Flux Composition Dependence

of Microstructure and Toughness of

Submerged Arc HSLA Weldments

CaF2-CaO-SiC>2 system fluxes produce good quality niobium

microalloyed HSLA weldments with very low oxygen content

BY C. B. DALLAM, S. LIU, AND D. L OLSON

ABSTRACT. Twenty eight reagent grade fused fluxes from the CaF2-CaO-Si02 sys­tem were used to produce bead-on-plate and double-V-groove submerged arc welds on a quenched-and-tempered nio­bium HSLA steel. An E70S3 welding wire was used with two different heat inputs —namely, 1.9 and 3.3 k j /mm (48.3 and 83.8 kj/in.) A niobium microalloyed steel was selected because of its fine grained microstructure, high yield strength, and high toughness at low tem­peratures. Fluxes from the CaF2-CaO-Si02 system were selected because of their low oxygen potential, and their ability to produce low oxygen (80-450 ppm) welds. Quantitative metallography and chemical analysis were performed on the welds. The chemical behavior of this flux system has been characterized with respect to manganese, silicon, niobium, and sulfur.

The lower heat input welds showed predominantly fine microstructure of acicular ferrite. At high oxygen content, a higher percentage of grain boundary fer­rite (ferrite veining) was observed. By reducing the oxygen in the weld metal, the amount of acicular ferrite was increased. With further reduction of weld metal oxygen, the main microstructural feature, instead of acicular ferrite, became bainite. Using higher heat input.

Based on paper presented at the 64th Annual A WS Convention held in Philadelphia, Pennsyl­vania, during April 24-29, 1983.

C B. DALLAM, S. LIU, and D. L. OLSON are with the Center for Welding Research, Department of Metallurgical Engineering, Col­orado School of Mines, Golden, Colorado.

the weld metal microstructure transition with oxygen level was not so clear.

In spite of the essentially similar optical microstructure and similar chemical com­position (other than oxygen), the mechanical properties of the various welds were observed to be very differ­ent. Toughness data (upper shelf energy and transition temperature) were found to correlate with weld metal oxygen content. The upper shelf energy decreased with increasing oxygen level in the weld metal.

Introduction and Background

HSLA steels were developed to achieve high yield strength and at the same time maintain a reasonable level of toughness with a minimum of alloying. Due to the possibility of increased design loading and strength to weight ratio, more and more structural applications using HSLA steels are being seen. Some examples are line pipes for gas and oil transportation and off-shore structures. The physical metallurgy of these microal­loyed steels for the optimization of their microstructure and properties has already been treated extensively in the literature (Refs. 1-4) and are not discussed in this paper.

Most applications of HSLA steels involve structures where welding is used. Two major consequences of this fabrica­tion process are the deterioration of the base metal properties due to the welding thermal cycles and the introduction of a solidification structure which is heteroge­neous (compared to the base metal) both chemically and microstructurally. In the fusion zone, the weld metal composition,

heat input and cooling rate, solidification characteristics, and reheating thermal cycles (in multiple pass welds) contribute to the final properties of the weld joint. Adjacent to the fusion zone is a thermally affected region (heat-affected zone, i.e., HAZ) within which the base metal micro-structure is altered by the high tempera­ture of the molten weld pool. Martensite or other low temperature transformation products may be formed impairing the toughness of these regions.

As noted below, under separate head­ing, the presence of oxygen can influence weld metal microstructure and proper­ties. With this in mind, the behavior of CaF2-CaO-Si02 flux systems was studied with the purpose of reporting on weld metal performance when using these fluxes on a niobium microalloyed steel.

Some Background

Typical Microstructure of C-Mn Steel Weldments

Several different microstructures may be obtained in the weld metal of low carbon microalloyed steels. They are grain boundary ferrite (BF), side plate ferrite (SPF), acicular ferrite (AF), upper bainite (AC), and micro-constitutents such as pearlite, cementite, and martensite.

Figure 1 shows some of the main constituents of a C-Mn steel weldment. A comparison of the various classifications of low carbon, low alloy steel weld metal microstructure is shown in Table 1.

Factors Affecting Weld Metal Microstructure

F.-jctors affecting weld metal toughness have been studied and acicular ferrite

140-s | MAY 1985

was found to be the constituent respon­sible for the high toughness (Refs. 6, 7). Acicular ferrite is formed intragranularly, resulting in randomly oriented short fer­rite needles with a basket weave feature. This interlocking nature, together with its fine grain size, provides the maximum resistance to crack propagation by cleav­age. For this reason, it has become increasingly important to understand the factors which would maximize the vol­ume fraction of acicular ferrite in the weld metal.

Weld metal composition (alloying ele­ments and oxygen), prior austenite grain size, and welding heat input (cooling rate) are the three main factors that determine the microstructure of a weld metal. It is shown (Ref. 8) that an increasing cooling rate progressively refines the resulting microstructure from grain boundary fer­rite to side plate ferrite, acicular ferrite, bainite, and eventually to martensite.

Alloying elements in the weld metal may come from the base metal, the welding electrode, and the welding fluxes. Hardenability agents such as man­ganese, chromium, and molybdenum will shift the austenite decomposition trans­formation to longer delay times. Superim­posing a cooling curve on the transfor­mation curves, one notices that the refin­ing of the final weld microstructure can be explained. This is shown schematically in Fig. 2. Composition control is necessary in order to maximize the volume fraction of acicular ferrite, because excessive alloying elements can cause the forma­tion of bainite and martensite.

A number of recent investigations (Refs. 9-13) indicate that oxygen affects the weld metal microstructure and the

>vru<M

S ::A:-

'Af 7 k~,

. V

AA

a

--m. !•;$-'>* V-.

I I g g f e • J u:> :•'•• • • 7- A a • ?»7SS""-"- '"' "- '*-• . --^ \ ;j 4 \ :. A AA':

*M?A :A: •

sl»>. >*•*c-t^-i '/.;•• A &• •'- -•' -7 A ;tV.ik•'*;.'",-•*-•.• 7^^~-;.*^.«" -'. '

ml <yw!U!Sm&-&*m?-m,m Fig. 1 — Typical microstructures in low carbon low alloy steel weldments showing: grain boundary ferrite (a); side plate ferrite (b); acicular ferrite (c); bainite (d)

**- • q

xrf" y

_j~&: ;-' J

a *<5$'y- :

"<'_J'V-, A'"

V : - • • • ' ' - - ^ •

•A^yM *. • w AS"-

> ,»s;« •0 m

j*r • • ^ i * '

Lt '4-At* /f

VAA i^v a "A 4^

%%

w - ' - . ^ V . : - L - ^

te r r . . -s __,*£-*-- r".**c"*

J^N^L--^ 77̂ /*"""".-'-r^-r-'l % . ;** _?-V-~it-

.-/>1s

^^Sfe:^1:>'S^^.5 /#^#<<^^ 50lJ

mechanical properties in the form of inclusions. These inclusions may be a product of the deoxidation process or result from some solid state transforma­tion (Ref. 14). Using Charpy testing data,

Ito and Nakanishi (Ref. 15) observed that both high and low weld metal oxygen content weldments (with >500 ppm or <200 ppm) showed poor impact proper­ties. Only the intermediate level oxygen

Table 1—Classification and

Dube, CA . Aronson, H.I.

Cochrane, R.C.*

Allotriomorphic (polygonal) ferrite*

-Primary and secondary

ferrite sideplates*

Intragranular ferrite plates*

Massive ferrite Pearlite Lath martensite Twinned martensite Retained austenite Upper (occasionally lower)

bainite

Terminology of Microstructures in Low C-Low Alloy Steel Weld Metal (Ref. 5).

Widgery, D.J.

