Effect of Electrochemical Reactions on
Submerged Arc Weld Metal Compositions
Weld metal composition is controlled by chemical reactions
in four separate areas during welding
BY J. H. KIM, R. H. FROST, D. L. OLSON AND M. BLANDER
ABSTRACT. The purpose of this work is to
investigate the relative influence of electrochemical and thermochemical reactions on the weld metal chemistry in a direct current submerged arc welding process. Chemical analyses were carried out
on the melted electrode tips, the detached droplets and the weld metal for
both electrode-positive (reverse) polarity
where the welding wire is anodic and
electrode-negative
(straight)
polarity
where the welding wire is cathodic. The
results suggest that both thermochemical
and electrochemical reactions are important in altering the composition of the
weld metal. The anodic electrochemical
reactions include the oxidation of iron and
alloy elements, and the discharge and
pickup of oxygen anions from the molten
flux. Cathodic electrochemical reactions
include the reduction of iron and alloy elements from the flux to the metal phase,
and the refining of oxygen. Thermochemical reactions occur and move the overall
composition toward the thermochemical
equilibrium.
Introduction
The submerged arc welding process
uses a protective flux cover with a consumable welding wire. The welding current is carried largely by the submerged
arc and to some extent by conduction in
the molten flux layer. Information on the
interaction between slag and weld metal
is significant for a better understanding of
the arc welding process. Although slag/
metal reactions are of importance in many
fusion welding processes and are often
referred to in the literature, comparatively
/. H. KIM, R. H FROST and D. L. OLSON are with
the Center for Welding Research, Department
of Metallurgical Engineering, Colorado School
of Mines, Golden, Colo. M. BLANDER is with
Argonne National Laboratory, Chemical Technology Division/Materials Science and Technology Program, Argonne, 111.
Paper presented at the 68th Annual AWS
Meeting, held March 22-27, 1987, in Chicago,
III.
446-s | DECEMBER 1990
little is known about the mechanisms that
control them.
The chemistry in steel making tends to
be dominated by thermochemical reactions because they generally involve large
surface areas, low current densities and
alternating current power. But a direct
current submerged arc welding process
involves small surface areas and high current densities. Therefore, it is expected
that electrochemical reactions, as well as
thermochemical reactions, will exert a significant influence over the final chemistry
of the weld.
The overall composition of the weld is
controlled by the composition of the
metal droplets that enter the weld pool
and by the amount of dilution of the weld
pool by the base plate. These compositions are controlled by chemical reactions
at four separate zones: the melted electrode tip, the detached droplet, the hot
weld pool immediately below the arc, and
the cooling and solidifying weld pool behind the arc. Thermochemical reactions
occur at all four zones and move the
chemistry in the direction of chemical
equilibrium. But electrochemical reactions
occur only at the melted electrode tip and
in the hot weld pool immediately below
the arc. The electrochemical reactions
would result from ionic conduction of a
portion of the welding current through
the molten slag layer. Therefore, only
thermochemical reactions are expected
to occur at the surface of the detached
KEY W O R D S
Submerged Arc Welding
Weld Metal Composition
Electrochemical Reaction
Thermochemical Reaction
Chemical Analyses
SAW
Alloy Elements
Flux Composition
Electrode Composition
Welding Speed Effect
droplets and solidifying weld pool since
these parts are no longer carrying current.
The most important chemical considerations for submerged arc welding include
the control of oxygen, oxidation losses of
alloy elements, and the pickup of undesirable elements from the slag. A number of
investigations have been made concerning the interaction between the slag and
the metal in submerged arc welding of
steel. Most of these investigations were
based solely on thermochemical reactions
(Refs. 1-9), although lately a few investigators have considered the electrochemical reactions that occur when direct current is used in welding. Frost (Ref. 10)
considered the different chemical effects
at the anode and cathode in electroslag
welding. Blander and Olson (Ref. 11) postulated an electrochemical mechanism for
the alteration of weld metal chemistry in
submerged arc welding.
Electrochemical and
Thermochemical Reactions
Electrochemical reactions could be an
important factor governing the chemistry
of weld metal in direct current welding.
Electrochemical effects can occur at the
interfaces of the slag with the hot weld
pool and the melted electrode tip. The
relative numbers of ions and electrons in
the slag govern the importance of these
effects, which could be major factors
controlling weld metal chemistry. When
liquid slag has an interface with the molten
metal, the electrochemical reactions are
relatively simple to understand. This type
of interface is present in submerged arc
welding. The possible anodic reactions include the oxidation of iron and alloy elements, and the discharge and pickup of
oxygen anions from the slag
M (metal) + n O 2 " (slag) =
M O „ (slag) + 2ne~
(D
where M is iron or an alloy element at the
electrode tip/slag or the weld pool/slag
interface. Thus, substantial oxidation
losses of alloy elements and pickup of
oxygen are expected at the anode. The
Table 1—Compositions of Steel Plate and
Welding Wire (wt-%)
Elements
C
Mn
Si
Mo
Cr
Ni
Al
Cu
Ti
P
S
o
N
Fe
Steel Plate
Welding Wire
0.18
1.25
0.05
0.03
0.077
0.06
0.004
0.036
0.001
0.018
0.031
0.002
0.004
balance
0.059
1.38
0.05
0.33
0.073
0.11
0.018
0.77
0.033
0.012
0.015
0.002
0.004
balance
possible cathodic reactions include reduction of metallic cations from the slag and,
to some extent, the refining of nonmetallic elements such as oxygen and sulfur
M 2 + (slag) + 2e~ = M (metal)
Si
4+
(slag) + 4e~ = Si (metal)
O (metal) + 2 e " = 0 2 ~ (slag)
(2)
(3)
(4)
where M and Si represent electrodeposited metals at the interface. The large disparity in areas between the electrode tip
and the weld pool results in different current densities. Current density is much
higher at the electrode tip interface than
at the weld pool interface. Thus, reactions
at the electrode tip may exert a greater
influence on the final weld metal chemistry than those at the weld pool.
