Ore Geology - 3
Ore Geology - 3
The aim of any phase equilibrium study is to simulate the natural ore-forming
environments in the laboratory. But such crucible-to-nature extrapolation is often
complicated and hindered by complexity of natural mineral systems, due the ‘plague
of metastability’.
Sulfide minerals pose some special problems due to differences with their non-
sulfide counterparts in terms of (a) crystal-chemical complexities, (b) faster reaction
kinetics and consequent non-quenchability and (c) dominant control of fS2 on phase
assemblages.
Marcasite: FeS2-X
Greigite: Fe3S4
Smythite: Fe9S11
Pyrrhotite: Fe1-XS
FeS: 1C, FFFFFFF….
Fe7S8: 4C, VFVFVFVF…
Fe11S12: 6C, VFFVFFV…
Fe9S10: 5C, VFVFFVFV…
2𝐹𝑒𝑆 + 𝑆2 = 2𝐹𝑒𝑆2
𝑃𝑜 2
∆𝐺 0 = 𝑅𝑇 ln 𝑓𝑆 2 𝑎𝐹𝑒𝑆
𝑃𝑜
ln 𝑓𝑆2 = ∆𝐺 0 Τ𝑅𝑇 − 2 ln 𝑎𝐹𝑒𝑆
𝑃𝑜 ഥ𝐹𝑒𝑆
𝑃𝑜 𝑓 𝑃𝑜 𝑃𝑜
𝑑 ln 𝑓𝑆 2 ∆𝐻0 𝑑 ln 𝑎𝐹𝑒𝑆 ∆𝐻0 𝐻 − 𝐻𝐹𝑒𝑆 𝜕 ln 𝑎𝐹𝑒𝑆 𝑑𝑁𝐹𝑒𝑆
= −2 = −2 −2 𝑃𝑜
𝑑 1Τ𝑇 𝑃;∆𝐺=0
𝑅 𝑑 1Τ𝑇 𝑃;∆𝐺=0
𝑅 𝑅 𝜕𝑁𝐹𝑒𝑆 𝑃;𝑇
𝑑 1Τ𝑇 𝑃;∆𝐺=0
2
𝑁𝐹𝑒𝑆 = −6.4867 × 10−4 log 𝑓𝑆 2 − 1.5453 × 10−2 log 𝑓𝑆 2 + 0.90989
𝜕 log 𝑓𝑆 2
= −85.83𝑁𝐹𝑒𝑆 + 70.83
𝜕 1000Τ𝑇 𝑁 𝐹𝑒𝑆
log 𝑓𝑆 2 𝑇, 𝑁𝐹𝑒𝑆
1000Τ𝑇 𝜕 log 𝑓𝑆 2 1000
= log 𝑓𝑆 2 1000°𝐾, 𝑁𝐹𝑒𝑆 + න 𝑑
1 𝜕 1000Τ𝑇 𝑁 𝐹𝑒𝑆
𝑇
𝑆𝑝
𝑃 𝑘𝑏𝑎𝑟𝑠 = 42.30 − 32.10 log 𝑋𝐹𝑒𝑆 ± 0.30
𝑆𝑝
𝑃 𝑘𝑏𝑎𝑟𝑠 = 27.982 log 𝛾𝐹𝑒𝑆
ሶ − 8.549 ± 0.5
𝑆𝑝 𝑆𝑝
𝛾𝐹𝑒𝑆
ሶ is given by 𝑑𝑎𝐹𝑒𝑆 Τ𝑑𝑋𝐹𝑒𝑆
Fe-As-S system
Fe-S-O system
Sulfide partial melting
Phase equilibria in Ore bearing systems
Here some general principles in studying synthetic sulfide and oxide systems of ore
forming significance are discussed. Application of laboratory data to natural assemblages
for quantitative interpretation in the form of phase diagrams requires recognition of
independent variables. The way to achieve this goal is through the Gibbs phase rule, a
fundamental statement in chemical thermodynamics, relating the number of stable phases
(p) in an equilibrating assemblages, to the number of independent components (C) and
the number of independent degrees of freedom (f).
Mathematically the phase rule is simply an accounting of the number of equations and
unknowns, while the identities of the unknowns are totally lost in accounting. ‘f’ is the
number of conditions that one is free to specify arbitrarily without contradicting some
previously specified condition of state.