Proeutectoid ferrite

-Lamellar component

(product)

• Acicular ferrite

-Pearlite Martensite

Abson, D.J.

Grain boundary ferrite

Polygonal ferrite Ferrite with aligned M-A-C

Acicular ferrite

-Ferrite-carbide aggregate Martensite M-A consitutent

Japanese Researchers'3'

Proeutectoid ferrite. Grain boundary ferrite. Polygonal ferrite

Polygonal ferrite Ferrite sideplate Lath-like ferrite

Acicular ferrite Needle-like ferrite Fine grained ferrite Granular ferrite Pearlite Martensite M-A constituent High-C martensite Upper bainite

Others'6 '

Proeutectoid ferrite. Grain boundary ferrite. Polygonal ferrite, Block ferrite

Ferrite islands Ferrite sideplate Lath ferrite Upper bainite Acicular ferrite Fine bainitic ferrite

—

Martensite M-A constituent Lath ferrite

(a) Nakanishi, M., Watanabe. I. I.. Kojima, T.. Mor i , N. et al, Kikuta, Y., et al. (b) Pargeter, R. I.; Watson, M . N.. Harrison, P. L . Farrar, R. A.; Levine, E„ Hill, D. C , Sawhill |r., |. M., Choi. C L., Hill, D. C , Clover, A. C , et al., Pacey, A. ).. Kerr, K. W., Abson, D. I.. Dolby, R. E., and Hart, P. H. M.

WELDING RESEARCH SUPPLEMENT 1141-s

£- —

High , -Oxygen '

\

/ / I 1 AC I j

1 A F \ | High A 3 i Hordenability

/ ' \ 1

M

log ( t )

Fig. 2 — Schematic continuous cooling transformation diagram of a low carbon low alloy steel weldment showing the effects of composition on weld metal phase transformations. In the dia­gram, "M" represents martensite, "AC" designates upper bainite (aligned ferrite with carbide), "AF" signifies acicular ferrite, and "BF" represents blocky ferrite

CoO 0.10 020 0.30 0.40 0.50 0.60 0.70

NCaF? ~ 0.80 0.90

CoF,

Fig. 3 —Partial ternary diagram of CaF2-CaO-Si02 system showing the nominal compositions of the fluxes used (open circles first set of welds; filled circles—both first and second set of welds)

welds (200 p p m to 500 ppm) p roduced tough acicular ferr i te structures. Kikuta ef al. (Ref. 16) repo r ted the same t rend for electroslag we ldments .

Cochrane and K i r k w o o d (Ref. 10) stud­ied the effects o f we ld metal oxygen on acicular ferr i te fo rmat ion by p roduc ing a series of we ld metals o f similar chemical compos i t ion , w i t h the except ion of oxy­gen level. Intermediate w e l d metal oxy­gen content (200-300 ppm) gave a pr i ­marily acicular ferr i te microstructure, wh i le side plate ferri te p redomina ted at high oxygen contents (500 ppm) . They conc luded that w e l d metal oxygen varia­t ion a f fec ted the inclusion size distr ibu­t ion and the surface energy o f the inclu­s ion/matr ix interface w h i c h resulted in di f ferent microstructures. Abson , Do lby and Hart (Ref. 9) observed the same kind o f microstructural evo lu t ion as Cochrane and K i r kwood . They also postu lated that oxygen-r ich inclusions can directly nucle­ate acicular ferr i te. In the absence o f these inclusions, lath t ype structures (bainite) wil l f o r m .

For higher oxygen systems ( > 5 0 0 ppm), Indacochea and O lson (Ref. 17) obta ined predominant ly grain boundary ferr i te and b locky ferr i te (intragranularly nucleated coarse ferr i te grains). Some ev idence of the direct nucleat ion o f the acicular ferr i te phase on inclusions was observed . Increasing w e l d metal oxygen content , and thus the amount of inclu­sions, should m o v e the t ransformat ion curve in Fig. 2 t o shorter delay times.

Harrison and Farrar (Ref. 18) repor ted that the austenite t o ferr i te t ransforma­t ion temperatures can be altered by the interact ion of inclusions w i t h austenite grain boundar ies. A high inclusion c o n ­centrat ion tended to reduce the austenite grain size, favor ing grain boundary ferr i te and side plate ferr i te fo rmat ion . Ferrante and Farrar (Ref. 19) de te rmined that the

inclusions cont ro l the austenite grain g r o w t h by pinning of the grain b o u n d ­aries, fo l low ing the Zener equat ion of prec ip i ta te-boundary interact ion. They also f o u n d that acicular ferr i te directly nucleates f r o m the inclusions.

Keville (Ref. 14) and Pargeter (Ref. 20) s h o w e d that the inclusions' size, distr ibu­t ion, and t ype (composi t ion) can be cor­related t o the microst ructure and the toughness of the we ld metal . Keville (Ref. 21) also indicated that the inclusions are of complex nature and that distinct geo­metrical shapes could be associated w i t h the t ype of we ld ing flux used. A change in inclusion shape was always observed w i th a change in the chemical compos i ­t ion o f the inclusion.

The types o f inclusions w e r e also cata­logued using particle analyzer scanning electron microscopy (Ref. 22). The idea that the dif ferential thermal cont ract ion of the inclusions and the austenite matrix dur ing cool ing of the we ldmen t may p rov ide favorable condi t ions for the fer­

rite nucleat ion is also being investigated (Ref. 23).

Screes of Oxygen in the Submerged Arc W ding Process

The four main sources of w e l d metal oxygc can be ident i f ied as the f lux, the base metal , the e lect rode we ld ing w i re , and the at1 osphere. In general, the oxy­gen and n i t rogen p ickup f r o m the a t m o ­sphere can L e considered as contamina­t ion and i ;itutes only a minor part o f the w e l d metal oxygen (Ref. 24). H o w e v ­er, it may become m o r e significant in welds made w i t h high basicity fluxes.

The lower viscosity o f high basicity fluxes may be the cause o f the increase in v e l d metal oxygen p ickup (Ref. 24). The t ase metal and we ld ing e lect rode c o n ­tain approximate ly 100 p p m oxygen each; the amount is de termined by the r, telting process. The amoun t that the base metal and e lec t rode cont r ibute is appreciably less than that generated f r o m the we ld ing flux. The major componen ts

Table 2—Chemical Compositions of Base Metals and Welding Electrode, %

C Mn Si P S Nb Al Cr Ni Mo Cu 0U> N<a>

Plate I

0.14 1.38 0.41 0.005 0.030 0.042 0.041 0.11 0.14 0.06 0.18

83 90

Plate II

0.09 1.40 0.24 0.013 0.004 0.033 0.024 0.13 0.13 0.04 0.20

E70S3 electrode

0.10 1.15 0.53 0.025 0.025

0 185 105

(a) ppm.

142-s I MAY 1985

Table 3—Welding Conditions for the 1st and 2nd Set of Experiments

1st

Potential, volts Current, amperes Travel speed, mm/min Heat Input, kj/mm

30 500 480 1.87

2nd

28 520 265 3.29

2.0

\

/

I / .125

1 1

\ .125

4.0

t .625

1

Fig. 4 —loint geometry of the double-V-grooved welds; measurements given in inches

of a flux are generally oxides, and a reasonably high oxygen potential of the flux would be expected. Therefore, the main source of oxygen in the submerged arc welding process is the flux. Some commercial flux systems based on FeO, MnO and SiC>2 may contribute up to 1000 ppm of oxygen to the weld metal.

