University
stringclasses 19
values | Text
stringlengths 458
20.7k
|
|---|---|
ADE | Chapter 7-Spatial Geochemical Models 253
According to Figure 7.11, there are high geometrical similarity among pair
geochemical haloes of Bi-Pb, Zn-Cu, S-Fe and As-Ag and the dispersal shape of Cd has
the lowest similarity level (98.33 %) with other geochemical haloes followed by the
geochemical halo pair As-Ag with 99.02 %.
Final partitions based on three groups are:
1. Group 1 (red lines): Bi, Sb, Pb, Zn, S, Fe and Cu,
2. Group 2 (blue line): As and Ag, and
3. Group 3 (green lines): Cd.
7.6 Summary
This chapter quantitatively evaluates the spatial distribution and geochemical
variability of the Western Mineralisation using image processing of 80 cross-sections for
10 three-dimensional block models. Apart from this research, no spatial modelling has
been used for identification of the geochemical zonation haloes for any of the orebodies in
the Broken Hill deposits.
The first aim of this study was to evaluate the spatial distribution of primary
geochemical haloes in the Western Mineralisation. The results show that the distribution
shape of Sb and Bi are stronger and broader than other elements at different depths and the
elements produced very strong primary haloes in all cross-sections. By contrast, Ag, As
and Cd revealed only weak haloes.
Secondly, this study used a combination of spatial analysis and concentration-area
method to separate the threshold concentrations from the geochemical background levels
for 10 elements. The outcome can be used for tracking anomalous area by surface
geochemical sampling and mapping of the element concentrations.
Finally, the geometrical properties (shape, orientation and dimension) and the scale
of spatial continuity of the geochemical haloes were compared quantitatively to delineate
distinct current zonation sequence in the Western Mineralisation. The results also suggest
Bi and Sb as potential pathfinder elements for the Western Mineralisation. |
ADE | CHAPTER 8
Spatial Models for Geochemistry, Geology and Geophysics of
the Western Mineralisation
8.1 Introduction
Selection of appropriate exploration techniques for distinguishing a hidden and
blind anomaly (true anomaly) from a false anomaly has always been difficult. In these
cases, exploration techniques can be improved by quantitative geological, geochemical and
geophysical studies of similar known types of ore deposits. In this chapter, in order to
improve exploration methods and geological information associated with the Western
Mineralisation, the following methods are conducted:
1. The spatial geochemical models of Chapter 7 are developed for:
1.1 Assessment of zonation sequence of 10 elements in axial sections.
1.2 Construction of an empirical formula of "Zonality Coefficient" (ZC) for the
Western Mineralisation type deposit in order to improve exploration techniques
for a hidden anomaly, and
1.3 Classification of anomalous concentrations for elements.
2. Different sections of sulphide minerals are studied for appraisal of their structural
variation, mineral zonation and alteration. The sections are also used to find their
similarities with the geochemical haloes of the orebody,
3. Different sections of magnetic susceptibility are studied to evaluate their spatial
relationships with the abundance of magnetic and non-magnetic pyrrhotite, other
sulphide minerals and geochemical haloes,
4. The average chemical composition of garnet samples collected from the Broken
Hill orebodies are analysed by correspondence analysis to show the relationship of
the Broken Hill orebodies with the chemical composition of garnet types,
5. 3D biplots are constructed for garnet samples of the Western Mineralisation to
show the chemical variation of garnet types in a spatial geometrical model, |
ADE | Chapter 8-Spatial Models for Geochemistry 255
6. Different sections of silicate minerals are studied for appraisal of their structural
variation and mineral zonation. The sections are also used to find their similarities
with the sulphide minerals, magnetic susceptibility and geochemical haloes of the
orebody,
7. Different sections of rock types are studied for appraisal of their structural
variation. The sections are also used to find their spatial relationships with the
silicate and sulphide minerals, magnetic susceptibility and the geochemical haloes
of the orebody, and
8. Different sections of sulphide textures are studied to evaluate their probability of
occurrence within the orebody. The sections are also used to find their spatial
relationships with the silicate and sulphide minerals, magnetic susceptibility, the
rock types and the geochemical haloes of the orebody.
8.2 Appraisal of sequence of the element concentrations in the axial halo zoning
The information about axial halo zoning for each mineralisation can be used for the
evaluation of the exposed level of the erosion for similar types of mineralisation associated
with depth (Beus & Grigorian 1977). Beus and Grigorian (1977) introduced several
empirical methods developed by Russian scientists for quantifying the zonation of haloes
in axial (vertical) sections.
One of the methods mostly used in geological research is the evaluation of the
average variations of element concentration ratios (e.g. Pb/U or Mo/U) as a function of
depth in vertical sections (e.g. Large, Bull & McGoldrick 2000). In this method, the
distribution shape of geochemical halos with depth is not considered.
Another method consists of a calculation of "Linear Productivity" (LP) that has
been given more emphasis, by calculating the LP, the width of haloes at different depths is
considered when characterising the relationship of geochemical haloes with the depth of
mineralisation. The LP is calculated as the product of width (in metres) of a geochemical
halo at a certain elevation and average element content (as percentage) at the same
elevation of the halo (Beus & Grigorian 1977). |
ADE | 258 Chapter 8-Spatial Models for Geochemistry
In order to compare LPs of geochemical haloes at different depths, Beus and
Grigorian (1977) suggested using the ratio of individual LP of the geochemical halo
element at a given elevation to the sum of LPs of the geochemical haloes of all elements at
that elevation to determine "Zonality Index" (ZI). In the next sections, the procedure for
calculating the ZI of geochemical haloes is outlined. The ZI represents the relative
concentration of elements at each elevation.
8.2.1 Calculations of ZI for the 10 geochemical haloes on the E-W axial sections
at N=2109 m
Tables 8.1 to 8.4 show procedures of computing of "Normalisation Coefficient"
(Cn), ZI and evaluating the zonation sequence of primary geochemical haloes for 10
element concentrations on the E-W axial section at N=2109 m.
The Cn is determined to normalise the values of LP of different element
concentrations so that their maximum LP values have the same order of magnitude basis.
For example, Pb has maximum LP value of 556.92 and Zn, Fe and S also have similar
orders of magnitude of LP (Table 8.1). However, the maximum LP values of Cd, Ag and
Sb need to be multiplied by 1000 to find the same order of magnitude of LP. Table 8.2
shows the amount of Cn for maximum LP values of 10 element concentrations. In order to
normalise LP values of each element at different elevations, the Cn value of each element
given in Table 8.2, should be multiplied by the LP values of the elements in all elevations
(Table 8.3).
The ZIs in Table 8.4 were calculated by dividing the normalised LP value of each
element in Table 8.3 at a certain elevation to the sum of LPs of that elevation. For example,
the ZI of Pb at level of 10198 is calculated by ratio of 59.76 to 909.771 (Table 8.3). Table
8.4 shows that the highest ZI of As occurs at elevation 10178 or at the upper level of the
geochemical haloes, whereas the maximum ZIs of Cu, Ag and Sb are at the lowest level of
the haloes. In these cases where in the zonation of haloes the order of elements with the
same position in a zonation sequence is not clear (e.g. Zn and S are both at the same
elevation of 10063 m in Table 8.4), the variability gradient1 [Equation (8.1)] of the
elements should be calculated to determine their real position in the sequence
(Beus & Grigorian 1977).
1Variability Index |
ADE | Chapter 8-Spatial Models for Geochemistry 261
Beus and Grigorian (1977, p. 96) suggested the empirical Equation (8.1) to
determine the order of the elements with the same sequence subject to their maximum ZI
being found at the upper or lower level (elevation) of the zonation of haloes:
n ZI
G max (8.1)
ZI
i i
Where G is the variability gradient, ZI is the maximum value of the zonality
max
index of a given element, ZI is the value of the zonality index in the ith elevation and n is
i
the number of elevations (exclusive of the elevations of maximum concentration of ZI ).
max
In Table 8.5, the G of Cu, Ag and Sb were calculated by Equation (8.1). Because the
elements are at the lowest elevation in Table 8.4, their sequence of elements was arranged
in the order of increasing G and this would be the other way round if the elements occurred
in the highest elevations (Beus & Grigorian 1977). The result of G G G
Sb Ag Cu
suggests the order of Sb-Ag-Cu from upward to downward direction of the zonation
sequence along the E-W axial section at N=2109 m.
Table 8.5: Calculations of G for Cu, Ag and Sb.
0.062899 0.062899 0.062899 0.062899 0.062899 0.062899 0.062899 0.062899
G
Cu 0.026220 0.028205 0.005811 0.015645 0.036080 0.019422 0.022497 0.026710
0.077699 0.077699 0.077699 0.077699 0.077699 0.077699 0.077699 0.077699
G
Ag 0.029546 0.066595 0.046487 0.049494 0.023110 0.047955 0.024996 0.052761
0.299696 0.299696 0.299696 0.299696 0.299696 0.299696 0.299696 0.299696
G
Sb 0.121348 0.238960 0.176377 0.287536 0.178279 0.258955 0.126231 0.247974
G 29.60
Cu
G 16.60 G G G
Ag Sb Ag Cu
G 12.88
Sb
According to Beus and Grigorian (1977), if the maximum ZI of several elements
occurs at a certain level in the middle horizon, then the difference between variability
gradients (ΔG = G - G ) between downward and upward directions are calculated where:
2 1
1. G is calculated for the downward direction from the level of the maximum ZI, and
2
2. G is calculated for the upward direction from the level of the maximum ZI.
1 |
ADE | 262 Chapter 8-Spatial Models for Geochemistry
The order of elements in zonation of haloes depends on the magnitude of ΔG. Each
element that contains a greater positiveΔG, will be placed further upward (to the left) in
the zonation of halo sequence and each element that contains lesser positive ΔG
will be located further downward (to the right) in the zonation of halo sequence. This will
be the other way round if the ΔG is a negative value. In this case, each element that
contains a greater negative ΔG will be placed further downward (to the right) in the
zonation. Table 8.6 shows the calculations of ΔG for pairs of S, Zn and pairs of Fe and Bi
in order to evaluate the sequence of primary zonation of the haloes in this axial section.
Table 8.6: Calculations of ΔG for pairs S, Zn and pairs Fe and Bi.
0.146467 0.146467 0.146467 0.146467 0.146467
G
S1 0.085253 0.131009 0.102086 0.138952 0.103526
0.146467 0.146467 0.146467
G
S2 0.064206 0.065422 0.098823
G G 6.00209716.7396202-0.737523
S2 S1
0.147296 0.147296 0.147296 0.147296 0.147296
G
Zn1 0.048638 0.114608 0.020285 0.101652 0.051461
0.147296 0.147296 0.147296
G
Zn2 0.029948 0.057672 0.068225
G G 9.631351-15.886293-6.254942
Zn2 Zn1
ΔG ΔG
S Zn
0.314642 0.314642 0.314642 0.314642 0.314642 0.314642 0.314642
G
Fe1 0.261779 0.292622 0.109799 0.115096 0.188318 0.115229 0.145302
0.314642
G
Fe2 0.097798
G G 3.21726714.44333011.226062
Fe2 Fe1
0.151885 0.151885 0.151885 0.151885 0.151885 0.151885 0.151885
G
Bi1 0.113878 0.034199 0.037338 0.044216 0.093512 0.084363 0.111760
0.151885
G
Bi2 0.054921
G G 2.7655018.06156015.296051
Bi2 Bi1
ΔG ΔG
Fe Bi |
ADE | 264 Chapter 8-Spatial Models for Geochemistry
*
Table 8.10: The normalised LP values for the 10 element concentrations at the selected
elevations.
Elevations
Elements Cn 9948 9928 9888 9868 9848
Pb 1 100.22 314.03 14.61 313.78 100.44
Zn 1 93.534 228.27 3.1122 368.01 18.216
Cd 100 34.02 110.16 N/A N/A N/A
Ag 10000 336 273 228 966 243.6
Sb 100 172.5 172.5 162 151.5 163.62
S 1 88.45 323.81 20.33 390.38 76.507
Fe 1 206.79 302.22 282.85 375.50 436.09
Cu 10 25.2 18 8.1 113.4 30.6
As 1000 588 588 105 319.2 730.8
Bi 100 121.5 132 132 153.12 121.5
Sum of LP values 1766.223 2461.996 956 3150.899 1921.368
*
Normalised LP value of each element = LP value of the element (Table 8.8) × Cn value of the element
(Table 8.9) for a certain elevation
The values of ZI in Table 8.11 were calculated by dividing the normalised LP value
of each element in Table 8.10 at a certain elevation to the sum of LP values of that
100.22
elevation. For example, the ZI of Pb at level of 9948 was calculated as (Table
1766.22
8.10).
Table 8.11: The values of ZI for 10 element concentrations and maximum ZI values
highlighted in red text.
Elements 9948 9928 9888 9868 9848
Pb 0.0567448 0.1275501 0.0152886 0.0995858 0.0522729
Zn 0.0529570 0.0927174 0.0032554 0.1167952 0.0094807
Cd 0.0192614 0.0447441 No data No data No data
Ag 0.1902364 0.1108856 0.2384931 0.3065791 0.1267846
Sb 0.0976660 0.0700650 0.1694556 0.0480815 0.0851580
S 0.0500797 0.1315244 0.0212635 0.1238954 0.0398191
Fe 0.1170820 0.1227564 0.2958633 0.1191729 0.2269682
Cu 0.0142677 0.0073111 0.0084727 0.0359897 0.0159261
As 0.3329137 0.2388305 0.1098323 0.1013044 0.3803539
Bi 0.0687908 0.0536150 0.1380749 0.0485956 0.0632361
Zonation sequence S-Pb-Cd Sb-Bi-Fe Ag-Cu-Zn As |
ADE | 266 Chapter 8-Spatial Models for Geochemistry
Table 8.13: Calculations of ΔG for Ag, Cu and Zn.
0.3065791 0.3065791 0.3065791
G
Ag1 0.2384931 0.1108856 0.1902364
0.3065791
G
Ag2 0.1267846
G G 2.418109-5.661876 3.243767
Ag2 Ag1
0.0359897 0.0359897 0.0359897
G ΔG ΔG ΔG
Cu1 0.0084727 0.0073111 0.0142677 Ag Cu Zn
0.0359897
G
Cu2 0.0159261
G G 2.2597882 11.692730 9.4329422
Cu2 Cu1
0.1167952 0.1167952 0.1167952
G
Zn1 0.0032554 0.0927174 0.0529570
0.1167952
G
Zn2 0.0094807
G G 12.319206 39.342193 27.0229867
Zn2 Zn1
Table 8.14: The zonation sequence of geochemical haloes on the E-W axial section at
N=1639 m.
Elevation 9948 9928 9888 9868 9848
Zonation sequence S-Pb-Cd Sb-Bi-Fe Ag-Cu-Zn As
8.2.3 Comparison of the ZIs between the E-W axial sections at N=2109 m at
N= 1639 m
Figure 8.3 was constructed by comparing the two ZIs in the above two axial
sections in order to evaluate the zonation sequence of the element concentrations from
elevation 10198 m to 9848 m (350 m depth) in the Western Mineralisation. The results are
outlined in Table 8.15. |
ADE | 268 Chapter 8-Spatial Models for Geochemistry
Figure 8.4: Cluster analysis for ZI values of 10 elements.
From Figure 8.4, some high percentage of similarities in the ZIs of 10 elements can
be outlined below:
1. Ag and As with 94 % similarity level,
2. Zn and S with 88.60 % similarity level,
3. Cd and Cu with 76.15 % similarity level, and
4. Fe and Bi with 75.29 % similarity level.
Figure 8.4 suggests that the ZI of Cd is more similar to the ZI of Cu and Sb that
tend to concentrate in the lower part of mineralisation. This may be because of decreasing
of the ZIs of Pb, Zn and S after level of 10063 m (Figure 8.3), while the ZI of Cd is
increasing after this level the same as Cu and Sb.
8.3 Exploration significance of the axial (vertical) zoning sequence (ZC)
The product of the indicator element concentrations in the lower part of the
Western Mineralisation over those of the upper part of the mineralisation can be used as
vertical ZC (Huang & Zhang 1989; Ziaii, Pouyan & Ziaei 2009) for the Western
Mineralisation or similar orebodies. The orientation of the ZC may be affected by |
ADE | Chapter 8-Spatial Models for Geochemistry 269
other factors, such as primary depositional zoning, secondary remobilisation or
reorientation due to folding and faulting.
The empirical product-ratios of ZC can be applied to the following issues:
1. Estimating the degree of denudation level of an anomaly,
2. Evaluating the exposure level of geochemical anomalies,
3. Recognising the upper anomalous halo from the lower anomalous halo, and
4. Predicting the presence of deeply buried Pb and Zn blind or hidden mineralisation
in potential areas where the concentrations of elements are more than their local
background levels.
According to Table 8.15, Cd, Pb, S, and Zn tend to concentrate in the upper part of
the mineralised zone, whereas As, Ag, Cu, Sb and Bi have a tendency to concentrate in the
lower part of the mineralisation. Based on the sequence of zonation in Table 8.15,
Equation (8.2) is suggested for the Western Mineralisation as ZC.
AgCuBiAs
ZC (8.2)
1 SPbZnCd
In Equation (8.2), Cd can be excluded from this calculation because three levels of
9888 m, 9868 m and 9848 m had no concentration values for Cd. Therefore, Equation (8.2)
is modified to take into account this situation for the calculation of mean element
concentrations at different levels [Equation (8.3)].
AgCuBiAs
ZC (8.3)
2 SPbZn
If it is considered that Bi is not a suitable element for the Equation (8.2) because of
its position in the middle of elevation in Table 8.15, Bi can be removed from Equation
(8.3) which can be modified to:
AgCuAs
ZC (8.4)
3 SPbZn |
ADE | 270 Chapter 8-Spatial Models for Geochemistry
The ZC values can also be interpreted as "Denudation Index" (DI) for a potential
deposit. Both Equations (8.3) and (8.4) can be considered as empirical product-ratios of DI
for Pb and Zn sulphide ore deposit similar to the Western Mineralisation. This is of value
for exploration because the Western Mineralisation does not crop out. The value of a DI
can be interpreted according to the following situations for a potential deposit:
1. If the DI < 1, it indicates the erosion level has not affected the upper part of the
anomaly. In this case, we have access to the upper part of anomaly and the
concentration of Pb, Zn and S will be higher in this part relative to concentrations
of As, Bi, Cu and Ag, therefore the value of DI will be less than one,
2. If the DI = 1, it shows that the erosion level has removed the upper part of the
anomaly or mineralisation and the rest of mineralisation may appear in depth, and
3. If the DI > 1, it means the erosion level has removed the upper and middle portion
of the anomaly. In this case, the lower part of anomaly is exposed and the
concentration of As, Bi, Cu and Ag will be greater than concentrations of Pb,
Zn and S. Therefore the DI will be more than one and any expectation of
mineralisation for that given area is very low if it was at the same erosional level.
Table 8.16 shows the variation of ZC and ZC at different elevations of the
2 3
Western Mineralisation. It is obvious that the values of ZC (or DI) are very much lower
than one, because the Western Mineralisation is an Zn and Pb sulphide orebody which
contains very much higher concentrations of S, Pb and Zn in comparison with the
concentrations of Ag, Cu, As and Bi.
According to Table 8.16, ZC (DI ) values vary between 0.01×10-10 and
2 2
317×10-10 in all levels and ZC (DI ) values vary between 0.002×10 -7 and 64×10 -7 in all
3 3
levels. The product-ratio of ZC or DI for the Western Mineralisation can be used for
evaluating similar Zn and Pb mineralisation in other parts of the Broken Hill Domain.
. |
ADE | 272 Chapter 8-Spatial Models for Geochemistry
Table 8.16 shows maximum ZC and ZC values at an elevation of 9948 m (the
2 3
greatest depth) of the cross-section of N=2109 m and maximum ZC and ZC values at an
2 3
elevation of 9868 m of the cross-section of N=1639 m. It should be noted that the distinct
(axial) vertical zonation sequence for the Western Mineralisation may not completely
reflect the complexity of the vertical zonation structure of the entire Broken Hill deposit.
In cross-sectional analysis, identification of the elements that tend to concentrate
below and above the mineralisation zone allow us to calculate an "Additive Index" (AI)
[Equation (8.5)] to improve exploration of other similar mineralisations (Beus & Grigorian
1977; Chen & Zhao 1998). Equation (8.5) is calculated by the ratio of the additive
concentration of elements that are higher in the zonation of halo (e.g. S, Pb, Zn and Cd as
numerator) to the additive concentration of elements that are lower in the zonation of halo
(e.g. As, Bi, Cu and Ag as denominator), i.e.
SPbZn
AI (8.5)
AgCuBiAs
The AI of the Western Mineralisation for elevations of Table 8.16 varies between
5.24 and 348.74. The AI also can be used for detection of a blind or deeply buried
mineralisation.
The Broken Hill Domain contains the world’s largest Zn-Pb-Ag deposit, the area
has undergone at east 12 km of denudation between the Palaeoproterozoic and the
Neoproterozoic and Phanerozoic denudation (Plimer 1984). There are hundreds of small
Broken Hill type deposits in the Domain in the Cues Formation, Parnell Formation,
Freyers Metasediments and Hores Gneiss. Compared with other major metallogenic
provinces (e.g. Mount Isa), it is inconceivable that the Broken Hill deposit is an orphan
and that other super-large Broken Hill-type deposits were also in the Willyama
Supergroup. Because the Western Mineralisation does not crop out, it is ideal for the study
of the ZC or DI and AI. If other large Broken Hill-type deposits are not exposed, these
indices will be useful for anomalous evaluation of an area where there may only be minor
sulphides or even quartz-gahnite and quartz-garnet lode horizon rocks. The closest
analogue to Broken Hill is the Cannington deposit (Queensland) in the Eastern Fold Belt.
An unmined exposed Broken Hill-type Zn-Pb-Ag deposit is known (i.e. Pegmont).
Cannington has the following characteristics (Davidson et al. 1989; Giles & Nutman 2003;
Laing 1998; Oliver et al. 2008): |
ADE | Chapter 8-Spatial Models for Geochemistry 273
1. Cannington is the same age as Broken Hill and occurs in the same lithological suite
of rocks including quartz-gahnite and quartz-garnet rocks suggesting the same ore-
forming processes,
2. It is far richer in silver and lead but lower in tonnage than Broken Hill,
3. It has abundant magnetite, a mineral absent from the Broken Hill deposit, and
4. It was discovered beneath Mesozoic cover rocks using gravity and magnetic
surveys.
Since discovery, a large geochemical dispersion halo has been detected at the
Mesozoic-Palaeoproterozoic unconformity and in Mesozoic sediments. If there is a
Broken Hill or Cannington-type deposit in the Broken Hill Domain at a depth of 400 m, it
would not be seen by the current geochemical, geological and geophysical methods used in
mineral exploration at Broken Hill. Such a deposit may be seen by using the DI or ZC and
AI developed in this study combined with deep electrical geophysics (e.g. Zeus method2).
If another Broken Hill orebody had been removed by weathering and erosion, the tail of a
geochemical anomaly may be present at the current level of erosion and may be expressed
in the DI or ZC and AI.
8.4 Classification of anomalous concentrations
For anomalous separation, eight different cross-sections are mapped for each
element of interest and the concentrations of each group of the cross-sections are shown by
colour indices (Figures 8.5 to 8.12). It should be noted that the major differences between
Figures 8.5 to 8.12 with the corresponding Figures 7.4 to 7.7 are their colour indices (their
range of element concentrations) and the scale of the diagrams, but the geometrical
characteristics of the corresponding geochemical haloes within the two groups of cross-
sections are the same.
2A proprietary induced polarization (IP) and resistivity technology. |
ADE | Chapter 8-Spatial Models for Geochemistry 279
The colour indices in Figures 8.5 or 8.6 classify anomalous concentrations for 10
elements. The anomalous range starts with a local threshold concentration to a relative high
concentration value within the plots. In order to enhance the contrast of colour resolution
within images, the relative maximum concentrations of each element were considered
instead of their respective absolute maximum concentrations. A deviation from lower
grade towards higher grade corresponds to concentration (enrichment) and from higher
grade toward lower grade means dispersion (depletion) for each element concentration.
The spatial classification of anomalous concentration of 10 elements provides the
opportunity to identify enrichment and depletion patterns for all elements simultaneously
for each portion of the Western Mineralisation and for detecting their underlying structural
relationships. In general, among Figures 8.5 to 8.12, there are some similarities between
concentration variations and anomalous areas among the following geochemical haloes:
1. As and Ag,
2. Sb and Bi, and
3. S, Fe, Cd, Cu and Zn.
In the Western Mineralisation, the extent and intensity of geochemical alteration
can be explained by diverse element concentrations in a different locus of anomaly in
Figures 8.5 to 8.12. The variations of anomalous concentrations within the plots express
the amount of mobility of the elements from depleted zones to an enriched area within the
Western Mineralisation. The different mobility of elements of the Western Mineralisation
may be attributed to various internal factors, including inherent physical and chemical
properties of the elements and numerous external factors such as:
1. Lithostratigraphy,
2. Chemical behaviour of the elements,
3. Variable chemical environments, and
4. Structural geology.
Because the Western Mineralisation will be concurrently mined from a number of
stopes, ore blending will minimise the high Bi and Sb in lead concentrate. Both these
elements are penalty elements. However, this study provides a prediction of concentration
and depletion of the ore elements (Pb and Zn), penalty elements (Bi, Sb and As) and bonus |
ADE | 280 Chapter 8-Spatial Models for Geochemistry
elements (Ag, Cd and Au). The zonation from Zn-rich to Pb-rich at depth (Figures 8.5 to
8.12) is an important conclusion. The mill design at Broken Hill is for “average” Western
Mineralisation ore derived from one down dip PQ hole. Although the behaviour of
Broken Hill ores in crushing, grinding and froth flotation mills has been documented after
some 200 Mt tonnes of sulphide ore treatment, this study shows that the early mill feed
Zn
will have a higher ratio than mill feed later in the mining operation. This may affect
Pb
recoveries as froth flotation circuits are specifically designed to treat ore of constant
mineralogy and chemistry. Figures 8.5 to 8.12 also show that the concentrations of Cu and
Fe have been elevated with depth and halo of Sb has higher concentration relative to Bi.
The variation of geochemical anomalies in different cross-sections (Figures. 8.5 to
8.12) may be used to distinguish different underlying geological structures (e.g. faults)
transgressing the Western Mineralisation. In geochemical plots, a fault may appear a linear
pattern or chained structure, or intrusive rocks may produce an arcuate shape (Cheng,
1999), or changes in rock type and geophysical properties may result in changes of the
overall shape of the mineralised zone. Table 8.17 outlines a range of concentrations for
anomaly; threshold and background for 10 elements in the Western Mineralisation resulted
from the colour indices of Figure 8.5 and Figure 7.4.
Table 8.17: The range of concentrations for anomaly, threshold and background for 10
elements of the Western Mineralisation.
Elements Background range Threshold range Anomalous range
Pb % , Zn % X* < 0.25 0.25 ≤ X ≤ 0.50 0.50 < X
Fe % X < 2 2 ≤ X ≤ 2.5 2.5 < X
S % X < 0.75 0.75 ≤ X ≤ 1 1 < X
Cu % X < 0.01 0.01 ≤ X ≤ 0.02 0.02 < X
Sb (ppm) X < 5 5 ≤ X ≤ 10 10 < X
Ag (ppm) X < 2.5 2.5 ≤ X ≤ 5 5 < X
Cd (ppm) X < 10 10 ≤ X ≤ 15 15 < X
Bi (ppm) X < 3 3 ≤ X ≤ 5 5 < X
As (ppm) X < 7 9 ≤ X ≤ 10 10 < X
*Concentration |
ADE | Chapter 8-Spatial Models for Geochemistry 281
8.5 Spatial variations of the sulphide minerals, magnetic susceptibility and specific
gravity in the Western Mineralisation
Figures 8.13 to 8.20 show the variation of sulphide mineral (vol. %), magnetic
susceptibility (SI) and specific gravity for the different sections of the Western
Mineralisation. Those Figures show some similarities in distribution patterns and spatial
continuity of sphalerite, AMS, MMS, chalcopyrite and pyrrhotite. Visual zonation of
pyrrhotite shows that the Western Mineralisation has a weak halo of pyrrhotite (Plimer
2006b) and this is consistent with the zonation of pyrrhotite in Figures 8.13 to 8.20 that is
narrower than the zonation of sphalerite.
The distribution pattern of galena is wider than that of sphalerite in all sections of
Figures 8.13 to 8.20 and this is comparable with distribution halo patterns of Pb and Zn.
Galena was distinctly enriched in the southern parts of the Western Mineralisation between
N=1550 m and N=1900 m whereas sphalerite is pronounced in the northern section of the
Western Mineralisation between N=1900 m and N=2350 m. Figures 8.13 and 8.14 further
support the dispersal pattern of elevated galena (vol. %) at the N=1639 m in comparison
with the N=2109 m and an antithetic pattern for sphalerite (vol. %). The variation of
specific gravity in the different cross-sections is small because there is not enough data for
reliable estimation. Since the sulphide-bearing rocks also contain dense minerals such as
garnet and gahnite, it is unlikely that changes in Western Mineralisation specific gravity
are correlated with grade.
The tonnage and grade of Broken Hill ores along the Main Line of Lode is in the
hinges of F3 folds and sulphide masses along axial plane cleavages or droppers are
common. Remobilisation and redistribution of sulphides during the Olarian and
Delamerian Orogenies at Broken Hill is well documented (Plimer 1984). The scale of this
local redistribution is not known. However, because the Western Mineralisation is on a
downward-facing F1 limb and has been folded in the F2 Broken Hill Antiform, Figures
8.13 to 8.20 show that the scale of remobilisation is less than the drill spacing for the
Western Mineralisation otherwise zoning would be spatially-related to structure.
Tables 8.18 and 8.19 outline the variation in abundance and dispersion of the
sulphide minerals and magnetic susceptibility from surface to the depth of the Western
Mineralisation based on observations of Figures 8.17 to 8.20. |
ADE | Chapter 8-Spatial Models for Geochemistry 289
Table 8.18: Variation model for sulphide minerals.
Sulphide minerals Variation
Sphalerite Enrichment at the middle depth section
Chalcopyrite Depletion with depth and decrease of distribution
Pyrite Enrichment with depth
Table 8.19: Variation model for magnetic susceptibility.
Magnetic
Variation Comments
susceptibility
Increase of conversion of pyrrhotite
I n c r e a se of the magnetic
with depth (i.e. changing from
AMS & MMS susceptibility at the
non-magnetic pyrrhotite to the
middle section of depth
magnetic pyrrhotite )
8.6 Correspondence analysis for garnet samples of the Broken Hill orebodies
In this section, before the study of spatial variation of silicate minerals in the
Western Mineralisation, the previous data sets of garnet samples were analysed statistically
and the results were interpreted. Garnet is conducted because it is one of the most common
accessory silicate minerals in the Broken Hill deposit. It has a grain size of 0.05-3 cm and
can be lilac, red, pink, orange and brown. General formula of garnet is A B (SiO ) and at
3 2 4 3
Broken Hill, it is a solid solution of the six common garnet end-members [almandine (Fe,
Al), grossular (Ca, Al), pyrope (Mg, Al), spessartine (Mn, Al), andradite (Ca, Fe) and
uvarovite (Ca, Cr)].
The Broken Hill garnet samples were collected by Kitchen (2001), Sproal (2001),
Tully (2002), Groombridge (2003), Patchett (2003) and Munyantwali (2006) and they were
analysed by EMPA. All samples were analysed on the same machine. Average
compositions of garnet types within the Broken Hill orebodies are presented in Table 8.20.
In this section, the Broken Hill orebodies are categorised based on their relation to the
average percent of garnet types using simple correspondence analysis. Table 8.20 shows
that the average percentage of almandine in the Western Mineralisation is higher than that
of the other orebodies. |
ADE | Chapter 8-Spatial Models for Geochemistry 293
8.6.1: Correspondence map for garnet samples of the Broken Hill orebodies
Figure 8.21 shows 95.11 percent3 of the total variation of garnet types within the
Broken Hill orebodies. Figure 8.21 shows that chemical composition of garnet types for A,
B and C Lodes, 1 Lens and the Western Mineralisation are more associated with almandine
and pyrope in comparison with other orebodies. Chemical composition of garnet types in 3
Lens is more related to spessartine and 2 Lens shows a greater relationship with chemical
composition of uvarovite and andradite relative to other garnet types.
Figure 8.21: Correspondence map in relation to dimensions 1 and 2 for garnet
(*the Western Mineralisation)
8.7: The 3D biplot of garnet types and samples of the Western Mineralisation
In this section, 543 garnet samples of Kitchen (2001), Patchett (2003) and
Munyantwali (2006) are visualized using 3D biplot model of the CoDaPack3D. The
information of introducing data into CoDaPack3D has been explained in Section 5.8.1 and
the data set of this section has been provided in supplementary file to this thesis. The 3D
biplot models (Figures 8.22 and 8.23) enable us to distinguish distribution patterns and the
major chemical composition of garnet types in the Western Mineralisation. Figures 8.22
and 8.23 show the distribution of 543 garnet samples (points) among three groups
including grossularite, andradite and the other four garnet types (i.e. spessartine,
almandine, uvarovite and pyrope).
3 Calculated by PPI (dimension 1) + PPI (dimension 2) from Table 8.23 |
ADE | Chapter 8-Spatial Models for Geochemistry 297
8.8: Spatial variations of the silicate minerals in the Western Mineralisation
Figures 8.24 to 8.31 display the variation of silicate mineral in different cross-
sections. In those Figures, the following minerals show great similarities in their
distribution patterns:
1. Gahnite and pink garnet, and
2. Blue quartz, green feldspar and hedenbergite.
In Figures 8.24 to 8.31, the following minerals show some similarities in their
distribution patterns:
1. Orange garnet and rhodonite, and
2. Red garnet and white quartz.
Some silicate minerals (Figures 8.24 to 8.31) have distribution patterns similar to
sulphide minerals, AMS, MMS and specific gravity (Figures 8.13 to 8.20). They are:
1. Green feldspar and arsenopyrite,
2. Rhodonite and pyrite,
3. Blue quartz, hedenbergite and galena,
4. Pink garnet, gahnite, sphalerite, pyrrhotite, AMS and MMS, and
5. White quartz and specific gravity.
Variations in abundance and dispersion of the silicate minerals from surface to
depth based on observations of Figures 8.28 to 8.31 are given in Table 8.25.
Table 8.25: Variation models for silicate minerals.