In addition to the electrochemical reactions, thermochemical reactions also play
an important role in controlling the weld
metal chemistry in submerged arc welding. The thermochemical reactions occur
very rapidly because of high temperature.
These thermochemical reactions include
deoxidation reactions, such as those encountered in steel making, and reactions
that lead to a closer approach to equilibrium between the flux and metal phase.
Examples of such reactions would be silicon pickup from a high silica flux or the
oxidation loss of transition elements
through a deoxidation reaction.
Si0 2 (slag) + 2M (metal) = Si (metal)
+ 2MO (slag)
FeO (slag) + M (metal) = Fe (metal)
+ M O (slag)
<5)
(6)
where M can be aluminum or calcium in
Reactions 5 and 6 and manganese in Reaction 6 for the flux system used in this research. The droplet forms at the electrode
tip and then travels through the molten
flux and plasma. The entire process occurs
in a few milliseconds (Refs. 12,13), and the
temperature of the droplet is very high.
Due to the high temperature, it is thermodynamically possible for chemical reac-
tions to occur. A number of investigations
have been made concerning the nature of
chemical reactions at the melted electrode tips and the detached droplets
(Refs. 5, 8, 12-18), and several investigators indicated that oxygen was transferred
to the metal in these t w o zones (Refs. 2,
3, 13, 14, 18). In the hot weld pool immediately below the arc, the detached droplets become "diluted" with molten metal
from the base plate. Although the slag/
metal interfacial contact area available for
reaction is much smaller than that of the
detached droplets, the strong turbulance
existing in the hot weld pool creates an
effective stirring of the liquid metal. Moreover, pickup of oxygen at this zone is favored by the increased time available for
reaction. It is believed that the pickup of
oxygen is the result of the oxygen solubility being exceeded in the hot weld pool.
The zone of the molten weld pool behind
the electrode starts to cool and solidify as
the electrode moves away from it. Upon
cooling of the metal in the weld pool from
a high temperature to the solidification
temperature, a supersaturation with respect to silicon and manganese deoxidation reactions occurs initially. Therefore,
silicon and manganese will react with dissolved oxygen, and then the composition
will move rapidly towards equilibrium.
Figure 1 shows the schematic of expected variations in oxygen and silicon
contents for an electrode-positive (reverse polarity) submerged arc weld due to
electrochemical and thermochemical reactions. The curve shows the oxygen and
silicon contents of steel in equilibrium with
solid silica at some temperature. Before
experiments were done, the composition
variation was expected to follow the
straight lines for an electrode-positive submerged arc weld. The anodic reaction at
the melted electrode tip causes some ox-
ygen pickup and silicon loss. The thermochemical reactions cause more oxygen
pickup at the melted electrode tip and the
detached droplet while the droplet is
passed through the molten flux. At the
cathodic weld pool, oxygen refining and
silica reduction were expected, and then
weld composition would vary with travel
speed due to the difference in the amount
of thermochemical deoxidation reaction.
As the welding speed decreases, the
amount of thermochemical deoxidation
reaction increases because the heat input
and the solidification time increase. Therefore, weld pool composition was expected to go to equilibrium composition
as the travel speed decreases.
Experimental Procedure
Direct
current
electrode-negative
(straight) and direct current electrodepositive (reverse) polarity submerged arc
welds were made with 2.38 mm (V32 in.)
commercial low-carbon steel welding
wire on ASTM A36 steel plates and a water-cooled pure copper plate. The compositions of the steel plate and the welding wire are given in Table 1, and the
composition of the commercial flux used
is shown in Table 2.
Figure 2 is a schematic drawing of a
cross-section of the submerged arc welding process showing the melted electrode
tips, the detached droplets and the base
plate for both electrode-positive and electrode-negative polarities. The melted electrode tips, detached droplets and weld
metal were collected and analyzed. The
water-cooled copper plate was used as a
base plate to collect the melted electrode
tips and the detached droplets. The welding process was operated at a constant
welding current of 585 A and a constant
potential of 28.5 V. The constant current
Expected Composition Variations
for a DCEP Submerged Arc Weld
Cathodic Weld Pool
Fig. 1 — Schematic of
expected variations
in oxygen and silicon
contents for an
electrode-positive
polarity submerged
arc weld.
i
a
Thermochemical
Reactions in
t h e Weld Pool
<D
&0
l>»
X
o
Thermochemical
Equilibrium
Silicon (wt.%)
WEEDING RESEARCH SUPPtEMENT 1447-s
Table 2—Composition of Welding Flux
(wt-%)
Substance
Amount
Si0 2
AI2O3
MgO
CaF2
CaO
MnO
Ti02
Na203
Fe 2 0 3
C
11.22
18.14
33.23
25.26
6.92
1.15
0.90
0.82
1.99
0.37
was obtained by adjusting the electrode
velocity, which was 72 mm/s (170 in./
min) for the anodic wire, and 99 mm/s
(235 in./min) for the cathodic wire.
The electrode tips were collected by
stopping the welding process, pulling the
electrode away from the weld pool and
cutting off the tip. The liquid metal droplets are released from the electrode tip
and travel through the molten flux. The
detached droplets were collected from
welds made at a high velocity over a water-cooled copper plate so that the droplets are suspended in the molten flux. The
detached droplets were extracted from
the ground slag by magnetic separation.
