MDSG 2004 Annual Meeting

January 6th and 7th 2004 at the University of

Leeds
Scientific Programme - Abstracts
|| MDSG 2004 home page || Scientific Programme ||

A new model for copper mineralization in the Polish Kupferschiefer

D.H.M. Alderton 1 , D.J. Blundell 1 , P. H. Karnkowski 2 , and D.G. Pearson 3

1 Dept of Geology, Royal Holloway, University of London , Egham

([email protected])

2 Institute of Geology, University of Warsaw , Al. Zwirki i Wigury 93, 02-089

Warsaw , Poland

3 Dept of Earth Sciences, Science Laboratories, South Road, University of

Durham, Durham DH1 3LE

Previously proposed genetic models (e.g. Jowett, Cathles et al.) have been
unable to account for the 350 million tonnes of copper metal associated with the
Kupferschiefer shales in the southwest of the Polish Basin, without an
unreasonably high concentration of Cu in the fluid transport system. A new
conceptual model is presented, based in particular on basin modelling and a
more detailed understanding of the structural, stratigraphic and thermal evolution
of the Rotliegend Basin is now possible using borehole information, vitrinite
reflectance data and seismic records. These data have demonstrated that a
localized high heat flow anomaly was present at the time of the main phase of
mineralization in the early Triassic. Furthermore, deep growth faults, which would
have penetrated to the basement, were probably also active at the time and
these could have acted as important fluid conduits. Helium isotope ratios of
present-day natural gas in this region suggest a mantle origin for the gas,
supporting the deep nature of these structures.

The mechanism for mineralization proposed here involves injection of hot brine
from a deep-seated, basement fault-fracture system into Rotliegend eolian
sandstone, with up-dip flow through this porous formation to the Lubin area
where chemical conditions in the immediately overlying Kupferschiefer shales
were conducive to the precipitation of the metals from the brine. Geochemical
considerations, including Os isotopic data, indicate that metals were derived by
leaching of crustal rocks, possibly Autunian volcanics and/or Carboniferous
sediments. The copper mineralization of the Lubin area is suggested to have
resulted from a large number of short pulses of fluid, with flow triggered by
coseismic strain associated with normal faulting. It is proposed that, within the
basement, a network of fractures with high aspect ratios opened within each
interseismic period under lateral tension and filled with brine. At the time of a
normal fault-related earthquake, when fault rupture occurred, coseismic strain
resulted in the closure of fractures in the vicinity of the fault and brine was
expelled. Calculations suggest that, with fluid pressure generated by the
coseismic strain close to lithostatic, the force from hot brine injected into the
Rotliegend sandstone was sufficient to drive the brine through the aquifer at up to
63 meters per year to the site of copper deposition in the Kupferschiefer shale.
Flow from each earthquake lasting, on average, for just over a year would have
resulted in a discharge of approximately 0.07 km 3 of fluid. If earthquakes and
normal faulting occurred at 100–200 year intervals, averaging 7,000 per million
years, 6,000 km 3 of fluid could have been been mobilized in a period of 12
million years. Assuming that this was a brine containing 60 ppm dissolved
copper, this volume would have been sufficient to account for the observed
quantity of copper in the southwest of the Polish Basin.

Fluid Evolution of the Mt. Leyshon Au Deposit, Queensland, Australia

Murray M. Allan 1 , Bruce W. D. Yardley 1 , and Gregg W. Morrison 2

1 School of Earth Sciences, University of Leeds, Leeds, W. Yorkshire, LS2 9JT,

[email protected] , [email protected]

2 Klondike Exploration Services, 7 Mary St., Townsville, Q 4810, Australia,

[email protected]

In general, epithermal and many porphyry ore deposits are emplaced at shallow
crustal levels where pressure, temperature, and oxidation state gradients in the
hydrothermal fluid can trigger ore mineral precipitation. A dominant mechanism
to facilitate these physicochemical changes is the development of fracture
connectivity between the hydrothermal system and the surface. Not only can
repeated fracturing and sealing give rise to transient fluctuations in pressure (and
to a lesser extent, temperature), but it can also allow mixing between magmatic
and meteoric fluids. However, many intrusion-associated ore deposits with a
history of brecciation and hydro-fracturing show very little evidence for meteoric
infiltration during mineralisation. We present a fluid model based on geological,
fluid inclusion, and stable isotope data (Morrison and Andrew, 1989) for the
porphyry-related, breccia-hosted Mt. Leyshon Au deposit, Queensland (ca. 290
Ma). Despite significant vertical clast transport in breccia pipes, the development
of hydro-fracturing in host rocks, and a shallow inferred palaeodepth (< 3 km),
Mt. Leyshon appears to represent an “isolated” magmatic-hydrothermal system,
with little or no surface water input (Morrison and Orr, 2000). An ongoing fluid
inclusion study, including cathodoluminescence (CL) imaging of quartz textures
and quantitative laser ablation ICP-MS analysis, is aimed at better resolving the
Mt. Leyshon deposit's fluid evolution in the framework of the existing geological
model. In particular, this study will test the possibility of groundwater incursion,
better resolve the physicochemical fluid conditions during mineralisation, and
investigate the metal budget of magmatic fluids.

The geology of the Mt. Leyshon Complex (Figure 1) is characterised by a

cylindrical pipe of hydrothermal breccia (MPBX) intruding basement granite and
metasediments. This breccia body is itself intruded by later breccias and
concomitant porphyry stocks and dikes. Economic mineralisation is mainly
hosted by the late stage Mount Leyshon Breccia (MLBX), and is related to a dike
swarm (LD) propagating from the southwest corner of the complex (SP).
Alteration in the Mt. Leyshon hydrothermal system progressed from early chloritic
and biotite-magnetite alteration, affecting the early breccia (MPBX), to phyllic
alteration of porphyry stocks, dikes, and their breccia envelopes. Although early
alteration was diffuse compared to the more focussed alteration haloes of later
porphyry intrusions nested within the early breccia, it is largely confined within
fractured margins of the Mt. Leyshon complex (defined roughly by the extent of
the MPBX). There is no broad halo of phyllic alteration beyond the complex
margins associated with early hydrothermal activity, suggesting that intrusion-
driven groundwater convection was limited. The evolution of the Mt. Leyshon
system is therefore characterised by multiple-stage, explosive release of over-
pressured fluid, exsolved from a magma body at depth, producing a column of
well-milled breccia, and assisting the ascent of porphyry bodies and metal-rich
fluids.

A domical roof to the Mt. Leyshon complex is partially preserved at the present-
day erosional surface as a biotite-altered, variably fractured, but relatively
impermeable cap of country rock lithologies. This suggests that breccia pipes
themselves probably did not penetrate the palaeosurface, although the possibility
of vertically extensive fracture networks or a volcanic edifice remains. Economic
Au and Ag mineralisation is hosted mainly by a later stage breccia, in a high-
grade envelope confined below the complex roof in the Mount Leyshon Breccia
(MLBX). This has lead Morrison (Morrison and Orr, 2000) to suggest that
hydrothermal circulation of auriferous fluids exsolved from late dikes (LD) was
limited by a strong permeability contrast at the breccia-host rock interface.

Fluid inclusion data also support a magmatically dominated fluid regime. Fluid
inclusions have been studied from 3 paragenetically distinct vein and breccia
cavity infillings. Early pre-ore stage quartz-molybdenite-pyrite-chalcopyrite veins
contain coexisting hypersaline and vapour-rich inclusions, suggesting formation
from a boiling magmatic fluid of moderate to high salinity and temperature (370-
470 ° C). Boiling is interpreted as a depressurisation phenomenon, and may
have arisen from either passive ascent of magmatic fluid or seismically triggered
depressurisation from lithostatic to near-hydrostatic pressure. Breccia infillings
containing quartz-K-feldspar-chlorite-carbonate-fluorite-pyrite-sphalerite occur
before the Au ore stage, and contain low to moderate salinity, moderate
temperature (320-410 ° C) fluids. CL images of quartz from the mineralised
breccia matrix show secondary fracture-sealed networks overprinting primary
euhedral growth zoning. There is no clear distinction between primary and
secondary fluid compositions, however, suggesting no major change in fluid
conditions between episodes of precipitation and fracturing. Ore stage sphalerite-
pyrite-quartz-Au veins also contain low to moderate salinity, moderate
temperature (350-400 ° C) fluids. These veins also contain detectable CO 2 ,
suggesting that Au may have been transported by a distinct CO 2 -charged
source.

Figure 1: Simplified geological map of the Mt. Leyshon Complex (Australian Map
Grid). Ore stage dacite dikes (LD) are likely responsible for gold transport from
an igneous source at the Southern Porphyry (SP) to the permeable breccia host
(MLBX). (after Morrison and Orr, 2000)

References

Morrison, G. W. and Orr, L. (2000) A review of exploration potential in Mount

Leyshon ML 101444 and Fenian EPM 10203, pp. 1-43. Klondike Exploration
Services.

Morrison, G. W. and Andrew, A. S. (1989), unpublished data.

A New Laboratory for LA-ICP-MS Analysis of Fluid Inclusions: Calibrations
using Solid, Liquid, and Fluid Inclusion Standards

Murray M. Allan 1 , LindaForbes 1 , Bruce W.D. Yardley 1 , David Banks 1 ,

Thomas J. Shepherd 2 , and Kirill Shmulovich 3

1 School of Earth Sciences, University of Leeds, Leeds, W.Yorkshire, LS2 9JT,

[email protected] , [email protected] ,
[email protected] , [email protected]

2 British Geological Survey, Keyworth, Nottingham NG12 5GG, [email protected]

3 Institute for Experimental Mineralogy, Chernogolovka, Russia, 142432,

[email protected]

We have established a Laser Ablation ICP-MS laboratory that is primarily

dedicated to the analysis of fluid inclusions. The laser is a 193 nm ArF Lambda
Physik Excimer laser, packaged as the Geolas Q plus system by Microlas. It is
coupled to an Agilent 7500c ICP-MS, with a collision cell to reduce interferences
on Ar masses. As a result, it is possible to analyse for Ca at m/z 40 and Fe at
m/z 56, although there is loss of sensitivity for Cl as a result.

Ablation takes place in a He gas flow, and the ablated material is mixed with Ar
make-up gas in a cyclone mixer before being introduced to the plasma. The
design of the ablation cell and mixer have been optimised to give a smooth,
stable signal during ablation of solids, and to broaden the peak arriving from a
single laser pulse to about 1 second, permitting precise multi-element analysis.

Calibration has been carried out using a combination of standard silicate glasses
(NIST 610, 612, 614, 616, and WRS 601), analysed with the sample stage slowly
moving beneath the laser, and standard multi-element solutions in glass capillary
tubes (labelled A, B, C). Capillaries are first breached with a 50 m m spot size,
and then steadily ablated with a 25 m m spot. The signal obtained during
progressive ablation for ca. 100 seconds is extremely stable, and at most
masses the %RSD over 20s blocks is around 1-3%.

Preliminary results have yielded linear calibrations for a wide range of elements
normalised to Na, including: B, Mg, K, Ca, Mn, Fe, Cu, Zn, Rb, Sr, Sn, Sb, Cs,
Ba, La, and Pb. In most cases, solutions in capillary tubes plot on the same
calibration as solid standards, within errors of a few percent, suggesting matrix-
independent sensitivity. Analyses of synthetic fluid inclusions in quartz
(containing solution B and a 5x dilution, B2) generally bracket the results from the
same solutions in capillaries, with the precision and detection limits dependent on
concentration and inclusion size. Figures 1 and 2 are calibration curves for K/Na
and Mn/Na, respectively, plotted with the results of synthetic fluid inclusion
analyses. Fluid inclusion results for several binary cations, including Ca, Mg, Sr,
Ba, and Mn (Figure 2) fit less well to their calibration curves as does K (Figure 1),
but the discrepancy appears to be linked to the depth of the inclusion from the
surface.
Fluids and hydrothermal alteration assemblages in a Devonian gold-
bearing hot-spring system, Rhynie, Scotland.

M. Baron 1 , S. Hillier 2 , C. M. Rice 1 , K. Czapnik 1 & J. Parnell 1 .

1. Department of Geology & Petroleum Geology, University of Aberdeen,

Aberdeen, AB24 3UE, U.K; 2. Macaulay Land Use Research Institute,
Craigiebuckler, Aberdeen, AB15 8QH, UK.

Hydrothermal mineralization and alteration at Rhynie in NE Scotland is the oldest

known example of a terrestrial low-sulphidation gold-bearing epithermal deposit.
Mineralization is concentrated along a low-angle extension fault (RFZ) which
separates a small outlier (1.5 km by 23 km) of Lower Devonian interbedded
sandstones, plant and arthropod-bearing sinters, shales, conglomerates,
andesitic lavas and tuffs from Caledonian basement comprising of Dalradian
metasediments, norites and minor serpentinites. The hydrothermally altered
rocks contain anomalous levels of elements, such as Au (up to 0.39 p.p.m.), As,
Sb, Hg, Mo and W, which are found in many modern day hot-spring systems.
Mineralization consists predominantly of breccias and veins filled with quartz, K-
feldspar, chert, calcite, pyrite and arsenian pyrite associated with the RFZ. In
contrast, away from the RFZ veins are filled by calcite and fluorite. Associated
pervasive alteration comprise of quartz, K-feldspar (adularia), calcite, hematite,
chlorite and mixed-interlayer illite/smectite. Textural and mineralogy evidence,
such as bladed quartz pseudomorphing calcite, banded chert and adularia,
suggest that boiling was an important process during mineralization. Gold occurs
in solid solution within arsenian pyrite and as sub-micron particles in highly
oxidized rocks. ? 18 O and ?D isotopes as well as noble gas chemistries show
that the mineralizing fluids were derived from both meteoric and magmatic
sources 1 . Detailed fluid inclusion microthermometry further reveals that in fact
several types of fluids were involved in mineralization in the Rhynie hydrothermal
system each with distinctive temperatures and compositions (see fig. 1). These
include:

1. Low to high temperature (T h 110 to 360 ° C), low salinity (0 to 2.9 wt.%
NaCl eq.) H 2 O-salt, heated meteoric fluids. This fluid was responsible for
mineralization associated with the RFZ and by its shear dominance was
probably the main gold-bearing fluid. The temperature and composition of
this fluid type is very similar to the fluid implicated in mineralization in
many other more recent Au-rich, low-sulphidation epithermal ore deposits
from other parts of the world 2 .
2. A very minor input of high temperature (T h 280 to 430 ° C), H 2 O-CO 2
-~6wt.%NaCl fluid. This fluid was also involved in mineralization along the
RFZ and was probably derived from andesitic magmatism. This fluid,
along with the low to high temperature, low salinity H 2 O-salt fluid, is only
recognised in the RFZ indicating that its migration was strongly controlled
by this fault zone.
3. Low temperature (<40 to 50 ° C), low salinity (0 to 3.5 wt.% NaCl eq.) H 2
O-salt fluids were involved in the formation of the plant and arthropod-
bearing sinters. ? 18 O,??D and noble gas composition of this fluid
indicates that it is of a meteoric origin.
4. Low temperature (T h 57 to 160 ° C), low to high salinity (0 to 18 wt.%
NaCl eq.) H 2 O-NaCl fluids. This fluid was responsible for mineralization
and cementation away from the RFZ in other areas of the basin. This fluid
is not recognised in the RFZ indicating that it was not involved in
mineralization along the RFZ or hot spring activity.

Detailed x-ray diffraction analysis in a traverse across the RFZ reveals zoned
hydrothermal alteration assemblages, which developed in response to
temperature gradients in the hydrothermal system (see fig. 1). These alteration
assemblages include:

1. A high temperature, gold-bearing K-feldspar-quartz-illite facies. In

comparison with other epithermal deposits this facies formed at
temperature between 250 and 350 ° C 3 . Fluid inclusions hosted within K-
feldspar yield very similar temperatures (T h of between 200 and 260 ° C).
2. A moderate temperature illite/smectite-quartz-K-feldspar-chlorite-calcite
facies. The presence of chlorite combined with the content of expandable
 layers in the illite/smectite fraction indicates that this facies formed at
temperatures between 150 and 200 ° C.
3. A low temperature chlorite-calcite-illite/smectite facies. This facies is
similar to that which characterises Devonian sediments elsewhere in NE
Scotland and corresponds to regional diagenetic temperatures in the
range of 100 to 150+ ° C 4 .

Subsequent faulting has juxtaposed the low temperature alteration facies against
the high temperature alteration facies. The bulk of the gold-bearing, high
temperature facies has been eroded indicating that the ore potential of this
deposit is low unless a portion of the zone has been preserved by downfaulting.

References

• Rice, C. M. & 13 others. A Devonian auriferous hot spring system, Rhynie,

Scotland. Journal of the Geological Society, London 152 , 229-250.

• Hedenquist, J. W. & Lowenstern, J. B. 1994. The role of magmas in the

formation of hydrothermal ore deposits. Nature 370 , 519-527.

• Inoue, A. 1995. Formation of clay minerals in hydrothermal environments. In

Velde B. (ed.) Origin and mineralogy of clays. Springer-Verlag, Berlin.

• Hillier, S. & Clayton, T. 1989. Illite/smectite diagenesis in Devonian lacustrine

mudrocks from northern Scotland and its relationship to organic maturity
indicators. Clay Minerals 24 , 181-196.
Figure 1. Subsurface reconstruction of the fluids and alteration assemblages in
the Rhynie hot spring system.

Re-evaluation of Flat-style mineralisation in the Northern Pennines—a

preliminary report

A.J. Benham 1* , J. Naden 1 and B. Young 2

1 British Geological Survey, Keyworth, Nottingham, NG12 5GG *email

[email protected]

2 British Geological Survey, West Mains Road, Edinburgh, EH9 3LA

The North Pennine Orefield (NPO) has produced in excess of 4 million tons of
lead as well as significant amounts of zinc, fluorite, barite and witherite (Dunham,
1990). It covers almost 4000 km 2 and is located in two fault-bound blocks: the
Alston Block to the north—the focus of this study—and the Askrigg Block to the
south, separated by a graben structure, the Stainmore Trough.
Figure 1 . Simplified geological map of the Alston Block in the Northern Pennines
showing localities studied (1. Tynebottom Mine; 2. Smallcleugh Mine; 3.
Wellheads Hush; 4. West Rigg open-cut; 5. Eastgate Cement Works quarry; 6.
Rogerley Mine; 7. Scordale (Hilton and Merton Mines)).

Mineralisation is located in a cyclical sequence of mostly flat-lying Dinantian

limestones, shales and sandstones and is hosted by veins and structures known
as “flats”. The veins are sub-vertical and best developed in the more competent
limestone and sandstone units, and locally in the Permo-Carboniferous dolerite
of the Whin Sill, where fault refraction creates open space. Commonly,
mineralisation extends into the wallrock to form sub-horizontal structures (flats)
that are thought to result from metasomatic replacement of wallrock limestone.
Veins and associated flats may be traced over several kilometres, and, though
the vein-style is the most obvious expression of mineralisation at surface, flats
have contributed significantly to the overall resources. At the orefield scale,
mineralisation is recorded at several stratigraphical levels with the Great
Limestone hosting many of the most productive orebodies. In addition, it is zoned
with respect to the main gangue phases. Deposits within the central region of the
Alston Block are dominated by fluorite, and surrounding this is a zone where
barite and/or witherite comprise the main gangue assemblage. The boundary
between these two zones is sharp and fluorite and barite are almost mutually
exclusive, with some notable exceptions (e.g. Hilton and Merton Mines—Figure
1)

As part of a wider research programme to develop improved mineral deposit

models, the British Geological Survey has recently begun re-evaluating
mineralisation in the NPO focussing on the genesis of flat-style ore bodies as
these have the best potential for additional resources.
Field observations at several localities (Figure 1) indicate a variety of
mineralisation styles. Common to all occurrences is widespread Fe-(Mg)-
metasomatism. This is manifest as pervasive ankeritisation and sideritisation of
the host limestones, which locally formed economic iron deposits (e.g. Carricks
and West Rigg Mines). However, at some localities Fe-metasomatism is also
associated with fracture mineralisation, for example at Wellheads Hush, where
centimetre-wide anastomising siderite veinlets penetrate pervasively ankeritised
limestone. Moreover, for flat-hosted Pb–Zn–Ba–F mineralisation, metasomatism
is not the only mineralisation mechanism. At a number of localities, at the metric
and smaller scales, significant mineralisation fills open space or involves veining
and hydrofracturing. At Rogerley Mine centimetric euhedral green fluorite lines
metre-scale sub-horizontal lenticular cavities developed in ankeritised Great
Limestone. Similarly at Smallcleugh, cubic galena (up to 5cm), locally intergrown
with abundant ankerite and calcite, can be seen lining small (1–2 m) cavities.
This style is passive, in the sense that sulphide mineralisation occurs in existing
open spaces and involves little tectonism. Conversely, at other localities
mineralisation is, in part, controlled by the brecciation and fracture filling of the
host rock. At Eastgate Quarry brecciation is characterised by centimetric angular
limestone clasts enclosed within a massive fluorite-rich matrix. It has very sharp
contacts with the host rock with little obvious metasomatic alteration of the
limestone. Brecciation is also seen at Tynebottom Mine. Here, angular, dark,
fine-grained, silicified limestone clasts are enclosed by coarsely crystalline white
quartz matrix with locally abundant iron sulphide. A similar situation, but with
different mineralogy, occurs at Scordale. Here, contact metamorphosed siltstone
(whetstone) caps the flat-style mineralisation within the Melmerby Scar
Limestone and, locally, centimetric angular brecciated siltstone blocks are
cemented by fluorite and barite. However, at the orefield scale this type of
mineralisation is not typical.