Derivation: If we represent the composition of all the phases in terms of Xi (X1, X2, X3…
etc), then there will be a Xi term of every component in every phase. Hence for ‘p’ phases
there will be ‘cp’ compositional variables. In general, there will be two additional
physical variables (P and T). Hence the total number of variables (unknowns) = cp + 2.
Since the sum of Xi in each phase = 1.0, then for ‘p’ phases, there will be p-number of
c
equations like ∑ X i = 1 . Transfer of chemical components takes place in the decreasing
i =1
1
Phase equilibria in Ore bearing systems
µ iα = µ iβ = ............µ ip (1)
For p-1 pairs, there will be c(p-1) number of the above equations making the total number
of equations = p+ c(p-1).
Now from the simple principle of linear algebra, the variance of the system of equations
= number of variables – number of equations
Hence f = cp + 2 - p-cp + c
f=c–p+2 (2)
It must be noted that the above phase rule is predicted at equilibrium with uniform P, T
and µI throughout and that the identity of each variable is lost in the process of account,
which can not be recovered by manipulation of the equations.
The major aim in studying phase equilibria is to extrapolate the stability relations of
synthetic equilibrium assemblages to natural ores. In this regard sulfide minerals pose
some special problems due to differences with their non-sulfide counterparts in terms of
(a) crystal-chemical complexities, (b) faster reaction kinetics and consequent non-
quenchability and (c) dominant control of fS2 on phase assemblages.
2
Phase equilibria in Ore bearing systems
Low isothermal compressibility is the reason behind the pressure insensitiveness of many
reactions involving sulfides.
3
Phase equilibria in Ore bearing systems
Unlike silicate systems, where P and T are the dominant variable for projecting phase
relations, in sulfide systems pressure, in general, has negligible effect, due to low
isothermal compressibility of major sulfides. On the contrary, sulfur fugacity (fS2) is the
most dominant intensive thermodynamic variable, after temperature that exerts control on
stability of sulfides.
The standard state of sulfur is conveniently chosen as ideal diatomic sulfur gas, S2, at one
bar fugacity and the temperature of interest. This state is used in spite of the fact that the
same is not physically attainable due to condensation of solid or liquid sulfur below
614°C. Therefore fS2 is numerically equal to PS2 (and aS2) in atmosphere. The choice of
this standard state is dictated because otherwise the sulfidation curves (discussed later)
will otherwise bend at the melting, transition and boiling point of the yellow colored
orthorhombic sulfur, stable at the normal P-T condition.
Non-aqueous system
Generally sulfide systems are experimented by the standard evacuated silica tube
technique, i.e., in ampules made of fused quartz tubes. Inside these ampules a S2- rich
vapor is a stable phase throughout the experiment. Hence pressure is fixed by the vapor
pressure of the condensed phases present. Theoretically such diagrams are polybaric, but
because most solids and liquids are relatively incompressible, the variation has little
effect on phase relation and the phase rule becomes f = c –p +1.
4
Phase equilibria in Ore bearing systems
P – T diagram
These diagrams are drawn on the basis of Clausius- Clapeyron’s eqn. Given by
dP ∆S ∆H
= = (3)
dT ∆V T∆V
Fig.2 P – insensitiveness
of sulfide-type reactions
in P - T space.
fS2 – T diagrams
Sulfide minerals can be easily represented by sulfidation reactions, involving S2, for their
formation. Therefore, fS2 becomes the driving force in sulfide petrology, after
temperature. Many sulfide equilibria can be represented by generalized sulfidation
reactions such as
5
Phase equilibria in Ore bearing systems
where MxSy+ and MxSy+1 are solids (x = 1, 2 ,3 and y = 1, 2, 3 etc.). If the solids are in
their standard state, their activities will be unity and the following relation hold good at P
=1bar.
∆H 0 ∆S 0
log fS2 = − (6)
2.303RT 2.303R
d log fS 2 ∆H 0
= (7)
d (1 / T ) 2.303R
From eqn. (7) it is evident that plots of log fS2 against 1/T for a sulfidation reaction will
be linear with a slope of ∆H°/2.303R (Fig.3). If one or both the solid phases (sulfides) are
not in their standard states, the sulfidation curve will plot as a curved line, as a
consequence of including their activities as well as fS2 in eqn. (4). Accordingly, eqn (7)
becomes
6
Phase equilibria in Ore bearing systems
fS2 –1/T diagrams can be used as “sulfidation grids” as emphasized by Barton (1974),
analogous to P-T or petrogenetic grids in metamorphic petrology, for deciphering
conditions of ore formation by outlining stability fields of naturally occurring
assemblages, in general. Some relevant sulfidation curves are given in Barton and
Skinner (1979) and numerical problems are furnished at the end of this chapter. It must
be emphasized here that although fS2 is the second most dominant intensive variable (after
temperature) in sulfide petrology, it is also the least known one. However, temperature
can be estimated by various tools (fluid inclusions, S-isotope etc) and fS2 can be thus
estimated as shown in Fig. 4.