The speed of quenching attained is sufficient to freeze the metal composition established at high temperature. Single pass,
bead-on-plate welds were made on steel
plates at several welding speeds (1.5534.67 mm/s) with constant voltage (28.5
V) and current (580 A) for both polarities.
The wire feed speed was adjusted to
maintain the constant current.
using a LECO interstitial analyzer, i.e.,
fusion in a graphite crucible under inert
atmosphere. A Baird — Atomic emission
spectrometer was used to determine manganese, silicon and other alloy elements in
weld metals, welding wire and electrode
tips. The electrode tips and wires were
flattened in a rolling mill to get enough
area for analysis. The detached droplets
were analyzed using the wavelength dispersive analyzer on a JEOL scanning electron microscope.
A metallurgical model by Their (Ref. 5)
is adopted to show the extent of element
transfer between the weld metal and flux.
Its application gives quantitative data for
the gain or loss of elements arising from
slag/metal reactions. The transfer efficiency into the weld can be expressed in
terms of the difference between analytical
and nominal compositions. The difference
between analytical and nominal compositions is defined as delta (A) quantity. The
analytical composition is obtained by
chemical analysis methods and the nominal composition can be obtained by calculations based on the detached droplet
and plate compositions and dilution. A
positive concentration change (A > 0) indicates a gain or pickup of a particular element, i.e., transfer of element from the
slag to the weld metal. A negative change
(A < 0) indicates a loss, i.e., transfer from
the weld metal to the slag.
of the high current densities, and thermochemical reactions are expected because
of the high temperature and the generally
large temperature-dependent differences
in chemical potentials of the various reactants and products in the flux and metal
phases.
An examination of the welding wire,
steel plate and flux compositions in Tables
1 and 2 shows that the wire and steel plate
have very low silicon and oxygen concentrations, and a relatively high manganese concentration, while the flux has a
high silica activity and a relatively low
manganese-oxide/iron-oxide ratio, which
is far from equilibrium with the welding
wire and steel plate. Thus, the manganese
content of the welding wire and steel
plate would be expected to drop as a result of thermochemical oxidation losses to
the flux, and the silicon and oxygen content would be expected to increase because of the reaction with the flux.
The reactions at the melted electrode
tip, the detached droplet and the weld
pool were considered. The chemical analysis results show that a substantial difference exists between the chemistries of
anode and cathode. These differences are
the result of electrochemical reactions.
Figure 3 shows the plot of the average
oxygen contents of the welding wire, the
melted electrode tips, and the detached
droplets for both electrode-negative and
electrode-positive
polarity
welding
modes. This plot shows a very low oxygen
Results and Discussion
content in the welding wire (20 ppm) and
a very significant oxygen pickup in the
The purpose of this investigation is to
melted electrode tips for both polarities.
consider the relative influence of thermoThe influence of thermochemical oxygen
chemical and electrochemical reactions
pickup is shown by the fact that significant
on the weld metal chemistry. Electrooxygen pickup is observed in both elecchemical reactions are expected because
Analyses for oxygen were carried out
trode-positive and electrode-negative
configurations. This excess oxygen came
from the surrounding atmosphere and
(-)
decomposition of oxide components in
the flux. The influence of electrochemical
reactions is shown by the fact that the oxwire
ygen content of the anode in the elec(2.4mm)
trode-positive polarity power mode (591
ppm) is over twice that of the cathode in
molten
arc
the electrode-negative polarity power
flux
plasma
mode (277 ppm). This oxygen content
difference is due to oxygen pickup at the
anode and oxygen refining at the cathode.
The real difference is somewhat less since
more wire is fed and melted at the cathode for a fixed current, thus diluting the
total electrochemical and thermochemical
effect at the cathode. If the electrochemical and thermochemical reactions were
considered as separate steps, the average
melted electrode tip oxygen concentration for the t w o polarities could be considered to crudely represent the thermochemical contribution, and the separation
of t w o concentrations from this mean
(a) Reverse Polarity (DCEP)
(b) S t r a i g h t Polarity (DCEN)
would represent the electrochemical efFig. 2 — Schematic of the cross-section of a submerged arc welding process showing the melted fects. However, the different wire feed
welding wire tips, the detached droplets and the base plate for both electrode-positive and elecrates cloud this interpretation. After the
trode-negative polarities.
(+)
448-s | DECEMBER 1990
0 6
1400
1200
1000
S
0. 5
EHM
1
|
53
Anode
Wire
Cathode
Z
tu
|
| Wire
EZH Anode
V////\ Cathode
2
a.
O
0.4
800
x
a
0 3
oc
<
o
o
600 -
0 2
400
o
0. I
'
200
•
0
0
Wire
Electrode
Tip
Wire
Droplet
Figures 4 a n d 5 s h o w the average silicon
a n d manganese c o n t e n t s in the initial
e l e c t r o d e , the m e l t e d e l e c t r o d e tip and
the d e t a c h e d d r o p l e t . The m e l t e d elect r o d e tips s h o w t h e increases in silicon
c o n c e n t r a t i o n (less increase at the a n o d e
a n d m o r e increase at the c a t h o d e ) a n d the
decreases in manganese c o n c e n t r a t i o n
(less decrease at the c a t h o d e a n d m o r e
decrease at the a n o d e ) for b o t h polarities.
