interfacial region (within ~100 μm distance from the interface) and
formation of some particles around the grain boundaries.
2) Formation of a (Nb, Mo)-rich phase along the grain boundaries beyond
~100 μm distance from the interface and precipitation of (Nb, Ti)-rich
in the bulk zone of the weld overlay region.
3) The dark etch band composition is similar to its as-welded condition
and formed due to different etching behaviour of the hard band.
5-4-1-3- Microstructural evolution after long term
thermal exposure
In this section results of the microstructure evaluation after long term thermal
exposure are presented. The experimental observations for the D128 sample
are described in the next section. Moreover, in order to understand the effect
of steel substrate on the evolution processes, the results of the WD128 sample
are shown in the following section.

The D128 sample
Figure 5-18(a) shows an optical microscopic image of the interfacial region in
the D128 sample in the etched condition. As seen, the dark etch band width
did not change significantly compared to that of the D16 sample (see Figure
5-13). However, due to coarsening of the precipitates along the grain
boundaries in the interfacial region, they appear more distinct in this sample
compared to those of the mid-term exposed samples.
Figure ‎5-18. Showing the weld overlay region microstructure in the D128 sample; (a)
optical microscope image in the etched condition indicates the grain boundaries in the
interfacial region, (b) BSE micrograph of the interfacial region in the un-etched
condition indicates the carbide precipitation area.

Figure 5-18(b) shows a BSE micrograph of the D128 sample in the un-etched
condition and indicates a region ~100 μm wide with carbide precipitates
adjacent to interface. It is seen that even after a long term ageing process; the
carbides precipitates along grain boundaries did not grow beyond this region.
A comparison of Figure 5-18(a) and (b) indicates that grain boundaries in the
etched condition beyond ~100 μm from the interface contain different
precipitates.
Higher magnification BSE micrographs from different regions in the weld
overlay are presented in Figure 5-19 and reveal that some bright precipitates
are present beyond the carbide precipitation region. A BSE micrograph from
the region at ~250 μm distant from the interface is shown in Figure 5-19(a)
and indicates irregular shape bright precipitates adjacent to some dark features
in this region. In other words, in this region carbides and precipitates with
higher atomic number are present together. These precipitates were probably
TCP phase which formed in the interdendritic locations and along the grain
boundaries in this region.
Figure 5-19(b) shows a BSE micrograph of the region at ~100 μm distant
from the interface. It is seen that at the end of the carbide precipitation region,
there is a different morphology for TCP precipitates. There are some acicular
precipitates adjacent to carbide precipitates which are different from those
observed in the bulk zone of the weld overlay region. Finally, Figure 5-19(c)
shows carbide precipitates along grain boundaries in the interfacial region. It
is seen that the population of these precipitates adjacent to interface is less
than other parts.

Figure ‎5-19. Showing different regions of the D128 sample; (a) the BSE micrograph at
~250‎ μm‎ distant from the interface indicates TCP precipitates along the grain
boundaries, (b) the‎ BSE‎ micrograph‎ at‎ ~100‎ μm‎ distant from the interface indicates
different TCP phases in this region and (c) the BSE micrograph from the interfacial
region indicates carbide precipitates along the grain boundaries.

EPMA analyses were carried out on the different locations of the weld overlay
region in the D128 sample. The average compositions of the mixing and the
bulk zone are similar to those of the as-welded sample. Moreover, some
elemental mappings were carried out on this sample. The first analysis was
carried out on a region at ~50 μm distant from the interface.
Figure 5-20(a) shows a BSE micrograph in this region which indicates the
analysed area. Figure 5-20(b) shows the elemental colour maps for this region
and due to larger size of carbide precipitates in this sample, the carbon map
shows higher concentrations in the regions which are correlated to the grain
boundary. Although the Cr map is also showing that grain boundaries are Cr-
rich, it is clear that Cr is uniformly distributed in the matrix of this region.
The Mo map shows two different precipitation behaviours; in the regions
which are correlated to the grain boundary there are higher concentrations of
Mo which is similar to the Cr map. Nevertheless, in the region which is
correlated to the bright precipitate there is a higher concentration of Mo which
is similar to Nb map. Moreover, there are some isolated, high concentration
regions of Ti and Nb which are correlated to the brighter contrast points in the
interdendritic regions.
The next EPMA was carried out on the region at ~100 μm distant from the
interface in the D128 sample where acicular precipitates are present. The
analysed area is shown in Figure 5-21(a) and the corresponding elemental
maps are presented in Figure 5-21(b). As seen there are two Mo, Nb-rich
points and one Nb, Ti-rich point.

Figure ‎5-20. Showing EPMA results for a‎region‎at‎~50‎μm‎distant‎from‎the‎interface in

the D128 sample; (a) the BSE micrograph indicates the analysed area, (b) The colour
maps for the elements.

Figure ‎5-21. Showing EPMA results for a‎region‎at‎~100‎μm‎distant‎from‎the‎interface‎in‎

the D128 sample; (a) the BSE micrograph indicates the analysed, (b) The colour maps
for the elements.

It is seen that two red points in the Mo map are correlated to the brighter
blocky particles in the BSE micrograph. Moreover, evaluation of the Nb map
indicates that although the Nb-rich regions are correlated to the Mo-rich
regions, the highly Nb-rich points are in different positions compared to those
of the Mo-rich points. Moreover, it is observed that some parts of the Mo-rich
regions in the bright particles are correlated to the Cr-rich regions.
Nevertheless, some of the bright precipitates are corresponding to regions
which are enriched in Nb and Ti.
The WD128 sample
Figure 5-22(a) shows an optical microscope image of the WD128 sample in
the etched condition which is different compared to that of the D128
microstructure (see Figure 5-18(a)). The dark etch band is not seen adjacent to
the interface; however, nano-indention hardness measurement results indicate
a higher hardness for the interfacial region compared to the bulk zone in the
WD128 sample. The grain boundaries in the interfacial region of the WD128
sample are less distinct compared those of the D128 sample.
An EDX line scan across the grain boundaries in the interfacial region of the
WD128 sample (see Figure 5-22(b)) reveals some (Mo, Nb)-rich precipitates
along the grain boundaries. Further investigation was carried out in the un-
etched condition. A BSE micrograph of the WD128 sample presented in
Figure 5-22(c) shows some bright contrast precipitates along the grain
boundaries in the interfacial region. There is no evidence for carbide
precipitation in this region which is different compared to the D128 sample
(see Figure 5-18(b)).

Figure ‎5-22. Showing the interfacial region in the WD128 sample; (a) an optical
microscope image in the etched condition indicates grain boundaries, (b) EDX line scan
results for the precipitates along the grain boundaries and (c) a BSE micrograph in the
un-etched condition indicates the TCP phase precipitates along the grain boundaries
close to the interface.

It is concluded that formation of the hard band is not related to presence of the
dark etch band and the steel substrate. A comparison of BSE micrographs for
the D128 and the WD128 samples indicates that presence of steel substrate
has a great influence on the type of precipitates along the grain boundaries in
the interfacial region. In the D128 sample, carbide precipitates are present
while TCP precipitates have formed along the grain boundaries in the WD128
sample.
Figure ‎5-23. Showing different regions in the bulk zone of the WD128 sample in the un-
etched condition (a) a BSE micrograph of a region‎at‎~250‎μm‎distant from the interface
indicates larger TCP precipitates at interdendritic regions, (b) a BSE micrograph of a
region‎ at‎ ~100‎ μm‎ distant from the interface indicates the smaller TCP precipitates
along grain boundaries.

Figure 5-23(a) shows a higher magnification BSE micrograph from a region at
~250 μm distant from the interface and shows the individual bright contrast
precipitates (TCP phase) which is similar to the D128 sample (see Figure
5-19(a)). A BSE micrograph from a region at ~100 μm distant from the
interface (presented in Figure 5-23(b)) shows some fine bright contrast
precipitates (TCP phase) along the grain boundaries which are smaller
compared to other regions. Moreover, the acicular bright precipitates did not
appear in this sample.
5-4-2- Microstructural evolution in the steel

region
As mentioned in chapter 4, there are five regions with different
microstructures in the steel region of the as-welded sample. The
microstructural evolution in each region is dependent upon the initial
microstructure. Three main regions are defined here for investigations of the
microstructural changes in the steel region.
The first and the second regions were chosen at ~100 μm and ~500 μm distant
from the interface which are in the near-HAZ and the far-HAZ regions
respectively. The third region was chosen at a ~1 mm distant from the inner
bore of tube which is in the central zone. Different analyses were carried out
on the microstructure of these regions to investigate the microstructural
evolution in each region. As described in section 5-2, different processes are
defined for changes in the microstructure of 2.25Cr1Mo steel region. These
processes and related samples are described in the following sections:

1) Tempering of the near-HAZ region is shown by the D1 and the D8
samples microstructure.
2) Soft zone formation is described by presentation of two mid-term
thermally exposed samples i.e. the D16 and D32 samples.
3) Re-hardening of the soft zone in the D64 sample and fine grain
formation in the D128 sample are shown together.
4) Formation of a network of carbides at the interface with increasing
exposure time is shown in the D32, the D64 and the D128 samples.
5-4-2-1- Tempering of the near-HAZ region
Figure 5-24(a1) shows an optical microscope image for the steel region in the
D1 sample in the etched condition which is similar to the as-welded sample
(see Figure 4-2(b)). Moreover, the locations of the main regions are shown in
this image. The optical microscope image of the D8 sample in the etched
condition is presented in Figure 5-24(a2). The overall view of the D8 sample
in this condition is similar to that of the D1 sample. However there are some
white/bright regions adjacent to the interface.
Figure 5-24(b1) and (b2) are SE micrographs from the interfacial region in the
D1 and the D8 samples in the etched condition. These images indicate some
precipitate free regions adjacent to the interface in the D8 sample. BSE
micrographs from the interfacial region of the D1 and the D8 samples in the
un-etched condition are shown in Figure 5-24(c1) and (c2) which indicate
formation of some fine bright precipitates in the D8 sample.

Figure ‎5-24. Showing different images of the D1 (1) and the D8 (2) samples; (a1, a2)
optical microscope images, (b1, b2) SE micrographs indicate the precipitate free regions
in the D8 sample and (c1, c2) BSE micrographs in un-etched condition.

Figure ‎5-25. Showing higher magnification optical microscope images from the D1 (1)
and the D8 (2) samples; (a1, a2) the near-HAZ region microstructure, (b1, b2)
microstructure in the far-HAZ region and (c1, c2) bimodal microstructure in the central
zone.

Higher magnification optical microscope images of the main regions in both
samples are presented in Figure 5-25. The near-HAZ region in the D1 sample
is shown in Figure 5-25(a1) and indicates a fine and uniform microstructure in
this region. Figure 5-25(a2) shows the same region in the D8 sample which is
not uniform, and adjacent to the interface, exhibits a coarser microstructure
compared to other regions. The far-HAZ region in the D1 and the D8 samples
are presented in Figure 5-25(b1) and (b2) respectively. The fine grains are
observed in this region of both samples. Finally, the higher magnification
optical microscope images from the central zone of the D1 and the D8
samples are shown in Figure 5-24(c1) and (c2) respectively. The bimodal
microstructure (ferrite grains and TM/B) of the central zone is seen in both
samples.
SE micrographs of the main regions in the etched condition for these samples
are presented in Figure 5-26. Figure 5-26(a1) shows the near-HAZ region
microstructure in the D1 sample and indicates uniform tempered structure.
The near-HAZ region microstructure in the as-welded sample is shown in
Figures 4-6(b) and (c) which exhibits a lath-like structure. After one day
exposure at 650˚C, the structure was changed to a tempered structure with fine
precipitates. The near-HAZ region in the D8 sample shown in Figure 5-26(a2)
indicates larger grains with fine precipitates. Figure 5-26(b1) shows the
microstructure of the far-HAZ region in the D1 sample which contains two
different types of precipitates named as coarse and fine precipitates. As
mentioned in section 4-3-4, in the as-welded sample, the far-HAZ region
microstructure is non-uniform with un-dissolved carbides.

Figure ‎5-26. Showing higher magnification SE micrographs for different regions in the
D1 (1) and the D8 (2) samples; (a1, a2) the near-HAZ region, (b1, b2) the far-HAZ
region and (c1, c2) the central zone.

After 1 day exposure, the tempered non-uniform microstructure is observed in
the far-HAZ region. Those regions with coarse carbides are related to the
former TM/B areas while the fine carbide dispersed in the matrix are related to
the former ferrite grains. Figure 5-26(b2) shows a SE micrograph from the far-
HAZ region in the D8 sample and indicates fine grains in this region.
Moreover, some carbides in this sample are coarser than to those observed in
the D1 sample. The microstructural evolution in the far-HAZ region is carbide
coarsening between the ferrite fine grains.
SE micrographs of the central zone in the D1 and the D8 samples are
presented in Figure 5-26(c1) and (c2) which are similar to the as-welded
sample. The bimodal microstructure is visible in these samples. However, the
grain boundaries are more distinct in the D8 sample compared to those of the
D1 sample. Moreover, some fine precipitates are seen within the ferrite grains
in the D8 sample.
Higher magnification BSE micrographs of the near-HAZ region in the D1 and
the D8 samples are shown in Figure 5-27(a1) and (a2). There are a few bright
acicular precipitates in the D8 sample. Figure 5-27(b1) shows a BSE
micrograph of the far-HAZ region in the D1 sample and there are some
regions where a few fine round precipitates are present which shows non-
uniform precipitation in the far-HAZ region after a day exposure. Figure
5-27(b2) shows the far-HAZ region in the D8 sample and there are more and
coarser bright precipitates compared to the D1 sample. The BSE micrographs
of the central zone in these samples are presented in Figure 5-27(c1) and (c2)
and show some precipitates on the ferrite grain boundaries in the D8 sample.

Figure ‎5-27. Showing higher magnification BSE micrographs for different regions in the
D1 (1) and the D8 (2) samples; (a1, a2) the near-HAZ region, (b1, b2) precipitates in the
far-HAZ regions and (c1, c2) fine precipitates on the grain boundaries in the central
zone of D8 sample.

5-4-2-2- Formation of soft zone
Microstructural analysis of thermally exposed samples reveals that after 12
days exposure, a soft zone with low hardness and a large grain structure has
formed adjacent to the interface. A soft zone was observed in samples aged up
to 32 days. In this section only the microstructures of the D16 and the D32
samples are presented. Figure 5-28(a1) shows an optical microscope image of
the interfacial region in the D16 sample in the etched condition where a white
zone ~300 μm wide is seen adjacent to the interface. Based on the hardness
test results (see section 5-3), this region has lower hardness values compared
to the other regions. The optical microscope image of the D32 sample in the
etched condition is shown in Figure 5-28(a2) which is similar to that of the
D16 sample.
Figure 5-28(b1) and (b2) show SE micrographs of the interfacial region in the
etched condition in these samples and indicate large grain structures with fine
precipitates. The soft zone in the D32 sample contains fewer and coarser
precipitates in the interfacial region compared to those of the D16 sample.
BSE micrographs of these samples in the un-etched condition shown in Figure
5-28(c1) and (c2), revealed a noticeable change in the microstructure. The
D16 sample microstructure is similar to the D8 sample. However, in the D32
sample some bright precipitates are visible which are non-uniformly
distributed in the near-HAZ region. The Mo-rich phase (possibly M6C
carbide) has a higher mean atomic number compared to that of the matrix
(Fe). Therefore bright precipitates can be attributed to the formation of the
Mo-rich carbide precipitates.

optical microscope images indicate the soft zone, (b1, b2) SE micrographs indicate the
large grains in the soft zone and (c1, c2) BSE micrographs indicate fine bright Mo-rich
carbide precipitates in the D32 sample.

Figure ‎5-29. Showing higher magnification optical microscope images from different
regions in the D16 (1) and D32 (2) samples in the etched condition; (a1, a2) show the soft
zone adjacent to the interface, (b1, b2) indicate fine grains in the far-HAZ region and
(c1, c2) the bimodal microstructure in the central zone.

Higher magnification optical microscope images of the D16 and the D32
samples are presented in Figure 5-29. The soft zones in these samples are
shown in Figure 5-29(a1) and (a2) which are seen to contain large grains (up
to several hundred microns in size). The microstructures of the far-HAZ
region are presented in Figure 5-24(b1) and (b2). The grain size in this region
is less than 10 μm and its microstructure is similar to that of the D8 sample.
The bimodal microstructure in the central zone of these samples is shown in
Figure 5-29(c1) and (c2) respectively.
SE images of the D16 and the D32 samples in the etched condition are shown
in Figure 5-30. SE micrographs of the soft zones in these samples are
presented in Figure 5-30(a1) and (a2) which show some fine precipitates
uniformly dispersed within the large grains. There are some coarse
precipitates in this region too. Grain boundaries in these samples are not
delineated by precipitates. There are more coarse precipitates in the D32 than
in the D16 sample. Moreover, the coarse precipitates are dispersed more
uniformly on the grain boundaries and within the grains in the D32 sample.
The far-HAZ region microstructures of these samples are presented in Figure
5-30(b1) and (b2) respectively. It seems that these images are similar to that of
the D8 sample and contain fine grains and some precipitates. Figure 5-30(c1)
and (c2) show the microstructure of the central zone in these samples. The
bimodal microstructure is clearly seen in these images too. However, the
precipitates on the ferrite grain boundaries are coarser in the D32 sample
compared to those in the D16 sample. Moreover, fine precipitates within the
ferrite grains are fewer and coarser in the D32 sample.

Figure ‎5-30. Showing higher magnification SE micrographs from different regions in the
D16 (1) and the D32 (2) samples in the etched condition; (a1, a2) precipitates in the near-
HAZ region, (b1, b2) microstructure in the far-HAZ region and (c1, c2) precipitates on
ferrite grain boundaries in the central zone.

BSE micrographs of the soft zone in the D16 and D32 samples are shown in
Figure 5-31(a1) and (a2). As seen, there are more bright precipitates and fewer
dark precipitates in the D32 sample compared to those of the D16 sample. A
comparison between the D8 sample (see Figure 5-27(a2)) and the D32 sample
(see Figure 5-31(a2)) shows that number and size of bright precipitates
increases with increasing exposure time. These elongated precipitates in the
near-HAZ region have higher mean atomic number compared to the matrix.
This indicates that Mo-rich precipitates (possibly M6C carbides) have grown
in the near-HAZ after 32 days exposure. The distribution of these precipitates
is not uniform in the matrix. In contrast, the dark precipitates in the D32
sample are fewer compared to those in the D16 sample. This indicates that
other types of carbides i.e. (Fe and/or Cr) carbides were dissolved in the
matrix with increasing exposure time.
Figure 5-31(b1) and (b2) show BSE micrographs of the far-HAZ regions in
these samples. Two different types of precipitates (bright and dark) are present
in these images. There is no significant difference between these micrographs.
It seems that the far-HAZ region microstructure did not change significantly
with increasing exposure time up to 32 days. The morphology of the bright
precipitates in the far-HAZ region is different compared to that of the near-
HAZ region in both samples. The size of elongated particles is up to several
microns in the near-HAZ region while in the far-HAZ region blocky
precipitates are very smaller in size. It seems that the fine precipitates in the
far-HAZ region did not grow with increasing exposure time, as their quantity
and size in the D32 sample is similar to those in the D16 sample.

Figure ‎5-31. Showing higher magnification BSE micrographs from different regions in
the D16 (1) and the D32 (2) samples; (a1, a2) precipitates in the near-HAZ region, (b1,
b2) indicate similar microstructure in the far-HAZ region and (c1, c2) dark precipitates
on the ferrite grain boundaries.

Finally, BSE micrographs of the central zone in these samples are seen in
Figure 5-31(c1) and (c2). It seems that some bright precipitates (possibly Mo-
rich M6C carbide) and some dark precipitates (possibly Fe/Cr-rich carbides)
are present along the ferrite grain boundaries in both samples. However, these
precipitates in the D32 sample are coarser compared to those of the D16
sample.
Further investigations were carried out on the D16 sample using EPMA to
examine the differences between the near-HAZ and the far-HAZ regions. The
elemental colour maps for these regions are presented in Figure 5-32. It should
be noted that the average of Cr and Mo in both regions are in the same level.
The Cr distribution in the soft zone is more uniform compared to that of the
far-HAZ region (see Figure 5-32(a1) and (a2)). However, the Mo map in the
far-HAZ region (see Figure 5-32(b2)) shows highly concentrated regions
while in the soft zone (see Figure 5-32(b1)), the Mo concentration is almost
uniform. The C maps for both points are presented in Figure 5-32(c1) and (c2)
and do not show any significant features due to its low concentration.
The elemental maps for the alloying elements indicate that Cr and Mo in the
far-HAZ region are concentrated in the precipitates. These maps confirm that
Mo-rich precipitates (possibly M6C carbide) are present in the far-HAZ region
of the D16 sample. However, considering the fully Tγ structure in the near-
HAZ region of the as-welded sample, the uniform Cr and Mo maps in the soft
zone of the D16 sample must be related to smaller precipitates after tempering
in this region. In other words, it is suggested that these did not grow
significantly even after 16 days exposure.

Figure ‎5-32. Showing EPMA elemental colour maps for alloying elements in the near-
HAZ region (1) and the far-HAZ region (2) of the D16 sample; (a1, a2) relatively
uniform Cr distribution in both regions, (b1, b2) Mo-rich points in the far-HAZ region
and (c1, c2) the C distribution maps which are not valid due to negative concentrations.