Silicate minerals Variations Comments
Between N=1550 m
Orange garnet Enrichment with depth
and N=1950 m
Depletion with depth and decrease of
Gahnite
distribution size with depth
Between N=1850 m
Red garnet Enrichment at the middle portion of depth
and N=2100 m
Between N=2150 m
Hedenbergite Enrichment with depth
and N=2450 m |
ADE | Chapter 8-Spatial Models for Geochemistry 311
1. Stringer, massive and laminated textures, and
2. Vein, network and brecciated textures.
There are some similarities in the structural distribution of the following sulphide
textures, silicate minerals (Figures 8.24 to 8.31), sulphide minerals and magnetic
susceptibility (Figures 8.13 to 8.20):
1. Massive, stringer, laminated, sphalerite, pyrite, pyrrhotite, chalcopyrite, AMS and
MMS,
2. Brecciated, network, vein, galena and arsenopyrite,
3. Massive, stringer, laminated, rhodonite, gahnite and pink garnet, and
4. Disseminated and white quartz.
There are some similarities in the structural distribution of the following textures
and rock types (Figures 8.32 to 8.36):
1. Massive, stringer, laminated, pegmatite and quartz garnetite,
2. Vein, network and brecciaed, metapsammite and metapsammopelite, and
3. Disseminated and metapelite.
Plimer (1984) stated that high-grade ores at Broken Hill have invariably brecciated
texture. In the Western Mineralisation, there is a visual correlation between brecciated ore
and high Pb and this has been used to design stope shapes. In this study, the distribution
pattern of vein, network and brecciated textures (Figures 8.37 to 8.41) show some
similarities with distribution shape of Pb (Figures 8.5 to 8.12). This indicates that the
textures control the grade of Pb in the Western Mineralisation. The distribution pattern of
stringer, massive and laminated textures (Figures 8.37 to 8.41) show some similarities with
distribution shape of Zn (Figures 8.5 to 8.12). This indicates that the textures control the
grade of Zn in the Western Mineralisation. |
ADE | 316 Chapter 8-Spatial Models for Geochemistry
identification of the geochemical halo, spatial mineralogical and lithological variation for
any of the Broken Hill orebodies.
The first aim of this chapter was to calculate the LP, ZI and ΔG of 10 elements in
axial sections of the Western Mineralisation in order to evaluate the zonation sequence of
the element concentrations as a function of depth. The final result shows that Cd, Pb, S and
Zn tend to concentrate in the upper part of the mineralised zone, whereas As, Ag, Cu and
Bi have a tendency to concentrate in the lower part of the mineralisation (Table 8.15).
AgCuBiAs
Consequently, the empirical product-ratio of was suggested as ZC or DI
SPbZn
for lead and zinc sulphide ore deposit similar to the Western Mineralisation. The value of
ZC for the Western Mineralisation varies between 0.01×10-10 and 317×10-10 that can be
compared with similar Zn and Pb ore mineralisation in other area.
The second aim of this chapter was to use structural analysis to provide a range of
colour indices between threshold level and relative local maximum anomaly for the 10
element concentrations. The result of anomalous separation shows several similarities
between anomalous areas of some group of elements including:
1. As and Ag,
2. Sb and Bi, and
3. S, Fe, Cd, Cu and Zn.
The correspondence map of average percent of garnet types within the Broken Hill
orebodies shows a close relationship of the Western Mineralisation to pyrope and
almandine end members in garnet. The 3D biplot shows the garnet samples of the Western
Mineralisation were affected highly by chemical variation of Ca and Al (grossular), Ca and
Fe (andradite) as well as Ca and Cr (uvarovite).
In final, the distribution patterns of elements, minerals, textures, rock types,
magnetic susceptibility and specific gravity were presented in different sections and they
were grouped based on their similar distribution patterns. |
ADE | CHAPTER 9
Conclusions
9.1 Introduction
This final chapter provides a summary of responses to unsolved problems and the
research questions of this dissertation that were presented in Chapter 1. It also covers the
significance of this research study and some suggestions for future similar research.
9.2 Responses to research questions of this dissertation
9.2.1 Maximizing information of core logging
The first purpose of this research was to maximize information of core logging in
the Western Mineralisation (unsolved problem "a" in Section 1.7). In order to solve this
problem, all available geological and geochemical data sets about these characteristics
from previous studies of the Broken Hill orebodies including the Western Mineralisation
were compiled and prepared for multivariate statistical analysis of this research. Moreover,
in this study, 30 important geological variables including sulphide and silicate minerals,
rock types and sulphide textures were investigated and quantified visually for 1,928
Western Mineralisation samples. The average and maximum magnetic susceptibility were
also measured and their specific gravity was recalculated for each metre sample.
Through this quantitative core logging and data processing, 63,6241 new points of
data information were generated and added to the previous assays of the Western
Mineralisation. However, because the assays of the Western Mineralisation had been
reported for different sample sizes, the theory of sample volume-variance relationship was
applied to all HQ and LTK60 assayed samples size in order to convert them to NQ sample
size. Also different lengths of measured assayed samples were recalculated for one metre
samples. This process of sample compositing performed for all assays of surface and
underground drill cores. In final, all quantitative geological, geochemical and geophysical
data sets were presented in a large data base (Excel format) named the "quantitative core
log data".
1 33 ×1928 |
ADE | 318 Chapter 9-Conclusions
9.2.2 Characteristics of the quantitative core log data of the Western Mineralisation
The second unsolved problem was about the conventional descriptive core logging
of the Western Mineralisation that is separate from mineral chemistry and geophysical data
as well as the processes undertaken at the Rasp Mine which did not allow integration of the
data for performing different statistical methods (unsolved problem "b" in Section 1.7).
The quantitative core logs data of this study incorporated 44 quantitative variables
including 24 geological variables (minerals and rocks), 7 textural variables, 3 geophysical
variables and 10 geochemical variables (element concentrations) for 1,928 samples of
surface drill cores and 8 geochemical variables for 1,231 samples of underground drill
cores.
The data set was defined for equal sample size that is a fundamental prerequisite for
performing statistical analyses. The spatial coordination of start, middle and end of each
metre core sample has been calculated regarding collar, survey and inclination of drill hole.
The quantitative core log data can be visualized directly in any 3D graphs and mining
software such as MicromineTM, VulcanTM and SurpacTM without the requirement of the
azimuth (bearing), dip and collar elevation of each drill core.
The quantitative core data was also used for providing high quality core log
diagrams with resolution of one metre. The first group of diagrams assists simultaneous
evaluation of the variation of 44 variables as a function of depth (Figures 2.8 to 2.13). The
second group shows simultaneous variation of Pb+Zn, galena+sphalerite; pyrrhotite,
magnetic susceptibility, specific gravity and sulphide textures as a function of depth
(Figures 4.13 to 4.17). The diagrams were used to detect internal-consistency among
variations of minerals, rock types, assays and sulphide textures in the Western
Mineralisation. The quantitative core log data allows the easy use of bar diagrams to
evaluate the trend variation of statistical parameters of each variable (e.g. element
concentration) within all surface and underground drill cores simultaneously. For example,
in bar diagrams of Chapter 3, the trend variation of mean and median values of galena,
sphalerite (Figure 3.11) and galena+sphalerite (Figure 3.12) in each sample were compared
by those of parameters in bar diagrams of Pb, Zn (Figure 3.3) and Pb+Zn (Figure 3.12)
respectively. |
ADE | Chapter 9- Conclusions 319
9.2.3 Application of classic statistics for the Western Mineralisation
The CBH Resources Ltd has been using geostatistics to determine if there is
enough ore at a sufficient grade to warrant economic extraction of the Western
Mineralisation, but these processes are unable to evaluate the scientific geological,
geochemical and geophysical characteristics by classic statistical analysis (unsolved
problem "c" in Section 1.7). The third purpose of this research study was to apply a variety
of standard univariate, bivariate and multivariate statistical methods to improve
quantitative understanding of mineral chemistry, mineral generations, geochemical,
geophysical and geological variations and their relationships.
Since the Broken Hill ore deposit has been well known already in terms of geology,
the geological, geochemical and geophysical features of the Western Mineralisation is
better understood at the conception of statistical models and numerical exploration
methodology. Moreover, the statistical results will be useful to determine potential areas
where ore awaits discovery.
In this dissertation, the univariate analyses of descriptive statistics were used to
understand which type of statistical parameters are important for each variable and
compare and contrast the descriptive statistical parameters of variables with each other
(Chapter 3).
Bivariate analyses including different parametric and non-parametric correlation
coefficients (Chapters 3, 4 and 5) were applied to measure the degree of relationships and
internal consistency among galena+sphalerite, pyrrhotite, magnetic susceptibility and
Pb+Zn or between rock types and Pb+Zn. These processes were not possible with
conventional core log information.
Statistical tests were used to evaluate significant differences of magnetic
susceptibility and pyrrhotite between two group samples containing galena+sphalerite and
other samples. Log-ratio transformation was used to open the closed percentage assays into
an unconstrained form of the real space. After the transformation, the initial skewed
distribution of elements changed to normal distribution. This process is useful for detection
of non-linear relationships using bivariate correlation coefficient. |
ADE | 320 Chapter 9-Conclusions
The linear multiple regressions were calculated for Pb and Zn concentrations
separately in order to identify their best predictor elements. The predictor elements can be
used as geochemical exploration guides for Pb and Zn ore sulphide types similar to the
Western Mineralisation (Section 5.5). Those predictor elements suggested by the linear
multiple regressions were calculated by geostatistical methods (Chapters 6 and 7) to
understand which of them are the most appropriate pathfinder elements of Pb and Zn in the
Western Mineralisation.
Cluster analysis was used to classify elements based on their percent of similarities
in concentration values or variogram parameters. The agglomerative algorithm and
correlation coefficient were used in this process and the results are useful for comparison
of the Western Mineralisation with other orebodies.
PCA was applied to 10 element concentrations to quantify the theories behind ore
sulphide genesis of the Western Mineralisation. The results of PCA showed strong
consistency with existing mineral paragenesis of the Western Mineralisation. PCA reduced
geochemical complexity of variation of ten elements into four multi-elemental
relationships (Section 5.7).
Simple correspondence analysis was applied to demonstrate in different maps the
relationships of different chemical composition of pyrrhotite, galena and sphalerite
samples with the Broken Hill orebodies. Simple correspondence analysis was also used to
visualize the relationships between the average percent of different garnet types and the
Broken Hill orebodies (Section 8.6).
The biplot of chemical composition of pyrrhotite, galena and sphalerite samples
were constructed to understand their mineral chemistry generations or their alterations in
the Western Mineralisation. The 3D biplot of average percent of garnet types was
constructed to understand major chemical compositions of garnet in the Western
Mineralisation (Section 8.7).
9.2.4 Application of geostatistics for the Western Mineralisation
There is a lack of spatial models for geological, geochemical and geophysical
features of the Western Mineralisation. This is because CBH Recourses Ltd uses |
ADE | Chapter 9- Conclusions 321
geostatistics to make spatial models for combination of Pb, Zn and Ag to estimate tonnage
and grade of the orebody (unsolved problem "d" in Section 1.7). This information alone
does not make an effective exploration tool for finding look-alike ore deposit styles within
similar prospective areas.
Although, the information of this part of the study may not result directly in a
decision to mine, it can be useful to find near-mine repetitions, extensions and continuity
of the existing orebody and the detection of primary geochemical haloes. Furthermore, it is
of use in regional exploration to evaluate a near miss or to use the AI, ZC and DI to
evaluate prospectively.
In order to address this problem, 136 directional variogram and down-hole
variogram models were calculated for 43 variables of the quantitative core logs. Their
variogram parameters were applied to construct 43 spatial block models using ordinary
kriging. The variogram models of the Western Mineralisation were validated using the
cross-validation technique. A few investigated minerals such as pyrite, arsenopyrite and
red garnet showed low variogram validation because the number of samples were very low
and their volume percentage changed between two or three numbers (e.g. between 1 % and
3 %).
The strike direction, plunge and plunge direction, dip and dip direction of the
Western Mineralisation was estimated by variogram analysis and geological observations.
From the variogram parameters, the optimal sampling grid was determined as an
exploration guide for tracking extension of geochemical haloes with surface sampling. The
variogram parameters of the Western Mineralisation were also compared with available
variogram parameters of the other lead and zinc sulphide ore deposits in order to classify
the ore styles (i.e. Sedex, MVT, vein and Irish deposit).
A block size of 20×20×10 m (with the volume of 4000 m3) was calculated as the
most suitable block size for kriging estimation of lead and zinc in the Western
Mineralisation. This block size also was considered for construction of the other 41
geological, geochemical and geophysical variables to enable comparison. For better
demonstration and interpretation of the 3D block models in this dissertation, each 3D block
model was intersected at different horizontal and vertical directions and during this
process, 424 cross-sections were generated from 43 spatial block models. |
ADE | 322 Chapter 9-Conclusions
The 80 cross-sections of 10 element concentrations were used for evaluating
dimension, orientation and anisotropies of geochemical halo (Chapter 7). For this purpose,
maximum lengths and widths of the haloes were measured in transverse, longitudinal and
axial sections in order to find the zonation sequence of the geochemical haloes through the
orebody. The results showed that Bi and Sb have greater dispersion of haloes relative to the
other 8 elements and they can be considered as geochemical pathfinders of the Western
Mineralisation.
The 10 geochemical haloes of the Western Mineralisation were plotted into two
groups of 80 cross-sections. One group of the cross-sections was distinguished by colour
indices between background and threshold levels and another group was distinguished by
colour indices between threshold and anomalous values. The colour index of individual
element within each group was adjusted by trial and error through the Geostatistics for
Windows software. In Chapter 8, the axial sections of geochemical haloes were used for
calculation of empirical formulas of the LP, ZI and ΔG (Beus & Grigorian 1977).
From this process, it was concluded that in axial sections, S, Pb, Zn and Cd tend to
concentrate in the upper part of the mineralised zone whereas Ag, Cu, Bi and As have a
tendency to concentrate in the lower part of the mineralisation. The result was used for
construction of an empirical formula of ZC2 or DI2 that has various exploration applications
(Section 8.3). The spatial distribution pattern of silicate and sulphide minerals, rock types,
sulphide textures and magnetic susceptibility were compared and contrasted with each
other to find their similarities and variations within the orebody.
9.3 Significance of this kind of research study
Numerical evaluation of geological, geochemical and geochemical characteristics
of the Western Mineralisation in the Broken Hill mine has great potential in the following
areas:
1. Extraction of more useful and quantitative geological and geophysical information
from core samples and integration of the information with assays,
AgCuBiAs
2
SPbZn |
ADE | Chapter 9- Conclusions 323
2. Evaluation of previous mineral chemistry data with multivariate statistical
methods,
3. Numerical mineralogy and petrology in this study were used to generate spatial
models similar to spatial geochemical models,
4. Information of geometrical distribution of the primary geochemical haloes can be
used in terms of tracking of secondary geochemical haloes, studies of regolith,
geobotany, hydrogeochemistry and biogeochemistry around the Broken Hill
orebodies,
5. Providing numerical information for abundant pyrrhotite, magnetic pyrrhotite and
magnetic signatures of the orebody,
6. Providing strong documentations based on the quantitative core log data,
classic and spatial statistical analysis, geological investigation, geochemical and
geophysical measurements which can be updated every time when new data is
available. This documentation can be used as numerical exploration guidance for
similar types of ore deposits or comparison of different characteristics of orebodies
for purposes of ore classification,
7. Providing sustainable information for an orebody because it uses the information of
core samples which have been preserved in core storage for a long time which can
be referred to in order to validate previous information or start of a new research,
8. The quantitative core log analysis is applicable and repeatable for the other Broken
Hill orebodies or other orebodies in other areas,
9. This kind of study does not add mining cost because it deals with previous
information and existing core samples of an orebody, and
10. This study also provides environmental information about the mechanism of
distribution and penetration of elements that may be released into groundwater and
soils in vicinity of the Broken Hill City because of its proximity to the Western
Mineralisation. A previous study by De Caritat et al. (2005) showed that
groundwater in the Broken Hill district contained lead content matching the Broken
Hill lead isotope signature. |
ADE | 324 Chapter 9-Conclusions
9.4 Suggestions for future research
This study was limited to the Western Mineralisation. Additional quantitative core
logging is needed for other orebodies of the Broken Hill Mine to identify specific
geochemical and mineralogical zonation patterns and spatial relationships of the orebodies
in a numerical framework. Future research in this area for the Broken Hill ore deposit can
develop the quality of numerical models in relation to geological and geochemical
information. Further research could investigate:
1. What the extent of association is among the Broken Hill orebodies in terms of
geochemical haloes, sulphide metal and magnetic mineral distribution, and
2. What the mineral/chemical zonation patterns are and which geochemical factors
control the Broken Hill orebodies.
This study was constrained to silicate mineral chemistry of a few samples collected
from limited drill cores within the Broken Hill orebodies. It is suggested that silicate
minerals such as garnet and gahnite are collected from a large range of drill holes within
the Broken Hill field and their chemical compositions are determined for construction of
spatial mineral chemistry models. Although gahnite and garnet are associated with Broken
Hill-types deposits, it is yet to be shown that the chemistry of these minerals can be used as
a vector for mineralisation. |
ADE | 1. INTRODUCTION
grasses are undergrowth. Sparse occurrence MinesfollowingtheGreatCobarMineintopro-
of Bimble Box (Eucalyptus popunea) with an ductionuntiltheearly20thcenturywere: Giril-
understory of woody shrubs is the dominating ambone 1881, Chesney 1887, Occidental 1889
faunaattheminesite(Lorrigan,2005). Overall, (later known as New Occidental), Cobar Gold
vegetationisrelativelysparse,withdenserveg- Mine 1890 (later known as New Cobar), Mt
etation found along creeks or pronounced inci- Drysdale 1893, Mt Pleasant 1895, Young Aus-
sions. tralian 1896, The Peak 1896, Mt Boppy 1898,
Queen Bee 1902, CSA 1905, Tinto 1906 and
1.2 Mining history in the Gladstone 1908 (Stegman and Stegman, 1996;
DepartmentofPrimaryIndustries,2007).
Cobar region∗
Advancing industrialisation, electrification
and the railway in July 1892 led to increased
Three important mining belts are distinguished
production and profitability. In 1889 the cop-
in the Cobar area: the Cobar Belt, the Can-
per price was £60 per ton, dropping down to
belego Belt and the Girilambone Belt. Out of
£39beforesteadilyincreasingto£75pertonin
those three mineral belts, the Cobar Belt is the
1899,withthecopperconsumptionbeingabove
mostimportantofthesebeltsandrepresentsthe
production by 1906. The copper price finally
largest accumulation of Phanerozoic base met-
peaked at over £100 per ton in 1907 (Stegman
alsinNewSouthWales. Allmajorminingcen-
andStegman,1996;DepartmentofPrimaryIn-
tres with important mineral deposits in the Co-
dustries,2007).
bar region, e.g. Cobar, Canbelego, Nymagee,
The fall of copper prices in 1908 lead to a
ShuttletonandMountHope,arehostedinEarly
halt of mining and depressed the flourishing
Devonian rocks of the Cobar Supergroup sedi-
exploration activities with particularly difficult
ments. Theleadingbasemetalandgoldproduc-
times during 1920s to 1950s. Equally, copper
ers are Endeavor, CSA, Great Cobar, Chesney,
price and demand fell by over 40% in 1919 af-
NewOccidentalandMtBoppyMines(Stegman
ter World War I, forcing Cobar mines to close.
andStegman,1996;DepartmentofPrimaryIn-
The only gleam of hope during this period was
dustries,2007).
astrongrenaissanceofgolddemandduringthe
ThefirstcommoditydiscoveredinCobarwas
economic depression in the 1930s. No ma-
copper in 1869 at the location of the Great
jor mines were in operation between 1952 and
Cobar Mine. This was followed by gold in
1965. The population of Cobar varied between
1871. Several other mineral occurrences were
a few hundred up to more than 10,000 during
exploredshortlyafterwhichhaveproducedsig-
the first quarter of the 19th century due to sig-
nificant quantities of copper, lead, zinc, sil-
nificantfluctuationsinproduction(Stegmanand
ver and gold, in addition to minor quantities
Stegman, 1996; Department of Primary Indus-
of by-products such as cadmium and antimony
tries,2007).
(Stegman and Stegman, 1996; Department of
In1962anewdeepshaftsinkingbeganatthe
PrimaryIndustries,2007).
CSA mine, full scale mine development com-
Theerectionoftheearliestsettlementsisonly
menced in 1964 and production re-commenced
poorlydocumentedbutprobablyoccurredinthe
in 1965. Since then, the CSA mine had been
early 1860s. It was initiated due to prospect-
seriouslyaffectedbyadropinmetalpricesand
ing and exploration activity, by extending set-
tlements along the Barwon and Darling Rivers. ∗.Summarisedafter:
Stegman, C.andStegman, T.(1996).Thehistoryof
Settlement was intensified at around 1869 ac-
miningintheCobarField.In:W.Cook,A.Ford,J.Mc-
companiedbywaterboreandwatertankinstal- DermottandP.Standish(eds.),TheCobarMineralField
lations. In the 1880s the metalliferous industry –A1996perspective,TheAustralasianInstituteofMin-
ingandMetallurgy,Melbourne,pp.3–39and
employed approximately 500 men and Cobar’s
Department of Primary Industries (2007). Cobar’s
population grew to 3,000 during this decade. mininghistory.PRIMEFACT,555,9
2 |
ADE | 1.2 Mining history in the Cobar region
themineclosedin1975andagainin1985. Af- shares on 23nd September 2010, CBH Re-
ter a mine closure in 1997/98 it was reopened sources Limited became a wholly owned sub-
byGlencorein1999andisnowoperatedbyCo- sidiaryofTohoZincCo.,Ltd,acompanylisted
barManagementPtyLtd–awhollyownedAus- ontheTokyoStockExchange. CBHResources
traliansubsidiaryofGlencoreInternationalAG was delisted from the ASX on 30th September
(Switzerland) (Stegman and Stegman, 1996; 2010(DepartmentofPrimaryIndustries,2007).
DepartmentofPrimaryIndustries,2007). Rather early in the mining history of Cobar,
ThePeakminingfield,comprisingThePeak, severalproductionproblemsanddifficultiesbe-
Perseverance, New Occidental, Chesney and came apparent. The sudden decrease of the ore
New Cobar Mines, are operated by Peak Gold gradefromtheinitial9to11%copperinthesu-
MinesPtyLtd,asubsidiaryofGoldcorpInc. pergene enriched zone down to 1 to 3% in the
The Elura Mine was initially discovered in unweathered primary mineralisation was prob-
1973 and operated by Electrolyte Zinc Co. of ably the most difficult. This problem was in-
Australasia Ltd before being taken over by tensifiedbyrecoveryissuesduetotheincreased
North Broken Hill Holdings Ltd in 1984/85 complexityinthesulphidemineralogy. Further-
whichinturnmergedwithRiotintoofAustralia more, the high consumption of timber for stop-
Ltd in 1988 forming Pasminco Ltd. Consoli- ing and furnaces led to major shortages in the
dated Broken Hill Ltd (now CBH Resources), timber supply. Severe dust storms, water sup-
anASXlistedcompany, boughttheEluramine plyissues,rockcreepandpartialminecollapses
from the Administrators of Pasminco Ltd and alsocausedseriousproblemsamongstthemin-
renamed it into the Endeavor Mine in 2003. ing community (Stegman and Stegman, 1996;
Upon acquisition of all CBH’s issued ordinary DepartmentofPrimaryIndustries,2007).
A
NOTE:
This figure/table/image has been removed
to comply with copyright regulations.
It is included in the print copy of the thesis
held by the University of Adelaide Library.
Figure1.2: EarlysettlementsinthetownshipofCobarinthelate1890s. MarshallSteet,nowadaysthemain
street,istotherightofthephoto. PhotographtakenfromStegmanandStegman(2002).
3 |
ADE | 1. INTRODUCTION
Summaryofownerships: • 1973 Electrolyte Zinc Co. of Australasia
Ltd,discoveryoftheEluradeposit,
• 1893 Great Cobar Syndicate organised by
Swanseasmelters(LongworthBros.),later • 1983 Electrolyte Zinc Co. of Australasia
GreatCobarLtd, Ltd,productioncommencedatElura,
• 1930 Occidental Ltd, later reorganised as • 1984/1985 North Broken Hill Holdings
NewOccidentalGoldMinesNL, Ltd, take-over of Electrolyte Zinc Co.
of Australasia Ltd and acquisition of the
• 1947 Enterprise Exploration Co. Pty Ltd, Eluramine,
asubsidiaryofZincCorporationLtd(later
Consolidated Zinc Pty Ltd), acquisition of • 1988 Pasminco Ltd, merger with North
CSA mine and extensive regional explo- Broken Hill Peko Ltd and Conzinc Riot-
ration (mapping of Cobar belt from CSA intoofAustraliaLtd,
toQueenBeein1: 4,800),
• 1992GoldenShamrockMines;acquisition
• 1955 Broken Hill South Ltd, secured an oftheCSAmine,
option over leases and other assets from
• 1999 Glencore International AG; acquisi-
NewOccidentalGoldMinesLtd,
tionoftheCSAmine,
• 1956 Broken Hill South Ltd created two
• 2003 Wheaton River Minerals Ltd (now
wholly owned subsidiaries, Cobar Mines
Goldcorp Inc); acquisition of the Peak
PtyLtdandCobarSouthPtyLtd,
MinesfromRioTinto,
• 1957 Broken Hill South purchased key
• 2003 Consolidated Broken Hill Resources
Cobar leases from New Occidental Gold
Ltd; acquisition of the Elura mine (re-
MinesLtd,
named in Endeavor mine) from Pasminco
• 1962CobarMinesPtyLtd;operatingCSA Ltd,
Mine,
• 2010 Toho Zinc Co. Ltd; Consolidated
• 1980 Conzinc Riotinto of Australia Ltd, Broken Hill Resources Ltd. is a wholly
purchasedCobarMinesPtyLtdandCobar ownedsubsidiary.
SouthPtyLtd,
4 |
ADE | 1.3 Discovery of the Elura deposit
1.3 Discovery of the Elura
deposit∗ A
NOTE:
Cobar’s mining field with its potential exten- This figure/table/image has been
sions of identified mineral resources repre- removed to comply with copyright
sented an excellent exploration target. It still regulations. It is included in the
does. In1971,ElectrolyticZincCoofAustrala- print copy of the thesis held by the
siasuccessfullyappliedfortwoExplorationLi- University of Adelaide Library.
cencesintheareaNWofCobar. Inpreviousex-
ploration work the geologists had assumed, the
Cobar Supergroup sediments, hosting the min-
Figure 1.3: Aerial photograph showing the foot-
eralisationoftheCSAmine, wouldcontinueto print of the Endeavor Mine (formerly known as
theNE,towardsMt. Drysdale. Anewhypothe- Elura Mine). Photograph supplied by CBH Re-
sourcesLtd.
sissuggestedthattheparticularlithologymight
aswellcontinueNandpossiblyNWundersoil
cover. Most of the known mineral occurrences netic surveys. A well defined gravity anomaly
in the Cobar area exhibit a magnetic anomaly. in addition to a weak conductive zone in the
Thatisduetotheoccurrenceofeitherpyrrhotite vicinity of the ground magnetic anomaly was
or magnetite. Subsequently, airborne magne- identified.
tometry was undertaken in order to test for po- InFebruary1974,thefirstdiamonddrillhole
tential extensions of the known mineralisation E1 intersected gossan and sulphide mineralisa-
attheCSAMine. tionatadownholedepthof102mand133.5m,
Several deep-seated and surface anomalies before being abandoned at 152.4m. After the
were identified. Despite the fact that Elura first drill hole, gossan was also found cropping
exhibited itself as a characteristic “bulls-eye” out at the surface. The second diamond drill
anomaly it was initially not considered to be holeE2,designedtotestthesourceofcombined
of great potential because it was not associated magnetic, gravity and induced polarisation, in-
withthehighlyprospectivehostlithologyofthe tersected85mofsulphidemineralisation. Ato-
CSA mineralisation. However, it was decided tal of 24 diamond drill holes and six wedged
later, to investigate all characteristic magnetic “daughter” holes were drilled during the main
anomaliesirrespectiveoftheirstratigraphicpo- exploration phase resulting in an indicated re-
sition. sourcepotentialofapproximately27Mtat8.4%
Groundmagneticfollow-upworkattheElura zinc, 5.6% lead and 139g/t silver. For the pur-
anomalycommencedinJanuary1973resulting pose of metallurgical bulk sampling, an explo-
in a single positive anomaly of approximately ration shaft was sunk to a depth of 165m be-
150nT and extending over 500m. The sub- tween 1976 and 1977. Mining by Electrolyte
sequent auger sampling revealed a strong Pb Zinc Co. of Australasia Ltd commenced at
anomaly with values up to 3,000ppm contin- ElurainMarch1983.
uing over a strike-length of 1,200m. A sam-
ple spacing of 5 to 10m on grid lines 50m
apart were used with samples taken at a depth ∗.Summarisedafter:
Schmidt, B. (1990). Elura zinc-lead-silver deposit, Co-
of 1.8m. The original strategy was to plan
bar. In: F. Hughes (ed.), Geology of the mineral de-
and drill the first two diamond holes based on posits of Australia and Papua New Guinea,vol.2,Aus-
themagneticsurveyandaugersamplingresults. tralasianInstituteofMiningandMetallurgy,Melbourne,
pp.1329–1336and
However,duetodelaysintheprocessingofex-
Davis, L. (1980). The Discovery of Elura and a brief
ploration tenements, the drilling program was summaryofsubsequentgeophysicaltestsatthedeposit.
In:D.Emerson(ed.),The Geophysics of the Elura Ore-
deferred in favour of gravity and electromag-
bodyCobar,NewSouthWales,Sydney,NSW2000:Aus-
tralianSocietyofExplorationGeophysicists,pp.5–9
5 |
ADE | 2. REGIONAL GEOLOGY
¨ New
N England
Orogen
Thomson
Orogen
S
Lachlan
Orogen Sydney
Delamarian
250 1,000
Kilometer Orogen
0 500
Figure 2.2: Regionalextentofthefourmainoro-
genicbeltsintheTasmanides(modifiedafterGray
andFoster(2004)).
Figure 2.1: The development of the Paleozoic
Tasman Fold Belt on the eastern margin of Gond- The Delamerian Supercycle com-
wanaafterVoset al.(2007). Theapproximateex- mencedataround750Maasaratherlong-lived
tentoftheLachlanFoldBeltisoutlinedinyellow.
rifting event when Rodinia began to disinte-
grate. This initial phase of the Delamerian
2005). The final cratonic stage was predomi- lasted for approximately 150 Ma, forming the
nantly characterised by sedimentation in a sag amagmatic Adelaide Rift Complex, followed
basin environment due to the plate boundary by the development of a major alkaline mag-
shifting towards the southwest pacific (Glen, matic rift system between 580 to 600 Ma.
2006). The change of tectonic plate movement at
An internal Permian-Triassic rift related around 520 Ma resulted in the establishment
forelandbasinsystem,consistingoftheBowen, of a convergent margin, with a subsequent
Gunnedah and Sydney sub-basins, separates change in magmatism and the formation of the
theolderfoldbeltsfromtheyoungest,theNew Kanmantoo Trough, a rift basin in back-arc
England Fold Belt (Scheibner, 1996). The position. The magmatic activity switched
development of this basin commenced in the from alkaline to mafic and ultramafic and is
east and progressed towards the west, and is reflected by the formation of boninitic fore-arc
dominatedbymolassicsediments. crust, which is preserved on the western part
of Tasmania and Victoria. The supercycle
TheTasmanidesevolvedtemporarilyinthree ended at approximately 510 to 505 Ma when
tectonic supercycles, as described by Glen arc and fore-arc rocks were accreted, followed
(2006). Each cycle is characterised by long- by extension, post-collision magmatism and
lived sedimentation and igneous activity, even- deformationattheendoftheCambrian.
tually ending in a rather short deformational
event, being the eponym for the cycle. The su- The Lachlan Supercycle started as an
percycles are the Delamerian (Neoproterozoic erosive event with the formation of molassic
to latest Cambrian), the Lachlan (Ordovician sediments on the Delamerian Orogen. How-
to Early Caboniferous) and the Hunter Bowen ever,thecyclewasmainlydominatedbyhighly
(Late Devonian to Late Triassic). Furthermore, convergent plate boundaries and the formation
the rather short lived Kanimblan cycle tempo- of the intraoceanic Macquarie Arc. During the
rally overlaps with the Hunter-Bowen super- Benambran Orogeny in the Late Ordovician to
cycle. These cycles described in Glen (2006) EarlySilurian,severalterranesconsistingoftur-
aresummarisedinthefollowing: bidites,shales,subductionrelatedblueschistfa-
ciesmetamorphicrocks,MORB-likemaficvol-
8 |
ADE | 2.1 The Tasmanides
canics and cherts were imbricated and accreted The Kanimblan Cycle, temporallyover-
togetherwiththeMacquarieArc,leadingtothe lapping with the Hunter-Bowen super-cycle
closureoftheoceanbasins. of the New England Orogen, mainly affected
The subsequent Tabberabberan cycle com- areas which are now located in NSW and in
menced as a major phase of increased tension Victoria. Thecycleischaracterisedbytheonset
on the eastern part of Gondwana in the Early ofnarrowriftsandtheformationofgranitesand
Silurianduetoarollbackofthesouthernpartof volcanics, of A-type in NSW, S- and I-type in
the Pacific Plate. This resulted in the develop- Victoria. This magmatic episode was followed
mentofanintraoceanicarc, ledtothedismem- byaperiodofsedimentdeposition,forminga3
beringofthepreviouslyaccretedMaquarieArc to 4 km thick cover of continental sedimentary
and initiated the formation of sedimentary rift rocks, extending across the Delamerian, the
andtrans-tensionalbasins, followedbytheem- Lachlan and the Thomson Orogen. The cycle
placement of I- and S-type granites. The cycle ended with regional deformation in the Early
peaked in the Tabberabberan Orogeny via the Carboniferous and the emplacement of post-
accretion of the Bendigo Terrane in the Middle tectonicgranites.
