MSS + Liquid in the System Fe-Ni-S

The Fe-Ni-S system is fundamental to understanding magmatic sulfide ore deposits. There has been a lot of work done on this system. Workers at the Geophysical Laboratory of the Carnegie Institution did the important early work, which Kullerud et al. summarized in their classic paper of 1969. This paper established the phase diagram of reference still in use by many geologists today. The metallurgists in Austin Chang's group at the University of Wisconsin rigorously and successfully quantified the thermodynamics of the Fe-Ni-S ternary, using an associated solution model for the liquid, coupled with a subregular solution model for the mss. Theirs is the work of reference in the Bulletin of Alloy Phase Diagrams (1982).
Unfortunately, Kullerud was not able to analyze his results with the electron microprobe, relying instead on carefully calibrated x-ray diffraction (XRD) methods. His tie-lines, which describe the partitioning of nickel between mss and liquid, can be shown to be egregiously wrong, and in some cases drawn in imaginatively by a draftsman with a ruler, in lieu of data.
Accurate tie-line data is crucial to any valid thermodynamic model for this system. The diagrams of Chang and Hsieh (1986) illustrate the power of such modeling, since they were able to get much closer to reality using a variety of data to calibrate their model, none of it being mss-liquid phase equilibrium data.
The liquid compositions on the sulfur-rich side (top half) of the diagram remain problematic. These are impossible to quench properly, and will be quite sensitive to total pressure. The correct answer at 1 bar is likely to be somewhere between the minimum bounds established by Ebel and Naldrett (1996), and the curve published by Chang and Hsieh (1986).

Fe-Ni-S at 1373K: comparison of results

Comparison of MSS-Liquid Tielines in Fe-Ni-S at 1373K
Sources of results, and method used to obtain them:
...Kullerud et al. (1969) analysis of XRD patterns of experiments
...Chang and Hsieh (1986) calculations using thermodynamic model
...Ebel and Naldrett (1996) microprobe analysis of experiments

The join between FeS (troilite, yellow square) and NiS (millerite) is illustrated in yellow for reference.
The stability field of MSS, and tie-lines to coexisting sulfide liquid, at 1373 K are illustrated here. The data for Kullerud et al. (1969) and Chang and Hsieh(1986) were read from their published figures. Sulfur contents in the Ebel and Naldrett (199) experimental liquids at high sulfur content are minimum bounds, because such liquids do not retain sulfur when quenched from high temperature.
Our new data on the partitioning of Ni between MSS and liquid establish the first accurate partition coefficients to be determined experimentally for this system. Similar data have been obtained for Cu.

----------------------------- -----------------------------

Experimental Results at 1373 K
The field of MSS extends to about 13.2 weight % nickel. The starting compositions are shown by triangles for experiments yielding homogeneous (one phase) results, which are shown as red squares. The light blue diamonds are starting compositions for two-phase results, which are shown as dark blue diamonds.
All the results are averages of many data points on each phase, which have the standard deviation shown by the large gold diamonds around each point. All the compositions have been normalized to total 100wt%. This figure illustrates the experimental data of Ebel and Naldrett (1969) in greater detail.
The tie-lines between MSS and Sulfide Liquid are revealed by the compositions of coexisting material in the experiments. Some bulk compositions in the interior of the MSS field were run to make sure the technique works properly, and the results (homogeneous MSS) show the accuracy of the microprobe technique and the reliability of the experimental method. Some experiments were run in duplicate, and they can be seen to yield results which are identical within one standard deviation of the microprobe results.

----------------------------- -----------------------------

Experimental Results at 1373 K

Detail of High Sulfur Experiments
The rest of this section is under construction.
Some experiments on the sulfur-rich side of the system, starting above 41 wt% S, yielded homogeneous sulfide liquid, but this liquid was missing sulfur. The reason for this is that sulfur forms vapor at the temperature of the experiment, and this vapor condenses on the wall of the tube when the experiment is quenched. Furthermore, some of the sulfur in the liquid at high temperature probably escapes the liquid as it is being quenched, regardless of how fast it is quenched. This sulfur loss is not a problem for experiments with less than about 39 wt% S, and it can be seen that the bulk compositions of these experiments lie very close to the tielines joining the results.

