Welding Underwater

7/21/2019 Welding Underwater

http://slidepdf.com/reader/full/welding-underwater-56dea864385e7 1/11

Journal of Materials Processing Technology 220 (2015) 76–86

Contents lists available at ScienceDirect

Journal of Materials Processing Technology

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

Microstructures and mechanical properties of dissimilar T91/347H

steel weldments

Rutash Mittal a,∗, Buta Singh Sidhu b

a Department of Mechanical Engineering, Malout Institute of Management& InformationTechnology, Malout, Dist. Sri Muktsar Sahib, 152107,

Punjab, Indiab Punjab TechnicalUniversity, Jalandhar-KapurthalaHighway, Kapurthala,Punjab, India

a r t i c l e i n f o

Article history:

Received 4 April 2014

Received in revised form 12 January 2015

Accepted 14 January 2015

Available online 23 January 2015

Keywords:

Dissimilar metal weldments

SMAW

GTAW

Mechanical characterization

Micro-hardness

Microstructures

a b s t r a c t

Shielded metal arc welding (SMAW), gas tungsten arc welding (GTAW) and their combination are

processed by using austenitic and nickel based diverse welding electrodes/filler wires to prepare the

dissimilar weldments. The comparative evaluation of an appropriate welding process and welding con-

sumable is based on microstructural features, micro-hardness variation, tensile testing and fracture

morphology. The martensitic morphology is found responsible for the higher micro-hardness of HAZ

of T91 side of all the weldments. Higher tensile strength (638.2 MPa) of GTAW, ERNiCr3 combination is

observed than other weldments. The fractography corroborates the highest ductility of GTAW, ERNiCr3

combination with an elongation of 28.33% and the ductility of all the combinations of weldments except

GTAW, ERNiCr3 is observed to be less than their base metals. Hence, it can be concluded that GTAW

process, using nickel based weld metal offered the better results for the dissimilar joint between T91and

347H.

© 2015 Elsevier B.V. All rights reserved.

Introduction

Dissimilar metal weldments are widely used in various products

in chemical,petrochemical,nuclear andpower industries (Hanand

Sun, 1994). Satyanarayana et al. (2005) have favoured the adoption

of dissimilar metal joints, as it provides feasible solutions for the

flexible design of the products by using each material efficiently.

In power generating industry, components working at elevated

temperature are made of stainless steels and those used at lower

temperature aremade of ferritic steels. Dissimilarmetal weldments

with cheaper steels in place of highly alloyed steels make a con-

siderable saving on cost. Ferritic steels are having well considered

mechanical properties, good thermal conductivity, good ductility

and austenitic steels bear’s good corrosion resistance, better creep

strength and high temperature stability of microstructure during

the service. Sun (1996) emphasized the adoption of this combina-

tion, based on both technical and economical reasons, along with

satisfactory service performance as well as considerable savings.

The dissimilar metal joints are inclined to frequent failures and

these failures are by and large credited to one or a greater amount

of the accompanying reasons reported by Joseph et al. (2005).

∗ Corresponding author. Tel.: +91 9876333349; fax: +91 1637264511.

E-mail addresses: [email protected] (R. Mittal), [email protected]

(B.S. Sidhu).

(a) Difference in mechanical properties across the weld joint anddifference in thecoefficient of thermal expansionof twometals and

resultingcreepat theinterface,(b) general alloyingissues ofthe two

distinctive base metals, such as brittle phase formation and dilu-

tion, (c)carbon migrationfromferriticsteelinto austenitic steel,(d)

preferential oxidation at the interface, (e) residual stresses present

in the weld joints and (f) service conditions and other factors.

Jones (1974) stressed on the selection of proper welding process

and filler material for dissimilar metal weldments as an essential

factor in view of difference in physical, chemical and mechanical

properties of the base metals involved. Weld ability and dissim-

ilar joint features without using PWHT between Inconel 657 and

310SS were studied by Naffakh et al. (2009). Characterization anal-

ysis identified Inconel A to be the best among the four filler metals.

