Oxygen Potential Measurement of Iron
Silicate Slag-Copper-Matte System
lop Pa 2 a~d lop P 52 vs. 1/T. The phase stability diagrams
for the qumary system of Cu-Fe-O-S-Si02 are illustrated on
the lop Pa 2 vs. lop P52 plane which diagram has been
presented by Yazawa2 •
2 . EXPERIMENTS
Yoichi Takeda
Associate Professor, Department of Materials Science and
Technology, Faculty of Engineering, Iwate University,
Ueda 4-3-5, Morioka 020, Japan.
Tel. 81-196-21-6367, Fax. 81-196-21-6373
ABSTRACT
'fl:ere a'.e many phase combination in the quinary
sys~e~ _mcludn1:g molten phase. Simple one is slag-metal
eqmhbnum which has been presented in another paper3 .
The equilibrium experiment among slag, matte and copper
:netal p~ase und~r iron saturation provided also useful
mformat10n for us . In the experiment oxygen potential was
measured_, and sulfur potential and the activity of copper
were estimated from metal composition. I present the
following three kind experimental data; equilibrium
between slag and matte in the Fe-S-0-SiO system under
iron saturation, equilibrium between slag a~d matte of the
Fe-Cu-S-O-Si02 system under 10% and 100% SO
atmosphere in a magnesia crucible, and equilibrium among
slag, matte and copper metal in a magnesia crucible.
2.1
The phase stability diagrams for the Cu-Fe-0-S-SiO
are illustrated at 1473 and 1573K. To collect th~
information for arranging the diagrams, several kinds of
equilibrium experiments such as slag-copper metal, slagmatte, and slag-matte-copper metal are performed with an
oxygen sensor to measure oxygen potential in the system.
The data are presented in the diagrams of log P02 or log P 52
vs. l/T and log P 02 vs. log P 52 •
1. INTRODUCTION
The quinary system of Cu-Fe-O-S-Si02 is essential
system in copper smelting. Concentrate usually contains
copper, iron oxygen and sulfur, and oxygen is also an
useful reactant from air. We add silica as flux to avoid the
mutual miscib\lity between slag and matte and to make slag
melt at smeltmg temperature. In copper matte smelting
furnace, sulfur reacts with oxygen and separates as sulfur
dioxide to gas phase, and iron oxidizes and separates to
slag phase, and consequently copper concentrates in matte
phase, that is further oxidized and converted to blister
copper. I have carried out several kinds of equilibrium
experiment including slag, matte or copper melt. In the
experiments immersing an oxygen sensor in the melt, I
have . measured oxygen potential in the system. The
expenmental data are available for illustrating a phase
stability. The phase stability diagram of the Cu-Fe-O-SSi02 give us important information to comprehend the
condition of copper smelting steps thermodynamically. In
order to arrange a fundamental experimental research for
copper smelting and analyze the experimental data, the
diagram is also helpful to us. Portrayal of the phase
relationships in the quinary system of Cu-Fe-O-S-Si02
requires consideration of 10 binary, 10 ternary and 5
quaternary sub-systems for which the information is
fragmentary in many cases 1 , especially in over quaternary.
We have performed several kind of equilibrium
experiments. I illustrate phase stability diagrams combining
the experimental data and information from literature.
Several ways to describe a phase stability diagram have
b~en proposed, and each of them has advantage and
disadvantage. I present the univariant equilibria including
the sub-system of the quinary system on the diagrams of
Slag-matte equilibrium in the Fe-S-O-Si0 2 system
Fe0-Si02 slag and Fe-S matte are melted in an iron
~rucibles at 14~3 and 1573K. Oxygen potential in the melt
1s meas~'.ed ";Ith an oxygen sensor. The slag and matte
comp?s1t10ns m_ the quenched samples are determined by
chemical analysis. Oxygen, sulfur and iron contents in the
m_atte phase, on mass per cent base, at 1473K are shown in
~1g. 1 and Fig. 2 against oxygen partial pressure. Silica,
lf?n and sulfur contents in the slag phase are also shown in
Fig. 3. Closed marks in the figure show the experimental
result for slag-matte equili~rium in the Fe-S-O-Si0 2 system.