EDX analyses were performed on the different precipitates in the interfacial
region of the D32 sample and the results are presented in Figure 5-33(a). As
seen in this image, two different precipitates are present in this region. There
are some Mo-rich precipitates in the near-HAZ region (soft zone) and some
(Cr, Mo)-rich precipitates adjacent to the interface. Figure 5-33(b) shows EDX
results in the far-HAZ region and shows there are some (Cr)-rich precipitates
in this region, in addition to (Mo)-rich precipitates.
Figure ‎5-33. Showing EDX analysis results in different regions of the D32 sample; (a) the
Mo-rich elongated precipitates in the near-HAZ region (soft zone) and (Cr, Mo)-rich
precipitates adjacent to the interface which is shown in higher magnification too, (b) the
Mo-rich and the Cr-rich precipitates in the far-HAZ region.

Further investigations were carried out on the D32 sample using EBSD in the
Oxford Instrument laboratory. The small grains in the far-HAZ region and
TM/B areas in the central zone can be easily observed with this process. The
large grains in the near-HAZ region (soft zone) and the fine grains in the far-
HAZ region are shown in Figure 5-34(a) and (b). The central zone is shown in
Figure 5-34(c) which indicates ferrite sub grains in the TM/B areas.
Figure ‎5-34. Showing EBSD results for the D32 sample indicates; (a) large grains in the
soft zone (b) fine grains in the far-HAZ region and (c) ferrite sub grains in TM/B areas.

5-4-2-3- Re-hardening and fine grain formation in
the near-HAZ region
Microstructural investigations were carried out on the D64 and the D128
samples and the results are presented in this section. Based on the hardness
test results shown in Figure 5-7(b), these samples have a similar hardness
profile. As mentioned in section 5-3, the soft zone which was formed in the
near-HAZ region in the mid-term thermally exposed sample was found to be
re-hardened in the D64 sample. Moreover, a little change was observed in the
D128 sample hardness compared to that of the D64 sample. This fact is related
to some changes in the microstructure which are shown schematically in
Figure 5-3.
The overall view images from the interfacial region in these samples are
shown in Figure 5-35. The optical microscope image of the interfacial region
in the D64 sample is presented in Figure 5-35(a1). As seen, the microstructure
in this sample is similar to that of the mid-term thermally exposed samples
(see Figure 5-28(a1) and (a2)). Although the hardness values of this region are
significantly increased, large grains are still seen in this image.
The optical microscope image of the interfacial region in the D128 sample is
shown in Figure 5-35(a2) and revealed a significant change in the
microstructure. As observed, the large ferrite grains were not seen in the near-
HAZ region after a long term thermal exposure which is a very unusual. SE
micrographs of the interfacial region in the etched condition in both samples
are shown in Figure 5-35(b1) and (b2) respectively.

optical microscope images indicate significant change in the near-HAZ region, (b1, b2)
SE micrographs indicate the grain size changing and (c1, c2) BSE micrographs indicate
different morphologies for bright precipitates in the near-HAZ region.

The SE micrograph of the D64 sample in the etched condition contains large
grains with some coarse and fine precipitates dispersed in the matrix.
However, the interfacial region in the D128 sample (see Figure 5-35(b2)) is
similar to the tempered structure of the D8 sample (see Figure 5-24(b2)). In
other words, the D128 sample microstructure contains fine grains and some
precipitates.
The BSE micrograph of the interfacial region of the D64 sample is presented
in Figure 5-35(c1) and is similar to that of the D32 sample (see Figure
5-28(c2)). As previously shown, the Mo-rich precipitates (possibly M6C
carbide) are present in the matrix. However, the morphology of these
precipitates in the D64 sample is different compared to those of the D32
sample. Moreover, the population and size of these precipitates did not
increase significantly. The BSE micrograph of the interfacial region in the
D128 sample in the un-etched condition contains Mo-rich precipitates (see
Figure 5-35(c2)). However, their morphology and distribution are significantly
different compared to those of the D64 sample.
Figure 5-36 shows higher magnification optical microscope images of the
main regions in these samples in the etched condition. It seems that all images
of the D64 sample presented in Figure 5-36 (a1), (b1) and (c1) are similar to
those of the mid-term thermally exposed samples. Nevertheless, higher
magnification optical microscope image of the near-HAZ region in the D128
sample (see Figure 5-36(a2)) shows a very unusual feature. It is seen that grain
size in this region is less than 10 μm which are very smaller compared to that
of the D64 sample. (i.e. there has been a substantial grain size reduction)

Figure ‎5-36. Showing higher magnification optical microscope images from different
regions in the D64 (1) and D128 (2) samples in the etched condition; (a1, a2) indicate the
grain size reduction in the near-HAZ region, (b1, b2) indicate similar far-HAZ regions
and (c1, c2) indicate the bimodal microstructure in the central zone.

Figure 5-36(b2) and (c2) show the far-HAZ region and the central zone
microstructures in the D128 sample respectively. There is no significant
difference between these images and those of the mid-term thermally exposed
samples. It is suggested that the bimodal microstructure remained largely
unchanged during these later stages of thermal exposure up to 128 days.
High magnification SE micrographs of the main regions in the etched
condition are presented in Figure 5-37. The first row shows the D64 sample
micrographs and they did not reveal any significant difference compared to
those of the D32 sample (see Figure 5-30(a2), (b2) and (c2)). The near-HAZ
region microstructure (see Figure 5-37(a1)) contains large grains with some
coarse and fine precipitates which are dispersed in the matrix. The far-HAZ
region microstructure is presented in Figure 5-37(b1) and shows fine grains.
Finally, the central zone microstructure is shown in Figure 5-37(c1) which
exhibits the bimodal microstructure. However, there are fewer and coarser
precipitates within the ferrite grains and larger precipitates along grain
boundaries in this sample compared to the D32 sample.
High magnification SE micrographs from the different regions of the D128
sample in the etched condition are presented in Figure 5-37(a2), (b2) and (c2).
The near-HAZ region microstructure is shown in Figure 5-37(a2) and contains
fine grains with some precipitates. This structure is similar to tempered
structure of the D8 sample (see Figure 5-27(a2)). There are no significant
differences between this sample and the D64 sample regarding the
microstructure of other regions. However, precipitates in the central zone have
grown with increasing exposure time.

Figure ‎5-37. Showing higher magnification SE micrographs from different regions in the
D64 (1) and the D128 (2) samples in the etched condition; (a1, a2) indicate the near-HAZ
region, (b1, b2) indicate the far-HAZ region and (c1, c2) indicate precipitates growth in
the central zone.

High magnification BSE micrographs from these regions in the D64 sample
are presented in Figure 5-38(a1), (b1) and (c1). In all of these micrographs,
fine dark precipitates are visible in the microstructure in addition to Mo-rich
bright precipitates. It is suggested that the dark precipitates are related to the
formation of Cr-rich precipitates (possibly M7C3 and M23C6) with lower mean
atomic number compared to the matrix. Moreover, the morphology of the
bright precipitates is different in these regions. The bright precipitates are
coarser in the near-HAZ region compared to those of the far-HAZ region (see
Figure 5-38(b1)). Moreover, these precipitates are present on the ferrite grain
boundaries in the central zone (see Figure 5-38(c1)).
The BSE micrographs from different regions of the D128 sample in the
polished condition are shown in Figure 5-38(a2), (b2) and (c2). The near-HAZ
region micrograph in the D128 sample (see Figure 5-38(a2)) is different
compared to that of the D64 sample and shows some chains of coarse bright
precipitates in some locations and some fine bright precipitates which are
dispersed in the matrix. Moreover, there are fewer coarse bright and fine dark
precipitates in the near-HAZ region of the D128 sample compared to those of
the D64 sample. The far-HAZ region and the central zone micrographs in the
D128 sample (see Figure 5-38(b2) and (c2)) contain similar features (bright
and dark precipitates). However, morphologies and distribution of these
precipitates are different compared to those of the near-HAZ region.
Generally, it is seen that dark precipitates are fewer in the D128 sample
compared to those of the D64 sample. This can be related to formation of (Cr,
Mo)-rich carbides with a mean atomic number close to that of the matrix.

Figure ‎5-38. Showing higher magnification BSE micrographs from different regions in
the D64 (1) and the D128 (2) samples; (a1, a2) indicate precipitates in the near-HAZ
region, (b1, b2) indicate the far-HAZ region precipitates and (c1, c2) indicate fewer dark
precipitates on the ferrite grain boundaries with increasing exposure time.

Comparison between BSE micrographs of the near-HAZ region in the D32
(see Figure 5-31(a2)) and the D64 (see Figure 5-38(a1)) samples shows that
the bright precipitates morphology changes from elongated to coarser irregular
particles. This means further growth of these precipitates with increasing
exposure time. Moreover, there are a lot of fine dark precipitates which are
uniformly dispersed in the matrix of the D64 sample which are related to Cr-
rich carbides and attributed to re-hardening of this region.
The presence of dark precipitates observed in the BSE micrographs of the
near-HAZ region is complex. In the D64 sample, there are more dark
precipitates compared to those in the D32 sample. However, these features are
fewer in the D128 sample than those observed in the D64 sample. It seems
that increasing exposure time results in precipitation of Cr-rich M23C6 in
addition to the former Cr-rich M7C3. However, these carbides convert to (Cr,
Mo)-rich M23C6 with a mean atomic number close to matrix which are not
seen in the BSE micrographs.
In order to identify these precipitates, further investigations were carried out
on the D128 sample using EBSD and the results are shown in Figure 5-39.
The fine grains in the steel region adjacent to interface are clearly visible in
the IPF (Inverse Pole Figure) map of this region (see Figure 5-39(a)). The
phase discrimination map of the interfacial region is shown in Figure 5-39(b)
and shows many mismatched (black) points. The precipitates contain some
points which match with Cr23C6 while other points match with Mo2Fe4C. Both
of these phases i.e. Cr23C6 and Mo2Fe4C have FCC structure; therefore it is
suggested that (Cr, Mo)-rich precipitates were formed in this region.

Figure ‎5-39. Showing EBSD results from different regions of the D128 sample; (a) IPF
map from the interfacial region indicates the interface and fine grain in this region, (b)
higher magnification phase discrimination map of the near-HAZ region indicates two
different phases in the precipitates and (c) higher magnification phase discrimination
map of the far-HAZ region indicates only M6C phase in the precipitates.

Figure 5-39(c) shows the phase discrimination map in the far-HAZ region, and
it is clear from BSE micrographs that finer precipitates are present in this
region. The precipitates in the far-HAZ region are mainly M6C with no
evidence for M23C6 phase. This issue is related to limitation of W filament
electron gun in producing phase discrimination maps.
Advanced microstructural examination
Based on the above observations, it was deemed necessary to identify the
bright and dark contrast precipitates of BSE images in the near-HAZ region of
the D64 and the D128 samples. Moreover, the unusual grain size reduction in
this region from the D64 to the D128 samples should be investigated;
specifically how this might be related to changes in the precipitates. Therefore
further analyses were carried out using ion beam imaging and FIB/TEM (see
section 3-5-4 and 3-5-5) on these samples in the Loughborough University.
The ion beam induced secondary electron (ISE) images of these samples are
presented in Figure 5-40. The interfacial regions in the D64 and D128 samples
are shown in Figure 5-40(a1) and (a2) respectively. A comparison of these
images shows a significant change in the grain size of the near-HAZ region.
Figure 5-40(b1) and (b2) show higher magnification ISE images from the
near-HAZ region after 1 raster in the D64 and the D128 samples respectively.
It is observed that there are a lot of fine precipitates in the D64 sample (see
Figure 5-40(b1)) and that these have grown with increasing exposure time to
128 days (see Figure 5-40(b2)). It is clear that further raster results in a
changing in channelling of ISE images. Therefore it is possible to observe the
grain boundaries.

Figure ‎5-40. Showing ion beam induced secondary images of the D64 (1) and the D128
(2) samples; (a1, a2) low magnification ISE images from the interfacial region indicate
grain size difference in the near-HAZ region, (b1, b2) higher magnification ISE images
of near-HAZ after 1 raster indicate precipitate growth with increasing exposure time
and (c1, c2) higher magnification ISE images of the same region after 3 raster indicate
precipitates along the grain boundaries in the D128 sample.

Higher magnification ISE images from the same regions after 3 raster in both
samples are presented in Figure 5-40(c1) and (c2). It is observed that the fine
precipitates in the D64 sample are scattered in the matrix. However, the
precipitates in the D128 sample are mostly along the grain boundaries in the
near-HAZ region.
Further investigation using FIB-TEM was carried out and the high angle
annular dark field (HAADF) scanning transmission electron microscope
(STEM) micrographs from two regions of these samples are presented in
Figure 5-41. The far-HAZ region of the D64 sample is shown in Figure
5-41(a1) and indicates fine grains and precipitates in this region. The near-
HAZ region of this sample are shown in Figure 5-41(b1) and (c1) with
different magnifications. It is observed that the grain boundaries are not
visible in the near-HAZ region of the D64 sample. However, the fine
precipitates (nano-particles) are clearly visible in both images.
Similarly the HAADF-STEM micrographs of the D128 sample are presented
in Figure 5-41(a2), (b2) and (c2). Figure 5-41(a1) shows the far-HAZ region
and indicates that it is similar to that of D64 sample. However, the low
magnification image of the near-HAZ region in the D128 sample (see Figure
5-41(b2) indicates grain boundaries. Moreover, the higher magnification
image of this region shows only one coarse precipitate (see Figure 5-41(c2)).
The EDS maps of the main elements (Fe, Cr and Mo) for Figure 5-41(c1) and
(c2) are presented in Figure 5-42. The Fe maps indicate the location of
precipitates in these regions and their size. Figure 5-42(a1) shows precipitate
size is less than 50 nm while Figure 5-42(a2) shows a ~200 nm particle.

Figure ‎5-41. Showing HAADF STEM micrographs of the D64 (1) and the D128 (2)
samples; (a1, a2) low magnification images from the far-HAZ region indicate similar
microsructure, (b1, b2) low magnification images from the near-HAZ region indicate
fine grain formation with increasing exposure time and (c1, c2) higher magnification
images from the near-HAZ region indicates precipitate growth with increasing exposure
time.

Comparison between Cr maps in the D64 (see Figure 5-42(b1)) and the D128
(see Figure 5-42 (b2)) samples indicates that Cr-rich precipitates grow in size
with increasing exposure time. It is observed that many fine Cr-rich
precipitates (less than ~50 nm size) are present in the D64 sample while there
is only one coarse Cr-rich precipitate (~200 nm size) in the D128 sample.
A comparison of Mo maps in both samples (see Figure 5-42(c1) and (c2)),
indicates the growth of Mo-rich precipitates with increasing exposure time.
Moreover, it is observed that coarse precipitate in the D128 sample is a (Cr,
Mo)-rich phase. It should be noted that Mo-rich part of this precipitate is
different from the Cr-rich part.
A comparison between the Cr and Mo maps in the D64 sample (Figure
5-42(b1) and (c1)) reveals that there are 3 types of precipitates in this sample.
It is seen that a number of precipitates are (Cr, Mo)-rich. However, there are
some precipitates which are either Cr-rich or Mo-rich precipitates. As such, it
is suggested that in the near-HAZ region of the D64 sample, fine Cr-rich and
(Cr, Mo)-rich precipitates are present in the microstructure in addition to
coarse Mo-rich precipitates in BSE micrographs (see Figure 5-38(a2)).
The fine (Cr, Mo)-rich precipitates are likely to be M23C6 carbide and their
presence causes re-hardening of the near-HAZ region (former soft zone) in the
D64 sample. Moreover, these precipitates have grown with increasing
exposure time and in the D128 sample only a coarser (Cr, Mo)-rich precipitate
(possibly M23C6 carbide) is present in the microstructure. It is observed that
these precipitates are present mainly on ferrite grain boundaries (see Figure
5-40 (c2)).

Figure ‎5-42. Showing EDS maps for key elements for the D64 (1) and the D128 (2)
samples with reference to Figures 5-42(c1) and (c2); (a1, a2) Fe maps indicate location of
precipitates and their growth with increasing exposure time, (b1, b2) Cr map for
precipitates and (c1, c2) Mo map for precipitates.

5-4-2-4- Formation of carbides at interface
The interfacial regions of the D32 and the D64 samples were analysed in the
Oxford Instruments labratory in order to identify the precipitates at the
interface and the results are shown in Figure 5-43. The low magnification BSE
micrographs of the interfacial region in these samples are presented in Figure
5-43(a1) and (a2) respectively. A comparison of these images shows that more
bright contrast precipitates exist at the interface of the D64 sample compared
to those of the D32 sample. Moreover, higher magnification micrographs from
the interface region indicate coarser precipitates in the D64 sample (Figure
5-43(b2)) compared to those of the D32 sample (Figure 5-43(b2)).
Further analyses were carried out using EBSD and EDS on the D64 sample.
The superimposed results of EBSD with the EDS for one precipitate adjacent
to the interface is presented in Figure 5-43(c) and shows a F.C.C structure
which is related to the Cr23C6 carbide. However, it is seen that some parts of
this precipitate did not match with the Cr23C6 structure. Analysis of the
mismatched parts with EDS shows a higher Mo concentration. Therefore, it is
suggested that these precipitates are (Cr, Mo)-rich carbide, possibly M23C6
carbide. Moreover, the carbide precipitates at the interface have grown with
increasing exposure time.
The precipitates at the interface in the D64 sample are easily identified by the
Cr and Mo peaks in the EDX line scans across the interface in this sample.
Figure 5-44(a) shows a BSE micrograph from the interfacial region in the D64
sample with Cr and Mo peaks adjacent to interface.

Figure ‎5-43. Showing precipitates at the interface region of the D32 (1) and the D64 (2)
samples; (a1, a2) low magnification BSE micrographs indicate precipitates growth with
increasing exposure time, (b1, b2) higher magnification BSE micrographs and (c) EBSD
analysis with EDS maps for one precipitate in the D64 sample.

Similar EDX analyses were carried out on the interfacial region of the D128
sample in the etched condition. The results confirm the presence of the (Cr,
Mo)-rich carbides adjacent to the interface (see Figure 5-44(b)). Moreover, the
SE micrograph in the etched condition indicates a network of these carbides
formed at the interface. A comparison of these images shows that Cr-rich
region width increased with extending exposure time.
Figure ‎5-44. Showing (Cr, Mo)-rich carbides in the D64 and the D128 samples; (a) BSE
micrograph indicate higher Cr and the Mo concentrations in the region adjacent to
interface in the D64 sample (b) SE micrograph in the etched condition indicates a
network of particles adjacent to the interface in the D128 sample in the Cr-rich region.

Advanced microstructural examination
Based on the above presented results, it was deemed that further analyses were
required for characterization of carbide precipitates at the interface region in
the D64 and the D128 samples. Therefore, further analyses were carried out
using FIB/TEM (see section 3-5-4) on these samples in the Loughborough
University. The HAADF STEM micrographs from the interface region in the
D64 and the D128 samples are presented in Figure 5-45(a1) and (a2)
respectively.
These dark field images show a chain of precipitates in both samples. The
EDS maps for the main elements in the carbide phase i.e. Cr and Mo in the
D64 sample are presented in Figure 5-45(b1) and (c1) and confirm that these
precipitates are (Cr, Mo)-rich and possibly are M23C6. The similar maps for
the D128 sample (see Figure 5-45(b2) and (c2)) indicate similar results for this
sample. These results are in accord with former results obtained with the
SEM.
Considering the EDS maps in both samples, it is clear that generally the Mo-
rich regions correspond to the Cr-rich regions; however, there are some points
in the precipitates that the Mo concentration is higher compared to other
points. A comparison of the EDS maps in theses samples shows that there are
slightly larger precipitates (~110 nm) in the D128 sample compared to (~90
nm) precipitates in the D64 sample. This fact could be related to the growth of
these precipitates with increasing exposure time. Moreover, these (Cr, Mo)-
rich precipitates are likely to be M23C6, as shown in the EBSD results of the
D64 sample (see Figure 5-43(c)).

Figure ‎5-45. Showing precipitates at the interface region in the D64 (1) and the D128 (2)
samples; (a1, a2) HAADF STEM micrographs from the interface, (b1, b2) the EDS map
for Cr indicate the growth of precipitates and (c1, c2) the EDS map for Mo.

In summary, the following transformations were observed during the ageing
of the weld overlaid tube at 650˚C:
1) In the weld overlay region:
a) Carbide particle formation and growth along grain boundaries within a
region ~100 μm wide adjacent to the interface. The width of this
region remained constant during thermal exposure.
b) Formation of a hard band ~30 μm wide adjacent to interface. The
hardness of this region may affect the tube service behaviour.
c) Precipitation and growth of TCP phases is different in different regions
of this material.
2) In the steel substrate:
a) Tempering of all regions of this material occurs at different rates
which could be related to the as-welded microstructure.
b) Decarburization of the steel occurs (formation of soft zone in the near-
HAZ region) which could affect the service behaviour.
c) Re-hardening of the near-HAZ region and fine grain formation in this
region is observed after 128 days. This unusual phenomenon may be
related to carbide transformation and carbon concentration in the
matrix.
d) Carbides precipitate at the interface and possibly cause a barrier to
carbon migration.