Jang et al. (2008) have examined the mechanical property variation

within the weld metal of Inconel 82/182 in the dissimilar joining

between low alloy steel and 316 stainless steel welded by SMAW

and GTAW processes. Sireesha et al. (2002) have used SMAW and

GTAW processes and Satyanarayana et al. (2005) have tried fric-

tion welding, whereas Sun (1996) has attempted dissimilar metal

weldments with laser beam welding for mechanical characteriza-

tion. Sudha et al. (2000) have identified the formation of soft and

hard zones in the dissimilar weldment of 9Cr-1Mo and2.25Cr-1Mo

steels. Bhaduri et al. (1994) and Raman and Tyagi (1994) have dis-

cussedabout the morphologyand distribution of precipitates in the

different sections of the weldment and concluded their difference

http://dx.doi.org/10.1016/j.jmatprotec.2015.01.008

0924-0136/© 2015 Elsevier B.V. All rights reserved.

http://localhost/var/www/apps/conversion/tmp/scratch_4/dx.doi.org/10.1016/j.jmatprotec.2015.01.008

http://www.sciencedirect.com/science/journal/09240136

http://www.elsevier.com/locate/jmatprotec

mailto:[email protected]






http://crossmark.crossref.org/dialog/?doi=10.1016/j.jmatprotec.2015.01.008&domain=pdf

http://www.elsevier.com/locate/jmatprotec

http://www.sciencedirect.com/science/journal/09240136




R. Mittal, B.S. Sidhu / Journal of Materials Processing Technology 220 (2015) 76–86 77

in the heat affected zone (HAZ) than those of base metal (BM)

and weld metal (WM). The precipitates revealed in the HAZ are

either parallelepiped or rod shaped which are typical shapes of the

M7C3 or M23C6 types of carbides respectively. Sudha et al. (2006)

reported the precipitation sequence and micro-chemistry of main

secondary phases of M23C6, M6C in the hard and of M2C inthesoft

zone in the weldments of 9Cr-1Mo steel and 2.25Cr-1Mo ferritic

steel that was post weld heat treated. The formation of hard HAZ

by virtue of presence of martensite phases in the weldment is an

important phenomenon, which needed to be studied from every

aspect. Foroozmehret al. (2011) havequoted theformation of mar-

tensite and austenite phases in ball milled powders through XRD

peaks. Detailed literature reviewrelated to formation of martensite

and its different crystal structures with respect to varying carbon

content has been published by Sherby Oleg et al. (2008). White

William (1992) has conveyed the corrosion and mechanical prop-

erty disintegration of dissimilar weldments and discussed about

the sensitization in the HAZ leading to inter-granular attack, weld

decayand preferential corrosion of secondary phases in multiphase

weld metal. The strength of the dissimilar weldments is generally

inferior to its base metals. Sireesha et al. (2000a) and Arivazhagan

et al. (2011) have reported the in-service failures of DMWs from

weld metal and HAZ respectively. Mohandas et al. (1999) have dis-

cussed thehigherstrength andductility of GTAW welds apparently

due to equiaxed fusion zone and protective nature of shielding

gas than SMAW welds for AISI430 ferritic steels. The formation of

martensitic layer leads to microstructural and mechanical prop-

erty variation across the weld interface with HAZ and is partly

responsible for the premature failure of the dissimilar welds at ele-

vated temperature (Dupont and Kusko, 2007). Viswanathan et al.

(1982) have reported about the failure of austenitic weld metal

at elevated temperatures for DMWs of martensitic and austenitic

steel. Shah Hosseini et al. (2011) also observed the better perfor-

mance of Inconel82 (Ni based filler material)than 310SS (austenitic

filler material) based on mechanical characterization features. The

superior performance of Inconel filler materials in DMW of 316LN

and alloy 800 is also corroborated by Sireesha et al. (2000a). The

authors have demonstrated the superiority of Ni based welds byvirtue of absence of precipitates at elevated service temperature.