Fig.4 log fS2 –1/T diagram showing several sulfidation reactions. The shaded
region indicates the maximum variation in fS2 for the assemblage pyrite +
chalcopyrite + magnetite + bismuthinite (Bi2S3) at temperature range of 250°–
300°C.
7
Phase equilibria in Ore bearing systems
fS2–fO2 diagrams
( f O 2 )2
From (C) ∆G° = –RTln K = –RT ln 3
(9)
( f S 2 )
( f )3
From (D) ∆G° = –RTln K = –RT ln O 2 4 (10)
( f S 2 )
∆G o = − RT ln K − ( P − 1)∆VS (11)
Given the standard free energy change of the above two reactions at any particular P-T,
eqns. (9) and (10) can be solved for using eqn. (11) at different values of fS2 and fO2, in
each case, and the phase boundaries can be plotted in an fO2- fS2 diagram. Some
calculation is given later.
Aqueous systems
Eh-pH diagrams
The diagrams are constructed for low temperature redox processes, on the basis of the
Nernst’s equation, which is given by
RT ∆G 0 RT
Eh = E° + ln K = + ln K (12)
nF nF nF
8
Phase equilibria in Ore bearing systems
where E° is the standard electrode potential, F = Faraday’s constant (23061 cal V-1), n =
number of electrons transferred in the redox reaction, K is the equilibrium constant and
∆G° is the standard free energy change of the reaction. At 25°C (298.15K) and 1 bar, eqn
(12) is reduced to
0.059
Eh = E° + log K (13)
n
Some worked out problems pertaining to construction of Eh-pH diagrams for low
temperature ore forming processes are given at the end of this chapter.
This is one of the most studied sulfide system, both in mineralogy and metallurgy. Pyrite
and pyrrhotite are the only two sulfide minerals to be called rock-forming minerals. The
Fe-S system is also an important bounding binary in many multi-component systems such
as Fe-Zn-S, Fe—Ni-S, Cu-Fe-S, Fe-As-S, Cu-Fe-Zn-S etc.).
The phase relation above 400°C is clear and straight forward (Fig. 5). The central
portion of the system is dominated by the large, high- temperature pyrrhotite solid
solution field constitutes extreme solid solution from stoichiometric FeS toward more S-
rich compositions. This high-T form with hexagonal NiAs-type structure accommodates
solid solution by random vacancies on the Fe sites within the lattice. Hence, the
composition of high temperature pyrrhotites, except for FeS, is best given as Fe1-XS. The
maximum thermal stability of the pyrrhotite solid solution at pressure = 1 bar is 1192°C,
above which it melts incongruently. The eutectic between pyrrhotite and Fe is at 988°C,
56% Fe. Brett and Bell (1969) found that the eutectic temperature is P-insensitive up to
30 kbar but then rises rapidly, reaching 1160°C at 100 kbar. All the possible minerals
(and phases) that are stable in the Fe-S system are given in table 1.
9
Phase equilibria in Ore bearing systems
Phase equilibria of the S-poor portion of the Fe-S system, unimportant for
terrestrial ore forming processes, do apply directly to lunar and meteoritic samples and
iron smelting. Rapid cooling of Fe-rich melts produces complex eutectic type
intergrowths comprising of cruciform Fe crystals in a FeS matrix, but during slow
cooling such textures are often transformed into anhedral masses with no evidence of
their original complexity. On the S-rich side of pyrrhotite, the minimum temperature of
existence of a sulfide liquid is 1088°C; above this temperature, there is a broad field of
liquid immiscibility comprising of a sulfide-rich and a sulfur-rich liquids.
Pyrite, the most abundant sulfide in the Earth’s crust, has a maximum thermal
stability of 742±1°C, above which it incongruently melts to pyrrhotite and an S-rich
liquid. The dT/dP slope of the above temperature is ∼14°C per kbar confining pressure.