Chapter6 … Calculation of phase equilibria using Thermo-calc software
Chapter 6
6- Calculation of phase equilibria in Ni-

based alloys and 2.25Cr1Mo steel using
Thermo-Calc software
6-1- Introduction
Thermo-Calc is a software package for multi-component phase diagram
calculation and, when combined with a suitable critically assessed
thermodynamic database, it can be used to predict equilibrium phase
formation in complex alloys such as those investigated in the present work.
Thermo-Calc is now available with a Windows interface and this version was
employed in this work which is referred to as TCW (Thermo-Calc Windows).

In this research, TCW was used to investigate the equilibrium phases in the
wrought IN625, weld overlay material and 2.25Cr1Mo steel. In the next
section, the principles of computer calculation of phase diagrams
(CALPHAD) will be presented and the implementation of this approach in
TCW will be reviewed. In the following section, the focus is on Thermo-Calc
software, the methodology employed and the available databases. A study of
the wrought IN625 composition is presented in the next section and then,
based on this section, the weld overlay material will be compared to the
wrought IN625. In the weld overlay section, both regions of the weld overlay
material (i.e. the bulk and the mixing zone) will be assessed separately.
A further application of the TCW software was to investigate non-equilibrium
solidification of the weld overlay based on the well known Scheil
solidification model. This was undertaken to have a better understanding of
phases likely to be present in the weld overlay material in the as-deposited
condition. The different regions of the weld overlay will be analysed in this
section too.
In the final section, the equilibrium stable and metastable phases are predicted
for the 2.25Cr1Mo steel. Based on these calculations, the carbide
transformations and changes in the co-existing ferrite will be described.
Finally, the effect of carbon content on the stable carbide will be introduced in
order to provide underpinning understanding of the carbide phase changes
which occur during decarburization of the near-HAZ region.

6-2- CALPHAD principles
The usual (temperature versus composition) phase diagram is a graphical
representation of the most stable phases in a certain chemical composition at a
given temperature as well as constant pressure. With these conditions these
stable phases correspond to a minimum of the system Gibbs free energy. The
Gibbs free energy of a system is defined by the equation as follow:
where H is the enthalpy (heat content), T the absolute temperature, and S the
entropy (measure of randomness) of the system. It is suitable to choose G as
the basic thermodynamic modelling function because a system at constant
temperature and pressure has the lowest possible value of the Gibbs free
energy if it is in stable equilibrium or phases are the most stable ones.
Phase transformations occur when one or more phases in an alloy system
change into a new phase or mixture of phases. This occurs because the initial
state of the alloy is unstable relative to the final one. This phenomena is
related to diffusion of atoms from one state to another one which occurs in
order to decrease the system Gibbs free energy [175]. Therefore, calculation
of stable phases in a system or its phase diagram is achieved by achieving the
minimum Gibbs free energy. If the Gibbs energies of all competing phases in
a system are known as a function of temperature and composition, this
calculation can be done numerically.
Practically, the thermodynamic functions can be determined from experiments
only for observable phases and these are valid only for their stability regions

in terms of temperature and composition ranges [176]. Therefore, choosing a
suitable mathematical expression is important for the thermodynamic
functions. These equations must be based on realistic physical models while
the interpolation of the experimental data as well as extrapolation outside the
experimentally ranges is being allowed.
A simple power series can represent the Gibbs energy of a system. There is no
natural zero point for the Gibbs free energy and it must always be given
relative to one reference. For example, it is expressed as following popular
equation:
The S0 is set to zero and then the last term is omitted. H0 (298.15) is indicated
by HSER for enthalpy of a stable element reference. Different states of a
stoichiometric compound can be shown with unary data. As their
compositions are defined, a mathematical formula is used for presenting their
fundamental properties such as enthalpy and entropy. These type of data are
expressed as a function of temperature in the form of the following equation:
Critical analyses and computer optimisation of experimental results and Gibbs
free energy as well as information from phase equilibria studies are needed for
calculation of the coefficients of this equation. Moreover, some theoretical
methods may also be used in this regard [161].
Where phases exist over a wide range of chemical compositions (for example,
solid solution phases in metallic materials) other mathematical models are

used to show the effect of composition changes on free energy. The ideal
solution is the simplest description of a solid solution phase which means that
the constituents are non-interacting and only mixed randomly together. The
free energy of a solution phase with two elements is described as follows:
where
The first term in this equation is related to the unary data of each element and
shows the free energy change because of presence of pure A and B atoms
together. Mixing of xA and xB mole fractions of atoms A and B in the solution
causes changes in the entropy of system. The second term is associated with
the configurational entropy and its effect on the Gibbs energy of system.
In practice, most binary systems are non-ideal and interactions between the
elements within the lattice should be considered. This difference from ideal is
expressed by the excess Gibbs energy which is shown in the following
equation by ΔGxsmix:
A large number of thermodynamic models have been developed for crystalline
phases and related systems such as the substitutional type and the multiple
sublattice models [155, 161] based on this general equation for free energy.
The main differences in these models are based on the definition of the excess
Gibbs free energy term. There are different sites for atoms in the crystalline
structure which cause an increase in the entropy of system. The free energy of

the substitutional model (ΔGm) for a multi-component system can be
represented by the following equation:
where xi is the mole fraction of component i, ΔG0i is the free energy of the
pure component i based on the unary data, T is the temperature and R is the
gas constant. Ωv is an interaction coefficient dependent on the value of v. In
the regular solution model, v=0 and in the sub-regular model v=0 and 1. In
practice v does not go up above 2 [155]. The ΔGxsmix term is associated with
interaction between the different elements on the first sublattice.
In real systems, there are two and/or more crystallographic sublattices and
interaction between these sublattices must be considered. The multiple
sublattice model is significantly complex because the ΔGxsmix term is related
to the interaction between components on the sublattice and between
sublattices themselves. These interactions must be considered in calculation of
the ΔG0 and the ΔGidealmix terms in addition to the ΔGxsmix term. Details of
calculation of these terms for a phase with two sublattices and two species in
each of them are presented elsewhere [177].
Once thermodynamic properties of various phases are defined, phase
equilibria can be calculated using software packages such as Thermo-Calc or
MTDATA which perform free energy minimization routines with the aid of
thermodynamic databases. All types of models require input of coefficients
which uniquely describe the properties of the various phases and these
coefficients are held in databases.

6-3- Thermo-Calc software
6-3-1- Methodology
Calculation of equilibrium phase fractions in a system and their compositions
as a function of temperature for multi-component materials is the main
application of Thermo-Calc software. The Windows version of Thermo-Calc
software (TCW) has a graphical user interface which makes calculations of
multi-component phase diagrams easy to perform. The calculation of stable
phases where temperature is varied can easily be shown by a single axis
diagram. In this diagram, mole fraction or mass fraction of each phase is
shown versus temperature. This is termed a multiphase diagram in this work.
Moreover, it is possible to analyse the effects of changing the chemical
composition on the stable phases at a given temperature in a single axis
diagram. For example, the effect of Fe content on the stable phases in the
wrought IN625 can be shown in this type of diagram. Different types of
“multiphase” diagrams were calculated for the wrought IN625, the weld
overlay material and 2.25Cr1Mo steel.
Diagrams for the equilibrium phases can be generated by the Thermo-Calc
software using a diagram map which contains two axes. In these phase
diagrams, the effect of chemical composition and temperature on the stable
phases can be plotted. By generating the diagram for equilibrium phases in a
material versus one specified element content, the phase formation at a given
temperature could be predicted. This data can be used for analysis of
experimental observations during the high temperature exposure experiments.

For example, the diagram for the equilibrium phases in the 2.25Cr1Mo steel
with carbon content and temperature can help in evaluation of the stable
phases in the steel resulting from changing carbon concentration. In other
words, the effects of the steel decarburizing on its stable phases can be
investigated.
Another application of the Thermo-Calc software is the calculation of phases
arising from non-equilibrium solidification using the Scheil-Gulliver model.
TCW contains a Scheil module for calculating the effect of microsegregation
on the solidification process in a multi-component alloy such as the weld
overlay material. The basic assumptions in the Scheil-Gulliver model are:
1) Fast diffusion in the liquid phase (homogenous composition in this
phase)
2) No diffusion of solute elements in the solid phase [160]
In this type of calculation, the chemical composition of the liquid and the solid
phases at the start of solidification are computed from the equilibrium
calculation. Then, for the next step, the resultant liquid composition is used for
the calculation of new liquid and solid phases using thermodynamic equations
in order to minimize the free energy of system. Therefore, the effect of
enrichment of alloy elements in the remaining liquid phase is considered for
calculation of the coexisting solid phase composition as temperature is
changed.

6-3-2- Multicomponent Databases
A thermodynamic software package is dependent upon accurate and validated
databases. There are many critically-assessed and high quality databases
available which cover a wide spectrum of materials including steels and
nickel-based alloys. An important step in using Thermo-Calc is choosing the
right database. The general database is SSOL4 which covers a large number of
elements whereas the TTNI7 database is suitable only for nickel-based alloys
and the TCFE6 database is designed for steels. The data sheets of these
databases are presented in Appendix 3.
The TCFE6 database was released in March 2008 by Thermo-Calc and
includes many new critical assessments of the thermodynamic data for
multicomponent system of various steel/Fe alloys. This database covers
complete and critical assessment of binary and ternary systems as well as the
iron corner of some higher order systems, within a 21 element framework.
Table ‎6-1.TCFE6 alloying elements and their maximum wt.%
Element Max (wt.%) Element Max (wt.%)

Al 5.0 N 5.0
B Trace Nb 5.0
C 7.0 Ni 20.0
Ca Trace O Trace
Co 20.0 P Trace
Cr 30.0 S Trace
Cu 5.0 Si 5.0
Mg Trace Ti 3.0
Mn 20.0 V 15.0
Mo 10 W 15.0

The database is applicable to various types of steels/Fe alloys with a minimum
50 wt.% Fe and for the alloying elements, the recommended composition
limits in weight percent are as Table 6-1. Sensible calculations cannot be
expected if all alloying elements are at their highest limits but considering the
2.25Cr1Mo steel composition (presented in Table 6-8), this database can be
used for equilibrium calculations in this material.
The TTNI7 database was specially created by Saunders [155-157] to answer
the requirements of dealing with Ni-based superalloys including IN718 and
IN625. This database has proved to provide excellent agreement with
experimental observations in a range of nickel based alloys [178].
Furthermore, it has also been used successfully in solidification modelling of
these alloys [156]. The database contains following 22 elements:
Ni, Al, Ca, Co, Cu, Fe, Hf, Mn, Mo, Nb, Re, Ru, Si, Ta, Ti, V, W, Zr, B, C, N
and O.
Considering the chemical composition of the wrought IN625 presented in
Table 6-2 and the weld overlay material (bulk and mixing zones) presented in
Table 6-4 and Table 6-6 respectively, the TTNI7 is an appropriate database for
them. Data sheets of these databases are presented in appendix No.3.
6-4- Calculation of phase equilibria in

wrought IN625
The results of calculations for the equilibrium phases in the wrought IN625
using Thermo-Calc software are presented in this section. As it is a well

known material, the Thermo-Calc results can be compared with the
experimental results. Moreover, it gives a good baseline for the calculation of
stable phases in the weld overlay material. The composition of wrought IN625
as obtained by spark emission spectroscopy is given in Table 6-2.
Table ‎6-2.Chemical composition of the wrought IN625 material
Element wt.% at.%

C 0.005 0.025
Al 0.1 0.22
Ti 0.21 0.26
Cr 21.9 25.2
Fe 3.18 3.4
Co 0.1 0.1
Ni Balance Balance
Cu 0.17 0.16
Nb 3.5 2.3
Mo 9.2 5.7
In this work, the “Elements” module was chosen for entering the initial
conditions (temperature, pressure and chemical composition). By choosing the
TTNI7 database for the wrought IN625, the chemical composition presented
in Table 6-2 was entered in the software. Moreover, referring to the thermal
exposure tests at 650˚C, additional conditions (T=650˚C, P=1 Bar) were
entered in software too. The equilibrium molar fractions of the phases at
650˚C are calculated as follows (in mol.%):
1) FCC-A1#1(γ) : Austenite (Ni solid solution) - 81.95
2) DELTA#1 (δ) : Ni3 (Nb0.88 Ti0.08 Mo0.04) - 8.97
3) Mu_PHASE#1 (μ) : Mo (Cr0.52 Ni0.47 Fe0.01)2 - 8.95
4) Carbide (M23C6): (Cr0.82 Mo0.13 Ni0.05)23C6 - 0.13

As expected for the wrought IN625 at 650˚C, the main phase is the austenite
(γ) or FCC-A1#1 which is the matrix and other phases are precipitates.
Because of the low (0.005 wt.%) carbon content, the stable Cr-rich M23C6
carbide phase has the lowest fraction in system. Two other phases in this
material can be classified as interemetallics. The (μ) is a Mo-rich phase, while
the (δ) is a Nb-rich phase.
The multiphase (molar fraction of phases) diagram versus temperature is a
convenient presentation of the changes in the stable phases. Figure 6-1(a)
shows a multiphase diagram of phase fraction (NPM) versus temperature for
the wrought IN625 alloy in the range of 500 to 1500˚C. This figure indicates
the temperature range for the equilibrium solidification in this material.
Moreover, the temperature range for precipitation of minor phases in the γ can
also be seen in this diagram.
It is clear that above 1000˚C and under equilibrium conditions, only austenite
(γ) phase exists in this material and below 1000˚C, (MC) precipitates from the
austenite (γ) phase. As seen in Figure 6-1(b) with decreasing temperature
other phases become stable in this material. The (δ) becomes stable at
temperatures below 900˚C while the (μ) is stable at temperatures below
850˚C. These phases are formed at the expense of matrix (γ) and so it can be
shown by following reaction:
where ppt. represents different secondary phases such as (MC), (δ) and (μ).

Figure ‎6-1. Showing plots of mole-fraction of phase (NPM) versus temperature for the
equilibrium phases in wrought IN625 (Table ‎4-2); (a) indicates equilibrium solidification
and precipitation, (b) indicates temperature range for precipitation of different phases
and (c) indicates carbide transformation temperature range .

Based on former experimental observations (see section 2-4-2-2), carbide
transformation occurs in the IN625 alloy and as seen in Figure 6-1(c) above
950˚C, the (MC) phase (line No.6) converted to the M6C carbide (line No.5).
In order to determine the chemical composition of the MC and M6C phases,
two separate single equilibrium calculations were carried out at 990˚C and
950˚C respectively. The results of the calculation at 990˚C show the following
equilibrium phases are present (in mole %):
1) FCC_A1#1 (γ) : Ni solid solution - 99.99
2) FCC_A1#1 (MC): (Nb0.96 Ti0.04)C - 0.01
While at 950˚C the equilibrium phases are as follows:
1) FCC_A1#1 (γ) : Ni solid solution with 99.99
2) Carbide (M6C): (Mo0.37 Ni0.33Cr0.24 Ti0.06)6 C with 0.01
The above results clearly indicate that the Nb-rich MC carbide transforms to
the Mo-rich M6C carbide. Moreover, the diagram shows that the Mo-rich M6C
molar fraction increased with decreasing temperature and finally transforms to
Cr-rich M23C6 at temperatures around ~800˚C.
As seen in Figure 6-1(b), in the temperature range of 500-800˚C four
equilibrium phases are present in this material. However, the mole fractions of
(δ) and (μ) phases increased while the mole fraction of the (γ) phase decreases
with equilibrium cooling. This behaviour can be attributed to the growth of
these phases. At a temperature just above 500˚C the Ni2M phase is predicted
to precipitate from the (γ) phase and its molar fraction increases with cooling.

There is evidence from published research (see section 2-4-2-3) that at
temperatures below 600˚C, the Ni2M phase precipitates and grows at the
expense of the (γ) phase by the following reaction:
The effect of changes in the chemical composition on the equilibrium phases
at a specified temperature can be investigated by a diagram which shows the
mole fractions of phases versus mass percent of one element. For example, the
effect of changing carbon and iron contents on the stable phases at 650˚C was
investigated by these diagrams. In these calculations, the carbon or iron
contents are changed by treating nickel as the balancing element and keeping
all other element contents constant.
The multiphase diagram of stable phases in the wrought IN625 versus carbon
mass percent at 650˚C is presented in Figure 6-2(a). The composition of
wrought IN625 contains 0.005 wt.% C; however, at the zero point of this
diagram, the composition contains zero percent carbon. The maximum of
carbon content was chosen for relevance to the experimental work. This
diagram shows that increasing the carbon content results in more M23C6
carbide at the expense of (γ) and (μ) phases. Increasing the carbon content has
negligible effect on the (δ) fraction while the (μ) phase is reduced due to
formation of the M23C6 carbide which contains a higher fraction of the Mo
atoms. Therefore it is suggested that increasing the carbon content at 650˚C
decreases the stability of the (μ) phase in the wrought IN625.
The effect of iron level was explored because this is relevant to intermixing in
the experimental samples due to weld deposition. The maximum iron level

was chosen for relevance to the experimental work. The multiphase diagram
of mole fraction of phases versus iron mass percent at 650˚C is presented in
Figure 6-2(b). Increasing the iron content to about 25 wt.% has no effect on
the carbide type and minimal effect on its molar fraction. However, above that
level the carbide transforms to MC with a little change in its molar fraction.
Increasing the Fe content up to 8 wt.% increases the mole fraction of (μ) but at
this level, the (σ) phase starts to appear and its fraction increases at the
expense of the (μ) phase, eventually replacing the latter above 13 wt.% Fe
Figure ‎6-2. Showing plots of mole-fraction of phases (NPM) versus composition for the
equilibrium phases in wrought IN625 (Table ‎6-2)‎at‎650˚C; (a) multiphase diagram for
varying the carbon (b) multiphase diagram for varying the iron content.

The results of calculation of equilibrium phases at 650˚C with different Fe
levels are shown in Table 6-3 and indicate changes in the stable phases and the
chemical composition of (μ) and (σ) phases. The mole fraction of (σ) phase
increases with increasing Fe content; however, above 33 wt.% Fe, the Laves
phase become stable and its mole fraction increases at the expense of (σ)
phase.
Table ‎6-3. Stable phases in the wrought IN625 with different Fe content at‎650˚C.
IN625 with 10 wt.% Fe IN625 with 15 wt.% Fe

Phase
Composition mol.% Composition mol.%
γ - 77.13 - 70.13
δ Ni3 (Nb0.88 Ti0.08 Mo0.04) 9.48 Ni3 (Nb0.88 Ti0.09 Mo0.03) 9.58
μ Mo (Cr0.52Ni0.43Fe0.05)2 9.26 - 0.00
σ Cr0.45 Ni0.25 Mo0.19 Fe0.11 4.01 Cr0.42 Ni0.22 Mo0.21 Fe0.15 20.17
M23C6 (Cr0.82 Mo0.13 Ni0.05)23C6 0.12 (Cr0.82 Mo0.13 Ni0.05)23C6 0.12
1.
The effects of chemical composition and temperature on the stable phases can
be investigated simultaneously with phase diagrams. Figure 6-3(a) shows the
phase diagram of the wrought IN625 with carbon while an enlarged diagram is
presented in Figure 6-3(b). The stability regions of stable phase are shown by
different lines. For example, the line No.2 shows the stability region of the
MC phase and its curvature indicates solid solution of carbon in the IN625 at
temperatures above 950˚C. Line No.4 shows the stability region of M6C phase
and indicates that increasing the carbon content results in stability of this type
of carbide at higher temperature. Finally, line No.3 shows the stability region
of M23C6 carbide and indicates that at carbon content more than 0.06 wt.%, it
becomes stable at higher temperatures.

Figure ‎6-3. Showing different phase diagrams for the wrought IN625 material (Table
‎6-2); (a) phase diagram of IN625 and carbon between‎ 400˚C‎ and‎ 1200˚C,‎ (b)‎ phase‎
diagram‎ of‎ IN625‎ and‎ carbon‎ between‎ 600˚C‎ and‎ 1000˚C,‎ (c)‎ phase‎ diagram‎ of‎ IN625‎
and‎iron‎between‎600˚C‎and‎1000˚C.‎

As shown in Figure 6-3(a) and (b) below 800˚C, the stable carbide is M23C6
and there is little solid solution of carbon in the IN625 alloy. The stability
regions of three different intermetallic phases in this material are shown by the
straight lines because carbon content does not have any significant effect on
these minor phases. The Ni2M phase (line No.1) is stable at temperatures
below 500˚C; however, line No.5 shows the stability region of (δ) phase
whereas line No.6 is related to the (μ) phase.
The phase diagram for the IN625 alloy with iron is presented in Figure 6-3(c)
and shows that above 850˚C, the (μ) phase is not stable while the (σ) phase is
stable. However, at 650˚C and iron content less than 8 wt.%, the (μ) phase is
stable. Between 8 and 13 wt.% iron, the (σ) phase becomes stable in addition
to the (μ) phase and at higher Fe content (>13 wt.% ), the (μ) phase is not
stable.
6-5- Calculation of phase equilibria in the

weld overlay material
The weld overlay is a nickel based similar on composition to alloy IN625 but
because of inter-mixing with the steel substrate during deposition, its chemical
composition is significantly different from that of the wrought IN625. Based
on the experimental observations presented in chapter 4, there are two
different regions in this material with different composition which have been
termed the bulk and the mixing zones. In this section, the effect of chemical
composition and temperature on the stable phases under equilibrium will be
described for these regions separately.