The transition joints between ferritic/austenitic steels suffer from a

mismatch in coefficient of thermal expansion, which causes reduc-

tion in weldment integrity, to overcome these issues Ni based

weldment have been suggested (Sireesha et al., 2000b). Bhaduri

et al. (1994) and Das et al. (2009) also reported the longer service

life of nickel based weld metals for steam generator applications.

Detailed review of microstructural evolution and high tempera-

ture failure of ferritic to austenitic dissimilar welds have been done

in detailed manner by Dupont (2012). Steep microstructural and

microstructural gradients, large variation of coefficient of expan-

sion, formation of carbides and preferred oxidation of ferritic steel

were reported to be the main reasons of failure.

However, no systematic work on characterization studies hasbeen conducted on the dissimilar joint of T91/347H by SMAW and

GTAW processes. As better results are being listed in the litera-

ture, regarding the nickel based filler metal, the work on dissimilar

metal weldments of T91 and 347H by using austenitic as well as

nickelbased filler material has been attempted. Foreach weldment

combination, detailed studies are conducted on microstructure

characterization, micro-hardness variation, tensile strength and

fracture morphology using X-ray diffraction (XRD), scanning elec-

tron microscopy (SEM)/energy dispersive spectroscopy (EDS).

Base materials andwelding processes

Two types of materials in the pipe form, ferritic boiler steel,

namely SA213T91 (T91) of dimensions 50.5mm (O.D.)×

5.59mm

(thickness) and austenitic steel AISI347H (347H) of dimensions

50.8mm (O.D.)×5.59mm (thickness) were used as base materi-

als. The pipes were machined to make V joint geometry having a

root gap as 2mm and the included angle of 75◦ for welding joint.

SA213T91 belongs to 9Cr-1Mo ferritic steel category and AISI347H

is from austenitic steel family. Then the test samples in the form

of flat strips by keeping the weld metal approximately in the mid-

dle were prepared from the dissimilar metal weldments. The test

specimens of the base metals of SA213T91 and AISI347H were

also additionally prepared of the same dimensions with the pur-

pose of comparison. The chemical composition of base metals and

welding electrodes/filler wires are presented in Table 1. In the

present investigation two welding electrodes, namely Rutox-B and

Rutox-Ast by SMAW process are used just to resemble the actual

weldments used in the boilers of G.N.D.T.P. (one of the power plant

in North India), Bathinda, Punjab, India. These first two combina-

tionsare experimented to simulate the actual industrialconditions.

The other two proposed combinations are used in accordance with

the suggestions of welding experts as well as literature. White

William, 1992, Wang et al.(2011) and Arivazhagan et al. (2011) and

other authors have quoted in favour of GTAW process to weld the

metals. Daset al.(2009) and Dupont (2012) along with other author

have concluded benefits of Ni based weld metal. Some authors

like Jang et al. (2008), Han and Sun (1994) and Sireesha et al.

(2002) have tried GTAW+ SMAW combination and shown better

output. These welding procedures along with welding electrodes

already used in the power plant of G.N.D.T.P. were selected along

with the proposed combinations to simulate the actual industrial

conditions. The purpose was improvement in the design of weld-

ment which is directly applicable and can be recommended with

comparison to the working industry. The present study is a part

of high temperature corrosion study of dissimilar weldments of

T91 and 347H, showing difference in the oxidation behaviour in

micro structurally varied regions. The weldments have not expe-

rienced any preheating or post weld heat treatment (PWHT), to

simulate the characterization of actual industrial weldments. The

details of welding processes, electrodes, filler wires and welding

specifications are listed in Table 2.