Devonian. Thiseventledtotheinversionofsed-
imentary basins and rifts, to the deformation of ThemetallogenyoftheTasmanFoldBelt,as
granitoids, and re-deformation and further im- described in Degeling et al. (1986), is mani-
bricationsofolderrocks. fold in respect to types and number of mineral
deposits. Only limited economic mineralisa-
The Hunter Bowen Supercycle, tion is found in the Kanmantoo Belt, mostly as
mainly reflected in the New England Orogen, stratabound copper deposits. During the Cam-
ischaracterisedbytheestablishmentofaconti- briantotheOrdovician,strataboundCuandMn
nental margin arc, a forearc basin and subduc- depositsformedintheLachlanFoldBeltwithin
tion related complexes to the east. The conver- avolcanicarcsetting. PorphyryCu-Audeposits
gent period lasted from Late Devonian to the originated during the latest stages of this tec-
Carboniferousandended,probablytriggeredby tonic setting in the Silurian, as well as post-
accretion of an intraoceanic arc, in exhumation accretionaryintheEarlyDevonian. Severalma-
of the subduction complex, imbrications and jor base metal deposits developed between the
uplift of the volcanic arc. The following exten- MiddleSilurianandEarlyDevonianduetorift-
sional period in Early Permian led to basin for- ingandvolcanism(Degelingetal.,1986). Con-
mations and granite intrusions and is believed temporaneous granite intrusions led to the for-
to reflect a rollback of the subduction zone. mationofSn-W,Moandbasemetal-Auminer-
TheformationoftheBowen-Gunnedah-Sydney alisation. Early to Middle Carboniferous felsic
Basin System, consisting of a series of rift and plutonic activity caused the genesis of Mo-W
trans-tensional basins, was initiated by a rift- and Sndeposits inthe easternpart ofthe Lach-
ing process west of the New England Orogen lan Fold belt. Progressing further east, epither-
inEarlyPermian. Duetoincreasedcrustalload malAu,associatedwithriftingandbimodalvol-
duringthecontinuingconvergenceandupliftof canism, formed in the Middle Devonian. Pre-
the arc, the basin converted to a foreland basin accretionary stratabound Cu and Mn, exhala-
in the Late Permian to Triassic. The collision tiveAu(EarlyDevoniantoLateCarboniferous),
with an intraoceanic arc in the Middle to Late andquartz-magnetitedeposits(LateCarbonifer-
Triassic lead to the termination of the Hunter- ous) are present in the New England Fold Belt.
Bowensuper-cycle. Furthermore, vein style deposits (Sn and Au-
As-Ag-Sb)originatedduetoS-typemagmatism
duringtheLateCarboniferoustotheEarlyPer-
mian.
9 |
ADE | 2. REGIONAL GEOLOGY
Figure 2.4: Map showing the
surface geology of New South
Wales. Yellow dashed line is the
approximate outline of the Lach-
lanFoldBelt. Mapisbasedonthe
1:5000000 Surface Geology map
ofAustralia.
extensional regime (˜505-495Ma). Contem- the continental margin (˜485Ma). Subduction
poraneously, oceanic crust developed off the zones developed on both sides of this back
Gondwana margin, subsequently forming the arc basin, east of the Delamerian Orogen and
basementoftheLachlanOrogen. inboardoftheMacquarieArc(˜460Ma).
Cratonisation of the Delamerian Orogen and Backarcbasinclosure(450to410Ma)
turbidite deposition in the Lachlan backarc The closure of the earlier described backarc
basin(490to460Ma) basin commenced due to double divergent sub-
At the start of this period, the Delamarian duction causing the development of extensive
Orogen was further affected by post orogenic thrust belts. Strong deformation (chevron-style
magmatism, followed by cooling and erosional folding) affected the turbititic sequence and
exhumation (˜490-480Ma). Enormous tur- fragments of oceanic crust were imbricated.
biditic fans were deposited within a marginal Extensive crustal thickening in the western
oceanicbasininadditiontominorquantitiesof Lachlan led to erosional exhumation of the
deep marine cherts (˜490-470Ma). The Mac- turbities.
quarie Volcanic Arc complex formed distal to
12 |
ADE | 2.2 The Lachlan Fold Belt
Docking, cratonisation of the western and Orogenies
central,andthedevelopmentofanAndean-type
A total of six orogenies, ranging for over
continental margin for the eastern Lachlan
100Ma, are interpreted to have affected the
FoldBelt(400to380Ma)
Lachlan Fold Belt (Gray and Foster, 1997), the
Duringthebasinclosure(finalisedat˜380Ma),
most important of them being the Benambran,
post-orogenic magmatism (˜400Ma) affected
theTabberabberanandtheKanimblanOrogeny
the western Lachlan Fold Belt, followed by
(Glen,2005). Allofthemarepredominantlyde-
amalgamation and cratonisation of the inner
finedbasedonunconformitiesintherockrecord
orogenic zone (western and central Lachlan
and/orchangesinsedimentationstyle,andmost
Fold Belt). Magmatism, metamporphism (high
ofthemarenamedaftertheirtypelocalities(ta-
temperature) and volcanism occurred along the
ble2.1).
Andean-type active continental margin in the
The timing of these six orogenic events is
easternLachlanFoldBelt.
shown in table 2.1, their regional extent in
figure 2.5. In the following, the orogenies
Rollback and post-orogenic extension
are summarised after Gray and Foster (1997),
(360to320Ma)
beginningwiththeoldest:
In-board convergence caused the inversion of
former extensional basins in the eastern Lach- The Benambran represents the oldest known
lan Orogen. The post-orogenic magmatism orogenic event in the Lachlan Fold Belt and
prolonged and the cratonisation of the Orogen occurred in the Early Silurian, mainly affected
wasfinalisedataround330Ma. the central parts of the belt. The orogeny
is characterised by at least two phases of
A
NOTE:
This figure/table/image has been removed
to comply with copyright regulations.
It is included in the print copy of the thesis
held by the University of Adelaide Library.
Table2.1: SummaryoftheorogeniceventsintheLachlanFoldBelt(modifiedafterGrayandFoster(1997)).
13 |
ADE | 2. REGIONAL GEOLOGY
deformation, accompanied by metamorphism
and intrusive events. East-west trending folds
and north-north-westerly striking structural
elementsarecharacteristic.
TheQuidonganOrogeny(lateEarlySilurian)
can’t be differentiated from the Benambran in
most of the Fold Belt, in particular in Victoria.
Thereforeitsregionalimportanceisuncertain.
The Bowning Orogeny is of Late Silurian to
Early Devonian age and caused wide spread
deformationintheeasternLachlanFoldBelt.
The Early Devonian Bindian Orogeny is
locally defined by an angular unconformity
betweenEarlyDevonianandSilurianvolcanics,
the latter characterised by intense foliation.
Regionally, however, the Bindian and the
Bowning Orogeny are indistinguishable, and
arenowviewedasthesameevent.
The Tabberabberan Orogeny was a major
period of regional deformation during the
MiddleDevonian,andistheorogenyregionally
Figure 2.5: Map of the Lachlan Fold Belt show-
most widely recognised in the Lachlan Fold ingtheapproximatearealextentofunconformities
Belt. This event was characterised by such a and regional deformation as well as the timing of
deformationalevents(modifiedafterGrayandFos-
prolonged intensity that subsequently the sedi-
ter(2004)).
mentary environment changed to non-marine.
The strength of deformation decreased towards
the northeast and caused open, close and tight
chevronfoldingofturbiditicrocksuites.
caused regionally extensive, although relative
weakdeformation. Generallyopen,locallytight
The Kanimblan Orogeny is the last event in
folding due to reactivation of older faults, are
thetectonicevolutionoftheLachlanFoldBelt.
themostimportantstructuralfeatures.
It occurred during the Early Carboniferous and
14 |
ADE | 2. REGIONAL GEOLOGY
schist facies, with the exception being hornfels truncated by major mylonitic strike-slip faults
contact zones in the vicinity of Devonian in- totheeastandthewest. Anultramafictomafic
trusions and unroofed amphibolite facies rocks melangezone,includingblueschistfaciescom-
alongtheMoystonFault,representingthewest- ponents,isdevelopedalongthewesternbound-
ernboundaryofthesubprovince. aryfault.
The Central Subprovince The Eastern Subprovince
Most of this subprovince is occupied by the In contrast to the other two subprovinces, here
metamorphic Wagga-Omeo Complex (figure folding is more open, characterised by an east-
2.6). The Complex is fault-bounded and wards trending fold tighteningaccompanied by
consists of complexly deformed Ordovician an increase in cleavage development. The sub-
metasediments, which experienced greenschist province is composed of anticlinorial and syn-
to upper amphibolite facies conditions. De- clinorial zones, bounded by east and west dip-
formation occurred during the Early to Middle pingreversefaultsandunderlainbycalcalkaline
Silurian accompanied or followed by the intru- volcanic arc basement. The prevailing litholo-
sion of large volumes of Silurian to Devonian gies are folded turbitites of Ordovician to Sil-
granites. The peak metamorphic conditions urian age. A Devonian belt of brittle thrusts is
are estimated at temperatures and pressures in developedintheeasternpartofthesubprovince,
theorderofapproximately700◦Cand3.5kbar, containing a melange of a Late Ordovician to
causing local anatexis and the formation of Early Silurian accretionary complex, in turn
migmatites. The subsequent erosional unroof- contained within a sequence of Late Cabrian
ingoftheWagga-OmeoComplexoccurreddur- to Early Ordovician cherts, basalts and tur-
ingtheMiddletoLateSilurianfollowedbythe bitites. Several foliated granites intrude along
emplacementasacrustalwedgeduringtheLate the thrust belt, causing relatively narrow low
Silurian to Early Devonian. The complex is pressure/high temperature metamorphic com-
A
NOTE:
This figure/table/image has been removed
to comply with copyright regulations.
It is included in the print copy of the thesis
held by the University of Adelaide Library.
Table 2.2: ImportantcharacteristicsofsubductionzonesintheLachlanFoldBelt(modifiedafterGrayand
Foster(2004))
16 |
ADE | 2.2 The Lachlan Fold Belt
plexes. Tectonic movement along these faults magma and that from partial crustal melting.
causedintensemultipledeformationofthewall Maficoceanicislandarcbasementandturbititic
rock, gneissic banding, mylonitic layering and rock suits are suggested as crustal components
secondaryfoliationinthegranites. affected by partial melting. Based on outcrop
pattern(e.g. shape,elongationandorientation),
2.2.4 Rifting induced basin devel- Foster and Gray (2000) suggested five major
opment and subsequent ig- graniteprovinces. Chappelletal.(1988)onthe
neous activity∗ other hand defined nine, potentially ten petro-
logical granite suites characterised by common
The Lachlan Orogen was affected by several chemical features reflecting their source rock
extensional episodes leading to the develop- composition. He linked the source variation to
ment of rift basins and half grabens, involv- blocksorterranesofcontinentaland/orsubcon-
ing granitic magmatism and silicic and bi- tinental lithosphere or basement. Variations in
modalvolcanism(FosterandGray,2000). Ma- their fertility in respect to precious and base
jor periods of basin formation occurred during metal content might have had significant influ-
the Early Silurian (˜440 to 430Ma) within the enceonthegenesisoforedeposits.
Wagga-Omeo-Complex (figure 2.6) and during
the Late Silurian to Early Devonian (˜420 to
2.2.5 Metallogeny of the Lachlan
400Ma) east of the belt , e.g. causing the de- Orogen†
velopmentoftheCobarBasin.
Granitesmainlyoccurinthecentralandeast- Diverse metallogenetic associations, signifi-
ern subprovinces and comprise up to 36% of cantly varying across the fold belt and com-
theexposedLachlanOrogen(figure2.7). Mag- prising more than 22 mineral commodities, are
matism occurred over a prolonged time (440 to found throughout the entire Lachlan Fold Belt
340Ma) culminating between 410 and 380Ma (Hough et al., 2007). The vast majority of
(Gray and Foster, 1997). According to Foster the mineral deposits formed between approx-
and Gray (2000), syntectonic plutons formed imately 450 to 360Ma. Many major por-
in the high-grade metamorphic complexes of phyry copper-gold and orogenic gold deposits
central and western subprovinces during 430 are interpreted to have formed during a metal-
and 415Ma (e.g. Bega batholiths), followed logenically significant period at approximately
by an eastwards directed migration of the mag- 440Ma.
matic front towards the coastal areas between From west to east, Bierlein et al. (2002)
405 and 390Ma. Contemporaneous with the
∗.Summarisedafter
latter period, numerous high-level plutons in-
Foster,D.andGray,D.(2000).EvolutionandStruc-
truded in the western, in addition to predomi- ture of the Lachlan Fold Belt (Orogen) of Eastern Aus-
nantly post-tectonic plutons in the central sub-
tralia.AnnualReviewsinEarthandPlanetarySciences,
28(1),47–80,
provinces. Post-tectonicgranites,interpretedto Gray, D. and Foster, D. (1997). Orogenic concepts-
be linked to the formation of caldera systems, application and definition: Lachlan Fold Belt, eastern
Australia. American Journal of Science, 297 (9), 859–
developed between 370 and 360Ma in central 891and
Victoria(FosterandGray,2000). Awiderange Chappell, B.,White, A.andHine, R.(1988).Gran-
ite provinces and basement terranes in the Lachlan
ofigneousrocks,rangingfrombasalt/gabbroto
FoldBelt,southeasternAustralia.AustralianJournalof
alkali-feldspargranitesoccur,withS-type(per- Earth Sciences,35(4),505–521
aluminous, containing cordierite) as well as I- †.Summarisedafter
Bierlein, F., Gray, D. and Foster, D. (2002). Met-
type(metaluminous,containinghornblendeand
allogenicrelationshipstotectonicevolution-theLachlan
biotite) geochemical signatures. The different Orogen,Australia.EarthandPlanetaryScienceLetters,
202(1),1–13and
types are interpreted to be caused by variations
Hough, M., Bierlein, F. and Wilde, A. (2007). A
in the degree of mixing between two magma review of the metallogeny and tectonics of the Lachlan
types: subduction related mantle-derived mafic Orogen.Mineralium Deposita,42(5),435–448
17 |
ADE | 2. REGIONAL GEOLOGY
defined five metallogenetic provinces: (1) tionarycollisionzone.
turbidite-hosted orogenic lode gold (±As-Sb), The sediment-hosted gold-copper-lead-zinc
(2) granite-related magmatic gold (±Mo-Sb- deposits of the central Lachlan Orogen (e.g.
Cu-Te), (3) epigenetic sediment-hosted base Elura, Cobar deposits and Queen Bee) occur in
metal and gold, (4) granite-hosted tin-tungsten turbidites containing variable quantities of vol-
and (5) porphyry copper-gold and volcanic- canics(Houghetal.,2007). Theyformedalong
hosted massive sulphides. Spatially, the de- or in the vicinity of the margin of sedimentary
posit types occurring throughout the Lachlan basins, developed in a rift-related extensional
reflect the geotectonic setting in which they and transpressional tectonic regime. Typically,
had formed (Hough et al., 2007). The west- the sulphide ore is characterised by a parage-
ern Lachlan Orogen is dominated by Silurian nesis consisting of base metal sulphides, mag-
to Devonian orogenic and magmatic gold de- netite,pyrite,pyrrhotite,goldand/orsilver,and
posits developed within an accretionary colli- occurs as massive lenses or subvertical pipes,
sion zone. In the central Lachlan Orogen, (a) stringerveins,orasstockwork. Alterationtypes
tin-tungstendeposits(Silurian)formedasacon- (e.g. silicification, chloritisation, sericite, car-
sequence of partial melting of the lower crust bonate, iron sulphide and iron oxide alteration)
in a subduction setting; (b) orogenic gold is andtheirintensitiesvarybetweendeposits. Fre-
found in an accretionary collision zone; and quently, mineralisation is hosted in high-strain
(c) rifting and accompanying bi-modal volcan- zones and truncated by deeply dipping faults.
ism led to the development of sediment-hosted Most ore bodies exhibit at least to some ex-
copper-gold and base metal deposits (Late Sil- tent a structural control and therefore have to
urian to Early Devonian). The eastern Lach- be syn-deformational and epigenetic in nature.
lan Orogen accommodates (a) porphyry gold- Theexacttimingoforeformationofseveralde-
copperdeposits(Ordovician)relatedtooceanic posits is, however, still debated, consequently,
island arc; (b) volcanic-hosted massive sul- theclassificationassyn-orepigeneticisequiv-
phide, sediment-hosted copper-gold and base ocal (Bierlein et al., 2002; Hough et al., 2007).
metal deposits of Late Silurian to Early Devo- Themostimportantsedimenthosteddepositsof
nian age in rift basins; and (c) orogenic gold theCobarregionaredescribedinmoredetailin
deposits(MiddletoLateDevonian)inanaccre- chapter2onpage31.
18 |
ADE | 2.3 The Cobar Basin
The geology of the Cobar Basin is presented the presence of syn-deformational faults. The
infigure2.9. Thebasinistruncatedontheeast- Cobar Basin is asymmetrical in shape, with the
ern margin as well as underlain by the Ordovi- westernedgesteeperthantheeasternedge,sug-
cian metasediments of the Girilambone Group gestingahigherimportanceofthewesternbasin
and by Late Silurian granitoids (Glen, 1990). edgeduringbasinopening.
Those multiply-deformed Ordovician metased-
iments, which experienced metamorphic con- 2.3.2 Stratigraphy∗
ditions between greenschist and lower amphi-
The earliest sediments deposited during syn-
bolites facies, were deposited in the back arc
rift stages on a rapidly subsiding and unstable
Wagga Basin (Suppel and Scheibner, 1990;
shelf are those of the Mouramba Group (Glen,
Glen,1990).
1990). The sediments were sourced from base-
The Girilambone Group is divided into two
mentrockstothesoutheastandeastandconsist
sub groups according to their metamorphic
of outwash fans and shallow water clastics, lo-
grade and rock types (Fergusson et al., 2005).
cally with minor felsic volcanics. The remain-
The western part, the Ballast Formation, con-
ingsyn-riftphaseischaracterisedbythedeposi-
sists of low grade metamorphic quartz-rich tur-
tionofturbiditicsedimentsandminorfelsicvol-
biditesandbeddedchertsofMiddletoLateOr-
canics of the Nurri Group and turbidites from
dovician age. The Ballast Formation forms the
the lower Amphitheatre Group (figure 2.10).
easternHinterlandtotheCobarbasinandthere-
The sediment source is the main difference be-
fore represents an important sediment source
tween those two units, with the Nurri Group
feeding into the basin during formation. Rocks
sediments sourced from the Girilambone base-
of the eastern part of the Girilambone Group
ment along the eastern basin margin, whereas
experiencedslightlyhighermetamorphicgrades
the lower Amphitheatre Group sediments de-
uptothelowestamphibolitefaciesandarecom-
rived from the northwest and west from the
posed of phyllites as well as psammitic and
Winduck Shelf, and to some extent from gran-
maficschists.
ites to the southwest (Glen, 1990). The dif-
The Silurian granites intruded at approxi-
ference in sediment source is also reflected by
mately 420Ma (figure 2.9) and are of post-
the occurrence of shallow water fossils in the
tectonic origin and S-type in character (Glen,
westerly-derived turbitides in contrast to the
1990). The Early Devonian rock sequences
fossilpoornatureofeasterly-derivedsediments
deposited in the composite Darling Basin and
from the Girilambone metasediments. Further-
its flanking shelves comprise the Cobar Su-
more, the Nurri Group is characterised by up-
pergroup (Suppel and Scheibner, 1990). The
wardfining,whereastheAmphitheatreGroupis
Kopyje Shelf is developed at the northern and
an upward coarsening unit commensurate with
eastern, and the Winduck Shelf at the west-
shallowwaterdeposition.
ern basin margin(figure 2.9). The former is
TheChesneyFormationrepresentsthelower
composed of siliciclastics and limestones of
Lochkovian age. Pragian storm sediments and
∗.Summarisedafter
littoralsandsarethemainlithologiesoccurring Glen, R. (1990). Formation and inversion of transten-
intheWinduckShelf(Glen,1990). sionalbasinsinthewesternpartoftheLachlanFoldBelt,
Australia,withemphasisontheCobarBasin.Journalof
TheCobarBasinispredominantlyfilledwith
Structural Geology,12(5-6),601–620,
turbiditic sediments sourced from the eastern, Scott, A.andPhillips, K.(1990).CSAcopper–lead–
zinc deposit, Cobar. In: F. Hughes (ed.), Geology of the
northern and western hinterland with only mi-
mineral deposits of Australia and Papua New Guinea,
nor occurrences of volcanic rocks, mostly as vol.2,Australasian Institute ofMiningandMetallurgy,
Melbourne,pp.1337–1344and
tuffunits. Themaximumthicknesshasbeenes-
Glen, R., Drummond, B., Goleby, B., Palmer, D.
timatedviadeep-crustalreflexionseismicatap- and Wake-Dyster, K. (1994). Structure of the Cobar
proximately6km(Glenetal.,1994). Thedepth Basin,NewSouthWales,basedonseismicreflectionpro-
filing.AustralianJournalofEarthSciences,41(4),341–
is variable and is to some extent controlled by 352
21 |
ADE | 2.3 The Cobar Basin
part of the Nurri Group (figure 2.10) and con- tency of mud-, silt- and sandstone beds indi-
sistsoflitharenite,minorconglomerateandsilt- cate a much more tectonically inactive deposi-
stone. It is overlain by the Great Cobar Slate tionalenvironment. Thegentlesubsidencedur-
(figure 2.10), composed of mud- and siltstone ingthesag-phaseofthebasinformationcaused
(Scott and Phillips, 1990). The lower Am- thedepositionoftheupward-shallowingpartof
phitheatre Group consists predominantly of the the Winduck Group (figure 2.10). They were
CSA Siltstone as the lower stratigraphic unit deposited over older Ordovician basement and
andisoverlainbythesandstone-dominatedBid- Silurian granites west of the Cobar Basin and
dabirra Formation (figure 2.10). The CSA Silt- consist of turbidites and shelf sediments (Glen,
stoneischaracterisedbyrhythmicallythinlyin- 1990).
terbedded mudstones and siltstones, and sand- The two-stage sedimentary characteristics of
stone beds at varying thicknesses occurring at theAmphitheatreGroupandtheoccurrencesof
mostly irregular intervals (Scott and Phillips, volcanic rock units in the early stage of basin
1990). development led to the interpretation of a rift
The deposition of the upper Amphitheatre related origin of the basins in the Cobar region
Group flags the commencement of the post- (Glen, 1990). However, the occurrence of vol-
rift phase of basin development (figure 2.10). canic units is more the exception than the rule
These turbiditic sediments are similar to those and they are commonly absent. Deep crustal
ofthelowerAmphitheatreGroup. However,the seismic reflection data indicates a mid-crustal
abruptchangeinbeddingthicknessandconsis- detachment formed between lower and upper
Figure2.10: StratigraphiccolumnoftheCobarBasinmodifiedafterGlen(1994);structuralzonesasdefined
byGlen(1990).
23 |
ADE | 2. REGIONAL GEOLOGY
crust due to extension (Glen et al., 1994). The inversion (figure 2.10). A general trend is ap-
interpreteddetachmentplanesurfacesnotatthe parent,withhighstrainzonesmainlyoccurring
basin margin but west of it. The tectonically on the eastern basin margin and decreasing to-
highlyactiveinitialstageofbasindevelopment, wardsthecentreandwest-wards. Anill-defined
accompanied by pronounced subsidence, was third zone, characterised by low strain, occurs
followedbyaquieterperiod. Glenetal.(1994) inthenorth-westernpartoftheCobarBasin.
concluded that the Cobar Basin had developed
asarampbasinandthetwo-stagesedimentation Structural Zone 1
characteristicscanalsobeinterpretedasaresult
The D1 high-strain zone is confined by the
ofaregionalthermaleventduringthebasinfor-
RookeryFaulttotheeast,theMyrt-ThuleFault
mation that caused increased subsidence rates
systemtothewestandbytheCrowlCreekFault
ontheeasternbasinmargin.
to the south. The following characteristic fea-
2.3.3 Structural setting∗ tures can be summarised for this southwards
wideningzone: (1)regionalintensesub-vertical
Several major syn-depositional fault systems S1 cleavage, overprinting an early cleavage
developedduringthebasinformationandmany obliquetobedding;(2)aregionaldown-dipex-
ofthosewerereactivatedduringbasininversion. tension lineation L1, meso- to macroscopic F1
The eastern margin of the Cobar Basin is de- folds. According to Glen (1990), these struc-
fined by the Rookery Fault as shown in figure tures developed due to right-lateral transpres-
2.10(Glen,1990). Thefaultactivitywasgreat- sion above a positive flower structure. The
estduringtheinitialstageofbasinformationin zone is highly imbricated and contains two in-
the Lochkovian. The interpretation of faults on ternalNNW-trendingfaults–theCobarandthe
thewesternbasinmarginisdifficultduetosyn- GreatChesneyFaults. ThelatterisasteeplyE-
rift sediments being covered by either post-rift dippingfaultandrepresentsthecontactbetween
orTertiarylithologies. However,itseemslikely, the folded and imbricated Great Cobar Slate in
thatthewesternmargincorrespondstotheJack- the footwall to the west and the folded Ches-
ermaroo Fault, separating shallow-water sedi- ney Formation in the hangingwall to the east.
ments on the west, from turbitites on the east To the south, the Great Chesney Fault is lost in
(Glen,1990). Thenorthernbasinmargin,which the Great Cobar Slate, whereupon the faulting
isinferredtobetheLittleTankFault,separates seemstohavebeentransferredsoutheast-wards
the northern Kopyje Shelf sediments from tur- to a zone of closely spaced ductile-brittle fault
bitites to the south. The Crowl Creek Fault array in the Peak area – the Great Peak Fault,
represents the southern boundary and separates Blue Shear and Lady Greaves Shear (Glen,
the siliciclastic Cobar Basin from the southern 1990).
Mount Hope Trough, dominated by deep-water Several mineral deposits are present in this
sediments and volcanics, respectively. Sudden structural zone. Most of deposits of the Co-
thickness changes in the Biddabirra Formation bar Gold Field, e.g. The Peak, New Occi-
and to a lesser extent in the CSA Siltstone dur- dental, Chesney and New Cobar are localised
ing Pragian times, suggests the development of along regional NNW-trending faults, the Great
intra-basinalfaultsystems(Glen,1990),e.g. the Chesney and the Great Peak Faults. The CSA
MyrtFault(figure2.10),theBundellaFault,the andtheGreatCobardepositsformedonthrusts
AmphitheatreFaultandtheBuckwarroonFault
∗.Summarisedafter
(figure 2.10). The most important faults are
Glen, R. (1990). Formation and inversion of transten-
summarisedafterGlen(1990)intable2.3. sionalbasinsinthewesternpartoftheLachlanFoldBelt,
Glen (1990) defined two zones of high and
Australia,withemphasisontheCobarBasin.Journalof
Structural Geology,12(5-6),601–620and
low strain defined by regional structures that Glen, R. (1995). Thrusts and thrust-associated min-
hadformedorwerereactivatedduringthebasin eralization in the Lachlan Orogen. Economic Geology,
90(6),1402
24 |
ADE | 2.3 The Cobar Basin
within imbricated plates, the Cobar and Ches- 2.3.4 Igneous activity within
neyplates,respectively(Glen,1995). and proximal to the Cobar
Basin∗
Structural Zone 2
Siliceousvolcanicsrepresentonlyaminorcom-
Most of the Cobar Basin was affected by much ponentoftheexposedCobarBasin. Severaltuff
lower strain and can be summarised in the D1 layers have been identified in the Elura deposit
low-strain zone, which in turn can be divided since its discovery. Thicknesses of those tuffs
into five sub-zones or blocks. Those are from vary,rangingbetween<1to˜20cm. Theydis-
north to south: the Bundella Block, the Mary- playtypicalsedimentaryfeaturesdevelopedun-
valeBlock, theOlinoBlockandtheBiddabirra der subaqueous conditions, i.e. a sharp lower
Block. The low-strain zone extends from the layercontactandanuppergradationalzoneinto
BuckwaroonFaultinthewesttotheMyrt-Thule mud-andsiltstone.
Fault system in the east and is confined by the The Silurian Tinderra Granite (410 ±10Ma;
Little Tank Fault in the north. The southern Pogson and Hilyard, 1981) is the only known
truncation is unclear. Two important structure intrusion proximal to the Elura deposit. The S-
sets were identified in this zone: (1) D2 struc- type granite is localised at the basin boundary
tures,includingvariablydevelopedS2cleavage at a distance of around 30km in north-easterly
andvariablyplunging;(2)uprightopenF2folds direction from the deposit (figure 2.9 and fig-
(variableinstrikedirection)whicharerefolding ure 2.10). In the vicinity of the CSA deposit, a
earlier F1 folds. F1 are accompanied by a sub- 30mthickchertybedisinterpretedtorepresent
verticalS1cleavageattimes(Glen,1990). a tuff unit (Scott and Phillips, 1990). Rhyolitic
The Bundella Block is located in the north- to rhyodacitic volcanics occur within the host
ernmost part of the low-strain zone. It is rocksequenceofthePeakdeposit(Hinmanand
confined by the Little Tank fault to the north Scott,1990).
and by the Bundella Fault to the south. Due By contrast, the southern extensions of the
to major thickening of the Biddabirra Forma- CobarBasin,theRastandtheMtHopeTroughs
tion, the Bundella Fault is interpreted as a re- (figure 2.8), are characterised by significant ig-
activated syn-depositional growth fault. The neous activity. The Ural Volcanics, contained
Bundella Block is internally characterised by within the Rast Trough around 200km south
WNW-trendingF1foldswithanaccompanying of Cobar, predominantly comprise felsic por-
subvertical S1 cleavage. These folds were af- phyric facies and are cross-cut by very minor
fected by a later deformation D2. The overall highlyalteredmaficandintermediatedykesand
occurrence of the S1 cleavage in the Bundalla intrusives. The deposition of these volcanic
Blockisasignificantdifferencecomparedwith rock suites occurred in a submarine environ-
otherareasinthelow-strainzone,andindicates ment. Concurrenteruptionswereboth,eruptive
increasedstraininthisparticularblock. During
∗.Summarisedafter
theD2event,theolderstructuralelementswere Scott, A.andPhillips, K.(1990).CSAcopper–lead–
folded into NE-trending upright F2 folds, with zinc deposit, Cobar. In: F. Hughes (ed.), Geology of the
mineral deposits of Australia and Papua New Guinea,
a locally occurring subvertical NE-trending S2
vol.2,Australasian Institute ofMiningandMetallurgy,
cleavage(Glen,1990). Melbourne,pp.1337–1344,
Hinman, M.andScott, A.(1990).ThePeakgoldde-
The Elura deposit is located in this block.
posit,Cobar.In:F.Hughes(ed.),Geologyofthemineral
Glen (1990) interpreted several blind thrusts depositsofAustraliaandPapuaNewGuinea,vol.2,Aus-
tralasianInstituteofMiningandMetallurgy,Melbourne,
throughout the Cobar Basin. Glen (1995) sug-
pp.1345–1351and
gested the deposit formed at the end of such a Bull, K., Crawford, A., McPhie, J., Newberry,
blindthrust,whichactedasamajorpathwayof R. and Meffre, S. (2008). Geochemistry, geochronol-
ogyandtectonicimplicationsofLateSilurian–EarlyDe-
ascendingmineralisingfluids.
vonian volcanic successions, Central Lachlan Orogen,
NewSouthWales.AustralianJournalofEarthSciences,
55(2),235–264
25 |
ADE | 2. REGIONAL GEOLOGY
and effusive in nature. Based on U-Pb zircon roclastics and minor mafic-intermediate cross-
ages, deposition is interpreted to have occurred cuttingdykesarepresent. TheMtHopeTrough
at approximately 411 to 395Ma (Bull et al., ishosttoseveralgraniticintrusions(e.g. Thule,
2008). Coan and Boohlahbone Granites). The most
The Mt Hope Volcanics occur within the prominent is the I-type Thule Granite, inter-
trough approximately 100km south of Cobar. preted to underlie the volcanic rock sequence.
Most of exposed felsic volcanic facies are rhy- An approximate age of the Mt Hope Volcanics
olitic to dacitic in composition, porphyritic to has been estimated at 411 to 400Ma. The age
aphantic and commonly feature spherulitic tex- isconstrainedbyK-Ardatingofthegraniteand
tures and flow-banding. They form sills, minor themarinefossil-bearingcoversequenceabove
dykes and lava flows, which in places are auto- the volcanics (Bull et al., 2008). The Boolah-
brecciated. Abundant, mostly syn-eruptive py- boneGraniteisanA-typegranite.
Fault Age Activity Type of fault
Separates shelf from basement; west-dipping. Eastern edge of Cobar basin
in the north, hinge zone overlapped by inter-fingering facies in south- Strike-slip,
Rookery Lochkovian
eastern corner. Infra-basinal conglomerates and slumped blocks of oblique-slip
limestone in basin sediments attribute to faulting.
Lochkovian
Little Tank Separates shelf from basin Extensional
(inferred)
Strike-slip
Separates basal Lochkovian shelf sediments from basal Lochkovian
Jackermaroo Lochkovian (inferred),
turbitites.
oblique-slip
West dipping. Dramatic thinning of Biddabirra Formation and CSA Strike-slip,
Myrt Pragian
Siltstone. Inactive before end of Biddabirra deposition oblique-slip
Bundella Pragian South dipping, confine Biddabirra formation to the south Extensional
Lochkovian
Crowl Creek to Pragian Separates southern volcanics from northern clastics Extensional
(inferred)
Buckwaroon Pragian Thickening of the Biddabarra Formation to the east Transfer fault
Table 2.3: Majorsyn-depositionalfaultsintheCobarBasinafter(Glen,1990).
