The Effect of Small Changes in Flux Basicity on
the Acicular Ferrite Content and Mechanical
Properties of Submerged Arc Weld Metal of
Navy HY-100 Steel
Oxygen content, inclusion size and acicular ferrite formation are all
influenced by the basicity in flux and filler metal
BY A. G. FOX, M. W. EAKES A N D G. L. FRANKE
ABSTRACT. The effect of small differ-
ences in flux basicity on the microstruc-
ture and mechanical properties of sub-
merged arc weld metal of Navy HY-100
steel has been investigated. It was found
that an increase in the flux basicity index
from 2.5 to 3.0 (calculated by the formula
proposed by Tuliani, Boniszewski and
Eaton, 1969, Weld. Met. and Fab. 37 (8):
327-329) led to a decrease in the total
oxygen content of the weld metal from
0.034 to 0.027 wt-%. This decrease in
weld metal oxygen content led to in-
creasing average nonmetallic inclusion
size and an increased propensity for aci-
cular ferrite formation, which were found
to increase both the strength and tough-
ness of the weld metal. These results sug-
gest that careful control of the chemistry
of both the filler metal and flux is neces-
sary to produce consistently high
strength and toughness in weld metal
during the submerged arc welding of HY-
100 steel.
Introduction
The origin of acicular ferrite in the fu-
sion zone of submerged arc weldments
on high-strength steels is very complex
and depends upon the chemical compo-
sition of the steel base plate and filler
A.G. FOX and M.W EAKESare with the Cen-
ter for Materials Science and Engineering,
United States Naval Postgraduate School,
Monterey, Calif. G. L. FRANKE is with Code
615 Welding Branch, Annapolis Detachment,
Carderock Division, Naval Surface Warfare
Center, Annapolis, Md.
330-s I OCTOBER 1996
metal, the composition of the flux used
during welding and the cooling rate of
the weld metal during the transformation
of the undercooled metastable austenite.
The strength and toughness of weld metal
improves as the amount of acicular fer-
rite increases due to its fine "basket
weave"microstructure, and so it is im-
portant to understand the mechanism of
its formation so that the volume fraction
of acicular ferrite can be maximized. This
topic has been extensively studied. For
recent reviews see Refs. 1-4. However,
as yet, the welding conditions appropri-
ate to maximizing the amount of acicu-
lar ferrite in Navy HY-100 steel during
submerged arc welding have yet to be es-
tablished.
The chemical composition of the filler
metal principally determines the final
composition of the weld metal, although
the composition of the base plate is im-
portant because of dilution effects. Also,
there may be transfer of elements from
the flux (Refs. 5, 6). In high-strength steels
KEY WORDS
HY-100
Submerged Arc Welding
CCT Diagram
Weld Metal Toughness
High-Strength Steel
Flux Basicity
Filler Metal
Acicular Ferrite
the alloying elements such as carbon,
nickel, chromium, copper and niobium
are present to give the required strength
levels, and a fortuitous outcome of their
presence is a continuous cooling trans-
formation (CCT) diagram with features
that mean that bainite (or, more correctly,
fine mixtures of lath ferrite, lath marten-
site and both interlath and "blocky" re-
tained austenite, which are best de-
scribed as granular ferrite in low-carbon
steels since carbides characteristic of bai-
nite are usually absent) is the major trans-
formation product in the fusion zone dur-
ing submerged arc welding of these
steels, provided suitable heat inputs and
preheat/interpass temperatures are cho-
sen during multipass welding. Since aci-
cular ferrite is also nucleated at similar
weld metal cooling rates (Refs. 7, 8),
some workers effectively describe it as
intragranularly nucleated bainite (Refs.
1,9, 10). It would appear that controlling
the weld thermal cycle so that conditions
are favorable for acicular ferrite nucle-
ation is not a problem, although bainite
(granular ferrite) and martensite are usu-
ally also formed. This subject will be dis-
cussed again later.
Once a suitable weld metal harden-
ability and cooling rate have been estab-
lished, the amount of acicular ferrite nu-
cleated will depend upon oxygen
content and the size, number, distribu-
tion and chemical composition of the
nonmetallic inclusions present (Refs.
1-4, 6-63, 69). Appropriate inclusions
appear to be in the size range 0.2-2.0 pm
with a mean size of about 0.4 IJm being
an optimum value (Refs. 1,2, 34). These
 inclusions usually contain manganese,
silicon, aluminum and titanium as their
major constituents and do not appear to
be exactly spherical but have a "faceted" O w

or slightly angular appearance. Some

workers have suggested that manganese
sulfide-containing inclusions are effec-
tive in nucleating acicular ferrite. For ex-
ample, Yamamoto, et al. (Ref. 40), have
argued that MnS, which sometimes
grows on the surfaces of the titanium
oxide-containing inclusions, is responsi- • °

ble for the formation of acicular ferrite, 20 pm •

although Chijiiwa, et al. (Ref. 42), and
Abson (Ref. 36) claim that the presence Fig. 1 - - Backscattered scanning electron mi- Fig. 2 - - Optical micrograph o f sample
crograph of a typical inclusion field in sample F296 showing acicular ferrite microstruc-
of sulfides leads to a reduction in the F296. ture.
amount of acicular ferrite. When no sul-
fur is present, it has been surmized that
the inclusions are oxides since they are considering the kinetics
composed of strongly deoxidizing ele- of inclusion formation;
ments, although Barbaro, et al. (Ref. 41 ), they regard MnO.AI203 ~" 0.025
as likely to be a major A
suggested that the titanium present in
suitable inclusions can be present as TiN. constituent in the inclu- ~ 0.02-
This is certainly possible in weld metals sions responsible for F296
with lower oxygen activity (Ref. 69). Al- acicular ferrite forma- I •
though appropriately sized inclusions tion, together with ~ 0.015-
which contain very little titanium and • Ai
AI203, SiO2, Ti305 and
whose composition is dominated by TiO 2. Babu, et al., also ~ 0.01- • TI
manganese, aluminum and silicon are indicated that TiO can
apparently capable of nucleating acicu- occur for certain weld
lar ferrite, their effectiveness appears to metal compositions i 0.005- A
• • |
be markedly improved as the titanium thus agreeing with with
content of the weld metal is increased Abson (Ref. 36) and o
above about 45 ppm (Refs. 1-3, 54). Evans (Ref. 52). These 250 32 3;0 375
Using transmission electron microscopy experimental results
Weld metal oxygen (ppm)
(TEM) and convergent beam electron dif- and the calculations of
fraction, Dowling, et al. (Ref. 31), have Babu, et al. (Ref. 58),
suggested that the faceted inclusions re- strongly suggest that the 3
sponsible for acicular ferrite in alu- oxidation state of the ti- ~"
minum-killed, low-titanium steels have tanium in these inclu- 2.5-
the MnO.AI203 (galaxite) structure. sions can vary and, if
However, more recently, Gregg and so, will depend on the ~ 2-
Bhadesia (Refs. 59, 60) have suggested weld metal composi- 8
that, on its own, MnO.AI203 is ineffec- tion, particularly with • •
respect to the major de- ._ • • • M~
tive in nucleating acicular ferrite. The © 1.5-
oxidizing elements, m
compositions and phases present in in- • Si
clusions containing Mn, AI, Si and Ti Mn, Si, AI and Ti, and ~ l-
F293
have not yet been thoroughly character- the welding process "~
ized because they are multiphase, too and its conditions. The ~ 0.5- • • 0~0 •
small for SEM microchemical analysis mechanism by which "e
and often too thick for TEM diffraction these inclusions nucle- ~" 0
analysis. As a result the oxidation state of ate acicular ferrite is un- 250 37s
the titanium in them is often unknown clear although four
and titanium has been variously reported have been suggested. Weld metal oxygen content (ppm)
as different oxides. For example, Abson These are :1 ) simple nu-
(Ref. 36) and Evans (Ref. 52) claim TiO cleation of acicular fer- Fig. 3 - - A - - Weld metal Zr, AI, Ti, Si and; B - - Mn vs. weld metal
for the welds they studied. Lau, et al. rite on an inert sub- oxygen.
(Refs. 37, 44), assumed TiO 2 in their strate, 2) epitaxial
work and Kluken and Grong (Ref. 47) growth of the ferrite on
chemical effects of the precipitation of
chose Ti203. Fox and Brothers (Ref. 62) suitably oriented crystallites in the inclu-
inclusions on the hardenability of the sur-
suggested two-phase mixtures of sions (lattice matching), 3) nucleation
rounding steel.
MnO.TiO 2 and MnO-AI203 could be re- arising from the strain energy associated
If classical heterogeneous nucleation
sponsible for the nucleation of acicular with the interfacial region between the
theory is considered to apply to the nu-
ferrite. Babu, et al. (Ref. 58), have devel- inclusions and the steel matrix resulting cleation of acicular ferrite from austenite,
oped a model to predict the nature of in- from cooling (the inclusions and steel then the inclusions will be less effective
clusions in submerged arc weld metal by matrix have significantly different coeffi- at nucleating acicular ferrite than grain
using equilibrium thermodynamics and cients of thermal expansion) and 4), boundary surfaces (Ref. 22). This is be-