Characterization of weldments

The dissimilar metal weldment combinations are portrayed for

optical microscopy to focus the optical microstructure of different

zones at diverse magnifications by utilizing an optical microscope

of Leica DM4000M at IIT Ropar, India. The standard metallographic

procedure has been adopted for obtaining the microstructures of

base metals, HAZs, weld metals and their interfaces. The sam-

ple preparation comprises polishing on different grades of emery

papers starting from 80 to 2000 grade and finally cloth polishing

with alumina paste on a rotating disc. Iteratively different combi-

nations of etching agents have been tediously applied to see themicrostructures as the evolution of structures is difficult in dis-

similar metal joints. The microstructure of 347H and its HAZ is

examined by Marble’s reagent {CuSo4 (4g) + HCl (20 ml) + distilled

water (20ml)} and of the T91, its HAZ and weld metals are treated

with fry’s reagent comprising {CuCl2 (5 g)+ HCl (40 m l) +ethyl

alcohol (25 ml) +distilled water (30ml)}. The Microhardness of

different sections of the weldment is carried out using Vickers’s

Microhardness tester of Wilson Instrument (an Instron company,

Model No. 402MVD) make at the regular interval of 0.5 mm across

the width of the sample covering all the regions of the weldment.

The tensile testing of the weldments has been done on the H25K-

S make of Hounsfield company tensile testing machine of 25kN

capacity fitted with a digital extensometer at Material testing lab-

oratory of Metallurgical and Materials Engineering Department of



78 R. Mittal, B.S. Sidhu / Journal of Materials Processing Technology 220 (2015) 76–86

Table 1

Chemical composition (wt.%) of the base metals and welding electrodes/wires.

Base material/welding electrode/wire C Cr Mn Mo Ni Si V Nb Cb Fe

SA213T91 0.0964 8.76 0.478 1.03 – 0.37 0.26 0.09 – Balance

AISI347H 0.0681 20.06 2.03 0.27 9.39 0.62 – 0.44 Balance

Rutox-B 0.03 19.8 1.40 – 10 0.40 – – – Balance

Rutox-Ast 0.03 19 1.20 10 0.45 – – 0.5 Balance

ERNiCr3 0.035 20 3.0 – 72.6 0.30 – – – 3.0

Table 2

Specifications of welding process, combinations and welding electrodes/wires.

S. no. Welding process used Specification of welding

electrode/filler wire used

Applied v oltage a nd c urrent Polarity a nd t he t otal n umber

of passes

1 SMAW (shielded metal arc

welding)

Rutox-B (D&H secheron)

AWS/SFA-5.4:E308L-16

3.15 mm dia. Electrode

Root Run:

V =15V, I =100A

Subsequent passes:

V = 15V, I =130A

DCEN, three

2 SMAW (shielded metal arc

welding)

Rutox-Ast (D&H secheron)

AWS/SFA-5.4:E347-16

3.15 mm dia. Electrode

Root Run:

V = 15V, I =100A

Subsequent passes:

V = 15V, I =130A

DCEN, three

3 GTAW (gas tungsten arc

welding), Shielding gas= argon,

flowrate= 10L/min. Purging

gas= argon,flow rate= 5 L/min.

ERNiCr3

AWS/SFA-5.11:ENiCrFe3

2.5mm dia. Filler wire

Root Run:

V =15V, I = 70A

Subsequent passes:

V =15V, I =100A

DCEN, three

4 GTAW*+ SMAW, *{shielding

gas= argon,flow

rate =10 L/min. purging

gas= argon,flow rate= 5L/min}

ERNiCr3 , 2.5mm dia. Filler

wire+ Rutox-Ast (D&H

Secheron), 3.15mm dia.

Electrode

Root Run:

V =15V, I = 70A

Subsequent passes:

V =15V, I =130A

DCEN, three (Root runwith

GTAW, further with SMAW)

Fig. 1. Dimensions of the tensile specimen (ASTM E8M-04).