The common occurrence of pyrite + pyrrhotite along with temperature dependence of
pyrrhotite composition in the above assemblage, led Arnold (1962) to propose the
pyrrhotite geothermometers (Fig.5). Arnold (op. cit) determined the pyrrhotite
composition by on the pyrite-pyrrhotite solvus by shift the d102 of hexagonal pyrrhotite.
The X-ray spacing curve was subsequently revised by Toulmin and Barton (1964) and
Yund and Hall (1969) resulting in the following expression.
10
Phase equilibria in Ore bearing systems
This method still remains as the most convenient accurate means of determining
pyrrhotite composition. However, utility of the pyrrhotite thermometry is only restricted
to samples where rapid equilibration between pyrite and pyrrhotites (eg. in basalts) takes
place. Otherwise, during slow cooling, the pyrrhotite composition slides down the solvus
(Fig.6) to a more Fe-rich composition. A further complication in most ore deposits is the
inversion to monoclinic pyrrhotite either by cooling below 254°C or as a low-temperature
oxidation product.
11
Phase equilibria in Ore bearing systems
Table 1 Minerals and phases in the central portion of the Fe–S system
12
Phase equilibria in Ore bearing systems
Phase relations at low temperature (< 350°C) in the Fe-S system are not convincingly
understood (Vaughan and Craig, 1978; 1997), in spite of some rigorous attempts (Fig.7,
modified from Kissin and Scott, 1982).
n FeS
Po
X FeS = (15)
n FeS + n S 2
Thus XFeSPo is equal to twice the atomic % Fe in pyrrhotite, divided by 100 (e.g., XFeS =
0.98 is equal to 49 atomic % Fe in pyrrhotite). Using electrum tarnishing method Barton
and Toulmin (op.cit.), determined the variation of fugacity of sulfur gas in equilibrium
with pyrrhotite over a wide temperature range (Fig.8) and furnished the following
expression (eqn. 16).
13
Phase equilibria in Ore bearing systems
𝑃𝑜
1000 𝑃𝑜 1⁄2
𝑙𝑜𝑔𝑓𝑆2 = �70.03 − 85.83𝑋𝐹𝑒𝑆� � − 1� + 39.3�1 − 0.9981𝑋𝐹𝑒𝑆� − 11.91 (16)
𝑇
The results are shown in a logfS2–1/T plot (Fig.8), in which the pyrrhotite field is
contoured in terms of XFeSPo. The most significant achievement of Barton and Toulmin
Po
(op.cit.) pertains to calculation of a FeS , using the Gibbs-Duhem equation for the FeS-S2
binary, which can be written as
X S2 = X S2
X S2
Hence, log a FeS = − ∫
X S 2 =0
X FeS
d log f S 2 (18)
Using eqn. (18), Toulmin and Barton (1964) deduced aFeS as follows
1000
log a FeS = 85.83 − 1 1 − X FeS
Po
(+ ln X FeS
Po
) (
+ 39.30 1 − 0.9981X FeS
Po
)
1/ 2
T
(
− 39.23 tanh −1 1 − 0.9981X FeS
Po
)
1/ 2
− 0.002
(19)
Fig.9.Variation in aFeS in
the FeS–S2 system.
14
Phase equilibria in Ore bearing systems
Following eqn. (18), isopleths of aFeS in both the pyrrhotite and pyrite fields are plotted,
and shown in Fig.9.
Fig.9 furnishes results that have important implications in sulfide petrology. Since
solid solubility in pyrrhotite is due to omission of Fe atoms, it has been commonly
believed that aFeS in pyrrhotite is very close to unity. But as seen from Fig.9, aFeS
decreases with decreasing Fe content. The sulfur-rich boundary of the pyrrhotite field is
defined by the reaction 2FeS+S2 = 2FeS2, but aFeS ≠ 1; it varies between 0.4 and 0.5. Scott
(1974) concluded that for pyrrhotite in equilibrium with pyrite below 500°C, aFeS is
constant at 0.48. Scott and Barnes (1971), considering the above sulfidation reaction,
In addition to phases encountered in the Fe-S systems, only new phases that appear in this
ternary are sphalerite and wurtzite polytypes. Sphalerite has a cubic closed packed (ccp)
structure in which every alternate tetrahedral site is empty. Iron, along with Mn and Cd
enters the structure by substituting for Zn and results in an increase in cell volume.
Because of its refractory nature and wide variation in Fe content, sphalerite has been tried
as a geochemical sensor for ore forming environment.