W E L D I N G RESEARCH SUPPLEMENT I 331-s

Table 1 I Welding Control Parameters Table 2 - - Chemistry of Base, Filler and Weld Metals

Plate thickness 25.4 mm Filler Base

Number of passes 21 or 24 Metal Plate F289 F292 F293 F295 F296
Wire type L-TEC 120
(HT 120022) C 0.081 0.157 0.062 0.062 0.056 0.064 0.064
Wire diameter 2.38 mm Mn 1.57 0.33 1.45 1.49 1.28 1.51 1.54
Current 500 A, DCRP Si 0.40 0.30 0.38 0.46 0.42 0.28 0.34
Voltage 35 V P 0.004 0.003 0.007 0.005 0.004 0.015 0.008
Weld speed 0.8 mm s-1 S 0.006 0.005 0.010 0.006 0.006 0.011 0.007
Heat input 2.177 kJ mm-1 Ni 2.25 2.79 2.33 2.56 2.51 2.34 2.34
Preheat/Interpass 121-135°C Mo 0.42 0.36 0.46 0.52 0.47 0.47 0.49
temperature Cr 0.28 1.28 0.25 0.43 0.40 0.47 0.50
Cooling rate 10-13 Cs 1 at 538°C V 0.001 0.006 0.003 0.003 0.002 0.004 0.003
AI 0.012 0.018 0.013 0.020 0.011 0.011 0.014
Ti 0.014 0.004 0.006 0.008 0.004 0.005 0.006
cause the inclusions present a convex Zr 0.012 0.00 0.003 0.003 0.003 0.002 0.004
curved interface in the austenite so that a Cu 0.011 0.097 0.020 0.023 0.017 0.021 0.026
larger amount of interfacial area has to be O 0.003 0.0054 0.030 0.027 0.034 0.035 0.032
created between the nucleus and parent N 0.004 0.016 0.007 0.006 0.006 0.009 0.006
B 0.004 0.001 0.004 0.003 0.004 0.004 0.001
phase per unit volume of nucleus. Also, H 0.00028 0.00011 0.00002 0.00004 0.00004 0.00003 0.00008
the energy of the interface between the
ferrite and the inclusion is likely to be
larger than would be the case when the sions as well as to the austenite. Grong, Ti20 3 case, depletion of Mn in the steel
ferrite nucleates on austenite grain et al. (Ref. 69), have followed up this near the interface reaction zone seemed
boundaries. The concept of lattice work and have made TEM observations to be associated with the formation of
matching leading to epitaxial growth has which show that T-AI20 3, MnO-AI203 bainite. In addition, Gregg and Bhadesia
been considered by several workers and TiN inclusions can be responsible for (Refs. 59, 60) studied the possibility that
(Refs. 12, 1 3, 15, 38) and has led to the the nucleation of acicular ferrite pro- epitaxial growth (lattice matching)
suggestion that TiN, TiO and other vided they are appropriately oriented so and/or lattice strain contribute to bainite
phases showing a small disregistry with that the lattice disregistry between them nucleation. This work indicated that lat-
appropriately oriented ferrite will be ef- and ferrite is small. In addition, they sug- tice matching theory alone may not be
fective in nucleating acicular ferrite. Un- gest that these inclusions must also be able to predict whether a phase will be
fortunately, some experimental data ap- oriented so that they have an orientation an effective nucleant of bainite, and thus
pear to suggest that the ferrite/inclusion relationship w h i c h allows the subse- acicular ferrite, and that there is appar-
orientation relationship tends to be ran- quently nucleated acicular ferrite to be ently no simple correlation between bai-
dom (Ref. 31), with the orientation rela- oriented within the Bain region with re- nite nucleation potency of a steel/inclu-
tionship between the acicular ferrite and spect to the prior austenite. With these sion interface and the thermal expansion
the parent austenite always being of the ideas and a simple probability analysis, difference between the metal and the ce-
Kurdjumov-Sachs or Nishiyama-Wasser- Grong, et al. (Ref. 69), proceeded to cal- ramic phases.
man type. However, Kluken, et al. (Refs. culate that only 12% of the total number The preceding discussion indicates
53, 57), have indicated that 5-ferrite can of inclusions that could nucleate acicu- that the overall chemical composition of
nucleate from the molten steel epitaxially lar ferrite actually do so and found the the steel base plate and welding wire
on inclusions so that the subsequently volume of an acicular ferrite lath to be 2 with respect to oxygen, manganese, sili-
formed acicular ferrite can retain a crys- x 10 -7 mm ~, which is in reasonable con, aluminum and titanium is very im-
tallographic relationship to the inclu- agreement with measured values (Ref. 1). portant. Manganese is usually present in
Recently, Gregg and significant amounts in high-strength
Bhadesia (Refs. 59, 60) steels and their filler metals (about 1.0 to
[] have suggested some inter- 1.5 wt-%) because it is an important
I esting chemical reactions strengthener, toughener and hardenabil-
p~
F296 that could be responsible ity agent. Silicon is usually maintained at
0.75 - for creating conditions levels below 0.4 wt-% in high-strength
suitable for the nucleation steels because it is usually considered to
0 of acicular ferrite. For ex- be detrimental to fracture toughness, al-
ample, they have shown though this low level of silicon still al-
0.5- that the TiO 2 at a lows this element to take part in the de-
steel/TiO2-bonded inter- oxidation process as discussed
face annealed at 1473 K previously. High-strength steels and their
can release oxygen into
weld ing wires are usually deoxidized
0,25 - the steel, possibly causing
O with aluminum and calcium treated so
[] local decarburization and
[] O that sulfur contents are kept below 60
thus increased stability of
ppm in order to enhance resistance to
ferrite near the interface
lamellar tearing and hydrogen-induced
0 ! I I due to reduced steel hard-
enability. These workers cracking. Titanium may be present either
250 275 300 325 350 375
also found that TiO and as an impurity in steel welding wire or it
Weld metal oxygen content (ppm) Ti203, like TiO2, were ef- may be added deliberately. A review of
fective at nucleating bai- previous work suggests that a steel weld
Fig. 4 1 Inclusion volume fraction vs. weld metal oxygen. nite. The mechanism is not metal AI + Ti content of about 300 ppm
yet clear although, in the with an AI/Ti ratio of about 2.0 is ideal for

332-s I OCTOBER 1996

 40 40
F289 average size = 0.32 pm F292 average size = 0.378 p m

30- 30 -

tno i'o- 1
I I I I I I I I l I l I I -- illi |

0.3 0-~ 0.7 0.9 1.1 1.3 1.5 0.~ 0.5 0.7 0.9 1.1 1.3 1.5
Inclusion size (microns) Inclusion Size (microns)

40 40
F293 average size = 0.322 F295 average size = 0.323 pm

30 - 30 -

qJ 20 - ~ 20 -

10- lo-

0 , ~--~' T 0 , , , , '7 , ,
0 0.3 0.5 0.7 0.9 1.1 1.3 1.5 0.3 0.5 0.7 0.9 1.1 1.3 1.5

Inclusion size (microns) Inclusion size (microns)

the nucleation of large amounts of acic- oxygen content

ular ferrite, although it would seem that much above 300 40
the a l u m i n u m is not necessary since ppm apparently
large amounts of acicular ferrite can be generates many
•:.
%. F296 average size = 0.423 pm
found in weldments even if the weld more small oxide
metal aluminum content is low provided inclusions (<0.2 30-
sufficient titanium is present (Ref. 3). Fi- vm), w h i c h can
nally, it would seem that the oxygen ac- pin austenite grain
tivity of the submergerd arc w e l d i n g boundaries and

liinoooo
(SAW) flux should be such that the final reduce prior
weld metal total oxygen content is austenite grain
around 200 to 300 ppm if large amounts size so that grain
of acicular ferrite are to be nucleated boundary-nucle-
10-
(Refs. 28, 63). This, of course, can be cor- ated microcon-
related with the comments concerning stituents form
the size, distribution and composition of rather than acicu-
the oxide inclusions made previously. lar ferrite (Refs. 1, 0 , I ! I I
.
I
. .
I
. .
I
Lower levels of total oxygen (< 200 ppm) 2, 32). Total oxy- 0 0.3 0.5 0.7 0.9 1.1 1.3 1.5
mean that insufficient inclusion numbers gen contents
are generated for the formation of acicu- around 250 to 300 Inclusion size (microns)
lar ferrite so that other microconstituents ppm are routinely
such as bainite or Widmanst~tten ferrite achieved in sub-
are formed preferentially. Increasing total merged arc weld- Fi,~. 5 - - Inclusion size distributions for all samples.