The requirement of low weld metal oxygen to achieve high toughness (high volume fraction of acicular ferrite) leads to the reformulation of fluxes attempting to reduce the oxygen transfer. One way is to substitute non-oxygen carriers (such as fluorides) for some of the oxides in the fluxes. Decreasing the amount of oxygen in the flux, a reduction of weld metal oxygen would be expected. A second way is to replace weak oxides (e.g., silica and ferrous oxide) partially by stronger oxides such as calcium oxide and magne­sium oxide. Since the Ca-O bonds are much stronger than the Si-O bonds, the degree of dissociation of calcium oxide in the molten flux would be less than that of silica, resulting in a lower weld metal oxygen content. Another possible alter­native is the addition of metal powders (e.g., aluminum, titanium) to the fluxes to reduce the weld metal oxygen pickup (Refs. 11,25,26).

Experimental Procedure

Twenty eight reagent grade fused fluxes were produced from the CaF2-CaO-Si02 system, using nominal compo­sition at every 10% increment on the ternary phase diagram —Fig. 3. The details of the flux preparation were reported in a previous paper (Ref. 27). Two quenched-and-tempered niobium microalloyed steel plates of similar grade were used to make the welds, and a %i in. (2.38 mm) diameter E70S3 welding electrode was used. The chemical com­position of the plates and the welding electrode are given in Table 2.

Two sets of experiments were per­formed:

1. Bead-on-plate welds were made on 4 X 1 0 X 0 . 5 in. ( 1 0 0 X 2 5 0 X 1 3 mm) plates (machined from plate I) with each one of the fluxes. Instead of using the flux hopper, a flux bin was set up along the torch path. The welding process parame­ters are shown in Table 3. The welds were used to study variables such as arc instability, bead morphology and pene­

tration, and contact angle for interfacial energy calculations.

Metallographic analysis and Rockwell hardness measurements were performed on each weldment. Chemical analyses were performed using a Baird Atomic Spectrovac Model 1000 Emission Spec­trometer. Weld metal oxygen, nitrogen, sulfur, and carbon contents were deter­mined using Leco Analyzers. The chemi­cal interaction between the molten flux and weld pool was also studied.

2. Double-V-groove welds were made on plates machined from the plate II. The dimensions of the welding plates were 4 X 1 1 X 0 . 6 2 5 in. (100 X 280 X 16 mm). Specific fluxes, indicated in Fig. 3, were used for this second set of welds. The joint design is shown in Fig. 4; the welding conditions are shown in Table 3.

The welds were radiographically examined before standard Charpy speci­mens were prepared. Approximately nine Charpy bars were machined out of each weld. The Charpy V-notch tough­ness measurements were done using a Tinius Olsen Charpy machine, with the testing temperatures ranging from —100 to 120°C (-148 to 248°F). At least eight temperatures were tested for each weld. Fractographic analyses were also done using AMR scanning electron micro­scope.

Carbon extraction replicas were made on some of the welds in order to exam­ine the inclusions and other second phases present in the welds. The replicas were examined using the Philips 200 transmission electron microscope.

In an attempt to characterize the crys­talline nature of the inclusions, an acid extraction was performed. Some drilling shavings from the weld were chemically decomposed with hydrochloric acid (10:1). To accelerate the process of dis­solution, the solution was heated to close to boiling with an argon cover gas. After the extraction, the fluid was filtered through a Millipore filtration set-up (0.025 micron pore size). The filtered particles were examined using a Philips x-ray dif-fractometer and spectrometer.

Results and Discussion

Chemical Behavior of the Fluxes

The final weld composition for a par­

ticular element is made up of contribu­tions from the welding electrode, the base metal, and the weld metal-flux reac­tions. The chemical behavior will be expressed in terms of a quantity called delta (A); this is the difference between the composition of a particular element determined analytically and the amount of that element which could be present in the weld if no elemental transfer from the weld pool to the flux or vice versa had occurred. The effects of dilution with the base metal were also taken into account. A negative value of delta means that, during welding, a particular element has been transferred from the weld metal to the slag; a positive delta would indicate the element going from the molten flux to the weld metal.

The chemical composition and the ele­mental delta values of the first set of welds are presented in Tables 4A, B and 5. The delta quantities of the elements were then plotted as a function of flux composition. Negative delta values were seen for manganese and niobium —Figs. 5 and 6. These delta quantities were also insensitive to flux compositional changes. This observation implies the activities of manganese and niobium are nearly con­stant for these systems.

High temperature activity data of MnO (Ref. 28) in the MnO-CaO-Si02 system indicate that the MnO activity is almost constant with 40% Si02 content in the flux providing the conditions of a con­stant delta manganese in the weld metal for that system. Comparing the two curves in Fig. 5, the same behavior would be expected for welds made with CaF2" CaO-Si02 fluxes, resulting in the constant delta manganese. A constant delta quan­tity in this case also suggests that small flux variation will not significantly alter the weld metal chemical composition.

The delta sulfur quantity is plotted as a function of the ratio CaO/Si02 in Fig. 7. The deviations were all negative, indicat­ing weld pool desulfurization. With increasing CaO-Si02, an increasing sulfur loss was observed. This is related to the oxygen potential of the molten flux in contact with the molten weld pool and can be explained by the desulfurization reaction:

and

5(metal) + f<?(metal) + O (slag) —

S (slag) + EeO(siag) 0)

WELDING RESEARCH SUPPLEMENT 1143-s

Table 4A—Chemical Composition of the 28 Welds Produced Using a Heat Input of 1.87 k j /mm, Wt-%,W

Flux O Mn Nb Cr Ni Al Mo Cu

1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28

0.10 0.10 0.11 0.11 0.10 0.09 0.12 0.10 0.10

— 0.11 0.10 0.10 0.10 0.10 0.11 0.09 0.10 0.11 0.09 0.11 0.12 0.10 0.09 0.09 0.10 0.09 0.09

122 140 58

183 215 87

206 222 154 229 266 234 231 213 88

326 265 248 107 301 268 215 250 318 293 154 353 457

116 91 92

101 104 96 92

102 100 116 78

87 86 83

117 81 95

195 93 85 98

160 82 97

114 100 97 98

0.97 0.98 1.03 0.99 0.87 0.92 0.87 0.96 1.04 0.75 0.92 0.96 1.01 1.05 0.97 0.77 0.87 1.00 1.03 0.93 0.90 1.00 1.00 0.92 0.98 0.98 0.89 0.98

0.47 0.52 0.36 0.49 0.47 0.31 0.59 0.47 0.43 0.35 0.56 0.51 0.46 0.43 0.20 0.61 0.48 0.49 0.39 0.49 0.49 0.53 0.34 0.47 0.43 0.38 0.44 0.46

0.016 0.018 0.022 0.020 0.019 0.011 0.021 0.022 0.014

— 0.022 0.022 0.021 0.013 0.014 0.023 0.021 0.015 0.011 0.022 0.025 0.012 0.011 0.022 0.020 0.017 0.022 0.019

0.030 0.031 0.032 0.028 0.025 0.027 0.026 0.026 0.030

— 0.025 0.030 0.028 0.029 0.020 0.028 0.027 0.030 0.031 0.028 0.031 0.027 0.027 0.027 0.030 0.030 0.029 0.028

0.012 0.014 0.022 0.017 0.008 0.011 0.013 0.014 0.016 0.025 0.015 0.019 0.017 0.019 0.008 0.014 0.013 0.015 0.015 0.015 0.015 0.014 0.012 0.016 0.016 0.015 0.015 0.019

0.06 0.07 0.06 0.07 0.06 0.06 0.06 0.06 0.07

-0.06 0.07 0.07 0.07 0.05 0.07 0.06 0.07 0.07 0.08 0.08 0.07 0.07 0.07 0.07 0.06 0.06 0.07

0.05 0.05 0.05 0.06 0.05 0.05 0.05 0.06 0.06

— 0.06 0.07 0.06 0.06 0.04 0.06 0.05 0.07 0.06 0.06 0.09 0.07 0.05 0.06 0.06 0.05 0.05 0.05

0.008 0.003 0.010 0.005 0.003 0.004 0.003 0.004 0.004

— 0.003 0.005 0.005 0.006 0.003 0.003 0.003 0.005 0.003 0.004 0.005 0.006 0.004 0.004 0.004 0.003 0.005 0.006

0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02

00.02 0.03 0.02 0.02 0.02 0.02 0.02 0.02 0.02

0.11 0.14 0.06 0.14 0.09 0.07 0.11 0.12 0.15

— 0.13 0.10 0.09 0.10 0.14 0.10 0.11 0.11 0.10 0.10 0.08 0.08 0.07 0.08 0.07 0.07 0.05 0.06

(a) Oxygen (O) and nitrogen (N) given in ppm.