At a number of localities, a common theme linking flat mineralisation is the

presence of impermeable layers, such as bedding-parallel clay seams, stylolites
and hardened siltstones. Although they are not necessarily widespread
throughout the orefield, these features suggest a host-rock control to the
distribution of flat-style mineralisation with impermeable layers influencing lateral
and vertical flow of mineralising fluids. At present, the role of ankeritisation and
sideritisation is unclear. It may be a wallrock alteration associated with the main
Pb–Zn–Ba–F mineralisation or an early ground preparation event—ankeritisation
and sideritisation of limestone involve volume reduction and the creation of void
space. A combination of sub-horizontal fluid pathways and open space created
by Fe-metasomatism may be a viable mechanism for locating Pb–Zn–Ba–F flat-
style mineralisation.

The conclusions of this preliminary study are:

• Several distinct styles of flat mineralisation occur: those where extensive

metasomatism is a major factor, (e.g. the Fe deposits and Smallcleugh Mine);
those where both metasomatism and open-space-filling are important (e.g.
Rogerley and to some extent, Smallcleugh Mine); and those where brecciation
and fracture filling are dominant locally resulting in what may be regarded as
“horizontal” vein fillings (e.g. Scordale).

• The relative importance of metasomatism versus open-space filling and

brecciation is unclear, but it is clear that flat mineralisation is produced by other
processes in addition to metasomatic replacement.

• Pervasive Fe-alteration of the limestones may have acted as ground

preparation for the later Pb–Zn–Ba–F mineralisation. This is supported by the
occurrence of Pb-, Zn-, Ba- and F-bearing minerals in the vuggy centres of
siderite veins and lining voids in ankeritised rock.

• Flat-lying impermeable layers and structures probably play a role in directing

the flow of mineralising fluids and flat formation.

Reference

Dunham K. C. (1990) Geology of the Northern Pennine Orefield. Volume 1: Tyne

to Stainmore. Economic Memoir Covering the Areas of 1:50000 and One-inch
Geological Sheets 19 and 25 and Parts of 13, 24, 26, 31, 32 (England and
Wales). British Geological Survey: HMSO

Potential for stratiform base-metal sulphides in SW England

A.J. Benham * , K. E. Rollin, F. M. McEvoy and A. G. Gunn

British Geological Survey, Keyworth, Nottingham, NG12 5GG *email

[email protected]

The British Geological Survey (BGS) has recently re-evaluated the potential for
stratiform base-metal sulphide deposits in Devon and east Cornwall (Rollin et al.,
2001) (Figure 1). On the basis of their stratigraphy and tectonic setting the Lower
Carboniferous strata between, and to the north of, Bodmin Moor and Dartmoor
are favourable targets for Iberian Pyrite Belt (IPB)-type deposits (Figure 2).
Similarly, the geological setting of the Middle to Upper Devonian sediments of
Exmoor are comparable with those of the Harz basin in Germany which hosts the
major polymetallic Sedex deposits at Rammelsberg and Meggen.
Figure 1. Location of areas mentioned in text. 1. Combe Martin; 2. Exmoor; 3.
East Dartmoor; 4. Central region; 5. Egloskerry

Figure 2. Stratigraphy of IPB compared with the area between Bodmin Moor and
Dartmoor, SW England

A large amount of exploration has been carried out in the target region by
commercial companies and by BGS, ranging from regional reconnaissance
surveys to detailed prospecting. Digital data from these surveys has been used
to assess the potential for the occurrence of stratiform sulphide deposits using
knowledge-based prospectivity analysis tools including Arc SDM (Spatial Data
Modeller) (Rollin et al., 2001). Prospective targets identified in this study include
the area between the Bodmin Moor and Dartmoor granites, parts of Exmoor and
east Dartmoor, each of which is discussed below.
The central area between Bodmin Moor and Dartmoor is prospective for IPB-type
deposits due to the following features: Lower Carboniferous strata, including
many formations with black shales; strong magnetic anomaly; many thrusts;
widespread stream-sediment anomalies for Mn and Zn; and the highest
concentration of known occurrences of stratiform mineralisation in S W England,
including prospects at Egloskerry, Bridestowe and Chillaton.

Other targets for stratiform sulphide mineralisation are present in the Exmoor
district. Minor stratabound base-metal occurrences are known at Lester Point, on
the coast at Combe Martin, as well as at other localities inland to the south-east,
together with drainage geochemical anomalies for Zn, Mn and Pb (Jones et al.,
1987). A distinct linear aeromagnetic anomaly, trending west-north-west to east-
south-east, extends for at least 40km following the boundary between the
Ilfracombe Slates and the Morte Slates. Two boreholes, previously drilled to test
this anomaly at Honeymead Farm (Jones et al., 1987), showed that the source
was probably pyrrhotite that had been produced by recrystallisation of syngenetic
pyrite.

Other targets are found in the east Dartmoor area where Lower Carboniferous
strata are overthrusted and are associated with minor positive gravity anomalies.
High contrast, multi-site Ba anomalies in the Teign Valley may reflect known vein
baryte mineralisation, while associated Mn anomalies may be due to Pb-Zn-Cu-
As-Ba mineralisation to the east of the River Teign. In the Teign Valley, near
Bridford and Bovey Tracey, previous geochemical and geophysical surveys by
BGS, followed by drilling at some sites, supported the existence of stratiform
sulphide mineralisation (Beer et al., 1992 ).

Additional targets occur in the Lower Carboniferous succession in the central

area between the Bodmin Moor and Dartmoor granites. Previous commercial
exploration at Egloskerry, near Launceston, in the early 1980s, including soil and
deep overburden sampling and drilling, identified stratabound galena and minor
sphalerite in Lower Carboniferous shales and mudstones of the Liddaton and
Crackington Formations. In order to evaluate the utility of high-resolution gravity
surveys in exploration for stratiform sulphide mineralisation in this environment
BGS carried out a detailed survey over an area of about 20 km 2 around
Egloskerry (Rollin et al ., 2002). The new work identified a positive residual
gravity anomaly that showed a strong correlation with geology and also with
anomalous Zn and Pb soil geochemistry.

New IP and high-resolution gravity surveys were also carried out over the Combe
Martin area and parts of Exmoor. At Combe Martin, where argentiferous galena
was mined for several centuries from predominantly vein-style mineralisation, a
low-amplitude positive gravity residual anomaly was identified along strike from
the former workings, approximately along the outcrop of the Lester Slates which
are the main host to the mineralisation. Farther east, the IP survey identified an
anomaly near the Honeymead borehole that may be related to pyrrhotite
identified in drillcore.

New geophysical investigations, following up targets identified in the prospectivity

analysis, have confirmed the potential of these areas for stratiform base-metal
sulphide mineralisation. Additional targets highlighted by the prospectivity
analysis remain untested. Furthermore, this study indicates that high-resolution
gravity and IP techniques have the potential for identifying concealed
mineralisation in this region.

References

Beer, K. E., Ball, T. K., Cooper, D. C., Evans, A. D., Jones, R. C., Rollin, K. E.,
and Tombs, J. M. C. (1992) Mineral investigations in the Teign Valley, Devon.
Part 2: base metals. Mineral Reconnaissance Programme Report of the British
Geological Survey, No 123.

Jones, R. C., Beer, K. E. and Tombs, J. M. C. (1987) Geochemical and

geophysical investigations in Exmoor and the Brendon Hills. Mineral
Reconnaissance Programme Report of the British Geological Survey, No 90.

Rollin, K. E., Gunn, A. G., Scrivener, R. C. and Shaw, M. H. (2001) Potential for
stratiform massive sulphide mineralisation in south-west England. British
Geological Survey Commissioned Report CR/01/240, DTI Minerals Programme
Publication No 9.

Rollin, K. E., Shaw, M. H., Benham, A. J., and Morgan, D. J. R. (2002) High-
resolution gravity surveys at Egloskerry and Combe Martin, SW England. British
Geological Survey, Internal Report . IR/02/195

Bonding and coordination of gold in arsenic rich pyrite: experimental

results

A.J. Bennett & L.G.Benning

School of Earth Sciences, University of Leeds, Leeds, LS2 9JT, UK.

[email protected]

Knowledge of the distribution, concentration and mechanisms of incorporation of

gold and arsenic into pyrite grains will significantly improve our understanding of
ore deposits containing large amounts of submicroscopic gold. Such knowledge
can lead to a more efficient method of extraction and recovery of gold in both
ancient and modern volcanogenic massive sulphide deposits.
The vast majority of studies of submicroscopic gold have examined natural
deposits from ancient massive sulphide deposits. In such cases the gold is
heterogeneously distributed and of low (up to 200ppm) concentrations, thus is
extremely difficult to analyse using conventional techniques. A novel approach
we have taken is to make synthetic pyrite grown in the laboratory. Onto this both
gold and arsenic can be sorbed or incorporated. Higher concentrations of gold
can be adsorbed to the pyrite grains using this method, thus allowing them to be
analysed using Synchrotron Extended X-Ray Absorption Spectroscopy (EXAFS)
and X-Ray Absorption Near Edge Spectroscopy (XANES). These techniques can
be used to determine both the oxidation state and local coordination of gold and
arsenic in pyrite.

An experimental system allowing mixing of two hot (up to 250 o C) fluids at

pressure and in an anaerobic environment has been built to mimic the mixing of
hydrothermal fluids. Results have shown that pyrite is a common material derived
from the mixing of either Fe 2+ or Fe 3+ solutions with slightly oxidised NaHS
solutions. Other common products of mixing at different solution ratios,
temperatures and pH values include pyrrhotite, greigite, marcasite and hematite.
The experimental set up was also used to sorb and coprecipitate gold and
arsenic onto the synthetic pyrite grains at high temperature (at concentrations of
500-1000ppm and 0.5-3wt% respectively). The particle size of the pyrite fraction
ranged from 200nm to 1.5µm, and the BET determined surface area was 5.10m
2 /g.

Transmission Electron Microscopy (TEM) has shown that sorption of arsenic onto
pyrite is not homogeneous. Fine grained pyrite adsorbed more arsenic onto its
surface than coarser grained and areas of an unidentified poorly-ordered iron
sulphide/oxide forms which sorbed the highest concentrations of arsenic.

X-ray absorption spectroscopy data has shown that arsenic is adsorbed onto the
surface of pyrite forming three sulphur bonds; also gold can become incorporated
into the structure of pyrite rather than being adsorbed onto the surface. In the
mixed As-Au samples the sorption of arsenic seems to inhibit the incorporation of
Au into the structure in comparison to the pyrite system containing only gold.

By using synthetic analogues of natural gold deposits we can begin to answer

some of the questions of how gold is associated with arsenic and becomes
incorporated into pyrite. Our microscopic and spectroscopic data provides an
initial insight into the possible processes involved in gold accumulation in a
modern hydrothermal environment.

The role of basinal fluids in the formation of ore deposits in Cumbria:

Hematite mineralisation in granite and limestone.
 Andrew Bennett and Stephanie Houston

The county of Cumbria lies at the eastern edge of the East Irish Sea Basin and is
host to a hematite ore deposit, which was economically important during the 19
th century. The hematite occurs in both the Dinantian limestones of the area and
in veins within the Caledonian Eskdale Granite. The origin of both deposits is of
much debate, but is thought to be related to the movement of hot, relatively
reducing, iron rich brines from the East Irish Sea Basin to the Carboniferous
Limestones where they met a cool, shallow, oxidising fluid. It has not yet been
established whether these two types of deposit were formed in the same
mineralisation event.

Fluid inclusion analyses were undertaken of samples of gangue minerals from

both the granite and limestone. Microthermometry revealed that the dominant
fluid in the limestone had a homogenisation temperature of 104 o C, salinity of c.
25% NaCl = 18wt% and CaCl = 9wt%. That is, a low temperature, high salinity
brine. The fluid in the granite samples was found to be completely different and is
high temperature (T h = 209 o C) and low salinity (NaCl wt% eq. = 6.0). Crush-
leach and laser ablation analysis confirmed these differences and showed that
only one fluid was involved in the formation of the limestone deposit, a bittern
brine, formed as a result of the evaporation of seawater. It revealed that the
granite-hosted fluids also appear to be derived from evaporation of seawater,
except one anomalous sample which has a ionic signature typical of those brines
derived from the dissolution of halite. It was discovered from literature review that
the iron content of the brines is probably directly related to both temperature and
oxidation state of the fluid. In addition, modelling results show that the best way
to cause hematite to precipitate is by a combination of a decrease in temperature
and an increase in the oxidation state of the system; this is most likely to be
brought about by fluid mixing.

This finding supports the theory of formation described above, that fluid mixing
occurred between a sulphatic groundwater and a warm, iron-rich, hypersaline
brine. It is proposed that fluid mixing or wall rock reaction are the most likely
modes of formation of both the granite- and limestone-hosted Cumbrian Hematite
deposits.

Recent Advances in Microanalytical Techniques Applied to Mineral Deposit

Research and Exploration

Robert J. Bodnar

Dept. of Geosciences, Virginia Tech, Blacksburg VA 24061 USA email:

[email protected]
Over the past several years our understanding of ore-forming processes in
magmatic-hydrothermal systems has improved significantly. Today we are able
to thoroughly characterize the chemical and isotopic compositions of ore-forming
fluids, including concentrations of metals in the fluids, as well as determine
absolute ages and durations of ore-forming events with a spatial resolution and
precision that was unprecedented a few decades ago. Improvements in modern
microanalytical techniques now permit analysis of metals and other elements at
parts per million levels or better in samples in the nanogram to picogram range,
and determination of isotopic compositions of areas of samples that are
measured in micrometers.

One might ask: Why do we want to analyze smaller samples? The answer lies in
the fact that ore formation is a dynamic, evolving process, with changing
temperature, pressure, fluid composition and mineral chemistry over the lifetime
of the ore deposit. Thus, analysis of a macroscopic (hand) sample of a rock from
an ore deposit will provide highly accurate and precise data on the chemistry of
that sample, but that chemistry represents the evolving chemistry integrated over
some relatively long and (perhaps) unknown period of time. Conversely, analysis
of a micro-sample from that same rock provides information concerning the
chemistry of the rock, and/or the fluids that have interacted with the rock, during
a much shorter portion of the rock history. It is this desire to understand the
manner in which the chemistry of the system changes during ore formation that is
the driving force for analyzing smaller and smaller samples.

Perhaps the question that has intrigued economic geologists more than any other
has been the concentration of metals in ore-forming fluids. The answer to this
question determines, to a large extent, the amounts of fluid and time required to
form a deposit. Recent developments involving laser ablation ICP-MS, PIXE,
PIGE and synchrotron-based XRF (SXRF) now permit metal contents of
individual fluid inclusions to be analyzed. Combined with careful petrography,
these data can be used to determine which of several different hydrothermal
fluids were transporting and depositing metals. Synchrotron-based XAFS
analysis offers the additional possibility of determining the speciation of metals in
solution. Analysis of melt inclusions and coeval fluid inclusions is providing
valuable information concerning metal contents of silicate melts associated with
magmatic-hydrothermal deposits, and mechanisms of transfer of metals from the
melt to the hydrothermal fluid. Raman analyses now permit quantitative
estimates of volatiles in ore-forming fluids.

Advances in instrumentation and techniques for stable and radiogenic isotopic

analysis of ore and gangue minerals and fluid inclusions have led to a more
complete understanding of sources of ore fluids and metals in magmatic-
hydrothermal deposits. SHRIMP analyses of individual sulfide phases have
revealed complex variability in sulfur isotopic composition in many deposit types.
Ages and durations of ore-forming events are now routinely determined based on
U-Pb, 40 Ar/ 39 Ar, and Re-Os isotopic systematics. Future collaborations
between geochronologists and ore-fluid geochemists are expected to better
constrain the temporal evolution of ore-fluid chemistry.

The “Calamine-type” Zn-Pb deposits in Belgium: a result of Mesozoic (?)

paleoweathering processes

Maria Boni, Luca Fedele (1) & Léon Dejonghe (2)

(1) Dipartimento di Geofisica e Vulcanologia, Università di Napoli Federico II, Via

Mezzocannone 8, 80134 Napoli, Italy ([email protected]), ([email protected])

(2) Geological Survey of Belgium, 13 Rue Jenner, B-1000 Brussels, Belgium

(Leon.Dejonghe @naturalsciences.be)

European non-sulfide Zn ore deposits represented the historical basis for zinc
mining and smelting industry. Based on their geological characteristics, they
have been assigned mainly to the “ Calamine ” group, derived from supergene
alteration of carbonate-hosted MVT and Sedex deposits (Large, 2001; Boni &
Large, 2003) .

The word “Calamine”, originates from the Belgian locality La Calamine where a
non-sulfide Zn deposit was continuously exploited from the 14 th century up to
1884, with a production of more than 600.000 tons of Zn metal. Since its
discovery, “Calamine” became a mining expression to describe a common
mixture of zinc-ores such as Zn-carbonate {smithsonite = ZnCO 3 } and Zn-
silicate {hemimorphite = Zn 4 (Si 2 O 7 )(OH) 2 .H 2 O}, or the assemblage of
hemimorphite, smithsonite, hydrozincite {Zn 5 (CO 3 ) 2 (OH) 6 } and willemite
{Zn 2 SiO 4 } in greyish, brownish or yellowish earthy and often crustified masses
mixed with red clays. “Calamines” are grey, yellow or even black in colour and
occur as massive rock masses or with concretionary, foliated or stalactitic
shapes. They could be microcrystalline- (so-called “amorphous” in the old
literature) or coarse-grained aggregates, breccia-cementing concretions or
vuggy, with internal dissolution cavities covered by small idiomorphic smithsonite
and hemimorphite crystals. The shape of the Zn-non sulfide concentrations
observed in most mines, ranges from bulk replacement bodies of host
carbonates and/or sulfides in the deepest levels, through amorphous and
microcrystalline aggregates, to the concretionary infilling of isolated druses and
cavities in the upper zones.

La Calamine is described in the scientific literature as a paleokarstic type of

deposit (Dejonghe, 1998), although its genesis as well as its age has never been
fully understood. Likewise, the age of the deep supergene alteration of many Zn-
Pb veins, exploited in the Belgian and German neighbouring mining districts
(whose characters have similarities with La Calamine) has never been clearly
determined. Our aim is to identify the geological constraints responsible for the
genesis of La Calamine deposit, as well as of the deep-reaching gossans of the
surrounding Zn-Pb vein-type deposits. Therefore, we examined the
paleoweathering stages in Belgium and Western Germany and their
metallogenetic implications. The thick kaolinized regolith remnants ranging in age
from Uppermost Jurassic to Lower Cretaceous, which developed on the post-
Variscan paleosurface in Cambrian to Silesian rocks of various compositions
(siliciclastic and carbonate), could be one of the surface expressions of the
weathering responsible for the non-sulfide Zn-ore deposits in Belgium and
Northern Eifel. The hypothesis of a strong paleoweathering period may be
carried out also because the level of the oxidation in most of the base metal
veins and associated flats of Eastern Belgium and Western Germany is
substantially deeper than the present-day water table. Upper Cretaceous
sediments, whose deposition had interrupted the weathering process,
conformably cover the regolith. Neither Caenozoic weathering, that was
responsible for halloysite accumulations at the bottom of giant cryptokarsts, nor
Quaternary landscape modelling, should have played in Belgium a major role in
the formation of the economic non-sulfide “Calamine-type” Zn-deposits.