This is e v i d e n c e that b o t h t h e r m o c h e m i c a l
a n d electrochemical reactions o c c u r r e d at
the m e l t e d e l e c t r o d e tip. T h e initial w e l d ing w i r e has a v e r y l o w silicon c o n t e n t ,
w h i l e the flux has a high silicon-oxide c o n tent. This causes the t h e r m o c h e m i c a l
p i c k u p of silicon f r o m the flux. T h e elect r o c h e m i c a l influence is significant, as indic a t e d by the fact that the c a t h o d i c elect r o d e tip silicon c o n t e n t is a b o u t 0.06 w t % higher than that of the anodic e l e c t r o d e
tip, a n d by t h e fact that t h e c a t h o d e f e e d
rate is higher than the a n o d e f e e d rate,
w h i c h means the t o t a l a m o u n t of silicon in
t h e c a t h o d i c e l e c t r o d e tip is relatively
higher than indicated in Fig. 4. Both ther-
Electrode
I
Droplet
OC
a.
O
_i
Ui
>
Ul
a
Tip
Fig. 4 —Average silicon contents of the initial wire, the electrode tips, and
the detached droplets for both electrode-negative (cathodic wire) and
electrode-positive (anodic wire) polarities.
Fig. 3—Average oxygen contents of the welding wire, the electrode tips
and the detached droplets for both electrode-negative (cathodic wire)
and electrode-positive (anodic wire) polarities.
m o l t e n d r o p l e t separates f r o m t h e w i r e ,
the electrochemical reaction ceases. But
the t h e r m o c h e m i c a l reaction continues
w h i l e the d r o p l e t is falling t h r o u g h the
m o l t e n slag. T h e r e f o r e , t h e d e t a c h e d
d r o p l e t s h o w s higher o x y g e n c o n t e n t
t h a n the m e l t e d e l e c t r o d e tip d u e t o the
increase in o x y g e n c o n t e n t b y the c o n t i n uous d e c o m p o s i t i o n of flux. Even t h o u g h
electrochemical reaction d o e s not occur
in d e t a c h e d d r o p l e t s , the results o f o x y g e n analysis still s h o w the electrochemical
reaction e f f e c t at t h e m e l t e d e l e c t r o d e
tips as indicated b y the d i f f e r e n c e in o x y gen c o n t e n t s b e t w e e n a n o d i c a n d cat h o d i c droplets.
;',',•
Ul
V)
Ul
m o c h e m i c a l a n d electrochemical reactions are also indicated in t h e case o f
manganese; h o w e v e r , t h e high manganese c o n t e n t in t h e e l e c t r o d e a n d the l o w
manganese-oxide
(MnO)/iron-oxide
(Fe203) ratio in t h e flux lead t o t h e r m o chemical manganese loss at the m e l t e d
e l e c t r o d e tip. T h e r e f o r e , manganese o x i d a t i o n a n d silicon r e d u c t i o n reactions are
m a j o r t h e r m o c h e m i c a l reactions in this
system. T h e changes in the silicon a n d
manganese c o n c e n t r a t i o n s f r o m the elect r o d e tip t o t h e d e t a c h e d d r o p l e t are
mostly t h e r m o c h e m i c a l . W i t h silicon, h o w e v e r , t h e r e is a decrease rather than an
e x p e c t e d increase in silicon c o n t e n t of t h e
d e t a c h e d d r o p l e t c o m p a r e d t o the m e l t e d
e l e c t r o d e tip, indicating that a large fract i o n of the silicon in the d r o p l e t has back
reacted w i t h m o r e n o b l e metal oxides in
the flux {e.g., Fe203). O n e o f t h e possible
driving forces f o r this reaction is related t o
the possibility that the d r o p l e t is at a l o w e r
t e m p e r a t u r e and has a higher o x y g e n
c o n t e n t than t h e tip. T h e average manganese c o n t e n t is f u r t h e r decreased b y
t h e r m o c h e m i c a l reactions w i t h m o r e n o ble metal oxides in t h e flux w h e n g o i n g
f r o m t h e e l e c t r o d e tip t o the d e t a c h e d
d r o p l e t , w h i c h falls t h r o u g h a n d reacts
w i t h the flux.
Figure 6 c o m p a r e s the d e t a c h e d d r o p let c o m p o s i t i o n s f o r the various alloy elements. Silicon, a l u m i n u m and manganese
3.0
2.5
I
1 Wire
ESS3
r
tn
UJ
z
<t
o
Cathode
2.0
Fig. 5 —Average
manganese contents
of the initial wire,
the electrode tips
and the detached
droplets for both
electrode-negative
(cathodic wire) and
electrode-positive
(anodic wire)
polarities.
o
tr
<
ui
CO
Ul
cc
z
Ul
S
a.
O
I
a
oc
<
ui
(/>
Ul
oc
a.
O
_i
UJ
>
4
Ul
in
ui
CE
z
I .5
ui
1.0
3>
2
a.
ui
a
u
cc
<
UJ
<n
0.5
Wire
Electrode
Tip
Droplet
W E L D I N G RESEARCH SUPPLEMENT 1449-s
2.50
0.50
DROPLET COMPOSITION
^
0.40
2.00
fc?
o
Anode
Wire
Cathode
0.30
1.50
DJ
in
<D
C
D
CT
C
D
<
0.20-
1.00
o
o
o 0.10 0
ILJI
Si
Al
Ni
-0.50
jtt
Cr
Ti
Mo
0
Mn
Fig. 6 Average droplet compositions for silicon, aluminum, nickel, chromium, molybdenum, titanium and manganese for electrode-negative and electrode-positive polarities compared with the
initial welding wire compositions.