K' = (%S) [%S]

(%FeO), (2)

w h e r e K ' is k n o w n as the equi l ibr ium index of the sulfur react ion, (%S)/[%S] is

the sulfur distr ibut ion ratio b e t w e e n the metal and the slag, and (%FeO) t is the total i ron ox ide content in the slag.

The increase of K ' , w i t h increasing basicity in Fig. 8, indicates that the sulfur is

being par t i t ioned m o r e to the slag. This translates into a negative delta sulfur quant i ty and agrees w i t h the t rend shown in Fig. 7. By increasing C a O activi­ty in the f lux, the sulfur wi l l be r e m o v e d

Table 4B—Flux Compositions for the Various Weld Number Designations

Table 5—Delta Quantities of Manganese, Silicon, Niobium, Sulfur for the 28 Welds Produced Using a Heat Input of 1.87 KJ/mm, %

Flux Number

1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28

%CaF2

100 90 90 80 80 80 70 70 70 70 60 60 60 60 60 50 50 50 50 40 40 40 40 30 30 30 20 10

%CaO

0 0

10 0

10 20 0

10 20 30

0 10 20 30 40 10 20 30 40 20 30 40 50 30 40 50 40 50

°»Si02

0 10 0

20 10 0

30 20 10 0

40 30 20 10 0

40 30 20 10 40 30 20 10 40 30 20 40 40

Wei

1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28

AMn ASi ANb

-0 .23 -0 .29 -0 .23 -0 .28 -0 .38 -0 .35 -0 .43 -0 .33 -0 .24 -0 .49 -0 .38 -0 .34 -0 .28 -0 .22 -0 .28 -0 .54 -0 .43 -0 .30 -0 .25 -0 .36 -0 .36 -0 .26 -0 .26 -0 .39 -0 .31 -0 .31 -0 .40 -0 .30

-0 .03 +0.05 -0 .11 +0.02 - 0 . 0 1 -0 .16 +0.14 -0 .01 -0 .03 -0 .13 +0.11 +0.06

0.00 -0 .04 -0 .18 +0.16 +0.03 +0.04 -0 .07 +0.03 +0.02 +0.05 -0 .13 +0.02 -0 .02 -0 .08 -0 .02

0.00

+0.002 -0 .008 -0 .001 -0.004 -0 .010 -0.010 -0.014 -0.012 -0 .008 +0.008 -0.012 -0 .009 -0 .009 -0.003 -0 .010 -0.015 -0.015 -0.012 -0 .008 -0 .010 -0.005 -0.007 -0 .008 -0.014 -0.010 -0 .010 -0.010 -0.005

AS

-0.013 -0.011 -0.006 -0.008 -0.009 -0.018 -0.005 -0.003 -0.012

0.000 -0.005 -0.004 -0.005 -0.013 -0.013 -0.005 -0.006 -0.012 -0.017 -0.004 -0.004 -0.015 -0.019 -0.004 -0.007 -0.014 -0.005 -0.006

1 4 4 - s l M A Y 1 9 8 5

10

0 8

0.6

0.4

o

i

c

<1

0.2

-0 .2

- 0 .4

-0 .6

-

-

-

1

0

o

o

D

o

1

•

1

o

1 1 —I

4 0 % S i 0 2

MnO-S i0 2 -CaF 2

0 /

o o

C a O - S i O z - CaF2

D / U

D i i i

I

-

p.

t

0.0 30

0.020

"5 0.010 — i

1 ° z ^ - 0.010

- 0 0 2 0

- 0 .030

-l 1 1 1 1 1 1 r

40 % Si02

o

J I I I 1_ J L 0 10 20 30 40 50 60 70 80

CaF2 IN FLUX (w/ol

Z <

0 .020

0.010

0

0.010

0.020

0.0 30

T '

-

-

1

i r

0

i i

T-

0

1

1 1

2 0 %

o u

1 1

!

Si02

O

1

-

-

o

•

-

1

2 0 , in

3 0 FLUX

4 0 ( W / o )

5 0 6 0 20 3 0 4 0

CaF 2 IN

5 0 6 0 70

FLUX (W/ . l

Fig. 5 —Delta manganese plotted as a function of weight-percent CaF2 for two different flux systems

Fig. 6 —Delta niobium plotted as a function of weight-percent CaF2 for different Si02 contents in the CaO-CaF2-SI02 flux system

more readily. This can occur by reducing

the silica content in the flux for the

CaF2-CaO-Si02 system.

Silicon had both positive and negative

delta values (gained or lost) as shown in

- 0 . 0 0 6

•0 .008

01 3

co

•0.010

• 0.0 I 2

• 0.0 I 4

•0 .0 16

• 0.0 I 8

0.020

Fig. 9. When the SiO? content is approxi­mately equal to the CaO content, no silicon transfer was observed. This corre­sponds to the zero delta silicon region (shaded region indicated in Fig. 9). Above

400

300

o u_ 3*

that region (%Si02/%CaO greater than one), there was an increase of silicon in the weld pool indicating that silicon was transferred from the flux to the weld. When %Si02/%CaO is less than one, a

<" "l 200

100

1

TYPE OF DATA A BASIC 0 - H A O.S.M. 0 BASIC 0 - H • EQUILIBRIUM

8 8

0

. 0 A

O

* A6. 0 "**•

, O A

A*° A A

<*

A .

A

A A

A

A

•

•

A

1

CaO S i 0 2

Fig. 7—Delta sulfur plotted as a function of basicity index CaO/Si02

p %CaO + 1.4 x %MoO B " % S i 0 2 + 0.84 x %P205

Fig. 8 — Variation of sulfur equilibrium index (K').

%CaO+ 1.4(%MgQ)

~ %Si02 + 0.84 (%P205) ' Data were collected from various sources using basic open hearth furnaces

WELDING RESEARCH SUPPLEMENT 1145-s

40%Si0 2 4 0 % SI02

CaO 10 20 30 40 50 60 70 80

%CaF2

9 0 CaF,

Fig. 9 —Partial ternary diagram of CaF2-CaO-Si02 system showing the composition regions where the weld metal gains or loses silicon. Notice the region of no silicon change, ASi = 0

0.30 0.70

CaO

0.10

0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 CaF2

Fig. 10- 1450°C (2642 °E) high temperature activity data ofSi02 in the CaF2-CaO-Si02 system (Ref. 30)

silicon loss was seen. The thermodynamic activity data (Ref. 30) shown in Fig. 10 explains the shaded region of no silicon change, since the silicon isoactivity lines lie parallel to this shaded region.

If it is desirable for the composition of the weld to be equal to the base metal composition, manganese and niobium can be introduced into the weld metal by either adjusting the flux composition or by alloy additions to the welding elec-

40%SiO

trodes. The silicon content, in the present case, can be controlled by simply adjust­ing the flux composition. The concept of controlling the weld metal composition by the proper combination of flux and filler metal (i.e., welding electrode) is very important, especially in cases where weld metal chemistry must be kept in narrow compositional ranges.

The results of the oxygen analysis on the weld beads are shown in Fig. 11.