I.I.T. Roorkee, India. For each type of reading, three test specimens

have been evaluated and average tensile strength is obtained, to

ensure repeatability. The basic dimensions of the tensile testing

specimen have been taken in accordance with the ASTM E8 stan-

dards as outlined in Fig. 1. Scanning electron microscope (SEM) of

JEOL Japan make having model JSM-6610LV equipped with EDS of

Oxford instrument facility having model number 51-ADD0013 is

used to study the microscopic and fracture modes of the samples.

X-ray diffraction facility using Instrument of PAN Alytical Model

X’Pert PROMPD made in the Netherlands is also used to have XRD

patterns ofthe specimen.Insteadof a singlescan ofthe whole weld-

ment as done by Arivazhagan et al. (2012), particular XRD scans of

the specific section i.e. HAZ, WM, etc. are performedfor the specific

information about the phases.

Results and discussion

Visual observations andmetallographic studies

The welded samples of all the four types of the dissimilar weld-

ments are demonstrated in Fig. 2, indicating that all the weldments

have a properpenetrationof filler electrode/wires. Thewidthof the

spread of the weld metal in the case of GTAW, ERNiCr3 is lesser in

comparison to other weldments.

Fig. 2. Different combinations of the dissimilar metal weldments.

White William (1992) enlisted that the characteristics of the

microstructures through weld zones, its size and extent of heat

affected zones of weldments will depend on the types of metals

being joined, whether or notthey areheat treatable, andthe classes

of welding or joining processes. The optical microstructures of

weldment of SMAW, Rutox-B combination are shown in Fig. 3a–l.The XRD patterns of different sections of all the combination of

weldments are displayed in Fig. 4a–f.

The microstructure of T91 BM shown in Fig. 3a comprises of

tempered lath martensite. The prior austenite grain boundaries, as

well as lath boundaries are decorated with precipitates (Laha et al.,

1995). The micro mechanism responsible for the formation of mar-

tensite structure on HAZ of T91 depends on the peak temperature

attained andcooling rate of that specified region. Depending on the

temperature of HAZto be above AC3or above AC1(i.e.betweenAC1

and AC3), austenitic or a mixture of ferritic and austenitic structure

gets formed which on cooling gets transformed into the marten-

sitic structure (Kou, 2003). The microstructures of the HAZ of T91

are segregated into two zones, namely fine grained heat affected

zone (FGHAZ) and coarse grained heat affected zone (CGHAZ) as




Fig. 3. Optical microstructures at different sections of SMAW, Rutox-B weldmentsof T91 and 347H.

shown in Fig.3b–f.The CGHAZ(Fig.3d) hasbeen just adjacent tothe

weld metal having large grain size, where as FGHAZ (Fig. 3c), just

after theT91 BM with lessergrainsize,having welding temperature

range between AC1 and AC3. The T91 CGHAZ depicts a martensitic

structurein thepresenceof some delta ferrite indicatedin Fig.3d–f,

just similar to the observations of Bhaduri et al. (1994). Pilling andRidley (1982) have also reported the formation of delta ferrite in

2.25Cr-1Mo steel. Delta ferrite is a chromium rich phase of Fe–Cr

alloys and may have formed due to depletion of Cr from the alloy

matrix. Depending on thepeak temperature andits cooling charac-

teristics of the thermal cycle, the phase of delta ferrite of T91 steel

can be as high as 35%. The presence of delta ferrite in addition to

martensite structure of HAZof T91 can facilitate the depletion of Crthrough precipitation and ultimatelyaffect the oxidationresistance

Fig.4. X-raydiffraction patternsfor differentcombinations of weldmentsof differentsectionsalong the weldments. (a) HAZof 347H,SMAW, Rutox-B (b)HAZ ofT91, SMAW,

Rutox-B (c)HAZ of T91, SMAW, Rutox-Ast(d) HAZ of T91, GTAW+ SMAW, weldment (e)HAZ of T91, GTAW, ERNiCr3 (f) WM of GTAW, ERNiCr3 weldment.