----------------------------- -----------------------------

References:

Chang, Y. A., and Hsieh, K-C. (1986) Thermochemical description of the ternary iron-nickel-sulphur system. In Nickel Metallurgy: volume 1: Extraction and Refining of Nickel, E. Ozberk and S.W. Marcuson, eds., Proceedings of the 25th annual conference of metallurgists, Canadian Institute of Mining and Metallurgy, p.248-277.

Ebel, D.S. and Naldrett, A.J. (1996) Fractional crystallization of sulfide ore liquids at high temperature. Economic Geology 91: 607-621.

Hsieh, K.C. Chang, Y.A., and Zhong, T. (1982) The Fe-Ni-S system above 700 C (iron-nickel-sulfur). Bulletin of Alloy Phase Diagrams, 3: 165-172.

Hsieh, K.-C., Vlach, K. C., and Chang, Y. A., 1987a, The Fe-Ni-S system I. A thermodynamic analysis of the phase equilibria and calculation of the phase diagram from 1173 to 1623 K: High Temperature Science, v. 23, p.17-37.

Kullerud, G., Yund, R.A., and Moh, G.H. (1969) Phase relations in the Cu-Fe-S, Cu-Ni-S and Fe-Ni-S systems. pp. 323-343 in Magmatic Ore Deposits (ed. H.D.B. Wilson), Economic Geology Publishing Co., Lancaster, PA.

Naldrett, A. J. (1989) Magmatic sulfide deposits. Oxford University Press, New York, 186 p.

Back to:
top of this page
D.S. Ebel home
Earth and Planetary Sciences at A.M.N.H.
A.M.N.H. home

Last revised: 26-Jan-2008 (DSE)

Comparison of MSS-Liquid Tielines in Fe-Ni-S at 1373K Sources of results, and method used to obtain them: ...Kullerud et al. (1969) analysis of XRD patterns of experiments ...Chang and Hsieh (1986) calculations using thermodynamic model ...Ebel and Naldrett (1996) microprobe analysis of experiments The join between FeS (troilite, yellow square) and NiS (millerite) is illustrated in yellow for reference.	The stability field of MSS, and tie-lines to coexisting sulfide liquid, at 1373 K are illustrated here. The data for Kullerud et al. (1969) and Chang and Hsieh(1986) were read from their published figures. Sulfur contents in the Ebel and Naldrett (199) experimental liquids at high sulfur content are minimum bounds, because such liquids do not retain sulfur when quenched from high temperature. Our new data on the partitioning of Ni between MSS and liquid establish the first accurate partition coefficients to be determined experimentally for this system. Similar data have been obtained for Cu.
-----------------------------	-----------------------------
Experimental Results at 1373 K The field of MSS extends to about 13.2 weight % nickel. The starting compositions are shown by triangles for experiments yielding homogeneous (one phase) results, which are shown as red squares. The light blue diamonds are starting compositions for two-phase results, which are shown as dark blue diamonds. All the results are averages of many data points on each phase, which have the standard deviation shown by the large gold diamonds around each point. All the compositions have been normalized to total 100wt%.	This figure illustrates the experimental data of Ebel and Naldrett (1969) in greater detail. The tie-lines between MSS and Sulfide Liquid are revealed by the compositions of coexisting material in the experiments. Some bulk compositions in the interior of the MSS field were run to make sure the technique works properly, and the results (homogeneous MSS) show the accuracy of the microprobe technique and the reliability of the experimental method. Some experiments were run in duplicate, and they can be seen to yield results which are identical within one standard deviation of the microprobe results.
-----------------------------	-----------------------------
Detail of High Sulfur Experiments The rest of this section is under construction.	Some experiments on the sulfur-rich side of the system, starting above 41 wt% S, yielded homogeneous sulfide liquid, but this liquid was missing sulfur. The reason for this is that sulfur forms vapor at the temperature of the experiment, and this vapor condenses on the wall of the tube when the experiment is quenched. Furthermore, some of the sulfur in the liquid at high temperature probably escapes the liquid as it is being quenched, regardless of how fast it is quenched. This sulfur loss is not a problem for experiments with less than about 39 wt% S, and it can be seen that the bulk compositions of these experiments lie very close to the tielines joining the results.
-----------------------------	-----------------------------