6-5-1- Stable phases in the Bulk zone
The chemical composition of the bulk zone in the weld overlay material as
obtained by spark emission spectroscopy is given in Table 6-4. The results of
calculations for the equilibrium phases using Thermo-Calc are presented in
this section. As the composition contains more Fe and C compared to those of
the wrought IN625, the Thermo-Calc results for this material can be compared
with the phases of the wrought IN625. The composition of stable phases and
their mole fractions in the weld overlay and the wrought IN625 at 650˚C are
presented in Table 6-5.
Table ‎6-4. Chemical composition of the weld overlay material in the bulk zone.
Element wt.% at.%

C 0.02 0.1
Ti 0.18 0.22
Cr 20.9 23.9
Fe 8.0 8.5
Ni Balance Balance
Nb 3.2 2.0
Mo 8.2 5.1
Table ‎6-5. Stable‎phases‎in‎the‎bulk‎zone‎of‎the‎weld‎overlay‎and‎the‎IN625‎at‎650˚C
IN625 Weld overlay

Phase
γ - 81.95 - 83.36
δ Ni3 (Nb0.88 Ti0.08 Mo0.04) 8.97 Ni3 (Nb0.87 Ti0.07 Mo0.06) 8.12
μ Mo (Cr0.52Ni0.47 Fe0.01)2 8.95 Mo (Cr0.49 Ni0.48 Fe0.03)2 8.04
M23C6 (Cr0.82 Mo0.13 Ni0.05)23C6 0.13 (Cr0.82 Mo0.13 Ni0.05)23C6 0.48
1.

A comparison between the phases in the weld overlay and IN625 at 650˚C
shows that the chemical composition of the intermetallic phases and their
molar fractions are almost similar. For example, the (μ) phase in weld overlay
contains more Fe atoms. The carbide molar fraction in the weld overlay is
almost four times that predicted in the wrought material; this ration is
approximately the same as the ratio of their carbon levels.
Figure 6-4(a) shows the multiphase diagram of (NPM) versus temperature for
the bulk zone in the range of 500 to 1500˚C. This figure indicates that the
temperature range for equilibrium solidification is approximately the same as
that of the wrought IN625 (see Figure 6-1(a)). However, the precipitation of
MC carbide occurs at temperatures above 1100˚C. The carbide transformation
process in this material is similar to that of the IN625 but the temperature
ranges of their stability are different as indicated in the following equations:
Figure 6-4(b) shows that higher carbon and iron contents in the weld overlay
material result in some changes in the stability of different phases compared to
those of the wrought IN625 (see Figure 6-1(b)). For example, the MC
becomes stable at higher temperature while M6C carbide transformation
temperature is lower compared to that of the wrought IN625 alloy.
Nevertheless, there is no significant difference between the precipitation
behaviour of the intermetallic phases i.e. (δ) and (μ) in these materials.

equilibrium phases in the bulk zone of the weld overlay material (Table ‎6-4); (a)
indicates equilibrium solidification and precipitation, (b) indicates temperature range
for carbide transformation and precipitation of intermetallic phases, (c) indicates
carbide‎ transformation‎ temperature‎ at‎ (~760˚C)‎ and‎ (σ)‎ phase‎ transformation‎ to‎ (μ)‎
phase‎at‎(~820˚C).

Moreover, the precipitation of the Ni2M phase is not seen in this range of
temperature. However, two different Cr-rich phases i.e. (σ) and α (Cr) phases
precipitate at temperatures ~830˚C and ~500˚C respectively. Figure 6-4(c)
shows the stability temperature range of the (σ) phase and as seen, it
transforms to the (μ) phase at ~820˚C.
6-5-1-1- Effect of carbon on phase equilibria
As seen in the experimental observations, high temperature exposure of the
overlaid tube at 650˚C results in microstructural changes due to carbon
diffusion. Therefore the effect of carbon content on the stable phases was
examined. To show this effect, a multiphase diagram involving changes in
carbon content at 650C along with a temperature – wt.% C phase diagram
were calculated. The multiphase diagram of stable phases in the bulk zone
versus carbon mass percent at 650˚C is presented in Figure 6-5(a). At the zero
point of this diagram, the composition is the same as Table 6-4 but with zero
percent carbon. This diagram shows that increasing the carbon content results
in an increase of carbide mole fraction at the expense of the (γ) and the (μ)
phases. Calculation of equilibrium phases at 650˚C with 0.12 wt.% carbon
indicates that the carbide mol.% increases to 2.85 while (μ) phase mol.%
decreases to 6.98 compared to values in Table 6-5.
Figure 6-5(b) shows the phase diagram of this material with varying carbon
level. Although the phases are similar to those of the wrought IN625 (see
Figure 6-3(b)), their stability regions are different. In this diagram there is a
stability region for the (σ) phase which is shown with line No.6 and increasing
the carbon content results in a reduction in the stability of this phase.

Figure ‎6-5. Showing different diagrams for the bulk zone of the weld overlay material
(Table ‎6-4); (a) the multiphase diagram versus‎ carbon‎ content‎ at‎ 650˚C‎ indicates‎
increasing‎the‎carbide‎mole‎fraction‎at‎the‎expense‎of‎the‎(γ)‎and‎the‎(μ)‎phases,‎ (b) the
phase‎ diagram‎with‎carbon‎between‎600˚C‎and‎1000˚C‎indicates‎stability‎region‎of‎the‎
Cr-rich‎(σ)‎phase.
6-5-2- Stable phases in the mixing zone
The Fe content in the mixing zone is higher compared to that of the bulk zone.
The definition of chemical composition in the mixing zone is complicated
because of a three dimensional interface between the steel and this zone. The
EDX results for elements (excluding C) in this region from several analyses at

a distance of ~10 μm from the interface were collected and their mean values
are presented in Table 6-6.
Table ‎6-6. Mean chemical composition of the weld overlay material in the mixing zone as
determine by EDX (excluding carbon)
Element wt.% at.%

C 0.02<x<0.1 0.1<x<0.5
Cr 17.4 19.7
Fe 24.8 26.1
Ni Balance Balance
Nb 3.1 2.0
Mo 7.0 4.3
Although the Fe level in this zone is greater than in standard nickel alloys, it is
believed that TTNI7 database is suitable for high Fe content nickel alloys
[178]. To allow simulation of interdiffusion of carbon into the mixing zone
during welding, a range of carbon levels is explored in the calculations. The
carbon content in this region is investigated in a range between the carbon
content of bulk zone (0.02wt.%) and steel (0.1wt%). The Thermo-Calc
calculation results for phase equilibria with different carbon content at 650˚C
are presented in Table 6-7.
Table ‎6-7. Stable phases‎in‎the‎mixing‎zone‎of‎the‎weld‎overlay‎at‎650˚C‎with‎different‎

carbon content.
C=0.02 wt.% C= 0.1wt.%

Phase
γ - 81.6 - 82.4
δ Ni3 (Nb0.97Mo0.03) 7.4 Ni3 (Nb0.97Mo0.03) 7.3
σ Cr0.37 Ni0.21 Mo0.24 Fe0.18 10.3 Cr0.37 Ni0.21 Mo0.24 Fe0.18 7.1
M6 C (Mo0.48Cr0.17Ni0.17Fe0.16Nb0.02)6C 0.7 (Mo0.48Cr0.17Ni0.17Fe0.16Nb0.02)6C 2.7
M23C6 - 0.00
1. (Cr0.8Mo0.13Ni0.04Fe0.03)23C6 0.5
1.

equilibrium phases in the mixing zone of the weld overlay material (Table ‎6-6); (a)
indicates equilibrium phases for 0.02 wt.% C, (b) indicates temperature range for
carbide transformation for 0.02 wt.% C and (c) indicates stable carbides and
transformation temperature for 0.1 wt.% C.

A comparison of the mole fraction of phases in Table 6-7 indicates that
increasing the carbon content in the mixing zone results in a higher carbide
content, with two different types forming at expense of the (σ) phase. The Cr-
rich M23C6 carbide becomes stable in addition to the Mo-rich M6C carbide.
The multiphase diagrams of NPM versus temperature for the mixing zone
with different carbon contents are presented in Figure 6-6. The multiphase
diagram with 0.02 wt.% carbon is shown in Figure 6-6(a). The equilibrium
solidification behaviour in this material is different compared to that of the
bulk zone (see Figure 6-4(a)). As seen in this diagram, at the final stages of
solidification, the MC phase becomes stable. It is suggested that equilibrium
solidification terminates with following eutectic reaction:
Considering 0.02 wt.% carbon (see Figure 6-6(b)), the M6C carbide is the
stable carbide at the temperature range of 620 to 690˚C while with 0.1 wt.%
carbon (see Figure 6-6 (c)), MC and M6C are stable above 660˚C and the M6C
and M23C6 carbides are stable below this temperature. The effect of carbon
content on the stable phases in the mixing zone can be easily shown using
multiphase diagram versus carbon content along with the phase diagram. The
multiphase diagram for stable phases versus carbon content at 650˚C in the
mixing zone is presented in Figure 6-7(a). An enlarged diagram presented in
Figure 6-7(b) indicates that with increasing carbon content, the mole fraction
of M6C carbide increases at the expense of the (σ) phase. Moreover, above
0.06 wt.% C, the M23C6 carbide becomes stable.

Figure ‎6-7. Showing different diagrams for the mixing zone of the weld overlay material
(Table ‎6-6); (a) the multiphase diagram NPM‎ versus‎ carbon‎ content‎ at‎ 650˚C,‎ (b)‎ a‎
enlarged‎ diagram‎ indicates‎ increasing‎ the‎ carbide‎ mole‎ fraction‎ at‎ the‎ expense‎ of‎ (σ)‎
phase and (c) phase diagram with carbon‎ mass‎ percent‎ between‎ 600‎ and‎ 1000˚C‎
indicates the stability regions of different carbides.

Figure 6-7(c) shows the phase diagram for the mixing zone with carbon
content. Although the carbide types in this diagram are similar to those of the
bulk zone, their stability regions are different (see Figure 6-5(b)). Moreover,
the stability region for the (σ) phase (line No.5) is significantly larger than that
in the bulk zone.
6-6- Non-equilibrium solidification of weld

overlay material
Rapid cooling after welding results in non-equilibrium solidification of weld
overlay material which affects phase formation at the final stages of
solidification. The Thermo-Calc Scheil module enables a diagram to be
plotted which shows mole fraction of all solid phases versus temperature.
6-6-1- Bulk zone
The non-equilibrium and equilibrium solidification behaviours for the
composition of the bulk zone (Table 6-4) are shown in Figure 6-8(a) and (b)
respectively. In non-equilibrium condition, the temperature range for
solidification is between 1370˚C and 1140˚C. In contrast, in the equilibrium
condition (Figure 6-8(b)) the temperature range of solidification is between
1370˚C and 1270˚C. Moreover, there are more secondary phases predicted to
form during non-equilibrium solidification than under equilibrium freezing.
The Scheil model shows that solidification starts at 1370˚C with following
reaction:
Along the Line No.2

This reaction causes enrichment of alloying elements in the remaining liquid
and at 1207˚C, the MC carbide starts to form with a eutectic-like reaction (as
follows) and therefore the solid phases are γ and MC along the Line No.3
Along the Line No.3
Point Temperature(˚C) Liquid mol.% Solid phases

A 1370 100 -
B 1207 12 γ
C 1148 7 γ + MC
D 1147 6 γ + MC+ δ
1137 0 γ + MC + δ +σ
Figure ‎6-8. Showing different diagrams for the bulk zone of the weld overlay material
(Table ‎6-4); (a) diagram indicates formation of different phases during final stages of the
solidification based on the Scheil model, (b) equilibrium solidification diagram.

Further non-equilibrium cooling results in Nb enrichment in the remaining
liquid and causes formation of the Nb-rich (δ) at ~1148˚C with the following
reaction:
Along the Line No.4
Finally, at ~1147˚C, the composition of the remaining liquid is capable for
formation of the Cr-rich (σ) phase which is shown with point D. Therefore the
final reaction is:
Along the Line No.5
Therefore, in the final state, different phases like (γ), (MC), (δ) and (σ) are
present in the as-solidified microstructure under non-equilibrium condition (no
diffusion in the solid state) while under equilibrium solidification, only (γ)
phase is stable. There is an option in the Scheil calculation for the elements to
be considered as fast diffusing elements in the solid state. Carbon is the only
element that can be considered as fast diffuser. However, calculation in this
conditon did not change the predicted phases in the Scheil model.
6-6-2- Mixing zone
The Scheil model calculation results for the mixing zone (Table 6-6) with low
carbon content (0.02 wt%) are shown in Figure 6-9(a). As seen in this
diagram, the temperature range for non–equilibrium solidification in this
material is similar, but secondary phases are different compared to those of the
bulk zone. In this composition, MC, Laves and (δ) were firstly formed and the
solidification terminates with formation of the (σ) and the (μ) phases.

A 1370 100 -
B 123 15 γ
C 1155 9 γ + MC
D 1147 5 γ + MC + Laves
E 1135 1 γ + MC + Laves + δ
F 1130 0 γ + M + Laves + δ +σ +μ
Figure ‎6-9.Showing different diagrams for the mixing zone of the weld overlay material
(Table ‎6-6) with 0.02 wt.% C; (a) diagram indicates formation of different phases during
final stages of the solidification based on the Scheil model, (b) equilibrium solidification
diagram.
The equilibrium solidification diagram presented in Figure 6-9(b) shows that
MC carbide is predicted to form at the final stages of solidification by a
eutectic like reaction. As seen in this diagram, at the final stages of

solidification, the mole fraction of liquid suddenly decreases to zero while (γ)
and (MC) carbide phases increase. Therefore, it is suggested that under
equilibrium conditions, the solidification terminates with following eutectic
reaction:
Considering carbon as a fast diffuser element has no effect on the results of
the Scheil model.

A 1370 100
B 1290 28 γ
C 1185 7 γ + MC
D 1150 5 γ + MC + σ
E 1135 3 γ + MC + σ + Laves
F 130 0 γ + MC +σ + Laves + δ +μ
Figure ‎6-10. Showing a diagram indicates formation of different phases during final
stages of the solidification based on the Scheil model for the mixing zone of the weld
overlay material (Table ‎6-6) with 0.1wt.% carbon .

Although the equilibrium solidification behaviour of the mixing zone is the
same for different carbon contents, the Scheil model results are different. The
Scheil model results for the mixing zone (Table 6-6) with a high (0.1 wt.%)
carbon content are shown in Figure 6-10. It is seen that secondary phases
during solidification under non-equilibrium condition are different compared
to those of the low carbon material. Increasing the carbon content causes
formation of (σ) phase after (MC) phase. At point D (point No.5), formation
of σ phase stopped and Laves phase formed from the remaining liquid which
is shown by line No.6.
6-7- Calculation of stable and metastable phase

equilibria in 2.25Cr1Mo steel
Analysis of the stable phases in the 2.25Cr1Mo steel (the substrate material of
the weld overlaid tube) is presented in this section. The chemical composition
of this steel is shown in Table 6-8. The thermodynamic calculations were
carried out using the TFEC6 database.
Table ‎6-8.Chemical composition of 2.25Cr1Mo steel substrate as determined by spark

emission spectroscopy
Element wt.% at.%

C 0.10 0.46
Si 0.24 0.48
Cr 2.33 2.5
Mn 0.43 0.43
Fe Balance Balance
Cu 0.11 0.1
Ni 0.1 0.1
Mo 0.92 0.53

6-7-1- Effect of temperature on the stable
phases
Thermo-Calc calculations for stable phases at 650˚C results in the following
phases being predicted to form (in mol. %):
1) BCC_A2#1 (α) : Ferrite - 97.77
2) M23C6#1 (M23C6 Carbide) : (Cr0.50Fe0.39Mo0.11)23C6 - 2.23
Referring to section 2-3-2, the stable carbide in this steel is dependent upon
the temperature. Therefore, a multiphase diagram of phase fraction versus
temperature is an appropriate method to display carbide type. The multiphase
diagrams of phase mole fraction (NPM) versus temperature with different
temperature ranges are presented in Figure 6-11. The equilibrium
solidification in low carbon steels terminates with the following peritectic
reaction:
Figure 6-11(a) shows this reaction in the temperature range between 1450 to
1470˚C. Moreover, the eutectoid transformation (as follows) at temperature
below 800˚C is indicated in this diagram.
Based on the experimental results, it is clear that this diagram shows the effect
of heating up during welding. In other words, the extent of HAZ can be related
to the stability region of the (γ) phase which is between ~ 850˚C and ~1450˚C.

equilibrium phases in the 2.25Cr1Mo steel (Table ‎6-8); (a) indicates equilibrium
peritectic solidification transformation and eutectoid carbide formation, (b) indicates
temperature range for stability of different carbides in the microstructure of steel.
6-7-2- Metastable phase equilibria
It is well known that the microstructure of the 2.25Cr1Mo steel in industrial
applications and after normalization is bainitic (see section 2-3-1) which is a
metastable structure. Moreover, as shown in section 2-3-2, the carbide type in
this steel changes during the ageing process and ageing at temperatures above
630˚C results in precipitation of the Cr-rich M23C6 carbide.

Based on the experimental observations of the as-welded sample (see section
4-3-4), the lath-like microstructure of the HAZ contains metastable carbide.
Therefore, it suggested that the initial carbide in this state is cementite. In
order to establish which metastable carbides could potentially form in the
microstructure, the stable carbide should be suspended from the calculation at
a specified temperature and this procedure repeated for all available carbides
in the database.
Thermo-calc software results show that in the 2.25Cr1Mo steel at 650˚C;
when Cr-rich M23C6 is not stable, the M7C3 and M6C are stable and instability
of M7C3 leads to stability of Cr-rich M2C. Moreover, instability of Cr-rich
M2C carbide results in the stability of (Cr, Fe)-rich M3C [CEMENTITE#1]
and (Fe, Mo)-rich M3C [KSI CARBIDE]. Finally, when KSI is not stable, it
transforms to CEMENTITE#1. Nevertheless, instability of M6C causes
stability of Mo-rich M2C; and instability of this carbide have a similar result.
Considering the chemical composition of these carbides from Thermo-Calc
calculations, the carbide transformation process can be related to the
enrichment of alloying elements (Cr and Mo) in the carbide phase. The
carbide transformation process can be summarized in the following reactions
where alpha represents coexisting B.C.C. iron.

It is known that at 650˚C, the Cr-rich M3C is not stable and above reactions
indicate how it transforms to different types of the Cr-rich and the Mo-rich
metastable carbides. It is clear that the Cr-rich carbides have different
metastable states compared to those of the Mo-rich carbides. Moreover,
changing the carbide type from the (Cr, Fe)-rich M3C to the equilibrium
M23C6 affects the coexisting ferrite composition as well as the molar fraction
of carbide which are shown in Table 6-9.
Table ‎6-9.Ferrite compositional changes with carbide transformation
Ferrite Cr(wt.%) Mo(wt.%) Carbide type Carbide

type mol.%
α1 1.68 0.91 (Cr-rich)M3C 1.83
α2 1.85 0.66 (Cr-rich)M3C+(Mo-rich)M3C 1.83
α3 1.81 0.38 (Cr-rich)M2C 1.37
α4 1.73 0.66 M7C3+(Mo-rich)M2C 1.52
α5 1.72 0.54 M7C3+M6C 1.72
α6 1.55 0.57 M23C6 2.23
6-7-3- Effect of carbon content on the stable
phase equilibria
The experimental results (Chapter 5) show that the near-HAZ region
decarburized during high temperature exposure at 650˚C. Thermo-Calc was
used to explore the effect of changing carbon content on the stable phases.

Figure ‎6-12. Showing different diagrams for the 2.25Cr1Mo steel (Table ‎6-8); (a) the
multiphase diagram versus‎carbon‎content‎at‎650˚C,‎(b) indicates changing carbide type
and increasing the M 23C6 carbide molar fraction with carbon content at‎650˚C‎and‎(c)‎
phase diagram‎with‎carbon‎between‎600˚C‎and‎950˚C‎indicates‎the‎ stability‎regions‎of‎
different carbides types.

The multiphase diagram of NPM versus carbon content at 650 C is shown in
Figure 6-12(a) and (b) which indicate that at low carbon levels, the stable
carbide is M6C which changes to the M23C6 with increasing carbon content.
The phase diagram of 2.25Cr1Mo steel with carbon is shown in Figure
6-12(c). It is seen that decarburizing of steel in the near-HAZ region results in
a changing in the carbide type and stability of M6C carbide at 650˚C. It is clear
that stability of M6C carbide affects the composition of coexisting ferrite.
A calculation was carried out with a similar composition but very low (0.005
wt.%) carbon content to analyse the ferrite composition. The results are as
follows in mol.%:
1) Ferrite : (α) [ Mo: 0.79 wt.%, Cr: 2.32 wt.%]- 99.84
2) M6C Carbide : (Mo0.55Fe0.37Cr0.06 Si0.02)6C - 0.16
In summary, Thermo-calc software is a useful tool for understanding of
equilibrium phases in different materials. Thermodynamic calculations can be
used to indicate the stability regions of interesting phases and the effects of
chemical composition and temperature on these regions. Moreover, the Scheil
module show solidification behaviour under non-equilibrium conditions and
phase formation during final stages of this process.