26 |
ADE | 2. REGIONAL GEOLOGY
Volcanic-hosted massive sulphide de- seafloor hydrothermal-seawater mixing are im-
∗
posits portant mechanisms for ore precipitation. Ex-
halites, such as banded iron formations and
ComparedtoMITandBHTdeposits,sustralian tourmalinites,precipitateontheseaflooraround
VHMS deposits are relatively small in size and theoutflowzone.
containaround16.52Mtzinc,5.03Mtleadand
19,730t silver. Some large examples are Mt
†
Broken Hill-type deposits
Lyell,RoseberyandHellyerinthewesternTas-
mania region, the Benambra district in Victo- Broken Hill and Cannington are economically
ria, Mt Morgan in Queensland and Woodlawn the most important examples of this deposit
inNewSouthWales. style. The BHT mineral systems are famous
Deposits of this style formed in submarine for their large size and their high grade meta-
back-arc basins, characterised by significant morphicore(amphibolitefaciesorhigher). The
volcanic activity, causing high heat flow sub- world’s largest Zn-Pb-Ag deposit, Broken Hill
sequently driving large hydrothermal systems. (Australia) has produced some 300Mt of high-
Observedorefluidtemperaturesareintheorder gradeorefor60Mtofmetalproducingrevenue
of200to350◦C.Mostlypolyphasesubvolcanic of $350billion. The wealth generated by the
intrusives, characterised by varying chemistry explotation of the orebody significantly aided
rangingfromtrondhjemitic,tonalitictogranitic the industrialisation process of Australia, cre-
and dioritic, represent the heat sources. The atedBrokenHillProprietaryCompanyLimited
metalsarepredominantlyleachedfromthevol- (BHP) in 1885 and financed the precursor to
canicsviacirculatingseawaterbrineswithami- Rio Tinto. The 1685Ma orebody hosted by
nor contribution of magmatic fluids. Sulphur metasediments comprises nine masses of sul-
is sourced from both, seawater and the leached phide rocks that were affected by multiphase
volcanicrocksequences. coeval high-grade metamorphism and intense
deformation in the Olarian Orogeny (1600Ma)
Massivesulphidelensesaredevelopedabove
and coeval lower grade metamorphism and de-
footwallalterationzonesdominatedbychlorite-
formationintheDelamerianOrogeny(500Ma).
and/or sericite and quartz alteration, and can
In 2006, all Australian BHT deposits com-
containingsignificantquantitiesofcopper. The
binedcontainanestimatedunminedresourceof
regionalalterationissemiconformable,indicat-
26.25Mtzinc,33.92Mtleadand65,600tsilver.
ing subhorizontal fluid flow direction. Trans-
gressive alteration is developed at deposit scale The generalised current genetic model for
as a consequence of ore forming fluids migrat- BHT ore forming systems includes syngenetic
ing and focussing along major fluid conduits or diagenetic deposition and syn- or post-
(e.g. synvolcanic faults). The volcanic host se- tectonic replacement. The mineralisation
quences are characterised by epidote, hematite, is hosted by metamorphic overprinted up-
albite and K-feldspar alteration. The extent of ward coarsening siliciclastics within sediment-
alteration varies in size ranging between 150
∗.Summarisedafter
to 500m. The apparent sulphide zoning is de-
Huston, D.,Stevens, B.,Southgate, P.,Muhling,
fined by temperature gradients with high-T as- P. and Wyborn, L. (2006). Australian Zn-Pb-Ag ore-
semblage (pyrite and Cu-bearing sulphides) at forming systems: A review and analysis. Economic Ge-
ology,101(6),1117
the base and lower-T mineral parageneses (Zn-
†.Summarisedafter
Pb-Ag sulphides and minor barite) at the top Huston, D.,Stevens, B.,Southgate, P.,Muhling,
of the deposits. The ore formation is con- P. and Wyborn, L. (2006). Australian Zn-Pb-Ag ore-
forming systems: A review and analysis. Economic Ge-
trolled by local structures, providing fluid fo-
ology,101(6),1117,
cusandpressuredropindilationalsites,aswell Plimer, I. (2006). Manganoan garnet rocks associated
withtheBrokenHillPb–Zn–Agorebody,Australia.Min-
as changes in the volcanic rock facies. Fur-
eralogy and Petrology,88(3),443–478and
thermore, replacement of host rocks and sub- personalcommunicationwithIanR.Plimer(2010)
28 |
ADE | 2.4 Australian Zn-Pb-Ag ore systems
dominated basins, which developed in a con- ken Hill orebody are interpreted as exhalites
tinental rift setting. These basins are charac- whereasstrikeequivalenttourmaliniteandpos-
terised by elevated geothermal gradients and sibly some quartz-gahnite rocks may have
were affected by igneous activity, such as the formed from stratal flow and metal precipita-
intrusions of high Fe-high Ti tholeiitic basalt tioninpsammiteaquifers. Ascendingfluids,in-
sills (1685Ma) at Broken Hill. The formation fluxing and replacing permeable psammitic se-
ofigneousrockstookplaceapproximatelycon- quences, formed stratiform sub-seafloor base-
temporaneous to ore formation. Interbedded metal mineralisation (e.g. Western Mineral-
pelite and psammite units constrained sulphur- isation in Broken Hill). Zones of minerali-
poor geothermal systems that leached evapor- sation are commonly enveloped by a quartz-
ites to provide hypersaline fluids, which subse- manganoan almandine-gahnite alteration as-
quently leached metals from deeper rock units. semblage(BrokenHill)orbyquartzosebiotite-
Reactivemineralsinbasalt,felsicvolcanicsand sillimaniteschistsandfeldspaticgarnet-bearing
felsic metasediments were altered and are now psammites (Cannington). In Broken Hill, alter-
represented by an almandine-quartz-sillimanite ationassociatedwithlead-dominatedorebodies
overprint. Metasediments stratigraphically be- ismostlyabsent.
low the Broken Hill orebody show extensive DeformationduringtheOlarianandDelame-
premetamorphic hydrothermal alteration and rian Orogenies remobilised the base metal sul-
now comprise blue quartz-manganoan garnet- phides into fold hinges and dilational jogs.
ferroan gahnite-plumbian orthoclase assem- Manganese, zinc and iron reacted with the en-
blages with minor and variable amounts of sul- closingmetasedimentstoformspessartine-and
phides. gahnite-rich lithologies. Remobilised ore has a
relativeenrichmentinPb,Ag,Cu,AsandSb.
Contrary to VHMS deposits, water was
sourcedfromintercontinentalfreshwaterlakes. Zinc,lead,ironandsilveroccurassulphides,
A sudden deepening of the basin caused an silicates, spinels, oxides or as native phases.
over-pressurisation of oxidised metal bearing Themineralparagenesisandthefactthatpyrite
low-sulphur hydrothermal fluids circulating in isabsentinBHTdepositssuggestsformationin
aquifers. Fluid migration, focus and, subse- a sulphur-poor environment. Sulphur isotopes
quentlythelocationofalterationandoreprecip- of various mineral phases are consistently at or
itation were mainly controlled by active struc- close to δ34S=0 . Strontium, lead and boron
tures enabling the breach of capping, imper- isotopes suggest(cid:104)mantle, crustal and evaporitic
meable rock units. Hydrothermal fluids re- sources. The genetic model proposed for the
leased from their aquifers caused transgres- BrokenHillbasemetaldepositinvolvesorefor-
sive sulphide-bearing footwall alteration zones. mationbyreplacementandlakefloorhydrother-
Those zones subsequently acted as major fluid mal sedimentation in a shallow fresh water rift
conduit for episodic pulses. Broken Hill’s C- lake.
Lodeformedalongsuchaconduitwherethees-
capingfluidsprecipitatedasironformationson ∗
Mississippi Valley type deposits
theseafloor,nowcomprisingathinfinelylami-
nated quartz-magnetite-spessartine-fluorapatite Several Australian MVT deposits are found
±hyalophane-sphalerite-gahnite-plumbian or- alongtheLennardshelfandtheBroomplatform
thoclase rock. Later, these fluids deposited in Western Australia, the largest of them being
stratiform base metal mineralisation, e.g. the the Admiral Bay deposit. The MVT deposits
zinc dominated B- and A-Lode, and No 1
Lens, as well as the stratigraphically higher
∗.Summarisedafter
lead dominated No 2 and No 3 Lenses in Huston, D.,Stevens, B.,Southgate, P.,Muhling,
P. and Wyborn, L. (2006). Australian Zn-Pb-Ag ore-
Broken Hill. Spessartine- and gahnite-bearing
forming systems: A review and analysis. Economic Ge-
rocks stratigraphically equivalent to the Bro- ology,101(6),1117
29 |
ADE | 2. REGIONAL GEOLOGY
arerelativelysmallinsizeandcontainatotalof hydrocarbons, fluorite, dolomite, calcite and
10.7Mtzinc,5.89Mtleadand4,888tsilver. barite.
The deposits are of epigenetic origin, strata-
bound and formed in intracratonic settings in Mount Isa-type deposits∗
sag or rift basins. Host lithologies are pre-
The four most important deposits of this style
dominantly marine platform carbonates. The
are HYC (Here’s Your Chance) deposit in the
estimated depth of ore formation ranges be-
McArthur region, and the Hilton-George Fis-
tween a few hundreds of metres and up to
cher,theMountIsaandtheCenturydepositsin
2km. The basinal rock sequences comprise
the western succession of the Mount Isa Inlier.
carbonates,mudstones,evaporates,andcalcare-
With77.09Mtzincand34.44Mtlead,theAus-
ous/carbonaceous shales and siltstones. The
tralianMITdepositscontaintheworld’slargest
basinstendtoshowweaklyreducingconditions.
base metal accumulations. The combined sil-
Basal permeable sandstone layers and reacti-
vercontentofallMITis61,890t,slightlylower
vated syn-depositional growth faults on basin
comparedtoBHTdeposits.
margins act as major fluid conduits. Due to the
Despite their similar large size compared to
historiccomplexityofintracratonicbasindevel-
BrokenHill-typedepositsandtheoccurrenceof
opment, source rocks are not easily identified.
volcanic rock suits within the basins in which
Basinalfluidsareeitherdrivenbythediagenetic
theyarehostedin,MITdepositsareinterpreted
compaction of the sediment column, hydrauli-
to have formed without immediate and co-
cally due to tectonic uplift, forced by orogenic
eval contribution of volcanism or magmatism.
compressionorduetogasexpulsionasaconse-
They are hosted by pyritic and carbonaceous
quenceofadropinsealevel. Metalwassourced
dolomitic siltstones developed within intracra-
andleachedfromthesiliciclasticsedimentsand
tonic extensional basins which formed due to
subsequentlytransportedviachloridecomplex-
tectonic activity along the North Australia cra-
ation. Seawaterand/orevaporiticsequencesare
ton. Rock types occurring in these basins are
interpretedaschloridesource. Sulphurisinter-
dominatedbysiliciclastics(laminateddolomitic
preted to be either transported via the basinal
siltstone and shale, quartzose sandy dolomite
brine as H S or sulphate, or sourced locally at
2
and sandstone), bimodal volcanics, felsic mag-
the site of ore precipitation from a second fluid
maticrocksuits,platformandrampcarbonates.
or from evaporites. Hydrocarbons and/or or-
Similarily to MVT deposits, temperatures of
ganic matter genetically linked to the ore for-
ore formation were relatively low (70-240◦C)
mationisinterpretedtobesourcedfromdeeper
andachievablewithouttheneedofanadditional
levels within the basin and not generated at the
heat sources such as granitic intrusions. Metal-
siteoforedeposition.
liferous fluids migrated along major sandstone
Mineralisation generally occurred during the
dominated aquifers contained within the basi-
latest stages of main subsidence prior to ex-
nal rock sequence and subsequently ascended
humation. Temperature conditions (80 to
along deep penetrating basin-scale faults. The
200◦C) reflect normal geothermal gradients
flow of these brines was controlled similar-
withoutmagmaticactivity. Sulphidedeposition
ily to MVT deposits, e.g. by over pressurisa-
occurred in or near reactivated basin-bounding
tion of aquifers caused by extension. Evapor-
faults as replacement of evaporates, hosted in
itic sequences are interpreted to represent the
hydrothermalsolutionbrecciasorinpermeable
sourceforsulphurandchloride. Volcanicrocks
reef carbonates. Most commonly, these depo-
contained in the basin sequence have been af-
sitional sites are truncated by sequences char-
acterised by low permeability (e.g. granites,
∗.Summarisedafter
shales). The alteration haloes are characterised Huston, D.,Stevens, B.,Southgate, P.,Muhling,
P. and Wyborn, L. (2006). Australian Zn-Pb-Ag ore-
by mineral assemblages typically composed of
forming systems: A review and analysis. Economic Ge-
quartz, siderite, gypsum, magnetite, marcasite, ology,101(6),1117
30 |
ADE | 2.4 Australian Zn-Pb-Ag ore systems
fected by hydrothermal alteration and are char- are similar, e.g. vertically elongated massive
acterised by losses of zinc, lead and copper. sulphide pipes, lenses or sulphides in stringer
These volcanics represent the most probable zones; (5) the mineralisation is exclusively dis-
metal sources. Thermal sulphate reduction is cordanttohostrocksequences;(6)geneticlinks
describedasthemostlikelyandmostimportant to igneous activity have not been identified in
mechanism for the generation of H S for base any of the known Cobar deposits. However,
2
metal formation. Alteration is typical potassic there are some differences. The obvious north-
innature,characterisedbymineralassemblages south trending change of ore mineral paragen-
eithercomposedoforthoclase-quartz±sericite esis, characterised by metal assemblages dom-
±hematite ±dolomite ±anatase ±barite, or inated by Pb-Zn-Ag ±Cu at the Elura deposit
chlorite-orthoclase-quartz. via Cu-Zn-Pb-Ag at the CSA to Cu & Au at
Syn-deformational and syngenetic models Peak, Chesney, New Cobar, New Occidental
havebeenproposedforMountIsa-typedeposits and Great Cobar, is indicative for changes in
in the past, however the age relationship be- fluid temperature, pH and redox condition, or,
tweenoreformationandthehostlithologiesare assuggestedbyLawrieandHinman(1998),for
stilldebated. Withanagediscrepancyof28Ma, more than one metal source. Another differ-
leadmodelagestudiesattheMountIsadeposit ence is the occurrence of magnetite in some
demonstrated that the host lithology is signifi- copper-gold systems, entirely absent in zinc-
cantlyolderthenthesulphidemineralisationit- lead-copper dominated deposits. This feature
self. Amongstmany,thisisprobablyoneofthe indicate changes in the fluid chemistry dur-
most important arguments supporting a diage- ing ore formation, e.g. redox condition or the
netic/epigeneticorigin. amount of dissolved sulphur as described in
Large (1977). The current genetic model of
theEluradepositalsoreferringtotheotherCo-
2.4.3 Deposits in the Cobar region
bar deposits will be discussed in section 3.2 on
page 50 and therefore is not described within
Several polymetallic base metal, copper-gold
this chapter. Important orebody features of the
and gold deposits are found in the Cobar area.
deposits beginning with the southernmost Peak
Schmidt(1990)estimatedatotalof85Mtofore
golddepositaresummarisedinthefollowing.
is contained in the known deposits with metal
contentsintheorderof2.6Mtzinc,1.6Mtlead,
∗
1.0Mtcopper,4000tsilverand70tgold. Since The Peak Gold deposit
that time, resources have grown as a conse-
The Peak gold and minor base metal deposit is
quenceofnearmineexplorationanddelineation
located 8km south-east of Cobar (figure 2.11)
drilling (no published data). The orebodies of
and is hosted by the Chesney Formation, part
the Cobar deposits show significantly different
of the Nurri Group (Hinman and Scott, 1990).
features compared to the other four major de-
Thedepositformedinanarrowhighstrainzone
posit types described in the previous chapter.
of strongly deformed host sediments. Siliceous
Thedepositsthemselves,excludingtheMcKin-
volcanics of rhyolitic to rhyodacitic composi-
nons deposit located towards the western basin
tion occur at depth. These rock types are vari-
margin,however,haveseveralcharacteristicsin
ably pseudobrecciated, partially chloritised and
common (Lawrie and Hinman, 1998): (1) all
deposits are hosted in siliciclastic marine tur-
∗.Summarisedafter
bidites of the Cobar Supergroup; (2) ore for- Hinman, M.andScott, A.(1990).ThePeakgoldde-
mation is strongly structurally controlled (di- posit,Cobar.In:F.Hughes(ed.),Geologyofthemineral
depositsofAustraliaandPapuaNewGuinea,vol.2,Aus-
lation, competency contrast, fault jog); (3) the
tralasianInstituteofMiningandMetallurgy,Melbourne,
ore bodies are subvertically orientated and lo- pp.1345–1351and
Lawrie, K.andHinman, M.(1998).Cobar-stylepoly-
cated in steeply-dipping high strain zones near
metallic Au-Cu-Ag-Pb-Zn deposits. AGSO Journal of
the eastern Cobar basin margin; (4) geometries Australian Geology and Geophysics,17,169–188
31 |
ADE | 2. REGIONAL GEOLOGY
show flow banding (Hinman and Scott, 1990). and developed due to strike slip reactivation
The mineralisation occurs adjacent to the ear- of thrust faults. A halo of pyrite surrounds
lier described volcanics as veins, fracture fill- the zones of mineralisation. Mineral paragene-
ing or in a disseminated manner. The orebody ses such as magnetite-Au-maldonite-nativeBi-
consistsofseveralzonedpodsandlenses,char- bismuthinite,chalcopyrite-pyrrhotite-pyriteand
acterisedbyaCu-richinnerzonebecomingPb- galena-sphalerite-pyrrhotitearefoundinthede-
Zndominatedtowardstherim(LawrieandHin- posit. Early silicification, quartz veining and a
man,1998). Onlyminormineralisationisfound syn-mineralising chloritisation are described as
in the volcanics themselves. Apparent cack- mainalterationfeatures.
seal features indicate formation in a dilational The sulphide-poor New Occidental deposit
environment. Hinman and Scott (1990) de- consists of a disc-shape orebody containing
scribedfivestylesofmineralisation,comprising six discrete ore lenses which are characterised
three lead-zinc and two copper dominated ones by varying metal ratios. The richest gold
in the Peak deposit: (1) zones of strong sili- zones are found in the upper parts of the
cification and disseminated sphalerite-galena; mineralisation as Au-maldonite-Bi-magnetite
(2) folded and boudinaged quartz-sphalerite- ore. Other typical mineral assemblages are
galena veins; (3) black banded silver-rich min- chalcopyrite-pyrrhotite-pyrite and galena-Au-
eralisation characterised by a paragenesis of sphalerite-pyrite-pyrrhotite. Theorebodyisepi-
chlorite-sphalerite-galena, and partially replac- genetic in nature and hosted within shears de-
ing style 1; (4) breccia-hosted high level gold veloped at zones of rheology contrasts. Alter-
chalcopyrite-pyrrhotite ±pyrite mineralisation ation is expressed as silicification, quartz veins
at the contact between the Nurri Group and and breccias, colloform banded quartz chlorite,
volcanics; (5) late quartz-chalcopyrite-iron sul- carbonateveiningandchloritisation.
phidevein-stylemineralisation. Threeverticalpipe-likeorelodescontinuefor
more than 1000m, and four smaller lenses are
found at the Great Cobar deposit. The deposit
∗
Other deposits in the Cobar Goldfield
is epigenetic and hosted in shear-zones. Three
The Cobar Goldfield consists of another four ore types are distinguished: (1) magnetite-
mineraldepositsinadditiontothePeakdeposit pyrhotite-chalcopyrite, hosted by silicified
(figure 2.11). The New Occidental represents and chloritised slate; (2) quartz-chalcopyrite-
a gold deposit whereas the remaining three are pyrrhotite±magnetite and (3) lead-zinc miner-
gold-copper systems. The ore bodies are sub- alisationwithaccompanyingchloritisation. The
vertical orientated and steeply plunging to the siliceousoretype(2)mainlyoccursatdepth.
north.
The Chesney deposit consists of two gold- The CSA base metal deposit†
rich pipe-like bodies connected by a copper-
Thepolymetalliccopper-lead-zincCSAdeposit
rich vein-style zone, and a thin lenticular
is located 11km NNW of Cobar (figure 2.11).
body. The orebody is epigenetic and hosted by
shear zones. The mineral paragenesis consists
∗.Summarisedafter
ofchalcopyrite-pyrrhotite-magnetite-nativeAu- Lawrie, K.andHinman, M.(1998).Cobar-stylepoly-
Ag-nativeBi-bismuthiniteandminorsphalerite- metallic Au-Cu-Ag-Pb-Zn deposits. AGSO Journal of
Australian Geology and Geophysics,17,169–188
galena-pyrite. Alteration is characterised by
†.Summarisedafter
early quartz veining and silicification followed Scott, A.andPhillips, K.(1990).CSAcopper–lead–
bychloritisation. zinc deposit, Cobar. In: F. Hughes (ed.), Geology of the
mineral deposits of Australia and Papua New Guinea,
The New Cobar deposit is characterised by
vol. 2,Australasian Instituteof Miningand Metallurgy,
fairly consistent metal ratios with an overall Melbourne,pp.1337–1344and
Lawrie, K.andHinman, M.(1998).Cobar-stylepoly-
increase of gold content towards the surface.
metallic Au-Cu-Ag-Pb-Zn deposits. AGSO Journal of
The four pipe-like orebodies are shear-hosted Australian Geology and Geophysics,17,169–188
32 |
ADE | 2.4 Australian Zn-Pb-Ag ore systems
The orebody is hosted by the CSA Siltstone, alteration halo is dominated by Fe-rich chlo-
which is a thinly bedded turbiditic rock se- rite (up to 50m wide zone) followed by a rel-
quence part of the Amphitheatre Group (Scott atively narrow zone of sericite alteration (Scott
and Phillips, 1990). The host rocks experi- and Phillips, 1990). A weak to moderate per-
enced greenschist metamorphism accompanied vasive silicification surrounds the entire miner-
by the development of a pronounced cleav- alisingsystem. Theorebodyconsistsofseveral
age. A prominent chert layer, approximately sub-parallel, vertical to sub-vertical (˜70◦ east-
30m thick and representing a tuff layer, is ward dipping) massive and sub-massive podi-
present within the sequence. The CSA Silt- form lenses and zones of strong sulphide vein-
stone features abundant sedimentary structures ing. Thesebodiesarelaterallydiscontinous,but
e.g. graded beds, load casts, ripple marks, etc. continueverticallyforover2km. Thedepositis
A very strong cleavage, obliterating the bed- subdividedintozonesofmineralisationaccord-
ding, is developed in the immediate vicinity of ingtooccurringoretypesandchangesofmetal
the orebodies (Scott and Phillips, 1990). Sev- contentsandratios. Averticalmetalzonationis
eral chloritic shears, containing abundant black expressedbyanincreaseofcopperanddecrease
Mg-rich chlorite, are developed throughout the of lead and zinc with depth (Lawrie and Hin-
deposit. Some of these shears are host to sul- man, 1998). Laterally, ore lenses are enriched
phides and talc. The mineralised zones are ge- inCuincorezoneswhereasZnandPbconcen-
netically linked to these shears and occur im- trationsincreasetowardsperipheralzones,sim-
mediately above them in the hangingwall. The ilarilyasobservedathePeakdeposit.
Figure 2.11: Geological map of the eastern margin of the Cobar Basin showing the
locations of important mineral deposits; map data is from the 1:250 000 Australian
NationalGridCobarSH/55-14mapsheet(Brunker,1969).
33 |
ADE | Chapter 3
The Elura Orebody
THE ELURA DEPOSIT is the largest mineral 3.1 Geological
resource in the Cobar region. It is located characteristics∗
at the north-eastern margin of the Cobar Basin
in the vicinity of a basement protrusion (figure
3.1.1 Host lithology
3.1). Atpresent,themineralisingsystemisstill
openalongstrikeandtowardsdepthanditsfull The orebody is hosted by the CSA Siltstone, a
dimension is yet to be identified. The current turbiditicsedimentsequencewhichisamember
geological model is defined by approximately oftheEarlyDevonianAmphitheatreGroup(see
2000 diamond drill holes in addition to under- figure3.1andfigure2.10Schmidt,1990).
groundmapping. Theseturbiditesconsistofinterbeddedsand-
Glen (1990) divided the Cobar Basin in a , silt- and mudstone layers in a (A-B)-C-D-E
high and a low strain zone, with the Elura de- Boumasequenceandshowavarietyofsoftsed-
positlocatedinthelatter(figure3.1). Fourma- iment deformational structures such as convo-
jor faults define the basin block hosting the de- lute bedding, slumps, load casts or flame struc-
posit (Glen et al., 1994): the Little Tank Fault tures. Thesiltstonebedsarepalegreyincolour
to the north, the Bundella Fault to the south, whereas the mudstone layers are dark grey to
theMyrtFaulttotheeastandtheBuckwarroon black. The regularity of this interbedding and
Faulttothewest(figure3.1). the relative amount of sandstone in compari-
The geology of the deposit was first de- sontothefinergrainedmud-andsiltstonebeds
scribedbySchmidt(1980);AdamsandSchmidt changes significantly. The variation of average
(1980); Archibald (1983). Since then, fur- greywackecontentanditsmaximumbedthick-
ther geologists, inter alia Taylor et al. (1984); nesswascalculatedbasedon6877loggingdata
De Roo (1989a,b); Schmidt (1990); Seccombe points out of 155 diamond drill holes through-
(1990); Lawrie and Hinman (1998); Webster out the deposit in the course of this study ( fig-
and Lutherborrow (1998); David (2008), stud- ure 3.2, data supplied by CBH Resources Ltd).
ied the base metal system. Outcrops in the The host rock contains a median of 20% sand-
vicinity of the deposit are sparse, subsequently stonewiththicknessesofmostofthesandstone
mostofthegeologyisinterpretedbasedondia-
∗.Thefollowingsummaryonthegeologicalcharacter-
mond drill hole data and from underground ex- isticsoftheEluraOrebodyisbasedonobservationsmade
posure. bytheauthorduringemploymentasgeologistattheEn-
deavor Mine by CBH Resources Ltd between 2006 and
2008inadditiontocitedworkofotherauthors. Itneeds
to be stressed, that acquired geological data, e.g. map-
ping,drillcorelogs,etc.,andimprovementsmadetothe
geologicalmodelduringthestatedtimeperiodhastobe
understoodasateameffortoftheentireGeologydepart-
mentoftheEndeavorMineandisnotexclusivelybased
ontheauthor’swork.
35 |
ADE | 3.1 Geological characteristics
A NOTE:
Brill (1988) investigated crystallographic char-
This figure/table/image has been
acteristics of illite and white mica in order to
removed to comply with copyright
estimate pressure conditions during metamor-
regulations. It is included in the
phism. Results showed a decrease in meta-
print copy of the thesis held by the
morphic intensity from anchizonal to epizonal
University of Adelaide Library.
in the south near Cobar to anchizonal at the
Figure 3.3: Example of a tuff bed in the Elura’s
Elura mine. An average metamorphic pressure
hostrocksequence. TuffshowninimageistheT2
of3kbarwascalculatedbasedonsilicacontent horizon from drill hole DE566 at 375m down-hole
of white micas in samples taken near the CSA depth. ImagesuppliedbyCBHResourcesLtd.
mine. This pressure indicates a lithostatic load
of approximately 11km and would require the
limestone, composed of reef rudstone, bound-
Late Devonian to Early Carboniferous Mulga
stone, coquina and bioclastic packstone, ac-
Downs Group to overlie the Cobar Supergroup
cording to the classification scheme of Dun-
sedimentsinthisparticulararea(Brill,1988).
ham (1962). Leevers (2000) suggested the reef
The Elura limestone is the deepest known
formed above a basement high because silici-
lithology at the deposit. The limestone is com-
fied CSA had been intersected to the east and
monly developed as a clast-supported foliated
the west at greater depths than the top of the
brecciathat,inplaces,hadbeenaffectedbyduc-
limestone complex. Carolan (1999) interpreted
tile deformation. It is composed of crystalline
thereefsequenceasbelongingtotheBrookong
or recrystallised limestone, crinoid stems, shell
Formation.
andcoralfragments, aswellasvariablequanti-
A deep drilling exploration program during
tiesofshale. Thelimestonecommonlycontains
2006/07 was designed to test the contact of the
thinstylolitesandquartzveins.
limestone for economic mineralisation. Fur-
David (2000) interpreted the limestone as an
thermore, it was aimed to investigate the true
insitupartoftheshelfsedimentsoftheKopyje
thicknessofthelimestone,thenatureofthecon-
Groupanddescribeditasanopenplatformreef
tact to and the type of the underlying rock se-
10200
10100
RL 10000 m
10000
9900
RL 9800 m
9800
9700
RL 9600 m
9600
9500
RL 9400 m
9400
9300
RL 9200 m
9200
m m m m m
9100
0 0 0 0 0
0 0 0 0 0
7 9 1 3 5
6 6 7 7 7
N N N N N
Avg. greywack component [%] Avg. maximum greywacke
bed thickness [mm]
Figure 3.2: Sedimentolgical variations of the host rock sequence as a potential control of the mineralising
system. Variation of average greywacke content and its maximum bed thickness, calculated based on 6877
logging data points out of 155 diamond drill holes throughout the deposit, is plotted vs reduced level depth
(RL). Minima of both values coincide with the approximate occurrances of the upper and lower laminated
unitsasproposedbyNicholsonetal.(2006). DiamonddrillholeloggingdatasuppliedbyCBHResourcesLtd.
37
]m[
LR
eniM
51 02 52 03 05 001 051 002 052 003 |
ADE | 3. THE ELURA OREBODY
Figure 3.4: Mine stratigraphy of the Elura deposit constructed for the main lode area at 6775N (local
mine grid). The semi-massive resource domain (red shape) and the modelled Elura Limestone contact (blue
shape) are show on the right side (facing north) in order to visualise the location of the orebody relative
to the stratigraphy. The yellow dashed line shows the approximate northing (6775N) corresponding to the
stratigraphy. Thestratigraphiccolumnisbasedonthegeologicalresoucemodel,suppliedbyCBHResources
Ltd.
quence. Theoveralldrillholeinformationden- back-reeflithofacies.
sity within the limestone is rather limited and Two very regular interbedded unites are
neither of these recent, nor any previous dia- found within the CSA host rock, the so called
monddrillholesintersectingthelimestoneever upperandlowerlaminatedunits(ULUandLLU
reached the footwall contact. Despite the de- respectively),bothapproximately50minthick-
posit being located in the vicinity of the basin ness and characterised by rather sparse and
margin,theregionalcorrelationofthelimestone thinlybeddedsandstone(˜15wt%). Theseunits
to the Kopyje Group is uncertain and the ex- were first identified and described by Nichol-
act stratigraphic position within the Cobar Su- son et al. (2006), suggesting that theyprobably
pergroup unknown. The Elura limestone may truncate the lower sheet-like mineralisation on
as well represent an allochtonous mass several eithersideatadepthof˜400and˜900mbelow
kilometres distant to the actual reef complex as surface. Thesouthernmostandlargestsulphide
partofanolisthostrom. bodypenetratestheULU,howeveralateralcon-
An approximately 100m thick transitional traction of the pyrrhotite rich inner ore zone is
rock sequence beneath the turbiditic host rocks noticeable. Both, average greywacke contents
conformably overlies the limestone. The se- and its maximum bed thicknesses, feature min-
quence consists of interbedded fossiliferous, ima at the approximate depths of the proposed
calcareousmud-andsiltstonewithminorsandy ULUandLLU(figure3.2).
beds. David (2008) interpreted them as distal
38 |
ADE | 3.1 Geological characteristics
3.1.2 Ore types and their zoning
RL 10000 m
The overall geometry of the Elura orebody can
be described as sheet-like, characterised by a
ML
RL 9800 m
general NNW orientation and a sub-vertical
plunge (see figure 3.17). The mineralisation
stretcheslaterally850malongstrikeandvaries RL 9600 m
z1 z2 z3 z4 z5
in vertical extent from almost 1000m in the z6
southern parts in the so called main lode area,
RL 9400 m
to 400 to 500m in northern pods. The sheet
itself consists of six grossly concentric to el-
RL 9200 m
lipsoidal sub-vertical pipes containing massive
m m m m m
and semi-massive sulphide mineralisation (fig-
0 0 0 0 0
0 0 0 0 0
ure3.5). Theyarefromsouthtonorththemain 7 9 1 3 5
6 6 7 7 7
lode(ML)andthefivenorthernpods(z1toz5). N N N N N
These pipes are 120m wide in the ML and 30 Figure3.5: Thedifferentmineralisedzonesofthe
Elura Orebody. Longitudinal view towards WSW
to 50m wide in z1 to z5. A smaller and rather
(˜245°). Redshapeisthesemi-massiveore(SiPy)
narrowlenticularsulphidebody,thez6pod(˜15 resourcedomain.
by20mwideandverticalextentof˜40m),rep-
resents the currently know northernmost eco-
a lesser extent, quartz (Schmidt, 1990). Chlo-
nomicmineralisation.
rite(Fe:[Fe+Mg+Mn]=0.81-0.98)andsericite
The orebody is discordant to the CSA host
contentsarevariableandoverallratherlow. Mi-
rock, the exeption being the contacts of the
nor minerals are barite, albite and barian-alkali
uppermost caps of the mineralisation where
feldspars.
they appear to be concordant. A strongly de-
Quartz occurs mainly cherty in silicified
formed and sheared wall rock zone with in-
host rock fragments and as post mineralisation
creasedquartzandcarbonateveiningenvelopes
quartz veining. Accessory detrital zircon, tour-
the orebody. Minor occurrences of stratiform
maline, apatite and kerogen are described by
mineralisation are predominantly composed of
Schmidt(1990).
pyrite with rare base metal sulphides and are
ThedifferentoretypesoccurringattheElura
preferablyhostedinsandstonebeds.
depositwerefirstdescribedbySchmidt(1980).
In decreasing order of abundance, the sul-
He defined three ore types: (1) Siliceous ore,
phide mineral paragenesis consists of iron sul-
(2) massive ore and (3) pyrrhotite ore. Later,
phides (median of 55wt%, ratio of pyrite to
Webster and Lutherborrow (1998) refined the
pyrrhotite varies between ore types), spha-
ore classification and introduced three addi-
lerite (median of 16wt%), galena (median of
tional ore categories. Ore classification is now
6.5wt%),chalcopyrite(medianof0.6wt%)and
also based on an economic point of view, i.e.
arsenopyrite (median of 0.15wt%). Silver-
on the contained metal as lead-zinc combined
bearing mineral phases (e.g. freibergite,
grade (wt% Pb+Zn). Economic massive and
argento-tetrahedrite,nativesilver)occurprefer-
siliceousoretypescontainequaltoormorethan
ably in the uppermost and peripheral miner-
10wt% wt% Pb+Zn and represent the main re-
alised areas as minor constituent. The sulphide
source. Massive ore is abbreviated as Po or
mineralogy has been investigated throughout
Py for pyrrhotite and pyrite dominated types.
the orebody as part of this thesis in great detail
Semi-massive ore is abbreviated analogue as
andisdescribedinchapter4onpage63.
SiPoorSiPy. Thebrecciaveinandstringerstyle
The gangue mineralogy predominantly con-
mineralisation, abbreviated as VEIN, contains
sists of carbonates such as siderite and
between 3 and 10wt% Pb+Zn. Whether eco-
ankerite(Fe:[Fe+Mg+Mn]=0.76-0.98)and,to
nomic or not strongly depends on the current
39 |
ADE | 3. THE ELURA OREBODY
metal prices. Accordingly, the sub-economic
weakly mineralised zones contain less than A
3wt% Pb+Zn but either sphalerite or galena NOTE:
must be present in order to be classified as this This figure/table/image has been
oretype(abbrev. asMinA).
removed to comply with copyright
When comparing pyrrhotite to pyrite domi-
regulations. It is included in the
nated ore types the former is characterised by
print copy of the thesis held by the
slightlyelevatedconcentrationlevelsofZn,Cu,
University of Adelaide Library.
Co, and Se, whereas the latter is relatively en-
riched in As, Ag, Sb, Au, Hg, Mo, Sn and Tl
(describedinchapter5onpage149).
Thesixcurrentlydistinguishedoretypesused Figure 3.6: Hand specimen of massive/semi-
massive pyrrhotitic ore. Left image side: Chlori-
in mine geology at the Elura deposit are modi-
tisedandpartiallysilicifiedwallrockfragmentcon-
fied versions of those defined by Webster and tainedinsulphidegroundmass;typicalappearance
Lutherborrow(1998)andareoutlinedbelow: ofsemi-massiveore. Rightimageside: massivesul-
phideorecomposedofcloseto100%sulphides. Im-
agesuppliedbyCBHResourcesLtd.