As further reference expenment Fe-S-0 matte is melted in
an iron crucible and oxygen potential is also measured. The
results are shown in the figures with open marks. As the
matte composition in the slag-matte equilibrium is very
close to the Fe-S-0 matte under iron saturation at a given
oxygen potential, sulfur potential in both system will be
close to each other. The sulfur partial pressure in Figure 2
is calculated by the Gibbs-Duhem integration for Fe-S-0
system under unity of iron activity. Steep increase in suifur
con~ent in the slag is_ ?bserved wi_th decreasing oxygen
partial pressure and s1hca content m slag, and slag and
matte does not separate at critical point of 10% Si02 • The
experimental results at 1573K for the same system are
plotted on Figs 4, 5 and 6.
2.2
Slag-matte equilibrium under S0 2 atmosphere
.. Slag a1:d matte in the Fe-Cu-S-O-Si02 system is
eqmhbrated with 10 or 100% S02 gas phase in a magnesia
crucible under controlled S2 partial pressure at 1473 and
~573K .. The oxygen potential in the system is measured by
1mmersmg an oxygen sensor. Magnesia content in slag
diss?l:'ed fro1:1 a 7rucible is 5 to 9%. The experimental
detail 1s descnbed m the other paper5 in this proceedings.
The oxygen partial presser, P a2 /atrn., is plotted against the
iron content in matte, {%Fe} on mass per cent base, as Fig .
7. Although sulfur partial pressure in the gas phase is
controlled,. it is difficult t? equilibrate molten phases with
gas phase_ m sulfur potentJal, hence sulfur partial pressure,
P 5zfatm., 1s calculated from the measured oxygen potential
by equation (3) and shown in Fig. 7.
1/2 Sz(g) + Oz(g)
11 G /J
0
= SOz(g)
= -362000 + 73.136T 6
(1)
(2)
MOLTEN SLAGS, FLUXES AND SALTS '97 CONFERENCE - 735
15
,..,.....
0
~
0
._..,
0
10
E
5
ro
--;,,
(/J
-6
Q
Si0 2 sat.
Ol
.2
0
30
Si0 2 sat.
25
0
20
FeO sat.
15
74
,..,.....
(1)
10
72
LL
~
Si02 sat.
::,__,
70
5
i •.
°'{.·.
OL.__JL_~~~--L~~~~-'--::-~~-·
-13.0
-12.5
-12.0
log P02 /atm.
Fig . 1.
Oxygen, and sulfur content~ in matte under ir?n
saturation at 1473K. Open crrcle shows those m
the Fe-S-0 system, and closed circle does for
the slag-matte equilibrium in the Fe-S-O-Si0 2
system.
log P02 /atm.
Fig. 2.
736 - MOLTEN SLAGS, FLUXES AND SALTS '97 CONFERENCE
Calculated sulfur partial pressure by the GibbsDuhem integration, the top figure, and iron cor_itent
in matte , the bottom figure, under iron saturat10n
at 1473K.
-12.5
-11.5
10
70
20
15
65
01
,..,...,
~
0
al
u..
(./)
~
0
0~
0
:::R
0
........,
(./)
~
60
55
10
5
~
5
iiiIll
0
Si0 2 sat
if)
0
(a)
10 I
-12.6
50
-12.1
log P 02 /atm.
Fig. 3.
(b)
0
-12.6
25
-12.1
log P02 /atm.
20
Silica and iron contents in slag, the left side
figure, and sulfur content in slag, the right side
figure, for slag- matte equilibrium in the Fe-S-0Si0 2 system under iron saturation at 1473K.
15
U)
:::R
0
._..,
logP0 /atm. = logP80 2 -1/2 logP82 -18908/T+3.82
(3)
5
Linear relations of logP0 2 or logP82 against log{%Fe} are
observed in Fig. 7. {%Fe} is related to copper content in
matte, {%Cu}, by equation (4) under higher S02
atmosphere.
{%Fe}= 65.5 - 0.824 {%Cu}
2.3
-12.0
(4)
-11.5
-11.0
log P 02 /atm
Slag-matte-copper metal equilibrium
Slag containing 18 to 21 % Si02 or 28 to 33% Si02 is
melted with matte and copper metal in a magnesia crucible
at 1573K. The oxygen partial pressure is measured with an
oxygen sensor and plotted against iron or copper content in
matte as shown in Fig. 8. In this phase e~uilibrium, sulfur
dioxide partial pressure increases from 10· atm. to 0.1 atm.
with increasing oxygen partial pressure.