Discussion and conclusions
Chapter 7
7- Discussion and conclusions
7-1- Introduction
In this chapter, the main aspects concerning the microstructural evolution
following high temperature exposure in the weld overlay and steel substrate
are discussed. The microstructural evolution in different regions (presented in
chapter 5) is discussed with reference to the as-welded condition. The size,
shape and type of precipitates in both materials are discussed in detail.
Emphasis is given to the carbon migration across the interface and its effect on
the phase transformation in both materials. The kinetics and thermodynamics
relating to nucleation, growth and phase transformation in both the weld
overlay and the steel are considered. The precipitate transformations in
different regions of the steel and the weld overlay and their effects on the
matrix microstructure are discussed.

This chapter is divided into four main sections; in the first section
microstructural changes in the steel region are discussed. In this section, a
model for carbon supply which considers formation of a carbide network at
the interface will be presented. The following section is related to the
microstructural evolution in the weld overlay and finally a summary will be
presented.
7-2- Microstructural evolution in the steel

substrate
As shown in section 2-3-2, the microstructure of 2.25Cr1Mo steel in the
normalized and tempered condition consists of almost equal portions of
bainite/tempered bainite and proeutectoid ferrite grains [66, 179]. During
ageing or service at high temperature, the morphology, distribution and type
of carbides in this steel will change depending on the time and temperature.
There are different arguments regarding equilibrium carbide and
transformation sequences during ageing. In this work, the equilibrium carbide
is considered to be M23C6 for both bainitic regions (as presented by Race [58])
and for ferrite grains (as presented by Yang [43]).
In this system, two essential issues should be considered for understanding the
microstructural evolution in the steel region following the high temperature
exposure.
1) Welding of the nickel alloy material on the steel substrate causes
microstructural changes in the HAZ region. Therefore, based on the
work of Thomson and Bhadeshia [60, 61], carbide transformation rates

in the HAZ following high temperature exposure is different to those
of the central zone.
2) The steel has a fusion bond to a nickel alloy with higher Cr and Mo
contents and carbon migration occurs across the interface during the
high temperature exposure experiments. The carbon migration may
cause further modifications to the carbide transformation process.
In this work, carbide transformation and carbon migration occurs
simultaneously. In following section, the carbide transformations in the central
zone and HAZ region are presented and effect of carbon migration is
discussed later with a carbon supply model.
7-2-1- Carbide transformation in the central
zone
The observations of the central zone reveal that there are some blocky
precipitates present in the TM/B areas and fine precipitates within the ferrite
grains. Although the hardness and etched micrographs of samples with
different ageing times are the same in this zone, there are significant
differences between precipitates in the BSE micrographs. The BSE
micrograph did not reveal any precipitates in the D1 sample. However, in the
D8 sample, some coarse blocky particles along grain boundaries with brighter
contrast were observed (see Figure 5-27). Increasing exposure time to 32 days
resulted in precipitation of some dark precipitates in addition to these bright
precipitates along grain boundaries (see Figure 5-30).

Referring to Figure 5-38, it is clear that dark and bright precipitates have
grown with increasing exposure time to 64 days. However, a comparison of
the BSE micrographs of the D64 and the D128 samples indicates that the
population of dark and bright precipitates have decreased with increasing
exposure time. Moreover, a comparison of the SE micrographs in the D8
(Figure 5-36) and the D128 samples (see Figure 5-37) indicates transformation
of a large number of needle shape precipitates to a few precipitates with more
rounded shapes within the ferrite grains.
Thermodynamic calculation for stable phases were used to predict carbide
transformation sequence in a similar manner to the work of Saroja [153].
Successively less stable carbides and their mole fraction (mol.%) were
determined using Thermo-Calc and TCFE6 database. The calculation results
for this steel in section 6-7 shows that the equilibrium carbide is
with 2.22 mol.% at 650˚C, and Figure 7-1(a) shows
the less stable carbides which are predicted with their mole fraction with the
least stable carbide being Cr-rich M3C.
These facts should be considered with the following important observations
concerning the carbide transformation process:
1) The M2C precipitates have an acicular shape and are very small in size
[53, 180].
2) It is reported that thermal ageing at 580 to 630˚C for durations less
than 10,000 hours could not lead the transformation of metastable
carbides to the equilibrium carbide [56]. Moreover, based on TTT

diagram of this steel presented by Robson and Bhadeshia [169, 181]
the M23C6 carbide is not seen before 1000 hours at 650˚C.
3) There are different arguments about stable carbides in the mid-term
aged samples (up to 500 hours at temperatures between 630˚C to
650˚C). Some argued that the stable carbide is Mo-rich M6C [44, 53,
130, 180]. However, others [43, 60, 170, 181] argued that Cr-rich
M7C3 and M23C6 carbides are the dominant carbides. Moreover,
metastable carbides formation affects the composition of the coexisting
ferrite [153, 170].
4) Some researchers believe that the Mo content in precipitates increases
with exposure time at 540˚C [44] while there are other reports that
stated that the Mo content in precipitates first increased with ageing
time and then decrease with long term ageing [46, 153] which caused
Mo replenishment in the coexisting ferrite [48].
Based on calculation in appendix No.2, the Mo-rich carbides (M2C and M6C)
have a higher mean atomic number than the matrix and are thus brighter in
BSE micrographs, while the Cr-rich carbides (M7C3 and M23C6) are darker
than matrix in BSE imaging. However, the mean atomic number of the alloyed
M23C6 carbide is close to that of matrix and therefore this carbide is not visible
in BSE micrographs. Moreover, due to presence of Mo atoms in (M3C)
carbides, these are not visible in BSE micrographs.
Bright precipitates with a blocky shape in the D8 sample are attributed to the
Mo-rich M6C carbide; fine acicular precipitates within ferrite grains are
suggested to be M2C carbide. Increasing the exposure time promotes

formation of Cr-rich M7C3 and Mo-rich M6C carbides. These carbides have
grown with increasing exposure time up to 64 days. However, in the D128
sample Cr-rich M7C3 carbides and Mo-rich M6C dissolved in the matrix and
the equilibrium alloyed M23C6 is formed at their expense.
Figure ‎7-1. Showing schematic diagrams for the carbide transformation in the central
zone; (a) indicates carbide transformation prediction using Thermo-Calc, (b) indicates
carbide transformation in TM/B areas with exposure time based on experimental
observations of the central zone.

The transformation of carbides within the ferrite grains is different and M2C
carbides transform directly to the M23C6 with increasing exposure time. A new
model considering both Cr-rich and Mo-rich carbides as metastable and M23C6
as the equilibrium carbide for TM/B areas is presented in Figure 7-1(b) and it
is clear that these sequences are not seen in the ferrite grains.
7-2-2- Carbide transformation in the HAZ

region
It is known that welding processes alter the microstructure in the heat affected
zone (HAZ) and generate a complex microstructure which results in a
significant variation in ageing behaviour [62]. Therefore, carbide
transformation in this region has different sequences compared to that of the
central zone. The microstructure of the HAZ is dependent on the heat input,
the welding procedure and the post weld heat treatment. The state of the HAZ
affected by welding process is shown schematically in Figure 7-2.
Bhadeshia reported that normalization (cooling form the austenite) in the bulk
material results in formation of bainite in this steel [42]. Moreover, Race [58]
and Bergquiest [64] and Laha [182] indicated that the weld metal of
2.25Cr1Mo has a fully bainitic microstructure. Nevertheless, it is reported that
metallographic analysis of the HAZ adjacent to the fusion boundary in the
dissimilar weld joints [63, 65] revealed a bainitic structure consisting of laths
of ferrite with aligned carbide precipitates between the laths with no pro-
eutectoid ferrite. There is a report which mentioned that the Mo-rich carbides
(M2C and M6C) are not seen in the HAZ after long term service [66].

The experimental observations of the as-welded sample confirmed that bainite
was formed in the HAZ region (Figure 4-7). The hardness profile can be
related to the carbon content in the new bainite which was formed in this
region after welding and is dependent upon austenitisation temperature.
Figure ‎7-2. Showing schematic diagrams for the HAZ microstructure affected by
welding; (a) indicates carbon content in the new bainite based on the hardness survey,
(b) different‎γ‎grains‎in‎the‎near-HAZ and the far-HAZ regions during welding process
(i.e. at elevated temperatures).
Figure 7-2(a) shows the carbon content in bainite based on hardness values
and the width of the near-HAZ region (Figure 5-4(b)). Referring to results on
TIG welding of 2.25Cr1Mo steel [62, 65], the fully bainitic region hardness is

about ~450 kgf mm-2 and the width of this region is about 300 μm in both
studies. However, the width of non-uniform bainite is reported to be from
about 300 μm [62] to zero[65].
Figure 7-2(b) is based on former research introducing prior coarse grain
austenite region adjacent to interface and prior fine grain austenite region
away from the weld interface [63, 149]. Moreover, it is reported that number
of carbides in the fine grain zone is higher than coarse grain zone and tend to
be coarser [63]. This diagram indicates differences in the HAZ region in the
as-welded sample. Therefore, it is suggested that a metastable structure of
bainite with different levels of carbon will vary in their decomposition.
Moreover, the presence of carbides in the far-HAZ region results in different
sequences in the carbide transformation.
The hardness profiles of the D1 and the D8 samples are comparable with other
research on 2.25Cr1Mo steel with an Inconel 182 nickel alloy filler. Nicholson
[140] reported the hardness after ageing at 630˚C decreased to a range of 220
to 230 kgf mm-2 within 200 h. Moreover, Bhaduri [143] mentioned that the
hardness values decreased to 200 ~ 210 kgf mm-2after 10 h ageing at 630˚C.
In the HAZ region of the D1 sample, the SE micrographs show a large number
of fine precipitates and very small ferrite grains (less than 2 μm). However,
BSE micrographs indicate that a few bright precipitates are present only in the
far-HAZ region. SE micrographs of the D8 sample indicate growth of ferrite
grains (about 10 μm) in the near-HAZ region and coarsening of precipitates in
the far-HAZ region. Moreover, BSE micrographs of the D8 sample show a
few acicular bright and fine dark precipitates in the near-HAZ region. The

number of these precipitates in the far-HAZ region is significantly higher. It
should be noted that some regions adjacent to the interface decarburized in the
D8 sample which will be discussed later.
The near-HAZ region decarburized very quickly between 8 days and 12 days
of exposure but changes in the far-HAZ region were not noticeable. The
hardness profile and microscopic observation of the D16 and the D32 samples
indicate that the decarburization width is equal to the near-HAZ region and
did not change with exposure time. The optical microscope and SE
micrographs of these samples in the far-HAZ region are similar to the D8
sample. However, BSE micrographs of the near-HAZ region in the D16 and
the D32 samples indicates that bright precipitates have grown and the dark
precipitates have become less with increasing exposure time.
Figure ‎7-3. Showing stable carbides in different regions of the steel with increasing
exposure time indicates different sequences in carbide transformation.
Finally, hardness surveys indicated that the near-HAZ region re-hardened in
the D64 sample and little changes occurs in the hardness profile with

increasing exposure time. Moreover, many fine dark and coarse bright
precipitates were observed in the BSE micrographs of this region in the D64
sample which become fewer in the D128 sample. These evidence confirmed
that carbide transformation sequences in the HAZ region are similar to those
of the central zone.
However, there are different rates in these sequences which are shown in
Figure 7-3. This diagram shows stable carbides for the different regions and
their transformation with exposure time based on observations and is in accord
agreed with the following former research:
1) Parameswaran [46] found that bainite dissolution rate is related to its
carbon content and higher carbon content shows more resistance to
dissolution.
2) There is a report that carbide transformation occurs at a slower rate in
a mixed microstructure of bainite and allotrimorphic ferrite compared
to that of a fully bainitic microstructure [60, 61].
3) The type of carbides in equilibrium is agreed with the Wada report
[168]. However, carbon content [46, 180] is an important factor in the
carbide transformation. For example, Thomson found that in low
carbon content, M7C3 carbide is stable at the expense of M3C [60].
Moreover, Natesan [180] reported that in low carbon content the Mo-
rich M6C carbide is stable.

It is known that these different carbides are affected by carbon migration
process and in the next section a new model is developed to explain
experimental results for decarburization of steel region.
7-2-3- Carbon migration
It is known that carbon migration occurs across the weld interface in the
dissimilar weld joint from the low alloy steel to the high Cr material [67, 128,
139, 143, 183, 184]. The temperature range of 500 to 700˚C results in
depletion of carbon in the area adjacent to the interface in the low alloy steel
which is called soft zone or carbon depleted zone (CDZ) [129, 131, 139]. The
soft zone is characterized by lower hardness and its width is varied depending
on the composition of materials as well as temperature and time of exposure.
Many researchers argued that growth of decarburized zone width (ηα) between
austenitic and ferritic steels joints [67, 128, 134, 139] follow parabolic kinetic
with time (t) according to following equation:
However, Race [58] reported that in ferritic steels joints, the decarburized
zone width growth is not a parabolic function of the time. The carbon
migration is a diffusion process and occurs according to the profile of
chemical potential of carbon [185] and is not related to the carbon
concentration. However, carbon concentration profile in the steel region, C(X,
t)(s) is used for showing of the decarburization process. Haung [67]
approximated the carbon concentration profile in the steel region by following
equation in a ferritic steel/nickel alloy system:

where C0(s) and Cf(s) are carbon concentrations in the initial and the
decarburized states and DC is the carbon diffusion coefficient and t is exposure
time. The flux of carbon out of the steel is equal the carbon flux into the weld
overlay (i.e. changes in both sides must satisfy following equation):
where C0(w) and Cf(w) are carbon concentration in the initial state and
carburized zone in the nickel alloy and DW is the carbon diffusion coefficient
in this material.
Carbon diffusion coefficient is an important parameter and is dependent upon
its concentration [185] and the presence of other alloying elements and
temperature [67]. The carbon diffusion coefficient (DC0) in ternary Fe-Cr-C is
reported as 1.05X10-11 m2/s while for the 2.25Cr1Mo steel is reported to be
1.58X10-11 m2/s. The apparent DC increased to 2.85x10-9 m2/s in the soft zone
with carbon concentration less than 0.002 at.% at 750˚C [185]. Pavlovsky
[127] reported experimental values for carbon diffusion coefficient in a
specific ferritic/austenitic steel dissimilar joint as a function of temperature
which is 3.16X10-14 m2/s at 650˚C. These different values of DC0 in this
system will be used to describe carbon supply model.
7-2-3-1- Thermodynamic prediction
Thermo-Calc calculation results show that the chemical potential of carbon in
the steel depends upon the carbide type and carbon content. These data are

shown schematically in Figure 7-4(a). As shown in the initial state where the
(Fe, Cr)-rich M3C is stable, carbon chemical potential is μ0(s)= -2.5E+04
while in the equilibrium state with alloyed M23C6 is μe(s)= -3.3E+04.
Moreover, in the low carbon content (0.005 wt.%) with M6C is μf(s)= -
4.1E+04. Nevertheless, the chemical potential of carbon in the bulk zone of
weld overlay region is μ0(w)= -8.26E+04 while in the mixing zone with the
same carbon content is -7.97E+04. Increasing the carbon content of mixing
zone to about 0.1 wt.% rises carbon chemical potential to μe(w)= -7.9E+04.
The carbon migration during welding is not considered and carbon
concentration profile in the as-welded condition is shown in Figure 7-4(b).
Since the carbon activity in the steel is higher than that of the weld overlay, it
is predicted that carbon atoms will diffuse from steel into this material.
Moreover, decarburization should continue indefinitely due to higher chemical
potential.
Figure ‎7-4. Showing schematic diagrams indicates carbon chemical potential and carbon
concentration in the as-welded sample.

7-2-3-2- Carbon supply model
Based on the experimental observation, the decarburized region growth in this
system has different rates. Moreover, there is a carbon migration across the
interface and the carburized zone includes a narrow region of the steel. After a
period of time, the carbon migration stopped at the interface while
redistribution of carbon within the steel region continued. In order to explain
the apparent inconsistencies between experimental results for decarburization
and the thermodynamic predictions and previously published research a new
model is developed based on the present carbides in different regions of steel
microstructure which is called the carbon supply model. The hardness profile
is considered to represent the carbon content in the steel and variation of
carbon chemical potential in this steel is shown schematically with correlation
to the carbide type.
Hardness surveys indicate that in the D8 sample, the hardness of the near-
HAZ region is still higher than that of the far-HAZ region and decarburization
is limited to small regions. However, it has been reported that exposure at
650˚C for times over 100 hours [168, 186] causes instability of M3C and M2C
carbides in this steel. This difference can be explained by nucleation of
alloyed M2C carbide in the near-HAZ region. Precipitation of this type of
carbide causes higher hardness values and prevents carbon migration in this
region.
In this condition decarburization occurs by dissolution of present coarse M3C
carbides and growth of metastable carbides in other regions. This process is

considered in the other models [169, 181, 187] and the rate of carbon supply is
low. The carbon concentration profile did not change significantly compared
to the as-welded condition (see Figure 7-5(a)). The calculated DC in this
condition is about 10 -16 m2/s which in the range of the diffusivity of C in this
steel reported in previous researches [127, 180].
Figure ‎7-5. Showing schematic diagrams indicates carbon chemical potential variation
and carbon concentration in different states; (a) in the D8 sample, (b) in the D12 sample,
(c) in the D32 sample and (d) in the D64 and the D128 samples. (Values are same as the
Figure 7-4)

The carbon concentration changes rapidly in the D12 sample (see Figure
7-5(b)). The nuclei of alloyed carbides in the near-HAZ region have formed in
the neighbourhood of other carbides. Therefore carbon supply is governed by
this region with DC = 10 -8 m2/s which is in the range of the diffusivity of C in
ferrite and the soft zone [185].
The chemical potential of carbon at the interface is μf(s) and there are different
slopes in the carbon potentials to ensure carbon diffusion. The decarburized
region width is equal to the near-HAZ region and carbon supply continues
until all M2C carbides transformed to other metastable carbides. The chemical
potential of carbon in other regions changes with increasing exposure time to
32 days (see Figure 7-5(c)).
DC in this condition is about 10-16 m2/s which means that carbon supply is
governed by other regions again. Figure 7-5(d) shows the carbon potential and
concentration in the D64 and D128 sample and chemical potential at border is
higher than steel region. This is related to formation of carbide network at the
interface which is shown schematically in Figure 5-3. EDX line scans over the
interface in these sample reveal that these particles are (Mo, Cr)-rich and
EBSD results indicate their structure is M23C6. The TEM work indicates that
increasing exposure time causes a little coarsening of the M23C6 carbides at
the interface.
Formation of carbides at interface is reported in former works of the aging of
dissimilar weld joints between 2.25Cr1Mo steel and different nickel alloys [6,
143, 149, 188, 189]. Moreover, it was reported [9, 143, 149] that an austenite
band formed parallel to interface which is related to diffusion of Ni and Cr

from the nickel alloy during bonding. The nickel rich austenite band has
different precipitation behaviour compared to that of other regions in the steel.
Interruption of carbon migration across the interface and carbon chemical
potential differences in the steel result in carbon supply to the soft zone from
the bulk steel and carbon concentration becomes even through the steel region.
The re-hardening of the near-HAZ region is related to formation of Cr-rich
carbides in this region which indicates that carbon content is increasing. Based
on thermodynamic calculations, the M23C6 carbide is the equilibrium carbide
in this steel and therefore increasing the exposure time results in stability of
alloyed M23C6 carbide in all regions. The slope of chemical potential line in
the steel region at equilibrium state will be zero.
7-2-4- Formation of fine ferrite grains in the

near-HAZ region of the D128 sample
Although the carbon depleted zone (CDZ) formation has been reported in
previous research on dissimilar welds, there is no information about the
microstructural evolution in this region after long term exposure. Albert [129]
reported that in the dissimilar welds between Cr-Mo steels and austenitic
stainless steels, the soft zone re-crystallized during the thermal exposure.
Moreover, Foret [139] analysed a dissimilar weld between two Cr-Mo-V
steels, and found that the re-crystallization of ferrite was observed in the CDZ
at temperatures above 700˚C.
Experimental observations of the near-HAZ region in the D64 and D128
samples indicates M23C6 carbides are formed at the expense of coarse Mo-rich
M6C carbides which indicates dissolution of Mo into the matrix. Moreover,

the IES images confirmed that in the D128 sample, precipitates in the near-
HAZ region are along the grains boundaries.
Thermodynamic calculation of the stable phases in this steel revealed that Mo
content (wt.%) in the coexisting ferrite with M23C6 in equilibrium 5.7X10-3
while it is 5.4X10-3 with metastable M6C and M7C3 carbides. Therefore, fine
grain ferrite formation could be related to the carbide transformation and the
related Mo replenishment in the coexisting ferrite, with new ferrite grains
being formed between carbide particles with following reaction.
Bhadeshia and Fujita [170] have shown similar reactions in the modelling of
alloy carbide sequence in power plant steels. Although there are
disagreements about the changes in Mo content in the carbides with exposure
time, the proposed reaction is in accord with the work of Jayan [48].
7-3- Microstructural evolution in the weld

overlay
Based on spark analysis, the weld overlay region is a nickel based alloy
similar to IN625 but which contains higher iron and carbon contents compared
to the standard IN625. The microstructural development in this nickel alloy
weldment can be significantly influenced by microsegregation and phase
transformation during non-equlibrium solidification because the dendritic
structure results in formation of intermetallic secondary solidification
constituents in the interdendritic regions.