Massive pyrrhotitic
mineralisation (Po)
3.11). Major parts of the upper main lode min-
The massive Po ore type contains in the order eralisation consist of the Py ore type also due
of65to95wt%sulphideswithsignificantcon- to the earlier described alteration of pyrrhotite
tent of pyrrhotite which, in places, may also tomarcasites. Quartzandsideritearethedomi-
represent the main iron sulphide phase (exam- nantganguephaseswithmedianconcentrations
ple shown in figure 3.6). The gangue miner- of4and12wt%,respectively.
alogy is dominated by siderite. Po commonly
occurs in the central core zones of the massive Siliceous pyrrhotitic (SiPo) &
sulphidepipes,howeverisnotrestrictedtothose siliceous pyritic (SiPy)
areas (figure 3.11). Schmidt (1990) argued, mineralisation
that retrograde alteration of primary hexago-
Siliceous ore types occur mostly but not ex-
nal pyrrhotite subsequent to regional metamor-
clusively at the margins of massive Po and
phism caused the formation of the magnetic
Py mineralisation (figure 3.11). They con-
monoclinic pyrrhotite variety. Prolonged ret-
tain highly silicified and variably chloritised
rogradealterationgeneratedabundantmarcasite
wall rock fragments accompanied by increased
in the upper and minor goethite in particular in
quartz as gangue phase and quartz veining at
peripheralzones. Silver-bearingmineralphases
variableintensities(figure3.6). Inplaces,these
do occur, but not as commonly as in pyrite and
remnant siltstone clasts are rounded but most
thesiliceousoretypes.
commonly are angular in shape and form brec-
ciaswithsulphides.
Massive pyritic
mineralisation (Py)
Breccia vein and stinger type
The main iron sulphide in the Py ore type is mineralisation (VEIN)
pyrite and/or marcasite. Pyrrhotite is gener-
The VEIN ore type envelopes the siliceous ore
allyabsentandonlyrarelyobservedasverymi-
and represents the interface to weakly miner-
nor constituent. The ore contains between 55
alised altered wall rock. The width is highly
and98wt%sulphidesandcommonlyoccursas
variable and rather narrow in places below one
anirregular,althoughconcentrichalosurround-
metre in the mid to upper parts of the orebody
ingthemassivepyrrhotitemineralisation(figure
butcanbeuptotensofmeterswide,particularly
40 |
ADE | 3.1 Geological characteristics
at depth (figure 3.11). The type of mineralisa-
tionischaracterisedbyreticulatesulphidevein-
ing, most commonly in a highly fractured and
silicifiedhostrock(figure3.7). Themineralas-
semblage consist of pyrite-pyrrhotite-sphalerite
with variable galena content, which may repre-
sent only a minor constituent. Significant chal- A
copyrite is associated with this ore type partic- NOTE:
ularly at greater depths. Visual discrimination This figure/table/image has been
between siliceous and vein style mineralisation removed to comply with copyright
can be difficult, hence an arbitrary boundary regulations. It is included in the
basedonacombinedPb+Zngradeisused. print copy of the thesis held by the
University of Adelaide Library.
Mineralised altered host rock (MinA)
MinA represents the outermost sub-economic
mineralised halo surrounding the orebody.
This ore type is predominantly developed in
the lower orebody and weaker or absent in
the upper main lode zone. There is associ-
ated siderite alteration and silicification at
Figure 3.7: Sulphide stringer in highly silicified
variable intensity. Sulphides, predominantly wall rock (VEIN type ore) in hand specimen. Sul-
phideveiniscomposedofpyrite,pyrrhotite,spha-
iron sulphides, sphalerite and minor galena
lerite, galena and subordinate chalcopyrite. Sand-
and chalcopyrite, occur either as thin sulphide stonebedsinwallrock(CSASiltstone)arestrongly
stringers or disseminations. Sandstone beds silicifiedasindicatedbythealmostchertyappear-
ance. Image width 12.5cm; sample location 290
are commonly the preferred host for sulphides
decline(290levelaccess). ImagesuppliedbyCBH
as they feature higher permeability compared ResourcesLtd.
to silt- and mudstone beds. In places, dissem-
inated sphalerite commonly replaces siderite
types with increasing mining depth, nowadays
spotting in mudstone rich sequences (figure
theyaretreatedsimilarlyandmodelledasasin-
3.8). The contact between MinA and VEIN is
gle domain. The transition from siliceous ore
gradational. A Pb+Zn combined grade is used
(SiPy and SiPo) via the VEIN breccia stringer
asdistinguishingfeature.
zone to the sub-economic minor mineralised
hostrock(MinA)isrelativelynarrow. Thetran-
sitionzonesurroundingthemassivemineralisa-
The zonation of the different ore types is
tion (Po and Py) can be as thin as 10 to 50cm
roughlyconcentricintheuppermainlodemin-
anduptoseveraltensofmetreswide,inpartic-
eralisation, however it becomes increasingly
ularat depth(figure3.7). DeRoo(1989b) sug-
complex and more irregular at depth. Figure
gestedthattheperipheralbrecciazones(mainly
3.11 shows the development of ore zonation in
VEIN and to a lesser extent SiPo and SiPy)
plan view at five different depth levels below
formedduetowallrockfragmentsbeingforced
surface. The contacts between massive Po and
orcollapsingintozonesofdilationandpressure
Py are commonly gradational. Po ore type oc-
relaxation.
cursalmostexclusivelyinthecorezonesofthe
The massive Py and Po core zones nar-
massivesulphidepipes. SiPoandSiPydomains
row towards north and with increasing depth
were modelled individually for the geological
(figure 3.11). By contrast, the width of the
and resource model in the past. Due to the
VEIN mineralisation surrounding the massive
increased geometrical complexity of those ore
pipes remains roughly constant with increas-
41 |
ADE | 3. THE ELURA OREBODY
A
NOTE:
This figure/table/image has been
removed to comply with copyright
regulations. It is included in the
print copy of the thesis held by the
University of Adelaide Library.
A
Figure 3.8: Sphaleriteinmudstonerichsequence NOTE:
preferably replacing siderite. Sample image from
This figure/table/image has been
drill hole NP915A; lower main lode area below
˜9260RL.ImagesuppliedbyCBHResourcesLtd. removed to comply with copyright
regulations. It is included in the
print copy of the thesis held by the
ing depth, connecting and enveloping individ-
University of Adelaide Library.
ual mineralised pods. At approximately 950m
below surfrace (9250mRL), the massive sul-
phide pipes contract further before they disap-
pear entirely (see plan view section 9300RL in
figure 3.11). However, VEIN ore continues to
greater depth hosted in hydraulic or hydrother-
mal breccias (figure 3.9)and extensive reticu-
lar vein systems in highly silicified CSA silt-
stone (figure 3.7). This change in mineralisa-
tion style is accompanied by a change in ore Figure3.9: Hydrothermalbrecciawithminorsul-
phides (MinA ore type) in strongly silicified host
mineralogy characterised by a slight increase
rock. Image width 3cm; sample image from drill
in quantities of chalcopyrite, the preferred oc- holeNP915A;lowermainlodeareabelow˜9260RL.
currence of pyrrhotite over pyrite and elevated ImagesuppliedbyCBHResourcesLtd.
iron content in sphalerite (visually identified –
darker in colour). Increased ore hardness due
veloped in Po and least in siliceous ore type.
to this strong silicification (figure 3.9) and the
Inperipheralorezones, itiscommonlycharac-
greater depth of this mineralised zone causes
terisedbyaconcentricorientationrelativetothe
significantly higher production costs. Hence,
orebody. A sub-horizontal rather fine sulphide
cut-off grades strongly depend on the current
layeringwasdescribedbySchmidt(1990). This
metal prices. Remnant angular chloritised wall
bandingisexpressedbyquartz-sideritefracture
rock fragments in semi-massive ore as well as
filling in pyrite-rich bands. These dilational
themineralisedbrecciaveinzonesindicatethat
structurespost-datethesulphidemineralisation.
fracturingwasamajormechanismforthedevel-
De Roo (1989b) linked the sulphide banding to
opment of zones of increased permeability act-
the three deformational events (D -D ; see ta-
2 4
ing as fluid pathways and as space for ore pre-
ble 3.1 and chapter 3 on page 45) causing in-
cipitation(figure3.9).
homogeneousintensefoliationoftheore. Sim-
Sub-vertical sulphide banding, defined by
ilarily, Schmidt (1990) explained the composi-
layers enriched in pyrite, pyrrhotite, galena or
tional layering by a cleavage formation event,
sphalerite, is present throughout the orebody.
either contemporaneously to ore formation or
The banding is found in all ore types, best de-
42 |
ADE | 3. THE ELURA OREBODY
overprintingslightlyoldermineralisation. andLutherborrow,1998). Sideritespottingalso
Webster and Lutherborrow (1998) described occurs adjacent to sulphide veins as shown in
three types of banding: (a) Discrete sphalerite figure3.12.
bandsatmillimetre-scalealternatingwithbands The size of siderite spots (<1mm-10mm)
enrichedinpyriteandpyrrhotite;(b)smearsand increases towards the deposit and they occur
streaksofsphaleritenotascontinuousandcon- preferentially in sandstone beds. The extent of
sistent as type (a) banding; and (c) intense, de- silicification between and enveloping the mas-
formational induced cleavage-style banding at sive sulphide pipes varies in width and inten-
the margins of the orebody. Bands are defined sity. However, it increases significantly below
by coarse-grained galena-forming veinlets ori- a depth of approximately 900m from surface.
entedparalleltothecleavageinwallrocks. Silicificationcanbeentirelyabsentintheupper
partsoftheorebody. Thezoneofintensesilici-
3.1.3 Wall rock alteration ficationatdepthcorrelateswiththechangeover
Chlorite, sericite, strong silicification and car-
bonate alteration in the form of pervasive
sideritespottingarethemostcommonalteration
features. Siderite alteration occurs predomi-
nantly on the eastern side of the deposit as a
halo up to a distance of 50m surrounding the
mineralisation. On the western side, the halo
ofsideritespottingisnotaspronouncedand,in
places, only extends for a few metres (Webster
A
NOTE:
This figure/table/image has been
removed to comply with copyright
A
regulations. It is included in the
NOTE:
print copy of the thesis held by the
This figure/table/image has been
University of Adelaide Library.
removed to comply with copyright
regulations. It is included in the
print copy of the thesis held by the
University of Adelaide Library.
Figure 3.12: Siderite spotting adjacent to sul-
phidevein. The sulphideveiniszoned, definedby
a marginal zone enriched in sphalerite, pyrite and
Figure 3.13: Pyritenodulesinstronglydeformed
sparrysiderite,andamassivegalenainternalzona.
zone of intense quartz-siderite veining. Pressure
Coarsegalenagrainsizesarepotentiallycausedby
shaddow, composed of quartz, is developed on
deformation-induced recrystallisation. Sandstone-
the left side of the pyrite concretion. Sandstone-
richfragmentsareintensilysilicified(lowerleftcor-
rich sequences are intensily silicified (lower image
ner). SampleimagefromdrillholeNP915A;lower
part). Image width is 3cm, sample from drill hole
mainlodeareabelow˜9260RL.Imagesuppliedby
NP915A;lowermainlodeareabelow˜9260RL.Im-
CBHResourcesLtd.
agesuppliedbyCBHResourcesLtd.
44 |
ADE | 3.1 Geological characteristics
from massive pipe-like to a vein/stringer type hydrothermalwhitemica(sericite),chloriteand
mineralisation. Strong pervasive silicification carbonates (siderite, ankerite and calcite) up to
altered the CSA siltstone to an almost cherty 80m distant to the orebody. Albitisation was
rock (figure 3.9 and figure 3.7). Webster and not detected via this analytical technique, how-
Lutherborrow (1998) identified a close spatial ever,accompanyingXRDanalysesshowedthat
relationship between the density of sulphide sodium alteration is rather weak and predomi-
veiningandtheintensityofsilicification. nantlyoccurredduringregionalmetamorphism.
CSA siltstone contains nodules of pyrite and Whitbread and Moore (2004) tried to iden-
minoreuhedralpyrite(figure3.13),moreabun- tify"vectorstoore"causedbyalterationviathe
dant in sandstone relative to silt- and mud- application of Pearce element ratio and isocon
stone beds. The pyrite content increases to- analysis and identified several element trends
wards the orebody indicating a genetic link characterisedbyconcentrationincreasetowards
(Schmidt, 1990). Sulphide replacement of the orebody: (1) elements known to form com-
wallrockoccurredalongmorepermeablesand- plexes with sulphur, e.g. Sb, Ag, As, Zn, Cu,
stonebedsseveralmetresdistantfromthemas- Pb, Ni, Cd and possibly Ba, (2) alteration of il-
sive sulphide body indicating migration of the litetomuscoviteisreflectedbyelementsincor-
metal-bearing fluid outwards into the country porated in muscovite, e.g. K, Rb, Cs and pos-
rock. Schmidt (1990) described mineralised sibly Ba, (3) siderite and ankerite contents in-
sandstone beds, containing a sulphide assem- crease towards the orebody subsequently caus-
blage of pyrite-sphalerite-galena-chalcopyrite- ingelevatedconcentrationsofCO andMn,(4)
2
tetrahedrite as far as 150m distant to the ore- increasesofLaandCetowardstheorebodyin-
body. dicate association to either carbonates or sheet
Intense chloritisation, ankeritisation, albiti- silicates,and(5)CaandSrfollowtheincreasing
sation and sericitisation in the innermost 20- trend,howeveraredepletedimmediatelyproxi-
25mwidealterationhalohasbeendescribedby mal to the orebody due to pronounced replace-
Schmidt (1990). This zone is characterised by mentofCaandSrbyFe.
elevatedconcentrationsofS,As,Ba,Hg,Pb,Sb
andZnwhereasMgO,CaOandMnaredepleted 3.1.4 Structural setting
(Tayloretal.,1984). FreshCSAsiltstoneprox-
The Elura orebody is highly discordant to the
imaltotheorebodyshowssimilarchemicalsig-
host rock and truncated, as well as intersected
natures compared to remote host rock samples
byseveralNNWandNNEtrendingsub-vertical
with the exception of elevated Ba and depleted
faults(figure3.15). Thesefaultsdefineanorth-
Mn.
north-westerly oriented transpressional struc-
A cryptic alteration halo was described by
turalhighstraincorridor,characterisedbyverti-
Schmidt (1990) featuring an increased carbon-
calandanoveralldextralstrike-slipmovement.
ate content and K O:Al O ratio accompanied
2 2 3
The corridor narrows in a NNW direction. The
by losses of Na O. He explained the chemical
2
most prominent of these faults is the "Western
modificationwiththedestructionofchlorite,al-
Shear" as the western truncation of the south-
bite and detrital components e.g. biotite and
ern parts of the orebody. This fault crosses the
iron-titanium oxides caused by acidic carbon
sheetofmineralisation,intersectingtheorebody
dioxide-rich fluids. Iron and Mg released dur-
between the z1 and z2 ore zones. Very poor
ing alteration led to the replacement of calcite
ground conditions along this structure caused
byankeriteandsiderite.
significant geotechnical issues and subsequent
Sun et al. (2001) used short-wave infrared
production losses in the past. The host rock
reflectance spectra (SWIR), obtained via a
is strongly folded and faulted in the vicinity of
portable infrared mineral analyser (PIMA), in
the orebody up to a distance of 30m (Schmidt,
order to investigate the alteration halo around
1990). Theintensityofdeformationrapidlyde-
the Elura orebody. In this study she identified
45 |
ADE | 3. THE ELURA OREBODY
Figure 3.15: Major structures
truncating, intersecting and partly
offsetting the Elura deposit; sec-
tionat9500RLdepthinplanview
(downwards). The spatial extent
oforetypesandlocationsoffaults
were modified from the geological
and resource model, supplied by
CBH Resource Ltd. Faults are
basedandmodelledfromrockqual-
ity designation (RQD) values from
diamonddrillholedata.
creases away from the sulphide body, charac- tionalsites,alongbendingfaults,ortheyreflect
terised by a much more open folding. Prox- rheological and competency contrasts between
imal to the orebody, the NNW-trending folds thesulphidesandthesiliciclastichostlithology,
form rather tight anticlines and synclines ac- whichis,inplaces,stronglysilicified.
companied, in places, by pronounced quartz De Roo (1989b) studied the southernmost
veining. S defined by De Roo (1989b) (table main lode zone (the northern ore bodies were
2
3.1)representsthedominantcleavageandisax- unknownatthattime)anddescribedfourdefor-
ial planar to those folds. According to Leevers mations (D to D ) and their associated struc-
1 4
(2000), folds in the host rock are symmetric on tural elements (table 3.1). He suggested that
theeasternsideoftheorebodyandasymmetric themainore-formingeventoccurredduringD
2
to the west with westerly-dipping fold hinges. with contemporaneous establishment of the al-
The NNE trending faults intersect the NNW- teration halo. The two sulphide rock apophy-
striking fault corridor and partly offset the ore- ses, developed in the upper main lode zone,
body (figure 3.15). These faults have similar are hosted in core zones of doubly-plunging
strike orientation as and are likely sympathetic anticlines. Deformation of the massive sul-
to the Buckwaroon fault approximately 2km phide body along ductile shears caused verti-
distant (figure 3.1), interpreted as reactivated cal stretching into its present shape accompa-
transferfault(Glenetal.,1996). Strongsilicifi- niedbythedevelopmentofsub-horizontaldila-
cationandquartzveiningoccuralongfaultsthat tionalstructures. Glen(1990)suggestedthatthe
truncate or offset the sulphide body. The ore- Elura mine is localised in an area where early
bodycomprisespinch-and-swellandboudinage WNW-orientated folds (F ) were refolded by
1
structures which are either developed in dila- NE-strikingfolds(F )causingtherotationofF
2 1
46 |
ADE | 3.1 Geological characteristics
intotheapparentNNWtrend. Hesuggestedthe wavesgeneratedduetolateralmovementalong
doubly plunging F folds identified by De Roo reversalornormalfaults.
2
(1989b)correspondtotheseF folds.
1
AccordingtostructuralstudiesdonebyWeb- 3.1.5 Regolith expression
ster and Lutherborrow (1998) the view that the
Surficial weathering
orebody is hosted within domes of a single
doubly-plunging anticline is far too simplistic.
The Cobar region is now an elevated palaeo-
Theysuggestthattheanticlinemayrepresentan
plain that has been exposed to surficial condi-
arrayofenechelonanticlines.
tions since at least Mesozoic times (McQueen,
David (2008) interpreted the anticline as
2004). Climatic cycles during the Cainozoic,
a fault-propagation fold developed within the
defined by warm humid conditions, were fol-
NNW-orientedfaultcorridorduetoreactivation
lowed by cooler more arid episodes that cre-
of faults during basin inversion. Furthermore
ateddeepchemicalweathering. Twomajorand
hesuggestedthreedifferentstraindomains: (1)
widespread periods of iron fixation occurred
Negative dilation zone with volume loss within
during the Early Palaeocene and the Middle
thelimestone,partiallyextendingintotheover-
Miocene and were identified based on palaeo-
laying clastic shelf sediments. The domain is
magneticdatingofferruginisedsaprolites(Mc-
characterised by pure shear and ductile defor-
Queen,2004).
mation. The E-W shortening based on stretch-
Three main soil types are found within the
ing lineations is estimated at approximately -
greater region around the Elura mine (Gibson
60%. (2) The neutral strain domain is inter-
et al., 2003). A loamy soil, 30 to 60cm thick
preted to occur at the lowermost extents of the
and red to brown in colour, represents the most
main sulphide mineralisation coinciding with
abundant type and formed from the underlying
the zone of pronounced breccia stringer style
saprolite in addition to a potential aeolian in-
mineralisation. Both compressional and exten-
put. Rare thin stony skeletal soils are only de-
sionaltexturesaredescribedand(3)Mostofthe
velopedabovethescarceremnantsofMesozoic
mineralisationformedwithinazoneofpositive
sedimentarycover. Clay-richsoil,grey,redand
dilation accompanied by volume increase. It
brownincolourandpredominantlyfoundonto-
hasbeensuggestedthatconical-shapedfracture
pographic highs, has been described by Gibson
zones developed as a consequence of shock-
A
NOTE:
This figure/table/image has been removed
to comply with copyright regulations.
It is included in the print copy of the thesis
held by the University of Adelaide Library.
Table 3.1: Summary of deformational events and their structural elements, either affecting
thehostlithologyoftheorebodyitself(DeRoo,1989b).
47 |
ADE | 3. THE ELURA OREBODY
etal.(2003)asthethirdsoiltype. orebody to depths greater than 800m (Reed,
AttheEluramine,upto1.5to2.0mthickred 2004). Taylor et al. (1984) distinguished two
top soil covers the weathered bed rock (Reed, typesofironstonegossan: (1)Insitudirectgos-
2004; Lorrigan, 2005). This surface layer con- san where the texture of the iron oxides devel-
tains abundant pisoliths of maghemite-bearing oped varies from sponge-like to massive and
iron-rich lag. The host rock is completely ox- (2)Solution-depositedgossanwhichoriginated
idised to a depth of approximately 80m below duetoprecipitationofironoxidesfromfluidsin
surface, coinciding with the depth of the cur- open space e.g. open joints or cavities. Acidic
rent water table (Taylor et al., 1984). Irreg- iron-rich fluids generated by the weathering of
ular partial weathering is controlled by struc- theorebodypartiallyinfiltratedthesurrounding
tures and fracture zones and extents to greater host rock and lead to local ferruginisation. The
depths up to 135m. The weathering formed contactbetweengossanandalteredhostrockis
bleachedsaprolite,whitetoyellowishincolour, verysharp. Themineralassemblageconsistsof
which can contain significant calcrete (Lorri- goethiteandhematiteasmajor,quartz,kaolinite
gan,2005). andbeudantite(PbFe+3(AsO )(SO )(OH) )
3 4 4 6
The saprolite is composed of quartz, mus- as minor constituents. The latter mineral phase
covite,K-feldspar,secondarykaolinite,goethite is responsible for elevated concentration levels
and hematite. Kaolinite and alunite are found ofPbandAsinthegossan(Tayloretal.,1984).
marginal to the orebody (5 to 10m) at even A supergene enrichment zone of six metre
greater depths than complete host rock oxida- thicknessdevelopedbetweenthegossanandthe
tion. Elemental leaching caused by weather- massive sulphide body (Taylor et al., 1984).
ing is reflected in pronounced losses of MgO, The supergene zone is composed of an equally
CaO, P O , Mn and Zn, and to a lesser extent thicksulphidezoneaboveprimarysulphideore
2 5
ofFe O ,K O,HgandCu(Tayloretal.,1984). and an overlying oxide zone. Later studies
2 3 2
Theselosseswereaccompaniedbyenrichments showed that the supergene oxide zone is with
ofSiO ,Al O ,BaandPbrelativetounweath- up to 15m significantly thicker than previously
2 2 3
ered CSA siltstone. Iron oxides commonly oc- thought(Scott,2003).
cur within the saprolite and are developed near The supergene sulphide zone comprises pri-
fracture zones or along bedding planes. They mary and secondary pyrite, galena, chalcopy-
form thin Liesegang or thicker bands and are rite, marcasite, digenite, bornite, anglesite and
characterised by elevated concentrations of As, beudantite (Taylor et al., 1984) and is char-
Bi, Cu, Pb, Zn, Co, Ni and Mn (Taylor et al., acterised by a vertical zonation form Pb-rich
1984). to Cu-rich at its bottom (Scott, 1994). The
occurrence of covellite is described by Scott
The Elura gossan and supergene min- (1994) and cerussite by Scott (2003). A de-
eralisation tailed study of the supergene mineralogy by
Leverett et al. (2005) unearthed the true com-
The gossan was initially exposed on the sur-
plexityofthemineralassemblageandidentified
face(Schmidt,1980)andwasburiedasaconse-
at least further 20 phases. The supergene sul-
quence of strong ground subsidence caused by
phide zone is enriched in Cu, Ag, Tl, Bi, Pb,
stopecollapsesin1996. Tayloretal.(1984)was
Hg and As. All element concentrations, except
thefirsttostudythegossanandsupergenemin-
As, are highest at the base of the layer and de-
eralisation in greater detail. The oxidation of
crease upwards (Taylor et al., 1984). Ba, Cd,
the orebody formed a dark brown to red gos-
Mo, Sn and Zn are generally depleted. Oc-
san to a depth of approximately 100m. To-
currences of rare oxychloride mineral phases
tal and partial oxidation occurred to a depth of
have been reported by Scott (1994). He iden-
80m and 100m, respectively. However, sul-
tifiedblixite(Pb O (OH) C )andmendipite
8 5 2 14
phide oxidation is found in cavities within the
(Pb Cl O ), mostly replacing anglesite and to
3 2 2
48 |
ADE | 3.1 Geological characteristics
a lesser extent cerussite. The relatively narrow deposit. Further lead is to some extent incor-
sulphide zone suggests a rather stable water ta- porated in resistant arsenates (Leverett et al.,
bleoveraprolongedtime(Tayloretal.,1984). 2005). The copper content is rather low and is
A distinct band, composed of "blue-black mostly associated with soluble mineral phases
sooty chalcosite", approximately 15cm in (e.g. malachite). Nostablezincmineralphases
thickness, is developed at the transition to the formed.
overlyingoxidisedzone(Scott,2003;Lorrigan, The surface expression of the deposit is gen-
2005). Minor digenite and enargite are also erally leached of many elements, except silica
presentinthisband. which is enriched. Taylor et al. (1984) iden-
The overlying oxidised zone consists of sul- tified As, Bi, Hg, Pb, and Sb as being impor-
phates, carbonates and arsenates. The zone tant near surface enrichments up to a depth of
is predominantly composed of several sec- 6m. However,thesesurfaceanomaliesareonly
ondary lead mineral phases, e.g. beudantite, developed above the southern main lode zone.
cerussite, mimetite (Pb (AsO ) Cl), nadorite No geochemical surface expression has been
5 4 3
(PbSbO Cl), lanarkite (Pb (SO )O), and to observed for the deeper northern pods, located
2 2 4
a lesser extent of goethite, quartz and barite approximately 400m below surface. These
(Taylor et al., 1984). Scott (2003) and Lor- anomalies are strongest developed around 50
rigan (2005) described a lowermost zone pre- to 100m distant to the orebody. Lead and ar-
dominantlycomposedofnativesilverandcassi- senic represent the most important anomalies
teriteinadditiontotheoccurrenceofhidalgoite for outlining the mineralisation. The more mo-
(PbAl (AsO )(SO )(OH) ). Thispartofthe bileelementscopperandzincshowmuchmore
3 4 4 6
supergene zone is characterised by enrichment diffuse and dispersed low-contrast anomalies.
ofPb,Ag,As,Ba,Hg,Mo,SbandSn. The Zn soil anomaly can be followed for more
than 1km to the south-west (Lorrigan, 2005).
Near surface geochemical anomalies Lead is preferably concentrated in coarser soil
fractions whereas zinc is anomalous in finer
The metal dispersion outwards from the ore-
fractions <63µm (Scott and Lorrigan, 2009).
body is a function of the stability of min-
Zinc concentrations within the weathered pro-
eral phases that interacted with surficial fluids.
file increase with depth towards the water table
RatherstablesecondaryPbmineralphases(e.g.
whereas copper and arsenic remain relatively
beudantite, mimetite) are common at the Elura
constant(Tayloretal.,1984).
49 |
ADE | 3.2 The current genetic model for the Elura deposit
the data and conclusions presented by De Roo David (2008) proposed a genetic model that
(1989b,a) he preferred the former. Ore forma- impliesanearlymineralisationinsemi-lithified
tionasreplacementwasinducedbyfluidtowall sediments during basin formation followed by
rock disequilibrium when the ascending metal deformation and modification during basin in-
bearing brine reached a certain level within the version. The author argued, that an epigenetic
stratigraphic column. Temporal and/or spa- origin, as proposed by several authors (inter
tial variability in fluid characteristics led to alia DeRoo(1989b,a);Schmidt(1990);Lawrie
the development of the pronounced zonation and Hinman (1998); Stegman (2001), is un-
of the orebody and alteration halo. He sug- likely due to: (a) the lithified nature of the host
gested furthermore that the different ore types rock sequence characterised by low permeabil-
at the Elura deposit had formed contempora- ity, and (b) seismic pumping through lithified
neous by replacement, progressively advanc- rock would create large zones of hydrothermal
ing upwards within the mineralising system. brecciation,whichdonotexistatElura.
Thepyrrhotite-dominatedcorezonesreflectad-
vanced replacement stages whereas peripheral 3.2.3 Age, metal source and phys-
siliceous ore represents are remnant early and iochemical fluid characteris-
immaturestage. tics
WebsterandLutherborrow(1998)scrutinised
Radiogenic isotopes
thetheoryofasingleanticlinehostingtheElura
orebody postulated by De Roo (1989b). Based
The age of the Cobar Supergroup sediments,
on structural data obtained from oriented core
the host rocks for all of the Cobar deposits, is
intersecting the NNW-striking fold axis the au-
poorlyconstrainedtothelatestSiluriantoEarly
thors concluded, that the simplistic view of the
Devonian based on paleontological chronology
northern pods being hosted in domes within a (Stegman, 2001). On the basis of 40Ar-39Ar
doubly plunging anticline is difficult to con-
whole rock dating Glen et al. (1992) estimated
firm. Apartfromanarrowzoneofintensedefor-
the peak cleavage formation event during basin
mation and cleavage at the immediate orebody
inversionhadoccurredbetween395-400Ma.
margin,theyobservedratheruniformstructural
Pb-Pb isotopic data of sulphides available
settings for the orebody as well as for the sur-
in literature (Pogson and Hilyard, 1981; Gul-
rounding country rock. Upon this observation
son, 1984; Carr et al., 1995; Sun, 2000; David,
the authors challenged the syn-deformational
2005), is very similar for all mineral deposits
replacement origin and suggested the possibil-
in the Cobar Basin. Due to their minor lead
ityoforeformationpriortodeformation. They
signature variations, Carr et al. (1995) pro-
suggested that competency and rheology con-
posed an isotopically uniform metal source or
trasts between host rocks and sulphides fo-
magmatic and hydrothermal processes causing
cussed strain and movement during later defor-
homogenisation (Stegman, 2001). A recent
mation represented in the earlier described in-
study by Mernagh (2007) confirmed previous
tensely deformed marginal zone. Furthermore,
Pb isotope data. Isotopic ratios compiled from
the authors speculated that stratigraphic deeper
several sources range between 18.108-18.157
sequences of CSA siltstone represent sources for206Pb/204Pb,15.614-15.672for207Pb/204Pb
for base metals whereas gold and copper min- and 38.232-38.450 for 208Pb/204Pb (David,
eralisation in the Cobar deposits is linked to
2008). Two lead sources are indicated by
strongly faulted basin margins. The more dis-
leadisotopesignaturesaccordingtoLawrieand
tantEluradepositcomparedtotheothercopper
Hinman (1998), a basement and a basin com-
andgolddominateddeposits,thereforecontains
ponent. Compared to Au-Cu deposits, those
onlyveryminorlevelsofcopperandalmostno
which are Pb-Zn dominated are characterised
gold(WebsterandLutherborrow,1998).
byaslightlymoreradiogenicleadsignature,in-
terpreted to reflect a predominance of a basi-
51 |
ADE | 3. THE ELURA OREBODY
nal metal source. A basement source is pro- mated at 440Ma to 415Ma (Pogson and Hil-
posed for Au-Cu deposits. Jiang et al. (2000) yard, 1981). Mernagh (2007) suggested that
speculatedthataninitialleadaccumulationhad 206Pb/204Pb signatures indicate an interaction
been facilitated by older granitic intrusion into with the Cambro-Ordovician sediments of the
the basement and was later remobilised by hy- GirilamboneGroup.
drothermal systems and released into the basin Thestrontiumisotopesignatureofsphalerite,
sequence. Sun (2000); Sun and Seccombe siderite,ankeriteandcalcitewasinvestigatedby
(2000) suggested a ternary mixing scenario for Sun (2000). Sphalerite and siderite, as mineral
ore lead sourced from I- and S-type granites as phases representing the ore-forming fluid, con-
well as from the Devonian Cobar Supergroup tain minor Sr (2.5-45.4ppm) at high 87Sr/86Sr
sediments. The difference in lead signature ratios (0.75501-0.83947). The highest Sr con-
could also reflect an increasing age of the Co- tent was observed in calcite from veins and
bar deposits from North to South as suggested limestone (260.3-550.9ppm) with substantially
byStegman(2001). lower87Sr/86Srratios(0.70936-0.72547). Vari-
David (2005) modelled a plumbo- ations of Sr concentration and 87Sr/86Sr ratios
chronological formation age of 411Ma for inalterationcarbonateswereinterpretedtorep-
the Elura deposit based on the Lachlan Orogen resent mixing of a hydrothermal and a marine
growth curve from Carr et al. (1995). David sedimentarystrontiumsource.
(2008) noticed significant differences in lead Sun (2000); Sun et al. (2000) dated sericite
isotope signatures (see figure 14, page 440 in contained in massive sulphide samples within
David,2008): (a)thelowermainmineralisation the Elura orebody via 40Ar-39Ar and identi-
(below level 6) is characterised by a rather fiedtwoeventsat385-389Maand376-379Ma.
uniform isotopic composition, (b) only one The former age resulted from cleavage-parallel
sulphidesamplefromthelimestonemineralisa- sericite and was interpreted as reflecting a ma-
tion as the lowest part of the Elura system was jor deformational event, challenging the basin
analysed. This sample is significantly more ra- inversion age suggested by Glen et al. (1992);
diogenic, expressedbyanelevated208Pb/204Pb Perkins et al. (1994). The younger age range
ratio whereas the 207Pb/204Pb and 206Pb/204Pb wassuggestedtorepresentthehydrothermalore
rations are the lowest values observed, and (c) formingevent,subsequenttopeakmetamorphic
samples from the upper main mineralisation conditionduringcontinuingbasininversion.