W E L D I N G RESEARCH SUPPLEMENT [ 333-s

 steel could be developed. If successful
Table 3 - - Flux Composition (wt-%), Basicity Index and Acicular Ferrite (AF) Content of each
this would allow the U.S. Navy to con-
of the Five Weld Metals
sistently produce submerged arc weld-
F289 F292 F293 F295 F296 ments of HY-100 steel with optimum me-
chanical properties.
AI203 18.42 18.57 14.00 16.95 15.62
Fe203 1.10 1.37 1.28 2.01 1.07
K20 1.00 0.69 1.32 0.39 1.36 Experimental Procedure
MgO 29.25 28.46 30.23 35.10 32.56
Na20 0.95 0.72 2.13 1.23 1.22 Five multipass submerged arc welds
CaO 10.29 12.58 18.44 8.13 9.99 designated F289, F292, F293, F295 and
CaF2 22.45 22.40 14.92 17.31 21.26 F296 were made with L-Tec 120S weld-
SiO2 14.87 13.63 17.63 15.64 14.71 ing wire and 25.4-mm-thick HY-100
TiO2 0.73 0.71 0.00 0.68 0.53 base plate using the conditions shown in
MnO 0.91 0.84 0.05 2.53 1.62
Table 1. A single V (45-deg angle) butt
ZrO2 0.03 0.03 0.00 0.03 0.06
Basicity Index 2.65 2.83 2.75 2.63 2.97 joint geometry supported from below by
AF content (vol.%) 43 65 41 27 78 an HY-100 backing bar was adopted with
a 12.7-mm root opening. Between 21
and 24 passes were necessary to com-
plete the weldment. Two or three passes
Table 4 - - Room Temperature Tensile Test Results for the Five Weld Metals were used in the first (root) layer and be-
tween six and seven layers with three to
Yield Strength Ultimate Tensile Reduction in four passes per layer were needed to fill
(MPa) Strength (MPa) Elongation (%) Area (%)
the groove without undercut at the top
F289 694 819 22.0 62.5 surface. The chemical compositions of
F292 752 857 19.5 63.0 the base plate, welding wire and result-
F293 687 790 22.5 69.5 ing weld metal were determined by
F295 705 823 21.0 63.5 Luvac laboratories using conventional
F296 825 873 16.5 39.5 analytical techniques as shown in Table
Mil. Spec. (min) 703 14 2. It should be noted that the analysis of
Mil. Spec. (max) 848
the weld metal was performed on pulled
tensile specimens (discussed below) so
Table 5 - - Charpy V-Notch Energy (CVNE) Data for the Five Weld Metals (The temperatures
that the mechanical properties data
of testing are indicated) could be directly correlated with weld
metal chemistry. Analysis of the metallic
Mil. Spec. elements and boron was performed by
F289 F292 F293 F295 F296 min. plasma emission, sulfur and carbon by
the combustion infrared method, oxygen
CVNE (J) 6.6 2.6 8.9 4.1 5.9 and nitrogen by inert gas fusion and hy-
at -151°C drogen by vacuum hot extraction. The
CVNE (J) 10.3 10.3 9.2 5.9 16.2
confidence limits reported by Luvac (in
at -118°C
CVNE (J) 28.5 23.3 14.2 21.0 34.5 wt-%) for each measurement were Ni
at -84°C 0.05, Cr, Mn 0.02, Si and Mo 0.01, AI
CVNE (J) 52.2 46.1 35.9 32.5 48.8 33.2 and P 0.002, C, S, V, Ti, Zr, Cu, O, N and
at -51 °C B 0.001 and H 0.00001. Different pro-
CVNE (J) 66.8 62.9 63.5 46.3 63.9 44.3 prietary fluxes were used for each weld-
at -18°C ment and these were chemically
CVNE (J) 77.3 67.4 76.3 55.9 63.4 analysed using x-ray fluorescence, in-
at 16°C ductively coupled plasma and optical
CVNE 0) 77.7 69.9 84.0 64.0 67.9
emission spectroscopy. This allowed a
at 49°C
CVNE (J) 77.3 70.1 90.6 67.5 70.4 good estimate of the chemical composi-
at 82°C tion of each flux. The major constituents
CVNE (J) 78.2 72.7 90.4 70.8 76.0 of these are shown in Table 3. It should
at 116°C be noted that the flux associated with
sample F293 contains small amounts of
cerium and lanthanum oxides. It is not
ing of high-strength steels by the use of a in steel weld metal (Ref. 54). clear why they are present but both are
flux with an appropriate composition The purpose of the present work was strong hydride-formers and so they could
and basicity index (Refs. 28 and 63). to carefully analyze the nonmetallic in- have been added deliberately in the hope
Consistent with these comments about clusions and microstructure in the fusion that La and Ce ions in the molten flux
grain size is the fact that boron promotes zone of five SAW multipass weldments would scavenge hydrogen. Samples for
the formation of acicular ferrite since B made with different proprietary fluxes on tensile and Charpy impact testing were
segregates to austenite grain boundaries, aluminum killed HY-100 steel using weld cut from the weldments to be represen-
lowers the boundary energy and thereby wire containing 120 ppm aluminum, 140 tative of multipass regions in the weld
favors acicular ferrite formation (Refs ppm titanium and 40 ppm boron. A spe- metal, i.e., machined from as near to the
1-3, 54). Indeed, weld metals containing cific aim was to elucidate the role of flux center of the weld metal as possible. The
about fifty ppm B and several hundred basicity on the nucleation of acicular fer- tensile samples were cut from all weld
ppm Ti have proved very effective at gen- rite in steel weld metal so that an appro- metal in the welding direction with
erating large amounts of acicular ferrite priate flux formulation for welding this gauge length 50.8 mm and diameter 12.8

334-s I O C T O B E R 1 9 9 6
 0.45
0.425 U

0
0.4-
o I O
i-

0.4- F296 .~.

.m

0.375 -
N
°~

= 0.35
.o
0.35 -

0.325
0 0 []
0.3 , , ,
0.3 I I I I 0 500 1000 15J00 2000 2500
250 275 300 325 350 375

Weld metal oxygen content (ppm) Weld oxygen/Inclusion volume fraction (ppm x I00)

Fi~. 6 - - Inclusion size vs. weld metal oxy,~en. The correlatk~n coet L Fi,~. 7 - - Inclusion size vs. normalized weld metal oxy,~ten (weld metal
ficient, t, tor the best linear fit is indicated (0.356) and it is c/ear that oxy,~en/inclusion vc~lume #action).
there is no linear conelation between these two variables, as discussed
in the text.