One of the peculiarities of La Calamine, is the economic abundance of willemite

(ZnSiO 4 ) among the non-sulfide minerals, a species that in the more recent
literature has been genetically related to hydrothermal processes (Brugger et al.,
2003; Hitzman et al., 2003). Willemite is then followed and partly replaced by
smithsonite and hemimorphite. We carried out preliminary microthermometric
measurements on fluid inclusions in willemite and found out that, among many
small monophase (empty?) inclusions, about 20% are liquid-rich two-phase
(L+V). They are allegedly primary, being isolated or in small clusters and not
related to any evident cleavage plans. Leakage was easily detected in many
inclusions which were avoided during microthermometry. Th in the willemite
crystals range from 70 to 190 °C and Tm data suggest a salinity close to 0 wt. %
NaCl equiv.

These high temperatures, if real, pose a problem to the supergene interpretation

of the Belgian “Calamine”. The only likely explanation we can suggest at this
stage, is to relate high temperatures of the willemites to the presence in the
mineralization area of an anomalous geothermal gradient during Cretaceous, so
far not recorded in NE Belgium, but well known in the Lower Saxonian Basin in
Germany (intrusion of the Bramsche Massif or deep burial followed by inversion,
Petmecky et al., 1999).

References

Boni, M. and Large, D., 2003, Nonsulfide zinc mineralization in Europe: an

overview. Economic Geology, v. 98, p. 715–729.
Brugger, J., McPhail, D.C., Wallace, M., and Waters, J., 2003, Formation of
willemite in hydrothermal environments. Economic Geology, v. 98, p. 819–835.

Dejonghe, L., 1998, Zinc-lead deposits of Belgium. Ore Geology Reviews, v. 12,
p. 329-354.

Hitzman, M.W., Reynolds, N.A., Sangster, D.F., Allen, C.R., and Carman, C.,
2003, Classification, genesis, and exploration guides for nonsulfide zinc deposits:
Economic Geology, v. 98, p. 685–714.

Large, D., 2001, The geology of non-sulphide zinc Deposits – an Overview:

Erzmetall, v. 54(5), p. 264-276.

Petmecky, S., Meier L., Reiser, H. and Littke, R., 1999, High thermal maturity in
the Lower Saxony Basin: intrusion or deep burial? Tectonophysics, v. 304, p.
317-344.

Chronology of the Iberian Pyrite Belt: Which time scale calibration provides
the best framework for deposit modelling and exploration?

Clive A. Boulter

School of Ocean & Earth Science, University of Southampton, Southampton

Oceanography Centre, Southampton, SO14 3ZH. E-Mail: [email protected]

Those looking on from the sidelines at research in the Iberian Pyrite Belt could be
forgiven for expressing puzzlement over the conflicting interpretations seen in
contemporaneous publications. Recently many lines of evidence have been
converging on the view that the remarkable accumulation of sulphide deposits in
the Iberian Pyrite Belt formed essentially at the same time very early in the
evolution of the host sequence (Boulter 1996; Boulter et al . 2001; Pereira et al .
1996; Oliveira et al . 1997; González et al . 2002). The reinterpretation of the
physical volcanology of the host Volcano-Sedimentary Complex has concluded
that the mineralization process at Rio Tinto was initiated prior to the onset of
magmatic activity (Boulter 1996) and at Tharsis (González et al . 2002) only a
few metres of volcanogenic material underlies the massive sulphide sheet. In
stark contrast to this trend to an early timing for the deposits all recent reviews
have championed the view that the major deposits in the Pyrite Belt formed in the
waning phase of the major/latest volcanic event and Barrie et al . (2002) consider
that “The massive sulphide deposits are invariably near the top of the felsic units
and where there are several felsic units, the VMS deposits are generally above
the uppermost one.” Further divergence from the consensus that was developing
is seen in the radiometric study (ID-TIMS) of Barrie et al . (2002) where three,
apparently distinct, episodes of massive sulphide formation were claimed.
Establishing the chronology of the belt is complicated by the absence of a
definitively calibrated time scale for events around the Devonian/Carboniferous
boundary. Current uncertainties introduce variations of up to 10 m.y. in the
estimates for stage boundaries in the interval being considered. The time scale
calibration adopted by Barrie et al . (2002) places the Devonian-Carboniferous
boundary at 362.5 Ma. At the other extreme, the chronometric calibration of the
relative time scale used in Streel et al . (2000) has the D-C boundary at 354 Ma.
From their radiometric data Barrie et al . (2002) claim an age range for the
volcanogenic host sequence [and massive sulphide mineralization] of 356 to 350
Ma. However they had already noted the biostratigraphic data from Aznalcollar
and Neves Corvo which give a latest Devonian, LN biozone, age (Pereira et al .
1996; Oliveira et al . 1997), and which, on their time scale calibration of choice,
give these deposits a numerical age close to 363 Ma. Incorporation of these
older ages would have more than doubled the age range of volcanic, and related
mineralization, activity quoted in the paper.

It is possible to reconcile these highly divergent positions by recognising that

voluminous post-mineralisation peperitic intrusions have dismembered the
components of the massive sulphide deposits (Boulter et al . 2001). These
intrusions have been emplaced above and below the sulphide sheet. In an
environment where such intrusions are common there is no longer a one-to-one
relationship between lithostratigraphic order and chronology. Simply because a
volcanogenic unit lies below a deposit cannot be taken as proof that it pre-dates
the mineralization process. For instance though the unit sampled for the TIMS
analysis at Aljustrel (Barrie et al . 2002) mainly lies below the sulphide sheet it
also intrudes into both the cherts and red mudrocks above the sulphides. Thus
the radiometric age records a post- rather than as claimed a pre-mineralisation
event.

Given the exceptional biostratigraphic control on major deposits widely dispersed

through the belt, plus the convergence of these data with the
palaeovolcanological evidence for mineralization early in the host rock evolution,
it is possible to invert the problem. Rather than using the radiometric data to
provide a Pyrite Belt chronology, the geological data can be used to assess the
accuracy of the radiometric results which are subject to several sources of error.
In this inverted scenario the Streel et al. (2000) chronometric calibration
produces a coherent pattern for the Pyrite Belt radiometric and the
biostratigraphic ages. Field relations show that both the Aljustrel and Nerva
radiometric ages (Dunning et al. 2002) are on post-mineralisation intrusions and
their dates agree with the proposal that all major deposits formed during the LN
biozone. Two outliers exist in the radiometric data; the Aznalcollar age which
must be in error because of the good palaeontological control and the Rio Tinto
age from Barrie et al. which may record the emplacement of the dominant post-
mineralisation intrusions and the hydrothermal activity generated as they
interacted with unconsolidated sediments. This latter event is sufficiently intense
to have created a semi-conformable alteration zone in the 80-90% of the
volcanogenic host that post-dates the sulphide sheets and imposed a distinctive
oxygen isotope signature on the dominant felsic intrusions. Overall the
comparison between the palaeontological and radiometric ages shows that the
true accuracy of the radiometric determination via the TIMS method, for
circumstances equivalent to the Pyrite Belt, is of the order of plus or minus three
million years. Geological processes will introduce much greater degrees of
uncertainty than those originating in laboratories.

Integration of a wide range of data supports the view that the major massive
sulphide deposits of the IPB formed in the same half million year interval very
soon after the initiation of the magmatic event. This means that deposit modelling
has to turn its focus away from magmatic associations and accept a major role
for tectonics in driving the hydrothermal convective system. The chronological
relations also show that metal sources are more probably to be found in the
upper crust than in the immediate volcanogenic host. Exploration needs to
consider the disruptive effects of the late intrusions on all aspects of the
mineralised system.

References

Barrie, C.T., Amelin, Y. & Pascual, E. 2002. U-Pb Geochronology of VMS

mineralisation in the Iberian Pyrite Belt. Mineralium Deposita , 37 , 684-703.

Boulter, C.A. 1996. Extensional tectonics and magmatism as drivers of

convection leading to Iberian Pyrite Belt massive sulphide deposits. Journal of
the Geological Society, London , 153 , 181-184.

Boulter, C.A., Soriano, C. & Zimman, P. 2001. The Iberian Pyrite belt: A
mineralised system dismembered by voluminous high-level sills. Terra Nova , 13
, 99-103.

Dunning, GR, Diez Montes, A, Matas, J, Martin Parra, LM, Almarza, J, Donaire,
M, (2002) Geocronologia U/Pb del volcanismo acido y granitoides de la Faja
Piritica Iberica (Zona Surportuguesa). Geogaceta 32: 127-130.

González , F., Moreno, C., Sáez, R. & Clayton, G. 2002. Ore genesis age of the
Tharsis Mining District (Iberian Pyrite Belt): a palynological approach. Journal of
the Geological Society, London , 159 , 229-232.

Oliveira, J.T., Pacheco, N., Carvalho, P. & Ferreira, A. 1997. The Neves Corvo
mine and the Paleozoic geology of southwest Portugal. In: Barriga, F.J.A.S. &
Carvalho, D. (eds) Geology and VMS deposits of the Iberian Pyrite Belt. Society
of Economic Geologists, Guidebook Series, 27 , 21-71.
Pereira, Z., Sáez, R., Pons, J.M., Oliveira, J.T. & Moreno, C. 1996. Edad
devónica (Estruniense) delas mineralizaciones de Aznalcóllar (Faja Pirítica
Ibérica) en base a palinologia. Geogaceta , 20 , 1609-1612.

Streel, M., Caputo, M.V., Loboziak, S. & Melo, J.H.G. 2000. Late Frasnian-
Famennian climates based on playnomorph analyses and the question of the
Late Devonian glaciations. Earth-Science Reviews , 52 , 121-173.

The NERC Isotope Community Support Facility (ICSF) at the Scottish

Universities Environmental Research Centre (SUERC)

Supporting Mineral Deposit Research in the UK and beyond.

Boyce, A.J., Fallick, A.E. & McDonald, A.

Isotope Geosciences Unit, Scottish Universities Environmental Research Centre,

East Kilbride, Glasgow G75 0QF

Do you need stable isotope analyses to support your research? If so, ICSF
offers access to SUERC's comprehensive stable isotope facilities for high
quality research in the broad field of Mineral Deposit research. Access is
mediated through a formal application to a NERC steering committee. The
application is not onerous, but demands clarity of the hypothesis under test, and
an understanding of how stable isotope analyses can make a telling contribution.
Advice by ICSF staff is given throughout the application procedure, if desired,
and we particularly encourage applications in support of Ph D student research.
For successful applications, all agreed isotopic analyses and training are free at
the point of delivery. Such analyses and ICSF input are costed and Principal
Investigators receive a NERC grant-in-kind which is RAE accountable. Contact
[email protected] for details on how to get started with ICSF – we look
forward to working with you at some point in the future! Here's a bit more about
our philosophy, practice and facilities.

Core contract funding is competitively won from the NERC, and primarily ICSF
meets the demand from the Geoscience Community for isotopic analyses and
training primarily in support of high quality research in the full life cycle of the
Applied Minerals research , from genetic studies informing exploration and
exploitation, through to remediation. This work falls largely into the NERC ENRI's
Natural Resource Management and Pollution and Waste . The 2000 NERC
response to the Fore sight consultation exercise - “ Energy and the Natural
Environment - A Way to Go ” - stated that the expertise of UK scientists in the
fields of resource exploration and exploitation, as well as remediation, was a
major benefit to meeting the future challenges posed by Sustainable
Development. Reflecting this, Sustainable Economies is one of NERC's core
priority areas in the NERC Strategy: Science for a sustainable future 2002-
2007 .

Central to any definition of Sustainable Economies is fulfilment of “needs”,

examining the ways in which the needs of society or individuals are met. A hard
fact is that people and societies need mineral commodities to prosper. By all
measures (see, for example, the Mining, Minerals and Sustainable Development
initiative: http://www.iied.org/mmsd/ ) the demand for minerals will rise inexorably
over the coming decades, in response to increased development. Of course,
there is also a drive that the exploitation of these resources is ecologically and
socially sensitive. There is not a single priority addressed the NERC Strategy that
does not need or revolve around the resources of natural minerals. From
approved projects addressing fundamental ore genesis, to dealing with
remediation strategies for acid mine drainage, ICSF is involved with collaborators
in shaping science that contributes to the heart of Sustainable Economies. ICSF
is committed to revealing and explaining the Earth's resources and supporting
their sustainable use in an environmentally conscious manner to provide
significant benefits for society. How do we address this?

Isotope data production and postgraduate student training in the technical

and interpretative use of light stable isotope (H, C, O & S) analyses lie at the core
of the service we deliver. Students gain access via ICSF to the international-
class suite of stable isotope systems at SUERC for analyses of fluids, minerals
and organic compounds, with access as appropriate to SUERC's comprehensive
radiogenic systems. Since 1998, over 90% of ICSF's approved projects have an
associated PhD student undergoing such intensive training in the principles and
practice of isotope geochemistry. ICSF core support will continue to centre on
this student training. In the past year, for example, 11 PhD students (3
NERC/CASE, 3 NERC, 1 EU, 1 DENI, 1 Industry, 1 Fermor Studentship, 1
University funded) received intensive training. In this way , ICSF continues to
support and promote the government's priority for the long term health of
the science base . Developing skilled people is a specific outcome of the
NERC strategy, and we work closely with PI's to ensure that the science tackled
is of a high standard, and that the student's experience is enjoyable and
rewarding.

ICSF has been at the forefront of development of new isotope systems for use by
the Community. NERC Scientific Services and Facilities have financially
contributed to these developments through capital input (typically shared with
SUERC). As a result, the range of isotope systems offered to the core
Community is second-to-none in the UK. For example:

• ICSF/SUERC hosts one of the world's first in situ laser S isotope system for
sulphides, and recently, helped develop the world's first in situ laser S isotope
system for anhydrite.
• One of only two UK sulphate oxygen systems.

• Total laser fluorination system for O isotope work on silicates and oxides,
allowing sample sizes down to 0.4mg of SiO 2 .

• The UK's only inclusion fluid H isotope system.

• Automated carbonate preparation system for rapid, high quality analyses of C

and O isotopes in all types of carbonate.

Output is primarily measured by (i) peer-reviewed publication , and (ii) the

production of motivated young scientists, trained to fulfil the strategic needs of
the UK, particularly (but not exclusively) in the field of Applied Minerals research.

From 1999-2003 ICSF has published 40 peer-reviewed papers in international

journals. ~80% in journals with impact factors over 1, and ~25% in journal with IF
>2, including:

11 papers in Economic Geology

4 papers in Mineralium Deposita

4 papers in Geochimica Cosmochimica Acta

2 papers in Geology

2 papers in Chemical Geology

All of these publications are collaborative, reflecting the support ICSF gives to
projects from beginning to end.

Since its inception in 1991, 92% of students from ICSF approved projects have
gone on to work in the mining/exploration/environmental industries (44%), or are
applying their skills in academia (48%).

ICSF therefore offers projects of high scientific quality the following:

• Comprehensive and flexible access to one of Europe's foremost isotope

laboratories;

• An excellent, professional training environment for PhD students and

postdocs;

• A proven commitment to quality, international-level peer-reviewed

publication.
The NERC Isotope Community Support Facility at SUERC welcomes all
enquiries – if we can help, just let us know by emailing as above, or calling Dr.
Adrian Boyce on - 1355 270 143. Check out our poster too!

REE mobility and mass balance during hydrothermal alteration of

carbonates: A case study from the Galmoy deposit, Ireland

H.L Crowther 1 , J.J Wilkinson 1 , T.Jeffries 2 and B.Coles 1

1 Department of Earth Science and Engineering, Imperial College London.

([email protected])

2 Mineralogy Department, Natural History Museum, London.

Introduction

The Rare Earth Elements (REE) are commonly used in many aspects of geology;
particularly igneous petrology, although as yet their application to ore deposit
studies has been surprisingly limited and their mobility under hydrothermal
conditions is poorly constrained. This investigation explores the effects of mass
change on geochemistry of carbonates formed during hydrothermal alteration in
a carbonate-hosted base metal ore deposit, and examines the the mobility of the
REE in this system.

Background

The Galmoy deposit is hosted by Carboniferous Waulsortian limestone, and is

enveloped by rock matrix breccias (RMB), widely recognised as being intimately
associated with economic sulphide mineralization in the Irish orefield. The black
dolomite matrix to these breccias is known to be of hydrothermal origin, formed
by dissolution-replacement (and possibly some open-space precipitation) within
the regionally-dolomitized Waulsortian precursor. Previous investigations
(Lowther et al., 1999) have suggested that a “Central Fissure” is the main fluid
conduit responsible for formation of the CW orebody at Galmoy.

Fourteen RMB samples were collected from drillcore from the CW orebody,
describing a SW-NE oriented traverse perpendicular to the central fissure. These
samples were located in the immediate hanging wall of the massive sulphide ore.
Several reference samples were also collected, including unaltered Waulsortian
limestone and Waulsortian dolomite. Black dolomite matrix and clast material
were separated from the RMB samples by micro-drilling and analysed for REE
and minor elements using ICP-MS and ICP-AES, respectively at the Natural
History Museum, London.
Results

A strong mutual positive correlation occurs between Ti, Al and K. Ti and Al are
high field-stength elements, commonly considered to be immobile in
hydrothermal systems. Their correlation with K suggests that all three elements
are rapidly incorporated into recrystallized aluminosilicates within the RMB
matrix, and are probably immobile on the scale of sampling.

Mass change values derived from these immobile element concentrations

indicate that in some samples, >90% mass loss has occurred during formation
and alteration of RMB matrices. The observed mass loss profile is symmetrical
about the central fissure. Maximum mass loss values occur at the centre of the
orebody. Such mass loss may have resulted in extensive collapse, perhaps
represented by the abundant white dolomite cemented breccias that form a
broad halo above and lateral to the RMB unit.

All samples associated with sulphide mineralization exhibit a marked increase in

REE content compared to the unaltered Waulsortian dolomite precursor. A slight
positive Eu anomaly is observed in all samples (Figure 1), which is most
pronounced directly above the central fissure.

Figure 1 Eu/Eu* either side of central fissure

(shown as dashed line). (Eu*=(Sm+Gd)/2)

Figure 2 REE retention patterns for hydrothermal

dolomite matrix samples

When considered in relation to the known mass loss in the system, it is apparent
that all samples have undergone significant REE depletion, with HREE more
effectively removed than LREE (Figure 2). Eu is preferentially retained in all
samples, probably due to the presence of Eu 2+ , which is more effectively bound
in the carbonate lattice than trivalent REEs, due to its similar size and charge to
Ca 2+ . Therefore the observed Eu anomaly directly above the central fissure
structure (Figure 1) is attributed to less effective redistribution of Eu by outward
moving hydrothermal fluids, rather than net addition of Eu. In theory, the
redistributed REEs may form a halo surrounding an orebody, potentially
providing a useful exploration indicator.

Similar redistribution patterns are observed for Mn, which is depleted proximal to
the central fissure and enriched towards the periphery of the CW orebody.

Acknowledgements: The authors would like to thank Arcon Mines for allowing
access to the Galmoy Mine. Imperial College London, University College Dublin,
The Natural History Museum, London and the Society of Economic Geologists
(BHP student grants), provided financial support.

Reference:

Lowther J.M., Balding A.B., Dunphy S., McEvoy F.M. and Bowden A.A. (1999)
Proc. 5th Biennial SGA Meeting, London, 1999. A.A. Balkema, Rotterdam, v. 2,
p. 881-884.