Fig. 4.
Oxygen, and sulfur contents in matte under iron
saturation at 1573K. Open circle shows those in
the Fe-S-0 system, and closed circle does for
the slag-matte equilibrium in the Fe-S-O-Si0 2
system.
3. PHASE STABILITY DIAGRAM
3.1
Univariant equilibria in the related system
There are many univariant equilibria in the Cu-Fe-0-
MOLTEN SLAGS, FLUXES AND SALTS '97 CONFERENCE - 737
-12.0
-11.5
30
-11.0
10
70
k%SiO,)
25
-5
(%Fe)
iii
(.)
65
:;::;
'5
,.---.
N
E
0
cu
Q)
(./)
--cf
~
60 ~
20
e.....
~
~
(./)
*-
5
Ol
.2
15
-6
55
J
10
-11.5
I
(/)
(a)
50
-11.0
log P02 /atm.
Fig. 6.
r
74
1ii
(/)
72
0
w
LL
70
•
log P02 /atm.
Fig. 5.
Calculated sulfur partial pressure by the GibbsDuhem integration, the top figure, and iron content
in matte , the bottom figure, under iron saturation
at 1573K.
~
()
ui
(b)
0
-11.5
-11 0
log P0 2 /atm.
Silica and iron contents in slag, the left side
figure, and sulfur content in slag, the right side
figure, for slag- matte equilibrium in the Fe-S-OSi02 system under iron saturation at 1573K.
S-Si02 quinary system and its sub-systems. Essential one
to comprehend copper smelting with iron silicate slag is
listed in Table I. For the sake of saving space in the table,
equilibrium phase is shown by molecular expression and
sometimes minor species in the phase is eliminated. The
oxygen partial pressures in univariant equilibria are shown
in Fig. 9 against reciprocal temperature, 1/T. The symbol
beside line in the figure correspond with that in Table I.
Marked h is univariant equilibrating solid iron with wustite
in Fe-0 binary , and j is an univariant equilibrium of solid
iron-silica-iron silicate slag in Fe-O-Si02 • There are many
univariant equilibria in the Cu-Fe-O-S-Si02 between h and
j, so the figure axes are expanded in this region as shown
in Fig. 10. The sulfur partial pressures in univariant
equilibria are also shown in Fig. 11.
3.2
Phase stability diagram under iron saturation
Phase stability diagrams in the Cu-Fe-O-S-Si02
system at 1573 and 1473K are illustrated in Figs 12 and 13.
In the figure solid iron is oxidized or sulfidized and
disappears on the upper area on the line connecting points g,
C, E,G and a. Under the line connecting points i, F and
G iron oxide in slag is reduced and slag phase disappears.
The line connecting D to E is critical that slag and matte do
not separate and form oxysulfide containing considerable
amount of oxygen and sulfur. Silica content in slag is
around 10 to 15% and copper content in matte is less than
15 at the critical composition. Copper content in matte
saturated copper metal, on the line of y, F , D, decreases
from 52 to 15% at 1573K and from 56% to 15 % at 1473K
with increasing oxygen partial pressure. Oxygen content in
the matte increases from O to 15%. Following nine molten
phase equilibria in the quinary system are possible under
solid iron saturation; slag, copper metal, matte, oxysulfidc,
slag-copper metal, oxysulfide-copper metal, slag-mattecopper metal, matte-copper metal, slag-matte.
738 - MOLTEN SLAGS, FLUXES AND SALTS '97 CONFERENCE
Table
Symbol
System
I
Oxygen and sulfur partial pressures for univariant equilibria
Condensed phase
log P02/atm.
in related systems
log P sz/atm.