Fig. 5. SEMmicrographsat differentsectionsof SMAW, Rutox-B weldments of T91 and 347H.

Table 3

Variation of elements along theHAZ of T91 and weld metal interface forSMAW, Rutox B weldment.

T91 HAZ WM (Rutox-B)

Elements 1 2 3 4 5 6 7

Cr 9.05 11.07 12.17 13.39 13.5 18.5 23.10

Mo 1.97 2.06 7.57 1.06 3.97 0.14 –

Fe 89.04 86.86 80.26 83.5 73.47 72.06 65.5

Ni – – – 1.97 9.01 9.25 11.74

Fig. 6. Optical microstructures at different sections of SMAW, Rutox-Astweldments of T91 and 347H.

Fig. 7. SEMmicrographsat differentsectionsof SMAW, Rutox-Ast weldments of T91 and 347H.

Table 4

Variation of elements along theHAZ of T91 and weld metal interface forSMAW, Rutox Ast weldment.

T91 HAZ WM (Rutox Ast)

Elements 1 2 3 4 5 6 7

Cr 9.25 12.03 12.5 16.06 11.15 18.48 16.48

Mo 3.08 2.45 1.10 5.00 – – –

Fe 87.67 85.5 86.03 78.6 8.5 73.06 76.34

Ni – – – – 3.85 8.46 7.88




Fig. 8. Optical microstructuresat differentsectionsof GTAW +SMAW,weldments of T91 and 347H.

Fig. 9. SEMmicrographsat differentsectionsof GTAW+ SMAW, weldmentsof T91 and 347H.

Table 5

Variationof elements along theHAZ of T91 and weld metal interface for GTAW +SMAW,weldment.

T91 HAZ WM

Elements 1 2 3 4 5 6 7

Cr 9.94 11.02 10.3 11.45 23.13 18.8 23.49

Mo 2.24 0.91 2.42 0.87 – – –

Fe 87.64 88.3 87.2 71.1 53.8 65.6 53.81

Ni – – – 13.06 23.06 15.5 13.63

Fig. 10. Optical microstructuresat differentsectionsof GTAW, ERNiCr3 weldments of T91 and 347H.




Fig. 11. SEMmicrographsat differentsectionsof GTAW ERNiCr3,weldments of T91 and 347H.

Table 6

Variation of elements along theHAZ of T91 and weld metal interface forGTAW, ERNiCr3 weldment.

T91 HAZ WM

Elements 1 2 3 4 5 6 7 8 9

Cr 9.82 10.8 12.59 14.51 15.71 11.94 17.46 18.95 16.25

Mo 1.84 2.38 2.49 1.21 1.78 – 1.12 1.39 1.72

Fe 85.3 85.4 82.4 77.92 78.15 10.29 6.25 7.2 5.91

Ni – 1.26 3.97 3.44 68.5 70.5 71.8 71.5

Fig. 12. Micro-hardness variation of combinations of all the weldment of (a) SMAW, Rutox-B (b) SMAW, Rutox-Ast (c) GTAW +SMAW, ERNiCr3+ Rutox Ast (d) GTAW,

ERNiCr3.




Fig. 13. Photograph of thetensile tested specimensof all thecombinations of weldment (a)SMAW, Rutox-B, (b) SMAW, Rutox-Ast, (c) GTAW+ SMAW, ERNiCr3 + Rutox Ast,

(d) GTAW, ERNiCr3.

The average value of micro-hardness of the weld metal is around

204 Hv and distribution of micro-hardness seems to be uniform

throughout the weld metal region. Similarly the micro-hardness

pattern in HAZ of 347H and 347H BM seems tobe uniform with an

average value to be around 175 Hv.