(abovelevel6)featurealineartrendoriginating
neartheclusterdefinedby(a)andadecreasein Stable isotopes
allthreeisotoperatiosapproaching207Pb/204Pb
and 206Pb/204Pb ratios of (b). The author Sulphur isotope studies on sulphide miner-
als form the Elura deposit have previously
explained these differences via variations in
been performed by Sun (1983); Seccombe
lead source mixing and concluded that initially
(1990); Sun (2000); David (2005); Mer-
theupperpartofthemineralisationwasformed
nagh (2007). The observed δ34S enrich-
via an undiluted mantle-derived fluid enriched
ment sequence, i.e. pyrite>sphalerite≈
in metals. Subsequently the lower part of the
pyrrhotite>chalcopyrite>galena, indicates
mineralisation formed from a mixed fluid,
precipitation conditions near equilibrium
characterised by a basinal fluid component and
(Ohmoto and Goldhaber, 1997). Those sul-
theinitialfluid.
phide phases yielded δ34S values between 4.7
Based on recent data from the Elura Mine,
and 12.6 with a mean of 8.1 (Seccombe,
Mernagh (2007) calculated a model age rang-
ing between 440 Ma and 410 Ma years and
1990). Se(cid:104)ccombe (1990) calculat(cid:104)ed an average
isotopic ore formation temperature of 275°C
identified at least four Pb isotope fractiona-
based on sphalerite-galena pairs, and inter-
tion trends. These ages coincides with gran-
preted this temperature as a minimum value
ite intrusion ages within the basement, esti-
52 |
ADE | 3.2 The current genetic model for the Elura deposit
due to potential isotopic re-equilibration below sions hosted by quartz and sphalerite by Sun
peak metamorphic conditions. Isotope data (2000);Jiangetal.(2000). Chloritesamplesas-
suggests a mixed sulphur source from early sociated with early pre-base metal mineralisa-
Devonian Cobar Supergroup sediments and tion exhibits isotopic signatures of a metamor-
underlying Ordovician metasediments of the phic fluid. The fluid linked to main base metal
GirilamboneGroup(Seccombe,1990). formationischaracterisedbyaclearshiftiniso-
Sun (2000) observed similar δ34S values for topic composition towards lower δ18O values.
ore sulphides, but in a slightly narrower range Sun (2000); Sun and Seccombe (2000); Jiang
(4.7to12.6 ,meanof8.1 ). Distinctlyheav- etal.(2000)concludedthat: (a)theoreforming
iersulphuri(cid:104)sotopesignature(cid:104)swereidentifiedfor fluidisofmeteoricoriginthatcirculateddeeply
syngenetic disseminated and framboidal pyrite with in basin and basement, thereby modify-
containedintheCSAhostrockrangingbetween ing its initial signature, (b) the isotopic data
9.5 and 20.1 (mean of 13.9 ). Sun (2000) indicates limited fluid-rock interactions at high
suggestedthe(cid:104)seδ34Svaluesrefl(cid:104)ectseawatersul- fluid fluxing rates without any magmatic con-
phatereductionassulphursource. Isotopictem- tribution to the hydrothermal system, and (c)
peratures calculated based on sphalerite-galena thedatashowedthattwocontrastingfluidswere
and pyrite-pyrrhotite pairs range between 220 presentatdifferentparageneticstagesandthere-
and 449°C. She concluded that the sulphur for foretimes,henceprecludingafluidmixingsce-
ore formation was sourced from host rocks and narioasamechanismfororeprecipitation.
a deeper crustal source. Stratigraphic horizons
within the sedimentary basin fill that contain
Fluid inclusion studies
syngenetic and diagenetic pyrite, in places, up
to significant quantaties, has been observed at
Seccombe (1990) was the first to study fluid
several localities (Stegman, 2001). Therefore,
inclusion hosted in quartz at the Elura de-
locally-derived sulphur seems highly likely as
posit. Inclusions had been visually recognised
importantsourceduringoreformation.
by Schmidt (1980). Analysed quartz grains
The significant spread of sulphur isotopes in
were either intergrown with sulphide minerals
addition to relatively high values at Elura com-
or sourced from quartz veins in the wall rock.
paredtootherCobardepositswhicharecharac-
Two-phase, liquid-rich primary fluid inclusions
terised by rather low and uniform values, may
of H O-CO -CH type were identified charac-
2 2 4
indicate a significant proportion of sulphur be-
terisedbyhigherCH /CO ratiosinoresamples
4 2
ing sourced from syngenetic/diagenetic pyrite
indicatingreducingconditionduringoreforma-
fromthehostrocks. JiangandSeccombe(2000)
tion. Neither N nor H S were detected. The
2 2
observedincreasesofδ34Sandtemperatureval-
uncorrected homogenisation temperature (T )
h
ues in the early sulphide paragenesis before
of fluid inclusions in massive and siliceous ore
dropping back to initial values and suggested
samples yielded two populations ranging be-
these variations are responsible for the spread
tween 150 to 231°C (mean 188°C) and 298 to
ofδ34Ssignatures.
354°C (mean 320°C). Fluid inclusions hosted
Investigation of carbon isotope signatures of
by quartz from extensional veins in wall rock
calcite,ankeriteandsideriteledtoasimilarcon-
arecharacterisedbyT between170and225°C
h
clusionasforthesulphurisotopes(Sun, 2000).
(mean 190°C). Seccombe (1990) argued that
Theδ13Cvariationsareproposedtoreflectmix-
duetomissinghightemperaturemineralassem-
ingofalowδ13Ccomponentsourcedfromcar-
blagesandresultsbasedonsulphurisotopether-
bonaceous metasediments with a higher δ13C
mometry, a pressure correction of the higher
component from the underlying limestone and
temperature fluid would insignificantly elevate
calcareoushostrocksequences.
the temperature and therefore is unnecessary.
Hydrogen and oxygen isotope analyses were
However,hesuggestedtheneedofpressurecor-
undertaken on chlorite, sericite and fluid inclu-
rections for the lower temperature fluid in the
53 |
ADE | 3.2 The current genetic model for the Elura deposit
4.2wt%NaCl (mean 3.2wt%NaCl ), and -28.27 and are more reducing than ore flu-
equiv equiv
respectively. In stage 4, temperature con- idsatthePeakMine(SunandSeccombe,2000).
ditions dropped to lower levels (mean T Sun(2000)arguedthatbecauseoflowandhigh
h
180°C and mean 4.0wt%NaCl ), sightly temperature fluids occurring exclusively in dif-
equiv
increasing in stage 5 (mean T 209°C ferent paragenetic stages, the two fluid mixing
h
and mean 4.6wt%NaCl ) before reach- model proposed by Lawrie and Hinman (1998)
equiv
ing maxima in both, temperature and salin- and Lawrie et al. (1999) couldn’t be substan-
ity, in stage 6. The last stage is char- tiated. She suggested ore formation was sup-
acterised by T between 130 and as high ported by decrease in temperature, rise in pH
h
as 423°C (mean 265°C) and salinities be- and an oxygen fugacity gradient characterised
tween 2.2 and as high as 63.4wt%NaCl byaninwarddirecteddecreasetowardsthecore
equiv
(mean 16.1wt%NaCl ). Sun and Sec- zonesofthemassivemineralisation. Highsalin-
equiv
combe (2000) suggested the mineralising hy- ities of fluids are related to significant concen-
drothermal system ramped up twice, once dur- trations of Ca2+ and Mg2+. The enrichment
ing peak metamorphism resulting in pyrite- ofthoseelements(subsequentlyincorporatedin
dominated mineralisation (stage 1 to 3), and siderite and ankerite) was established via alter-
during a second later phase as the main ore ation of calcite and dolomite contained in the
forming event (stage 4 to 6). Fluid inclu- hostrocksequence.
sions related to pre-mineralisation quartz vein- Jiang et al. (2000) also studied inclusions
ingaremarkedbyaNa+>K+>Ca2+>Cl−> of the ore-forming fluid in sphalerite and iden-
Mg2+> F− ±hydrocarbons (CH /CO between tified moderate to high salinities ranging be-
4
0.27 and 1.58) chemical signature, whereas tween 5 and 41.7wt%NaCl and T be-
equiv h
fluid inclusions associated with the main min- tween 200 and 343°C. Fluids trapped in spha-
eralising event shift to a composition charac- lerite are weakly acidic to alkaline (pH 5.72 to
terised by Mg2+> Ca2+> Na+> Cl−+F−> 7.51) and are enriched in F−,Cl− and CH rel-
4
K+ ±hydrocarbons, characterised by increased ativetothosecontainedininclusionshostedby
CH levels (CH /CO between 1.05 and 8.16). quartz.
4 4
Calculated logfO2 values range between -58.69
55 |
ADE | 3. THE ELURA OREBODY
3.3 Research objectives and
subsequentlydirectlyinfluencingthemineeco-
methodology nomics, the mineralogy and more importantly
themineralchemistryandwholerockgeochem-
istry are responsible for the quality of the min-
At the Elura Mine, three different metal con-
eralconcentrate. Thatis,becauseof: (a)worth-
centrates are produced in its mineral process-
lessmineralspeciesareseparatedtogetherwith
ing facility: zinc concentrate as the main prod-
ore minerals, passively lowering the metal re-
uct, lead-silver concentrate as secondary prod-
covery rate; (b) penalty elements, for instance
uct and copper concentrate as by-product. In
bismuthintheleadconcentrate,activelylower-
the past, the metal recoveries during the min-
ingthequalityandsubsequentlythevalueofthe
eralprocessingstageattheEluraMinewereand
mineralconcentrate.
stillarefluctuatingasafunctionoforefromdif-
Despite the obvious importance of those ore
ferent locations throughout the orebody being
characteristics mentioned above, very little is
minedandmixedbeforedifferentialfrothflota-
known about them in general and basically
tion.
nothingisknownabouttheir3-dimensionalspa-
The tertiary zinc-lead-copper flotation pro-
tial variability. This project will investigate the
cess is a complex and highly sensitive min-
geochemical, mineralogical, mineral chemical
eral separation technique. The flotation recov-
andtexturalzonationoftheEluradepositbased
ery rate and the quality of the mineral concen-
on a large set of samples taken throughout the
tratesarestronglyaffectedbychangesinwhole
orebody. Byfillingthisgapinknowledge,asig-
rock geochemistry, ore mineralogy and min-
nificantcontributiontoabetterunderstandingof
eral chemistry, in addition to the textural char-
theoreformingprocesseswillbemade,andwill
acteristics (e.g. grain size, grain shape, inter-
in particular be a great benefit to the ore pro-
growth, inclusions, ex-solutions, etc.) of the
cessing (i.e. milling, screening, flotation, etc.)
massive sulphide ore. Apart from regular met-
attheEluraMine. Furthermore,adetailedstudy
allurgical test work on material from the flota-
at stope scale is undertaken in order to investi-
tion feed and tailings, only limited studies on
gate local spatial changes of ore characteristics
geo-metallurgical ore characteristics had been
in the three dimensions, which have the poten-
undertakensinceproductioncommenced.
tialofbeingsignificantforthemineralflotation
Thesilverrecovery(floatedasabonusmetal process in respect to concentrate recovery rate
within the lead floatation circuit) never ex- andquality.
ceeded values above approximately 50%. The
Although several research and mine geolo-
3-dimensional silver grade distribution is very
gists have studied and investigated the orebody
well known and is modelled within the re-
in the past, the genesis of the Elura orebody
source block model. However, the silver oc-
remains contentious. Almost as many differ-
currence e.g. as discrete silver mineral phases,
ent genetic models were proposed as scientists
as solid-solution or as impurities in other sul-
had studied the deposit. As shown in the pre-
phidephases,isonlypoorlyunderstood. Conse-
vious section, fundamentally two opposing ge-
quently,thelackofthisknowledgeseemslikely
neticmodelshavebeenproposedinthepast: (1)
to represent the explanation for the low silver
a syngenetic, remobilisation model; and (2) an
recovery rates. The copper concentrate repre-
epigenetic syn-deformational model. Recently,
sents only a minor by-product. However, due
David (2008) suggested significant similarities
to indicated increases of copper concentrations
to Irish base metal deposits and proposed a di-
within the Elura orebody as a function of min-
agenetic model. Therefore, based on a detailed
ing depth, the copper floatation process might
studyofthemineralogy,texture,mineralchem-
becomeincreasinglyimportantinnearfuture.
istry and geochemistry on a large selection of
Apart from the influence of the ore mineral- samples, the ore genesis of the Elura deposit
ogy for ore to concentrate metal recovery and willbeinvestigatedinthisstudy.
56 |
ADE | 3.3 Research objectives and methodology
Alargesetofdata(inparticularconsistingof entire orebody (see chapter 3 on page 59). In
fluid inclusion, in addition to some stable and ordertoaddresstheproblemsandquestionsout-
radiogenic isotope data) had already been ac- lined above, the following aspects and objec-
quired during the last three decades. It is un- tivesweredefinedforthisresearchproject.
fortunate that these studies were rarely based
on an adequately large sample set, considering
3.3.1 Petrography and texture of
the deposit’s large size and its pronounced het-
the massive sulphide miner-
erogeneity. Accordingtothepreviouslyformu-
alisation
lated first research objectives, this study pre-
dominantly focuses on the economic propor- A detailed knowledge about ore characteristics
tion of the mineralising system (i.e. massive is of great importance for a successful mineral
and semi-massive ore types) and on sulphide flotation process and for the production of a
mineral phases. It is aimed to answer genesis- high quality mineral concentrate. Those char-
related questions by focussing on, for exam- acteristicscaneitherdirectlyorindirectlyinflu-
ple, different isotopic systems and/or analyti- encethemineproductionandthemineralrecov-
cal techniques, not utilised in previous studies. ery rate during the flotation process. Amongst
However, some investigations, already under- the important petrographic and textural charac-
takeninthepast(e.g. analysesoffluidinclusion teristics are: grain size (e.g. influences milling
hosted in sphalerite), will be performed in or- time), shape (e.g. elongated grains poten-
dertoconfirmthelegacydataand/ortoimprove tially inter-grown with different minerals spec-
verticaldataresolutionovertheentiredepthex- imen negatively affecting the flotation proper-
tentoftheorebody. ties), intergrowths, inclusions, exsolution, sul-
The second research objective aims at im- phidebanding,etc. Lastbutnotleast,themodal
proving the knowledge of: (a) the temperature percentage of the sulphide mineral phases is of
andpressureconditionsduring,and(b)thetime great importance, in particular the content of
of the ore formation and/or emplacement. The pyrrhotite. Pyrrhotite is a highly reactive min-
integration of mineralogical, geochemical, iso- eral phase and easily oxidises on ore piles un-
topicandthermodynamicdatamayevenhelpto derground and on the run of mine stockpile.
gainabetterunderstandingof: (a)theproperties This in situ oxidation creates consolidated ore
oftheoreformingfluid(s),and(b)thepotential (production losses and delays) and sulphuric
metalsources. acidwhichaffectsthepHintheflotationcircuit
It should be stressed that the entire Cobar andhencelowersmineralrecoveryrates.
Basin represents an enormous accumulation of Reflected and transmitted light ore mi-
lead-zinc-silver-copper-gold base metal occur- croscopy, followed by particle mineral analy-
rences. Coeval structurally similar deposits in sis and field scans via the application of the
Cobar region have to be genetically linked. An QEMSCAN® technique had be used to inves-
aimofthisstudyistofilltheknowledgegapsin tigatethosepetrographicandtexturalcharacter-
respect to ore characteristics and the genesis of isticsmentionedabove.
theEluradeposit. Byincreasingtheunderstand-
ingofoneoredeposit,thegeneticmodelofthe
3.3.2 Mineral chemistry of sul-
entire metallogenic province will be enhanced,
phide ore phases
directlylinkingintotheimprovementofmineral
explorationtechniques. Similar to petrographic and textural ore char-
The research project was carried out in con- acteristics, the mineral chemistry of sulphide
junction with the metallurgical department of phases is very important for the mineral flota-
the Endeavor Mine. An extensive sampling tion technique. That is because minor changes
campaignhadbeenexecutedforthepurposeof can result in significantly different mineral
a good overall 3-dimensional coverage of the flotation characteristics. A profound knowl-
57 |
ADE | 3. THE ELURA OREBODY
edge of the ore mineral and gangue mineral sediments. Rare earth elements (REE) are in-
chemistryisalsoimportantforasuccessfulap- corporated in minerals like zircon and mon-
plication of the QEMSCAN® technique. The azite which are highly refractory. These min-
trace element composition of sulphide min- eralsareknowntobenotaffectedbyhydrother-
erals can furthermore indicate potential metal mal alteration during ore formation and there-
sourcesandthecompositionoforeformingflu- foreshouldbepreservedwithinthemassiveand
ids. Chlorite and arsenopyrite geothermome- semi-massive sulphide mineralisation. REE in
trywillbeusedtofurtherconstraintemperature combination with trace element concentrations
conditionsduringthemineralisingevent. Spha- and their distribution will be used in order to
lerite geobarometry will help to estimate pres- investigate mechanisms of ore precipitation. In
sureseitheroforeformationordeformation. contrast,hydrothermalcarbonates(e.g. siderite)
If penalty elements have been identified via are also known to incorporate REEs to some
wholerockgeochemistry,theknowledgeofthe extent, therefore should reflect the REE signa-
mineral chemistry will help to identify the host ture of the ore forming fluid. By comparing
mineral of those elements and the nature of REE distribution patterns from unaltered coun-
their occurrence, e.g. as solid solutions, ex- try rock with those from orebody samples, in-
solutions, inclusions, or as individual mineral formation about the extent of replacement as
phases. Similar to penalty elements, the host well as the ratio between replacement and ore
mineralsandtheoccurrenceofsilverwillbede- precipitationinopenspaces(e.g. fractures,cav-
terminedviamineralchemicalinvestigations. ities)canbedetermined. REEsmightalsohelp
to investigate physio-chemical features of the
oreformingfluid.
3.3.3 Whole rock geochemistry
Major and trace element geochemistry
Major and trace element geochemistry will be
Platinum group element geochemistry
used to acquire information about composi-
tionalvariabilityofdifferentoretypes,theorig-
Amassivesulphidesedimenthostedbasemetal
inalhostsedimentsandwillbeusedtoreconcile
deposit is under "normal" circumstances not
geochemistry to mineralogy. The analysis of
a deposit style where someone would expect
trace element concentrations will contribute to
to find high concentrations of platinum group
the classification of the host lithology and will
elements (PGE) and therefore were not investi-
enable the identification of alteration/ore form-
gatedattheEluradepositinthepast. However,
ingprocessesandpotentiallytheoriginofmetal
PGEsarequitemobileundercertainconditions
bearing fluid and metal source. Furthermore,
such as under an oxidising environment or via
traceelementgeochemistrywillbeusedtoiden-
transport as chloride complexes (Wood, 2002).
tifypenaltyelements(e.g. bismuth)whichhave
As a trial, the analysis of sulphide samples
the potential of negatively affecting the quality
was proposed for this research project (approx-
of the mineral concentrate. The distribution of
imately 5 to 7 samples). If PGEs have been
other elements might have an influence on the
identified and are at sufficient concentrations,
mineralprocessing.
thefollowinginformationcanbeacquired:
Rare Earth Element geochemistry
(a) Timing of ore formation via Re-Os
Currently, the epigenetic syn-deformational geochronology
model represents the genetic model with the
highestacceptancewithintheresearchcommu- (b) Fluid and metal source based on PGE
nity. Inthistheory,theoredepositionisaccom- distribution pattern and Re-Os isotopic signa-
panied by significant replacement of the host ture.
58 |
ADE | 3.4 Sampling
3.3.4 Fluid characteristics and unexplored until the late 1990s. Upon inven-
timing of the ore forming tion of the multiple collector inductively cou-
event pled plasma mass spectrometry (MC-ICP-MS),
a significantly higher analytical accuracy was
In the past, the deposits in the Cobar Basin made possible. A breakthrough in zinc iso-
including the Elura deposit were dated based tope analytic was achieved by Maréchal et al.
on Pb-Pb and Ar-Ar techniques resulting in a (1999) who, for the first time, was able to de-
significant age discrepancy for the genesis of termine zinc isotope fractionation in nature on
the Elura deposit in the order of 35Ma (based various samples, including geological materi-
on oldest age defined by David (2005) and als. Sincethen,thestableisotopesystematicsof
youngest age by Sun (2000); Sun et al. (2000). zinc gained increasingly more attention. How-
Inordertobetterconstraintheageoftheminer- ever, the Zn isotope data pool for ore forming
alising event it is proposed to use isotopic sys- systemsisstillrathersmallanditsisotopicfrac-
temspreviouslynotapplied. Re-Osgeochronol- tionationmechanismsfarfromwellunderstood.
ogy, which was successfully used for age de- Its use for discrimination between different de-
termination on several sediment hosted base posit types and for identification of important
metal deposits, such as the Century deposit genetic mechanisms is therefore limited at this
(Keaysetal.,2006),theTennantCreekdeposits pointintime. Nevertheless,asatrialandinor-
(McInnes et al., 2008) or the Red Dog deposit der to make an effort to increase the isotopic
(Morelli et al., 2004). Keays et al. (2006) for data available to the research community, zinc
instance was able to identify a two stage ori- isotopesignaturesofsphaleriteareinvestigated
gin of the Century deposit, characterised by an aspartofthisPhDstudy.
early pre-concentration of ore-forming metals,
followed by a later transportation and deposi-
3.4 Sampling
tionviabasinalbrines. Alternatively,Rb-Sriso-
topesinsphaleritehavethepotentialtodirectly
datetheoreformationevent(e.g.Brannonetal. The main strategy for the sampling campaign
(1992); Nakai et al. (1993); Christensen et al. was to: (a) achieve a good three dimensional
(1995b,a); Schneider et al. (2003)). This tech- coverage of the entire orebody, (b) take sam-
niqueisstillquiteexperimental,however,ithad ples over the entire longitudinal and in partic-
been successfully applied on many occasions ular depth extent of the mineralisation, and (c)
andaccuratelydeterminedformationageofsul- sample each individual pipe versus depth with
phidedeposits. high sample density particularly in the main
Investigations of fluid inclusions hosted in lode zone that represents the largest metal ac-
sphalerite will determine fluid temperature, cumulation. Onlyfreshsamplesweretaken,i.e.
salinity and composition, and will particularly that were not affected by supergene processes.
focus on their variability as a function of depth The two year life of mine plan from August
withinthemineralisingsystem. 2008wasconsideredduringthesamplinginor-
dertomakedirectuseofthescientificoutcome
Stable isotope signatures of hydrogen, car-
of this study for potential production and min-
bon, oxygen and sulphur of various minerals
eral processing related questions or problems.
in the Elura orebody have been studied in the
The achieved spatial sample coverage is visu-
past, some of them quite extensively. Rosman
alisedinfigure3.17.
(1972) was the first person to investigate the
isotopic and elemental abundances of zinc via Duetothepipe-likegeometryoftheindivid-
thermal ionisation mass spectroscopy. Because ualmassivesulphidebodies,whichisrathernar-
of high analytical errors as a consequence of rowinseveralzoneswithinthenorthernpodsin
the very minor mass dependant isotope frac- particular having an east-west direction, a de-
tionation of zinc, the isotopic system remained tailedsamplingoftheirlateralzonationwasnot
59 |
ADE | Chapter 4
Ore mineralogy and petrography
DETAILED microscopy was undertaken to
describetexturalorecharacteristics,iden-
tifythemineralphasesandtheirgrainsizevari-
abilityandtodescribetheparageneticandtem-
poral association. Electron microprobe analy-
sis (EMPA) was used to determine the major
and minor element composition of major sul-
phide mineral phases and their compositional
variationswithrespecttodifferentoretypesand
depth within the orebody (sample locations are
shown in figure 4.1). Trace element concentra-
tions, with particular emphasis on penalty ele-
mentsthathaveapoordetectionlimitbyEMPA
andweredeterminedbyLA-ICP-MS.
4.1 Microscopic
observations
Figure4.1: EMPAsamplelocations;sampleswere
selectedinordertocovertheentiredepthextentof
Macroscopic observations from underground the orebody, all ore zones and ore types. Longitu-
dinalviewtowardsWSW(˜245°). Yellowshapeis
exposures and diamond drill core (DDH)
thestringertypeore(VEIN)resourcedomain.
throughout the orebody show significant
changesinmodalabundancedemonstratingthe
highly heterogeneous nature of the ore. This mine areas. Within the stringer vein type
variability is present on metre scales down to mineralisation at depth, sphalerite appears
scales in the order of a few millimetres. The to be darker in colour, chalcopyrite tends to
most noticeable sulphide texture is a distinct occurs in greater abundances and the quantity
sulphide layering or banding on millimetre to ofpyrrhotiteexceedsthatofpyrite.
centimetre scale defined by elongated areas Iron sulphides (i.e. pyrite and pyrrhotite) are
enriched in sphalerite and pyrite, others by the most abundant sulphide mineral phases at
galena and, if present, pyrrhotite and minor theEluradeposit. Thepresenceofpyrrhotite,in
chalcopyrite. Large patches, zones or veins additiontotheoccurrenceandquantityofrem-
significantly enriched in commonly coarse nantwallrockfragmentsandquartzareusedto
grained galena (in the order of a few millime- definedifferentoretypes. Basemetalsulphides
tres in diameter) are encountered in several occurinthefollowingdecreasingorderofabun-
63 |
ADE | 4. ORE MINERALOGY AND PETROGRAPHY
dance: sphalerite>galena>chalcopyrite. are commonly tectonised, as well as replaced
Ironcarbonatesandquartzrepresentthemost andintrudedbybasemetalsulphides. Intrusion
important non-sulphide gangue (NSG) mineral and initiation of replacement by base metal
phases. The former preferably but not ex- sulphidesalonggrainboundariesarecommon.
clusively occurs in massive ore types (Po and
Py). Quartz gangue is predominantly observed
in siliceous semi-massive ore types and mas- CataclasticpyriteB
sive pyritic ore (SiPy, SiPo and Py). Vari- (pre-toearlysyn-basemetal)
ablequantitiesofalkali-feldspar,albite,chlorite
Tectonised pyrite B is sub- to euhedral in
andsericitearerelatedtometasomatismofwall
shape and commonly strongly corroded at
rockasaninsitualterationoralterationoffrag-
grain margins. Abundant inclusions of base
mentsembeddedwithinsulphidegroundmass.
metals, pyrrhotite and minor NSG phases are
Selected microphotographs and back scatter
observed. These mineral phases intrude along
electron (BSE) images are presented in the fol-
fractures and grain boundaries, subsequently
lowing. Further images are shown in appendix
replacing pyrite. Grain sizes of pyrite B range
A from page 270 (photomicrographs) and ap-
between10and500µmwithavisualmedianof
pendixAfrompage311(BSEimages).
approximately 50 to 100µm. In places, pyrite
B is intergrown with pyrite A forming large
4.1.1 Sulphide gangue phases
aggregatesofmillimetresizebutalsooccursas
isolatedporphyroblastswithinagroundmassof
Pyrite
pyrrhotiteandbasemetalsulphides. Micropho-
Various pyrite species were observed and tographs are shown in figure 4.2 (image b and
grouped according to their genetic nature and c)andfigure4.3(imagecandd).
timing in respect to base metal mineralisation.
Distant to the deposit, the CSA siltstone
itself contains euhedral and framboidal syn- Syn-basemetalpyriteCvarieties
genetic/diagenetic pyrite at concentrations
Late syn- base metal pyrite C, in places,
ranging between 0.25-0.7wt% (Taylor et al.,
intimately intergrown with magnetite in a
1984). These pyrite varieties were not investi-
symplectic-like texture, occurs exclusively in
gatedinthecourseofthisstudy.
assemblages with pyrrhotite (image e in figure
4.3). This pyrite C variety is characterised by
grain sizes of approximately 5 to 20µm and
Colloform,framboidalandcloudypyriteA
occurs as discrete grains commonly within a
(pre-basemetal)
groundmass of pyrrhotite or may overgrow as
Type A pyrite represents the earliest sulphide wellasreplacepyriteB.Magnetiteiscommonly
mineralisation. Microphotographs are shown veryfinegrainedintheorderofafewmicrons.
in figure 4.2 (image a and b) and figure 4.3 In places, larger grains of approximately 10-
(image a and b). Fine-grained cloudy pyrite 20µm are present. Minor marcasite replacing
is commonly in the order of a few microns pyrrhotitemayoccurinthevicinityofpyriteC.
in size and forms large euhedral aggregates A second pyrite C variety is found in pyrite
containing abundant inclusion of NSG. The dominated ore types. It is anhedral in shape,
anhedralcolloformpyritegenerallyformslarge commonlyformsgrainsbetweenafewmicrons
dense and rather homogeneous aggregates and up to ˜50µm and contains abundant inclu-
where grain sizes are oblique. Subordinate sions of NSG phases and base metal sulphides.
framboidal pyrite is observed as isolated grains If present, pyrite C is commonly overgrowing
or incorporated in larger pyrite aggregates of pyriteAandB(imagefinfigure4.3).
other pyrite varieties. These pyrite aggregates
64 |
ADE | 4.1 Microscopic observations
Pyrrhotite
LaterecrystallisedpyriteD(post-basemetal)
Pyrrhotite is the dominant iron sulphide phase
Euhedral pyrite D is entirely inclusion free, incorezonesoftheEluraorebody. Thosemas-
characterised by small grain sizes in the order sive pyrrhotitic ore zones are commonly sur-
ofapproximately5-20µmandismostlyassoci- rounded by semi-massive siliceous pyrrhotitic
atedwithbasemetals,predominantlysphalerite. or massive pyritic ore. The occurrence of
This pyrite type is completely unfractured and pyrrhotiteismanifold:
commonly overgrows earlier pyrite generations (1) The vast majority represents the ground-
A-C or is associated with NSG (image a, b and mass of massive and semi-massive siliceous
finfigure4.3). pyrrhotitic ore. It is composed of equant sub-
to euhedral grains forming a granular texture.
Sizes of those pyrrhotite grains range between
20 to 50µm and commonly feature 120° triple
(a) (b)
Py B
Ag-Tet Cpy
Sph
Py A
(c) (d)
FeOOH
Gn
Py B
Sph
Sph
Fe-carb
Figure 4.2: Back scatter electron images of pyrite and its replacement: (a) Fine grained
galenaintrudingandreplacingcolloformpyriteAalonggrainboundariesandfractures;minor
argentiantetrahedriteisintergrownwithgalena;chalcopyriteoccursinterstitialtopyrite(sam-
pleDE398;Pyoretype;9775mRL;MLzone). (b)AcolloformpyriteAfragmentintrudedand
replaced by sphalerite; pyrite A is partially overgrown by sub- to euhedral pyrite B (sample
DE174-1-A; Po ore type; 10082mRL; ML zone). (c) Sphalerite, commonly intergrown with
galena, is overgrowing and replacing strongly tectonised pyrite B (sample DE174-1-A; Po ore
type; 10082mRL; ML zone). (d) Zone of framboidal and sub- to euhedral pyrite contained
in weakly mineralised and altered wall rock. It is uncertain whether these pyrite grains are
of hydrothermal or diagenetic origin. Both pyrite varieties are partially replaced by goethite.
Minor low-iron sphalerite occurs interstitial to pyrite or within the strongly chloritised wall
rock. Abundantironcarbonateisobservedinterstitialtopyrite(sampleNP948-2-A;MinAore
type;9185mRL;MLzone).
65 |
ADE | 4.1 Microscopic observations
junctions. In places, pyrrhotite appears to be 200 µm
replacedbybasemetals.
(2) Abundant inclusions of pyrrhotite are
predominantly hosted by sphalerite and are
commonly aligned along its crystallographic
planes. Otherwise, pyrrhotite occurs along
fractures in cataclastic pyrite type A and B.
Grain sizes vary between a few microns and
up to 50µm. Small grains in sphalerite may
representexsolvedgrainswhereascoarsegrains
are likely to be either linked to annealing of
former tiny exsolved grains or are inclusions Figure 4.4: Sulphidebandingisdefinedbyzones
oftightlyintergrownsphalerite(mediumgrey)and
causedbycontemporaneousmineralgrowth.
galena(lightgrey)interlayeredwithzonesenriched
in marcasite (light yellow greenish); galena (light
(3) The remaining pyrrhotite is generally in- grey)preferablyoccursinthevicinityofthemarc-
asitezone;euhedralpyriteB(lightyellow)isfound
timately intergrown with base metals in a in both layers. The sample is classified as Py ore
myrmekitic-liketexture,clearlyindicatingcon- type. Oretextureistypicalformassivepyrrhotitic
ore. Priortomarcasiteformation,thesamplecon-
temporaneous growth of base metal phases and
tained significant pyrrhotite and would therefore
pyrrhotite. Grain sizes vary largely and range needtobeclassifiedasPoingeneticterms;sample
DE174-2,Pyoretype,10102mRL,MLzone.
betweenafewmicronsandupto100µm.
casite formation is clearly linked to iron car-
Marcasite
bonate veins and commonly progresses along
Marcasite is observed in variable quantities in grain boundaries of iron carbonates. Occasion-
severalsamplesthroughouttheentiredepthex- ally,marcasiteappearstobeassociatedwiththe
tent of the orebody. In the uppermost parts of formation of pyrite. It is not clear whether the
the orebody, former massive and semi-massive transformation pyrrhotite > marcasite > pyrite
pyrrhotite-rich ore types are strongly altered, took place or the formation of both phases are
most likely due to supergene descending flu- geneticallyunrelated.
ids. Most of this pyrrhotite is completely
transformed into marcasite in a typical worm-
Arsenopyrite
or lamellae-like texture. A distinct marcasite
banding is present in places where pyrrhotite Arsenopyrite is present in all paragenic stages,
occurred in a layered texture (i.e. pseudo- oretypesandthroughouttheentireorebody,but
morphic texture, example shown in figure 4.4). occurs in noticeable larger quantities towards
Marcasitefillsinterstitialspacesbetweenpyrite theupperpartsoftheEluraorebody. Grainsizes
AandB.Thisisacommontextureofunaltered vary significantly and range between a few mi-
pyrrhotiteindicatingmarcasiteformationdueto cronsanduptoaslargeas300µm. Smallgrains
alteration of pyrrhotite. Furthermore, marca- tend to be anhedral in shape whereas larger
siteispreservedassmallinclusionsinsphalerite grainsaregenerallysub-toeuhedral. Cataclas-
shielded from alteration fluids, whereas former tic textures are commonly observed. Most ar-
pyrrhotite intimately intergrown with galena is senopyrite is incorporated in large aggregates
completelytransformed. Withincreasingdepth, of pyrite, intergrown with pyrite A and B, but
pyrrhotitetomarcasite transformation becomes is also found intergrown with syn-base metal
progressively incomplete and rather variable, pyrite C. In other places, euhedral arsenopyrite
commonly appearing in a “birds-eye-like” tex- is contained within iron carbonate veins. Some
ture. Insampleswherealterationisminor,mar- largersub-toeuhedralarsenopyritegrainscarry
67 |
ADE | 4. ORE MINERALOGY AND PETROGRAPHY
intergrownwithpyrrhotiteandotherbasemetal
sulphides in a myrmekitic-like texture. Com-
monly, pronounced sphalerite replacements of
pyriteingeneral,andvughorfracturefillingsin
pyriteAandpyriteB,respectively,areobserved
(figure4.6andimageainfigure4.3).