s h o w higher c o n c e n t r a t i o n s in the cathodic d r o p l e t s than in the anodic d r o p lets. This is largely caused b y oxidation
losses t o the flux at the a n o d e and electrochemical r e d u c t i o n f r o m the flux at the
c a t h o d e d u r i n g the reaction at the m e l t e d
e l e c t r o d e tip. T h e differences for nickel,
titanium a n d c h r o m i u m are of the o r d e r of
t h e analytical uncertainties a n d are thus
inconclusive. M o l y b d e n u m s h o w s the o p posite t r e n d . That is, the c o n c e n t r a t i o n in
the a n o d e is higher t h a n that o f the catho d e . This b e h a v i o r is difficult t o explain in
terms of m o l y b d e n u m r e a c t i o n alone, b u t
if iron reaction is also taken into a c c o u n t ,
it can b e explained easily. Iron is the m a jor e l e c t r o d e constituent and it is m o r e
easily o x i d i z e d t h a n m o l y b d e n u m . T h e r e f o r e , the loss of iron at t h e a n o d e and the
gain of i r o n at t h e c a t h o d e are relatively
greater than those o f m o l y b d e n u m , a n d
these cause an increase in m o l y b d e n u m
c o n c e n t r a t i o n at the a n o d e and a d e crease at the c a t h o d e . Thus, this b e h a v i o r
of m o l y b d e n u m is also the result o f electrochemical reaction.
Figures 7 t h r o u g h 14 s h o w the results of
chemical analyses of the a n o d e a n d catho d e w e l d metal f o r silicon, manganese,
a l u m i n u m , m o l y b d e n u m and o x y g e n as a
f u n c t i o n of w e l d i n g s p e e d . T h e results for
the a n o d e a n d c a t h o d e of each element
are p l o t t e d t o g e t h e r . These results s h o w
that electrochemical reactions cause substantial changes in w e l d metal chemistry.
Figure 7 illustrates t h e variations in delta
silicon d u e t o polarity. High transient c o n c e n t r a t i o n of o x y g e n m a y exist in the liquid metal at t h e hot w e l d p o o l as a result
of high solubility at e l e v a t e d t e m p e r a t u r e .
O n s l o w c o o l i n g d o w n t o the solidification
t e m p e r a t u r e , this will lead t o s p o n t a n e o u s
reaction a n d losses of dissolved silicon.
The silicon loss must be d u e t o the f o r m a t i o n o f s o m e d e o x i d a t i o n reaction p r o d uct. It is generally a c c e p t e d that f o r m a t i o n
of silicon o x i d e (Si02) takes place w h e n
sufficient o x y g e n is available. But silicon
loss is r e c o v e r e d b y electrochemical r e d u c t i o n reaction at the c a t h o d e a n d i n creased at t h e a n o d e b y electrochemical
oxidation reaction. T h e r e f o r e , silicon
s h o w s m u c h loss at the a n o d e , and a little
gain or almost n o change at the c a t h o d e .
Figure 8 s h o w s the variations in delta
manganese d u e t o polarity. Similar t o silic o n , o x i d a t i o n o f manganese is f a v o r e d b y
the l o w e r t e m p e r a t u r e prevailing in the
c o o l e r part o f the w e l d p o o l . Here, m a n ganese will react w i t h dissolved o x y g e n t o
f o r m slags that exhibit a large loss of m a n ganese in b o t h a n o d e a n d c a t h o d e w e l d
metal. T h e reason w h y manganese s h o w s
extensive losses in b o t h cases is because
the flux has a little manganese o x i d e
( M n O ) a n d the base plate has a high m a n ganese c o n t e n t . T h e r e f o r e , manganese
can b e o x i d i z e d easily. A n o t h e r reason f o r
manganese loss can be a result of t h e
e v a p o r a t i o n because v a p o r pressure i n creases strongly w i t h t e m p e r a t u r e . But the
difference in delta manganese d u e t o p o larity s h o w s t h e effect of electrochemical
reaction. That is, the c a t h o d e s h o w s less
loss than the a n o d e because of the elect r o c h e m i c a l r e d u c t i o n reaction at the catho d e , and the electrochemical o x i d a t i o n
reaction at the a n o d e .
Variations in delta a l u m i n u m are s h o w n
in Fig. 9. A l u m i n u m , like silicon and m a n ganese, is also o x i d i z e d w h i l e the t e m p e r ature of t h e w e l d p o o l is l o w e r e d t o t h e
0.15
Manganese Change in the Weld Pool
Silicon Change in Weld Pool
0.00
0.10
Cathode
-0.05
6?
% -0.10
0.05
"5
Cathode
<*A
a)
171
0)
0.00
-0.15
o
cn
L -0.20
D
-0.05 Anode
<
-0.25 -
-0.10
-0.30
-0.15
I | I I I | I I I | I I I | I I I | I r I i I I I I I—r—i—]—I—i—r-
0.0
0.4
0.8
1.2
Travel
1.6
2.0
Speed
2.4
2.8
(cm/s)
3.2
3.6
Fig. 7 — Gain or loss of silicon due to the reactions in the weld pool for
electrode-negative and electrode-positive polarities as a function of
welding speed.
450-s | DECEMBER
1990
-0.35
Anode
0.0
i i i I i i i l i i i I i i i I i i i l i i i I i i i I i i i I i i
0.4
0.8
1.2
Travel
1.6
2.0
Speed
2.4
2.8
(cm/s)
3.2
3.6
Fig. 8 — Gain or loss of manganese due to the reactions in the weld pool
for electrode-negative and electrode-positive polarities as a function of
welding speed.
0.15
0.15
Molybdenum Change in the Weld Pool
Aluminum Change in the Weld Pool
0.10 -
0.10
»?
fc?
"J
'—'
E
j?