him and for each weldment two tensile Table 6 - - Charpy V-Notch Fracture Appearance Data for the Five Weld Metals, the
tests were performed using a Instron ten- Percentage Brittle Fracture (%BF) and the Temperatures of Testing
sile testing nlachine. Standard (10 mm 2
cross-section) Charpy samples were cut F289 F292 F293 F295 F296
so that the 2 - m m - d e e p 45-deg notch
w o u k i progress through the center of the %BF at 95 100 100 100 95
-151 °C
weld metal in the welding direction dur- %BF at 90 100 100 90 80
ing testing and three tests were per- _118oc
formed at each of the temperatures %BF at 67.5 87.5 83.3 63.3 60
shown in Table 5. The results of the room 84°C
temperature tensile testing (averaged for %BF at 44 46 54 53.3 30
the two tests for each sample) are shown -51 °C
in Table 4 and the Charpy impact testing %BF at 30 14 28 36.7 10
(averaged for the three tests at each tenl- -18°C
perature) in Table 5 together with the mil- %BF at 5 (1 12.5 10 3
16°C
itary specifications for the mechanical %BF at 1.67 0 1.67 1.67 2.5
properties of Navy HY-100 steel. The 49oC
fracture appearance data resulting from %8F at 0 0 0 0 0
the Charpy tests is given in Table 6 (per- 82°C
cent brittle fracture) and the fracture ap- %BF at 0 0 0 0 0
pearance transition temperatures (FATT) 116°C
(50% brittle fracture) and the ductile-to-
brittle transition temperatures (DBTT)
( m i d p o i n t between upper and l o w e r Table 7 J Charpy V-Notch Transition Temperature Data for the Five Weld Metals Calculated
from the Data in Tables 5 and 6
shelves) deduced from Tables 5 and 6 are
presented in Table 7.
F289 F292 F293 F295 F296
Cross-section samples of each weld-
ment were cut and polished down to one DBTT (C) -64 -65 -62 -42 -78
I]m diamond and areas corresponding to FATT (C) -54 -54 -49 -47 -73
the positions from which the mechanical
properties test pieces were machined
were identified. For subsequent micro-
scopical examination every attempt was Table 8 - - Average Inclusion Composition (atomic %) for each Sample Determined by Energy
made to examine those regions where Dispersive X-ray Analysis in the Transmission Electron Microscope
the Charpy notch was likely to have been
(i.e., in the center of the weld metal). F289 F292 F293 F295 F296
These areas were examined in the un-
Mn 38.40 25.05 36.95 46.48 38.42
etched condition in a Cambridge $200 Ti 8.41 12.16 5.68 5.42 7.96
s(anning ele{tron microscope (SEM) op- AI 28.69 50.87 25.15 20.29 29.77
erating in the backscattered imaging Si 23.08 9.77 30.54 26.87 21.9(1
Zr 1.41 2.18 1.69 0.94 1.95

W E L D I N G RESEARCH SUPPLEMENT I 335-s

 F296 [] []

2.9 2.9 - r = 0.86~

¢g
q~

2.8 - 2.8"

t~
2.7 - 2."/-

[] []
[]
2.6 . , , , 2.6
250 275 300 325 350 375 0 $00 1000 15100 2(;00 2500
Weld oxygen/Inclusion volume fraction (ppm x 100)
Weld metal oxygen content (ppm)

Fig. 8 - - Basicity index vs. weld metal oxygen. There is no linear cor- Fig. 9 - - Basicity Index vs. normalized weld metal oxygen.
relation between these two variables (r = 0.294), as discussed in the text.

mode at 20 kV in order to determine the
size distribution and volume fraction of
the nonmetallic inclusions. This was ac-
complished by counting and sizing the
inclusions in 100 random fields observed
at a magnification of 7040 X with a work-
ing distance of 9 mm. Use of the SEM in
this way allowed inclusions as small as
50 nm to be included in the count. A typ-
ical inclusion field is shown in Fig. 1. The
samples were then etched with 5% Nital
and optical micrographs were taken from
the same areas from which the inclusion
counts were taken - - Fig. 2. From these
the amount of acicular ferrite was deter-
mined by the linear intercept method ap-
plied to ten fields taken at a magnifica-
tion of 500X with ten random lines
analyzed per field. The acicular ferrite
TIME
Fig. 10 - - CCT diagram for AX90 and AX110 filler metals (Ref. 67).
content for each weld metal is given in
Table 3. The balance of the microstruc-
ture in each case was found to be pre-
dominantly bainite, which is consistent
with other microstructural observations
on high-strength steels containing acicu-
lar ferrite (Refs. 1, 43, 50). Attempts to
modernize this procedure using ad-
vanced image processing software
proved difficult since the boundaries be-
tween the microstructural constituents
were not well defined. After optical ex-
amination, the etched samples were
coated with about 20 nm of carbon in a
Fullam EFFA evaporator and then etched
again (with Nital) to obtain extraction
replicas from the same areas which had
been examined optically and by SEM;
these were placed on copper grids and
io
( SECONOS)
examined in a JEOL 100 CX transmission
electron microscope (TEM) equipped
with a Kevex Quantum light element x-
ray detector. This allowed a quantitative
chemical analysis of the elements with
atomic number greater than 11 in the in-
clusions and a qualitative appraisal of the
light elements present. For each sample,
forty inclusions over the whole size range
were chemically analyzed and very little
variation in the chemical compositions
within each set of forty was found.
Results and Discussion
From the flux compositions shown in
Table 3, the basicity indices (BI) of each
flux were calculated by putting the
weight percentages of each flux con-
stituent into the following formula (Ref.
64):
CaO + C a F 2 +MgO + N a 2 0 +
BI= K20+Li20+O'5(MnO+FeO)
Si02 +0"5( AI203 +Ti02 +Zr02) (1)
It should be noted that F293 contained
very little MnO compared with the other
fluxes and, as discussed by Palm (Ref. 5),
low flux MnO can lead to Mn losses in
steel weld metal. This effect does not
seem to depend on flux basicity. To avoid
such losses in Navy HY-100 steel about
1.0 wt-% MnO is needed in the flux for-
mulation. However, it should be noted
that excessive additions of MnO to the
flux can lead to increased levels of man-
ganese in HY-100 weld metal which may
be undesirable (Ref. 68).
The overall chemical analyses of the

336-s I OCTOBER 1996

base plate, welding wire and weld metal
are shown in Table 2. It is important to
note that the welding wire contains 120
ppm AI, 140 ppm Ti and 4 ppm B, as well
as larger amounts of Si and Mn and thus
has the potential to generate significant
amounts of acicular ferrite provided the
flux basicity, heat input and preheat/in-
terpass temperatures are chosen cor-
rectly. It is clear that the weld metal as-
sociated with flux F293 (low in MnO) has
a significantly lower Mn content com-
pared with the other weld metals, as ex-
pected, otherwise these have composi-
tions consistent with the welding wire
composition after dilution to a certain ex-
tent by the base plate. Since deoxidation
is responsible for inclusions and hence
the potential nucleation of intragranular
acicular ferrite, it is important to examine
the behavior of the strong "trace" deoxi-
dants Zr, AI, Ti, Si and Mn as a function
of total oxygen content. These trends are
acteristics which are unpredictable, to
say the least, in submerged arc weld
metal because of the turbulence of the
weld pool, the rapid solidification asso-
ciated with welding and the effects of
multiple passes. It is also possible that
the inclusions are not distributed uni-
formly and that many more fields need
to be studied, although each experiment
was performed twice (thrice for F296) in
different weld metal regions. Further-
more, because equilibrium is not Fig. 11 - - Transmission electron micrograph of
achieved during welding the partition- an extraction replica taken from the weld metal
ing of the oxygen between that dissolved of sample F296 showing acicular ferrite nucle-
in the metal and that present as inclu- ated by an inclusion.
sions may differ
for different welds
produced using
the same welding ~'
conditions. ~ 50. 0
As well as in- "" F296 •
clusion volume
fraction, inclu-
~'.
,10- I O AI

O Ti
plotted in Fig. 3A and B. Since oxygen sion size is impor-
activities are very high in SAW weld tant and, accord- C 30- [] O • O Zr
metal, manganese also takes part in the ing to Fleck, et al. • • O A
• Q • Si
deoxidation process. However, the over- (Ref. 32), the in- ~ 20-
all Mn content of the weld metal seems clusion size in .~ • Mn
to be little affected by oxygen content as weld metals "~
n lO-
shown in Fig. 3B, although sample F293 should decrease ..~ 0 0
O0
has a low Mn content due to the low as the oxygen " 0 0 0 Oo
'~ 0
MnO content of the flux. The Zr, AI, Ti content increases ~ As 375
and Si contents all appear to decrease in submerged arc
with increasing oxygen content (with the weld metal. The Weld metal oxygen (ppm)
exception of F296, which appears to inclusion size dis-
have AI, Ti and Zr contents somewhat tributions for all Fig. 12 - - Inclusion Mn, Si, AI, Ti and Zr vs. weld metal oxygen.
higher than its oxygen content of 320 the samples are
ppm would suggest) and this occurs since shown in Fig. 5
increasing the oxygen content leads to and the average
increased consumption of deoxidants in inclusion size is
the molten state to form oxides. Some of plotted vs. oxygen
these deoxidation products float up into content in Fig. 6.
the slag while others remain as non- Between 270 and
metallic inclusions. 300 ppm oxygen
The nature of the size, distribution the average inclu-
and chemical analysis of the nonmetallic sion size drops
inclusions in the weld metal proved very from 0.38 pm to
complex. For example, the inclusion vol- about 0.32 lam
ume fraction should be expected to in- and then stays J _ - ~ - ° ' ~ T" " " . \ w,,.J~,*~t,

crease with increasing oxygen content. fairly constant ex-

However, Fig. 4 shows that this is not the cept for sample
case. In fact there appears to be no clear F296 which has
relation between volume fraction of in- an oxygen con- ,-
clusions and the oxygen content, and the tent of 320 ppm
amount of inclusions in the weld metal and a high aver-
associated with flux F296 is very much age inclusion size
higher than the others. It should be of 0.42 lam. It
pointed out that the inclusion count for should be remem-
sample F296 was performed three times bered that F296
in the center of three different weld cross had a very high
sections and the variation in the inclu- inclusion volume
sion volume fraction between the three fraction and a Fig. 13 - - MnO.SiO2.AI203 ternary phase diagram adapted from Ref. 14
observations was found to be minimal. simple calcula- showing the different inclusion compositions obtained in steels killed
The reason for this erratic volume frac- tion based on the with varying amounts of Mn, AI and 5i. The open squares indicate the
tion behavior must almost certainly be inclusions being data of the present work.
associated with inclusion float-out char- spherical and as-

WELDING RESEARCH SUPPLEMENT I 337-s

 80 80
F296 ~ra

70- 70-
mm
[]
w
N
60-
o,m

50-
5o-

40-
40-
30-
[]
30-
I-1
20 ! !