The micro-hardness profile of the weldment produced by the

combination of GTAW and SMAW is presented in Fig. 12c. An

increase in the micro-hardness value can be seen while mov-ing from weld metal to HAZ of T91 side. The sharp increase in

micro-hardness in the HAZ of the T91 has been correlated to

the martensitic morphology (Fig. 8b–d) having average micro-

hardness to be around 420 Hv. The morphology of HAZ of T91 side

seems to be needle like martensitic structure, depicted in the SEM

micrograph of Fig. 9b andc. The presence of carbides i.e. (Cr,Fe)7C3,

Cr7C3 along with Ni-Cr-Fe and martensite phases in the HAZ of

T91 validated in XRD analysis (Fig. 4d), can also observed to be the

cause of higher micro-hardness. Yajiang et al. (2002) and Tavares

et al.(2002) havealso observed thepresence of martensite phase in

weld metal of 9Cr-1Mo steel. The average value of micro-hardness

of the weld metal is 170 Hv having uniform distribution through-

out the weld metal area, but a gradual increase in its value being

observed in theHAZ andBM of the347H. This increase in themicro-

hardness of HAZ of 347H side may be attributed to the presence of

carbides, as indicated in the optical micrographs in Fig. 8g and h.The micro-hardness profile of the weldment of GTAW, ERNiCr3

given in Fig. 12d presentsa sharpincrease in itsvalue, while moving

from weld metal to HAZ of T91 side. The average micro-hardness

of the HAZ is around 440Hv. An increase in the micro-hardness,

while moving from weld metal towards T91 BM may be attributed

to the needle shaped morphology of martensite and delta ferrite

of HAZ in Fig. 10c and d and SEM micrograph in Fig. 11b and c.

(Cr,Fe)7C3, Cr7C3, Ni-Cr-Fe along with martensite phases observed

Table 7

Tensile strength, micro-hardness variation and fractography of different combinations of the weldment.

Sample/joint

specifications of

T91-347H type

Yield strength

(MPa)

Ultimate

tensile strength

(MPa)

Percentage

elongation

Fracture

location

Maximum

micro-hardness on

weldment (Vickers), Hv

Tensile fractography

SMAW, Rutox-B No yielding

–

480 15 Weld metal 421 Parabolic shape dimple formation,

small tearing ridge, quasi-cleavage

fracture

SMAW, Rutox-Ast 404.3 580 22.27 BM of T91 450 Layered tearing ridge, quasi cleavage

fracture

GTAW+ SMAW,

ERNiCr3 +Rutox-Ast

427.6 508 17.63 Weld metal 450 Ductile fracture with micro-void

coalescence, small sized dimple

formation

GTAW, ERNiCr3 376 638.20 28.33 BM of T91 476 Fine and uniform dimples, some large

dimples surrounded by small dimples,

ductile fracture

Base Metal T91 573.6 624 25.71 – – Layered tearing ridge, quasi cleavage

fracture

Base Metal 347H 340.7 628 54.24 – – Ductile Fracture, small shallow

dimples indicative of high ductility and

tensile strength




Fig. 14. SEMfractographyof thetensile tested specimensof base metals anddissimilar weldments, (a) T91 BM, (b) 347H BM, (c) SMAW, Rutox-B, (d) SMAW, Rutox-Ast, (e)

GTAW +SMAW, ERNiCr3 + Rutox-Ast, (f) GTAW, ERNiCr3.

in theparticular XRDpattern of Fig.4e can alsobe a factorforhigher

micro-hardness. The formation of various phases as identified by

the specific XRD analysis (Fig. 4f) might have contributed to the

observed increase in micro-hardness of the weld metal with an

average value to be around 214Hv. A gradual increase of micro-

hardness observed in the HAZ and 347H BM, might be by virtue

of re-crystallization of the austenitic structure of the 347H side

(Fig. 10 j).