Refernce
a,-a2
Cu-0
Cu(I), Cu20(I)
a,
Cu-0
Critical point at 1608K
a2
Cu-0
Cu(I) , Cu 20(I), Cu 20(s); Monotectic at 1489K
ai-a,
Cu-0
Cu(I) , Cu20(s)
10.130-21000/T
7
b
Cu-O-Si02
Cu(I) , SiOi{s), (Cu20-Si02)slag
3.70-11630/T
8
c
Cu-Fe-O-Si0 2
Cu(I) , SiOi{s), Fe,O,(s), (Cu20 -FeO,-Si02)slag
d
Cu-Fe-0-SiO,
Cu(I), SiO,(s), Fe,O,(s), (FeO,-Si0 2)slag
e
Fe-O-Si02
Si02(S), Fe30,(s). (FeO,-Si02)slag
Fe-0
5.629-14290/T
25.11-48840/T
9
FeO(s) , Fe30,(s)
15.08-35810/T
10
g
Fe-Cu-O-Si0 2
Fe(s) , FeO(s), Cu(I), (FeO-Si02)slag
6.45-26920/T
h
Fe-0
Fe(s), FeO(s)
Fe(s), Si02(s), Cu(I), (FeO-Si02) slag
6.77-27520/T
2.94-22500/T
11
Fe-Cu-O-Si0 2
Fe-O-Si0 2
Fe(s), Si02(s), (FeO-SiO,)slag
3.67-23810/T
10
a
Fe-S
Fe(s), (Fe-S)matte
2.93-12360/T
IH\
13,-1\
Cu-S
Cu(I), Cu 2S(I)
4. 76-15900/T
Cu-S
Cu(I) , Cu2S(s)
y
Fe-Cu-S
Fe(s) , Cu(I) , (Cu-Fe-S)matte
A
Fe-S-O-Si0 2
Si0 2(s), (Fe-S-O)matte, slag, 1 atm. S02
2.60-16200/T
2.39-5320/T
B
A'
Fe-S-0-SiO,
Si0 2(s), (Fe-S-O)matte, slag, 0.1 atm . S02
1 .95-16200/T
1.23-4630/T
Fe-Cu-S-O-Si02
Si0 2(s), Cu(I), Cu 2S(I), slag, 1 aim. S02
1.86-11570/T
4.76-15900/T
B'
Fe-Cu-S-0-SiO,
SiO,(s). Cu(I) , Cu,S(I), slag, 0.1 atm. SO,
c
Fe-S-0
D
E
12
3 .67-14660/T 13, 14, 15
3.65-15740/T
4.76-15900/T
Fe(s) , FeO(s), (Fe-S-O)matte
68.9-27710/T
Fe-Cu-S-0-SiO,
Fe(s), Cu(I) , matte , slag , oxysulfide; Critical point
4.84-24 770 /T
2. 73-13890/T
Fe-S-0-SiO,
Fe(s), matte, slag, oxysulfide; Critical point
4.80-25230/T
3.34-13430/T
F
Fe-Cu-S-O-Si02
Fe(s), Cu(I) , SiO,(s), matte, slag
3.93-24070/T
3.72-15050/T
G
Fe-S-O-Si0 2
Fe(s) , Si0 2(s), matte, slag
5.18-26160//T
-3.53-4490/T
2 .83- 12500/T
MOLTEN SLAGS, FLUXES AND SALTS '97 CONFERENCE - 739
{%Cu}
-1
?""o/'
•
E
-3
Ol
-4
'o,,s-·
1473
1373
b
-,,f-,
"s'i
,:,0>~
(/)
_Q
1573
'o,,s-·
-2
Ii
a_
Temp./K
75 70 60 50 30 0
80
-4
(?c,O~
-,,f-,
e
"s'i.
-5
-6
-5
-6
E -6
I§
cf
-7
Ol
_Q
-8
A
-8
-9
-1
B
E
0
2
I§
N
a':
log {%Fe}
Ol
Fig. 7.
Oxygen and sulfur partial pressure for slag-matte
equilibrium with 0.1 or 1 atm. S0 2 pressure at
1473 or 1573K against iron content in matte.
Silica saturated FeOx-Si0 2 slag and matte are
melted in a magnesia crucible.
_Q
-10
h
{%Cu}
80
75
70 60
40
-6
0
-12
-7
E -8
rn
--.
N
a':
-9
Ol
.Q
-10
-14 _....__ _ ____,.___ _ _ _.___ _ _
6.0
6.5
7.0
7.5
~
-11
-12
104/T
-1
2
0
log {%Fe}
Fig. 8.
Fig. 9.
Oxygen partial pressure for slag-matte-copper
metal equilibrium at 1573K. Fe0x-Si02 slag
containing 18 to 21 % Si0 2 or 28 to 32% Si0 2 ,
matte and copper metal are melted in a magnesia
crucible.
740 - MOLTEN SLAGS, FLUXES AND SALTS '97 CONFERENCE
Oxygen partial pressure for univariant equilibria
in the Fe-Cu-S-O-Si0 2 and its sub-systems.