4 0 % Si02

CoO 10 20 30 40 50 60 70 80 90

CaF.

% CaF2

Fig. 11-Partial ternary diagram of CaF2-CaO-Si02 system showing the variation of weld metal oxygen as a function of flux composition

Welds made with silica free fluxes showed the lowest oxygen contents (<100 ppm). Increasing the silica content in the flux, the weld metal oxygen was found to increase. To a lesser degree, an increase in CaO content also increased the oxygen content in the weld metal.

Traditionally, weld metal oxygen and mechanical properties had been related to the basicity index. Tuliani et al. (Ref. 31) showed that higher basicity fluxes resulted in lower weld metal oxygen, sulfur, and silicon. Weld metal toughness was also said to improve with increasing basicity index (Ref. 32). However, it is still questionable whether the parameter basicity index expresses any fundamental relationship. The flux basicity index (B.I.) equation proposed by Tuliani (Ref. 31) is:

B.I. = [CaO + CaF2 + MgO + BaO + Sr O + K 20 + Na20 + Li20 + ¥l (MnO + FeO)]/[Si02 + '/2 (Al203 + T i0 2

+ Zr02) ] (3)

Eagar (Ref. 24) modified this relationship by omitting the CaF2 term. Using the equation developed by Eagar (Ref. 24), the weld metal oxygen contents were plotted as a function of basicity index — Fig. 12. The data points showed large scattering without following the usual trend as indicated by the solid curve proposed by other authors (Refs. 24, 31). However, if the weld metal oxygen con­tent were plotted using all the terms in equation (3), the data points fell smoothly in the band as seen in Fig. 13 and report­ed in the literature (Ref. 24). This compli­cates the interpretation of the relation­ship between the basicity index and weld metal oxygen. It is also known that weld metal oxygen content increases with an increasing heat input.

Microstructure

A definite trend of microstructural vari­ation was observed within the series of weld metals. For the purpose of analysis, the following discussion will be based upon three particular weldments. Figure 14 shows the light micrographs of the three weldments. Their oxygen contents are 350, 250, and 140 ppm, respectively (each containing similar manganese and silicon contents).

The two main features seen in the high oxygen weld metal are grain boundary ferrite and acicular ferrite. The grain boundary ferrite occurred in the form of continuous veining along the prior aus­tenite grain boundaries, and constituted around 20 to 30% of the whole struc­ture—Fig. 14A.

By reducing the oxygen in the weld metal to 250 ppm, a refining of the microstructure was seen. There was a significant decrease of grain boundary ferrite veining. Quantitative metallogra-

146-s | M A Y 1985

1000

E Q. O.

UJ CD >-X o

< i -UJ 5

1 0 0 0

8 0 0

6 0 0

4 0 0

2 0 0

af ter T Eager (24)

o Ca0 -CaF 2 -S i0 2

0*->- V- 1

BASICITY INDEX

Fig. 12 - Variation of weld metal oxygen content with basicity index omitting the CaF2 term

E a. a.

2 LU

o O

LU C5 >-X o

800 -

6 0 0

4 0 0

LU

o _ l Ul

5 200

( ref .24)

T%-O O

V. I

BASICITY INDEX

Fig. 13 — Variation of weld metal oxygen content with basicity index including the CaF2 term

phy s h o w e d that acicular ferr i te consti­tu ted more than 90% of the microstruc­ture—Fig. 14B. Decreasing the oxygen fur ther to 140 p p m , a finer acicular ferri te was not ob ta ined ; instead long and aligned ferr i te laths w e r e f o r m e d w i t h an aspect ratio o f approximate ly 1 0 : 1 to 1 2 : 1 . Carbide precip i tat ion cou ld be seen b e t w e e n the ferr i te plates and we re thus classified as bainite —Fig. 14C.

The results dep ic ted in Fig. 14 suggest that, for a given flux system, there is an o p t i m u m level of we ld metal oxygen produc ing the max imum amount of acic­ular ferr i te. In the CaF 2 -CaO-Si0 2 flux system, the o p t i m u m w e l d metal oxygen

content ranged f r o m 200-250 p p m corre­sponding to a microstructure approx i ­mately 90% acicular ferr i te.

Physical Properties

The w e l d penet ra t ion, interfacial ten ­sions, and arc stabilities w e r e quant i ta­t ively measured and analyzed. The results have been repo r ted in a prev ious paper (Ref. 27), and are not discussed here.

Recommended Fluxes from the Fluorspar-Lime-Silica System

From the results of the first set o f we lds , impor tant conclusions can be

reached w i th respect t o the applicabil i ty o f CaF2-CaO-Si02 flux system. There is a flux composi t ional region wh i ch s h o w e d very g o o d per fo rmance dur ing we ld ing ; this appears as the shaded area in Fig. 15. Fluxes o n the CaF2-CaO side of the terna­ry system w o u l d not be advisable because of its lack of a silicate ne two rk .

W e l d poo l p ro tec t ion and the w e l d surface quali ty w e r e observed to be poor . The region of the CaF2-Si02 side should be avo ided because of the t w o liquids region. Flux behavior may be unpredictable. We lds made in the upper left region should be avo ided in the we ld ing of HSLA steels because the

Fig. 14 — Microstructural variation with weld metal oxygen content: A - 350 ppm of oxygen, mixed microstructure of grain boundary ferrite and acicular ferrite; B- 260 ppm of oxygen, microstructure predominantly acicular ferrite; C- 107 ppm of oxygen, microstructure predominantly upper bainite

W E L D I N G RESEARCH SUPPLEMENT 1147-s

4 0 % S i 0 2

2 L I Q U I D S R E G I O N

C a O 6 0 7 0 8 0 9 0

%CaF 0

Fig. 15 — Partial ternary diagram of CaF2-CaO-Si02 system showing as the shaded region the flux compositions which resulted in good quality welds

"h ighe r " w e l d metal oxygen content ( low c o m p a r e d to many welds made w i th o ther flux systems) w o u l d yield a larger amount o f grain boundary ferr i te and side plate ferri te than desired fo r the max imum toughness. Regions to the left side ough t t o be avo ided because o f the steep ridges in the melt ing temperature of the fluxes.

Toughness

From the indicated region in Fig. 9, nine fluxes cou ld be selected along the zero delta silicon quant i ty line for the second set of welds. The chemical compos i t ion o f the second pass (the un tempered bead) of this second set o f we ldments was de te rmined and presented in Table 6.

This cr i ter ion of flux selection p r o v e d adequate, because w i t h the except ion of oxygen, the welds s h o w e d similar c o m ­posit ion. A higher heat input of 3.3 k j / m m (83.8 k j / in . ) was uti l ized to achieve deeper penetrat ion such that the Charpy specimens w e r e made entirely o f we ld metal. Due to the s lower cool ing rate of these we ldments , the microstructural variat ion was not as clear as the l ower heat input we lds . High magnif icat ion light micrographs s h o w e d that slightly f iner acicular ferr i te was associated w i t h l owe r

oxygen welds. Howeve r , no significant d i f ference in the microstructure cou ld be found to distinguish the welds. In spite of the similar microstructure o f the di f ferent welds, their toughness measurements s h o w e d qui te di f ferent results. The upper shelf energies and the duct i le br i t t le t ran­sition temperatures (DBTT) of the w e l d ­ments are g iven in Table 7.