Tensile testing and fracture analysis

Thetensile testing of allthe base metalsand jointsformulatedby

using different welding techniques and filler electrodes/wires have

been made and presented in Fig. 13. The ultimate tensile strength,

yield strength and percentage elongation along with features of

fractography has been presentedin tabular formin Table 7. The ten-

sile strength so obtained show that maximum tensile strength of

638.2MPa,possessed bythe specimen produced byGTAW,ERNiCr3

combination followed by 580MPa of SMAW, Rutox-Ast combina-

tion. The lower heat evolved in the GTAW, ERNiCr3 (I = 7 0 & 100 A )

weldment than other welded combinations have produced better

joint integrity. Kumar andShahi (2011) and Mohandas et al. (1999)

have recommended lower heat input process to attain good tensilestrength, ductility and joint integrity as higher heat input can pro-

duce coarsening in HAZ and weld metal. Han and Sun (1994) have

quoted that for dissimilar metal joints, it is common practice that

mechanical properties of the joints should not be worse than those

of the inferior base metal. The tensile properties (tensile strength

and elongation), micro-hardness of the joint depends on chemical

composition and microstructure of the weld metal. Formation of

fully austenitic vermicular shaped microstructure of weld metal of

GTAW, ERNiCr3 due to thepresence of higher percentage of Ni than

the weld metals of other weldments led to enhancement of ductil-

ityand tensile strength, which ensures its safe application in steam

generator circuits. The joints have broken from the T91BM in case

of GTAW, ERNiCr3 and SMAW, Rutox-Ast combination, whereas

weldments of SMAW, Rutox-B and GTAW+ SMAW combination,

have fractured from the weld metal. It has been observed that

the percentage elongation of all the dissimilar weldments except

GTAW, ERNiCr3 is less than the percentage elongation of their

respective base metals; hence the weld metals produced by differ-

ent combinations, except GTAW, ERNiCr3 are less ductile than the

base metals i.e. T91 and 347H. The greater ductility and strength of

GTAW, ERNiCr3 weldment as compared to other combinations of

weldments, can be attributed to the equi-axed and austenitic mor-

phology ofthe weld metalin thegas tungstenarc welds,alsoto inertgas shielding. Less inclusion content has been observed in welds

produced withGTAW process thanSMAW process, whichpromotes

better joint strength and integrity (Dehmolaei et al., 2008). The

general low ductility of the welds of all the combinations except

GTAW, ERNiCr3 compared to that of initial base metals may be by

virtueof thecastmicrostructure ofthe fusionzone(Mohandaset al.,

1999).

The fractured surfaces of the tensile tested specimens were

analyzed using scanning electron microscopy and presented in

Fig. 14a–f. Dimples of varying size and shape were observed in all

the fractured surfaces, which indicate the major fracturing mecha-

nism to be ductile. Blach et al.(2011) have reported theinitiation of

the formation of dimples at secondaryphase particles, which even-

tually resulted in different morphology of each studied dimplesfracture, according to its own particular microstructural character-

istics. Thefractureof weldmentof SMAW, Rutox-B fromweld metal

reveals a parabolic dimple formation with the river line pattern of

quasi-cleavage fracture along with a small area of tearing ridge for-

mation (Fig. 14c). The fracture of weldment of SMAW, Rutox-Ast is

from the T91 BM just close to hard HAZ of T91 side, which reveals a

mixed mode of cleavage fracture having layered facets of fracture

as can be seen in Fig. 14d. In the fractography of GTAW+ SMAW

combination, ductile fracture with micro-void coalesces with rela-

tively smaller sizeddimple formationis observed in Fig. 14e. Finally

the GTAW, ERNiCr3 weldment fracture is from T91 BM, which

comprises fine and uniform dimples in which larger dimples are

surrounded by the cluster of fine dimples having a ductile fracture

as demonstrated in Fig. 14f. The fractography of the base metal T91



http://refhub.elsevier.com/S0924-0136(15)00010-2/sbref0200










































































































































































































































































































































































































































































































































































































































































































































































































































































































Welding Underwater

Documents