0 500 1000 1500 2000 2500

20 I I I I
250 275 300 325 350 375
Weld oxygen/inclusion volume fraction (ppm x 100
Weld metal oxygen (ppm)

Fig. 14 - - A c i c u l a r ferrite content vs. w e l d metal oxy,~en. Fig. 15 - - A c i c u l a r ferrite content vs. n o r m a l i z e d w e l d m e t a l oxygen.

80 80

70

60
o/ 70-

6o-
r-i

t.. 50- • 50-

40- 40-
I

.2
0-
[]
30 X
ra
°
20 I I I I 20 , , ,

0.3 0.325 0.35 0.375 0.4 0.425 2.6 2.7 2.8 2.9 3

Average Inclusion Size (microns) Baslcity Index

Fig,. 16 - - A c i c u l a r ferrite content vs. inclusion size. Fig. 17 - - A c i c u l a r f~rrite content vs. basicity index.

suming F296 to have an inclusion vol-
ume fraction which is an average of the
other four samples (0.00187) predicts the
oxygen content of F296 to be 290 ppm.
This suggests that if the float-out rate in
F296 had been similar to the other sam-
ples then its oxygen content would have
been of this order. This is consistent with
the weld metal total AI, Zr, Ti and Si con-
tents for F296 discussed earlier. A simple
way of dealing with the effect of these er-
ratic float-out rates is to normalize the
weld metal oxygen content by dividing it
by inclusion volume fraction. The rea-
soning for this comes from a simple ex-
pression relating oxygen content (wt-%
O) and oxide inclusion volume fraction,
Vf, in the weldments which can be of the
form
wt-% O = AVf + (wt-% O)dissolve d (2)
where A should be a constant if the weld-
ing conditions for all the weldments stud-
ied are the same and the sulfur in the in-
clusions is ignored (Ref. 1). In the present
work the value of A for the weld metal as-
sociated with F296 is very different than
F289 through F295. However, we can
normalize with respect to A by ignoring
the dissolved oxygen, (wt-% O) dis-
solved, (this is usally small compared
with the amount present as inclusions)
and divide wt-% O by Vf. This is impor-
tant since acicular ferrite is nucleated by
inclusions and the weld metal acicular
ferrite content will increase with increas-
ing inclusion volume fraction and de-