W i t h an increasing w e l d metal oxygen content , the upper shelf energy absorbed at f racture decreased. This rela­t ionship is shown in Fig. 16. In the upper shelf region, the Charpy specimens frac­ture by duct i le failure m o d e , absorbing a max imum amount of energy wh i ch is related to the magni tude and distr ibut ion of strain at notch roo t . It is mainly inf lu­

enced by the presence of inclusions (populat ion, spatial d istr ibut ion, and shape). Due to the very l o w solubility of oxygen in BCC i ron, the total amount of oxygen in solut ion is negligible. At r o o m tempera ture , oxygen exists in the c o m ­b ined f o r m as inclusions (e.g., oxides, silicates, aluminates, oxysulf ides, etc.). Based o n this argument , we ld metal oxy­gen content cou ld be used as an estimate indicating the amount o f inclusions in the we ld metal as a correlat ing parameter. Fundamentally, h o w e v e r , inclusion c o n ­tent should be the m o r e appropr ia te variable t o be compared .

Transmission e lect ron micrographs o f selected carbon extract ion replicas are shown in Fig. 17. High we ld metal oxygen

Table 7—The Upper Shelf Energies (USE) and the 100 Joules Transition Temperatures (TT10oj) of the Nine Welds Produced Using a Heat Input of 3.3 k j /mm

Weld

1 2 3 4 5 6 7

Flux

5 8 9

13 17 18 21 25 27

USE, joules

215 220 195 190 200 195 190 185 165

TTiooj,°C

- 3 - 3 + 15 + 18 - 3 - 1 0 +20 +6 + 18

Table 6—Chemical Composition of the Second Set of Weldments—Heat Input 3.3 k j /mm, %<a>

Weld

1 2 3 4 5 6 7

Flux

5 8 9

13 17 18 21 25 27

0.085 0.081 0.088 0.085 0.079 0.086 0.085 0.088 0.079

Mn

1.20 1.09 1.27 1.17 1.11 1.25 1.14 1.19 1.12

Si

0.40 0.46 0.41 0.46 0.47 0.42 0.44 0.44 0.51

Nb

0.038 0.022 0.023 0.021 0.022 0.025 0.022 0.023 0.019

Cu

0.19 0.12 0.14 0.12 0.12 0.13 0.12 0.12 0.12

O

205 240 195 391 373 264 345 338 432

N

108 80 74 86 72 88 76 81 75

(a) Oxygen (O) and nitrogen (N) given in ppm.

1 4 8 - s | M A Y 1985

~ 2 2 0

> 2 0 0 CD rr Ul z Ul

I cn

CE Ul

Q.

= >

180

160

E70S-3 Lukens Frostline

3 . 2 k J / m m , 16mm plate *

• • iai iamr*^"Tt ^ aa> ^ Jfc«* _

* * <w»* -Si

180 260 3 4 0 4 2 0 500

OXYGEN ( p p m )

Fig. 76 - Variation of the upper shelf energies (USE) of the Charpy specimens with weld metal oxygen content showing a decreasing USE with increasing oxygen

samples s h o w e d higher inclusion content than the l ower oxygen welds. Finer part i ­cles and larger size variat ion w e r e also seen to be associated w i t h the higher oxygen welds. Particles' sizes ranged f r o m 0.05 to 1 m ic ron , * and various geometr ical shapes we re also observed. This all seems to suppor t , t o a certain extent , the prev ious statement that the

*1 micron = 10~3 mm = 0.00004 in.

®

If- j ** * p

Fig. 17 —Carbon extraction replicas of weld metal showing the effects of weld metal oxy­gen content on the distribution of inclusions: A — 432 ppm oxygen weld metal; B — 205 ppm oxygen weld metal

w e l d metal oxygen can be used to indi­cate the amount of inclusions in the w e l d metal for correlat ion purposes.

To fur ther analyze the inclusions effect on the w e l d metal toughness, the highest and lowest w e l d metal oxygen samples w e r e chosen. Their Charpy V-no tch energy transit ion curves as a funct ion o f temperatures are shown in Fig. 18. Cor re­sponding scanning electron micrographs of the f racture surfaces at the upper shelf and lower shelf regions are s h o w n in Figs. 19 and 20, respectively.

At the upper shelf regions, dimples cou ld be seen in the f racture surface of b o t h the samples, indicating ducti le fail­ure by the mic rovo id f racture mecha­nism. The di f ference b e t w e e n the t w o samples is not so much in the number o f

Fig. 19 —Scanning electron micrographs of the Charpy specimens fracture surfaces at 97 "C (206.8 "F) high testing temperatures showing ductile type failure with dimples: A — 432 ppm oxygen weld metal; B - 205 ppm oxygen weld metal

dimples, but in the amount of strain associated w i t h each dimple. The micro-vo id f racture mechanism can be descr ibed as being c o m p o s e d of three stages. The first stage is the initiation of a m ic rovo id wh ich occurs by the decohe-sion of a particle f r o m the matrix material, o r by the f racture o f a second phase material. The second stage o f the micro-vo id mode l is the g r o w t h o f the d imple wh ich is independent o f inclusion part i -

• - 205 ppm OXYGEN

•V - 432 ppm OXYGEN

0 2 0 4 0

TEMPERATURE ( » C )

Fig. 18-Charpy V-notch energy curves as a function of temperature for the highest and lowest weld metal oxygen content welds produced with 20%CaF2-40%CaO-40%SiO2 and 80%CaF2-10%CaO-10%SiO2 fluxes

WELDING RESEARCH SUPPLEMENT 1149-s

Fig. 20 —Scanning electron micrographs of the Charpy specimens fracture surfaces at —55°C (—67°F) low testing temperatures showing cleavage type failure: A - 432 ppm oxygen weld metal; B — 205 ppm oxygen weld metal

cles. The third stage is the coalescence of the individual microvoids.

As the material between the micro-voids becomes necked, the process depends essentially on the properties of the matrix only. Due to the many fine particles in the high oxygen content welds, the stress-strain behavior of the matrix material between the larger inclu­sions is modified. The total amount of uniform strain to failure is reduced. Therefore, the material with more parti­cles (high oxygen content) has a lower total uniform strain to failure, and not much strain will be associated with the ductile dimples.

This is reflected in a lower absorbed energy at failure —Fig. 19A. In the lower oxygen weldments, the inclusions con­tent is lower. Fewer microvoid nucleation sites would be expected, requiring more energy to fracture the material. This results in a large degree of plastic defor­mation at the ductile dimples —Fig. 19B.

On the other hand, brittle failure in the lower shelf region is more microstructure dependent. It is influenced mainly by the grain size and precipitation hardening of the matrix. An examination of the two fracture surfaces revealed flat fracture facets — Fig. 20. The size of these facets is at the order of 20-30 microns, corre­sponding to the allotriomorphic ferrite grain size, suggesting that the cracks tend to follow the path which gives the largest uninterrupted route. This is also observed by Ebden and Weatherly (Ref. 33).

The energy difference observed between the high and low oxygen welds may be explained by the presence of small regions of ductility between the flat fracture facets. They were more fre­quently observed in the low oxygen welds. This indicates that the crack has to change directions during propagation throughout the material, and while doing so, a small amount of ductility is seen. This is related to the finer grain size (thus, more grain boundaries) observed in the lower oxygen weld metal.

A minimum transition temperature was found in the range of 200-300 ppm of oxygen. This agrees with the upper shelf energy data shown previously. The same behavior has been observed and report­ed by Devillers et al. (Ref. 23). When analyzing these toughness data and com­paring with the metallographic and frac-tographic results, the optimum weld met­al oxygen content for the CaF2-CaO-Si02

system was determined to be in the range of 200-300 ppm.

Particle Extraction

In an attempt to study the chemical and crystalline nature of the particles, an acid dissolution technique was used. The collected residue was analyzed using x-ray spectrometry and diffractometry. Peaks for several elements including man­ganese, silicon and niobium were seen. However, no diffraction pattern could be obtained. This seems to suggest that the inclusion particles could be amorphous even though they showed definite facets. The negative results using the x-ray tech­niques could also be due to the extremely small quantity of residue obtained being insufficient for the characterization.