5
0.05
Anode
0.05
o
0.00
%
A
E
A
A
0.00
TJ
A
Cathode
>-•
< -0.05 -
°
O
o
o o
o o
-0.15
-0.10 -
-0.15
i l i i i l i i i l i i i l i i i l i i i I i i i I i i i l i i
0.0
0.4
0.8
1.2
Travel
1.6
2.0
Speed
2.4
2.8
(cm/s)
3.2
Figure 10 s h o w s the variations in delta
m o l y b d e n u m d u e t o polarity. M o l y b d e -
| I I I | I I J 1 I I I ; I I I | i I I [ I i
0.0
3.6
Fig. 9 — Gain or loss of aluminum due to the reactions in the weld pool
for electrode-negative and electrode-positive polarities as a function of
welding speed.
solidification t e m p e r a t u r e . A l u m i n u m o x ide (AI2O3) is stable at high t e m p e r a t u r e
rather than silicon o x i d e (Si02> a n d m a n ganese o x i d e ( M n O ) , as indicated in the
Ellingham
diagram. Therefore,
even
t h o u g h the flux contains a large a m o u n t of
aluminum o x i d e , it is difficult t o pick u p the
a l u m i n u m f r o m the slag b y e l e c t r o c h e m i cal r e d u c t i o n reaction at the c a t h o d e w e l d
p o o l . So, t h e results s h o w the a l u m i n u m
losses e v e n at t h e c a t h o d e w e l d p o o l . But
t h e lesser loss at the c a t h o d e is the
e v i d e n c e f o r the electrochemical reaction.
-0.05
<
Anode
-0.10
0.4
0.8
1.2
Travel
1.6
2.0
Speed
1 i i i | i I I |
2.4
2.8
(cm/s)
3.2
3.6
Fig. 10 — Gain or loss oi molybdenum due to the reactions in the weld
pool for electrode-negative and electrode-positive polarities as a function of welding speed.
n u m changes s h o w the o p p o s i t e t r e n d as
in the m e l t e d e l e c t r o d e tip. This b e h a v i o r
of m o l y b d e n u m is the result of iron losses
at t h e a n o d e and p i c k u p at the c a t h o d e as
discussed previously.
Figures 11 t h r o u g h 13 s h o w the variations in delta nickel, c h r o m i u m a n d titan i u m d u e t o polarity. Those elements also
s h o w a little electrochemical effect. But
t h e differences b e t w e e n a n o d e and catho d e are of the o r d e r of t h e analytical u n certainties and are thus inconclusive.
Figure 14 s h o w s the variations in delta
o x y g e n d u e t o polarity. O x y g e n losses are
s h o w n at b o t h c a t h o d e and a n o d e , a n d
the c a t h o d e s h o w s less loss than the
a n o d e . This is t h e result of t h e difference
in the s l a g / m e t a l interfacial contact area
available f o r d e o x i d a t i o n reaction. T h e
area of the c a t h o d e is smaller than that of
t h e a n o d e . T h e r e f o r e , the separation of
d e o x i d a t i o n p r o d u c t s at the a n o d e is easier than that at t h e c a t h o d e . Thus, electrochemical effect may be s w e p t o u t b y
t h e r m o c h e m i c a l reaction.
Figure 15 is the c o m p o s i t i o n m a p for
b o t h electrode-negative and electrodepositive polarity s u b m e r g e d arc w e l d i n g ,
s h o w i n g the variations in o x y g e n and silic o n c o n t e n t s d u r i n g w e l d i n g . T h e curves
are the plots of the silicon a n d o x y g e n
c o n t e n t s of steel in equilibrium w i t h solid
0.15
0.15
Chromium Change in the Weld Pool
Nickel Change in the Weld Pool
0.10
0.10
K °05
%
—
a)
o
Cathode
"5
0.05 -
Cathode
Cathode
tb
-J&g, A A —A—
Qp OQ o
0.00
o
A A
O O
"
0
A
0
0.00
A
Anode
Anode
-0.05
<-> - 0 . 0 5
-0.10
-0.10
-0.15
0.0
0.4
0.8
1.2
Travel
i i I
1.6
2.0
Speed
A
o o
-0.15
2.4
2.8
(cm/s)
3.2
3.6
Fig. 11 — Gain or loss of nickel due to the reactions in the weld pool for
electrode-negative and electrode-positive polarities as a function of
welding speed.
i I i i i
0.0
0.4
0.8
1.2
Travel
1.6
2.0
Speed
2.4
2.8
(cm/s)
3.2
3.6
Fig. 12 — Gain or loss of chromium due to the reactions in the weld pool
for electrode-negative and electrode-positive polarities as a function of
welding speed.
W E L D I N G RESEARCH S U P P L E M E N T 1451-s
0.15
100
Oxygen Change in the Weld Pool
Titanium Change in Weld Pool
0.10
Cathode
E
0.05
Cathode
E
A^
0.00
$$
A
Anode
CD
aP'
cn
c
o
P
-100-
Q.
Q.