Kullerud (1953) first proposed that the Fe content in sphalerite (in equilibrium
Sp
with pyrrhotite) could be used as a geothermometer, with the basic assumption that a FeS
does not depart very much from unity. Later from the work of Toulmin and Barton
(1964) it was confirmed that the underlying assumption of Kullerud was incorrect (Fig. );
hence the geothermometric concept became invalid.
Fig. 10 shows the schematic isothermal section of the ZnS-FeS-S portion of the
system, where extensive solid solution of FeS in ZnS and its temperature dependence on the
coexisting Fe- sulfide(s) is seen. Fig. 11 shows the T-X projection of the ZnS-rich portion of
the system ZnS-FeS-S, from the S- apex onto the ZnS-FeS binary. At 742°C, in the T - X
section of Zn-Fe-S system, pyrite incongruently breaks down to Po + S- rich liquid (Fig.11).
15
Phase equilibria in Ore bearing systems
From this Point (I), three invariant curves emanate: (1) Sp + Po + S (1), (2) Sp + Py + Po and
(3) Sp + Py + S (1).
Fig. 10 Isothermal 1-bar phase relations in the Fe-Zn-S system. The sphalerite and
pyrrhotite solid solutions are shown as the bold line on the FeS-ZnS and Fe-S
binaries respectively. Note the tie lines with various Fe- sulfides at different
temperatures. See text for discussion.
16
Phase equilibria in Ore bearing systems
Fig. 11 T-X projection of the Zn-S rich portion of the system FeS-ZnS-S at
one bar pressure.
Fig. 12 Activity of FeS in
sphalerite as a function of its
composition for a temperature
range ∼225°C – 1100°C.
17
Phase equilibria in Ore bearing systems
Assemblage Sp + Po
Sp
a FeS − X FeS
Sp
relation : From the work of Barton and Toulmin (1966) it was established that
Sp Sp
a FeS is a function of X FeS over a wide range to temperature, (except for very high Fe-rich Sp)
as given by the relation
aFeS = 0.0257 (mole % FeS) – 0.00014 (mole % FeS) (20)
Hence sphalerite composition is a reliable indicator of aFeS (Fig. 12).
Sp
X FeS as a function of T and fS2 (in equilibrium with pyrrhotite) has been determined by Po-
indicator method as shown in Fig.13 . A least square regression (trend) surface fitted to the
experimental data gives the following relation
15900.5
Mole % FeS in Sp = 72.26695 − + 0.01448 log f S 2
T
108 7205.5
− 0.38918 2 − log f S 2 − 0.34486 (log f S 2 ) (esd = 1.7) (21)
2
T T
The effect of pressure on the isopleths is complicated due to rise in γ FeS
Sp
with pressure making
Fig. 13 logfS2 -T- X diagram for the Fe-Zn-S system, showing the composition of
sphalerite in equilibrium with Fe- sulfides. Dashed isopleths in the pyrite field
theoretically calculated, using eqn. (23), given in the text later.
18
Phase equilibria in Ore bearing systems
Sp
ess X FeS at the same temperature. Therefore, without additional information on ∫S2 and
pressure, the isopleths shown in Fig 13 have little use in geothermometry. Even if the P-
effect can be calculated, pyrrhotite might have suffered post-depositional changes, making it
an unreliable indicator of fS2.
Assemblage Sp + Py + Po
Addition of pyrite to the assemblage sphalerite + hex. pyrrhotite buffers aFeS (due to the
reaction 2FeS + S2 = 2FeS2) and overcomes the problem of having to preserve pyrrhotite
composition from high temperature. For this reason and the widespread occurrence of
this assemblage in nature, the T - X relationship is concentrated on the Sp + Py + HPo
solvus (Fig.11). The solvus remains vertical below 600°C at 20.7 ± 0.6 mole % FeS.
Hence, for the major portion of geological temperature range this assemblage can not be
used as a geothermometer.
Sphalerite Geobarometer
Sp
VFeS is large, compared to VZnS as evidenced by substantial increase in cell edge of sphalerite
with Fe content (Fig. 14). As a consequence, we should expect that in a system where aFeS is
buffered by Py + Hpo, sphalerite should become less Fe-rich with increasing pressure. Scott
(1973) measured the composition of sphalerite on the pyrite-pyrrhotite solvus by
hydrothermal recrystallization experiment over a wide range of P-T and found that sphalerite
isobars do indeed shift to lower FeS contents as shown in Fig. 15 and is given by the relation
(Hutchinson and Scott, 1980).