Integration of spatial geophysical data by geostatistical simulation

 P. A. Dowd & E. Pardo-Igúzquiza

Department of Mining and Mineral Engineering, University of Leeds, Leeds LS2

9JT

[email protected]

Studies of the mechanical, hydrogeological and structural characteristics of an

underground rock volume are often based on two or more types of data. For
example, characteristics can be measured directly on drill cores of rock or
indirectly by wireline geophysical logs along boreholes or from three-dimensional
seismic surveys. There are usually qualitative and quantitative differences among
the different types of data. For example, core data provide direct measurements
of the variable of interest whereas wireline logs and seismic surveys provided
indirect measurements; the quality and reliability of wireline data may be
significantly better than seismic data. For qualitative or/and quantitative reasons
one variable is chosen as the primary variable and the others are designated
secondary variables. The purpose of a data fusion procedure in this context is to
produce a complete as possible image of the variable of interest (e.g., porosity,
hydraulic conductivity) by incorporating and integrating all available data. One
possibility is to estimate the variable of interest using a multivariate procedure,
such as cokriging, which also provides an evaluation of the uncertainty of the
estimate. One difficulty with this approach is that the estimated image is much
smoother than reality. Geostatistical simulation is an efficient alternative that can
be used to generate a large number of stochastic images, each of which is a
possible realisation of reality that reproduces the spatial variability observed in
the experimental data and reproduces the experimental data at the sampled
locations. Data integration is accomplished by requiring the simulated images to
be coherent with the experimental data provided by the secondary variables. The
authors describe the methodology of data fusion by simulation and illustrate its
application by merging wireline acoustic impedance data (primary data)
measured along boreholes with acoustic impedance data from a 3D seismic
survey (secondary data). The wireline acoustic impedance data are more
accurate and reliable measurements than those provided by the 3D seismic
survey but they are available from only nine boreholes that represent only a small
proportion of the total rock volume. Although the 3D seismic acoustic impedance
data are of significantly lower quality they provide a complete coverage of the 3D
rock volume. Integrating the two types of data should provide more realistic
images, and hence images that are more useful in practical applications, than
images obtained from using only one type of data.
Fig. 1: Two-dimensional tomogram between two boreholes in the Sellafield
Potential Repository Zone

Fig. 2: Geostatistically simulated images of acoustic impedance over the same

area as that shown in Fig. 1. The geostatistical images have integrated acoustic
impedance data from wireline and seismic sources, accounting for the relatively
higher measurement errors in the latter. The simulations provide a means of
quantifying the uncertainty in models based on sparse and/or error prone data as
wells as a means of visualising the associated risk.

A Powerful New Magnetic Derivative Method for Mapping Mineral Prospects

J Derek Fairhead 1,2 , Bruno Verduzco 2 and Chris MacKenzie 3

1. GETECH, School of Earth Sciences, University of Leeds, LEEDS LS2 9JT UK.
[email protected]

2. School of Earth Sciences, University of Leeds

3. BAFEX Exploration (Pty.) Ltd. [email protected]

 
Mapping the spatial extent and boundary locations of magnetic bodies using
aeromagnetic data usually requires the data to be initially transformed i.e.
‘Reduced to Pole' (RTP), ‘Reduced to Equator' (RTE) or pseudo gravity, prior to
deriving the Total Horizontal Derivative (THDR). The maxima of the THDR help
to define the edges of magnetic bodies. The field transformation by necessity
assumes that all magnetic bodies have induced magnetisation. If this assumption
is incorrect then the inferences on edge detection can be in error. A derivative
method considered to overcome this assumption, and thus able to be applied
simultaneously to bodies containing remanent and induced magnetisation, is the
Analytic Signal. This derivative method has major practical limitations and is not
generally used.

This contribution demonstrates that the Tilt derivative overcomes the short
comings of the Analytic Signal and has several major advantages in structurally
mapping the spatial extent of mineralised bodies with internally varying
magnetisation. The theory behind the method will be show and demonstrated on
simple 2D models to show that the Tilt derivative can be used in various ways to
be more effective substitute for AGC, vertical derivative and edge detection
filters. The method is then be applied to high resolution aeromagnetic data
covering part of Namibia. Here the results of applying this new
mapping/interpretation technique may have significant implications on our
understanding of the mineralisation that occurs within the BAFEX Erindi gold
prospect.

High resolution orebody definition using 3D-seismics: Examples from the

Bushveld Platinum deposits.

Steve Freeman, Graham Stuart 1 , Schalk Kock 2 , and Peter Mellowship 3

1 RDR Ltd., Dept. of Earth Science, Leeds University, UK.

[email protected]

2 Dept. of Earth Sciences, Leeds University, UK.

3 Impala Platinum, Rustenburg, RSA.

3D-seismics data has been used to define the Merensky and UG2 platinum reefs
across the entire Impala Platinum lease ground. This resource represents a
major proportion of the worlds predicted demand for the next several decades.
The 3D seismic reflection acquisition, interpretation and modelling programme
has been used to define shaft placements, decline programmes, ore delineation
borehole patterns right through to haulage locations, stope panel geometries and
extraction scheduling. A several year work programme of interpretation,
reprocessing, and advanced geological modelling has been used to define the
orebody geometry. As an example of the benefits that the seismic programme
has yielded;

 Faults with offsets well below 10m of throw have been routinely mapped at
1000-2000m below surface,
 Continuous dip panel blocks have been outlined and targeted for
extraction, whereas structurally complex ground has been delineated and
removed from the current extraction plans.

This current initiative represents one of the most integrated applications of 3D

seismic reflection data in mine planning currently within the industry. The
programme also represents the merging of front line academic and software
technological developments with direct commercial drivers.

Clear economic benefits have been delivered by 3D-seismics in the deeper parts
of the Bushveld complex. The massive potential for this technique in other strata-
form / strat-bound deposits remains as yet either under utilised or untested.

Multi-scale fluid flow path analysis: Why there is a lot of zinc at Navan

S. Freeman 1 , C. Everett 2 , J. Wilkinson 2 , A. Boyce 3 , J. Cleverley 4 , A.

Barnicoat 1 , B. Yardley 4 , G. Phillips 1 , P. Jones 1 , C. Graham 5 , J. Ashton 6
& R. Blakeman 7

1 RDR, School of Earth Sciences, University of Leeds, Leeds LS2 9JT

2 Department of Earth Science and Engineering, Imperial College London

3 Isotope Geosciences Unit, SUERC, East Kilbride, Glasgow G75 0QF

4 School of Earth Sciences, University of Leeds

5 Department of Geology and Geophysics, University of Edinburgh

6 Outokumpu-Tara Mines Ltd, Knockumber, Navan, Co Meath, Ireland

7 Division of Earth Sciences, Department of Geography and Topographic

Science, University of Glasgow

This project funded through the NERC Micro to Macro programme looked at fluid
flow at a variety of scales to investigate the genesis of the Navan deposit. An
integrated study was carried out of field relations, especially the links between
grade and structure, petrography, stable isotope distributions and fluid
composition. Coupled physical and chemical modelling was used to investigate
the controls on the siting of the deposit.

The ore body is locally stratabound but also extends along major faults that
originated during the Carboniferous. Detailed grade mapping reveals that the
highest grade ores pick out a fracture pattern that predates the main faults
however, and is immediately earlier. The orientations indicate reactivation of
Caledonian trends in the adjacent basement. While the preponderance of
isotopically light S in the ore bodies is well known, detailed investigations have
recently shown that ore minerals in the major faults have a heavier signature.
Fluid inclusion studies provide evidence, as at other Irish orebodies, of mixing of
a hotter, less saline fluid with a cool brine.

The overall chemical model for the origin of the orebody is that a hot, metal-rich
fluid with a minor content of heavy sulphur, interacted with a cool bittern brine
generated at the Carboniferous sea bed due to evaporation, and containing
biogenic sulphide, which then sank through the sediment column. Our results
show that effective mixing of the fluids occurred prior to the development of major
through going faults, in the precursor fracture mash, that permitted more
complete mixing. Once the main faults developed they provided a fast conduit for
seismic expulsion of metal-rich fluid, which precipitated only a small part of its
metal load with the minor, heavy sulphide carried in the same fluid.

Genesis of the Angouran Zn-Pb deposit, NW Iran

H. A. Gilg 1 , M. Boni 2 , G. Balassone 3 , C. Allen 4 , F. Moore 5

1 Lehrstuhl für Ingenieur-Geologie, Technische Universität München, Germany

2 Dipartimento Geofisica & Vulcanologia, Universitá di Napoli, Italy

3 Dipartimento Scienze della Terra, Universitá di Napoli, Italy

4 TeckCominco American, Spokane, U.S.A.

5 Geological Department, University of Shiraz, Iran

The Angouran Zn-Pb deposit is situated in the Zanjan Province, NW Iran,

approximately 450 km from Tehran. It is one of the largest and highest-grade
non-sulfide zinc deposits in the world, and additionally hosts a very high grade
sulfide ore body. The resources in 1999 were estimated at 13.5 Mt of non-sulfide
ore with 26.4% Zn and 4.5% Pb, and 3.2 Mt of sulfide ore at 37.0% Zn and
2.32% Pb, plus 2.0 Mt of mixed sulfide-carbonate ore at 31.2% Zn and 4.1% Pb.
The carbonate ores are currently exploited in an open pit around 2700 m above
sea level.

Angouran is located within a metamorphic complex in the central Sanandaj-Sirjan

Zone of the Zagros collisional belt close to the Urumieh-Dokhtar magmatic arc.
The metamorphic rocks comprise a series of amphibolites, micaschists, and
marbles of supposed Precambrian (to Cambrian) age and greenschist to
amphibolite facies conditions. The crystalline complex shows internal thrusting
and is thrusted onto Tertiary sediments at its western margin. The basement
rocks are unconformably overlain by Miocene intermediate to acidic pyroclastic
and intrusive igneous rocks, and by a succession of sedimentary deposits. The
latter consist mainly of shallow marine limestones of the Aquitanian Qom
Formation and evaporite-bearing, red marls and sandstones of the Upper Red
Formation. Quaternary travertine terraces complete the stratigraphic section.

The sulfide orebody (600m long, 200-400m wide and up to 50m thick) is more or
less tabular, restricted to the lithological boundary between footwall micaschists
and hanging wall marbles of the metamorphic complex, and is situated in the
crest of an anticlinal structure that plunges southward at 30°. The sulfide ore
body is capped by up to 200m thick Zn carbonate ores that protrude discordantly
into the hanging wall marble. Both ore types are closely associated to breccias
especially at three major fault zones that flank the orebody.

Sulfide ores are generally massive and often breccia-textured containing clasts of
micaschists and marbles, as well as clasts of earlier sulfide ores indicating
multiple brecciation events during mineralization. The dominant ore mineral is
Fe-poor sphalerite often showing oscillatory zoning. Sulfide mineralization cuts
the schistosity of wall rocks. Galena, very little pyrite, and traces of Ni-Co
arsenides complete the sulfide paragenesis. Accessory gangue minerals are
euhedral quartz, muscovite, dolomite and lath-shaped anhydrite (often as relic
ghost textures). Sulfide ores are characterized by high Ag (210 ppm), As (760
ppm), Sb (300 ppm), Hg (20 ppm) , Co (400 ppm) and Ni (350 ppm) contents.
Honey-brown sphalerite contains abundant primary two-phase fluid inclusions
with homogenization temperatures of 100 to 150°C. Low first (-45°C), hydrohalite
(-35°C) and ice melting (-24 to -26°C) temperatures indicate Ca-Na-Cl brines with
high salinity (23-25 wt.% TDS) and Ca dominance. Preliminary crush-leach
analyses (D. Banks, Univ. Leeds) show very high and variable Br/Cl ratios
suggesting fluid mixing and probably the participation of highly evaporated sea
water. Sulfur isotope compositions of sphalerite and galena are homogeneous at
+7 ± 1 ‰ (V-CDT). Reconnaissance Pb isotope data are also homogeneous and
cluster around 18.921 ( 206 Pb/ 204 Pb), 15.700 ( 207 Pb/ 204 Pb), and 38.908
( 208 Pb/ 204 Pb). These isotope ratios indicate an upper crustal Pb source and
young (Cenozoic) Pb-model ages. Sulfide ores unambiguously postdate regional
metamorphism of the wall rocks , which according to preliminary 40 Ar- 39 Ar
data on sericite has a Miocene age (~20 Ma). Sulfide ores cannot be classified
as SEDEX or VMS type mineralization. The simple paragenesis, presence of
anhydrite laths, abundance of breccias, Fe-poor character, and especially the
fluid inclusion characteristics are suggestive of MVT -type mineralization.
However, homogeneous sulfur isotope values and the characteristic trace
element signature might indicate participation of magmatic components to the
ore-forming hydrothermal system.

The non-sulfide ores can be subdivided into two paragenetic stages, an early and
predominant stage I carbonate(-sulfide) ore and a late stage II carbonate(-oxide)
ore. Stage I carbonate ores are dominated by smithsonite showing a variety of
textures from massive to brecciated, from botryoidal to euhedral dog tooth-
shaped, from vuggy to dense. This mineralization stage occurs also within sulfide
ores filling cavities of former anhydrite laths or replacing marble clasts in the
sulfide ores without any apparent oxidation of the sulfides. Stage I smithsonites
exhibit abundant tiny, euhedral arsenopyrite, and less abundant quartz, galena,
pyrite and greenockite crystals along specific growth zones. In some areas, the
stage I smithsonite-arsenopyrite assemblage seems to replace earlier sulfide
ores, in others, especially at the flanking fault zones of the orebody, direct
replacement of marbles by zebra-textured smithsonite-calcite ores can be
observed. Stage I carbonate ores are enriched in Pb (7%), As (6100 ppm), and
Sb (570 ppm) with respect to sulfide ores but have similarly high Ni and Co
contents. The co-precipitation of sulfides, the absence of oxides and the
presence of reduced As species exclude the formation of stage I carbonate ores
in a supergene oxidation environment. The stability of smithsonite plus quartz
implies maximum formation temperatures of ~100°C and high CO 2 fugacities.
The paragenesis smithsonite plus arsenopyrite is stable under these conditions
in reducing environments at a pH of ~8 to 9, thus very close to the sphalerite
stability field. Highly variable oxygen isotope values of stage I smithsonites (18.3
to 23.6 ‰ V-SMOW) and very high and homogeneous d 13 C values (+5 ± 1 ‰
V-PDB) are not typical for supergene smithsonites, but suggest formation in a
subsurface hydrothermal environment. We note that the travertine deposits very
close to the Angouran mine have identical, heavy carbon isotope values (+5.4 ‰)
and are rich in Zn (970 ppm), Pb (124 ppm) and As (124 ppm).

Stage II carbonate ores are found as colloform bands in veins crosscutting

massive stage I carbonate ores, or as euhedral crystals in vugs and open
fractures. However, these ores amount only to a very minor part of the non-
sulfide ore. Stage I I carbonate ores are characterized by the assemblage
smithsonite, hemimorphite, mimetite, goethite, hematite, Mn-oxides, calcite and
cerussite. Oxygen isotope values of stage II smithsonites are homogeneous at
+25 ± 1 ‰ and d 13 C values quite variable from +5 to -0.8 ‰, thus clearly
distinct from stage I smithsonites. The isotope characteristics and mineral
paragenesis of stage II carbonate ores, including the presence of arsenates and
oxides identif y them as classic supergene oxidation ores.

Calcites occurs in and around the Zn carbonate orebody in a great variety of

forms, such as euhedral crystals in vugs, massive sparry fissure fillings, but also
as colloform crusts in karst-like cavities in the marble. The stable isotope
compositions of these calcites show a negative correlation ranging from d 18 O
values of 16 ‰ and d 13 C values of +5 ‰ to d 18 O values of 22 ‰ and d 13 C
values of -2 ‰. The isotope variation of calcites can be interpreted as mixing of
13 C-rich hydrothermal fluids with cold, 12 C-rich ground waters.

Our comprehensive geological, petrographic and geochemical study clearly

shows that two superimposed, but chemically contrasting hydrothermal systems
shaped the Angouran Zn-Pb deposit. An early hot (>100°C), sulfur- and salt-rich,
but CO 2 -poor fluid formed the sulfide orebody of mixed MVT and low-
temperature CRD character. In contrast, a cooler (<100°C), reduced, As-, Pb-
and CO 2 -rich, but probably salt-poor meteoric hydrothermal fluid deposited the
major part of the Zn carbonate ores. We suggest that low-temperature
hydrothermal carbonatization of sulfide ores at Angouran is related to the
subsurface activity of a travertine-depositing system.

Base metal sulphide geochemistry of the Merensky Reef, Bushveld

Complex, South Africa

D. A. Holwell 1 , R. G. Cawthorn 2 , J. C. Ø. Andersen 3

1 School of Earth, Ocean and Planetary Sciences, Cardiff University, Main

Building, Park Place, Cardiff, CF10 3YE, UK. [email protected].

2 School of Geosciences, University of the Witwatersrand, PO Wits 2050, South

Africa. [email protected].

3 Camborne School of Mines, University of Exeter, Redruth, Cornwall, TR15

3SE, UK. [email protected].

The Bushveld Complex, South Africa, hosts over half the world's reserves and of
platinum group element (PGE) deposits (Cawthorn, 1999). The best known of the
PGE deposits of the Bushveld Complex is the stratiform Merensky Reef. The reef
is comprised of a pyroxenite, commonly pegmatoidal with variable amounts of
chromite, with PGE hosted by, or spatially associated with an assemblage of
interstitial base metal sulphides. Despite its huge economic significance, the
processes involved in concentrating the PGE are still debated, and there are
currently four generalised models. The first involves an immiscible sulphide liquid
separating from silicate magma during a magma mixing event. The sulphide
droplets scavenge base metals and PGE as they settle to the floor of the magma
chamber to form the reef (Campbell et al , 1983). The second model involves
PGE and sulphur being scavenged from the underlying cumulates by upward
percolating interstitial Cl-rich fluids, with a subsequent deposition at a chemical
reaction front (Vermaak, 1976, Boudreau & Meurer, 1999). The third model
suggests that clusters of PGE atoms suspended in the melt were collected by
carriers such as an immiscible sulphide liquid or by chromite sinking to the floor
of the magma chamber (Tredoux et al, 1995). The forth model involves PGE
being scavenged by a fluid from primary sulphides precipitated in the Critical
Zone rocks. The fluid reaches the crystal-magma interface and redissolves,
causing an immiscible sulphide, rich in PGE, to form (Boudreau, 1999).

This study investigates the base metal sulphide geochemistry in two distinctly
different parts of the Merensky Reef. Samples were taken from Impala Platinum
Holdings Ltd.'s Impala Mine, in the western Bushveld, and Marula Mine, in the
eastern Bushveld. The aim is to determine whether the sulphide assemblage has
a magmatic signature, or whether there is any evidence of later fluid overprint. It
is possible to predict the Ni and Cu content of a magmatic sulphide liquid if
certain factors are known: namely the partition coefficient of the metal between
silicate magma and sulphide liquid, the initial concentration of the metal in the
magma, and the magma/sulphide liquid ratio (R-factor), though this is of lesser
importance for Ni and Cu. For a magmatic sulphide separating from an upper
Critical Zone magma at the level of the Merensky Reef, it can be estimated that
the resulting sulphide liquid would have contained 10%Ni, 5%Cu, ~30%S and
55%Fe (Naldrett, 1981), giving S:Ni:Cu ratios of around 6:2:1. Hydrothermal
signatures would vary highly depending on the nature of the fluid and cannot be
easily predicted.

Six core sections from Impala Mine were analysed and found to have average
Ni/Cu ratios of 1.8 and S/Cu ratios of 6.2, in line with that expected for a
magmatic sulphide. The ratios show a slight enrichment of Cu with height,
possibly indicating fractionation of the sulphide liquid. PGE abundances mirror
those of Cu, Ni and S. There is no evidence of any hydrothermal or metasomatic
overprinting. Thus it can be assumed that at Impala Mine, the sulphides in the
Merensky Reef formed by separation of a sulphide liquid from silicate magma,
with no evidence of further alteration processes.