-—'-200C
5^-300X
O
Anode
-0.05
-400
-0.10 -
-0.15
0.0
-500
0.4
0.8
1.2
Travel
1.6
2.0
Speed
2.4
2.8
(cm/s)
3.2
-600-|-n
3.6
0.0
Fig. 13 — Gain or loss of titanium due to the reactions in the weld pool
for electrode-negative and electrode-positive polarities as a function of
welding speed.
silica at 1 6 0 0 ° C (2912°F) a n d 1 8 0 0 ° C
(3272°F). T h e c o m p o s i t i o n s of w e l d i n g
w i r e , m e l t e d e l e c t r o d e tips, d e t a c h e d
droplets and w e l d metal are p l o t t e d o n
the diagram f o r b o t h polarities. T h e w i r e
has l o w silicon a n d o x y g e n c o n t e n t s . Both
anodic a n d c a t h o d i c e l e c t r o d e tips s h o w
an increase in silicon c o n t e n t d u e t o the
o x i d a t i o n o f i r o n a n d manganese, a n d in
oxygen content due to decomposition of
flux and c o n t a m i n a t i o n f r o m the a t m o sphere. But the cathodic e l e c t r o d e tip
s h o w s a greater increase in silicon a n d less
increase in o x y g e n , d u e t o t h e e l e c t r o chemical silicon r e d u c t i o n a n d o x y g e n refining reactions. W h i l e the d e t a c h e d d r o p lets are falling t h r o u g h the m o l t e n flux, the
droplets pick u p the o x y g e n b y flux d e c o m p o s i t i o n a n d c o n t a m i n a t i o n f r o m the
a t m o s p h e r e . T h e r e f o r e , the o x y g e n c o n -
Fig. 15-The
composition map for
direct current
electrode-negative
(DCEN) and direct
current
electrode-positive
(DCEP) polarity
submerged arc
welding showing the
variations in oxygen
and si/icon contents
during welding.
0.20
i i i i i i i i i i i i i i i i i i i i i i i i | i i i
0.4
Conclusions
1) B o t h e l e c t r o c h e m i c a l a n d t h e r m o chemical reactions are significant in subm e r g e d arc w e l d i n g .
2) T h e t h e r m o c h e m i c a l reactions m o v e
the flux a n d w e l d metal c o m p o s i t i o n s t o w a r d chemical equilibrium. In this study,
manganese activity w a s higher in t h e
metal than in the flux; t h e r e f o r e , manga-
0.15
DCEP
0.10
DCEN
roplet
x
O
2.4
2.8
(cm/s)
3.2
3.6
nese oxidation loss f r o m t h e metal t o t h e
flux w a s o b s e r v e d . Silicon activity w a s
higher in t h e flux; t h e r e f o r e , a t h e r m o chemical silicon p i c k u p b y the w e l d metal
was observed.
3) T h e electrochemical reactions at the
a n o d e include o x i d a t i o n losses of alloy elements t o the flux and the discharge a n d
p i c k u p of o x y g e n anions f r o m t h e flux.
T h e electrochemical reactions at the catho d e include the r e d u c t i o n of metal ions
f r o m the flux a n d the refining of o x y g e n .
4) T h e r m o c h e m i c a l a n d e l e c t r o c h e m i cal c o m p o s i t i o n changes are greater at a
l o w than at a high w e l d i n g s p e e d . Electrochemical reactions are e n h a n c e d b y
higher, total current f l o w per unit v o l u m e
of w e l d m e t a l . T h e r m o c h e m i c a l reactions
at a l o w w e l d i n g speed are e n h a n c e d b y
higher t e m p e r a t u r e s a n d longer reaction
time b e f o r e solidification.
5) C o m p o s i t i o n paths f o r the silicono x y g e n equilibrium are different for direct
current e l e c t r o d e - p o s i t i v e (reverse polarity) a n d direct current e l e c t r o d e - n e g a t i v e
(straight polarity) w e l d s .
Composition Maps for DCEP and DCEN
Submerged Arc Welding Processes
^
1.6
2.0
Speed
Fig. 14 — Gain or loss of oxygen due to the reactions in the weld pool for
electrode-negative and electrode-positive polarities as a function of
welding speed.
tents in the droplets are higher than those
in t h e e l e c t r o d e tips f o r b o t h polarities.
T h e w e l d metal s h o w s decreases in silicon
and o x y g e n c o n t e n t s b y d e o x i d a t i o n r e action a n d c o m p o s i t i o n variation d u e t o
w e l d i n g s p e e d . As the w e l d i n g speed d e creases, the c o m p o s i t i o n of the w e l d
metal a p p r o a c h e s the equilibrium c o m p o sition.
%
0.8
1.2
Travel
6) T h e total gain or loss of alloy elements in w e l d metal is influenced b y t h e
c o m p o s i t i o n s of flux, e l e c t r o d e and base
plate, a n d b y w e l d i n g process conditions.
Further experiments w i t h synthetic
fluxes, w h i c h are chosen t o minimize
t h e r m o c h e m i c a l reactions, are p l a n n e d
and should help t o b e t t e r define the relative i m p o r t a n c e of electrochemical reactions.
0.05
Acknowledgments
0.00
0.00
0.10
Silicon
452-s I DECEMBER 1 9 9 0
0.20
(wt.%)
0.30
J. H. Kim, R. H. Frost and D. L. O l s o n
wish t o a c k n o w l e d g e t h e s u p p o r t of the
U.S. A r m y Research O f f i c e at the C o l o r a d o School o f M i n e s , a n d M . Blander ack n o w l e d g e s the s u p p o r t o f the O f f i c e o f
Naval Research at A r g o n n e National Laboratory under Navy Order No. N0001485-F-0097.
References
1. Belton, C. R., Moore, T.)., and Tankins, E.
S. 1963. Slag-metal reactions in submerged-arc
welding. Welding journal 42(7): 289-s to 297-s.
2. Chai, C. S„ and Eagar, T. W . 1982. Slagmetal reactions in binary CaF2-metal oxide
welding fluxes. Welding journal 61(7): 229-s to
232-s.
3. Chai, C. S„ and Eagar, T. W . 1981. Slagmetal equilibrium during submerged arc welding. Metall. Trans. B 12B(9): 539 to 547.