Galena
The most pronounced grain size variation of
all identified mineral phases is exhibited by
galena (figure 4.6). Grains as tiny as a few mi-
Figure 4.5: Euhedral zoned arsenopyrite with a crons contrast with occurrences of large mono-
homogeneous and corroded core; grain is weakly
mineralaggregatesorpatchesuptotensofcen-
tectonisedandcontainsabundantinclusionsofiron
carbonate; sample 500-ML-E-A, SiPy ore type, timetres in size. These patches are composed
9505mRL,MLzone. of grains with maximum sizes in the order of
a few hundred microns. Grains are strongly
deformed in places as indicated by bending of
abundant inclusions of pyrite, pyrrhotite, base
cleavage planes. Galena occurs intimately in-
metal sulphides and siderite. Replacements of
tergrown with other base metal sulphides and
preferably large sub- to euhedral arsenopyrite
pyrrhotite (figure 4.10), fills fractures in pyrite,
grains by base metal sulphides and pyrrhotite
or occurs in interstitial spaces of pyrite (figure
are common. An alternating and rather thin
4.12). Fine-grainedgalena(2-10µm)ispresent
growthzonationparalleltothecrystallographic
inhighlyalteredwallrockfragments(image(a)
planesisobservedinseveralarsenopyritegrains
infigure4.14)andoccasionallyasinclusionsin
(figure 4.5). A second zonation pattern, char-
sphalerite.
acterised by a more irregular zonation, is over-
printingtheaforementionedpattern.
Chalcopyrite
4.1.2 Base metal and silver phases
Apartfromtetrahedrite,minorchalcopyriterep-
resents the only copper-bearing mineral phase
Sphalerite
within the orebody. An increase in its quantity
Sphalerite is the most abundant base metal sul- towards greater depths within the deposit is a
phide in the Elura deposit and is characterised characteristic feature. Fine grained chalcopy-
by changes in iron content as indicated by ritepredominantlyoccursasinclusionsinspha-
its colour (between light brown-yellowish to lerite. These inclusions are likely to represent
dark brown). Low-iron sphalerite is primarily exsolutionsifpresentastinyrandomlyoriented
present in the upper parts of the orebody as grains smaller than or in the order of 1µm.
well as in those of minor mineralised wall rock Lamellae or elongated grains, predominantly
(MinA). Similar to all other sulphide phases, occurring along crystallographic or cleavage
grain sizes vary largely and are in the order of planes of sphalerite, associated with pyrrhotite
afewmicronsanduptoapproximately400µm at times, are interpreted as intergrowths or re-
(figure4.6). Sphaleriteisveryimpure,contain- placements. Roundishdroplet-likechalcopyrite
ing abundant inclusions of siderite, chalcopy- grains with sizes ranging between 5 to 10µm
rite, pyrite, pyrrhotite, andoccasionallygalena. arelikelycausedbyannealingoftheaforemen-
Sphalerite textures are variable, e.g. occurring tioned finer grained varieties. In general, chal-
asbands, elongatedlenses,interstitialtopyrite, copyrite preferentially occurs in the vicinity of
coarse and well defined crystals or intimately ironcarbonatesand/orquartz,characteristically
68 |
ADE | 4.1 Microscopic observations
(Py and SiPy) in the upper 100-150m of the
main lode zone. The uppermost zones of
the northern pods (z1 to z5) are enriched in
tetrahedrite associated with galena in a simi-
lar manner as observed in the upper main lode
zone. However, a significant grain size reduc-
tion down to maxima of around 15-20µm is
common in these zones. The preferred associ-
ationwithgalenaisnotobservedintheremain-
deroftheorebodyweremanifoldmineralasso-
ciations and intergrowths are present, e.g. with
pyrite, chalcopyrite, galena, siderite and quartz
Apy (figure4.7). Thesetetrahedritegrainsaregener-
allyintheorderofafewmicronsanduptoap-
proximately 15µm in size. Occasionally, triple
junctionsbetweensphalerite, galena, chalcopy-
rite and tetrahedrite are present, either indicat-
Sph ingcontemporaneousmineralgrowthorrecrys-
Py A tallisation. Alteration of tetrahedrite to native
Py B silverandchalcopyriteisoccasionallyobserved
withinsamplesfromtheupperorezones. Den-
dritic native silver occurs in marginal areas of
altered tetrahedrite grains. Whereas internally,
silver forms tiny inclusions or is intimately in-
tergrownwithtetrahedriteinareticular-liketex-
Figure 4.6: Significant grain size variation of
sphalerite and galena: Galena occurs either as ture. These intimate intergrowths cause a pro-
coarse patches or as thin veinlets intruding pyrite
nouncedgrainyappearanceinbackscatterelec-
alongfracturesorgrainboundaries(upperimage);
sampleDE008-2,Pooretype,9993mRL,MLzone. tronimaging. Otherzonesinalteredtetrahedrite
Fine grained sphalerite is progressively replacing nowconsistofchalcopyrite. Thealterationpat-
a compound of fine grained pyrite A (lower im-
tern is diffuse with tetrahedrite progressively
age); sub- to euhedral pyrite B is only replaced at
grain margins; fractured arsenopyrite is intruded transformedtochalcopyriteand/ornativesilver
bygalena(white);sampleDE174-1-A,Pooretype,
(seesection4.2.2frompage93).
10082mRL,MLzone.
4.1.3 Non-sulphide gangue phases
indicating an affinity to these NSG phases (im-
and altered wall rock frag-
age c in figure 4.14). Large patches of chal-
ments
copyrite up to 400µm in size may be present
preferablyinsiliceousoretypes(SiPyandSiPo)
Iron carbonate is the most abundant non-
orinbreccia-stringertypeore(VEIN).Intimate
sulphide gangue mineral phase in pyrrhotitic
intergrowthsofchalcopyritewithgalenaand/or
and in most of pyritic ore. A general increase
pyrrhotitearecommon.
of quartz gangue is obvious towards periph-
eralzones,particularlyinsemi-massiveore(Py
and Po). Quartz content in massive pyrrhotitic
Tetrahedrite and native silver
ore is very low. Changes in colour and refrac-
Coarse tetrahedrite with sizes up to approxi- tion of iron carbonates observed in transmitted
mately200µmisalmostexclusivelyassociated lightindicatevariabilityoftheirchemistry. The
with galena (figure 4.7). It predominantly oc- chemical variability of iron carbonate phases
curs in massive and semi-massive pyritic ore wasqualitativelydeterminedviaenergydisper-
69 |
ADE | 4.1 Microscopic observations
tion, replacement of these porphyroblasts by sionally,tetrahedriteisassociatedwithgoethite.
sphalerite is observed (figure 3.8). Subordinate Barite is a common, but a minor mineral con-
alkali-feldspar and sodium-rich plagioclase is stituent (figure 4.8). Galena may be partially
recognised in altered in situ wall rock or wall alteredtocerussiteintheuppermostpartsofthe
rock fragments embedded in sulphide ground- orebody affected by supergene processes (im-
mass. Feldspars commonly exhibit numerous agedinfigure4.8).
tiny inclusions of sericite, calcite and chlorite,
asaresultofalteration. Subordinated magnetite is exclusively asso-
ciated and intimately intergrown with pyrite
Minorgoethiteoccurrencesarepresentinall type C. The co-genetic association between
ore types but have been observed particularly pyrrhotite, pyrite and magnetite is very rare.
in those of pyritic dominance (figure 4.8). It Magnetitegrainsareverysmall(commonlybe-
preferably replaces siderite along grain bound- low10to15µandmaybemixedupwithspha-
aries and fractures mostly as coarser anhedral lerite and or Fe-hydroxides. In order to ensure
radial mineral aggregates (several tens of mi- correct visual identification, mineral composi-
crons) or thin veinlets in the order of a few mi- tion of the supposed magnetite shown in figure
crons. Minor sub- to euhedral goethite inter- 4.3 (image e) was determined via electron mi-
grown with iron carbonate is observed in veins croprobe analyses (see section 4.2 from page
or as interstitial filling between pyrite. Occa- 91).
(a) (b)
40 µm 40 µm
(c) (d)
Gn & Cpy
Sph
Gn
Brt
Py
Gt & Chl
Cer Brt
Qz
Fe-carb
Figure 4.8: Various non-sulphide gangue (NSG) phases in reflected (a) and transmitted (b) micropho-
tographs(sampleDE398,Pyoretype,9775mRL,MLzone),aswellasinBSEimages(imagescandd;sample
DE306, Py ore type, 9713mRL, ML zone). Goethite (centre of images a and b, medium grey in RL, reddish
inTL,bothparallelnicols)isreplacingironcarbonate(black). Forcomparison,lowironsphaleriteoccurson
therightimagemargin, featuringalmostsimilarreflectanceand colourinRLbutshowsonlyweakyellowish
colouring in TL. (c) Quartz, iron-carbonates and chlorite are the most common NSG phases. Minor occur-
rances of goethite and barite are observed, the former partially replacing pyrite. (d) Alteration of galena to
cerussiteislimitedtotheuppermostpartsoftheorebody.
71 |
ADE | 4. ORE MINERALOGY AND PETROGRAPHY
4.1.4 Ore type specific character- texture, present exclusively in samples of mas-
istics and textures sivepyrrhotiticore.
Sphalerite is commonly medium to dark
Massive pyrrhotitic ore (Po)
brownincolourandratherimpureduetoabun-
Significantpyrrhotitequantitiesandpyriteofall dant inclusions comprising NSG (i.e. predomi-
varietiesarepresentinthisoretype(typicalex- nantlyironcarbonate),pyrrhotite,minorgalena
amplesshowninfigure4.10). Inminegeology, and chalcopyrite. The abundance of inclusions
ore classification is based on visual identifica- is variable. Chalcopyrite inclusions in spha-
tion of pyrrhotite and the empirically identified lerite are more abundant compared to pyrite
intensityofsamplemagnetismviahandmagnet. dominated ore. In samples featuring sulphide
Pyrrhotitecontentinthisoretypevariessignif- banding, layers significantly enriched in spha-
icantly. Classificationisnotstrictlyconstrained lerite are observed. Galena is commonly as-
andneedstobeunderstoodasdiffusetransition sociated with pyrrhotite or occurs interstitial
from pyritic to pyrrhotitic ore. However, most to pyrite and is characterised by a pronounced
investigated thin sections of the latter ore type grainsizevariability. Chalcopyriteispreferably
containatleast5to10wt%pyrrhotite, somein associatedwithironcarbonate. Otherwiseitoc-
excess of 50wt%. Pyrrhotite commonly repre- curs as inclusions in sphalerite or interstitial to
sents the sulphide groundmass or is intimately pyrite. Tetrahedrite is rarely observed micro-
intergrownwithbasemetalsulphides,inplaces, scopically and appears in very small grain size
in an annealed texture. These tight intergrows (i.e. 5to10µm).
are characteristic for this ore type and may be General features of sulphide banding are
very fine at grain sizes below 10µm. In areas defined by layers enriched in: (1) pyrite
ofintensesulphidebanding,pyrrhotiteiseither with pronounced occurrence of galena and/or
intimately intergrown with galena and/or chal- chalcopyrite in interstitial spaces; (2) galena,
copyrite,oritformsalmostmonomineraliclay- pyrrhotite and occasionally chalcopyrite; and
ers. (3)inclusion-richsphalerite. Layers(2)and(3)
Partialtransformationofpyrrhotitetomarca- are either rather continuous, or occur as elon-
site is present in several samples and observed gatedlentoidalbodies,orreminiscentofpinch-
even in the lowermost part of the ore body at and-swell textures. A mylonitic texture is ob-
depths of almost up to 1,000m below surface. served in samples featuring intense sulphide
In contrast, complete transformation is limited banding. This texture is characterised by the
to the uppermost areas in the main lode zone. occurrence of rounded clasts of sphalerite ei-
A spatial relationship between occurrences of ther isolated in layers composed of pyrrhotite
symplectic pyrite C with beginning marcasite andgalenaorwithinsphalerite-richlayers.
alteration is indicated in some samples. Fur-
thermore, partial pyrrhotite-marcasite transfor-
Massive pyritic ore (Py)
mation is also present in transitional zones be-
tweenpyrrhotiteandpyritedominatedorezone. Pyriteisbyfarthemostdominantironsulphide
Pyrite abundances vary, with type B signifi- mineral species, and represents the major sul-
cantlyexceedingthatofpyritetypeA.PyriteC phidephaseingeneral(typicalexamplesshown
and D commonly represent minor constituents. in figure 4.12). Locally, traces of pyrrhotite
Pyrite B is strongly corroded and, in places, may be observed. Former pyrrhotite affected
features a pronounced cataclastic texture. The by complete transformation to marcasite is ob-
grains contain abundant inclusions of, and are served in samples taken from the uppermost
intensely replaced by base metal sulphides in central parts of the main lode zone. These ar-
additiontopyrrhotite. Thesemineralphasesare eas are classified as pyritic ore type, although
present along fractures. Pyrite C is intimately in genetic terms, they would need to be viewed
intergrown with magnetite in a symplectic-like aspyrrhotite-richore. Tracesofpyrrhotitemay
72 |
ADE | 4.1 Microscopic observations
(a) Figure 4.10: Typical examples of massive
200 µm pyrrhotitic ore and its textural characteristics
(reflected light images). (a) Pyrite B (yellow)
dominates over pyrite type A and occur in a
pyrrhotite (pink) rich zones. In this image,
only pyrite B is observed, which is fractured,
stronglycorrodedandcontainsabundantinclu-
sionsofsphalerite(mediumgrey),chalcopyrite
(yellow) and galena (light grey). Pyrrhotite
is intimately intergrown with sphalerite. A
weak sulphide banding is observed, indicated
by the elongated sphalerite rich zone; sample
NP776-2, 9334mRL, z3 zone. (b) Intimate
myrmekiticintergrowthsofsphalerite(medium
(b) grey),galena(lightgrey)andpyrrhotite(pink);
euhedral to subhedral pyrite B is corroded
200 µm
andcommonlycontainsinclusionsofbasemet-
als; sample NP245-1, 9613mRL, z4 zone. (c)
Weakly developed sulphide banding is defined
by elongated grains of galena (light grey) in a
matrixconsistingofpyrrhotite(pink);rounded
clasts of sphalerite (medium grey) without
preferred orientation. Minor iron carbonate
(medium-darkgrey)andquartz(darkgrey)as
NSGphases;sampleNP767-2-A,9354mRL,z4
zone.
(c)
200 µm
be preserved in sphalerite. All pyrite types are in narrow fractures. Most sphalerite is coarser
present with pyrite A and B as the most abun- compared to galena. It occurs as bands and
dant species, and within those types, pyrite B lenses at grain sizes in the order of a few to
subordinate to pyrite A. Very fine grained syn- several tens of microns but may be larger than
base metal pyrite C is likely to have formed in 200µm. The colour of sphalerite is medium
lowquantities,however,microscopicidentifica- to dark brown. A shift to slightly lower iron
tion and quantification is difficult. Similarly to content compared to sphalerite in Po ore is in-
pyrrhotitic ore, pyrite D is common but occurs dicated. Tiny inclusions of chalcopyrite are
in minor quantities. Pyrite A and B are gen- occasionally present in very minor quantities.
erally intruded and replaced by base metal sul- Galena is present as large patches in excess of
phides, either along fractures or grain bound- 200µm but may be as fine as less than one mi-
aries. In places, primary colloform or fram- croninsize. Chalcopyriteconcentrationisgen-
boidaltexturesarepreserved. erallylowbuthighlyvariable,absentinmostar-
Grain sizes of base metal sulphides vary and eas but relatively enriched in others. If present
mayberatherfine(sub-micronto10µm)ifoc- it is associated with galena but preferentially
curring in interstitial space between pyrite or occurs in zones rich in NSG (quartz/carbonate
73 |
ADE | 4. ORE MINERALOGY AND PETROGRAPHY
Figure 4.12: Typical examples of mas- (a)
sive pyritic ore and its textural characteris- 200 µm
tics (reflected light images). (a) Tectonised
compounds of pyrite A (yellow) intergrown
with abundant sub- to anhedral arsenopyrite
(whitish) and intruded as well as replaced by
sphalerite (medium grey) and galena (light
grey). Bothphasesarealmostperpendicularto
each other. Subordinate euhedral to subhedral
pyrite B at rather small grain sizes; quartz oc-
curs in minor quantities; sample CAF-1LS-1-1,
10135mRL, ML zone. (b) Left side of image:
zoneoftectonisedcoarseeuhedralpyriteB(yel-
low)inaNSGgroundmasspredominantlycom-
posedofironcarbonate(darkgrey)andsubor- (b)
dinatequartz(black);minorsphalerite(medium
200 µm
grey)andgalena(lightgrey)occurinterstitially
to pyrite or intrudes and partially replaces it.
Right side of image: zone is predominantly
composed of cloudy pyrite A, in places, over-
grownbyanhedralpyriteCandeuhedtalpyrite
D. Minor occurrences of tectonised subhedral
pyrite B, partially replaced by base metal sul-
phides. Arsenopyrite (whitish) is intergrown
with pyrite; sample DE008-3, 9937mRL, ML
zone. (c) Tectonised fragment of rather com-
pact pyrite (light yellow) composed of pyrite
typeA.Fractureisfilledbysphalerite(medium
grey), chalcopyrite (yellow) and galena (light
(c)
grey); ground mass consists of finer grained
cloudy pyrite type A and minor sub- to eu- 200 µm
hedral pyrite type B and D. Base metals oc-
cur interstitial to pyrite A and C; minor type
D pyrite occurs in sphalerite; abundant ar-
senopyrite (whitish); quartz (black) as none
sulphide gangue (NSG) phase; sample DE007,
9959,mRL,MLzone.
veins). Tetrahedriteismostcommoninpyritic- pronounced when compared to pyrrhotitic ore.
dominated ore as trace mineral phase. In the The nature of preferred orientation of layers or
upperandperipheralpartsofthemainlodezone lensesenrichedindifferentbasemetalsulphides
it is almost exclusively associated with galena, is controlled by brittle deformation, whereas
exhibitingcommonlylargegrainsizesintheor- sulphide banding in Po ore preferably exhibits
der of 200µm. In most other zones within the ductilefeatures. Thebrittlenatureofpyritede-
orebody,asignificantgrainsizereductiondown formationisexpressedbyanalmostperpendic-
to approximately 5-20µm is observed. Here, ularalignmentbetweensphalerite-richlayersor
tetrahedrite is not necessarily associated with lenses, and dilational fractures in pyrite filled
galena but is observed to be intergrown with with galena. Galena appears to be deformed
sphalerite and chalcopyrite. Pyrrhotite may be in some areas. Sphalerite rich layers are im-
associated with tetrahedrite in transitional ore pure and contain abundant inclusions of pyrite,
zonesclosetomassivepyrrhotiticore. NSG(predominantlyironcarbonate)andminor
galena.
Locally-developedsulphidebandingisnotas
74 |
ADE | 4.1 Microscopic observations
(a) Figure 4.14: Microphotographs of semi-
massive ore types: (a) Highly altered wall
200 µm
rock fragment in SiPy ore type. The frag-
ment consists of iron carbonate, chlorite and
sericite(differentshadesofdarkishgrey)asal-
terationphases. Veryfinegrainedgalena(light
grey) probably intrudes in space created due
to volume loss during alteration. Abundant
sphalerite(mediumgrey)containsinclusionsof
galenaandpyrrhotite(pink). PyriteBisfrac-
tured and commonly corroded, intruded and
replaced by base metal sulphides. Minor fine
grained euhedral pyrite D is observed; sam-
pleNP910-1-A,9244mRL,MLzone. (b)Inti-
mateintergrowthsofpyrrhotite(pinkish)with
(b)
sphalerite (medium grey) and minor galena
40 µm (light grey). NSG phase is predominantly
subhedral quartz (black) and late crosscutting
iron carbonate veins (dark grey); sample 665-
ML1-1-A, 9668mRL, ML zone. (c) Quartz
carbonate veining in VEIN ore type. Spha-
lerite(mediumgrey),chalcopyrite(yellow)and
galena (light grey) are hosted by a vein com-
posed of quartz, iron carbonates and minor
pyrite;sampleNP955-1,9607mRL,MLzone.
(c)
200 µm
Semi-massive siliceous ore (SiPy and preferably occurs within or proximal to NSG.
SiPo) and breccia-stringer type min- Locally,pyrrhotiteiscompletelytransformedto
eralisation (VEIN) marcasite. These zones are now classified as
SiPybuttexturesaresimilartopyrrhotitedom-
Themaindifferencebetweenthesiliceoussemi- inated ore types. Pyrrhotite to marcasite trans-
massive analogues of the previously outlined formation appears to be intensified in siliceous
massive sulphide ore types are higher quartz compared to massive pyrrhotitic ore. SiPy is
content and the occurrence of wall rock frag- dominatedbyvariablyfracturedpyriteAandB.
ments. Thosefragmentsareincorporatedinthe Wall rock fragments are generally strongly
sulphide groundmass and commonly strongly altered. Sericite and chlorite are abundant but
alteredandreplaced. Oretexturesandsulphide appear to occur preferably in former mud- and
pargenesesaresimilar(figure4.14). siltstone rich layers. Iron carbonates and base
Base metal sulphides in SiPo ore are com- metal sulphides are commonly present in sand-
monly intimately intergrown. An- to subhedral stonerichlayers,partiallytocompletelyreplac-
quartz and iron carbonate are the most impor- ing quartz. The fragments may be, in places,
tantNSGphasesinbothoretypes. Chalcopyrite strongly silicified in particular in semi-massive
75 |
ADE | 4. ORE MINERALOGY AND PETROGRAPHY
oreatgreaterdepths. eralphase,wasfracturedand,inplaces,signifi-
The borderline-economic breccia-stringer cantlytectonised. Primarymineralintergrowths
type mineralisation (VEIN) is characterised by werelostinmanyareasasaconsequenceofde-
reticularquartz-ironcarbonate-sulphideveining formational induced remobilisation, thus con-
hosted by brecciated and strongly altered CSA cealing the initial mineral paragenesis. Nev-
wall rock. The shell of VEIN surrounding ertheless, several observations were made that
semi-massive and massive ore is rather narrow indicate the sequence of mineral formation and
in most zones of the orebody (see chapter 3 aresummarisedbelow. Themineralparagenesis
on page 39). At depth below approximately oftheEluradepositisshowninfigure4.15.
9300mRL the zone becomes more prominent
accompanied by pronounced and widespread • Sphalerite, galena and chalcopyrite are
silicification and iron carbonate alteration. The clearly associated and commonly inti-
sulphide stringers contain abundant chalcopy- mately intergrown with pyrrhotite which
rite in these zones, particularly beneath the is by far the most abundant iron sulphide
massivesulphidepipes. phase during the main stage of base metal
The prevalence of iron sulphide species and formation. Intergrowths of tetrahedrite
the base metal sulphide content vary between with galena are common, thus contempo-
locations. The intensity of alteration generally raneous.
increasestowardszonesofintensifyingsulphide
• Pre-base metal pyrite A is fine grained,
veining. Proximal to those veins, former CSA
fractured and occurs as colloform, fram-
siltstone wall rock is entirely replaced by iron
boidal and cloudy varieties. Replacement
carbonates,quartz,sericiteandchlorite. Altered
of pyrite A by sphalerite and galena has
wall rock fragments are occasionally observed
beenobservedbutnoprimaryintergrowths
within these veins and are hosted by a matrix
or mineral associations are present. This
composedofsulphidephases.
pyrite variety clearly formed prior to the
Contentsofpyrrhotite,chalcopyriteandiron-
main stage of base metal precipitation,
richsphaleriteareelevatedinstingeroreproxi-
most likely from a lower temperature hy-
maltoandbelowofthesub-verticaloreshoots.
drothermalfluid.
Furtherdistanttothesecorezones,ironinspha-
lerite decrease and pyrrhotite is commonly ab- • Coarse cataclastic pyrite B is observed
sent. Similarily,theintensityofwallrockalter- withinpyrite(PyandSiPy)and,inplaces,
ationdecreases. Chalcopyritemaystilloccurin in pyrrhotite dominated ore zones (Po
relativelylargequantities,evenratherdistantto and SiPo). Strong replacement by and
core zones. Pyrite content is variable but gen- inclusions of base metal sulphides and
erallylowandtypeBismoreabundantthanA. pyrrhotitewereobserved. Mostinclusions
Pyriteisstronglycorrodedandreplacedbybase appear to be not primary. They are inter-
metalsulphidesand,inplaces,appearsrounded. preted to represent internal replacements
as the metal bearing fluid migrated along
abundant fractures and fissures within the
4.1.5 Mineral paragenesis
cataclastic pyrite B. It needs to be kept in
Themineralparagenesisrelativetothedifferent mind that the formation of large ore bod-
pyritevarietiesobservedduringmicroscopicin- ies such as Elura is a consequence of a
vestigations is described. It is obvious that the prolongedandcontinuesprocesswherethe
sulphideorebodyhasbeenmodifiedduringde- hydrothermal system evolves over time,
formation. Many sulphide mineral phases, in characterised by changes in fluid temper-
particular galena, pyrrhotite, and chalcopyrite ature and chemistry. The formation of
and to a lesser extent sphalerite, were strongly pyrite B is interpreted to have occurred
remobilised. Incontrast,pyrite,asabrittlemin- in a period when the fluid temperature in-
76 |
ADE | 4.1 Microscopic observations
Hydrothermal stages
Mineral phase Pre-base metal stage(1) Main-base metal stage Post-base metal stage(2)
Pyrite type A
Pyrite type B
Pyrite type C
Pyrite type D
Arsenopyrite
Pyrrhotite
Sphalerite
Galena
Chalcopyrite
Argentian-tetrahedrite
Quartz
Ankerite
Siderite
Chlorite and sericite
Native silver
Marcasite
GGooeetthhiittee
(1)Commencment of hydrothermal activity at the site of ore formation via early prominent pyrite alteration.
(2)Fading of hydrothermal activity, deformational induced sulphide mobilisation and alteration via meteoric fluids.
Figure 4.15: Mineral paragenesis and hydrothermal stages of the Elura deposit are shown relative to the
differentpyritevarietiesobservedduringmicroscopicinvestigations.
creased and when base metal concentra- • Pyrite D is entirely free of inclusions,
tions began to increase. The main stage unfractured and commonly overgrowths
of base metal formation is dominated by pyrite varieties A and B. Pyrite D forma-
pyrrhotite. Thus, pyrite B formed pre- to tion is interpreted to have occurred after
earlybasemetalmineralisation. thebasemetaleventduringalaterfluidin-
flux event accompanied by the recrystalli-
sationofearlierpyrite.
• Minor pyrite C is overgrowing pyrite A
and B in pyrite dominated ore zones. It • Meteoric fluids that interacted with sul-
containsabundantinclusionsofbasemetal phides within the shallow supergene zone
sulphides. In pyrrhotite dominated ore, became acidic and caused transformation
pyrite C is rare and, in places, intimately ofpyrrhotitetomarcasite. Partialtransfor-
intergrown with magnetite. This pyrite mationisobservedevenwithinthedeepest
variety is post pyrite A and B and con- parts of the orebody. No primary marca-
temporaneous with base metal precipita- site has been noticed. Minor alteration of
tion. Due to its low abundance, pyrite C tetrahedritetonativesilverwithinprimary
formation is interpreted to have occurred sulphidemineralisation isalso likelyto be
during the latest stages of the main base caused by meteoric fluids that descended
metal formation event and prior to its ces- fromsurfacealongfracturezones.
sation. The presence of rare magnetite-
pyriteintergrowthscloselyassociatedwith • Arsenopyrite is associated with pyrite A
pyrrhotiteindicatesachangingfluidchem- andBaswellaswithbasemetalsulphides.
istry towards higher oxygen fugacity dur- Itformedduringtheearlypyritedominated
ingtheverylateststagesoforeformation. hydrothermal stage and during the main
77 |
ADE | 4.2 Electron microprobe analyses
4.2 Electron microprobe 4.2.1 Compositional variability of
analyses sulphides throughout the
Elura orebody
THE chemical compositions of major sul-
Sphalerite
phide mineral phases in various ore types
were determined using election microprobe Macroscopic and microscopic observations in-
analysis (EMPA). Major sulphide phases were dicate significant chemical variability with re-
analysed in 44 samples (see figure 4.1 for sam- specttoironcontentinsolidsolutionwithzinc.
ple locations). Seven additional samples were Abundant inclusions of several other mineral
choseninordertoincreasethedatasetforanal- phases (e.g. chalcopyrite, pyrrhotite, pyrite,
yses of galena. Chlorite composition was de- NSG, etc.) were encountered, potentially neg-
terminedon9samplesandusedforgeothermo- atively effecting analytical quality. All EMPA
metric calculations. All samples were polished results of sphalerite (total of 275) were scru-
thin sections with analyses performed at the tinised based on totals and stoichiometry, 252
Department of Applied Geosciences and Geo- consideredasvalid(summarisedintable4.2).
physics, University of Leoben. The selected Optical identifications were confirmed by
sampleswerefromlocationsthroughouttheen- EMPA. Iron content in sphalerite significantly
tire orebody in order to establish data coverage varies and ranges between 2.41 and 8.22wt%.
ofalloretypesandorezones,althoughprimar- High-iron sphalerite is common in the lower
ily focussing on the entire depth extent of the mineralisation whereas sphalerite from the up-
main lode zone (table 4.1; figure 4.1). Samples perpartoftheMLzoneandtheveinstylemin-
were also selected from the rather small north- eralisationfurtherdistanttotheorebodyischar-
ernmostz6pod(2)andfromthewesternminer- acterised by less iron substituting for zinc (fig-
alisation(2). ure4.17). Thethreedeepestsamplesweretaken
EMPA were conducted on a JEOL Super- fromdeepestzonesofbutperipheraltotheore-
probe JXA 8200 at 20kV acceleration voltage body,thusfeaturelowerironcontent.
and10nAprobecurrentforsulphidephasesand Consistently high iron content with a me-
15kV acceleration voltage at the same probe dian of approximately 7wt% is observed in
currentforchloriteanalyses. Theachievedmin- pyrrhotite-dominated ore types (figure 4.16).
imum, average and maximum lower detection The chemical composition of sphalerite con-
limits are summarised in appendix A on page tained in pyritic ore types as well as those in
332. Analyses are corrected via ZAF and PRZ
methods. Standards, peak position and corre-
No. of analyses
sponding element concentrations used for cali-
Po 72
brationaregivenintable4.9. Py 88
SiPo 29
9
SiPy 35
The entire data set of high quality EMPA re-
VEIN/MinA 28
8
sults, i.e. correct stoichiometry and total ele-
7
ment concentration, is presented in appendix A
onpage332. 6
5
4
Lithology ML z1 z2 z3 z4 z5 z6 W-min
Po 6 3 3 1 2 1 2 - 3
Py 11 1 1 2 2 1 - -
2
SiPo 3 1 - - 1 - - 1
SiPy 1 2 - 1 1 - - 1 1
Vein 3 - - - - - - - Po Py SiPo SiPy VEIN/MinA
Total 24 7 4 4 6 2 2 2 Ore type
Figure 4.16: Box and Whisker plot showing the
Table4.1: Numberofsamplestakenfromdifferent contentofironincorporatedinsphaleritecontained
orezonesandoretypes. indifferentoretypes.
79
eF
]%.tw[ |
ADE | 4.2 Electron microprobe analyses
thisstudyfocusesontheinvestigationandiden- Qualitative wavelength dispersive element
tificationofthesourceforBi. DiscreteBimin- mapping indicates that Bi is predominantly
eral phases are reported to occur in other de- evenly distributed within galena as shown (fig-
positsintheregion. ure 4.22; Sample NP549-1-B, SiPo ore type,
NopotentialBiphaseswereencountereddur- 9323mRL,MLzone). Thisdistributionandthe
ingmicroscopicobservations. Consideringthat similar atomic radius of Pb and Bi suggest its
Bi is enriched in the lead concentrate and the occurrence as solid solution rather than micro
fact that no other Bi mineral phases are identi- inclusions. However,fourtinyareaselevatedin
fied, the possibility of Bi enrichment in galena Bi are present, each around 1 to 3µm in size
seems self-evident. The concentration of Bi (figure4.22). Theseareascoincidewithenrich-
in galena, its distribution and spatial variability ments of either Ag or Sb, suggesting the pres-
throughouttheorebodywasoneoftheareasin- enceofsulfosaltinclusionsorintergrowths. Oc-
vestigated. currences of discrete Bi mineral phases were
investigated by element mapping of a larger
A total of 441 points were analysed, 407
section area of one sample (Sample NP910-2-
deemed to be of high analytical quality based
B, SiPy ore type, 9237mRL, ML zone). The
on stoichiometry and total element concentra-
sample was selected based on whole rock geo-
tions (table 4.3). Bismuth is above the lower
chemistry that showed high levels of Bi, i.e.
detection limit in 85% of all analyses at a me-
171ppm. AllareaselevatedinBiconcentration
dian concentration of 0.28wt% and as high as
coincidewithgalena(figure4.21). Nopotential
0.85wt%. A clear enrichment trend of Bi in-
Bi-phasesareobserved.
corporated in galena with increasing depth be-
lowsurfaceispresent(figure4.20). Thedeepest
twosamplesdonotfitthistrend. Thesesamples
arelowgradeore(MinAandVEIN)andarenot Po
10200 Py
directly part of the massive sulphide orebody.
SiPo
ThepronouncedvariabilityofBiconcentrations SiPy
VEIN/MinA
ingalena suggeststhattheseconcentrations are
10000
not artificial, i.e. caused by interference with
lead intensity peaks. Galena seems to contain
higherlevelsofBiinpyrrhotitic-dominatedore
types(figure4.19). 9800
No. of analyses 9600
Po 135
1.0 Py 136
SiPo 40
0.9 SiPy 26
VEIN/MinA 7
0.8 9400
0.7
0.6
0.5
9200
0.4
0.3
0.2
0.1 9000
0.0 0.2 0.4 0.6 0.8
Po Py SiPo SiPy VEIN/MinA Bi [wt.%]
Ore type Figure 4.20: Bismuth detected in galena via
Figure 4.19: Box and Whisker plot showing the EMPA and plotted as a function of depth below
content of Bi detected in galena of different ore surface. Plotted concentrations are average values
types. ofindividualsamples.
81
iB
]%.tw[
]m[
LR
eniM |
ADE | 4.2 Electron microprobe analyses
Chalcopyrite Major Traces and impurities
n = 187 Cu Fe S Mn Co Ni Zn As Se Ag Cd In Sn Sb Te Au Hg Pb Bi
Max [wt%] 35.25 30.95 35.62 0.09 0.10 0.10 2.78 0.20 0.11 0.28 0.06 0.04 0.09 0.11 0.06 0.21 0.42 0.38 0.14
Mean [wt%] 34.53 29.92 34.78 0.05 0.04 0.04 0.31 0.08 0.06 0.06 0.04 0.03 0.05 0.04 0.04 0.08 0.21 0.14 0.06
Median [wt%] 34.59 29.88 34.83 0.05 0.04 0.03 0.10 0.07 0.05 0.05 0.04 0.03 0.05 0.04 0.05 0.08 0.20 0.12 0.06
Min [wt%] 32.71 28.66 33.91 0.03 0.02 0.02 0.03 0.03 0.04 0.02 0.03 0.02 0.04 0.02 0.02 0.04 0.11 0.06 0.04
n analysed [%] 100 100 100 40 100 83 100 100 83 100 100 40 42 100 83 83 57 100 100
n > LDL [%] 100 100 100 17 92 20 81 38 10 47 14 11 30 16 4 21 37 17 43
Avg. LDL [ppm] 324 235 220 312 199 233 340 328 382 280 605 161 341 381 313 451 830 416 361
Table 4.4: SummaryofchalcopyritecompositiondeterminedviaEMPA.