Conclusions

The results of this investigation led to the following conclusions.

1. The CaF2-Ca0-Si02 flux system is found to produce good quality niobium microalloyed HSLA steel weldments with very low oxygen content (100-460 ppm).

2. Manganese and niobium showed negative delta values during welding, indicating a loss from the weld pool to the slag.

3. There is a region of zero delta silicon where weld pool chemistry can be easily controlled.

4. The negative delta sulfur values indicated that the CaF2-CaO-Si02 flux system was effective in sulfur control.

5. The optimum weld metal oxygen content for this flux system ranged from 200-300 ppm, corresponding to a micro-structure of approximately 90% acicular ferrite.

6. The refinement of the weld metal microstructure due to oxygen decrease

resulted in an improvement of the weld metal toughness.

7. High weld metal oxygen welds showed high inclusion contents, but the inclusions were observed to be finer in size.

8. In the lower oxygen content welds, a larger strain was found to associate with the upper shelf ductile dimples lead­ing to a higher energy absorbed at frac­ture.

9. The lower shelf failure was charac­terized by cleavage fracture with small regions of ductility between the flat frac­ture facets which account for the energy difference between high oxygen and low oxygen welds.

Acknowledgments

The authors acknowledge the research support of the United States Army Research Office, the fellowship support of one of the authors by the Conselho Nacional De Pesquisa e Desenvolvimento (Brasil), and the materials and equipment support of Lukens Steel Company and Hobart Brothers Company.

References

1. Cohen, M„ and Hansen, S. S. 1979. Microstructural control in microalloyed steels. ASTM STP 672, pp. 34-52.

2. Meyer, L, and DeBoer, H. 1977. (Jan). HSLA plate metallurgy, alloying: normalizing, controlled rolling. Journal of Metals: 17-23.

3. DeArdo, A. J., and Brown, E. 1977. (Jan). Hot rolling behavior of austenite microalloyed with vanadium and nitrogen, journal of Metals: 26-29.

4. Boyd, J. D. 1977. (March). Microstructure and toughness in high-strength linepipe steels. Symposium on Toughness Characterization and Specifications for HSLA and Structural Steels, pp. 216-235. Atlanta, Georgia: Metallur­gical Society of AIME.

5. Committee of Welding Metallurgy of Japan Welding Society. 1983. Classification of microstructures in low-C low alloy steel weld metal and terminology. IIW DOC IX-1282-83.

6. Dolby, R. E. 1976. Factors controlling weld toughness — the present position, part II—weld metals. The Welding Institute Research report 14/1976/M.

7. Clover, A. G., McGrath, J. T., and Eaton, N. F. 1977. Fracture toughness of submerged arc weld metal. Symposium on Toughness Characterization and Specifications for HSLA and Structural Steels, pp. 143-160. Atlanta, Georgia: Metallurgical Society of AIME.

8. Glover, A. G., McGrath, J. T„ Tinkler, M. J. and Weatherly, G. C. 1977. The influence of cooling rate and composition on weld metal microstructure in a C/Mn and HSLA steel. Welding lournal 56 (9): 267-s to 273-s.

9. Abson, D. )., Dolby, R. E., and Hart, P. H. M. 1978. The role of non-metallic inclusions in ferrite nucleations in carbon steel weld metals. Intl. Conf. on Trends in Steel and Consumables for Welding, paper 25. The Welding Institute, London.

10. Cochrane, R. C, and Kirkwood, P. R. 1978. The effect of oxygen on weld metal microstructure. Intl. Conf. on Trends in Steels

150-s|MAY 1985

and Consumables for Welding. The Welding Institute, London.

11. North, T. H., Bell, H. B., Koukabi, A., and Craig, I. 1979. Notch toughness of low oxygen content submerged arc deposits. Welding lournal 58 (6): 343-s to 353-s.

12. Ricks, R. A., Howell, P. R., and Barrite, G. J. 1982. The nature of acicular ferrite in HSLA steel weld metals, lournal of Materials Science 17: 732-740.

13. Kayali, E. S„ Corbett, J. M., and Kerr, H. W. 1983. Observations on inclusions and acic­ular ferrite nucleation in submerged arc HSLA welds, /ournal of Materials Science Letters: 2: 123-128.

14. Keville, B. R., and Cochrane, R. C. 1981. The influence of inclusion morphology on the microstructure and toughness of submerged arc weldments. Conf. Proceedings, "Steels for line pipe and pipeline fittings." London: The Metals Society.

15. Ito, Y„ and Nakanishi, M. 1976. Study on Charpy impact properties of weld metal with submerged arc welding. The Sumitomo Search 15:42.

16. Kikuta, Y., Araki, T., Honda, K., and Sakahira, S. 1980 (Oct.). The metallurgical properties of electroslag weld metal using CeF3 addition wire. Conf. Proceedings, Weld­ing research in the 1980's, Session B, pp. 131-136. Osaka, Japan: Japan Welding Research Institute.

17. Indacochea, J. E., and Olson, D. L. 1983. (Dec). Relationship of weld metal microstruc­ture and penetration to weld metal oxygen content. /. Materials for Energy Systems 5 (3): 139-148.

18. Harrison, P. L., and Farrar, R. A. (1981). Influence of oxygen-rich inclusions on the

austenite to ferrite phase transformation in high-strength low-alloy (HSLA) steel weld met­als, lournal of Materials Science 16: 2218-2226.

19. Ferrante, M „ and Farrar, R. A. 1982. The role of oxygen rich inclusions in determining the microstructure of weld metal deposits. lournal of Materials Science 17: 3293-3298.

20. Pargeter, R. J. 1981. Investigation of submerged arc weld metal inclusions. The Welding Institute research report no. 151/ 1981.

21. Keville, B. R. 1983. Preliminary observa­tions of the type, shape, and distribution of inclusions found in submerged arc weldments. Welding lournal 62 (9): 253-s to 260-s.

22. Ahlblom, B., Bergstrom, U., Hannerz, N. E., and Werlefors, I. 1983 (Nov). Influence of welding parameters on nitrogen content and microstructure of submerged-arc weld metal. First Intl. Conf. on The Effect of Residual, Impurity, and Microalloying Elements on Weld­ability and Weld Properties. The Welding Insti­tute, London.

23. Devillers, L., Kaplan, D „ Marandet, B., Ribes, A., and Riboud, P. V. 1983 (Nov.). The effects of low concentrations of source ele­ments on the toughness of submerged-arc welded C-Mn steel welds. Conf. Proc. "The Effects of Residual, Impurity, and Microalloying Elements on Weldability and Welded Proper­ties," P11-P1-11. London: The Welding Insti­tute.

24. Eagar, T. W. 1978. Sources of weld metal oxygen contamination during sub­merged arc welding. Welding journal 57 (3): 76-s to 80-s.

25. Craig, I., North, T. H., and Bell, H. B. 1978. Deoxidation during submerged arc

welding. Int. Conf. on "Trends in Steels and Consumables for Welding." Paper 21, pp. 249-263. London: The Welding Institute.

26. Koukabi, A., North, T. H., and Bell, H. B. 1978. Flux formulation, sulphur, oxygen and rare-earth additions in submerged arc welding. Intl. Conf. on "Trends in Steels and Consuma­bles for Welding," Paper 22. London: The Welding Institute.

27. Liu, S., Dallam, C. B„ and Olson, D. L. 1982 (Mar). Performance of the CaF2CaO-Si02 system as a submerged arc welding flux for a niobium based HSLA steel. AWS, ASM, and ORNL Joint Intl. Conf. on Welding Tech­nology for Energy Applications, Gatlinburg, Tenn.