The significant feature is the vertical slope of the isobars which remain so until they encounter
the slope reversal of pyrite-pyrrhotite solvus (upper stability limit of Mpo) which is below
300OC at 5 Kbar. Therefore this geobarometer is T-independent within much of the
geologically important P-T ranges (Fig. 15).
19
Phase equilibria in Ore bearing systems
a b
Fig. 15 Sphalerite geobarometry shown as T –X sections along the FeS-ZnS join of the
system ZnS-FeS-S for the sphalerite + pyrite + hex. pyrrhotite equilibrium as a function
of pressure. T-independent portion of the isobars in (b) is shaded.
20
Phase equilibria in Ore bearing systems
Assemblage Sp + Py
For the assemblage sphalerite + pyrite, the fS2 – T relationship can be calculated using the
following reaction
FeS (in Sp) + 0.5 S2 = FeS2
For the above equation, considering a value of 2.3 for γ FeS
Sp
Scott and Barnes (1971) gave the
following relation
Sp
log X FeS = 6.65 –7340/T – 1/2 log fS2 (23)
Using equation (23), and from the Fe-content in sphalerite (in association with pyrite), fS2 – T
curves can be constructed (Fig. 13).
21
Phase equilibria in Ore bearing systems
Assemblage Sp + Tr + Fe
Sphalerite composition, buffered by troilite (Tr) and Fe (2Fe + S2 = 2FeS) have been
measured by Barton and Toulmin (1966). The buffered assemblage fixes aFeS at unity for all
temperature. The effect of pressure on Sp + Tr solves is large, as shown in Fig. 17.
d ln γ FeS
Sp
∂ ln γ FeS
Sp
V − Sp − V −Tr
Again =[ ]T , X Sp = FeS
dP ∂P FeS
RT
1 da FeS
Tr
For troilite, a FeS = 1, γ FeS
Tr
= Tr
and =0
X FeS dP
Sp
dX FeS (V − Sp − V −Tr ) X FeS
Sp
Hence, = − FeS
dP RT
− Sp
(VFeS − V −Tr ) X FeS
Sp
Sp
Or dX FeS = −[ ]dP (27)
RT
The integral form of eqn. (27) can be written as
(V − V ) * ( X FeS
Sp
)
Sp
( X FeS ) P2 − ( X FeS
Sp
) P1 = −[ ]( P2 − P )1 (28)
RT
P1 + P2
Where P1 and P2 are low and high pressure respectively and * refers to P = . The
2
shift in the isobars of the Sp + Tr + Fe solves is shown in Fig.17. However, its
applications are only limited to meteorites. Some meteorites contain small sphalerite
blebs within nearly pure troilite nodules surrounded by kamacite (Fe, Ni). Schwarcz and
22
Phase equilibria in Ore bearing systems
Scott (1974) and Schwarcz et al (1975) used eqn. (28)/ Fig. 17 as a paleomanometer to
estimate the size of the parent asteroids that spawned the meteorites.
The phases occurring in the system Cu-Fe-S are found in many geological environments and
in lunar and meteoritic materials. Phase relation is not properly understood in spite of
extensive work, primarily due to, large number of phases, extensive solid solution, non-
stoichiometry, non-quenchability and metastability. Fig. 18 shows the different phases
reported in this system.
23
Phase equilibria in Ore bearing systems
Phase equilibria at 400°C and above (Figs. 19 and 20) are dominated by three solid solutions,
which are (1) chalcocite - digenite - bornite (2) intermediate solid solution (Iss) and (3)
pyrrhotite
24
Phase equilibria in Ore bearing systems
solid solution. The Cc - Di - Bn solid solution represents a ternary extension of high-Dg solid
solution (in the Cu-S system) into the Fe-apex of the Cu-Fe-S ternary, towards composition
more Fe-rich and S-rich than the stoichiometric bornite (Cu5FeS4). However, compositional
gradation exists between Fe-poor (chalcocite) and Fe-rich (bornite) end members, which is
readily identified in striking difference in their optical properties. The Iss field encompasses
several minerals in the central portion of the ternary (talnakhite, mooihoekite and haycockite)
the absence of complete solid solution between Cc-Di-Bn and the Iss, in spite of their
similarity of structure and cell size, suggests differences in the type of bonding. The Iss has a
cell edge of 5.4Å, sphalerite-type, fcc unit cell whereas chalcopyrite cell is same excepting
doubling in the C-axis direction. The pyrrhotite solid solution is the ternary extension of high
temperature Fe1–XS, the Cu-content is temperature dependent, with a maximum of 4.5%
(Yund and Kullerud, 1966).