(a) (b)
(c)
Figure 4.22: Traceelementmappingingalena;SampleNP549-1-B(SiPo,9323mRL,MLzone);(a)
BSEimage,(b)reflectedlightmicrophotographand(c)wavelenght-dispersive-spectra(WDS)element
mapping (c); minor tiny concentration highs of Bi were identified, some of them correspond to Ag
and/or Sb and are likely to be caused by micro impurities of argentian tetrahedrite; concentration
levels of Cd, Te and Hg are indicated to be homogeneously enriched throughout the galena grain
without any apparent zonation; weak enrichments of Cd and Hg as well as depletion of Te coincide
withgrindingdamageofthethinsections’surfaceandareartefacts.
83 |
ADE | 4.2 Electron microprobe analyses
Po Po
10200 Py 10200 Py
SiPo SiPo
SiPy SiPy
VEIN/MinA VEIN/MinA
10000 10000
9800 9800
9600 9600
9400 9400
9200 9200
9000 9000
0.05 0.10 0.15 0.20 0.05 0.10 0.15 0.20
Au [wt.%] Ag [wt.%]
Figure 4.24: Gold detected in chalcopyrite via Figure 4.25: Silver detected in pyrite via EMPA
EMPA and plotted as a function of depth below and plotted as a function of depth below surface.
surface. Plotted concentrations are average values Plotted concentrations are average values of indi-
ofindividualsamples. vidualsamples.
Iron sulphides, arsenopyrite & mag-
Figure 4.23 shows the variability of atomic ra-
netite
tios between Co, Ni and As for pyrite type A
(red),typeB(green)andundifferentiatedpyrite
Pyrite
analyses (grey). No clustering of type A and B
Fourpyritegenerationswereidentifiedfrommi- is indicated. Cobalt concentrations were above
croscopic investigations. Pyrite types A and C thelowerdetectionlimit(avg.LDL190ppm)in
are generally fine grained, very impure or in- all pyrite analyses at a median concentration of
timately and finely intergrown with other sul- 0.07wt% and as high as 0.26wt%. Nickel is
phide mineral phases. EMPA results of high present at a median concentration of 0.04wt%
qualityweredifficulttoachieveforthesepyrite based on 27% of analyses above the detection
varieties. PyriteBiscoarsegrained,commonly limit. Pyrite of all ore types contain arsenic at
intrudedandreplacedbyothersulphidephases. rather constant median concentrations ranging
Post-basemetalpyriteDwasnotinvestigatedin between0.07and0.1wt%. Arseniclevelsfrom
detailaspartofthisstudy. pyritic massive ore samples, however, feature
a significant greater variability and maximum
No compositional changes related to differ-
concentration of 2.14wt% as shown in figure
ent pyrite types could be identified (table 4.5)
4.27.
from191highqualityanalysesoutoftotaldata
set of 265. Most of the rejected 28% of analy- Silver and Au are observed at median con-
ses are from impure and fine grained pyrite A. centrations of 0.05 and 0.06wt%, respectively.
85
]m[
LR
eniM
]m[
LR
eniM |
ADE | 4.2 Electron microprobe analyses
Their erratic concentrations (20% above LDL contrast,nickelisonlypresentinapproximately
forAgand24%forAu)indicatetheoccurrence a quarter of all results at lower concentrations
as micro-inclusions. Silver concentrations (median0.03wt%,maximum0.07wt%). Ame-
tend to be slightly elevated in pyrite from dianAsconcentrationof0.07wt%isdefinedby
siliceous semi-massive ore (SiPo and SiPy) 53% of all analyses above the detection limit
(figure 4.28). A weak Ag enrichment trend (averagedetectionlimit348ppm).
towards the surface is present (figure 4.25). GoldandAgwererarelydetectedatconcen-
Maximum concentrations for Ag and Au were trations up to 0.17 and 0.10wt%, respectively.
at 0.19wt% and 0.20wt%, respectively. Bis- Due to the erratic nature of their occurrence,
muth is detected in almost 60% of all analyses these elements are likely to be linked to mi-
at a median concentration of 0.07wt%. High cro inclusions. Bismuth is more common and
Bi concentrations preferably occur in pyrite present in above 50% of the analyses at a me-
contained in pyrrhotitic ore types (figure 4.29). dianconcentrationof0.07wt%. Thedetermina-
Other element concentrations are sporadic and tionofHginpyrrhotiteshowedaconcentration
arelikelytobecausedbymicro-inclusions. range between 0.12 and 0.36wt%. Some 31%
of the analyses are above the average detection
limitof899ppm.
Elements such as Mn, Se, Cd, Sn, Sb and Te
Pyrrhotite
weredetectedatlowconcentrationsclosetothe
Pyrrhotite is the main iron sulphide phase as- detectionlimit. Thepronouncedfluctuatingna-
sociated with base metal sulphides. It is com- tureoftheseelementsandbasemetals(i.e. Zn,
monly free of inclusions and other impurities Pb, Cu) detected in pyrrhotite is indicative of
butmaybeintimatelyintergrownwithothersul- being caused by micro inclusion of base metal
phidephases. Mostofanalysesareofgoodan- sulphides rather than lattice substitution. Nei-
alytical quality with 142 results selected out of ther, correlation trends with depth nor affinities
atotalof152(table4.6). Itisbeyondthescope to particular ore types are identified for any of
of this thesis to study the stoichiometry and theanalysedtraceelements.
crystallographyofpyrrhotiteattheEluraMine.
Therefore a general overview is given in the
following. The calculated median stoichiome-
try is Fe S (n=142) with an Fe-deficiency x
0.83
equal to 0.17 (average x=0.16) (figure 4.30).
This deviation is described by Anthony et al.
(1990)asthemaximumFe-deficiencyobserved
in naturally-occurring pyrrhotite. Although
other sources, e.g. Back et al. (2008), sug-
gests0.2asamaximum. Itisknownfromfield
observations that pyrrhotite is commonly mag-
netic, a feature of monoclinic pyrrhotite Fe S
7 8
(x=0.125). Despite the indication of two Fe-
deficienctpopulations(figure4.30), nocorrela-
tion with depth nor an affinity to particular ore
typesororezoneswereidentified.
Figure 4.30: BoxandWhiskerplot
Cobalt was identified as the most abundant
showingtheiron-deficiencyrangeob-
trace element incorporated in pyrrhotite. All servedinpyrrhotite.
analyses are above the average detection limit
of188ppm. Themedianandmaximumconcen-
trationsare0.09and0.17wt%,respectively. By
87 |
ADE | 4.2 Electron microprobe analyses
Arsenopyrite Major Traces and impurities
n = 84 Fe As S Co Ni Sb Au Cu Zn Ag Cd Pb Bi
Max [wt%] 35.51 46.65 23.53 2.27 0.05 5.36 0.23 0.09 2.49 0.07 0.07 0.32 0.12
Mean [wt%] 33.92 44.29 20.32 0.18 0.04 0.92 0.08 0.05 0.39 0.05 0.04 0.11 0.06
Median [wt%] 33.94 44.60 20.14 0.07 0.03 0.67 0.07 0.05 0.26 0.05 0.04 0.08 0.06
Min [wt%] 32.23 38.21 18.23 0.03 0.03 0.03 0.04 0.03 0.04 0.04 0.03 0.05 0.03
n analysed [%] 100 100 100 100 87 100 87 100 100 100 100 100 100
n > LDL [%] 100 100 100 98 22 98 37 18 79 13 31 15 21
Avg. LDL [ppm] 249 628 226 203 253 422 411 349 345 318 453 381 351
Table 4.7: SummaryofarsenopyritecompositiondeterminedviaEMPA.
Figure 4.33: The wave length dispersive spectra (WDS) elemental mapping identified elevated antimony
and cobalt concentrations defining the growth zonation whereas the distributions of sulphur and arsenic are
responsibleforchangesintheBSEintensitydefiningthesecondzonationpattern.
89 |
ADE | 4.2 Electron microprobe analyses
4.2.2 Silver phases – their classifi- The phase is currently solely defined based on
cation and alteration mineral chemistry. Crystallographic character-
istics based on X-ray powder pattern are yet
Background
to be determined, subsequently its approval by
Tetrahedriteandtennantitesulfosaltshavebeen theInternationalMineralogicalAssociationasa
described as the most common Ag-bearing discretemineralspeciesisunsettled.
mineral phases identified in the Elura ore- The mineral phase stylotypite with the for-
body (e.g. Lawrie and Hinman, 1998; Taylor mula
etal.,1984;Leverettetal.,2005;David,2008). (Ag,Cu,Fe,Zn) SbS
3 3
These phases are particularly enriched in up-
per and peripheral mineralised zones within was first described and named by Franz Ko-
the orebody. The chemical variability of Ag- bell in 1865 and was accepted as a discrete
phases is investigated via EMP as part of this mineral phase until 1952. Milton and Axel-
study, mainly focussing on minerals part of rod(1951)reviewedpublisheddataofstylotyp-
the tetrahedrite-tennantite solid-solution series. ite and showed that all samples are either tetra-
The standards and peak positions used for cali- hedrite or a mixture of tetrahedrite with other
brationaregivenintable4.10. known mineral phases (e.g. jamesonite). He
Thesolid-solutionseriesisdefinedbytheend concludedthatstylotypiteasanindividualmin-
memberstetrahedrite(Cu Sb S )andtennan- eralspeciesshouldbediscredited.
12 4 13
tite (Cu As S ), with Sb and As featuring a
12 4 13
complete mutual lattice substitution. Johnson
Analytical results
etal.(1986)definedageneralformulabasedon
analyses of a large number of natural and syn- Throughout most of the orebody, sulfosalt
thetic tetrahedrite samples (eq. 4.1). This for- phases occur as small grains, in the order of
mula implies a maximum of six Ag atoms per a few microns, but occur as larger grains to-
formula unit may be incorporated in the struc- wardstheuppermostandperipheralareas. Sam-
ture. However, tetrahedrite is commonly non- ples from those areas were preferably selected
stoichiometric and rarely fits the ideal mineral for EMPA. In order to achieve good analytical
formula(e.g.Johnsonetal.,1988). results, only grains of at least 7 to 10µm in
Freibergite is a Ag-rich sulfosalt species size were analysed. A total of 97 EMPA anal-
chemicallysimilartotetrahedrite. Itsstatusasa yses of sulfosalt and related alteration phases
discretemineralphaseis,however,stilldisputed were undertaken on twelve samples. Five anal-
(Moeloetal.,2008). Themaincharacteristicof yses were rejected due to poor analytical qual-
freibergite is the decrease of S with increasing ity. Figure4.36showsthecalculatedAg/(Ag+Cu)
substitution of Ag for Cu above approximately and Zn/(Zn+Fe+Hg+Cd) versus Sb/(Sb+As+Bi+Te) ra-
4apfu (Moelo et al., 2008). The general for- tios. The observed Sb/(Sb+As+Bi+Te) values are
mulaisdefinedineq. 4.2. Thecrystalstructural ratherconstantandrangebetween0.95and0.98
formulaformaximumAgcontentis with a median of 0.97. The occurrence of
tennantite or other arsenic-dominated sulfosalt
Ag Cu (Fe,Zn) Sb S .
6 4.44 1.56 4 12.09 mineralphaseswasnotverifiedinthecourseof
Mineral phases that contain less than 4apfu
this study. The Ag/(Ag+Cu) values vary signifi-
cantly between 0.20 and 0.60 with a median of
AgshouldbenamedAg-richtetrahedriterather
0.35. A weak inverse dependency between Zn
thanfreibergite(Moeloetal.,2008).
Argentotetrahedrite is suggested as the Sb-
and Sb is present. Overall, the Zn/(Zn+Fe+Hg+Cd)
ratio ranges between 0.05 and 0.34 with a me-
richanaloguetoargentotennantitewiththeideal
dianof0.21.
formula
Commonly, the mineral formula for tetra-
Ag (Fe,Zn) Sb S . hedriteiscalculatedbasedonSequalto13apfu.
10 2 4 13
93 |
ADE | 4.2 Electron microprobe analyses
4.2.3 Chlorite chemistry with all Fe assumed to be ferrous. Most
calculated compositions feature negative va-
Ore mineralogy varies as a function of depth cancies ranging between -0.44 and 0.14apfu
within the sulphide mineralisation at the Elura (table 4.15). Nine chlorite analyses in three
orebody. Thisvariabilityisreflectedbyseveral samples(table4.15)arecharacterisedbyatotal
elementconcentrationtrendscausedbychanges cation occupancy in the octahedral position
in modal mineralogy and mineral chemistry. of less than 12.1apfu and fall within the
These trends are also observed in whole rock commonly observed vacancy range (de Caritat
geochemistry (see chapter 5 on page 139). et al., 1993). If vacancies range between 0
These changes may be linked to vertical tem- and 2apfu, chlorite is tri-trioctahedral (Zane
perature gradients within the depositional en- and Weiss, 1998). The over-occupancy of oc-
vironment. Chlorite compositions were deter- tahedralcoordinatedcationsiseithercausedby:
mined on eight thin-section samples from dif-
ferent depth levels of the orebody. Different (a) ferric iron not considered during calcu-
chloritegeothermometerswereusedinorderto lation because all Fe is considered as ferrous
estimateformationtemperatures. (deCaritatetal.,1993),or
A total of 67 chlorite analyses were per-
formed via EMPA (summarised in table 4.15). (b) chlorites are sub-microscopically inter-
Used mineral standards and conditions are grown with other alumino-silicate mineral
showen in table 4.14. The total element con- phases which lack Na, K and Ca (e.g. vermi-
centrationsrangebetween83.37and91.56wt% culite).
with a median of 89.39wt%. These values
are consistent with typical chlorite composi- Chlorite contains minor Zn at a median con-
tions (e.g. Deer et al., 1992). The analy- centration of 0.28wt%, low concentrations of
ses were checked for Na, K and Ca con- Mn(medianof0.07wt%)andFe/(Fe+Mg)values
(cid:80)
centrations ( Na,K,Ca) in order to investi- ranging between 0.75 and 1.00. Classification
gate sub-microscopic intergrowths with other of chlorite was performed via the scheme de-
silicates (e.g. muscovite, feldspar or clay fined by Hey (1954). Most chlorite composi-
mineral phases). The maximum combined tionsfallwithintheripidolitefield,somecross-
atomic percentage of these elements is very ingoverintothedaphniteandthepseudothurin-
low at 0.03at%, equal to 0.1wt%. More gitefields(figure4.39). Theoveroccupancyof
than 70% of chlorite analyses fall below theoctahedralpositionmaybeindicativeoffer-
(cid:80)
Na,K,Ca=0.01at% indicating the presence ricironincorporatedinthelattice, andthusox-
ofpurechlorite. idation. If these chlorite varieties are oxidised
Chlorite compositions vary significantly and theycanbeclassifiedasthuringite(Hey,1954).
are controlled by fluid conditions such as f ,
O2
pH,temperature,Fe/(Fe/Mg),andthecomposition Chlorite geothermometer
of the host rock (de Caritat et al., 1993). Ele-
Severaldifferentchloritegeothermometershave
ments in both octahedral and tetrahedral posi-
been proposed in the past. Most of them are
tions are affected by these compositional vari-
empirically determined and based on the tem-
abilities. Chloritecommonlyhasvacancies((cid:114))
perature dependent changes in aluminium con-
attheoctahedralsitetypicallyrangingbetween
tent at the tetrahedral position and octahedral
-0.1 and 1.1 based on O (OH) (de Caritat
20 16
vacancy changes (e.g. Cathelineau and Nieva,
et al., 1993). The generalised mineral formula
1985; Cathelineau, 1988; Hillier and Velde,
forchloriteisgivenineq. 4.9(ZaneandWeiss,
1991). Three groups can be defined according
1998).
todifferentapproaches.
Chlorite stoichiometry was calculated based
on a total of 28 anhydrous oxygen atoms
101 |
ADE | 4.2 Electron microprobe analyses
4.13 (abbrev. GT-2), Kranidiotis and MacLean alising system, and to investigate temperature
(1987) given in eq. 4.14 (abbrev. GT-3) and trends versus depth below surface. The tem-
ZangandFyfe(1995)givenineq. 4.15(abbrev. perature estimates range between 314°C and
GT-4)wereusedtocalculatechloriteformation 343°C, based on the 25 and 75% percentiles.
temperaturesinthecourseofthisstudy. The median is calculated at 328°C. These val-
Temperature estimates for the individual ues are within the temperature range proposed
geothermometers vary (figure 4.40). Geother- for the main base metal precipitating fluid as
mometer GT-1, GT-2 and GT-3 give similar re- suggestedbasedonfluidinclusionstudies. Con-
sults for individual chlorite analyses with an sequently, it can be concluded that chlorite
average standard deviation of 14.9°C and me- geothermometry results give correct tempera-
dian values of 398, 388 and 386°C, respec- tureestimates.
tively. Temperatures calculated via GT-4 are A systematic temperature change with in-
consistentlylowerwithanaveragedifferenceof creasing depth is recorded for samples taken
117°C(σ=14°C)andamedianof271°C. from the lower mineralisation (figure 4.41).
GT-3 and GT-4 estimates are characterised Temperature estimates based on chlorite anal-
by the lowest standard deviations of analyses yses of one sample from the upper mineralisa-
within individual samples at 12.6 and 11.3°C, tion is at the higher end of the estimated range.
respectively, and will therefore be preferred A clear drop is observed in the lower min-
over GT-1 and GT-2. The maximum temper- eralisation. Maximum temperature is around
ature of GT-3 (419°C) and the minimum tem- 9700mRL, it then decreases with increasing
perature of GT-4 (209°C) are commensurate depth, before increasing again up to maximum
with fluid inclusion homogenisation tempera- temperatures in the lowermost parts of the ore-
turesfromearlierstudies(seechapter3onpage body. This trend roughly correlates with a
50). The main difference between both GT-3 changing host lithology as described in chap-
and GT-4 is the power of the Fe/(Fe+Mg) correc- ter 3 on page 35. Most chlorite geothermome-
tion term. Thus, these estimates may repre-
sent extreme values. Average values were cal- 1100220000
culated on the basis of those two GTs, in or-
1100110000
dertoachievethemostlikelytemperaturerange
of chlorite crystallisation for the entire miner- 1100000000
99990000
99880000
99770000
99660000
99550000
9400
9300
9200
9100
Figure 4.40: Box and Whisker plot comparing Mean Temp. (GT-3/4)
temperature estimates of different geothermome- [°C]
ters; geothermometric calculatations are given in
eq. 4.10 for GT-1, eq. 4.11 for GT-2, eq. 4.12 for Figure 4.41: Chloritetemperaturevariabilityvs.
GT-3andeq. 4.13forGT-4. depth.
105
]]mm[[
LLRR
eenniiMM
002 052 003 053 004 |
ADE | 4. ORE MINERALOGY AND PETROGRAPHY
tersareempiricallybased. Itiswellknownthat but pyrite is present within the host rocks and
chlorite composition can be sensitive to many is a major mineral within the orebody. The
factors, not solely to temperature (e.g. de Car- ore-forming fluid certainly would have inter-
itat et al., 1993). Changes in fluid conditions actedwiththewallrockduringascentandwith
(e.g. f ,pH,temperature)and/orhostlithology pre-enriched pyrite at the site of ore precipi-
O2
(e.g. Fe/Mg values) are amongst those factors. tation. Thus it can be assumed that the a
FeS
Temperaturevaluesshouldalwaysbetakenwith of the ore forming fluid is at or close to the
cautionandonlyusedincombinationwithalter- pyrrhotite-pyrite buffer assemblage. Therefore,
nativemethods. Someauthorsevenrefusechlo- sphaleritewasinequilibriumwithbothironsul-
ritegeothermometryentirely. However,temper- phides at the time of precipitation. However,
ature estimates calculated in the course of this re-equilibration is highly probable to have oc-
studyareinagreementwithresultspresentedin curred during the transformation of hexagonal
other studies using different methods, e.g fluid to monoclinic pyrrhotite. During deformation
inclusions studies. The calculated temperature oftheorebody,remobilisationandrecrystallisa-
dependency as a function of depth is related to tion of sulphides occurred. Ore textures such
a change in host lithology. This change may as sulphide banding clearly indicate disequi-
reflect either changing crystallisation tempera- librium of base metal sulphides with iron sul-
tures, changes in fluid characteristics or is in- phides. Consideringtheseobservations,itisun-
ducedbythechangingchemicalcompositionof likelythattheapplicationofthesphaleritegeo-
thehostlithology. barometer will result in correct pressure esti-
mates. Nevertheless,anobviousandsystematic
change of iron content in sphalerite is present
4.2.4 Application of the sphalerite
throughout the orebody that needs explanation
geobarometer
(section4.2.1onpage79).
Pressurewascalculatedviatheequation
There is a large literature on sphalerite in-
corporating Fe in the lattice structure and its
P = 42.30−log(mole%FeS )
Sph
dependence on temperature, sulphur fugacity
and pressure (e.g. Barton and Toulmin, 1966; (HutchisonandScott,1981)forsphaleritecom-
ScottandBarnes,1971;ScottandKissin,1973; positions as described in section 4.2.1 on page
Hutchison and Scott, 1981). If the activity of 79(completesphaleritedatasetisshowninap-
FeS (a ) is buffered by sphalerite coexisting pendix A section A.4.2). The results range be-
FeS
with pyrite and hexagonal pyrrhotite, tempera- tween 6.6 and 23.5kbar, and are, as expected,
ture dependency is negligible at up to approxi- fartoohighastobeplausible. Despitetheabso-
mately 530°C. Under these circumstances, the lute pressure values being wrong, their relative
Fe content in sphalerite is strongly dependant changesinrespecttooretypesanddepthwithin
ontheeffectivepressureduringmineralgrowth, the orebody seem significant and are thus dis-
enablingitsutilisationforgeobarometriccalcu- cussedinthefollowing.
lations. Figure 4.42 shows pressure estimates
Microscopicobservationsandmineralchem- and their variations for different ore types.
ical investigations as part of this study clearly Pyrrhotitic dominated ore types (Po and SiPo)
showed that: (1) sphalerite and pyrrhotite are show rather uniform values at a combined
co-genetic; (2) pyrite varieties occur through- median of approximately 8.8kbar. Results for
out the paragenetic sequence and in the wall other ore types feature a significant variation
rock sequence; (3) initial hexagonal pyrrhotite with a combined median of 14.9kbar. The
was retrogressed to the monoclinic polymorph; pronouncedspreadinpressureismostlycaused
and(4)mostoftheorebodyhasbeendeformed bysphaleriteanalysesfromtheupperapophysis
at varying intensities. Pyrrhotite is the most of the orebody in pyritic ore (Py and SiPy),
abundant syn-base metal iron sulphide species featuring the lowest Fe content of all analyses.
106 |
ADE | 4.2 Electron microprobe analyses
By plotting calculated pressures vs. depth
within the orebody, two groups were defined
(figure4.43). Thepressurevaluesforthosetwo
26
groups are approximately 19.1 and 8.9kbar.
24
The latter value is close to the median of 22
pyrrhotite-richore. 20
Pressure calculations based on the sphalerite 18
16
geobarometer were unsuccessful and resulted
14
in far too high and meaningless pressure es-
12
timates, confirming microscopic observations
10
of ore texture, indicating that sphalerite and 8
pyrrhotite are not in equilibrium with pyrite. 6
4
The observed variability of iron incorporated Po Py SiPo SiPy VEIN/MinA
in sphalerite may to some extent be controlled Ore type
Figure 4.42: Box and Whisker plot showing the
bydeformational-inducedre-crystallisationand
calculatedpressureestimatesfordifferentoretypes.
subsequent re-equilibration with adjacent iron
sulphidephases.
Twoironpopulationswereidentifiedinmas-
sive pyritic ore with the high-Fe one featur-
ing similar concentrations as sphalerite in mas-
sive pyrrhotitic ore (see Figure 4.42; pressure
1100220000 SSiiPPyy
and Fe content in sphalerite feature an inverse UUppppeerr
SSiiPPoo
relation). Some ore zones now classified as 1100110000
PPyy
pyriticwouldneedto,ingeneticterms,beclas-
sified as pyrrhotitic ore. This is because initial 1100000000 PPoo
pyrrhotite has been altered to marcasite. Such
99990000
mis-classified ore zones (in genetic terms) may
containhigh-Fesphaleritetypicalforpyrrhotite 99880000
dominated ore types, causing the two Fe pop-
ulations. Low Fe-sphalerite may have formed 99770000
exclusively via replacement of pyrite in zone
99660000
wherepyrrhotiteisabsent.
But neither the deformational mechanism,
9500
nor the mis-classification of ore types suffi-
ciently explains the systematic shift from low- 9400 Lower
Fe sphalerite in the upper main lode zone to mineralisation
9300
consistent high-Fe sphalerite in lower zones of
the orebody (see figure 4.43; pressure and iron
9200
contentinsphaleritefeatureaninverserelation),
5 10 15 20 25
independent of whether pyrite or pyrrhotite is Pressure [kbar]
the prevailing iron sulphide species. The only
Figure 4.43: Pressure estimates as a function of
feasibleexplanationseemsachangeinthefluid
depth. Plotted data are median values sphalerite
chemistrythatprecipitatedsphaleriteinthefirst analysesofindividualsamples,errorbarsarebased
on minimum and maximum values; sphalerite in
place as the hydrothermal system evolved. The
samples from the upper mineralisation feature a
precipitation of significant pyrrhotite from the shifttohigherpressureestimatescomparedtothose
hydrothermalfluidmayhavecausedadecrease fromthelowermineralisation.
of its FeS activity, subsequently less FeS is
availablefortheincorporationwithsphalerite.
107
]]mm[[
LLRR
eenniiMM
erusserP
]rabk[ |
ADE | 4. ORE MINERALOGY AND PETROGRAPHY
4.3 Trace element
composition derived
from LA-ICP-MS
MAJORandminorelementcompositionhas
been investigated via electron micro-
probe analysis (EMPA) for all important sul-
phidemineralphasesoccurringintheEluraore-
body. Severalsignificantelementaltrendswere
identified (see section 4.2.1 on page 79). In or-
der to validate these trends and to further in-
vestigate trace element concentrations in spha-
lerite, galena, pyrite and pyrrhotite down to
concentration levels not detectable via EMPA,
Laser Ablation Inductively Coupled Plasma
Mass Spectrometry (LA-ICP-MS) was used. It
is known from literature, that the above min-
eralphasesarecapableofincorporatingadiver-
sity of elements (e.g. Blackburn and Schwen-
deman, 1977; Foord and Shawe, 1989; Sharp
Figure 4.44: Sample locations for LA-ICP-MS
and Buseck, 1993; Huston et al., 1995; Cook
study; samples cover most of the depth extent of
etal.,2009;Largeetal.,2009;Yeetal.,2011). theEluraorebody;sampleswereselectedprimarily
The elements are either incorporated via stoi- focussingelevatedbismuthconcentrationsobserved
viaEMPAorwholerockgeochemistry. Numberof
chiometric or non-stoichiometric lattice substi- samples is 11. Longitudinal view towards WSW
tution or as evenly distributed micro impurities (˜245°). Yellow shape is the stringer type ore
(VEIN)resourcedomain.
atnanometrescale.
The main focus in the course of this study
is to identify the mineral hosts for important wereselectedfortraceelementanalysisofspha-
smelter penalty elements (e.g. Bi, Hg, As, lerite, galena, pyrite and pyrrhotite. Pyrrhotite
Mn, etc.), the potential occurrences of uniden- is present in three of the samples. Galena was
tifiedbeneficialelements(e.g. Ininsphalerite), analysed in two additional samples. These two
and to investigate the amount of Ag incorpo- sampleswereselectedbasedonhighBiconcen-
rated in sulphides other than sulfosalt phases, trationsidentifiedviawholerockgeochemistry.
in particular in galena. Where significant el- Thesamplesarepreparedasuncoveredpolished
∗
ement enrichments were identified, an attempt thicksections .
was made to find evidence of the nature and
mechanisms of element incorporation. Trace
4.3.1 Analytical background
element concentrations as well as their spatial
variability throughout the mineralisation may Trace element analyses were carried out us-
indicate changes in physio-chemical fluid char- ing the LA-ICP-MS system at Adelaide Mi-
acteristicsormetalsource. croscopy, University of Adelaide. The instru-
mentsetupconsistsofahighperformanceNew
Investigatedsamplesroughlycoverthedepth
WaveNd:Yag213nmUVlaserablationsystem
extentoftheEluraorebody(figure4.44). Sam-
attached to an Agilent 7500cs quadrupole ICP-
ple selection focussed primarily on elevated Bi
concentrations observed via EMPA or whole
∗.Sample preparation was carried out at the Depart-
rock geochemistry (see section 4.2.1 on page
ment of Applied Geosciences and Geophysics; chair of
79 and chapter 5 on page 149). Nine samples ResourceMineralogy;UniversityofLeoben,Austria.
108 |
ADE | 4. ORE MINERALOGY AND PETROGRAPHY
4.3.2 Overview of analytical Most Cu is likely associated with chalcopyrite.
results Goldisnotclearlylinkedtoanymineralphase,
althoughrelativecloserelationtoFesuggesting
Time-resolved intensity profiles of a total of
pyriteasthemostprobablehostmineralphase.
156 analyses were scrutinised for potential im-
Severalcorrelationsandtrendsarepresentfor
purities and 26 LA-ICP-MS analyses rejected
element concentration and counts per second
due to heavy contamination with other mineral
(cps) intensity data and are presented as cor-
phases. Element concentrations, 1-σ errors
relation matrices and as time-resolved intensity
and lower detection limits of the remaining
profile diagrams in appendix A on page 379.
data set of 130 analyses in addition to their
Bycomparingcpsintensitytrendsoftheinves-
corresponding time-resolved intensity profiles
tigated mineral phases (figure 4.57), potential
aregiveninappendixAonpage379.
trace element contaminations caused by micro-
impurities may be identified, and will be dis-
∗
Clusteranalysis ofcountspersecondintensity
cussed for individual mineral phases in the fol-
data of sphalerite, galena and pyrite was used
lowing.
to investigate trace element affinities to certain
Vanadium and Cr were consistently detected
mineral phases (see figure 4.45). The elements
atlowconcentrationsintheorderofafewparts
Sb,Ag,Tl,Sn,Se,BiandTearecloselyrelated
per million in all investigated mineral phases.
toPb,thusindicatinggalenaastheirhostphase.
Due to the insignificance of these elements for
Hg,Cd,In,GaandMnarelinkedtoZnsuggest-
this study, their very low concentrations, in ad-
ingincorporationinsphalerite. Elementsfeatur-
dition to the lack of data and studies available
ing affinities to Fe are Co, Ni, and As, indica-
in literature, their occurrence is not further dis-
tive for their association with pyrite, or to po-
cussed.
tentialimpuritiesofarsenopyrite. Accordingto
theclusteranalysis,CuappearslinkedtoAsand
4.3.3 Trace elements in sphalerite
Fe,butitsrealnatureofoccurrenceisuncertain.
There is a literature and comprehensive data
availableontraceelementsinsphaleriteandthis
has been comprehensively compiled in Cook
et al. (2009). Sphalerite readily incorporates a
significantnumberofelementsinsolidsolution.
Undoubtedly, the three most common and best
investigated ones are iron, manganese and cad-
mium. Thesetwoelementsareincorporatedvia
stoichiometric lattice substitution in the man-
ner:
Cd2+ (cid:10) Zn2+
orgeneralised
M2+ (cid:10) Zn2+
where M is predominantly Mn, Fe, Co and Cd.
MinorquantitiesofCumaybeincorporatedvia
thismechanismandpotentiallyNi(Cooketal.,
Figure 4.45: Dendrogram showing trace element
affinities to certain mineral phases based on clus- 2009). Other mono-, di-, tri- and tetravalent
ter analysis∗. Elements indicated to be related to
galena are framed red, those related to sphalerite ∗.Spearman’s correlation coefficients are used as at-
framed blue, and orange for pyrite. Dashed green tribute distance measure between elements. Cluster
framegroupselementswithoutclearrelation. analysisisperformedaftertheapproachofWard.
110 |
ADE | 4.3 Trace element composition derived from LA-ICP-MS
ThemechanismsofTl,HgandZnincorpora- able with concentrations ranging between 332
tionarenotdescribedintheliterature. Thallium to2,339ppmand94to2,385ppm,respectively.
incorporation is likely to be established similar AconversetrendcomparedtoBiisobservedfor
to the aforementioned coupled substitution re- Sb with decreasing concentrations with depth,
action along with Ag and Cu, or with Bi, Sb whereas Ag appears to be enriched towards
and As. That is because Tl may feature mono- the upper and the lower parts of the mineral-
or trivalent charge. Divalent Tl does not oc-
cur in nature. Zinc and Hg are most probably
directly substituted for Pb in solid solution, al-
though Hg, also occurring in monovalent state,
mayalsotakepartinthecouplesubstitutionre-
action along with Ag or Cu. Selenium substi-
tutes for S within the solid solution series with
theend-membersclausthalite(PbSe). Nosolid
solutionseriesexistsbetweengalenaandaltaite
(PbTe).
Several elements consistently feature
plateau-like time-resolved intensity profiles in
galena (example shown in figure 4.50). Those
elements are Se, Ag, Cd, In, Sn, Sb, Te, Tl and
Bi. The average trace element concentrations
calculated on the basis of representative analy-
ses of eleven samples are summarised in table
4.18.
The most enriched elements are Bi, Sb and
Ag. Bismuth concentrations vary significantly
and range between 0.05 and 5,645ppm. A
significant increase in Bi concentration is no-
ticed towards deeper parts of the orebody (fig-
ure 4.48). Silver and Sb are not quite as vari-
Figure 4.48: Compositional variability of galena vs. depth. Sele-
nium and tellurium concentrations increase with increasing depth. An
inverse trend is observed for antimony and thallium. Silver content is
elevated in galena contained in samples from deep as well as shallow
parts of the orebody. One sample miss-fits most trends (symbol filled
red). This stringer ore sample (NP950-3) was taken at some distance
almost 200m below the orebody, thus it may not immediately support
orfeaturetrendsobservedwithinthemainmineralisation.
115 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.