Two core sections from Marula Mine were also analysed and yielded average
Ni/Cu ratios of 1.5 and a S/Cu ratio of only 4.1, indicating significant sulphur loss
from the ideal magmatic sulphide liquid, or that the sulphides formed under
different conditions. Sulphur can be lost from a sulphide melt by oxidation, which
typically results in the formation of hydrous silicates, bornite and magnetite.
Secondary amphiboles are found in the Marula Mine samples overprinting
primary orthopyroxene and sulphides, however, no bornite or magnetite is
present. In contrast to what would be expected with an oxidising fluid, pyrite is
much more abundant than pyrrhotite. A “normal” Merensky Reef sulphide
assemblage, typified by that at Impala, contains 44% pyrrhotite, 34% pentlandite,
17% chalcopyrite and 4% pyrite. The Marula Mine assemblage is made up of
27% pyrrhotite, 36% pentlandite, 20% chalcopyrite and 17% pyrite. Also
associated with the sulphides are appreciable amounts of interstitial, Fe-rich
chromite. Iron loss from pyrrhotite to chromite via the relationship Naldrett and
Lehmann (1987):

4 / 3 Fe 2 O 3 (spinel) + 1 / 3 FeS à Fe 3 O 4 (spinel) + 1 / 6 S 2

could explain the increase in pyrite over pyrrhotite, though not the apparent loss
in sulphur. This was possibly caused by metasomatic fluids which formed the
overprint of hydrous minerals. These fluids mobilised Cu, obliterating any
fractionation trends analogous to those present in the Impala sections, that may
have been present in the Marula Mine sections, and produced veins and fracture-
fills of sulphide, rich in chalcopyrite. As at Impala Mine, PGE abundances mirror
those of Cu, Ni and S.

The evidence from this study is consistent with the model involving Merensky
Reef sulphides forming from an immiscible sulphide liquid separating from
silicate magma. Due to the intimate association of PGE with the unaltered
magmatic sulphides at Impala Mine, it is likely the PGE were collected by this
sulphide liquid. In terms of the Boudreau mode, scavenging of PGE from the
footwall ought to have also scavenged Cu and Ni, yielding anomalously high Cu
and Ni in the magma and resultant sulphide. In the Marula Mine profiles, there is
evidence for a probable post-magmatic hydrothermal event that has affected the
primary sulphide composition, but such evidence is lacking in the Impala Mine
profiles. Despite this difference, the PGE contents in the two settings are
comparable, and there is no evidence for addition or redistribution of PGE by
such fluids.

References:

Boudreau, A. E., 1999. J. Pet, 40 , 755-772.

Boudreau, A. E. and Meurer, W. P., 1999. Cont. Min. Pet. 134 , 174-185.

Campbell, I. H., Naldrett, A. J. and Barnes, S. J., 1983. J. Petrology, 24 , 133-

165.

Cawthorn, R. G., 1999. S. Afr. J. Sci., 95 , 481-489.

Naldrett, A. J., 1981. in: Cabri, L. J. (ed). Can. Inst. Min. Metall., Montreal,
Canada, 23 , 197-232.

Naldrett, A. J. and Lehmann, J. 1987. in: Prichard et al, Geo-Platinum '87, 113-
143.

Tredoux, M., Lindsay, N. M., Davies, G. and McDonald, I., 1995. S. Af. J. Geol.,
98 , 157-167.
Vermaak, C. F., 1976. Econ. Geol. 71 , 1270-1298.

Mineralisation potential mapping for ophiolite-hosted volcanic massive

sulphide (VHMS) deposits, Cyprus

Jowitt, S.M. 1 , Naden, J. 2 , Nicolaides, S. 3 , Bateson, L. 2 , Rollin, K.E. 2 and

Gunn, A.G. 2 .

1 Camborne School of Mines, Redruth, Cornwall, UK, TR15 3SE

([email protected])

2 British Geological Survey, Kingsley Dunham Centre, Keyworth, Nottingham,

NG12 5GG, UK ([email protected])

3 Geological Survey Department, Ministry of Agriculture and Natural Resources,

1415, Nicosia, Cyprus ([email protected])

The island of Cyprus, and its well-preserved Troodos ophiolite sequence, has
been an important source of copper since pre-Roman times. However, with the
diminishing market for pyrite used in sulphur extraction, the mining and
exploration industry on the island was in decline until the late 1990's. With the
reopening of mining operations in the Skouriotissa region a new phase of
exploration for VHMS copper deposits began. This exploration programme,
unlike previous programmes, aims to discover buried deposits underneath the
autochthonous sedimentary cover of the ophiolite sequence. This project seeks
to determine whether realistic regional scale exploration targets for surface and
buried VHMS mineralisation in the Troodos ophiolite sequence can be generated
using spatial data from a variety of sources.
Figure 1 Zones of high copper mineralisation potential on Cyprus compared to
current and former massive sulphide and stockwork mineralisation mining
districts. (A = Ampelikou / Apliki / Agios Epiphanos Soleas area, B = Agios
Georgious Kafkallou area, C = Klirou / Malounta area, D = Agia Varvara
Lefkosias / Sia / Mathiatis area, E = Lythrodontas area, F = Arakapas /
Dierouna / Eptagoneia area, G = Asgata area, H = Mandria Lemesou area.)

Using a GIS-based prospectivity analysis and a weights-of-evidence

methodology (e.g. Kemp et al. , 2001; Tangestani and Moore, 1998) in
conjunction with legacy data and recently available data sources, it is possible to
define realistic areas of high mineralisation potential. These areas of high
potential can then be considered as targets for future, detailed, exploration work.
Geological data, lithological contact data, Bouguer gravity anomaly data, fault
data, gossan occurrence data, ASTER satellite lineations and band ratio images
and Landsat Thematic Mapper band ratio images were all found to have a spatial
relationship with mineralisation. ASTER satellite data was ground-truthed using
PIMA analyses from mine sites, gossans and two deep boreholes. This ground-
truthing enabled the identification and differentiation of 6 individual band ratios
relating to alteration associated with economic mineralisation.

These datasets were used as evidential themes (maps) against the distribution of
known VHMS-related mineral occurrences (training point theme) to establish
areas of high mineralisation potential. Seventeen datasets produced from the
data within the GIS had identifiable relationships with mineralisation, and binary
maps showing areas of prospectivity and non-prospectivity for VHMS
mineralisation were produced. Finally, each of these datasets were integrated
into one final prospectivity map, identifying eight areas of high mineralisation
potential (Figure 1). Three of these areas are outlined as possible targets for
future exploration due to their size, lack of previously known mineral occurrences
and proximity to known mineral occurrences and mining districts. This work has
demonstrated the applicability of GIS-based approaches to exploration on
Cyprus and could prove to be a useful tool for determining areas for future
exploration work.

Although exploration geologists working on Cyprus already know most of the

areas defined as having high mineralisation potential, further work should enable
the identification of target areas for blind VHMS mineralisation. The targeting of
blind deposits is an area of research and exploration ignored until recently on
Cyprus. Further development of the GIS system will aim not only to incorporate
more data, such as geochemistry and geophysics, but also to consider
alternative prospectivity methodologies, such as fuzzy logic. In addition, the
deposit model for Cyprus should be updated, as it is probably no longer
appropriate to consider the Troodos as a simple subduction zone ophiolite.

Furthermore, the GIS methodology used ensures that updating of probability

maps is straightforward once additional data is in a suitable format.

Acknowledgements:

This work forms part of an M.Sc. dissertation by SMJ undertaken at the

Geological Survey Department (Cyprus) and The British Geological Survey (UK),
who allowed access to in-house computing facilities and a variety of digital data.
In addition, Hellenic Mining Company, Cyprus and Eastern Mediterranean
Minerals are also thanked for giving access to prospects and mines on Cyprus.
SMJ acknowledges the financial support of a NERC studentship funding the
M.Sc. at Camborne School of Mines.

References:

Kemp, LD., Bonham-Carter, GF., Raines, GL. and Looney, CG., 2001. Arc-SDM:
Arcview extension for spatial data modelling using weights of evidence, logistic
regression, fuzzy logic and neural network analysis.
http://ntserv.gis.nrcan.gc.ca/sdm/ .

Tangestani, MH. and Moore, F., 2001. Porphyry copper potential mapping using
the weights-of evidence model in a GIS, northern Shahr-e-Babak, Iran. Australian
Journal of Earth Sciences, 48: 695-701.
Mineral textures of the Laisvall Sandstone-Hosted Pb-Zn deposit and their
implications for the Pre-ore permeability.

Lucks, T.J. 1 Barnicoat, A. 2 Freeman, S. 3 Knipe, R. 3

1 University of Leeds, England. Contact: [email protected]

2 Geoscience Australia, PO Box 378, Canberra Act 2601, Australia.

3 RDR, University of Leeds, England.

Previous studies of the Laisvall deposit (Rickard et al. 1979; Linblom 1982) have
neglected to identify and describe the microscopic pre-ore environment of the
Laisvall sandstones. This study identifies, by comparison of the ore hosting units
to those of uneconomic value, critical factors involved in developing a successful
host for mineralisation.

The Laisvall sandstone-hosted lead-zinc deposit is located along the present day
eastern margin to the Swedish Caledonides. The ore stage galena and sphalerite
reside within the pore spaces of two quartz arenite horizons within the
autochthonous sediments of the Laisvall Formation.

Prior to mineralisation the Laisvall sands developed as “normal” quartz arenites

undergoing burial/ diagenesis to a depth of 3-4 kilometres (Lucks et al. 2002).
During diagenesis a number of cements formed related to the primary mineralogy
of the unit in question. These cements significantly altered the porosity/
permeability structure of the units. Of the pre-ore cements to develop, feldspar
authigensis was the most prolific and if the Laisvall detrital feldspar content had
been as high as the neighbouring Maiva deposit it would have been unlikely that
the Laisvall deposit would have become economic.

The onset of mineralisation was marked by a period of quartz dissolution

affecting the host rock, increasing porosity and permeability. In extreme cases
producing dissolution pathways through sandstone, initiating a positive feed back
loop.

The Tjalek Member, a 8-10m thick unit separating the Upper and Lower
Sandstone ore bodies is typically un-mineralised, although composing an almost
identical primary mineralogy to the ore hosting units. The development of quartz
and illite cements prior to mineralisation reduced porosity and pore space
connectivity, plugging pore throats.

The petrology of the surrounding stratigraphy to the ore hosting horizons reveals
significant implications for possible paths for large scale fluid flow supplying the
Laisvall ore body. In particular, the underlying Saivatj Member (shale) and the
Ackerselet Formation (arkose) are unsuitable conduits due to their poor
permeabilities affected by diagenetic cementation.

This implies the fluids involved in mineralisation were either resident within the
sandstones, transported laterally along the ore hosting horizons, formed from de-
watering of the overlying silts, sourced along fractures linked to the basement
rocks or leaked directly into the sandstone at the point of origin. The next step to
understanding the genesis of the Laisvall deposit is combining this information
with structural data, grade modelling and isotope geochemistry.

References:

Rickard, D. T. Wildèn, M. Y. Marinder, N. E. Donelly T. H. (1979) “Studies on the

genesis of the Laisvall sandstone lead-zinc deposit, Sweden”. Economic
Geology. Vol. 74, No. 5. PP. 1255-1285.

Linblom, S. (1982) “Fluid inclusion studies of the Liasvall sandstone lead-zinc

deposit, Sweden”. Doctoral thesis. Stockholms Universitets Geologiska Institution
N;r 252.

Lucks, T. J. Barnicoat, A. Freeman (2002) “Porosity and permeability controls

within sandstone lithologies of the Laisvall Pb-Zn, N. Sweden”. Trans. Inst. Min.
Min. App. E. S., 226.

Contribution of infrared microthermometry to fluid inclusions studies in

opaque minerals: Hematite-hosted fluid inclusions from Au–Pd–Pt-bearing
quartz–hematite veins, Quadrilátero Ferrífero, Minas Gerais, Brazil

Volker Lüders

GeoForschungsZentrum Potsdam, D-14473 Potsdam, Germany

Fluid inclusions yield important information for economic geology studies. The
fluid composition, pressure, and temperature prevailing at the time of ore
formation, especially in hydrothermal deposits, can be estimated, and the history
of fluid evolution and mineral formation can be reconstructed. Fluid inclusion data
can provide information not only for questions of ore deposit genesis and for ore
deposit modeling, but are also useful for prospecting and exploration, e. g., fluid
inclusion studies can assist in the determination and delineation of individual
mineralization and mineral provinces. Usually, fluid inclusions are studied in
transparent gangue minerals such as quartz, carbonates, or fluorite because
most ore minerals occurring in ore deposits are opaque for visible light.
Therefore, the assumption is made that the mineral associations (sulfides,
oxides, gangue minerals) within an occurrence deposited co-genetically when
using fluid inclusion data from gangue minerals to infer the P-T-X conditions of
ore deposition. On the other hand, it was found that transparent ore minerals
such as bright sphalerite locally contain fluids of differing composition than the
associated gangue minerals. This must be taken into consideration for
interpretation of the composition of fluids in transparent gangue minerals.

During the past 15 years it was demonstrated that some opaque ore minerals
such as wolframite, stibnite, pyrite, enargite or hematite which is normally opaque
for visible light show transparency near infrared light radiation and fluid inclusions
therein can be studied by IR microthermometry (e.g. Campbell and Panter; 1990;
Mancano and Campbell, 1995; Lüders, 1996; Lüders and Ziemann, 1999; Lüders
et al., 1999; Kouzmanov et al., 2002). In some cases the results of fluid inclusion
studies in quartz and ore minerals disprove the inferences from petrographic
studies that quartz and ore minerals deposited cogenetically.

This study presents fluid inclusion data of specular hematite from veins from four
iron ore deposits in the Quadrilátero Ferrífero: the barren veins of Córrego do
Feijão and Fábrica, which are located in the low-strain zone, and the Au-Pd–Pt-
bearing veins of Gongo Soco and Itabira. The Quadrilátero Ferrífero is a world-
class mineral province (Fe, Au, Mn) comprising a Palaeoproterozoic clastic-
sedimentary sequence (Minas Supergroup), an Archaean greenstone belt
sequence (Rio das Velhas Supergroup) and granite-gneiss domes. The Minas
Supergroup hosts a thick itabirite unit (banded iron formation) with high-grade
hematite ores. Presumedly, much of the primary (lode) gold from Minas Gerais
came from itabirite-hosted, sulphide-free, specular hematite-rich, palladian gold-
bearing ( jacutinga ) veins. Examples of this vein-style mineralization are found in
the Gongo Soco and Itabira iron ore deposits, in which gold is mined as by-
product. The jacutinga -style gold deposits are situated in the high-strain domain
of the Quadrilátero Ferrífero, where the host rock (itabirite) shows a pervasive
foliation commonly attributed to the 0.8–06 Ga Brasiliano Orogeny. The
metamorphic grade increases eastwards from lower greenschist facies (Gongo
Soco) to lower amphibolite facies (Itabira). In contrast, quartz-specularite veins
from the western and south-western low-strain zone of the Quadrilátero Ferrífero
are barren.Quartz–iron oxide isotope thermometry in the foliated itabirite and iron
ore (Hoefs et al., 1982) indicated higher temperatures than those suggested by
the mineral assemblages. Both in Itabira and Gongo Soco, the jacutinga veins
cross-cut the foliated itabirite. However, Olivo et al. (1995) considered that the
Itabira jacutinga veins were synchronous with the peak metamorphic
temperatures, with quartz–iron oxide model temperatures as high as 660°C.

Fluid inclusions in quartz from quartz-hematite veins cutting Palaeoproterozoic

itabirite in the Ouro Fino syncline area near Capanema, about 25 km south from
Gongo Soco, have been studied in detail by Boiron et al. (1999) . These fluids
have high sulphate, as well as high B, Li and F contents. They are oxidizing and
characterized by low Na/K and Cl/SO 4 ratios. Only a few total homogenization
temperatures were obtained in the range of 104 to 160°C for aqueous fluid
inclusions, and between 250 and 280°C for aqueous-carbonic ones. Br/Cl ratios
indicate an evolved seawater origin from which halite had precipitated (Boiron et
al., 1999).

F luid inclusions hosted in hematite from barren veins as well as from the Au-Pd-
Pt-bearing veins of Gongo Soco are always two-phase aqueous and often
contain multiple daughter minerals of unknown composition. These inclusions
occur isolated within the studied samples and, therefore, they are assumed to be
of primary origin. They show either rounded or irregular forms and size up to 100
(!) microns. Melting of ice was observed in this type of inclusions in the
temperature range between -9.8 and -4.5°C. First melting is approximately
between -30 and -27°C. The homogenization temperatures (Th) range between
150 and 228°C. The lowest Th values were measured in samples from Gongo
Soco, whereas the highest Th values were measured in samples from a barren
vein at Fábrica. Melting of daughter minerals hosted in type I inclusions were not
observed even upon heating to temperatures as high as 450°C. Samples from
Itabira host two different types of primary inclusions. Besides aqueous inclusions
with at least one daughter (?) crystal, aqueous carbonic inclusions are also very
frequent in the studied samples. Carbonic inclusions often show hexagonal or
rounded shapes and sizes up to 35 microns. Fluid inclusions hosted in hematite
samples from Itabira have higher Th values and lower ice melting temperatures
compared to samples from the other studied sites. Carbonic inclusions show
melting of solid CO 2 between -62.9 and -56.7°C and CO 2 homogenization
temperatures to the liquid phase between 17.9 and 22.4°C, and often decrepitate
prior to total homogenization. Total homogenization temperatures are as high as
450°C.

Infrared microthermometric studies of specular hematite-hosted fluid inclusions

from quartz-hematite veins in the Quadrilátero Ferrífero yielded distinct formation
conditions in different parts of the study area. In the north-eastern part (Itabira),
Au–Pd–Pt-bearing hematite-quartz veins, which crosscut the host itabirite,
display a complex fluid chemistry. The presence of CO 2 and high
homogenization temperatures let assume a retrograde metamorphic formation as
suggested for similar veins from Ouro Fino (Boiron et al., 1999). In contrast, Au–
Pd-bearing hematite-quartz veins at Gongo Soco seem to have deposited from
saline formation waters at considerable lower temperatures. The fluids trapped in
barren veins within the western and southern lower-strain zone are similar to
those at Gongo Soco, but obviously did not transport Au, Pd and Pt. The reason
for this is still unclear. It can only be speculated that in the high-strain domain,
fluids could have had access (through faults and reactivated structures) to source
rocks to leach them more effectively than in the low-strain zone.

References

Boiron, M-C, Moisette, A., Cathelineau, M., Banks, D., Monnin, C., Dubessy, J.
(1999): Detailled determination of paleofluid chemistry: an integrated study of
sulphate-volatile rich brines and aquo-carbonic fluids in quartz veins from Ouro
Fino (Brazil). Chemical Geology , 154, 179-192.

Campbell A.R. and Panter K.S. (1990): Comparison of fluid inclusions in

coexisting (cogenetic ?) wolframite, cassiterite and quartz from St. Michel´s
Mount and Cligga Head, Cornwall, England: Geochimica et Cosmochimica Acta ,
54: 673-681.

Hoefs, J., Müller, G., Schuster, A. K. (1982): Polymetamorphic relations in iron

ores from the Iron Quadrangle, Brazil: the correlation of oxygen isotope
variations with deformation history. Contributions to Mineralogy and Petrology ,
79, 241-251.

Olivo, G. R., Gauthier, M., Bardoux, M., Sá, E. L., Fonseca, J. T. F., Santana, F.
C. (1995): Palladium-bearing gold deposit hosted by Proterozoic Lake Superior-
type iron-formation at the Cauê iron mine, Itabira district, southern São Francisco
craton, Brazil: Geologic and structural controls. Economic Geology , 90, 118-134.

Kouzmanov K., Bailly L., Ramboz C., Rouer O. and Bény J.-M. (2002):
Morphology, origin and infrared microthermometry of fluid inclusions in pyrite
from the Radka epithermal copper deposit, Srednogorie zone, Bulgaria.
Mineralium Deposita 37: 599-613.

Lüders V. (1996): Contribution of infrared microscopy to fluid inclusion studies in

some opaque minerals (wolframite, stibnite, bournonite); metallogenic
implications. Economic Geology , 91: 1462-1468

Lüders V. and Ziemann M. (1999): Possibilities and limits of infrared light

microthermometry applied to studies of pyrite-hosted fluid inclusions. Chemical
Geology , 154: 169-178.