Temp./K
Temp./K
1673
1573
1573
1473
1473
1373
-1r-~~"7'-~r-~---'~--,,-~---'-~--,
A
B-----
-2
-3
-11
E
......
~N
E
cf
-4
~
a:"'
OJ
_Q
a
OJ
_Q
-12
-5
-6
-13'--~~~~~~~'--~~~~~~--'
6.0
6.5
-7
7.0
-8'--~~~~'--~~~~L-~~~__J
Fig. 10. Oxygen partial pressure for univariant equilibria
in the Fe-Cu-S-O-Si0 2 and its sub-systems under
solid iron saturation.
6.0
6.5
7.0
7.5
4
10 / T
Fig. 11. Sulfur partial pressure for univariant equilibria
in the Fe-Cu-S-O-Si0 2 and its sub-systems.
3.3
Phase stability diagram in copper smelting condition
Phase stability diagrams in more wide range of
oxygen and sulfur partial pressures are illustrated in Figs
14, 15 and 16 to comprehend copper smelting condition
thermodynamically. The lines which show constant copper
content in matte are based on the information of slag-matte
equilibrium experiment melted in a magnesia crucible. Fig.
14 and Fig. 16 are illustrated under the condition of silica
saturation at 1573 and 1473K, respectively, and Fig. 15 is
under 20% Si02 in slag at 1573K. Decrease in silica
content in slag, increase in iron content, leads to high
oxygen pressure at given copper content in slag and sulfur
dioxide partial pressure. Fig. 14, that is under silica
saturation at 1573K, is not essentially different from that
presented by Yaz.awa2 because the affinities between the
components in the quinary system are unique and decided
by Heaven. If we see some difference in the figures, that is
due to an experimental error or human mistake. We can get
large amount of important information from a phase
stability diagram. It is necessary for well understanding the
diagram to contrast its composition diagram.
4. CONCLUSION
Oxygen and sulfur partial pressures for univariant
equilibria in the Cu-Fe-O-S-Si02 system and its related
sub-system are presented . The phase stability diagrams for
the Cu-Fe-O-S-Si02 system, under silica saturation and
20 % Si02 at 1473 and 1573K, are illustrated to
comprehend copper smelting condition.
MOLTEN SLAGS, FLUXES AND SALTS '97 CONFERENCE - 741
E
E
i--s;o,sat.
1§
N
cf
1§
-11.5
N
+
(Fe·S-0)
matte
Fe(s)+Cu(l)+Si0 2 (s)
Ol
0
cf
-12.5
Si0 2(s)
t
(Fe-S-0)
matte
Ol
_Q
-13.0
-12 .0
{%Cu} 50
30
20 10 0
50 40 30 20 10 0
{%Cu}
a
-6.5
-6.0
-13.5 .___ _ ____.__ ___.____..................._.___,_.___._._---I
-6.0
-5.5
-7.0
-6.5
-5.5
log Ps 2/atm.
log Ps 2/atm.
Fig. 12. Phase stability diagram for the Fe-Cu-S-0-SiO,
system under solid iron saturation at 1573K. -
Fig. 13. Phase stability diagram for the Fe-Cu-S-O-Si0 2
system under solid iron saturation at 1473K.
(Cu 20-Si0 2)slag
-3
b + SiO,(s)
-3
slag+Cu 2S(f)+Si0 2(s)
(Cu 20-Fe0,-Si0 2 )slag+Cu(f)+Si0,(s)
-4
-4
Fe 30 4 (s)
-5
-5
+(Fe-{;u-S-O)matte
-6
-6
E
1§
E
-7
1u
--cf
N
N
cf
Ol
-8
-8
0)
_Q
.Q
0.1
-9
-9
log Pso2
-10
,
-11
'G
G
Fe(s)+(Fe-{;u-S-O)matte
-12
+Si0 2(s)
~
_5
10
10 .,
-13L----l~-L.ILI......J.....:~.L----l.~---'-~...,___.
a...._
-8
-7
-6
-5
-4
-3
-2
\
-12
-
...............
5
-13L-~..J_~-'-Yi!~'l.....Ja_1~0~·'_._~---''--~-'--~1_0~
· -""
-8
-1
-7
-6
-5
-4
-3
-2
-1
log P52 /atm.
log P 52 /atm.