338-s I O C T O B E R 1996
crease with decreasing average inclLision
size (and therefore increasing oxygen
content according to Fig. 6). As a result,
a plot of inclusion size vs. this normal-
ized oxygen content (Fig. 7) shows ex-
cellent correlation between these two
parameters. This scrutiny of inclusion
size is important because Fig. 5 clearly
shows that reductions in average inclu-
sion size arise as a result of increasing
numbers of inclusions with sizes less
than 0.2 tim. As discussed by Fleck, e t a l .
(Ref. 32), and Liu and Olson (Ref. 34), it
is these small inclusions that are appar-
ently responsible for pinning prior
austenite grain boundaries leading to
smaller austenite grain size and thus re-
duced incidence of intragranular acicu-
lar ferrite.
Although the basicity of the welding
flux is a very subjective measure of the
likely oxygen activity in the weld metal,
the weld metal oxygen should, in princi-
ple, decrease as the basicity increases.
However, examination of the work of Tu-
liani, et al. (Ref. 64), and Dallam, et al.
(Ref. 63), suggests that such a variation is
small and difficult to determine for ba-
sicity indices greater than about 1. This is
no doubt because of the scatter in the
weld metal oxygen contents and inclu-
sion volume fractions discussed previ-
ously. This is illustrated in Fig. 8, which
is a plot of basicity index (BI) vs. weld
metal oxygen for the welds studied in the
present work, and it is clear that there is
very little correlation between these vari-
ables, with sample F296, which has the
abnormally high inclusion volume frac-
tion, really going against the trend of de-
creasing BI with increasing weld metal
oxygen. However, a plot of BI vs. the nor-
malized oxygen content (Fig. 9) shows a
high degree of correlation; this finding
suggests that it is important to take into
account in some way the erratic float-out
rate of the nonmetallic inclusions in
order to develop a quantitative relation-
ship between welding parameters, flux
basicity, inclusion oxygen content, total
oxygen content and, ultimately, acicular
ferrite content. Figure 9 also suggests that
the oxygen content of the inclusions is in-
creasing with increasing basicity index,
which would imply that oxides such as
TiO 2 and AI203 are increasing at the ex-
pense of oxides such as MnO as the oxy-
gen content decreases. This is difficult to
prove because of the problem of quanti-
fying the oxygen content of the inclu-
sions by x-ray analysis in the TEM, as
pointed out previously. However, as dis-
cussed in the next paragraph, there is
clear evidence for an increase in inclu-
sion MnO content with increasing weld
metal oxygen.
As indicated in the introduction, in
addition to the presence of inclusions
-40
G" -so "'",
e~_
-60
• /
-7o
-80 3;
% Acicular Ferrite
Fi,~. 18 Yield strength and ductile/brittle transition temperature vs. acicu-
lar t~rrite content.
with appropriate size distribution and
chemical composition, the possible for-
mation of acicular ferrite depends on the
nature of the CCT curve for the welding
wire and the cooling rate through the
austenite transformation region. Simple
calculations based on the approximate
equation developed by Christensen and
Simonsen (Ref. 65)
At8 5 = 5tiE
indicate that the time for the weld metal
to cool from 800°C to 500°C, At 8 5, for
the weldments studied in the present
work is around 1] s. Liu and Indacochea
(Ref. 66) have indicated that 15 s is ideal
for nucleating optimum amounts of aci-
cular ferrite in HSLA steels. In Equation
3, 11 is the welding efficiency (N0.95 for
SAW) and E the gross heat input in kJ
m m J. Figure 10 (Ref. 67), which shows
the CCT behavior of AX110 weld metal
and which is very similar to that used in
the present work, indicates that a cool-
ing rate of 11s between 800°C and
500°C would lead to the formation of
bainite in the absence of appropriate in-
clusions for acicular ferrite formation.
Consequently, it would seem that the
weld thermal cyle for the fusion zones of
the weldments studied in the present
work is close to ideal for acicular ferrite
formation. The term acicular ferrite may,
in fact, be a misnomer for high-strength
• DBTT
• YS
850
•
%%%
750 ~
~'%%.
" """'/ -1oo
65O
4; 5; 6; 7; 80
steels since the intragranularly nucle-
ated ferrite in HY-IO0 and similar steels
has a much smaller aspect ratio and a
finer morphology than that observed in
lower strength varieties as shown in the
optical and transmission electron micro-
graphs of Figs. 2 and 11. The TEM with
EDX chemical analyses of the weld
metal inclusions in the present work in-
dicate that these contain MnO, Al203,
(3) SiO2, TixOy (x and y unknown) and a
small amount of ZrO 2 with occasionally
a trace of MnS. The average inclusion
composition for each sample (excluding
oxygen) is given in Table 8. Careful
scrutiny of the light element (atomic
number, Z < 11 ) region of the energy dis-
persive x-ray spectra of the inclusions
failed to detect the presence of any ni-
trogen. A plot of Mn, Si, AI, Ti and Zr in
the inclusions is shown in Fig. 12, and it
can be seen that the amount of strong de-
oxidants (AI, Zr and Ti) decreases with
increasing oxygen content while the
amount of the weaker deoxidants (Si and
Mn) increases. This is consistent with the
fact that, as the oxygen activity rises,
weaker deoxidants such as Mn and Si are
more likely to take part in the deoxida-
tion process. Once again, F296 appears
to have more AI, Ti and Zr in the inclu-
sions than would be expected for an oxy-
gen content of 320 ppm. This suggestion
reinforces the assertion that the oxygen
content of F296 weld metal would have
been somewhat less than 300 ppm if the
W E L D I N G RESEARCH SUPPLEMENT [ 339-s
inclusion float-out rate had been similar
to the other samples. Previous experi-
ments in our laboratory (Ref. 62), which
were performed on sample F296 of the
present work, suggest that these inclu-
sions are two-phase mixtures comprising
galaxite MnO.AI203 and pyrophanite,
MnO-TiO 2. If a mass balance is per-
formed on our inclusion compositions,
assuming that all the Ti is tied up as
MnO.TiO2, and the composition of the
remaining ternary phase calculated then
these occur at the positions shown on
the MnO-SiO2-AI203 phase diagram - -
Fig. 13. Also shown on this phase dia-
gram are the inclusion data of Kiessling
and Lange (Ref. 14) which were drawn
from many sources. Their work suggests
that compositions of weld metal inclu-
sions all lie close to a line between alu-
mina and rhodonite, particularly for
weld metal containing minimal amounts
of titanium. Our inclusion results also lie
close to this line (for the Ti-free inclusion
phase) and it would seem that the lower
the oxygen content of the weld metal,
the more likely are the inclusions to lie
near the alumina corner of the MnO-
SiO2-AI203 phase diagram. This is not
surprising as AI is the strongest deoxi-
dant of the three. If MnO.TiO 2 is present
then this phase could be responsible for
the nucleation of acicular ferrite since
several studies have shown that Ti-con-
taining inclusions are most effective at
nucleating acicular ferrite.
There has been much speculation
about how the inclusions responsible for
acicular ferrite develop. Uncertainty ex-
ists because they are formed by a non-
equilibrium process and the experimen-
tal data suggest that they vary in
composition and types of phase present
according to the steel type and the activ-
ities of O, AI, Ti, Si and Mn in the weld
metal. A simplistic approach would be to
ignore the possible effects of nitrogen
(since its activity is low compared with
oxygen in the present work) and then as-
sume that, during inclusion formation,
the oxide components of the inclusions
develop in the following order: AI203,
TixOy, SiO 2 and MnO. This merely
places the oxide components of the in-
clusions in order of increasing free en-
ergy of formation and appears to cor-
rectly predict the oxides present in the
inclusions but not the phases. A thermo-
dynamic calculation is required which
takes into account the dissolution of oxy-
gen, stability of all the possible oxides
and the nucleation and growth of the in-
clusions during the thermal cycle associ-
ated with welding. Babu, et al. (Ref. 58),
have recently performed calculations of
this type for some steel compositions and
analyzed the inclusion phase and com-
position data and found some agreement
340-s [ OCTOBER 1996
between experiment and theory. For ex-
ample, for a submerged arc steel weld
metal of composition (wt-%) C-0.095, Si-
0.31, Mn-1.59, Ti-0.003, AI-0.017 and
0-0.024 welded using a power of 1.865
kJ mm -1, Babu, et al., predict that the fol-
lowing oxides will develop in the fol-
lowing order in the inclusions as the tem-
perature decreases: AI203, MnO.AI203,
FeO.AI203, Ti305, TiO2, SiO 2 and MnO.
It should be noted that the formation of
TiO was not predicted. Unfortunately,
Babu, et al., did not consider the possi-
bility that the oxide phases that devel-
oped had time to react and, as discussed
by Fox and Brothers (Ref. 62), the mor-
phology of the inclusions responsible for
acicular ferrite is very close to spherical
with only slight faceting. This suggests
they could have been molten at some
stage during their formation and pro-
gressed nearer to equlibrium than pre-
dicted by Babu, et al., so that fewer
phases developed during cooling. In-
deed, Fox and Brothers only found two
phases (a manganese alumino-silicate
and pyrophanite, MnO.TiO 2) in the in-
clusions reponsible for acicular ferrite in
sample F296. In addition, there may be
chemical reactions of the sort discussed
by Gregg and Bhadesia (Refs. 59, 60) tak-