28. Abraham, K. P., Davies, M. W., and Richardson, F. D. 1960. journal of Iron and Steel Institute 196: 82-89.

29. U.S. Steel Corporation. 1971. The mak­ing, shaping, and treating of steels. 9th edition, pp. 392.

30. Sommerville, I. D., and Kay, D. A. R. 1971. Activity determination in the CaF2-CaO-Si02 system at 1450C. Met. Trans 2 (6): 1727-1732.

31. Tuliani, S. S., Boniszewski, T., and Eaton, N. F. 1969 (Aug.). Notch toughness of com­mercial submerged arc weld metal. Welding and Metal Fabrication: 27.

32. Garland, I. G., and Kirkwood, P. R. 1976 (Apr.) A reappraisal of the relationship between flux basicity and mechanical proper­ties in submerged-arc welding. Welding and Metal Fabrication: 217:224.

33. Ebden, ]. R., and Weatherly, G. C. (1983). Weld metal fracture modes in a series of HSLA steel weldments. Canadian Metallurgi­cal Quarterly 22 (2): 149-155.

WRC Bulletin 298 September 1984

Long-Range Plan for Pressure-Vessel Research—Seventh Edition By the Pressure Vessel Research Committee

Every three years, the PVRC Long-Range Plan is up-dated. The Sixth Edition was widely distributed for review and comment. Up-dated problem areas have been suggested by ASME, API, EPRI and other organizations. Most of the problems in the Sixth Edition have been modified to meet current needs, and a number of new problems have been added to this Seventh Edition.

The list of "PVRC Research Problems" is comprised of 58 research topics, divided into three groups relating to the three divisions of PVRC; i.e., materials, design and fabrication. Each project is outlined briefly in a project description, giving the title, statement of problem and objectives, current status and action proposed.

Because of budget limitations, PVRC will not be able to investigate all of these problems in the foreseeable future. Therefore, the cooperation and efforts of other groups in studying these areas is invited. If work is planned on one of the problems, PVRC should be informed in order to avoid duplication.

Publication of this bulletin was sponsored by the Pressure Vessel Research Committee of the Welding Research Council. The price of WRC Bulletin 298 is $14.00 per copy, plus $5.00 for postage and handling. Orders should be sent with payment to the Welding Research Council, Room 1301, 345 E. 47th St., New York, NY 10017.

WELDING RESEARCH SUPPLEMENT 1151-s

AN INVITATION TO

AUTHORS Gentlemen:

The American Welding Society will hold its 67th Annual Convention and AWS Welding Show in Atlanta, Georgia, during April 13-18, 1986. One of the most important events of our 67th Annual Convention will be its Professional Program.

It is indeed a pleasure to invite you as Authors to be participants in the Professional Program of our 67th Annual Convention. On this occasion, the Society is offering an opportunity to Authors to bring the results of outstanding work on their part to the attention of our entire membership, the welding industry, and the nation's metalworking industries.

To this end, the Society's Technical Papers Committee will be happy to receive your application for participation in our 67th Annual Convention Professional Program; the Committee is inviting 500-word summaries for, basically, two categories of papers:

1. Applied Technology — unusual industrial or field applications of welding, new process or significant equipment developments, surfacing, unique welding case histories, education, safety and health, cost studies —also related topics such as nondestructive testing of weldments as well as maintenance and repair.

2. Research Oriented — results of significant laboratory research and/or development projects, welding metallurgy, weldability studies, weld cracking or fracture, new test methods, arc physics —also related topics such as fracture mechanics.

To apply for participation in our 67th Annual Convention Professional Program, please complete both sides of the Author Application Form on the facing page. Also, please prepare a 500 to 1,000 word summary of what you intend to say in your paper and mail with the completed form to AWS.

The Technical Papers Committee will screen all Author applications and summaries (also the manuscripts of completed papers if included), and we expect to notify Authors during late October or early November concerning acceptance. Completed Author Application Forms and accompanying 500-word summaries must be mailed by August 1, 1985, to ensure consideration for the 67th Annual Convention Professional Program.

Please note that, as it screens Author Applications and summaries, the Technical Papers Committee will be looking for: (1) the newness of information and need for the information in its field, (2) technical accuracy, (3) clarity of presentation, and (4) adaptability for oral presentation (a paper consisting basically of complicated tables would not be suitable for oral presentation).

Sincerely yours,

f^OD.lC^^L

P. W. Ramsey Executive Director

May 1, 1985

AMERICAN WELDING SOCIETY P.O. Box 351040, Miami Florida 33135

152-s|MAY 1985

AUTHOR APPLICATION FORM - F O R -

67th Annual AWS Convention Atlanta, Georgia, April 13-18,1986

COMPLETE BOTH SIDES AND MAIL WITH500-1000 WORD

SUMMARY TO:

American Welding Society P.O. Box 351040

Miami, Florida 33135

Date Mailed

Author's Name Check how addressed: Mr. • Ms. D

Title or Position Dr. • Other

Company or Organization

Mailing Address

City State Zip Telephone: Area Number

If there are to be joint

Authors, give name(s)

of other author(s)

Name

Address

Name

Address

PROPOSED TITLE (10 words or less):

SUBJECT CLASSIFICATION: Classify your paper by placing a check mark in the two appropriate boxes:

Applied Technology • Research Oriented • Other • /An Original Contribution • A Review • Progress Report •

SUMMARY: • Type, double spaced, a summary of not less than 500—but preferably not more than 1000 — words and attach to this

form. • Be sure to give sufficient information to enable the Technical Papers Committee to obtain a clear idea of content of the

proposed paper; confine background to about 100 words. Also be sure to emphasize Results and Conclusions since this material, together with information on what is NEW, will have a very important bearing on the decision of the Technical Papers Committee.

• If complete manuscript is available, in addition to summary, please attach two copies of this form. • Application Form and summary must be postmarked not later than August 1, 1985, to ensure consideration for inclusion

on the 67th Annual AWS Convention Professional Program.

MANUSCRIPT DEADLINES: • All manuscripts shouldbe in the hands of the Technical Papers Committee not later than March 1,1986. If received prior to

November 30, 1985, every effort will be made to publish them in advance of meeting. • It is expected that the Committee's selections will be announced sometime in October or November 1985. • If your paper is made a part of the program, which of the following manuscript deadlines will you be able to meet?

November 30, 1985 D January 31, 1986 • March 1, 1986 •

PRESENTATION A N D PUBLICATION OF PAPERS: • Has material in this paper been previously presented in meeting or published?

YesD N o D When?

Where?

• Following presentation at the 67th Annual AWS Convention, would you accept invitations to present this paper before , WS Sections? Yes • No D

• Papers accepted for presentation become the property of the Society with original publication rights assigned to the Welding Journal.

RETURN TO AWS HEADQUARTERS. MUST BE POSTMARKEDUOl LATER THAN AUGUST 1,1985, TO ENSURE CONSIDERATION.

Author's Signature

Author(s) of Proposed Paper:

SUMMARY CODE

(For AWS Headquarters Use Only)

Title of Paper (may be abbreviated)

TO AUTHOR(S): Please briefly answer questions in spaces provided below—also please note that answers given below are required in addition to 500 to 1,000 word Summary. Please use typewriter.

(1) WHY WAS THE WORK DONE?

(2) WHAT WAS DONE?

(3) WHAT WAS FOUND?

(4) WHAT ARE YOUR MOST IMPORTANT CONCLUSIONS?

(5) TO WHOM DO YOU THINK YOUR PAPER WILL BE IMPORTANT? HOW?

OTHER COMMENT (Optional)

RETURN TO AWS HEADQUARTERS. MUST BE POSTMARKED NOT LATER THAN AUGUST 1, 1985, TO ENSURE CONSIDERATION.

Be sure to attach 500-1,000 word Summary when mailing completed Author's Application Form to AWS Headquarters.