Fig. 20 Phase relations in the central portion of the Cu-Fe-S system at 400°C and 300°C
(schematic).
The high temperature solid solution fields shrink with decreasing temperature (Fig.21)
and cation ordering produces nearly stoichiometric phases at low temperature. On cooling
below 557°C, chalcopyrite appears as an ordered (tetragonal) phase. Above this temperature
the mineral transforms to a cubic Iss phase (and small amount of pyrite). The orthorhombic
polymorph of cubanite (CuFe2S3) is only stable up to ≈210°C (Cabri et al., 1973) above which
it inverts to cubic Iss. Natural intergrowths of chalcopyrite-bornite, chalcopyrite-cubanite;
and talnakhite (Cu9Fe8S16), mooihoekite (Cu9Fe9S16) and haycockite (Cu4Fe5S8) are likely
formed as decomposition products of initially deposited high temperature Iss.
25
Phase equilibria in Ore bearing systems
At 1000°C, the Fe–Ni–S system is dominated by a large liquid field (Fig. 21a) and a solid
solution field extending from pyrrhotite (Fe1-xS) towards nickel monosulfide (Ni1-xS). At
temperatures <992°C, the monosulfide solid solution (mss) becomes complete and
dominate the central portion of the system. At 650°C, as shown in Fig. 21b, the mss
coexists with (Fe,Ni)S2 or (Ni,Fe)S2 on the S-rich side or (Ni,Fe)3±xS2 or γ– or α–(Fe,Ni)
on the metal rich side. On further cooling, the mss filed narrows down and pentlandite
appears below 610°C on the metal rich side and violarite on the S-rich side (Fig. 21c).
Pentlandite is therefore entirely a subsolidus phase, formed by exsolution of the mss.
The other phases in the system at 400°C are disulfides (Ni7S6 and Ni3S2), that have
limited solid solubility with various (Ni,Fe) alloys (Fig. 21c). The mss begins to unmix
below 300°C and within the uncertainties, the schematic phase relation at 250°C is shown
in Fig. 21d. Here the mss is shown breaking into three segments near 200°C. Below 212°
± 13°C, the pentlandite-pyrite tie line is established. Schematic phase diagram at 25°C is
shown in Fig. 21d, solely from the study of natural assemblages. Although the pentlandite-
violarite natural assemblage is observed, it is generally metastable. Hence, the preferred tie
lines are shown by joining millerite with pentlandite and pyrite.
In the Fe-Ni-S system, kinetics of pentlandite exsolution from mss has been a
subject of in depth study. Pentlandite can form below 610°C by exsolution from mss, but
the onset of unmixing depends on the mss bulk composition in relation to equilibrium mss-
pentlandite solvus (Fig. 22). For example, a composition just within the mss at 600°C will,
in theory, begin to exsolve pentlandite at temperature little below 600°C. At such high
temperatures, diffusion rates will be too high but the driving force for nucleation that
depends on the degree of undercooling below the solvus will be small. Under such
26
Phase equilibria in Ore bearing systems
a
c d
e
Fig.21 Phase relations in the Fe-Ni-S systems
shown by isothermal sections at 1000°C (a),
650°C (b), 400°C (c), 250°C (d) and 25°C
(e). Abbreviations: mss = monosulfide solid
solution, tr = toilte (FeS), vs = vaesite (NiS2),
v-p = violarite-polydimite (FeNi2S4-Ni3S4),
pentlandite (Ni,Fe)9S8, ml = millerite (NiS),
m-po = monoclinic pyrrhotite (Fe7S8), h-po =
hexagonal pyrrhotite (Fe1-xS), gs =
godlevskite (Ni7S6), heazlewwodite (Ni3S2).
27
Phase equilibria in Ore bearing systems
conditions Ni (and necessary additional Fe) will diffuse to grain boundaries where
heterogeneous nucleation will lead to the formation of granular pentlandite at interstices
between grain and along grain boundaries (Fig. 23a). Slow cooling from these high
temperatures (or annealing) will generate this pentlandite structure. At the other extreme,
where equilibrium phase diagram shows that exsolution should not start until a much
lower temperature (at lower diffusion rates) or where cooling is rapid, a greater degree of
undercooling is necessary to cause unmixing. Here, homogeneous nucleation may lead to
oriented flames (Fig. 23b).