Fig. 14. Phase stability diagram for the Fe-Cu-S-O-Si0 2
system under solid silica saturation at 1573K.
slag+Fe(s)+(Fe-{;u-S-O)matte
!
Fig. 15. Phase stability diagram for the Fe-Cu-S-0-Si0 2
system with 20% Si0 2 slag at 1573K.
742 - MOLTEN SLAGS, FLUXES AND SALTS '97 CONFERENCE
-4
8.
U. Kuxmann and K. Kurre, "Die Mischungsliicke im
Kupfer-Sauerstoff und ihre Beeinflussung durch die
Oxide CaO, Si0 2 , Al2 0 3 , MgO· Al20 3 und ZrOz",
ErzrnetalI, Vol. 21, 1968, pp.199-209.
9.
E.J. Michal and R. Schuhmann, "Thermodynamics of
Iron-Silicate Slags: Slags Saturated with Solid Silica",
Trans . AIME, Vol. 194, 1952, pp. 723-728.
b
(Cu 20-FeO,..Si0 2)slag+Cu(l)+SiO,(s)
Fe 30 4 (s)
-6
-7
10. E. Schiirmann and G. Kraume, "Reduktionsgleeichgewichte im System Fe-Fe 2 0 3 -Ca0 oberhalb 1050°C'',
Arck. Eisenhiittenwes, Vol. 47, No. 9, 1976, pp.471476.
E
m
-..
rf
Ol
-9
11. G.G. Charette and S.N. Flengas, "Thermodynamic
Properties of the Oxides of Fe, Ni, Pb, Cu and Mn, by
EMF Measurements", J. Electrochem. Soc., Vol. 115 ,
No. 8, 1968, pp.796-804.
.2
-10
-11
12. J. Niemela and P. Taskinen, "Activities and Phase
Equilibria in Cu-S Melts by EMF Techniques", SraruL
J. Metal!., Vol. 13, 1984, pp.382-390.
-12
-13
13. W.A. Krivsky and R. Schuhmann, "Thermodynamics
of the Cu-Fe-S System at Matte Smelting
Temperatures", Trans. AIME, Vol. 209, 1957, pp.
988-981.
log Ps2/atm.
Fig. 16. Phase stability diagram for the Fe-Cu-S-O-Si0 2
system under solid silica saturation at 1473K.
15. J. Koh and A. Yazawa, "Thermodynamic Properties
of the Cu-S, Fe-Sand Cu-Fe-S Systems", Bulletin of
the Research Institute of Mineral Dressing_and_
Metallurgy, Tohoku University, Vol. 38, No . 2, 1982,
pp. 107-118.
REFERENCES
1.
J.F. Elliott, "Phase Relationships in the
Pyrometallurgy of Copper", Metal!. Trans., Vol. 7B,
March, 1976, pp.17-33.
2.
A. Yazawa, "Thermodynamic Considerations of
Copper Smelting", Can. Met. Quart., Vol. 13, No.3,
1974, pp.443-453.
3.
Y. Takeda, "The Effects of Basicity on Oxidic
Dissolution of Copper in Slag", Metaliurgjcal Processes
for the Early Twenty-First Century, TMS, Vol. 1,
1994, pp.453-466.
4.
14. Y.A. Chang, J.P. Neumann and U.V. Choudary,
"Phase Diagrams and Thermodynamic Properties of
Ternary Copper-Sulfur-Metal Systems". INC.RA_
Monograph VII, NSRDS, 1979, pp.58-88.
Y. Takeda, "Oxidic and Sulfidic Dissolution of Copper
in Matte Smelting Slag", Proceedings of 4th
International Conference on Molten Slags and Fluxes,
ISIJ, 1992, pp.584-589.
5.
Y. Takeda, "Copper Solubility in Matte Smelting
Slag", 5th International Conference on Molten Slags,
Fluxes and Salts '97, ISS, 1997.
6.
J.P. Coughlin, "Contributions to the Data on
Theoretical Metallurgy", Bureau of Mines, Bulletin
542.
7.
Y. Kayahara, "Thermodynamic Study of the Liquid
Cu-0 system", Trans JIM, Vol. 22, No. 7, 1981,
pp.493-500 .
MOLTEN SLAGS, FLUXES AND SALTS '97 CONFERENCE - 743