The intermixing between substrate and the overlay material during welding
results in a variation of Fe and C contents in the weld overlay material.
Therefore, solidification and ageing behaviour of different regions are various.
Finally, there is a metallurgical bond to the steel which has a higher carbon
activity and carbon migration from the steel substrate causes further changes
in the microstructure of this material.
7-3-1- Microstructure in the as-welded

sample
Phase formation during solidification in the nickel based alloys including
IN625 [69, 85] in the bulk solidification [69, 71, 85] and in weld metal [33,
74, 77, 79, 81-83, 150, 190, 191] have been thoroughly investigated in
previoius research. It is reported that the dominating solidification reaction is
the enrichment of the remaining interdendritic liquid in niobium, and the
consequent formation of Laves phase and/or (Nb, Ti)-rich MC carbide during
final stages of solidification [71]. It is reported that the residual liquid
solidifies with a eutectic-like reaction [84] and that the phase formation is
strongly influenced by the C/Nb ratio.
Cieslak [70] showed that carbon addition causes formation of more MC
carbide at the expense of Laves phase. The solidification of the nickel alloy
weldment is generally similar to the above mentioned process. Although the
welding wires or electrodes are produced with no intermetallic phases, after
welding TCP phases appear in the microstructure [74, 81, 192]. The size and
type of the terminal eutectic-like constituents in IN625 weldments is affected

by heat input during welding [33] and cooling rate; however, the amount of
carbide present remains at constant level [83].
The experimental observations indicate that weld overlay material has a
rapidly solidified microstructure and consists of columnar grains originating
from the substrate (see Figure 4-9). Within the grains are finely spaced
dendrites which indicate a general loss of homogeneity (Figure 4-10).
Moreover, EPMA results (Figure 4-13) confirmed the formation of (Nb, Ti)-
rich points in the bulk zone. There is no evidence for formation of phases in
the mixing zone.
Thermodynamic calculations reveal that phase formation during non-
equilibrium solidification of the weld overlay depends on Fe and C content. In
the bulk zone, MC, (δ) and (σ) phases are predicted to form (see Figure 6-8).
The MC and δ are Nb-rich while the σ is a (Cr, Mo)-rich phase. Increasing the
Fe content with the same level of carbon in the mixing zone causes formation
of Laves phase (see Figure 6-9).
However, addition of carbon delays formation of Laves phase by formation of
σ phase (see Figure 6-10). These phases must be present in the interdendritic
regions of the as-welded sample. The (Nb, Ti)-rich points are related to the
MC and/or δ phases, and cooling rate has a critical affect on formation of
these phases. Higher cooling rate in the mixing zone will result in the absence
of δ phase in this region. However, it is believed that there are some fine
carbide precipitates in this region.

7-3-2- Microstrcutral evolution following high

temperature exposure
7-3-2-1- Bulk zone hardening
Different studies have dealt with the effects of aging at temperatures in the
range of 600 to 750°C on the wrought IN625 microstructure [87, 88, 90, 98,
101, 103, 104, 106-108, 114, 115, 117, 118] and weld overlay material [95,
119, 150]. In this condition, strengthening is brought about by precipitation of
the metastable (γ‟‟) phase with a body centred tetragonal DO22 structure [69,
97, 98]. It is reported that (γ‟‟) phase transform to orthorhombic (δ) phase with
DOa structure on prolonged aging [99].
Experimental observation indicates that hardening of the bulk zone in the D1
sample (Figure 5-4) continues with increasing exposure time (Figure 5-5).
However, exposure times of more than 8 days causes little change in the bulk
zone hardness. The hardness values in the as-welded condition are well
matched with those of the weld metal [119] and rapidly solidified IN625 [82]
from previous research. The hardness values after 8 days of thermal exposure
are also well matched with those of the weld metal [95] and wrought IN625
[69].
As such, it may be argued that the hardening of the bulk zone with thermal
exposure at 650˚C occurs by precipitation of (γ‟‟) phase. This phase has a
coherent structure and cannot be seen in this study. However, needles of the
(δ) phase become visible after prolong ageing in the D128 sample in some
regions which can related to transformation of (γ‟‟) to (δ) phase.

7-3-2-2- Mixing zone hardening
Formation of a narrow hard band adjacent to the interface in the weld metal,
after heat exposure in the range of 600˚C to 700˚C has been reported in
previous research in dissimilar weld joints between steels and Ni alloys [6, 7,
9, 66]. It has been argued that the increasing in hardness is related to the
precipitation of carbides on the higher Cr side of the interface.
Observations of the D8 sample in this work indicates hardening of the mixing
zone and increasing exposure time to 16 days results in formation of a hard
band ~30 μm wide. Increasing exposure time causes only small changes, both
in hardness values or hard band wide. Nevertheless, it is notable that the hard
band was also formed in the free standing weld overlay sample (WD128).
Figure ‎7-6. Showing Schaeffler diagram and indicating Cr equivalent for the mixing
zone composition and stability of austenite in this region.
Therefore it is suggested that carbon migration has no effect on hard band
formation, which indicates that it is not related to formation of carbides. The
Schaeffler diagram (see Figure 7-6) indicates that (γ) austenite is stable in the

mixing zone. Thermodynamic calculation indicates that the (σ) phase is stable
at 650˚C in this region (Figure 6-6). Therefore it is suggested that hard band
formation is related to the precipitation of the hard intermetallic (σ) phase in
this zone
7-3-2-3- Phase precipitation in the interfacial
region
It is known that some carbides and intermetallic phases are normally
precipitated in the IN625 alloy during either aging or service at temperatures
in the range of 600˚C to 750˚C in wrought [93, 193, 194] and as-solidified
conditions [91, 96, 97, 99, 102, 195]. It is reported [104, 119] that below
700˚C, the M23C6 carbide is the most stable type of carbide in the IN625 alloy.
Moreover, precipitation of Mo-rich (μ) phase in this alloy has also been
reported [92].
Observations indicate that some veins were formed in the interfacial region
following thermal exposure tests. These veins are ~100 μm in length and
increasing exposure time has a little effect on the length of them. BSE
micrographs indicated that these veins are associated with carbides in the
microstructure and that the carbide population increased with increasing
exposure time around the grain boundaries. The EPMA results of the D64 and
the D128 samples confirm that in the interfacial region, (Cr, Mo)-rich carbide
precipitates are present. Analysis of the WD128 sample reveals that
precipitation of carbides in this region is related to carbon migration from the
steel.

Thermodynamic calculations for the bulk zone show that at 650˚C, the (Cr,
Mo)-rich M23C6 carbide is stable. However, there are different stable carbides
with changing Fe content, and in the mixing zone the Mo-rich M6C is stable.
Increasing the carbon content results in a higher carbide mol.% at the expense
of other phases in the bulk zone while in the mixing zone causes the stability
of (Cr, Mo)-rich M23C6. These results are similar to those reported in other
works based on thermodynamic calculations [103, 142].
Therefore, it is suggested that MC carbides formed during the final stage of
solidification transform to the (Cr, Mo)-rich M23C6 in the neighbourhood of
grain boundaries in the interfacial region. Carbon migration across the
interface results in growth of these carbide precipitates. However, the constant
length of the carbide precipitation region in the weld overlay (veins) can be
related to the formation of other phases.
Evaluation of the weld overlay region microstructure in different samples
indicates that some intermetallic precipitates are present at the grain
boundaries which were identified as TCP phases. These precipitates are (Mo,
Nb)-rich and were formed in the interdendritic regions. However, EPMA
results reveal that the Nb-rich areas are different from the Mo-rich areas.
Thermodynamic calculations show that although (μ) phase is not formed
during non-equilibrium solidification, this phase is stable at 650˚C with Fe
contents of less than 8 wt.% (in addition to Nb-rich (δ) phase). Increasing the
Fe content results in formation of (σ) phase at the expense of the (μ) phase and
above 15wt.%, (σ) phase is stable with Nb-rich (δ) phase.

Therefore, beyond the ~100 μm distance and due to lower Fe content, the
(Mo, Cr)-rich (μ) phase becomes stable and formation of this phase decreases
the chemical potential of Cr and Mo and prevents carbide formation in this
area. However, this phase formed very close to the Nb-rich (δ) in the
interdendritic regions and so it is difficult to distinguish these phases from
each other. It can be said that both these phases are co-exist in these regions.
In the D128 sample some acicular bright precipitates become visible in the
microstructure very close to carbides which are suggested to be the (δ) phase.
In this region, the Cr and Mo atoms formed carbides and as such, the Nb
atoms are free to form the Ni3Nb phase. The (δ) phase was also seen in the
WD128 sample along the grain boundaries close to interface because (μ)
phase is not stable at this region due to high Fe content.
7-4- Summary
The precipitation behaviour following thermal exposure, in various regions of
the weld overlaid tube, are different due to variation of as-welded
microstructure and composition. Moreover carbon migration causes further
changes in the interfacial region due to wide differences in its chemical
potential in the 2.25Cr1Mo steel and the nickel alloy.
The main process in the early stages of thermal exposure is dissolution of
metastable carbides in the steel region and formation of carbides at the
interface and along the grain boundaries of the interfacial region. However,
different microstructures in the steel region cause different rates of dissolution
and precipitation sequences which are considered in a new model to explain
microstructural evolution in the different regions of steel. It is clear that the

width of the HAZ zone is related to heat input during welding. Therefore
choosing a welding process with a narrower HAZ could be useful in changing
the width of the soft zone during service.
Formation of a carbide network at the interface interrupts carbon supply
across the interface. However, there is carbon redistribution in the steel region
which causes further changes in the microstructure of the near-HAZ region.
The re-hardening of this region and the formation of fine ferrite grains is
explained by formation of fine Cr-rich carbides and Mo replenishment in the
coexisting ferrite. Due to the above observations, carbon migration cannot
continue indefinitely and it is observed that the mechanical properties and
microstructure of the central and the far-HAZ zones remain almost unchanged
at 650˚C. Therefore, it is suggested that increasing the steam temperature to
above 450˚C in the power plants is possible. However, the technical issues in
this regard should be considered.
The Fe/Mo ratio is the major parameter in formation of different phases in the
microstructure of weld overlay region. Increasing distance from the interface
with lower Fe/Mo ratio results in formation of the (μ) intermetallic phase at
the expense of carbide precipitates which are stable at higher ratio along the
grain boundaries. This fact indicates that the steel substrate has some effect on
the first 100 μm of the weld overlay material. Therefore, it is suggested that by
consideration of corrosion rate and it relation to the thickness, the weld
overlay thickness can be reduced which means reduction of costs for
manufacturing of the bimetallic tube.

Based on the results of this research, it is clear that formation of the hard band
is not related to the carbide formation as it also formed in the WD128 sample.
Although formation of this hard band layer has not a significant effect on the
hook stress, due to the temperature fluctuation and 3 dimensional forces in the
tube, the presence of a hard band adjacent to the soft zone could degrade
mechanical properties of the tube during its service life.
7-5- Future work
The first important issue is to determine the realistic temperature profile in the
wall thickness of a super heater tube in the service condition. This thermal
analysis helps to understand the effect of increasing service temperature on the
different regions of this tube especially at interface.
The second issue is that the aim of applying a weld overlay on ferritic steel is
to increase the corrosion resistance in the severe operating conditions such as
WtE boilers. The corrosion resistance is related to the microstructure and it is
necessary to analyse this property of the weld overlay. It is clear that the weld
overlay material contains about 9 wt% Fe which may cause different corrosion
behaviour compared to that of the standard wrought Inconel 625 with less than
5 wt% Fe content.
The third issue is to understand the affects of the hard band and the soft zone
on the mechanical properties under service conditions. Therefore mechanical
tests need to be done on different samples.

References
8- References
1. Adamiec, J., "High temperature corrosion of power boiler components

cladded with nickel alloys" Materials Characterization, 2009. 60(10):
p. 1093-1099.
2. Lee, S.H. ; Themelis, N.J. ; Castaidi, M.J., "High-temperature

corrosion in waste-to-energy boilers" Journal of Thermal Spray
Technology, 2007. 16(1): p. 104-110.
3. Kawahara, Y., "Development and application of high-temperature

corrosion-resistant materials and coatings for advanced waste-to-
energy plants" Materials at High Temperatures, 1997. 14(3): p. 261-
268.
4. Cane, B.J. ; Fidler, R.S., "The Effect of Microstructure and Grain Size
on the Creep and Rupture Properties of 2.25Cr--Mo and 9Cr--Mo
Steels", in International Conference on Ferritic Steels for Fast Reactor
Steam Generators. 1978: London. p. 193-199.
5. Orr, J. ; Beckitt, F.R. ; Fawkes, G.D., "The Physical Metallurgy of

Chromium--Molybdenum Steels for Fast Boilers", in International
Conference on Ferritic Steels for Fast Reactor Steam Generators.
1978: London. p. 91-109.
6. Nicholson, R.D., "Effect of aging on interfacial structures of nickel-

based transition joints" Metals Technology, 1984. 11(MAR): p. 115-
124.
7. Lopez, B. ; Gutierrez, I. ; Urcola, J.J., "Microstructural analysis of

steel nickel alloy clad interfaces" Materials Science and Technology,
1996. 12(1): p. 45-55.
8. Uusitalo, M.A. ; Vuoristo, P.M.J. ; Mantyla, T.A., "High temperature

corrosion of coatings and boiler steels in oxidizing chlorine-containing
atmosphere" Materials Science and Engineering A-Structural
Materials Properties Microstructure and Processing, 2003. 346(1-2): p.
168-177.
9. Lopez, B. ; Gomez, X. ; Echeberria, J. ; Gutierrez, I. ; Urcola, J.J.,

"Interface analysis on diffusion bonded bimetallic composites .1.
Extended solid solubility", in Cmmc 96 - Proceedings of the First
International Conference on Ceramic and Metal Matrix Composites,
Pts 1 and 2, M. Fuentes, J.M. MartinezEsnaola, and A.M. Daniel,
Editors. 1997, Trans Tech Publications: Clausthal Zellerfe. p. 695-702.
10. Lai, G.Y. ; Blogg, N., "Unifuse overlay cladding for surface protection
against corrosion and erosion/corrosion in power boilers and waste
heat recovery", in OMMI. 2004. p. 1-19.

References
11. Kawahara, Y., "Application of high temperature corrosion-resistant

materials and coatings under severe corrosive environment in waste-
to-energy boilers" Journal of Thermal Spray Technology, 2007. 16(2):
p. 202-213.
12. www.wte.org. [cited.
13. Kiser, S.D. ; Hinshaw, E.B. ; Orsini, T., "Extending the life of fossil
fired boiler tubing with cladding of nickel based alloy materials", in
NACE International CORROSION/2006, 61st Annual Conference &
Exposition. 2006, NACE International, P.O. Box 218340, Houston,
TX, 77218, USA,: San Diego, CA; USA.
14. Sidky, P.S. ; Hocking, M.G., "Review of inorganic coatings and

coating processes for reducing wear and corrosion" British Corrosion
Journal, 1999. 34(3): p. 171-183.
15. Stringer, J., "Coatings in the electricity supply industry: past, present,
and opportunities for the future" Surface & Coatings Technology,
1998. 108(1-3): p. 1-9.
16. Wyatt, L.M., "Materials tchnology in power generation in UK"

Materials Science and Engineering, 1971. 7(5): p. 237-&.
17. Viswanathan, R. ; Henry, J.F. ; Tanzosh, J. ; Stanko, G. ;

Shingledecker, J. ; Vitalis, B. ; Purgert, R., "US program on materials
technology for ultra-supercritical coal power plants" journal of
Materials Engineering and Performance, 2005. 14(3): p. 281-292.
18. Viswanathan, R. ; Purgert, R. ; Rao, U., "Materials technology for

advanced coal power plants", in 1st International Conference on
Super-High Strength Steels. 2005: Rome, Italy.
19. Uusitalo, M. ; Kaipiainen, M. ; Vuoristo, P. ; Mantyla, T., "High

temperature erosion-corrosion of superheater materials and coatings
in chlorine containing environments", in High Temperature Corrosion
and Protection of Materials 5, Pts 1 and 2, R. Streiff, et al., Editors.
2001, Trans Tech Publications Ltd: Zurich-Uetikon. p. 475-482.

corrosion of coatings and boiler steels in reducing chlorine-containing
atmosphere" Surface & Coatings Technology, 2002. 161(2-3): p. 275-
285.

corrosion of coatings and boiler steels below chlorine-containing salt
deposits" Corrosion Science, 2004. 46(6): p. 1311-1331.
22. Montgomery, M. ; Biede, O. ; Larsen, O.H., "Experiences with Inconel

625 in biomass and waste incineration plants", in High-Temperature
Oxidation and Corrosion 2005, S. Taniguchi, et al., Editors. 2005,
Trans Tech Publications Ltd: Stafa-Zurich. p. 523-530.

References
23. Cha, S.C. ; Wolpert, P., "High-temperature erosion and corrosion

measurement of thermally sprayed materials" Advanced Engineering
Materials, 2003. 5(4): p. 213-217.
24. Vollertsen, F. ; Partes, K. ; Meijer, J., "State of the art of laser

hardening and cladding", in Third International WLT-conference on
Lasers in Manufacturing, E. Beyer, et al., Editors. 2005: Munich,
Germany.
25. Lai, G. ; Hulsizer, P., "Corrosion control by modern weld overlay

technology", in 14th International Corrosion Congress (ICC). 1999:
Cape Town; South Africa.
26. Thornton, C.E. ; Cooper, C., "Welding and corrosion performance of

INCO-WELD 686CPT filler metal in Waste-To-Energy power plants".
2003, Special Metals Welding Products Co.
27. Deacon, R.M. ; DuPont, J.N. ; Marder, A.R., "High temperature

corrosion resistance of candidate nickel-based weld overlay alloys in a
low NOx environment" Materials Science and Engineering A-Structural
Materials Properties Microstructure and Processing, 2007. 460: p. 392-
402.
28. Amador, P. ; Lai, G. "Application of unifuse overlay tubes in the

convection section of waste-to-energy boilers." in 11th Annual North
American Waste to Energy Conference. 2003. New York: Amer. Soc.
Mechanical Engineers.
29. Ganesan, P. ; Renteria, C.M. ; Crum, J.R., "Versatile corrosion

resistance of INCONEL alloy 625 in various aqueous and chemical
processing environments", in Superalloys 718, 625 and Various
Derivatives, E.A. Loria, Editor. 1991, The Minerals, Metals and
Materials Society (TMS): Pittsburgh, PA. p. 663-680.
30. Capitamescu, D., "Alloy 625 weld overlays for offshore and onshore
projects", in Superalloys 718, 625 and Various Derivatives, E.A.
Loria, Editor. 1991, The Minerals, Metals and Materials Society
(TMS): Pittsburgh, PA. p. 821-835.
31. Smith, G.D. ; Tillack, D.J. ; Patel, S.J., "Alloy 625 - Impressive
past/significant presence/awesome future", in Superalloys 718, 625,
706 and Various Derivatives, E.A. Loria, Editor. 2001, The Minerals,
Metals and Materials Society (TMS): Warrendale, PA. p. 35-46.
32. Lai, G. ; Van Lienden, C. ; Hou, P.Y., "Erosion/corrosion behavior of

weld overlays in HCl containing atmospheres", in Corrosion/2001.
2001, NACE International: Houston, TX.
33. Wilson, I.L.W. ; Gourley, R.G. ; Walkosak, R.M. ; Bruck, G.J., "The
effect of heat input on microstructure and cracking in alloy 625 weld
overlays", in Superalloys 718, 625 and Various Derivatives, E.A.

References

34. Oh, Y.K. ; Devletian, J.H. ; Chen, S.J., "Low-dilution electrosalg

cladding for shipbuilding" Welding Journal, 1990. 69(8): p. 37-44.
35. Bates, D., "Cost effective maintenance by spiral welding" Welding and
Metal Fabrication, 1983. 51(8): p. 440.
36. Wang, X.L. ; Payzant, E.A. ; Taljat, B. ; Hubbard, C.R. ; Keiser, J.R. ;
Jirinec, M.J., "Experimental determination of the residual stresses in a
spiral weld overlay tube" Materials Science and Engineering A-
Structural Materials Properties Microstructure and Processing, 1997.
232(1-2): p. 31-38.
37. Siman, A. ; Shelton, J., "Feedwater heater application of T-22 tubing",

Bobcockpower Thermal Engineering International Inc. .
38. Xu, H. ; Huang, X.K. ; Yang, Z. ; Pan, J.T., "Property changes of

12Cr1MoV with 10CrMo910 main steam piping after long-term high-
temperature service" Engineering Failure Analysis, 2003. 10(2): p.
245-250.
39. Matsumoto, I. ; Wignarajah, S. ; Shinoda, T. ; Yamada, S.,

"Microstructure and Mechanical Properties of 2.25Cr1Mo and
0.50Mo Heat Resisting Steels After Long Service" Transaction of the
Japan Welding Society, 1983. 14(1): p. 34-39.
40. Song, S.H. ; Faulkner, R.G. ; Flewitt, P.E.J., "Quenching and

tempering-induced molybdenum segregation to grain boundaries in a
2.25Cr-1Mo steel" Materials Science and Engineering A-Structural
23-27.
41. Gojic, M. ; Kosec, L. ; Matkovic, P., "The effect of tempering

temperature on mechanical properties and microstructure of low alloy
Cr and CrMo steel" Journal of Materials Science, 1998. 33(2): p. 395-
403.
42. Bhadeshia, H.K.D.H., "Bainite in steels : transformations,

microstructure and properties ". 2001, London: Institute of Materials.
43. Yang, J.R. ; Huang, C.Y. ; Yang, C.N. ; Horng, J.L., "Microstructural
examination of 2.25Cr-1Mo steel steam pipes after extended service"
Materials Characterization, 1993. 30(2): p. 75-88.
44. Mitchell, D.R.G. ; Small, R., "Microstructural evolution in seven

2.25Cr-1Mo superheater outlet headers resulting from service
exposure" Science and Technology of Welding and Joining, 2001.
6(3): p. 168-176.