Lüders V., Gutzmer J. and Beukes N.J. (1999): Fluid inclusion studies in
cogenetic hematite, hausmannite, and gangue minerals from high-grade
manganese ores in the Kalahari Manganese Field, South Africa. Economic
Geology , 94: 589-596.

Mancano D.P. and Campbell A.R. (1995): Microthermometry of enargite-hosted

fluid inclusions from Lepanto, Phillipines, high-sulfidation Cu-Au deposit.
Geochimica et Cosmochimica Acta , 59: 3909-3916.

Spatial characteristics of distribution of gold mineralisation in Archaean

cratons.

Alexander Mikahailov 1,2 and Paul Dirks 2

1 SRK Consulting, Windsor Court, 1-3 Windsor Place, Cardiff, CF10 3BX, UK

2 School of Geosciences, Univ. of the Witwatersrand, PO Wit s2050,

Johannesburg, RSA

Knowledge of spatial parameters that influence the distribution of mineralisation

should be used as the basis for planning exploration activities, and for
understanding the genetic models that control mineralisation. The study
presented here concentrates on gold occurrences in Archaean cratons in Africa,
and is aimed at gaining a better understanding of the spatial parameters that
determine gold distribution patterns in order to develop a system that defines the
scale-dependent characteristics of mineralised objects. This will lead to an ability
to better plan exploration activity.

The spatial distribution of mineral occurrences can be described by a

hierarchically organised range of mineralised objects: e.g. payshoot (ore shoot)-
ore body – mineral deposit - ore field – mining camp - ore district – metallogenic
province – metallogenic tectonic belt. For particular commodities attempts have
been made to assess the specific dimensions of mineralised objects at different
hierarchy levels. These dimensions are important since they link in with the scale
of investigation at which mineral exploration takes place. For lode gold deposits
that are largely structurally controlled, an understanding of the relationship
between the hierarchy levels of mineralisation and scale-specific structural
features is vital.

In a number of research works the existence of large-scale structural corridors in

the mantle lithosphere, and their influence on distribution of mineralisation has
been demonstrated. But these works give just indications of spatial relations of
mineralisation patterns and regional “hidden” structures. For better understanding
of such type of relationships a first step could be a quantitative evaluation of
spatial distribution of gold mineralisation - scale dependant clustering and
relations of clusters at different hierarchy levels. To understand the dimension
and distribution pattern of mineralised objects at different hierarchy levels is
important, because these parameters determine the manner in which a mineral
exploration programme has to be conducted.

Most of our research is based on a comprehensive Gold Database developed for

the Zimbabwe Craton. The Zimbabwe Gold Database contains the accurate
location (+/- 50 m) of more than 6000 gold occurrences with more than 200
attributes fields for each occurrence describing locations, gold production and
grade; associated minerals and metals; type, shape parameters and orientation
of the ore bodies; associated intrusions and their orientations; associated
geological structures; alteration assemblages and other parameters. The
Zimbabwe gold database clearly demonstrates that gold occurrences are
clustered in different ways at different scale or hierarchy levels.
For quantitative evaluation of distribution of gold mineralisation a number of
spatial analysis techniques were used – block counting, nearest neighbourhood
analysis, density analysis, “line length” and buffer analysis. Spatial parameters
obtained from the Zimbabwe database were compared with gold occurrences
distribution from other similar datasets. Although the study presented here is
based on Archaean lode gold deposits in Africa, we believe that the
methodologies are relevant to controls on gold mineralisation in Archaean
terrains in other continents. Results of this research could be used for planning of
regional exploration programmes and target selections.

"Who does mining need and who needs mining?"

Mark Moody-Stuart

Anglo American plc, 20, Carlton House Terrace, London SW1Y 5AN
[email protected]

To a geologist or an economist the answer to the second part of the question in

the title would be “almost everyone”, for as was agreed by the United Nations
2002 World Summit on Sustainable Development (WSSD) in its Plan of Action “
Mining, minerals and metals are important to the economic and social
development of many countries. Minerals are essential for modern living.” The
WSSD rightly made its support conditional on the mining industry taking fully into
account the environmental and social impacts of its operations. Opposition to
mining comes not merely from those of the NIMBY and BANANA (Build
Absolutely Nothing Anywhere Near Anything) tendencies, but from a wide range
of people concerned about its impact on local communities, on indigenous
people, on biodiversity and so on. And even when all of these issues can be
addressed satisfactorily, many thoughtful people point to the fact that many
countries which have gained significant income from natural resources do not
seem to have been able to apply this income effectively in the interests of the
country at large and that significant economic and social distortions have arisen –
the so-called “resource curse”

A geologist commencing a career in mining Victorian times did so in a society

which believed that the development and utilisation of natural resources was an
essential element of progress. Even in the sixties, when suspicion and vilification
of transnational corporations was at a previous zenith and the Club of Rome had
just published its report “The Limits to Growth”, developing nations were founding
large state industries with the help of significant international debt and concern
was more about the potential future scarcity of resources than on the methods of
extraction and the utilisation of the revenue generated. Any young person
considering entering the mining industry today would be well advised to have
thought through some of these issues before being sandbagged with them by
fellow students in the pub or at meals or meetings. And it would be sensible for
universities to help students to think through these issues in a way that is
integrated with other teaching. Earth Sciences departments might well find that
programmes in this area had appeal way beyond the confines of the department
– and there could be merit in addressing the issue in part together with
colleagues in economics or social studies.

Suppose that a person has convinced themselves that mining done in a

responsible way and within sound overall governance frameworks can be a
contribution to sustainable development by converting natural resources into
other forms of capital – not simply through the generation of financial wealth but
through development of physical and social infrastructure, building human
capacity. We could then turn to the first half of the question of the title - what sort
of person is needed to develop such an industry?

The fundamental basis is a sound grasp of the science. If a solid grounding in

geology and mineralogy is not there, you are not going to get it in industry. The
emphasis and specialisation is less important than the sound base. A second
critical factor is to have a sound numerical basis, with an understanding of
statistics. It would also be difficult to call someone soundly based if they had no
grounding in basic geophysical techniques, and some understanding of the
chemistry (and biology) involved not just in the formation of mineral deposits and
their treatment and processing, but in their weathering and that of waste streams.
One would hope that a good degree or postgraduate work would lead through
natural curiosity to building out different areas of knowledge and demonstrate an
ability to apply science to the solution of problems, rather than simply
demonstrating ability to absorb facts. A project topic chosen and pursued with
passion and method, even if it did not deliver any notable results, is not such a
bad analogue for some future experiences in the mining industry.

Many of the mineral deposits of the world lie in developing countries and in
remote environments. It is unlikely that a geologist would feel at home in a mining
company if the limit of their experience was Cornwall or the Lake District. A
geologist will find themselves working with colleagues of all backgrounds and
nationalities as companies recruit and train people from countries with major
mining industries and from their main operating areas. Some experience of
another culture is practically a sine qua non – including an almost instinctive
acknowledgement that no one philosophy, culture or way of life provides all the
answers. If the mining industry is to address some of the challenges to its
legitimacy, we need people who can listen carefully to what at first hearing might
sound like ill informed or illogical positions or perceptions, but which if one is
open in one's listening and thinking one can recognise and address their point of
origin and also often acknowledge their considerable validity.

The reason that the detail of the degree is less important than the sound
grounding is because when someone comes into the mining industry, there will
almost certainly be a lot of specific learning that needs to be done to draw on the
corporate store of expertise in a particular mineral or metal. Much of the work is
done in multidisciplinary teams, or at very least working with one or two others
experienced in that particular area. Every member of a multidisciplinary team
brings their own skills – whether it be in geology, resource estimation, mining
engineering, process engineering, economics, or logistics. Whatever your team
membership is based on, you had better be good at it and at working in a team,
or else the whole team fails. That skill is your team membership ticket. But the
joy of working in such teams is the ability to acquire some of the expertise of
others, or at least an understanding of the basics so that you can communicate
and integrate more effectively. This is an essential part of ongoing further
personal development. There may also be opportunities to take courses later in
some, but this will be very much easier if you have familiarity from a
multidisciplinary team.

In my experience, when people join a large company or organisation, they often

have very little idea of where their development will take them. Very often those
with wider experience or different backgrounds may see alternative directions of
development or jobs of which you yourself are barely aware. If one is offered a
change of direction, and it offers new learning or experience, it is wise to seize it,
even if it may not be very convenient nor what you had thought of. Not only do
you stand a chance of learning something, but you are probably doing someone
a favour by filling a job which they think that you can do. And if you do it well,
both they and you benefit. And if you are not offered it, maybe you should go
looking for it. For conversely, whatever you are doing, in industry or academia, if
after two or three years you feel that you are no longer widening or deepening
your knowledge and experience – give it up and do something else.

Now return to the question of the title. If the mining industry is to thrive and have
legitimacy, from the very first activity in an area – exploration – we need to be
sure that we are engaging with the people of the area. They may be excited at
the opportunities, but they are likely to be worried about the implications. What
will a mine look like if it happens? Will there be work? What about schools and
training? What are the environmental risks? Will their agriculture or fishing be
disturbed? Will their way of life be disrupted by incomers? We need people in the
industry, who apart from doing their own job are interested in these questions
themselves, who can think of ways of addressing some of the fears and who are
interested in arguments for mutual benefit. Who have the ability to explain their
own science in straightforward ways, who listen openly and who are honest
about the implications. We also need people who can look at our own processes
and well established ways with new eyes, and make changes such that
operations have lower impact, less footprint, use waste as feedstock and so on.
As projects go forward we need people who can reach agreements on what the
best indicators are to reflect whether or not industry is delivering on its promises,
and who can publish these indicators in easily understandable forms. We also
need people who have a geologists ability to look at processes over time, to see
what may happen to communities, services and physical legacies decades down
the line when the mine may no longer be there. And who regard all of this, not as
a pain in the backside, but as a truly important and integral part of the mineral
resource industry.

This is a long but very worthwhile job - our children and grandchildren will still be
at it. We need people with both vision and persistence.

The Genesis of Magmatic Ni-Cu Sulfide Deposits: What we know and what
we don't know

Anthony J. Naldrett

Department of Geology, University of Toronto, Toronto, Canada M5S 3B1

(present address: 15, Apers Ave., Woking, Surrey, GU22 9NB;

[email protected])

Key requirements for

the formation of a magmatic sulfide deposit include (1) a magma containing an
adequate concentration of Ni (> 75 ppm Ni), (2) the development of droplets of
immiscible sulfide, and their mixing with a sufficient quantity of magma (a
magma/sulfide mass ratio > 300), and (3) a mechanism for concentrating the
sulfides is a restricted space. Excepting the deposits at Sudbury, the other major
deposits of the world provide consistent answers as to how these requirements
are met.

Major deposits are compared in terms of their total resources of Ni in Figure 1.

Noril'sk and Sudbury are pre-eminent. The Merensky Reef, Great Dyke and
Platreef are only economic because of their high content of PGE, dance to a
different drummer, and are not discussed further here. The deposits of the Duluth
complex are too low in grade to be economic under present circumstances,
although they constitute a major resource. The remaining major deposits/camps
(> 1 mt tonnes of contained Ni metal) comprise, in decreasing order of contained
Ni, Jinchuan (5.5 mt), Pechenga (4.0 mt), Thompson (3.5 mt), Mt Keith (2.9 mt),
Voisey's Bay (2.2 mt) and Kambalda (1.9 mt). With regard to requirement (1)
MgNos. and Ni contents of typical magmas responsible for some major deposits
are given in Table 1. In some cases the compositional data are based on
chilled/spinifex-textured samples, in others they have been calculated from
olivine compositions corrected for the trapped liquid shift. Most magmas
(excepting that at Voisey's Bay) are relatively primitive, but all, except for those
responsible for some komatiite-related deposits, have undergone fractionation
since leaving their source. However, the relatively elevated Ni concentrations
indicate that the fractionation was limited in extent. All of the cited examples
relate to magmas that have been introduced in a rifting environment. The
conclusion is that any relatively primitive, Ni-rich magma has the potential to form
a magmatic sulfide deposit, provided that other requirements are satisfied.

Considering requirement (2), sulfide immiscibility, Wendlandt (1982) and

Mavrogenes and O'Neill (1999) have demonstrated that a mafic magma's ability
to dissolve sulfide increases with decrease in pressure ( i.e. as it rises). Thus
most magmas traversing a large thickness of depleted

Table 1 Estimates of magma compositions

Deposit/Camp Magma Type MgNo. Ni ppm Source of data

Kambalda (also Mt Peridotitic 0.84 1500 Lesher et al. (1999)

Keith) Komatiite

Thompson Komatiite 0.73* 500-800 Lesher (personal

+ communication,
2003)

Pechenga Ferropicrite 0.67 790 Melezhik et al. (1995)

Jinchuan Magnesian 0.61* 173* Chai and Naldrett

Basalt (1992)
Noril'sk (last magma Tholeiite 0.56 116 Naldrett (2003)
involved)

Voisey's Bay Troctolite 0.45* 160* Li and Naldrett (1999)

Duluth Complex Troctolite/ol. 0.45* 150* Miller, J.D. and

gabbro Ripley, E.M., (1996)

* Calculated from olivine – (D Ni ) ol-liq = 10, (Fe/Mg) ol = 0.3x(Fe/Mg) liq ; + (D

Ni ) ol-liq = 4

lithosphere will arrive in the crust unsaturated in sulfide. MORB is the most
common exception to this generalisation, but in this case the vertical distance
between site of separation from the source and point of emplacement is
generally small, so the pressure effect is minimised. In Table 2, the sulfur isotopic
composition of the ores and nature of the country rocks are summarised for
many of the deposits cited in Figure 1. Interaction of the host magma with
country rocks has been established

Table 2 d 34 S and type of country rock

Deposit/Camp d 34 S Contaminated Country rock Source of data

Magma

Kambalda Yes, but not Sulfidic interflow Lesher et al.

throughout sed (1999), Ripley and
Li (2003)

Thompson 0 to +9 Yes, but only S/Se. Pelitic, sulfidic Bleeker (1990),

No trace element schist Eckstrand (1989)
data as yet
Brugmann et al.
Pechenga -3 to +8 Yes, principally g Sulfidic siltstone, (2000)
Os phyllite
Smolkin (1997)

Jinchuan -2 to +5 Not established Quartzo- Chai and Naldrett

feldspathic gneiss (1992), Ripley et
al. (2003)

Noril'sk +6 to Yes Shales, Duzhikov et al.

+16 evaporites (1992), Grinenko
(1985)
 Amelin et al.
Voisey's Bay -4 to +2 Yes Pelitic/psammitic (2000)
sulfidic gneiss
Ripley et al. (2002)

Duluth -2 to Yes Sulfidic Fe Ripley (1999)

Complex +20 formation and
slate

on the basis of one or more of trace element and Sr, Nd, Pb, Os, O and C
isotopic data in all cases save for Jinchuan, and d 34 S and g Os values indicate
that country-rock sulfur is commonly involved. It is becoming increasingly evident
that crustal contamination, generally involving the addition of sulfur, is a
requirement for forming an Ni-Cu sulfide deposit. What is not established is
whether contamination alone, without addition of sulfur, is sufficient to cause
sulfide immiscibilty, nor is it clear how country rock sulfur is transferred into the
magma.

With regard to requirement (3), of the deposits cited in Tables 1 and 2, the
Kambalda deposits occur in lava flow channels; those at Thompson and
Pechenga occur in sills that are co-magmatic with and believed to be feeders for
overlying volcanism; those at Noril'sk are located in tubular conduits through
which several generations of magma have flowed to form the overlying flood
basalts; and those at Voisey's Bay occur both within a vertical dyke linking a
lower magma chamber, in which sulfide immiscibility developed, with an upper
chamber into which both the original magma and subsequent new magma was
flushed, and at the entrance of the dyke with the upper chamber. At Jinchuan,
economic concentrations of sulfide occur within the funnel-shaped entrants of
feeders to the overlying magma. It is only at Duluth, where there is no evidence
of focused magma flow, that the mineralization occurs as clouds of sulfide that
are too dispersed to be economic. At both Noril'sk and Voisey's Bay, the original
sulfides to form were too low in tenor to be economic, but the passage of fresh
magma along the ore-bearing conduits has upgraded the tenors to economic
levels. While the importance of hydrodynamic processes in concentrating sulfides
from magma carrying dispersed droplets, the physics of how and where this
occurs is far from understood.

A Review of Regional Geochemical Surveys in Northern Ireland with

Implications for the Future

Miller O'Prey

Geological Survey of Northern Ireland, 20 College Gardens, Belfast, BT9 6BS,

[email protected]
Three regional geochemical surveys have been completed in Northern Ireland,
which have concentrated on stream sediment sampling. The first survey was the
only one to cover the entire country and was conducted by the Applied
Geochemistry Research Group at Imperial College London between 1969-1973.
This was followed by geochemical surveying, over the Dalradian lithologies of
Northern Ireland, undertaken by staff from GSNI between 1973-1975. The most
recent survey was completed by staff from the British Geological Survey (BGS)
between 1994-1996 and covered the western half of Northern Ireland.

The N.E.R.C. funded Imperial College Survey was directed by Professor John
Webb and produced the Provisional Geochemical Atlas of Northern Ireland
published in 1973 as well as PhD theses by Butt and Young. In total 4,832
stream sediment samples were collected, representing a sample density of just
less than 1/ sq. mile. The sampling programme was completed in one summer
field season and sampled locations just upstream from road bridges. Typically
100g of sediment was collected from tributaries where the upstream catchment
area was less than two sq. miles. The samples were analysed for 20 elements
including base (but not precious) metals by spectrographic and chemical
methods. Analytical and sampling errors varied considerably although the ranges
of the sampling and spectrographic analytical errors were comparable.

A series of grey scale maps were produced for each element based on arbitrary
concentration intervals and also some percentile intervals. In addition a series of
subtractive colour-mixing images were developed to display co-variations for
three element suites. The data was smoothed using a moving average technique
to reduce noise resulting from sampling and analytical errors. The quality of the
analysis and map production was state of the art at its time with the colour-mixing
plots reflecting the geological map of Northern Ireland.

Examples of correlation between geochemical patterns and solid geology

identified included the Mourne granites and the Antrim basalts. Spurious
anomalies resulted from co precipitation with iron and manganese in streams
draining acid soils in highland areas, e.g. the Sperrin Mountains and also in part
at least related to contamination in the old lead mining districts of Co. Down and
Co. Armagh. Otherwise cultural contamination was minor.

The survey, and in particular the thesis by Butt, identified two areas showing
anomalies possibly indicating base metal mineralisation near Keady in Co.
Armagh (unrelated to historic mine workings) and near Newtownstewart in Co.
Tyrone. Further drainage and soil sampling was completed in these areas and
the results were discussed in relation to the solid and superficial geology and the
different dispersion mechanisms present. This produced a number of conclusions
to guide later surveys and highlighted the importance of understanding the effect
of fluvio-glacial cover and the different dispersion mechanisms.
The survey was a technical success especially with respect to the rapid coverage
and the scale of sampling. Also in terms of exploration activity in Northern
Ireland, it is interesting to note the significant increase in land holdings on the
year the PhD theses were completed (and in subsequent years). On reviewing
the dataset one of the most significant images is that for arsenic. This shows
anomalies in the areas surrounding the then undiscovered Curraghinalt and
Cavanacaw deposits as well as the Co. Armagh mineralisation currently being
explored.

The second survey was funded by the then Department for Economic
Development (DED) and completed by John Arthurs of GSNI who conducted
reconnaissance geological and multi-element drainage surveys over the areas of
outcropping Dalradian lithologies of the Sperrin Mountains and northeast Antrim.
This resulted in two Mineral Potential Assessment Reports published in 1976.
BGS sampling techniques were adopted with a sample density of 1 per 2sq. km
and collection of 1.5kg wet-screened stream sediment, heavy mineral
concentrate and a 30cc water sample. The sediment samples were analysed for
up to 16 elements, including base metals, uranium and silver but excluding gold
and the stream waters were analysed for uranium. The Sperrins area was the
subject of more rigorous study with the panned concentrates analysed for Cu, Zn
and Ba and a more detailed drainage survey also completed over 34 anomalous
areas, followed up by some soil sampling.