ing place near the steel inclusion inter-
faces after the steel is solidified so that
MnO-TiO 2 could have formed as a result
of a TixOy phase in the inclusions react-
ing with the oxygen and manganese in
the steel in their vicinity to form
MnO.TiO 2. Such a reaction would cer-
tainly favor the formation of acicular fer-
rite since the steel around the inclusions
would have reduced hardenability and
thus an increased propensity for ferrite
formation.
It is accordingly clear that weld metal
oxygen content defines the size distribu-
tion and chemistry of the inclusions re-
sponsible for acicular ferrite and work by
Dallam, et al. (Ref. 63), using the CaO-
SiO2-CaF 2 flux system during sub-
merged arc welding of Nb-microalloyed
steel suggests that the optimum oxygen
content for the nucleation of a maximum
amount of acicular ferrite is about 260
ppm. As discussed previously, this and
other work suggests that for higher oxy-
gen contents more small (<0.2 pm) in-
clusions are formed which pin prior
austenite grain boundaries, giving a
smaller austenite grain size and leading
to grain-boundary-nucleated mi-
crostructures. Weld metal oxygen con-
tents significantly less than 260 ppm are
likely to have insufficient inclusion num-
bers to generate large amounts of acicu-
lar ferrite. Figures 4, 5 and 6 are consis-
tent with this trend in that the fraction of
smaller inclusions increases with in-
creasing oxygen content (taking inclu-
sion volume fraction into account). The
exception is F296 which has the lowest
fraction of sub-0.2 pm inclusions but has
an inordinately high inclusion volume
fraction so that its weld metal oxygen
content is somewhat higher at 320 ppm.
These results suggest that increasing
weld metal oxygen should lead to de-
creasing acicular ferrite content for our
samples and indeed this is the case ex-
cept for sample F296, as shown in Fig.
14. However, when the acicular ferrite
content is plotted against weld metal
oxygen normalized to inclusion volume
fraction (calculated in the same manner
as before) then there is a very strong cor-
relation (Fig. 15) as there is when acicu-
lar ferrite content is plotted against aver-
age inclusion size - - Fig. 16. As
mentioned previously, this suggests that
if the inclusion float-out rate for F296
had been similar to the other weld met-
als then its weld metal oxygen content
would have been somewhat lower than
300 ppm. The preceding arguments in-
dicate that there should be a strong cor-
relation between increasing basicity
index and acicular ferrite content and
Fig. 17 shows that this is indeed the case.
This finding compares very favorably
with the work of Bhatti, et al. (Ref. 28),
who obtained similar results for a cal-
cium-treated C-Mn microalloyed steel.
Finally, it is important to illustrate the
effect of acicular ferrite on the strength
and toughness of the weld metal. As
shown in Table 4, samples F289 and
F293 marginally failed to meet the yield
strength requirement for Navy HY-100
steel because both of these had lost man-
ganese during welding due to the use of
a lower basicity flux and, in the case of
F293, because of a low flux MnO con-
tent. In addition, F289 and F293 only
contained 43 and 41 vol-% acicular fer-
rite, respectively. All the weld metals
met specification for elongation and
only one (F295) failed to meet toughness
requirements and this failure was mar-
ginal as shown in Table 5. This is not sur-
prising since F295 contained the lowest
amount of acicular ferrite at 27 vol-%.
Samples F292 and F296 easily met spec-
ifications in all respects and this is easy
to understand since they lost less alloy-
ing elements during welding (higher ba-
sicity fluxes) and contained 65 and 78
vol-% acicular ferrite, respectively. Fig-
ure 18 summarizes these comments and
shows that the yield strength is markedly
increased and that the ductile/brittle
transformation temperature is dramati-
cally reduced by increasing the acicular
ferrite content of the samples studied in
the present work. This of course is the
reason it is desirable to maximize the
amount of this microconstituent as both
strength and toughness are improved by
its presence.
Conclusions and Recommendations
The conclusions and recommenda-
tions to be drawn from this work are as
follows:
1) Small changes in flux basicity can
lead to marked changes in the amount of
acicular ferrite in submerged arc weld
metal of Navy HY-100 steel. This occurs
because the acicular ferrite content de-
pends critically on the size, volume frac-
tion and composition of the oxide inclu-
sions present in the weld metal. The use
of flux c o m p o s i t i o n s similar to those
shown in Table 3 for F292 or F296 for
submerged arc welding of Navy HY-100
steel with L-Tec 1 20S welding wire with
the composition shown in Table 2 w o u l d
appear to lead to appropriate inclusion
contents consistent with maximizing the
amount of acicular ferrite present. Using
the definition of basicity due to Tuliani, et
al. (Ref. 64), these fluxes have basicities
in the range 2.83 to 2.97. Other work by
the authors (Ref. 68) indicates that the
M n O content of the flux should be about
1.0 wt-% in order to minimize losses of
manganese by o x i d a t i o n effects other
than those due to basicity.
2) It may be desirable to have slightly
greater levels of Ti and B than the
amounts present in the L-Tec 120S wire
used in the present work. This suggestion
is supported by all previous w o r k de-
scribed in the introduction. However, it
is probably difficult to produce welding
wires with closely controlled contents of
the trace elements Ti, AI and B. It is, of
course, possible that other batches of L-
Tec 120S wire may have compositions
nearer the optimum for maximizing the
a m o u n t of acicular ferrite in HY-100
weld metal with fluxes F292 and F296.
Acknowledgments
The authors w o u l d like to thank the
Base Commanders of the Naval Post-
graduate School, Monterey, Calif., and
the Annapolis Detachment of the Carde-
rock Division, Naval Surface Warfare
Center, Annapolis, Md., for the provision
of financial support and laboratory facil-
ities. The authors w o u l d also like to thank
Commander D. G. Brothers and Lt. Com-
mander K. W. Kettell for their help with
the analysis of the results.
References
1. Bhadesia, H. K. D. H. 1992. Bainite in
Steels, the Institute of Materials, Chapter 10,
London, U.K.
2. Edwards, G. R., and Liu, S. 1990. Recent
developments in HSLA steel welding. Pro-
ceedings of the First United States-Japan Sym-
posium on Advances in Welding Metallurgy.
Cosponsored by the American Welding Soci-
ety, the Japanese Welding Society and the
JapaneseWelding Engineering Society, held at
San Francisco, Calif., and at Yokohama, Japan,
pp. 213-292.
3. Grong, O., and Matlock, D. K. 1986.
Microstructural developments in mild and low
alloy steel weld metals. International Metals
Review 31 (1): 27-48.
4. Abson, D. J., and Pargeter, R. J. 1986.
Factors influencing as-deposited strength, mi-
crostructure and toughness of manual metal
arc welds suitable for C-Mn steel fabrications.
International Metals Review 31 (4): 141-194.
5. Palm, J. H. 1972. How fluxes determine
the metallurgical properties of submerged arc
welds. Welding Journal 51 (7): 358-s to 360-s.
6. Davis, M. L. E., Pargeter, R. J., and Bai-
ley, N. 1983. Effects of titanium and boron ad-
ditions to submerged arc welding fluxes. Metal
Construction 15: 338-344.
7. Abson, D. J., Dolby, R. E., and Hart, P.
H. M. 1979. The role of non-metallic inclu-
sions in ferrite nucleation in carbon steel weld
metals. Proceedings on Trends in Steels and
Consumables for Welding Conference, pp,
75-101. The Welding Institute, U.K.
8. Kou, S. 1987. Welding Metallurgy,
Chapter 9., John Wiley and Sons, New York,
N.Y.
9. Yang, }. R., and Bhadesia, H. K. D. H.
1987. Proceedings of the International Con-
ference on the Welding Metallurgy of Struc-
tural Steels, pp. 549 563, TMS-AIME, War-
rendale, Pa.
10. Strangwood, M., and Bhadesia, H. K.
D. H. 1987. The mechanism of acicular ferrite
formation in steel weld deposits. Advances in
Welding Science and Technology, pp.
209-213, ASM International, Materials Park,
Ohio.
11. Brooksbank, D., and Andrews, K. W.
1972. Stressfields around inclusions and their
relation to mechanical properties. Journal of
the Iron and Steel Institute. 210: 246-255.
12. Ito, Y., and Nakanishi, M. 1976. A
study of Charpy impact properties of weld
metals with submerged arc welding. Sumit-
omo Search. 15: 42-62.
13. Kanazawa, S., Nakashima, A.,
Okamoto, K., and Kanaya, K. 1976. Improve-
ment of weld fusion zone toughness by TiN.
Transactions of the Iron and Steel Institute of
Japan. 16: 486~-95.
14. Kiessling, R., and Lange, N. 1978.
Non-Metallic Inclusions in Steel. The Metals
Society.
15. Koukabi, A. A., North, T. H., and Bell,
H. B. 1978. Flux formation, sulfur, oxygen and
rare earth additions in submerged arc welding.
Proceedings of the Trends in Steels and Con-
sumables for Welding Conference, pp.
281-297, The Welding Institute, Abington,
U.K.
16. Barrite, G. S., and Edmonds, D. V.
1981. The microstructure and toughness of
HSLA steel weld metals. Proceedings on Ad-
vances in the Physical Metallurgy of Steels
Conference, pp. 126-135, The Metals Society.
17. Cochrane, R. C., and Kelville, B. R.
1981. Influence of inclusion morphology on
microstructure and toughness of submerged
arc weldments. Proceedings of the Steels for
Line Pipe and Pipeline Fittings Conference,
pp. 51-60, the Metals Society.
W E L D I N G RESEARCH SUPPLEMENT I 341-s
18. Harrison, I~ L., and Farrar, R. A. 1981.
Influence of oxygen-rich inclusions on tile
austenite-to-ferrite phase transformations in