28
Phase equilibria in Ore bearing systems
a b
Fig. 23 Two textural varieties of Ni- bearing pyrrhotite ores: (a) granular pentlandite in
pyrrhotite and (b) flame-like intergrowth of pentlandite in pyrrhotite. See text for
discussion.
The Fe – As – S system
The Fe-As-S system contains only one ternary phase, arsenopyrite, FeAsS. Although less
abundant than many Fe, Cu-Fe, and Fe-Ni sulfides, arsenopyrite is a common,
paragenetically early accessory mineral in hydrothermal ore deposits and has the great
advantage of being relatively refractory. The general topology of the Fe-As-S system has
29
Phase equilibria in Ore bearing systems
been studied by Clark (1960), Kretschmar and Scott (1976). Barton (1969) derived
thermodynamic data for many of the binary and ternary phases. Of most interest is the
compositional variation of arsenopyrite, i.e., in terms of As/S ratio on the FeS2 (pyrite) –
FeAS2 (löllingite) join, and its dependence on formation (or equilibration) temperatures.
Phase relations in the Fe-As-S system are schematically shown over temperature ranges
688° to 363°C in Fig. 24. In the high temperature portion, arsenopyrite coexists with
pyrrhotite, löllingite, As, or an (As,S) liquid (represented by cooling by an arsenic sulfide)
in assemblages #1 through 4 in Fig. 24. Between 491° and 363°C, arsenopyrite can coexist
with pyrite, pyrrhotite, löllingite, As, or (As,S) liquid in the assemblages #5 through 8.
30
Phase equilibria in Ore bearing systems
Since arsenopyrite composition is a function of both temperature and fS2, the same
can be used a geothermometer, if arsenopyrite occurs in an fS2- buffered assemblage
(labeled as 1 to 8 in Fig. 24). The common ƒS2 – buffered univariant assemblages include
Asp + Lo + Po, Asp + Py + Po, Asp + Py + As, in which atomic %As in arsenopyrite
uniquely fix both T and fS2 (Fig. 25). On the other hand, ƒS2 – unbuffered assemblages
furnish a range in T and fS2. Composition of arsenopyrite can be readily determined using
d(131) X-ray spacing curve following eqn. (29), proposed by Kretschmar and Scott (1976).
However, the mineral is commonly zoned and spot analysis is necessary.
The effect of pressure on composition of arsenopyrite buffered by two other phases is not
too large and can be ignored for low-P hydrothermal deposits, but has to be taken into
account for in metamorphosed ore deposits.
The Fe – Zn – As – S system
31
Phase equilibria in Ore bearing systems
The Cu – Fe – Zn – S system
Mineral assemblages representative of the Cu-Fe-Zn-S system are more common in
hydrothermal ore deposits than those in the arsenide- bearing system but, in contrast,
commonly undergo post-depositional reequilibration. Not only the Cu- and Cu-Fe-sulfides
readjust their composition during cooling, presence of minerals such as chalcopyrite promotes
changes in the otherwise refractory sphalerite. A very common feature in Zn- and Cu-bearing
hydrothermal veins, volcanogenic, and metamorphosed massive sulfide deposits is the texture
of dispersed blebs and rods of chalcopyrite in sphalerite (Fig. 27a). The term ‘chalcopyrite
32
Phase equilibria in Ore bearing systems
disease’, first introduced by Barton (1978) and described in greater detail by Bethke and
Barton (1987), has become a very common feature in sulfide petrology. This texture has been
notionally thought to be an exsolution product. However, experimental studies in the Cu-Fe-
Zn-S system (Hutchison and Scott, 1980; Wiggins and Craig, 1980) have demonstrated that
up to 500°C, solubility of chalcopyrite in sphalerite is negligible. Further, natural sphalerites
containing chalcopyrite disease occur in ores such as carbonate- hosted Pb-Zn ores (100°–
150°C) and unmetamorphosed VMS ores (200°–300°C). Hence, the existing interpretations of
for origin of chalcopyrite disease are by (i) epitaxial growth during formation of sphalerite or
(ii) selective replacement of Fe- rich zones in sphalerite by a Cu-bearing chloride fluid,
following a representative reaction (eqn. 30)
a b
Sphalerite occurs as stars or crosses, texturally termed sphalerite stars (Fig. 27b), which has
been interpreted to have been formed due to exsolution, because of appreciable solubility of
ZnS in CuFeS2.
33