References
45. Chang, H.J. ; Kai, J.J., "The effects of thermal treatment on the
microstructure and tensile properties of 2.25Cr1Mo Steel" Scripta
Metallurgica et Materialia, 1990. 24(11): p. 2101-2106.
46. Parameswaran, P. ; Vijayalakshmi, M. ; Shankar, P. ; Raghunathan,

V.S., "Influence of carbon content on microstructure and tempering
behaviour of 2-1/4Cr1Mo steel" Journal of Materials Science, 1993.
28(20): p. 5426-5434.
47. Mitchell, D.R.G. ; Ball, C.J., "A quantitative X-ray diffraction and
analytical electron microscopy study of service-exposed 2.25Cr-1Mo
steels" Materials Characterization, 2001. 47(1): p. 17-26.
48. Jayan, V. ; Khan, M.Y. ; Hussain, M., "X-ray investigation of solid

solution partitioning in 2.25Cr-1Mo steel after extended elevated
temperature service in power station" Materials Science and
Technology, 2003. 19(11): p. 1546-1552.
49. Yang, H.T. ; Kim, S., "A study on the mechanical strength change of
2.25Cr-1Mo steel by thermal aging" Materials Science and
Engineering A-Structural Materials Properties Microstructure and
Processing, 2001. 319: p. 316-320.
50. Lonsdale, D. ; Flewitt, P.E.J., "Damage accumulation and

microstructural changes occurring during the creep of a 2
1/4%Cr1%Mo steel" Materials Science and Engineering, 1979. 39(2):
p. 217-229.
51. Ahila, S. ; Radhakrishnan, V.M. ; Iyer, S.R., "Effect of temperature on

the microstructure of 2.25Cr1Mo steel during creep testing" Journal of
Materials Science Letters, 1993. 12(16): p. 1296-1298.
52. Gope, N. ; Chatterjee, A. ; Mukherjee, T. ; Sarma, D.S., "Influence of

long-term aging and superimposed creep stress on the microstructure
of 2.25Cr1Mo steel" Metallurgical Transactions A-Physical Metallurgy
and Materials Science, 1993. 24(2): p. 315-326.
53. Abdellatif, A.M. ; Corbett, J.M. ; Taplin, D.M.R., "Analysis of

carbides formed during accelerated aging of 2-25Cr-1Mo steel" Metal
Science, 1982. 16(2): p. 90-96.
54. Cheruvu, N.S., "Degradation of mechanical properties of Cr-Mo-V

and 2.25Cr-1Mo steel components after long-term service at elevated
temperatures" Metallurgical Transactions A-Physical Metallurgy and
Materials Science, 1989. 20(1): p. 87-97.
55. Baker, R.G. ; Nutting, J., "The tempering of a 2.25%Cr–1%Mo steel

after quenching and normalizing" Journal of the Iron and Steel
Institute., 1959. 208: p. 257-268.
56. Jones, W.B. ; Vandeavyle, J.A., "Substructure and strengthening in

2.25Cr1Mo steel at elevated temperatures" Metallurgical Transactions

References
A-Physical Metallurgy and Materials Science, 1980. 11(8): p. 1275-

1286.
57. Pigrova, G. ; Sedov, V. ; Archakov, Y., "Carbide transformations in

Cr-Mo steels in the process of long-term ageing and operation" Metal
Science and Heat Treatment 1997. 39(9): p. 371-375.
58. Race, J.M.,"Carbon diffusion across dissimilar steel

welds",1992,Ph.D.,Materials Science and Metallurgy St. John's
College
59. Klueh, R.L. ; Swindeman, R.W., "Thw microstructure and mechanical

properties of a modified 2.25Cr1Mo steel" Metallurgical Transactions
1034.
60. Thomson, R.C. ; Bhadeshia, H.K.D.H., "Changes in chemical

composition of carbides in 2.25Cr-1Mo power plant steel. part 1.
Bainitic microstructure" Materials Science and Technology, 1994.
10(3): p. 193-203.
61. Thomson, R.C. ; Bhadeshia, H.K.D.H., "Changes in chemical

composition of carbides in 2.25Cr-1Mo power plant steel. part 2.
Mixed microstructure" Materials Science and Technology, 1994.
10(3): p. 205-208.
62. Parameswaran, P. ; Paul, V.T. ; Vijayalakshmi, M. ; Raghunathan,

V.S., "Microstructural evolution in a single pass autogenously welded
2.25Cr-1Mo steel" Transactions of the Indian Institute of Metals, 2004.
57(3): p. 253-264.
63. Miranda, R.M. ; Fortes, M.A., "Austenite grain growth, microstructure

and hardness in the heat-affected-zone of a 2.25Cr1Mo steel"
Materials Science and Engineering A-Structural Materials Properties
Microstructure and Processing, 1989. A108(1-2): p. 1-8.
64. Bergquist, E.L. ; Karlsson, L. ; Thuvander, M. ; Keehan, E.,

"Microstructure and properties of post weld heat treated 2.25Cr1Mo
weld metal", in 7th International Conference: Trends in Welding
Research. 2005, ASM International: Pine Mountain, GA; USA. p. 963-
968.
65. Rao, R.V.S. ; Parameswaran, P. ; Dayal, R.K., "Microstructural

modification in a laser-treated 2.25Cr-1Mo steel" Journal of Materials
Engineering and Performance, 2001. 10(5): p. 515-520.
66. Laha, K. ; Chandravathi, K.S. ; Rao, K.B.S. ; Mannan, S.L. ; Sastry,

D.H., "An assessment of creep deformation and fracture behavior of
2.25Cr-1Mo similar and dissimilar weld joints" Metallurgical and
Materials Transactions A-Physical Metallurgy and Materials Science,
2001. 32(1): p. 115-124.

References
67. Buchmayer, B. ; Cerjak, H. ; Witwer, M. ; Kiekaldy, J.S., "Carbon

diffusion and microstructure in dissimilar Cr-Mo-V welds and their
influence on the mechanical properties", in Recent Trends in Welding
Science and Technology. 1989, ASM International, Materials Park,
Ohio 44073, USA Gatlinburg, Tennessee; USA. p. 237-242.
68. Pilling, J. ; Ridley, N., "Tempering of 2.25%Cr1%Mo low-carbon

steels" Metallurgical Transactions A-Physical Metallurgy and
69. Eiselstein, H.L. ; Tillack, D.J., "The invention and definition of alloy
625", in Superalloys 718, 625 and Various Derivatives, E.A. Loria,
Editor. 1991, The Minerals, Metals and Materials Society (TMS):
Pittsburgh, PA. p. 1-13.
70. Cieslak, M.J. ; Headley, T.J. ; Kollie, T. ; Romig, A.D.J., "A melting
and solidification study of alloy 625" Metallurgical Transactions A-
Physical Metallurgy and Materials Science., 1988. 19A(9): p. 2319-
2331.
71. Floreen, S. ; Fuchs, G.E. ; Yang, W.J., "The metallurgy of alloy 625",
in Superalloys 718, 625, 706 and Various Derivatives, E.A. Loria,
Editor. 1994, The Minerals, Metals and Materials Society (TMS):
Warrendale, PA. p. 13-37.
72. Atamert, S. ; Bhadeshia, H.K.D.H., "Nickel based hardfacing alloys

for high-temperature applications" Materials Science and Technology,
1989. 5(12): p. 1220-1228.
73. Luer, K. ; DuPont, J. ; Marder, A. ; Skelonis, C., "Corrosion fatigue of

alloy 625 weld claddings in combustion environments" Materials at
High Temperatures, 2001. 18(1): p. 11-19.
74. Cieslak, M.J. ; Knorovsky, G.A. ; Headley, T.J. ; Romig, A.D., "The
use of new PHACOMP in understanding the solidification
microstructure of nickel-base alloy weld metal" Metallurgical
Transactions A-Physical Metallurgy and Materials Science, 1986.
17(12): p. 2107-2115.
75. Thompson, R.G. ; Mayo, D.E. ; Radhakrishnan, B., "The relationship

between carbon content, microstructure, and intergranular liquation
in cast nickel alloy-718" Metallurgical Transactions A-Physical
Metallurgy and Materials Science, 1991. 22(2): p. 557-567.
76. Cieslak, M.J. ; Headley, T.J. ; Romig, A.D., "The welding metallurgy
of hastelloy alloy-C-4, alloy-C-22, and alloy-C-276" Metallurgical
17(11): p. 2035-2047.
77. Cieslak, M.J., "The welding and solidification metallurgy of alloy 625"
Welding Journal., 1991. 70(2): p. 49s-56s.

References
78. Cieslak, M.J. ; Headley, T.J. ; Frank, R.B., "The welding metallurgy of
custom age 625 Plus alloy" Welding Journal, 1989. 68(12): p. S473-
S482.
79. Yoon, B.H. ; Ahn, Y.S. ; Lee, C.H., "The effect of dilution on HAZ
liquation cracking in PTAW Ni-base superalloys overlay deposit" The
Iron and Steel Institute of Japan International, 2002. 42(2): p. 178-183.
80. Cooper, K.P. ; Slebodnick, P. ; Thomas, E.D., "Seawater corrosion

behavior of laser surface modified Inconel 625 alloy" Materials
Science and Engineering A-Structural Materials Properties
Microstructure and Processing, 1996. 206(1): p. 138-149.
81. Ogborn, J.S. ; Olson, D.L. ; Cieslak, M.J., "Influence of solidification

on the microstructural evolution of nickel base weld metal" Materials
Microstructure and Processing, 1995. 203(1-2): p. 134-139.
82. Dey, G.K. ; Albert, S. ; Srivastava, D. ; Sundararaman, M. ;

Mukhopadhyay, P., "Microstructural studies on rapidly solidified
INCONEL-625" Materials Science and Engineering A-Structural
Materials Properties Microstructure and Processing, 1989. 119: p. 175-
184.
83. Tinoco, J. ; Fredriksson, H., "Solidification of a modified Inconel 625

alloy under different cooling rates" High Temperature Materials and
Processes, 2004. 23(1): p. 13-24.
84. Hojerslev, C. ; Tiedje, N. ; Hald, J., "Segregation effects and phase

developments during solidification of alloy 625", in Solidification and
Gravity Iv, R. Roosz, M. Rettenmayr, and Z. Gacsi, Editors. 2006,
Trans Tech Publications Ltd: Zurich-Uetikon. p. 373-378.
85. Barker, J.F. ; Cox, J.D. ; Margolin, E., "INCONEL 625 - An alloy for
steam and gas turbines" Metal Progress, 1968. 93(4): p. 91-&.
86. Matthewss, S.J., "Thermal Stability of Solid solutioned Strengthened

High Performance Alloys", in 3rd International Superalloy
Symposium. 1976: Seven Springs, PA.
87. Radavich, J.F. ; Fort, A., "Effects of long-time exposure in alloy-625 at

1200°F, 1400°F and 1600°F", in Superalloys 718, 625, 706 and
Various Derivatives, E.A. Loria, Editor. 1994, The Minerals, Metals
and Materials Society (TMS): Warrendale, PA. p. 635-647.
88. Kohler, M., "Effect of the elevated-temperature-precipitation in alloy

625 on preoperites and microstructure", in Superalloys 718, 625, 706
and Various Derivatives, E.A. Loria, Editor. 1991, The Minerals,
Metals & Materials Society(TMS): Pittsburgh, PA. p. 363-374.
89. Heubner, U. ; Kohler, M., "Effect of carbon content and other

variables on the yield strength, ductility and creep preoperties of

References
alloy-625", in Superalloys 718, 625, 706 and Various Derivatives,

E.A. Loria, Editor. 1994, The Minerals, Metals and Materials Society
90. Kohler, M. ; Heubner, U., "Time-Temperature-Sensitization and Time-

Temperature-Precipitation behavior of alloy 625", in CORROSION
96. 1996, NACE International: Denver, Co. p. 427/1-427/10.
91. Tawancy, H.M. ; Allam, I.M. ; Abbas, N.M., "Effect of Ni3Nb

precipitation on the corrosion resistance of INCONEL alloy 625"
Journal of Materials Science Letters, 1990. 9(3): p. 343-347.
92. Tawancy, H.M., "Precipitation characteristics of mu-phase in wrought

nickel-base alloys and its effect on their properties" Journal of
93. Ferrer, L. ; Pieraggi, B. ; Uginet, J.F., "Microstructural evolution

during thermomechanical processing of alloy 625", in Superalloys
718, 625 and Various Derivatives, E.A. Loria, Editor. 1991, The
Minerals, Metals and Materials Society (TMS): Pittsburgh, PA. p. 217-
229.
94. Comprelli, F.A. ; Wolff, U.E., "Stability of high nickel alloys in

superheated steam". 1963, General Electric: san jose.
95. Cortial, F. ; Corrieu, J.M. ; Vernotloier, C., "Heat treatments of weld

alloy 625 - Influence on the microstructure, mechanical properties and
corrosion resistance", in Superalloys 718, 625, 706 and Various
96. Kirman, I., "Precipitation in Fe-Ni-Cr-Nb system" Journal of the Iron

and Steel Institute, 1969. 207: p. 1612-1618.
97. Bohm, H. ; Ehrlich, K. ; Kramer, K.H., "Precipitation behaviour of

Nickel-alloy INCONEL-625" Metall, 1970. 24(2): p. 139-.
98. Sundararaman, M. ; Kishore, R. ; Mukhopadhyay, P., "Some aspects of

the heterogeneous precipitation of the metastable-gamma'' phase in
alloy 625", in Superalloys 718, 625, 706 and Various Derivatives, E.A.
99. Sundararaman, M. ; Mukhopadhyay, P. ; Banerjee, S., "Precipitation

of the delta-Ni3Nb phase in 2 nickel-base superalloys" Metallurgical
19(3): p. 453-465.
100. Mittra, J. ; Dubey, J.S. ; Kulkarni, U.D. ; Dey, G.K., "Role of

dislocation density in raising the stage II work-hardening rate of Alloy
625" Materials Science and Engineering A-Structural Materials
Properties Microstructure and Processing, 2009. 512(1-2): p. 87-91.

References
101. Sundararaman, M. ; Mukhopadhyay, P. ; Banerjee, S., "Carbide

precipitation in nickel base superalloys 718 and 625 and their effect
on mechanical properties", in Superalloys 718, 625, 706 and Various
102. Dahlen, M. ; Fischmister, H., "Carbide precipitation in superalloys",

in Superalloys 1980, Metals Park, OH, American Society for Metals.
103. Evans, N.D. ; Maziasz, P.J. ; Shingledecker, J.P. ; Yamamoto, Y.,

"Microstructure evolution of alloy 625 foil and sheet during creep at
750 degrees C" Materials Science and Engineering A-Structural
412-420.
104. Rai, S.K. ; Kumar, A. ; Shankar, V. ; Jayakumar, T. ; Rao, K.B.S. ;

Raj, B., "Characterization of microstructures in Inconel 625 using X-
ray diffraction peak broadening and lattice parameter measurements"
Scripta Materialia, 2004. 51(1): p. 59-63.
105. Muzyka, D.R., "Controlling microstructures and properties of

superalloys via use of precipitated phases" Metals Engineering
Quarterly, 1971. 11(4): p. 12-&.
106. Shankar, V. ; Rao, K.B.S. ; Mannan, S.L., "Microstructure and

mechanical properties of Inconel 625 superalloy" Journal of Nuclear
Materials, 2001. 288(2-3): p. 222-232.
107. Mathew, M.D. ; Rao, K.B.S. ; Mannan, S.L., "Creep properties of

service-exposed Alloy 625 after re-solution annealing treatment"
108. Mathew, M.D. ; Parameswaran, P. ; Rao, K.B.S., "Microstructural

changes in alloy 625 during high temperature creep" Materials
Characterization, 2008. 59(5): p. 508-513.
109. Vernotloier, C. ; Cortial, F., "Influence of heat treatment on

microstructure, mechanical properties and corrosion behaviour of
alloy-625 forged rod", in Superalloys 718, 625, 706 and Various
110. Corrieu, J.M. ; Vernotloier, C. ; Cortial, F., "Influence of heat

treatments on corrosion behaviour of alloy-625 forged rod", in
Superalloys 718, 625, 706 and Various Derivatives, E.A. Loria, Editor.
1994, The Minerals, Metals and Materials Society (TMS): Pittsburgh,
PA. p. 795-806.
111. Vandervoort, G.F. ; Bowman, J.W. ; Frank, R.B., "Microstructural

characterization of custom age 625-PLUS(R) alloy", in Superalloys

References
718, 625, 706 and Various Derivatives, E.A. Loria, Editor. 1994, The
Minerals, Metals and Materials Society (TMS): Pittsburgh, PA. p. 489-
498.
112. Edris, H.,"Studies on high-velocity oxy-fuel sprayed coating of Inconel

625 and Ni-Cr3C2",1997,Ph.D.,M3 School,University of nottingham
113. Jones, R.H., "Stress-corrosion cracking". 1992: ASM International.
114. Sundararaman, M. ; Kumar, L. ; Prasad, G.E. ; Mukhopadhyay, P. ;

Banerjee, S., "Precipitation of an intermetallic phase with Pt2Mo-type
structure in alloy 625" Metallurgical and Materials Transactions A-
Physical Metallurgy and Materials Science, 1999. 30(1): p. 41-52.
115. Sundararaman, M. ; Mukhopadhyay, P. ; Banerjee, S., "Influence of

intermetallic phase precipitation during prolonged service in Alloy
625 on its properties", in Superalloys 718, 625, 706 and Various
116. alloys, U.P.-C.-r.n.-c.-m.
117. Palanichamy, P. ; Mathew, M.D. ; Latha, S. ; Jayakumar, T. ; Bhanu,

K. ; Rao, S.K. ; Mannan, S.L. ; Raj, B., "Assessing microstructural
changes in alloy 625 using ultrasonic waves and correlation with
tensile properties" Scripta Materialia, 2001. 45(9): p. 1025-1030.
118. Mittra, J. ; Dubey, J.S. ; Banerjee, S., "Acoustic emission technique

used for detecting early stages of precipitation during aging of Inconel
625" Scripta Materialia, 2003. 49(12): p. 1209-1214.
119. Cortial, F. ; Corrieu, J.M. ; Vernotloier, C., "Influence of heat

treatments on microstructure, mechanical propeties, and corrosion
resistance of weld alloy 625" Metallurgical and Materials Transactions
1286.
120. Savage, W.F. ; Nippes, E.F. ; Szekeres, E.S., "Study of weld interface
phenomena in a low-alloy steel" Welding Journal, 1976. 55(9): p.
S260-S268.
121. Cui, Y. ; Xu, C.L. ; Han, Q., "Effect of ultrasonic vibration on unmixed
zone formation" Scripta Materialia, 2006. 55(11): p. 975-978.
122. Sopousek, J. ; Million, B., "Carbon profile analysis of Fe-Cr-C/Fe-Cr-

Ni-C diffusion joints" Kovove Materialy-(Metallic Materials), 2003.
41(2): p. 118-126.
123. Li, G.F. ; Charles, E.A. ; Congleton, J., "Effect of post weld heat
treatment on stress corrosion cracking of a low alloy steel to stainless
steel transition weld" Corrosion Science, 2001. 43(10): p. 1963-1983.