4. Indacochea, ). E., Blander, M., Christensen, N., and Olson, D. L. 1985. Chemical reactions during submerged arc welding with FeOMnO-Si0 2 fluxes. Metall. Trans. B 16B(6): 237 to
245.
5. Their, H. 1980. Metallurgical reactions in
submerged arc welding. Proc. Intl. Conf. on
Weld Pool Chemistry and Metallurgy. The
Welding Institute, London, England, pp. 271 to
278.
6. North, T. H., Bell, H. B., Nowicki, A., and
Craig, I. 1978. Slag/metal interaction, oxygen
and toughness in submerged arc welding.
Welding lournal 57(3): 63-s to 75-s.
7. Davis, M. L. E„ and Coe, F. R. 1977. The
chemistry of submerged arc welding fluxes.
The Welding Institute. 3 9 / 1 9 7 7 / M .
8. Mitra, U. 1984. Kinetics of slag metal reactions during submerged arc welding of steel.
Ph.D. dissertation. MIT, Cambridge, Mass.
9. Blander, M., and Olson, D. L. 1984. Thermodynamic and kinetic factors in the pyrochemistry of submerged arc flux welding of
iron-based alloys. Proc. Intl. Symp. on Metallurgical Slags and Fluxes. The Met. Soc. of AIME.
10. Frost, R. H., Olson, D. L, and Edwards,
C. R. 1983. The influence of electrochemical
reactions on the chemistry of the electroslag
welding process. Engineering foundation conferences. Modeling of Casting and Welding
Processes II. eds. |. A. Dantzig and I. T. Berry.
The Met. Soc. of AIME.
11. Blander, M., and Olson, D. L. 1986. Electrochemical effects on weld pool chemistry in
submerged arc and D. C. electroslag welding.
Proc. Intl. Conf. on Trends in Welding Research.
12. Pokhodnya, I. K„ and Kostenko, B. A.
1965. Fusion of electrode metal and its interaction with the slag during submerged arc welding. Avt. Svarka 10: 16 to 22.
13. Potapov, N. N., and Lyubavski, K. V.
1971. Oxygen content of weld metal deposited
by automatic submerged arc welding. Svar.
Proiz. 1: 11 to 15.
14. North, T. H. 1977. The distribution of
manganese between slag and metal during
submerged arc welding. Welding Research
Abroad 23: 2 to 13.
15. Potapov, N. N., and Lyubavski, K. V.
1971. Interaction between the metal and slag in
the reaction zone during submerged arc welding. Svar. Proiz. 7: 9 to 11.
16. Norin, P. A., and Malyshev, N. I. 1982.
Losses of manganese from electrode droplets
in arc welding in air. Svar. Proiz. 2: 21 to 33.
17. Grong, O., and Christensen, N. 1982.
Factors controlling weld metal chemistry. Final
Report Contract No. DATA 376-81-C-0309.
European Research Office of the U.S. Army.
18. Lau, T „ Weatherly, G. C , and McLean,
A. 1985. The source of oxygen and nitrogen
contamination in submerged arc welding using
CaO-Al203 based fluxes. Welding
lournal
64(12): 343-s to 347-s.
WRC Bulletin 352
April 1990
In October 1987, t h e PVRC Steering and Technical C o m m i t t e e s on Piping Systems established a task
g r o u p on independent support m o t i o n (ISM) t o evaluate t h e technical merits of using t h e ISM m e t h o d of
spectral analysis in t h e design and analysis of nuclear power plant piping systems.
The results of t h e task group evaluation c u l m i n a t e d in a unanimous technical position t h a t t h e ISM
m e t h o d of spectral seismic analysis provides m o r e accurate and generally less conservative response
predictions t h a n t h e c o m m o n l y accepted envelope response spectra (ERS) m e t h o d , and are r e p o r t e d in
this WRC Bulletin. The price of WRC Bulletin 3 5 2 is $ 2 5 . 0 0 per copy, plus $ 5 . 0 0 for U.S., or $ 1 0 . 0 0 for
overseas, postage and handling. Orders should be sent w i t h p a y m e n t t o t h e Welding Research Council,
3 4 5 E. 4 7 t h St., Room 1 3 0 1 , New York, NY 1 0 0 1 7 .
WRC Bulletin 354
June 1990
The t w o papers contained in this bulletin provide definitive information concerning t h e elevated t e m perature r u p t u r e behavior of 2 V 4 C r - l M o weld metals.
( 1 ) Failure Analysis of a Service-Exposed Hot Reheat Steam Line in a Utility Steam
Plant
By C. D. Lundin, K. K. Khan, 0. Yang, S. Hilton and W. Zielke
( 2 ) The Influence of Flux Composition of the Elevated Temperature Properties of Cr-Mo
Submerged Arc Weldments
By J. F. Henry, F. V. Ellis and C. D. Lundin
The first paper gives a detailed metallurgical failure analysis of cracking in a longitudinally welded h o t
reheat pipe w i t h 184,000 hours of operation at 1 0 5 0 ° F . The second paper defines t h e role of t h e welding
flux in submerged arc welding of 2 V 4 C r - l M o steel.
Publication of this r e p o r t was sponsored by t h e Steering and Technical C o m m i t t e e s on Piping Systems
of t h e Pressure Vessel Research Council of t h e Welding Research Council. The price of WRC Bulletin 3 5 4
is $ 5 0 . 0 0 per copy, plus $5.00 for U.S. and $ 1 0 . 0 0 for overseas postage and handling. Orders should be
sent w i t h p a y m e n t to t h e Welding Research Council, 3 4 5 E. 4 7 t h St., Room 1 3 0 1 , New York, NY 1 0 0 1 7 .
WELDING RESEARCH SUPPLEMENT 1453-s