The results of the reconnaissance stream sediment sampling were plotted for
each element on maps of the two areas, which lead to the definition of a number
of locations assessed as having good potential for base metal and barium
mineralisation (the main purpose of the study). However, Arthurs also highlighted
the prospectivity of the area in relation to gold rich deposits - although he felt the
terrane was most prospective for VMS style deposits with some later
hydrothermal reworking. The close relationship between arsenic in stream
sediments and gold observed in panned concentrates was also noted. This
observation has been confirmed by recent metallogenic studies, which showed a
positive correlation between gold and arsenic in mineralised bedrock.

Technically the survey was a success and it formed the basis of exploration
programmes into the 1980s. Indeed it was at Arthurs' recommendation that the
then exploration manager for Ulster Base Metals licenced much of the area and
initiated a programme that resulted in the discovery of the Curraghinalt gold
deposit and a number of other prospects. A significant number of the anomalous
locations highlighted are now known to contain bedrock gold and/or base metal
mineralisation, most notably at Golan Burn and Glenlark.

The most recent survey was co-funded by the Department for Enterprise, Trade
and Investment (Northern Ireland), the Department for the Environment (Northern
Ireland) and the BGS. Using the BGS G-Base methodology 2908 sites were
visited (approx 1 sample per 2.15sq. km) and 3 filtered and 3 unfiltered stream
waters, 100g of stream sediment (screened to –150 µm fraction) and 50g of
heavy mineral concentrate were collected at each site. Baseline chemical data
were obtained for 33 elements in stream sediments and 42 parameters in stream
waters along with a locational and observational database containing information
such as heavy mineral occurrences and land use at sample sites.

The stream sediment and stream water results were plotted for each parameter
as full colour images using modern imaging software. Similar to Webb, this study
found a strong correlation between certain elements and the underlying geology
and again the affect of thick glacial cover was highlighted. Both the Curraghinalt
and Cavanacaw gold deposits were identified, confirming the sensitivity and
efficiency of the technique. In addition, a number of new areas were highlighted
as having potential to host mineral deposits, one of which was followed up in
1998 and 1999. The extreme west of Co. Tyrone and in particular the
Newtownstewart area was the subject of more detailed stream sediment
sampling and some soil sampling. The results were plotted in a GIS environment
and a number of targets for further work were identified.

The mineral exploration application of this database has been heavily utilised by
companies exploring in the region. In particular Tournigan Gold Corporation used
the dataset (mainly arsenic stream sediments and gold observations from
panned concentrates) to select areas for their regional exploration programme in
2002.

In summary, regional geochemical surveys can play a significant role in

improving the knowledge of the geology and mineral potential of an area, which
in turn can lead to increases in exploration activity and deposit discovery. Any
future surveys that can build on these datasets will be of great value to Northern
Ireland.

Trace metal and halogen content (Cl, Br, I) of individual fluid inclusions
using synchrotron X-ray microfluorescence

Pascal Philippot

Institut de Physique du Globe de Paris, Laboratoire de Géosciences Marines, 4

place Jussieu, case 89, 75252 Paris cedex 05, France, [email protected]

Fluid inclusions represent the only relic of past movements of fluids in the Earth's
crust, and a knowledge of the trace-element composition of individual fluid
inclusions is central to our understanding of the formation of economic and
energetic resources. Since 6 years, we have developed an experimental protocol
aimed at analyzing single fluid inclusions using the X-ray microfluorescence set
up installed on line ID22 of the European Synchrotron Research Facility (ESRF).
Calibration procedures accounting for absorption by the host mineral and the
geometry and volume of the inclusion fluids were developed. High-spatial
resolution fluorescence X-ray maps and micro-fluotomography reconstruction of
individual fluid inclusions were documented for several samples. Concentration
and spatial distribution of major (Cl, K, Ca, Mn, Fe) and trace elements (Ti, Cu,
Ni, Zn, V, As, Ba, Br, Rb, Sr, Au, Ce, Hf, Nb, Zr, Th, Pb, U) were determined in
individual fluid inclusions from a variety of case studies (Brusson and Pont de
Rastel gold deposits, Chivor emerald deposit, Streltsov uranium deposit, and
Dunbar oil field). Recently, halogen content (Cl, Br and I) were determined in
highly saline brines from the Dabie shan and western Alps ultra-high pressure
terranes. Interesting perspectives include in situ analysis of homogenized fluid
inclusions facilitating quantification of bulk inclusion fluids and most importantly
allowing determining the speciation of dissolved species at the time of trapping
using micro-X-ray Absorption Near-Edge Structure (micro-XANES) analysis.

The above mentioned studies show the great potential of synchrotron techniques
for characterizing quantitatively the composition and speciation of ore fluids
during trapping. A major drawback, however, concerns beamtime availability for
performing statistically relevant fluid inclusion studies (typically several tens to
100 measurements). Owing to the automated protocol progressively installed at
ESRF, the first geologically relevant study was recently achieved for a
reasonable amount of time (48 - 72 hours beamtime allocation depending of the
size of the inclusions, down to 8-10 µm, and heterogeneity of the host matrix).
Results were obtained on Archaean quartz segregations filling inter pillow basalts
from the North Pole hydrothermal system (Pilbara, western Australia).

The sources of gold in orogenic gold deposits: Insights from the Otago Schist,
New Zealand.

Iain K. Pitcairn 1 , Damon A.H. Teagle 1 , Dave Craw 2

1 Southampton Oceanography Centre, University of Southampton, SO14-3ZH (

[email protected] )

2 University of Otago, PO Box 56, Dunedin, New Zealand

 Orogenic gold deposits occur within
deformed metamorphic terranes, dating from the mid-Archean to the Tertiary.
They have contributed over 20% of the global gold production and are lode
sources for most of the world's placer deposits (Groves et al. 2003). Despite their
economic significance, the sources of metals, and the fluids responsible for their
deposition are not well understood.

The Otago Schist in the South Island of New Zealand is Mesozoic

metasedimentary belt, exhibiting multiple generations of quartz-carbonate veins,
some of which have been mineralised with gold. The schists show little
compositional variation, are comparatively young and are exposed from
unmetamorphosed protolith to amphibolite grade rocks. As such they are an ideal
laboratory for the elucidation of the sources of fluids and metals required for the
formation of orogenic gold deposits.

The Otago Schist is composed of

three meta-sedimentary terranes, the Torlesse, Caples and Aspiring, which
comprise metagreywackes and metapelites with minor metavolcanic rock
horizons. The metamorphic grade ranges from prehnite – pumpelyite facies to
upper greenschist (450 ° C, 8-10 kbar; Fig.1). Uplift along the Alpine Fault from
the Miocene to Recent has exposed the Alpine Schist, a deeper crustal section of
the Otago Schist with higher metamorphic grade from chlorite zone greenschist
facies (450 ° C, 6-8 kbar) to garnet-oligoclase zone amphibolite facies (Fig.1;
600-650 ° C, 10 kbar; Mortimer 2000). Macraes in eastern Otago is the largest
ore deposit in the region with reserves for 110t of Au (deRonde et al. 2000) and
enrichments in As, Au, Sb, W, Mo, and Bi. These elements are suggested as
being sourced internally, leached from large volumes of the host schist by
metamorphic fluids (Henley et al. 1976; McKeag et al. 1989), or brought in from
an external source such as mantle or granitic metal-rich fluids (deRonde et al.
2000).

A large set of samples representing the variation in metamorphic grade, lithology

and structural setting has been analysed for a full suite of elements including
those enriched at Macraes and other ore deposits. The rationale is to define
“regions” of rock be it a specific metamorphic grade, rock-type or structural
setting, that is depleted in these elements relative to other areas and therefore
could be a source for elements now enriched in mineralised areas. Elements with
low crustal abundance such as gold (0.5-1.5 ppb) require analytical methods with
very low detection limits if areas of depletion are to be clearly recognised. A new
method to analyse rocks for gold has been developed with a detection limit of 10
ppt, thus allowing observation of gold mobility in a metamorphic belt.

Elemental concentration within the schist is controlled by protolith composition,

deformation and fluid flow regime, and metamorphic temperature with associated
mineral assemblage. The brittle - ductile transition (BDT) within the crust has
been identified at 300-350 ° C with predominantly ductile behaviour occurring at
temperatures greater ~ 450 ° C within the Otago and Alpine Schist (Sibson 1977,
Holm et al. 1989). Figure 2 shows concentrations of arsenic and gold in
metamorphic rocks from the Otago and Alpine Schists. Both elements are highly
enriched in samples from Macraes gold deposit relative to unmineralised schist.
There is a clear pattern of depletion of arsenic relative to unmetamorphosed
protolith values (100 ° C), at metamorphic temperatures greater than 450 ° C.
Although less pronounced than that of arsenic, gold also shows depletions in
lower crustal rocks relative to protolith values (Fig 2). Depletions in Caples
terrane samples are more pronounced than in the Torlesse terrane suggesting
that gold may have been sourced from fertile lithologies within the schist pile.
Fluids such as those released by metamorphic dehydration reactions along grain
boundaries in ductile regimes have slower flow rates than in brittle regimes
allowing greater interaction between the fluid and rock and leaching of
incompatible elements by the hydrothermal fluid. Once uplifted into the BDT
region, brittle fracture allows mineralised hydrothermal fluids to

escape upwards where ore deposition may occur due to changes in temperature,
pressure, redox etc. An example of this type of deposition is seen at Macraes
were ore veins were deposited near to the BDT at about 300 - 350 ° C in both
ductile and brittle structures (Teagle et al. 1990, Craw et al. 1999).
 Simple mass balance calculations
suggest that 1km 3 of garnet-oligoclase grade schist with an average
composition of 0.5ppm As and 0.2ppb Au has had 10000 tonnes of As and 0.5
tonnes of Au leached from it by hydrothermal fluids. Gold is heterogeneously
distributed within rocks and this mass balance estimate would be significantly
greater if protoplacer concentrations are considered rather than just barren
protolith. There is a considerable volume of upper greenschist to amphibolite
grade rock underlying the Otago Schist and presumably other greenschist-hosted
orogenic gold terranes. The depletion of gold and arsenic reported here provides
evidence for a local source of metals hydrothermally leached from the host rock
by fluids most likely of metamorphic origin.

References

Craw D., Windle S. J., Angus P. V., 1999 Gold mineralisation without quartz
veins in a ductile-brittle shear zone, Macraes Mine, Otago Schist, New Zealand.
Min Deposita, 34, 382-394

deRonde C. E. J., Faure K., Bray C. J., Whitford D. J., 2000 Round Hill Shear
Zone-Hosted Gold Deposit, Macraes Flat, Otago, New Zealand: Evidence of a
magmatic Ore Fluid. Econ Geol, 95, 1025-1048
Groves, D.I., Goldfarb, R.J., Robert, F., Hart, C.J.R., 2003 Gold Deposits in
Metamorphic Belts: Overview of Current Understanding, Outstanding Problems,
Future Research, and Exploration Significance. Econ Geol, 98, 1-29

Henley, R.W., Norris, R.J., Paterson, C.J., 1976 Multistage ore genesis in the
New Zealand geosyncline: a history of post-metamorphic lode emplacement,
Min. Deposita, 11, 180-196

Holm D. K., Norris R. J., Craw D., 1989 Brittle and ductile deformation in a zone
of rapid uplift: Central Southern Alps, New Zealand. Tectonics, 8(2), 153-168

McKeag S. A., Craw D., Norris R. J., 1989 Origin and deposition of graphitic
schist-hosted metamorphogenic Au-W deposit, Macraes, East Otago, New
Zealand. Min Deposita, 24, 124-131

Mortimer N., 2000 Metamorphic discontinuities in orogenic belts: example of the

garnet-biotite-albite zone in The Otago Schist, New Zealand. Int J Earth Sci, 89,
295-306

Sibson R. H., 1977 Fault rocks and fault mechanisms. Geol Soc Lond J., 133,
191-213

Teagle D.A.H., Norris R.J., Craw D., 1990 Structural controls on gold-bearing
quartz mineralisation in a duplex thrust system, Hyde-Macraes shear zone,
Otago Schist,New Zealand. Econ Geol 85: 1711- 1719

Experimental Studies of Ore Depositing Systems-new data, new insights

and future research directions

T.M. Seward

Institut für Mineralogie und Petrographie, ETH Zürich, Switzerland

Hydrothermal fluids occur ubiquitously throughout the Earth's crust and upper
mantle and are involved in magmatic processes, metamorphism, tectonics and
fault movement and the formation of ore deposits. They are responsible for the
distribution of enormous amounts of material throughout the crust and facilitate
the transport of both heat and many chemical components to the Earth's surface
via hydrothermal fluids which discharge both subaerially and onto the seafloor.
These fluids are multicomponent electrolyte solutions in which the main salt
component is generally sodium chloride with lesser amounts of other alkali metal
and alkali earth metal salts together with volatile molecular species and weak
acid/base components. Trace components comprise much of the periodic table.
The chemistry of the transport and deposition of trace components such as
transition metals and lanthanides is still relatively poorly known. An
understanding of the formation of hydrothermal ore deposits is fundamentally
premised on a knowledge of the chemistry of metal complex equilibria, mineral
precipitation reactions, surface chemistry, and element partitioning in near-
magmatic fluid (phase separating systems) at extreme conditions of temperature
and pressure.

These processes will be discussed in relation to our current state-of-knowledge

in conjunction with newly available experimental data pertaining to liquid and
volatile phase complex equilibria, mineral precipitation reactions and surface
adsorption chemistry.

Regional fluid inclusion studies of the iron oxide-copper-gold (IOCG)

district of Norbotten, Sweden : Investigating the link between Fe oxide and
copper mineralisation.

M.P. Smith 1 , S. Gleeson 2 and C.D. Storey 1 .

1 AGRU, School of the Environment, University of Brighton , Brighton .

E-mail: [email protected], [email protected]

2 Department of Earth and Atmospheric Sciences, University of Alberta ,

Edmonton , Alberta , Canada . E-mail: [email protected]

Introduction.

Hydrothermal iron oxide-related deposits are currently an extremely attractive

target for mineral exploration, as a result of the attention paid to the class of Fe
oxide-copper-gold (IOCG) deposits proposed by Hitzman et al., (1992). In this
study we report the initial results of an investigation into the chemistry of fluids
associated with iron oxide and copper deposits in the Norbotten area, Sweden ,
funded by the EU-RDF Georange programme. The magnetite-hematite-apatite
deposits of the region are proposed to be end-members of this genetically linked
class of ore deposits, with the Cu-Au deposits of the area (including
Pahtohavaara and Aitik) representing the copper-rich portions of the system. In
order for the IOCG classification to be useful in general terms, the proposed link
between these systems must be tested, and the exact nature of the link
established. This will allow a fuller understanding of the genesis of the deposits
and the nature of fluids responsible for mineralisation and provide new guides for
exploration.
The samples investigated in this study come the from the Kirunavaara and
Luossavaara magnetite-apatite bodies, the magnetite-hematite-apatite bodies of
the Per Geiger Ores, and the near-by copper-gold deposit at Pahtohavaara. All
these deposits are hosted in Proterozoic rocks, the Fe-oxide ores in the
intermediate to acid volcanics of the Porphyry Group, and the Pahtohavaara
deposit in the underlying basic volcanics of the Greenstone group (Bergman et
al., 2001). A sample of hematite bearing quartz vein from within a sheared
portion of the Porphyry group was also analysed. Additional samples are also
being studied from the Gruvbergite Fe-oxide-copper prospect to the south near
Svappavaara, from the Malmberget magnetite-hematite-apatite body, and from
deposits associated with the Nautanen Deformation Zone (NDZ) at Aitik,
Nautanen, Ferrum and Liikavaara.

Figure 1: Preliminary microthermometric data on fluid inclusions from Fe oxide

and Copper deposits in the Norbotten district, Sweden .

Fluid inclusion microthermometry

All the samples studied contain hypersaline brine inclusions. Within the Fe-oxide
deposits vein quartz typically contains liquid water (Lw) plus vapour (V) plus
halite (Sh) inclusions, whilst quartz associated with Cu-deposits hosts inclusions
with a more complex daughter mineral assemblage, commonly including
carbonates. Within the Fe-oxide deposits homogenisation temperatures
commonly range from 90-123°C, whilst halite dissolution temperatures range
from 200-300°C, corresponding to a salinity range of 32 to 38wt. % NaCl eq.
Hematite bearing quartz veins from shear zones in the surrounding sediments In
the copper deposits of the Kiruna area T h typically ranges from 110-200°C, with
halite dissolution temperatures ranging from 300-450°C, corresponding to a
salinity range of 38 to 47°C. These are typically accompanied by vapour-rich
inclusions with melting temperatures in the range -58.7 to -59.6°C, and T h in the
range -2.2 to -14.9°C, corresponding to low density liquid CO 2 possibly with
minor methane. The inclusion assemblage from the Aitik deposit is considerably
more complex, with coexisting Lw+Sh+V, Lw+Lc+V and Lc+V inclusions. These
samples are currently in the process of being investigated but suggest
considerable differences to the greenstone-hosted deposits of the Kiruna area.

Discussion and Conclusions

The initial microthermometric analyses presented here clearly show the

association of Fe-oxide deposits with extremely saline brines, trapped at high
temperatures and pressures, with minimum pressures in the range 2-3kbar at
halite dissolution. The potential of such brines to carry extremely high
concentrations of iron in solution is undoubted. The data from the Fe-oxide
bodies fall in a range distinct from the copper mineralisation, but with major
overlaps between deposits suggesting a similar fluid was associated with Fe-
oxide mineralisation in each case. Previous workers have attributed the formation
of the Kirunavaara body to purely magmatic processes, to exhalitive deposition of
magnetite form hydrothermal solution, and to hydrothermal replacement
processes. Our data clearly argue against an exhalitive origin for the ore, given
the high pressures inferred for mineralisation. The quartz veins studied occur
cutting both the wall rocks and the Fe-oxide ore bodies themselves, and so must
be late-stage to post- the ore forming events. However, the assemblage of
minerals co-existing within the veins and the ore, plus widespread textural
evidence, suggest a link between the two and support a hydrothermal
replacement origin for the ore bodies.

Veins from copper deposits and in the surrounding stratigraphic sequence are
distinct from those within the Fe-oxide deposits. Fluid inclusions from quartz
veins within shear zones surrounding the Kiruna area ore bodies show similar
salinities to those in the ore bodies, but a range of homogenisation temperatures
extending to much higher temperatures. This can almost certainly be attributed to
active deformation syn- to post-quartz vein formation. All the copper deposit
hosted inclusions so far analysed show a much higher salinity than those
associated with the Fe ores, and so require either a different fluid, or a process of
modification to the Fe-oxide related fluids that may well be linked to copper
deposition. Such a mechanism is provided by the presence of CO 2 -rich fluid
inclusions in quartz, alongside hyper-saline brines in many of the samples so far
analysed. The addition to the fluid of CO 2 derived from interaction with
sediments within the stratigraphy, led to fluid immiscibility and consequent
increase in the salinity of the residual brine. This may have played a fundamental
role in the formation of significant Cu deposits in the region. Preliminary data
from the Aitik deposit are distinct from all other samples so far analysed, and
suggest different fluid processes in this deposit, and possibly others associated
with the NDZ.

The evolutionary link proposed between fluids associated with Fe-oxide

deposition and copper ore deposition remains to be tested, as does the
distinction between the Aitik deposit and other Cu prospects and mines in the
area. Data on the anion and cation chemistry of the fluids, and their chlorine
isotope composition are currently being acquired at the University of Alberta . A
parallel study is investigating the timing of the different ore forming events, to
determine if there is a temporal link between Fe-oxide and Cu mineralisation.
This will contribute new constraints to the debate over the origin of mineralising
fluids with IOCG systems, with particular relevance to exploration within
Norbotten and the rest of the Fennoscandian Shield.

References

Bergman, S., Kübler, L. and Martinsson, O. (2001) Description of regional

geological and geophysical maps of northern Norrbotten County (east of the
Caledonian Orogen). Sveriges Geologiska Undersökning Ba 56.