References
124. Diez, F.M. ; Liu, S., "Compositional boundary between primary

austenitic and ferritic Mn-Cr and Ni-Cr steel weld metals" Materials
Microstructure and Processing, 2007. 452: p. 1-7.
125. Gómez, X. ; Echeberría, J., "Microstructure and mechanical properties

of low alloy steel T11-austenitic stainless steel 347H bimetallic tubes"
Materials Science and Technology, 2000. 16: p. 187-193.
126. Barnhouse, E.J. ; Lippold, J.C., "Microstructure/property relationships

in dissimilar welds between duplex stainless steels and carbon steels"
Welding Journal, 1998. 77(12): p. 477S-487S.
127. Pavlovsky, J. ; Million, B. ; Ciha, K. ; Stransky, K., "Carbon

redistribution between an austenitic cladding and a ferritic steel for
pressure-vessels of a nuclear-reactor" Materials Science and
Processing, 1991. 149(1): p. 105-110.
128. Huang, M.L. ; Wang, L., "Carbon migration in 5Cr-0.5Mo/21Cr-12Ni

dissimilar metal welds" Metallurgical and Materials Transactions A-
Physical Metallurgy and Materials Science, 1998. 29(12): p. 3037-
3046.
129. Albert, S.K. ; Gill, T.P.S. ; Tyagi, A.K. ; Mannan, S.L. ; Kulkarni, S.D.
; Rodriguez, P., "Soft zone formation in dissimilar welds between two
Cr-Mo steels" Welding Journal, 1997. 76(3): p. S135-S142.
130. Race, J.M. ; Bhadeshia, H.K.D.H., "Precipitation sequences during

carburization of Cr-Mo steel" Materials Science and Technology,
1992. 8(10): p. 875-882.
131. Sudha, C. ; Paul, V.T. ; Terrance, A.L.E. ; Saroja, S. ; Vijayalakshmi,

M., "Microstructure and microchemistry of hard zone in dissimilar
weldments of Cr-Mo steels" Welding Journal, 2006. 85(4): p. 71S-80S.
132. Kozeschnik, E. ; Polt, P. ; Brett, S. ; Buchmayr, B., "Dissimilar

2.25Cr/9Cr and 2Cr/0.5CrMoV steel welds - Part 1: Characterisation
of weld zone and numerical simulation" Science and Technology of
Welding and Joining, 2002. 7(2): p. 65-68.
133. Kozeschnik, E. ; Warbichler, P. ; Letofsky-Papst, I. ; Brett, S. ;

Buchmayr, B., "Dissimilar 2.25Cr/9Cr and 2Cr/0.5CrMoV steel welds
- Part 2: Identification of precipitates" Science and Technology of
Welding and Joining, 2002. 7(2): p. 69-76.
134. Laha, K. ; Latha, S. ; Rao, K.B.S. ; Mannan, S.L. ; Sastry, D.H.,

"Comparison of creep behaviour of 2.25Cr-1Mo/9Cr-1Mo dissimilar
weld joint with its base and weld metals" Materials Science and
Technology, 2001. 17(10): p. 1265-1272.

References
135. Sudha, C. ; Terrance, A.L.E. ; Albert, S.K. ; Vijayalakshmi, M.,

"Systematic study of formation of soft and hard zones in the dissimilar
weldments of Cr-Mo steels" Journal of Nuclear Materials, 2002. 302(2-
3): p. 193-205.
136. Kozeschnik, E. ; Polt, P. ; Warbichler, P. ; Letofsky-Papst, I. ; Brett, S.

; Buchmayr, B., "The microstructure of dissimilar chromium steel
welds during PWHT and inservice exposure", in Mathematical
Modelling of Weld Phenomena 6, H. Cerjak, Editor. 2002, Maney
Publishing: Leeds. p. 323-336.
137. Moorthy, V. ; Vaidyanathan, S. ; Laha, K. ; Jayakumar, T. ; Bhanu

Sankara Rao, K. ; Raj, B., "Evaluation of microstructures in 2.25Cr-
1Mo and 9Cr-1Mo steel weldments using magnetic Barkhausen noise"
138. Luo, G.M. ; Wu, J.S. ; Meng, Q.S., "Microstructural evolution on the
T91 dissimilar metal joints during creep rupture tests" Journal of
Materials Science & Technology, 2004. 20(4): p. 383-386.
139. Foret, R. ; Zlamal, B. ; Sopousek, J., "Structural stability of dissimilar

weld between two Cr-Mo-V steels - Results are presented of an
experimental study and modeling of the structural stability of 6Cr-Mo-
V 8-3-2 and X12Cr-Mo-Nb 10-1 steels in a temperature interval from
600 to 900C" Welding Journal, 2006. 85(10): p. 211S-217S.
140. Nicholson, R.D., "Effect of post weld heat teatment on development of

interfacial structures in nickel-based transition joints" Materials
Science and Technology, 1985. 1(3): p. 227-233.
141. Naffakh, H. ; Shamanian, M. ; Ashrafizadeh, F., "Influence of artificial

aging on microstructure and mechanical properties of dissimilar welds
between 310 stainless steel and INCONEL 657" Metallurgical and
Materials Transactions A-Physical Metallurgy and Materials Science,
2008. 39A(10): p. 2403-2415.
142. Lippold, J.C. ; Sowards, J.W. ; Murray, G.M. ; Alexandrov, B.T. ;

Ramirez, A.J., "Weld solidification cracking in solid-solution
strengthened Ni-base filler metals", in Hot Cracking Phenomena in
Welds II, T. Bollinghaus, et al., Editors. 2008, Springer-Verlag Berlin:
Berlin. p. 147-170.
143. Bhaduri, A.K. ; Seetharaman, V. ; Venkadesan, S., "Effect of aging on

the interfacial microstructure and mechanical properties of a alloy
800/2.25Cr1Mo steel joint" Zeitschrift fur Metallkunde, 1989. 80(9):
p. 630-634.
144. Gómez, X. ; Echeberria, J., "Microstructure and mechanical properties

of carbon steel A210-superalloy Sanicro 28 bimetallic tubes" Materials

References

145. Sireesha, M. ; Albert, S.K. ; Sundaresan, S., "Metallurgical changes

and mechanical behaviour during high temperature aging of welds
between Alloy 800 and 316LN austenitic stainless steel" Materials
Science and Technology, 2003. 19(10): p. 1411-1417.
146. Sireesha, M. ; Albert, S.K. ; Sundaresan, S., "Influence of high-

temperature and mechanical properties of exposure on the
microstructure dissimilar metal welds between modified 9Cr-1Mo steel
and alloy 800" Metallurgical and Materials Transactions a-Physical
Metallurgy and Materials Science, 2005. 36A(6): p. 1495-1506.
147. Naffakh, H. ; Shamanian, M. ; Ashrafizadeh, F., "Microstructural

evolutions in dissimilar welds between AISI 310 austenitic stainless
steel and Inconel 657" Journal of Materials Science, 2010. 45(10): p.
2564-2573.
148. Sireesha, M. ; Albert, S.K. ; Sundaresan, S., "Thermal cycling of

transition joints between modified 9Cr-lMo steel and Alloy 800 for
steam generator application" International Journal of Pressure Vessels
and Piping, 2002. 79(12): p. 819-827.
149. Ayer, R. ; Mueller, R.R. ; Leta, D.P. ; Sisak, W.J., "Phase

transformations at steel-IN625 clad interfaces" Metallurgical
20(4): p. 665-681.
150. DuPont, J.N., "Solidification of an Alloy 625 weld overlay"

Metallurgical and Materials Transactions A-Physical Metallurgy and
151. Schmid-Fetzer, R. ; Andersson, D. ; Chevalier, P.Y. ; Eleno, L. ;

Fabrichnaya, O. ; Kattner, U.R. ; Sundman, B. ; Wang, C. ; Watson, A.
; Zabdyr, L. ; Zinkevich, M., "Assessment techniques, database design
and software facilities for thermodynamics and diffusion" Calphad,
2007. 31(1): p. 38-52.
152. Engstrom, A. ; Hoglund, L. ; Agren, J., "Computer-simulation of

diffusion in multiphase systems" Metallurgical and Materials
Transactions a-Physical Metallurgy and Materials Science, 1994.
25(6): p. 1127-1134.
153. Saroja, S. ; Parameswaran, P. ; Vijayalakshmi, M. ; Raghunathan,

V.S., "Prediction of microstructural states in Cr-Mo steels using phase
evolution diagrams" Acta Metallurgica et Materialia, 1995. 43(8): p.
2985-3000.
154. Schneider, A. ; Inden, G., "Simulation of the kinetics of precipitation

reactions in ferritic steels" Acta Materialia, 2005. 53(2): p. 519-531.

References
155. Saunders, N., "Phase diagram calculations for Ni-based superalloys",

in Superalloys 1996, R.D. Kissinger, et al., Editors. 1996, The
Minerals, Metals & Materials Society(TMS): Warrendale, PA. p. 101-
110.
156. Saunders, N. ; Fahrmann, M. ; Small, C.J., "The application of

CALPHAD calculations to Ni-based superalloys", in Superalloys 2000,
T.M. Pollock, et al., Editors. 2000, Minerals, Metals & Materials Soc:
Warrendale. p. 803-811.
157. Saunders, N. ; Guo, Z. ; Li, X. ; Miodownik, A.P. ; Schille, J.P.,

"Modelling the material properties and behaviour of Ni-based
superalloys", in Superalloys 2004, K.A. Green, et al., Editors. 2004,
The Minerals, Metals and Materials Society (TMS): Warrendale, PA.
p. 849-858.
158. Warnken, N. ; Ma, D. ; Drevermann, A. ; Reed, R.C. ; Fries, S.G. ;

Steinbach, I., "Phase-field modelling of as-cast microstructure
evolution in nickel-based superalloys" Acta Materialia, 2009. 57(19):
p. 5862-5875.
159. Turchi, P.E.A. ; Kaufman, L. ; Liu, Z.K., "Modeling of Ni-Cr-Mo

based alloys: Part I-phase stability" Calphad, 2006. 30(1): p. 70-87.
160. Andersson, J.O. ; Helander, T. ; Hoglund, L.H. ; Shi, P.F. ; Sundman,

B., "THERMO-CALC & DICTRA, computational tools for materials
science" Calphad, 2002. 26(2): p. 273-312.
161. Butler, C.J.,"The solidification characteristics of titanium aluminides

",1995,Ph.D.,M3 School,University of Nottingham
162. Sundman, B. ; Jansson, B. ; Andersson, J.-O., "The Thermo-Calc

databank system" Calphad, 1985. 9(2): p. 153-190.
163. Borgenstam, A. ; Engstrom, A. ; Hoglund, L. ; Agren, J., "DICTRA, a

tool for simulation of diffusional transformations in alloys" Journal of
Phase Equilibria, 2000. 21(3): p. 269-280.
164. Miller, M.K. ; Babu, S.S. ; Burke, M.G., "Comparison of the phase
compositions in Alloy 718 measured by atom probe tomography and
predicted by thermodynamic calculations" Materials Science and
Processing, 2002. 327(1): p. 84-88.
165. Tancret, F., "Thermo-Calc and Dictra simulation of constitutional

liquation of gamma prime (gamma') during welding of Ni base
superalloys" Computational Materials Science, 2007. 41(1): p. 13-19.
166. Million, B. ; Bacilek, K. ; Kucera, J. ; Michalicka, P. ; Rek, A. ;

Stransky, K., "Carbon diffusion and thermodynamic characteristics in
chromium steels" Zeitschrift fur Metallkunde, 1995. 86(10): p. 706-
712.

References
167. Million, B. ; Kucera, J. ; Michalicka, P., "The influence of silicon on

carbon redistribution in steel weldments" Materials Science and
Processing, 1995. 190(1-2): p. 247-252.
168. Wada, H., "Thermodynamic properties of carbides in 2.25Cr1Mo steel

at 985K" Metallurgical Transactions A-Physical Metallurgy and
169. Robson, J.D. ; Bhadeshia, H.K.D.H., "Modelling precipitation

sequences in power plant steels .2. Application of kinetic theory"
Materials Science and Technology, 1997. 13(8): p. 640-644.
170. Fujita, N. ; Bhadeshia, H.K.D.H., "Modelling simultaneous alloy

carbide sequence in power plant steels" The Iron and Steel Institute of
Japan International, 2002. 42(7): p. 760-769.
171. Helander, T. ; Agren, J. ; Nilsson, J.O., "An experimental and

theoretical investigation of diffusion across a joint of two
multicomponent steels" The Iron and Steel Institute of Japan
International, 1997. 37(11): p. 1139-1145.
172. Engstrom, A. ; Hoglund, L. ; Agren, J., "Computer-simulation of

carburization in multiphase systems", in Proceedings of the Second
Asm Heat Treatment and Surface Engineering Conference in Europe,
Pts 1 and 2, E.J. Mittemeijer, Editor. 1994, Transtec Publications Ltd:
Zurich-Uetikon. p. 725-730.
173. Reed, S.J.B., "Electron microprobe analysis". 1997: Cambridge

University Press.
174. McCartney, D.G.,"Studies on cellular and dendritic

growth",1982,DPhill,University of Oxford
175. Porter, D.A. ; Easterling, K.E., "Phase transformations in metals and

alloys". 1992: Nelson Thornes.
176. Guttmann, V., "Phase stability in high temperature alloys". 1981:

Applied Science Publishers.
177. Dupin, N. ; Sundman, B., "A thermodynamic database for Ni-base

superalloys" Scandinavian Journal of Metallurgy, 2001. 30(3): p. 184-
192.
178. Saunders, N. ; Guo, Z. ; Miodownik, A.P. ; Schille, J.P., "Modelling

the material properties and behaviour of Ni- and NiFe-based
superalloys", in Superalloys 718, 625, 706 and Derivatives,
Proceedings. 2005, Minerals, Metals & Materials Soc: Warrendale. p.
571-580.
179. Bueno, L.D., "Effect of oxidation on creep data: Part 1-Comparison

between some constant load creep results in air and vacuum on

References
21/4Cr-1Mo steel from 600 to 700 C" Materials at High Temperatures,

2008. 25(3): p. 213-221.
180. Natesan, K. ; Chopra, O.K. ; Kassner, T.F., "Compatibility of Fe-2 1/4

wt%Cr-1 wt%Mo steel in a sodium environment" Nuclear Technology,
1976. 28(3): p. 441-451.
181. Robson, J.D. ; Bhadeshia, H.K.D.H., "Kinetics of precipitation

reactions in ferritic power plant steels", in Microstructural Stability of
Creep Resistant Alloys for High Temperature Plant Applications, A.
Strang, J. Cawley, and G.W. Greenwood, Editors. 1998, Inst Materials:
London. p. 395-404.
182. Laha, K. ; Chandravathi, K.S. ; Rao, K.B.S. ; Mannan, S.L. ; Sastry,

D.H., "Prediction of creep deformation and rupture behaviour of
2.25Cr-1Mo weld joint" International Journal of Pressure Vessels and
Piping, 2000. 77(12): p. 761-769.
183. Chopra, O.K. ; Natesan, K. ; Kassner, T.F., "Carbon and nitrogen

transfer in Fe-9Cr-Mo ferritic steels exposed to a sodium
environment" Journal of Nuclear Materials, 1981. 96(3): p. 269-284.
184. Kucera, J. ; Million, B. ; stransky, K., "Stationary and quasistationary

models of carbon redistribution in austenitic steel weldments. I.
Ternary systems" Czechoslovak Journal of Physics B, 1985. 35: p.
1355-1361.
185. Sudha, C. ; Anand, R. ; Saroja, S. ; Vijayalakshmi, M., "Evaluation of

concentration dependant diffusion coefficients of carbon in a
dissimilar joint of ferritic steels" Transactions of the Indian Institute of
Metals, 2010. 63(4): p. 739-744.
186. Robson, J.D. ; Bhadeshia, H.K.D.H., "Kinetics of precipitation in

power plant steels" Calphad, 1996. 20(4): p. 447-460.
187. Robson, J.D. ; Bhadeshia, H.K.D.H., "Modelling precipitation

sequences in power plant steels .1. Kinetic theory" Materials Science
and Technology, 1997. 13(8): p. 631-639.
188. Bhaduri, A.K. ; Venkadesan, S. ; Rodriguez, P. ; Mukunda, P.G.,

"Combined effects of post-weld heat-treatment and aging on alloy
800/2.25Cr1Mo steel joint" Materials Science and Technology, 1991.
7(11): p. 1051-1056.
189. Bhaduri, A.K. ; Ray, S.K. ; Rodriguez, P., "New toughness parameters
from tensile test for optimizing post weld heat treatment of alloy
800/2.25Cr1Mo steel joint" Materials Science and Technology, 1993.
9(12): p. 1133-1136.
190. Ojo, O.A. ; Richards, N.L. ; Chaturvedi, M.C., "Microstructural study

of weld fusion zone of TIG welded IN 738LC nickel-based superalloy"
Scripta Materialia, 2004. 51(7): p. 683-688.

References
191. Gurumoorthy, K. ; Kamaraj, M. ; Rao, K.P. ; Venugopal, S.,

"Microstructure and wear characteristics of nickel based hardfacing
alloys deposited by plasma transferred arc welding" Materials Science
and Technology, 2006. 22(8): p. 975-980.
192. Thompson, R.G., "Microfissuring of alloy-718 in the weld heat-

affected-zone" Journal of Metals, 1988. 40(7): p. 44-48.
193. Tawancy, H.M., "Thermal-stability of an Ni-Cr-Mo-Nb alloy"

Materials Characterization, 1992. 28(3): p. 221-240.
194. Thomas, C. ; Tait, P., "The performance of alloy 625 in long-term

intermediate temperature applications" International Journal of
Pressure Vessels and Piping, 1994. 59(1-3): p. 41-49.
195. Tawancy, H.M., "Precipitation of NiMo in a Ni-Mo Base alloy"

Journal of Materials Science, 1980. 15(10): p. 2597-2604.

Appendixes
9- Appendix No.1: Optical emission spectroscopy

Appendixes

Appendixes

Appendixes
Appendix No.2: Mean atomic number calculations

1) Steel
2) Weld overlay

Appendixes
Appendix No.3: Database description forms

Appendixes

Appendixes

Nickel-Based Alloy Overlay On A 2.25Cr1Mo PDF

Uploaded by

Copyright:

Available Formats

Nickel-Based Alloy Overlay On A 2.25Cr1Mo PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Nickel-Based Alloy Overlay On A 2.25Cr1Mo PDF

Uploaded by

Copyright:

Available Formats

Saghafifar, Hassan (2011) Microstructural stability of a

nickel-based alloy overlay on a 2.25Cr1Mo steel

Access from the University of Nottingham repository:

Copyright and reuse:

For more information, please contact [email protected]

overlay on a 2.25Cr1Mo steel substrate

Hassan Saghafifar, M.Sc.

Thesis submitted for the degree of

Department of Mechanical, Materials, and Manufacturing

Ni-based superalloy weld overlays are widely used in electricity generating

Hardness survey results indicate significant changes in the interfacial region

Microstructural evolution in the steel region is related to the carbide

Experimental observations confirmed that a network of alloyed M23C6 carbide

Microstructural stability of a nickel-based alloy on a 2.25Cr1Mo steel substrate...i

2-1- Introduction ........................................................................................................... 5

2-2- General Background ............................................................................................... 6

2-2-1- Energy challenges and WtE plants .......................................................................... 6

2-2-2- Materials requirements for boiler in WtE plants .................................................... 7

2-2-4- Alloy 625 weld overlay and welding processes .................................................... 10

2-3- 2.25Cr1Mo steel ................................................................................................... 12

2-3-1- Physical metallurgy of 2.25Cr1Mo steel ............................................................... 13

2-3-2- Carbide transformation in 2.25Cr1Mo microstructure at high temperature ....... 14

2-3-3- Effects of welding on the 2.25Cr1Mo microstructure .......................................... 17

2-4- Alloy 625, a nickel based super alloy .................................................................... 19

2-4-1-1- Solidification of Alloy 625 ............................................................................ 20

2-4-1-2- Solidification of weld deposit of Alloy 625 .................................................. 23

2-4-1-4- Effect of dilution .......................................................................................... 25

2-4-2-1- Precipitation Hardening Phases .................................................................. 27

Microstructural stability of a nickel-based alloy on a 2.25Cr1Mo steel substrate...ii

2-4-2-2- Carbides and carbo-nitrides ........................................................................ 29

2-4-2-3- Topologically Closed Packed (TCP) phases .................................................. 32

2-4-3- TTT diagrams for wrought Alloy 625 ..................................................................... 34

2-4-4- Precipitation behaviour of weld deposit .............................................................. 36

2-5- Cladding and dissimilar metal welding ................................................................. 37

2-5-1- Interface and its different regions ........................................................................ 38

2-5-2- Interface and Diffusion ......................................................................................... 40

2-6- Microstructural evolution in dissimilar weld joints ............................................... 41

2-6-2- Steel/nickel alloys and formation of carbide arrays ............................................. 42

2-7- Evaluation of microstructural evolution using thermodynamic calculation .......... 46

2-7-1- Thermo-Calc software .......................................................................................... 47

2-7-2- Modelling of microstructure evolution in Ni-based alloys ................................... 48

2-7-3- Prediction of microstructural states in Cr-Mo steels ............................................ 49

2-7-4- Thermodynamic aspects of carbon migration in dissimilar weld joints ............... 50

2-8- Summary .............................................................................................................. 50

3- Experimental procedure ................................................................... 52

3-1- Introduction ......................................................................................................... 52

3-2- Materials .............................................................................................................. 53

3-4- Sample Preparation .............................................................................................. 58

Microstructural stability of a nickel-based alloy on a 2.25Cr1Mo steel substrate...iii

3-4-1- Composition analysis using spark emission spectroscopy .................................... 58

3-4-2- Sample preparation for optical and SE microscopy .............................................. 59

3-4-3- FIB-TEM sample preparation ................................................................................ 60

3-5- Principles of characterization methods ................................................................. 63

3-5-1- Optical emission spectroscopy ............................................................................. 63

3-5-2- Optical Microscopy ............................................................................................... 63

3-5-3- Scanning Electron Microscopy (SEM) ................................................................... 64

3-5-4- Transmission Electron Microscopy (TEM)............................................................. 65

3-5-5- Ion beam imaging ................................................................................................. 66

3-5-6- Energy Dispersive X-ray Analysis (EDX) ................................................................. 66