high-strength-low-alloy (HSLA) steel weld
metals. Journal of Materials Science 16:
2218-2226.
19. Ferrante, M., and Farrar, R. A. 1982.
The role of oxygen-rich inclusions in deter-
mining the rnicrostructure of weld metal de-
posits. Journal of Materials Science 17:
3293-3298.
20. Ito, Y., Nakanishi, M., and Komizo, Y.
]982. Metals Construction 14: 472-478.
21. Kelville, B. R, and Cochrane, R. C.
1982. Factors (ontrolling the microstructure
and toughness of submerged arc weldments.
Proceedings of the Welding Technology "82
Conference, pp. 263-269, The Australian
Welding Institute.
22. Ricks, R. A., Howell, P. R., and Barrite,
G. S. 1982. The nature of acicular ferrite in
HSLA steel weld metals. Journal of Materials
Science, 17: 732-740.
23. Ahlblom, B., Bergstrom, H., Hannerz,
N. E., and Werlefors I. 1983. Influence of
welding parameters on nitrogen content and
microstructure of submerged arc weld metal.
Proceedings of the Effects of Residual, Impu-
rity and Mk roalloying Elements on Weldabil-
ity and Weld Properties Con ference, Paper 38,
the Welding Institute, Abington, U.K.
24. Cochrane, R. C., Ward, J. L., and
Kelville, B. R. 1983. The influence of deoxi-
dation and/or desuifurization practice on the
weld metal toughness of high dilution welds.
Proceedings of the Eft~cts of Residual, Impu-
rity and Microalloying Elements on Weldabil-
ity and Weh-tProperties Conference, Paper 16,
the Welding Institute, Abington, U.K.
25. Devillers, L., Kaplan, D., Marandet, B.,
Ribes, A., and Riboud, P. V. 1983. The effect
of low level concentrations of some elements
on the toughness of submerged arc welded C-
Mn steels. Proceedings of the Effects of Resid-
ual, Impurity and Microalloying Elements on
Weldability and Weld Properties Conterence.
Paper 1, The Welding Institute, Abington, U.K.
26. Kayali, E. S., Corbett, J. M., and Kerr,
H. W. 1983. Observations on inclusions and
acicular ferrite nucleation in submerged arc
HSLA welds. Journal of Materials Science Let-
ters 2:123-126.
27. Saggesse,M. E., Bhatti, A. R., Hawkins,
D. N., and Whiteman, J. A. 1983. Factors in-
fluencing inclusion chemistry and microstruc-
ture in submerged arc welds. Proceedings of
the Effects of Residual, Impurity and Microal-
Ioying Elements on Weldability and Weld
Properties Conference, Paper 15, The Welding
Institute, Abington, U.K.
28. Bhatti, A. R., Saggese,M. E., Hawkins,
D. N., Whiteman, J. A., and Golding, M. S.
1984. Analysis of inclusions in submerged arc
welds in microalloyed steels. Welding Journal
63(7): 224-s to 230-s.
29. Court, S. G., and Pollard, G. 1985. Mi-
croanalysis of weld metal inclusions. Journal
of Material Science Letters 4: 427-430.
30. Oldland, R. B. 1985. The influence of
aluminum and nitrogen on the microstructure
and properties of single-pass submerged arc
welds. Australian Welding Research 14:
44-56.
31. Dowling, J. M., Corbett, J. M., and Kerr,
H. W. 1986. Observations on inclusions and
acicular ferrite nucleation in submerged arc
HSLA welds in high strength low alloy steels.
Metallurgical Transactions A 17:1611-1623.
32. Fleck, N. A., Grong O., Edwards, G.
R., and Matlock, D. K. 1986. The role of filler
metal wire and flux composition in submerged
arc weld metal transformation kinetics. Weld-
ing Journal 65(5): 113-s to 121 -s.
33. Horii Y., Ohkita, S., Wakabayashi, M.
and Namura, M. 1987. Development of Weld-
ing Materials for Low Temperature Service. In-
ternational institute for Welding Document II-
A-7 1487.
34. Liu, S., and Olson, D. L. 1986. The role
of inclusions in controlling HSLA steel weld
microstructures. Welding Journal 65 (6): 139-
s to 149-s.
35. Thewlis, G. 1986. The influence of
pipe plate and consumable chemistry on the
composition, microstructure and toughness of
weld metal. Prodeedings of Welding and Per-
formance of Pipelines Conference, Paper 9,
The Welding Institute, Abington, U.K.
36. Abson, D. J. 1987. Small Particles in
Weld Metals - - A Review. International Insti-
tute for Welding Document IXJ-122-87.
37. Lau, T. W., Sadowsky, M. M., North, T.
H., and Weatherly, C. G. 1987. Effect of nitro-
gen on the toughness of HSLA weld deposits.
Proceedings of an International Conference
on the Welding Metallurgy of Structural Steels,
pp. 349-365, The Metallurgical Society, War-
rendale, Pa.
38. Mills, A. R., Thewlis, G., and White-
man, J. A. 1987. The nature of inclusions in
steel weld metals and their influence on the
formation of acicular ferrite. Materials Science
and Technology 3:1051-1061.
39. Pargeter, R. J. 1987. Investigation of
submerged arc weld metal inclusions. Pro-
ceedings of an International Conference on
the Welding Metallurgy of Structural Steels,
pp. 393-414, The Metallurgical Society, War-
rendale, Pa.
40. Yamamoto, K., Matsuda, S., Haze, T.,
Chijiiwa, R., and Mimura, H. 1987. Proceed-
ings of a Conference on Residual and Un-
specified Elements in Steel, pp. 1-24, ASM In-
ternational, Materials Park, Ohio.
41. Barbaro, F. J., Edwards, R. H., and East-
erling, K. E. 1988. The composition and mor-
phology of non-metallic inclusions in HSLA
steel weld metals. Proceedings of the 7th Na-
tional Conference of the Australian X-Ray
Analysis Association, pp. 1-14.
42. Chijiiwa, R., Tamehiro, H., Hirai, H.,
Matsuda, H., and Mimura, H. 1988. Proceed-
ings of a Conference on Offshore Mechanics
and Artic Engineering, pp. 1-8.
342-s I O C T O B E R 1996
43. Horii, Y., Wakabayashi, M., Okhita, S.,
and Namura, M. 1988. Studyofthe Toughness
of Large Heat Input Weld Metal for Low Tem-
perature Service TMCP Steel. Nippon Steel
Technical Report No. 37, pp. 1-9, the Nippon
Steel Co., Japan.
44. Lau, T. W., Sadowsky, M. M., North, T.
H., and Weatherly, C. G. 1987. Effect of nitro-
gen on the toughness of HSLA weld deposits.
Materials Science and Technology 4: 52-61.
45. Sugden, A. A. B., and Bhadesia, H.K.
D. H. 1988. The nonuniform distribution of in-
clusions in low-alloy steel weld deposits. Met-
allurgical Transactions A 19: 669-674.
46. Dallam, C. B., and Olson, D. L. 1989.
The effects of stress and grain size on weld
metal ferrite formation. Welding Journal 68
(5): 198-s to 205-s.
47. Kluken, A. O., and Grong, 0 . 1989.
Mechanisms of inclusion formation in AI-Ti-
Si-Mn deoxidized steel weld metals. Metallur-
gical Transactions A 20:1335-1349.
49. Kluken, A. O., and Grong, 0 . 1989.
Proceedings of a Conference on Recent Trends
in Welding Science and Technology, pp.
781-786, ASM International, Materials Park,
Ohio.
50. Babu, S. S., and Bhadesia, H. K. D. H.
1990. Transition from bainite to acicular fer-
rite in reheated Fe-Cr-C weld deposits. Mate-
rials Science and Technology 6:1005-1020.
51. Es-Souni, M., and Beaven, P. A. 1990.
Microanalysis of inclusion/matrix interfaces in
weld metals. International Institute of Welding
Document II-A-8 15-90.
52. Evans, G. M. 1992. The effect of tita-
nium in SMA C-Mn steel multipass deposits.
Welding Journal 71 (12): 447-s to 454-s.
53. Kluken, A. O., Grong, O., and Rorvik,
G. 1990. Metallurgical Transactions A 21:
2047-2058.
54. Oh, D. W., OIson, D. L., and Frost, R.
H. 1990. The influence of boron and titanium
on low carbon steel weld metal. Welding Jour-
nal 69(4): 151 -s to 158-s.
55. Ringer, S. R., Barbaro, F., Krauklis P.,
and Easterling, K. 1990. Proceedings of a Con-
ference on Microstructure Control to Achieve
Properties in Modern Steels, pp. 1-6, Institute
of Metals and Materials, Australasia, Mel-
bourne.
56. Thewlis, G. 1990. Transformation Ki-
netics of Submerged Arc Weld Metal, British
Steel Internal Report IXJ, 165-90:1-11.
57. Kluken, A. O., Grong, 0., and Hjelen,
J. 1991. The origin of transformation textures
in steel weld metals containing acicular fer-
rite. Metallurgical Transactions A 22:
657-663.
58. Babu, S. S., David, S. A., Vitek, J. M.,
Kundra, K., and DebRoy, T. 1995. Develop-
ment of macro- and microstructures of C-Mn
low alloy steel welds - - inclusion formation.
Materials Science and Technology 11:
186-199.
59. Gregg, J. M., and Bhadesia, H. K. D. H.
1994. Bainite nucleation from mineral sur-
faces. Acta Metallurgica 42: 3321-3330.
60. Gregg, J. M., and Bhadesia, H. K. D. H.
1994. Titanium-rich mineral phases and the
nucleation of bainite. Metallurgical Transac-
tions. In press.
61. Lee, J-L. 1994. Evaluation of the nu-
cleation potential of intragranular acicular fer-
rite in steel weldments. Acta Metallurgica et
Materiala 42(10): 3291-3298.
62. Fox, A. G., and Brothers, D. G. 1995.
The role of titanium in the non-metallic inclu-
sions which nucleate acicular ferrite in the
submerged arc weld (SAW) fusion zones of
Navy H Y 100 steel. Scripta Metallurgica et Ma-
teriala 32 (7):1061-1066.
63. Dallam, C. B., Liu, S., and OIson, D. L.
1985. Flux composition dependence of mi-
crostructure and toughness of submerged arc
HSLA weldments. Welding Journal 64 (5):
140-s to 151 -s.
64. Tuliani, S.S., Boniszewski, T., and
Eaton, N. F. 1969. Notch toughness of com-
mercial submerged arc weld metal. Welding
Metal and Fabrication 37(8): 327-329.
65. Christensen, N., and Simonsen, T.
1981. Transformation behavior of weld metal.
Scandinavian Journal of Metallurgy 10(4):
147-156.
66. Liu, S., and Indacochea, J. E. 1990.
Control of chemical composition and mi-
crostructure in low-carbon microalloyed steel
weldments. Welding Theory and Practice,
Chapter 4, pp. 117-148, North Holland Press.
67. Lyttle, J. E., Dorschu, K. E., and
Fragetta, W. A. 1969. Welding Journal48 (11 ):
493-s to 498-s.
68. Fox A. G., and Franke G. L. Unpub-
lished work.
69. Grong, (D., Kluken, A. O., Nylund, H.
K., Dons, A. L., and Hjelen, J. 1995. Catalyst
effects in heterogeneous nucleation of acicu-
lar ferrite. Metallurgical Transactions A 26:
525-534.