Hitzman, M.W., Oreskes, N., and Einaudi, M.T. (1992) Geological characteristics
and tectonic setting of Proterozoic iron oxide (Cu-U-Au-REE) deposits.
Precambrian Research , 58 , 241-287.

Metal Transport and Deposition in Hydrothermal Veins revealed by 213nm

UV Laser Ablation Microanalysis of single Fluid Inclusions

B. Stoffell 1 , J. J. Wilkinson 1 and T. E. Jeffries 2

1 Department of Earth Science and Engineering, Imperial College London ,

South Kensington Campus, Exhibition Road, London SW7 2AZ , UK . E-mail:
[email protected]

2 Department of Mineralogy, The Natural History Museum , Cromwell Road,

London SW7 5BD , UK .

213nm UV laser ablation provides a powerful and cost-effective new method for
the chemical microanalysis of individual fluid inclusions. Analyses of groups of
cogenetic primary and secondary inclusions yield typical relative standard
deviations for absolute element concentrations of around 20%, comparable with
those obtained by excimer laser ablation. Results are consistent with Ca
concentrations independently estimated from microthermometry, previously
published bulk chemical analyses, and give good charge balance.

Pb(-Zn) mineralized ‘crosscourse' veins from Southwest England are known to

have formed from low temperature brines, expelled from Permo-Triassic basins
adjacent to the Cornubian peninsula, probably during the Triassic. Analyses of
primary fluid inclusions across growth-zoned quartz crystals from one of these
veins illustrate temporal changes in orefluid chemistry. Relatively short timescale
decreases in the concentrations of Cl, K, Ca, Ba, Sr, Pb and Zn of up to two
orders of magnitude are superimposed on gradual systematic increases in K, Ca,
Sr, Ba and Zn. The relationships between the major cations and chloride deviate
from trends typically observed in basinal brines. This suggests that mineral
dissolution and precipitation reactions during flow out of the basin are an
important control of compositional variation on the timescale of vein growth,
rather than evolution of the fluid reservoir.

Large, relatively short time-scale drops in element concentrations are attributable

to mineral precipitation events and indicate the operation of processes that
rapidly drive the brines towards supersaturation. Initial galena precipitation was
due to reduction as oxidized fluids encountered the first strongly reducing
lithology on their flow path out of the basin source region. Later co-precipitation
of sulfides (galena, sphalerite) with sulfates (anhydrite, barite) was driven by
intermittent mixing with dilute, near neutral, reducing fluids. The common lack of
direct spatial association between fluid chemical excursions and mineral
precipitates suggests fluid inclusions may commonly trap depleted fluids after
mineral precipitation has already occurred some distance upstream, rather than
reflecting precipitation events in their immediate vicinity.

The primary ore-forming fluid contains elevated Pb and Zn, and displays a high
Pb/Zn ratio. The Pb-rich tenor of the vein deposits is therefore primarily
controlled by the metal content of the basin porefluids rather than aquifer
geochemistry or selective precipitation mechanisms and is ultimately a function
of the geochemistry of the basin source.

In-Situ LA-ICP-MS dating of titanite: a new tool for mineral deposit studies
and application to the Kiruna Iron Oxide deposit

Craig Storey 1,2,* , Martin Smith 1-2 & Teresa Jeffries 2

1 School of the Environment, University of Brighton , Lewes Road , Brighton ,

BN2 4GJ

[email protected]

2 Mineralogy Dept., The Natural History Museum , Cromwell Road , London ,

SW7 5BD

U-Pb geochronology has been the mainstay for dating key geologic events
throughout much of the history of the Earth System and much energy has been
expended into improving accuracy, precision and spatial resolution of such
analyses. At present it is possible to date areas as small as 20-30µm within U-
rich phases such as zircon and monazite by ion-probes such as the SHRIMP and
Cameca 1270. However, these methods have difficulty in dealing with matrices
containing multi-cation species due to unpredictable sputtering characteristics
and, therefore, complex or inconsistent calibration. Effectively, this rules out
efficient and rapid U-Pb measurements on many useful phases such as titanite
and allanite. Recent improvements in laser ablation (LA) of geologic materials
means that it is now possible to ablate small volumes of material with a small
particle size in a controlled manner, using UV lasers such as frequency
quintupled Nd:YAG lasers operating at 213nm or Excimer lasers operating at
193nm in a mixed Ar:He carrier gas. Coupled to an ICP-MS, these laser
innovations have been shown to drastically reduce elemental fractionation in the
plasma and allow improved reproducibility in isotope measurement. Moreover, it
has been demonstrated that standard and unknown zircons can be dated by U-
Pb isotope ratio measurement with a spatial resolution similar to ion probe
techniques and an accuracy and precision at least as good as these techniques.
Whilst zircon is undoubtedly a useful mineral for accurate and precise U-Pb
geochronology, its applicability is limited by a lack of understanding of its reaction
history and its behaviour under different P-T-x conditions. Furthermore, for the
sake of the present audience, it is extremely difficult to unequivocally
demonstrate zircon growth or recrystallisation under hydrothermal conditions and
be able to link such phenomena to discrete events in the P-T-t-x history.

In this study we turn our attention to titanite, a mineral which is abundant in

metamorphic, igneous and hydrothermal environments, can be readily related to
the petrogenesis of these systems, readily accepts U and Th into its lattice and
has a high closure temperature for Pb. Titanite has been dated by ID-TIMS
methods for many years, but generally quite a large amount of material is
required for dissolution in order to separate enough U-Pb for measurement,
effectively ruling out routine measurements on single crystals or parts of single
crystals by this method. In this study we present data from our ongoing efforts to
date titanite in-situ by LA-ICP-MS and show that reproducible U-Pb
measurements can be made on titanites of various ages and compositions, an
important first step. We also demonstrate that the zircon geostandard 91500 can
be used to calibrate U-Pb measurements in titanite by this method, negating the
necessity of finding a suitable titanite standard. Nevertheless, we also present
data on a potential U-Pb titanite standard that we have begun to characterise.

The LA-ICP-MS data were performed at the Natural History Museum, London ,
on polished grains mounted in a resin block and on polished thick sections, which
were imaged and characterised in an SEM prior to analysis. The laser used was
a New-Wave UP213 (213nm), coupled to a PlasmaQuad 3 ICP-MS with an S-
option; we used a mixed Ar:He carrier gas for all analyses.

To demonstrate the accuracy and precision of this method, we present U-Pb

titanite data from a Scottish pegmatite, for which we have reliable ID-TIMS data,
and show that we can calculate a correct age. Furthermore, we use this sample
to demonstrate how we can unravel U-Pb systematics on the inter-grain scale
using this method in order to ascertain how and if a common Pb correction needs
to be applied. In the case of this particular sample, we demonstrate that
individual grains contain hetereogeneous abundance of common Pb, but of a
homogeneous common Pb composition. Meanwhile, the grains have a
homogeneous radiogenic Pb composition meaning that they represent a single
age population, which can be visualised on Conventional and Tera-Wasserburg
U-Pb concordia plots (Figs 1 & 2). The age in this case is calculated by a
regression to concordia which effectively is exactly the same as applying a
common Pb correction based on a known composition (i.e. from another co-
genetic phase) or a model composition (e.g. Stacey and Kramers, 1975). The
efficacy of this method is borne out as the calculated LA-ICPMS regressed age
(432 + 13 Ma) is identical to that calculated by TIMS (437 + 6 Ma) and,
furthermore, the uncorrected (for common Pb) TIMS analyses plot along the
same regression line.

This method has been developed in order to attempt to date magmatic, tectono-
metamorphic, hydrothermal and mineralization events in the Norrbotten Ore
Province of North Sweden and is funded by Georange, an EU-RDF grant.
Titanite is a common phase in many of the exposed and drilled rocks in this
region and its petrographic context is often unequivocal. Moreover, individual
grains are often large enough to see in hand specimen, and certainly in thin
section under low power optics. We have made a series of polished thick
sections (c.150µm) from samples containing coarse titanite, the obvious
advantage being that we can ensure that the petrographic context remains intact
following in-situ dating of the grains. Here, we report the first attempt at such
dating from amygdaloidal syenites in the footwall of the Luossovaara Fe-Oxide
deposit. The amygdules are filled by titanite + magnetite and so we can be sure
that the ages will reflect a minimum age of extrusion and, most likely, the age of
Fe-Oxide mineralization. A similar sample has been dated by Romer et al. (1994)
by ID-TIMS at 1876 + 9 Ma and the nearby Kirunavaara Fe-Oxide deposit (the
largest underground mine in the world) is dated at 1880-1900 Ma by U-Pb zircon
ID-TIMS (Cliff et al., 1990). BSE images demonstrate a dark older component
within coarse individual grains, which has been altered by a patchy BSE-bright
component, which is best explained as hydrothermal alteration. The older
component is relatively extremely enriched in U compared to the replacive BSE-
bright component, representing a clear chemical distinction between the two
areas. We present data from intra-grain analyses within these two components
which clearly demonstrate a hiatus of >150 Ma, with the older, volumetrically
inferior component yielding a U-Pb concordant age of c.2100 Ma and the
younger component at c. 1930 Ma. These data question the present knowledge
of the age of magmatism in this region since the syenites are correlated with the
Porphyry Group of supracrustals, which are currently believed to be 1.91-1.88 Ga
old (Bergman et al. 2001). Based on age alone these syenites could be
genetically related to the stratigraphically underlying 2.2-1.95 Ga Greenstone
Group, otherwise the stratigraphy in the Norrbotten region requires revision. We
intend to test this hypothesis further by in-situ dating of titanite from other key
samples in the region.

References
Bergman, S., Kübler, L. and Martinsson, O. 2001. Description of regional
geological and geophysical maps of northern Norrbotten county (east of the
Caledonian orogen). Sveriges geologiska undersökning Ba 56 .

Cliff, R.A., Rickard, D. and Blake, K. 1990. Isotope systematics of the Kiruna
magnetite ores, Sweden : Part 1. Age of the ore. Economic Geology , 85 : 1770-
1776.

Romer, R.L., Martinsson, O. and Perdahl, J.-A. 1994. Geochronology of the

Kiruna iron ores and hydrothermal alterations. Economic Geology , 89 : 1249-
1261.

Stacey, J.S. and Kramers, J.D. 1975. Approximation of terrestrial lead isotope
evolution by a two-stage model. Earth and Planetary Science Letters , 26 : 207-
221.

Direct dating of Precious Metals Using 131 Te Decay.

H. V. Thomas, Dr. J. Gilmour.

Earth Sciences Dept, University of Manchester , Oxford Rd , Manchester , M13

9PL . [email protected]., [email protected].
This project involves the development of a new geochronometer using double
beta decay ( bb -decay) of tellurium to xenon, giving the potential to date the ore
forming minerals directly rather than dating the wallrock or alteration minerals.

Telluride minerals are very widespread and occur (in low concentrations) in
nearly all types of ore deposit. They are especially prevalent in hydrothermal ore
deposits where silver or gold-bearing tellurides host the mineralization. They
occur all across the globe and can be found throughout geological time.

Excesses of 130 Xe ( 130 Xe * ) accumulate in tellurium rich minerals from the

double beta decay of 130 Te 1 . Therefore, an age can be measured as an
excess of 131 Xe ( 131 Xe * ) produced by neutron capture on 130 Te in a
reactor. Thus a single xenon isotope analysis (i.e. of 130 Xe * / 131 Xe * ) can
determine the ratio of daughter to parent in a sample. This technique is
analogous to the Ar-Ar refinement of K-Ar dating where the Te-Xe age
represents the time of last closure to xenon loss 2 . Step-heating experiments
have yielded well defined peak releases between 400-600 O C, suggesting that
the ages reflect the end of the last heating event that produced temperatures in
excess of these values.

The telluride minerals are analyzed in Manchester in a unique, state of the art,
ultra sensitive resonance ionization time of flight mass spectrometer: RELAX
(Refrigerator Enhanced Laser Analyzer for Xenon) developed by Dr J. Gilmour
3.

This project aims to refine the technique and hopefully constrain the half life of
130 Te decay to 130 Xe. This will be achieved by analyzing many more samples
form throughout the globe and geological time. Other aspects to consider may be
the different retention of Xe in different telluride phases, the probability of
resetting and the possibility that differing telluride ages may be an effect of the
differing modes of deposition in ore deposits.

References:

1 Kirsten et al. (1968); Phys. Rev. Lett. 20, 1300-1303.

2 Gilmour et al (1999); In: Proceedings 5 Bicentennial SGA Meeting and 10

Quadrennial IAGOD meeting, London . Balkmena , Netherlands , 1259-1261.

3 Gilmour et al. (1994); Rev Sci. Inst. 65, 617-625.

Understanding the 3-Dimensional Fault Geometry of a Vein-Hosted Fluorite

Deposit: The Basis for Realistic Numerical Fluid Flow Simulations
 R.H. Tomlinson and S.K. Matthai

Department of Earth Science & Engineering, Royal School of Mines, Imperial

College , London . ( [email protected] )

Hydrothermal fluid flow dynamics within faults and associated ore deposition is a
complex and coupled process incorporating the advection/convection of a
(saline) fluid, heat transfer and fluid-rock interactions within a 3-dimensional
heterogeneous geologic medium. Consequently, numerical simulations can help
us to qualitatively and quantitatively understand theses processes. To date, most
simulations tend to be of generic and over-simplistic ‘base-cases' as opposed to
deposit-specific, making it difficult to extrapolate results from the simulations to
the deposit. In order to simulate a particular mineralising system, a good
understanding of its structural and mineralogical evolution is required so that
realistic boundary conditions can be ascribed and so that the model is a
physically accurate representation of the natural system.

Fluid flow simulations are to be based on the operational fault-hosted Silius

fluorite (Pb-Zn-Ba) deposit in south-eastern Sardinia . They will incorporate a
highly realistic 3-dimensional geometrical representation of the fault which hosts
the mineralisation, as determined from the digitisation of the mine level plans.
From underground structural observations, the fault is interpreted to be a dextral
strike slip fault, with this interpretation being confirmed by the larger-scale
features of the 3-dimensional fault geometry. Faulting and mineralisation, which
cross-cut bi-modal Permian dykes are interpreted to be contemporaneous and
cyclical. Two main veins are found in the deposit distinguished by their differing
parageneses. The (older) San Giorgio vein contains fluorite, quartz, chalcedony,
(minor) calcite, sphalerite and marcasite whilst the (younger) San Guiseppe vein
contains fluorite, (abundant) calcite and galena with some barite (mostly found in
the higher levels of the mine).

The mineralisation is zoned and so the spatial position of the different zones will
be representative of a general flow regime within that part of the fault. For
example, provisional geochemical modelling of the system shows that barite
precipitates when the brine is diluted with meteoric water. As barite is located at
the top of the deposit, this may indicate that meteoric water is entering the
system from above and that the position of the barite represents a certain fluid
salinity (range). Similarly, the location of quartz/chalcedony is interpreted to
represent a region of fluid up-flow.

The realistic fault geometry will therefore form the basis for 3-dimensional
numerical fluid flow simulations. It is hoped that as well as accurately modelling
the Sillius mineralising system, further insights into the dynamics of hydrothermal
systems will be gained such as system lifespan and transiency.
 A Nd-Sr isotope study of the Navan base metal deposit: support for the
hydrothermal convection cell model

R.D. Walshaw 1 and J.F. Menuge 2

1 BNFL Particle Alliance , Institute of Particle Science & Engineering, University

of Leeds , LS2 9JT (e-mail: [email protected])

2 Department of Geology, University College Dublin, Belfield, Dublin 4, Ireland

(e-mail: [email protected]).

The giant Navan Zn-Pb deposit of County Meath , Ireland , is hosted in

Courceyan limestones. It comprises a series of stacked stratabound lenses
divided into three major fault bounded zones (Ashton et al. 1986). The ore is
generally thickest adjacent to faults and dolomitised/argillaceous horizons, which
appear to have acted as conduits and barriers, respectively. It appears to have
formed primarily by replacement of host limestones, though with some open
space and late fracture-filling mineralization. Ore minerals comprise massive
sphalerite and galena, accompanied by iron sulphides, calcite, dolomite and
barite. It is inferred to lie just north of the Caledonian Iapetus Suture (e.g. Phillips
2001).

The timing of mineralization, and its relationship to normal faulting with which it is
spatially associated, continue to be the subject of debate. Based on the
relationship between mineralization and faulting, Blakeman et al. (2002) argued
that mineralization began early in the rifting history of the area, i.e. no later than
Chadian times. On the other hand, Peace and Wallace (2000) concluded from a
diagenetic study that all ore is epigenetic and post-Arundian in age. Symons et
al. (2002) have concluded on the basis of a palaeomagnetic study that the ores
were formed during the late Arundian to early Asbian. Whatever the exact age,
the main mineralization almost certainly occurred during the Lower
Carboniferous.

Nd and Sr isotope analysis of hydrothermal gangue minerals spanning the

paragenetic sequence in the main Navan ore body reveals a systematic evolution
in the mean composition of the mineralizing fluid. Early fluid represented by
replacive dolomite exhibits the lowest initial 87 Sr/ 86 Sr ratio (0.7083 – 0.7086),
closest to that of the host limestone and to Lower Carboniferous seawater, and
the highest 143 Nd/ 144 Nd ratio (0.51161 – 0.51176). Later generations of
dolomite, barite and calcite, which encompass sulphide precipitation, tend to
have higher initial 87 Sr/ 86 Sr ratios (maximum 0.7105), often exceeding that of
any likely seawater input, and 143 Nd/ 144 Nd ratios ranging down to 0.51157.
All initial Nd isotope ratios are too low to have been acquired solely from the host
limestone.
The data are interpreted to indicate mixing of early limestone-buffered brine with
a hydrothermal fluid which had passed through sub-Carboniferous rocks. The
143 Nd/ 144 Nd ratio of this basement-derived fluid is too low to have been
imparted by flow through the Devonian Old Red Sandstone of the Munster Basin
of southern Ireland, as required in models of regional fluid flow in response to
Hercynian uplift (Hitzman 1995). Irrespective of whether or not such regional fluid
flow occurred, the hydrothermal Nd must, by a process of elimination, have been
derived from sub-Devonian rocks. A similar conclusion was reached on the basis
of Pb isotope data, e.g. O'Keeffe (1986).

In broad terms, these conclusions support the local hydrothermal convection cell
model of Russell (1986). In this model, brines ultimately of surface origin
penetrated to a depth of several kilometers, leaching metals from the rocks
through which they passed. Metals were subsequently precipitated in carbonate
rocks at sites of mixing with cooler, sulphur-rich fluids. The corollary is that the
initial Nd isotope ratios of the hydrothermal minerals so formed constrain the
crustal residence age of the leached rocks.

New data from presumed Ordovician volcanic, volcanosedimentary and

sedimentary rocks from core beneath the Navan ore body have 143 Nd/ 144 Nd
ratios at the time of Navan mineralization of 0.51184 – 0.51217. These ratios,
and ratios of Lower Palaeozoic volcanosedimentary rocks of England , are all
significantly higher than the lowest values seen in Navan hydrothermal gangue
(0.51157), strongly suggesting that Lower Palaeozoic rocks cannot alone
account for the lowest initial Nd isotope ratios in the Navan gangue. Mills et al.
(1987) drew similar conclusions based on the Pb isotope compositions of Navan
galena. The only known rocks occurring northwest of the Iapetus Suture in
Ireland which might supply this very unradiogenic Nd are the Palaeoproterozoic
Mullet Gneisses of the Annagh Gneiss Complex of north Mayo (Menuge and
Daly 1990) and the metasedimentary rocks of the Neoproterozoic Dalradian
Supergroup (Daly and Menuge 1989). As the Navan deposit lies immediately
north of the Iapetus Suture, the implication is that the Laurentian margin includes
Precambrian basement in addition to the volcanosedimentary rocks exposed in
the Longford-Down massif.

References

Ashton J.H. et al. (1986) In: Andrew C.J. et al. (eds) Geology and genesis of
mineral deposits in Ireland . IAEG Spec. Publ., 243-280.

Blakeman R.J. et al. (2002) Econ. Geol., 97, 73-91. Daly J.S. and Menuge J.F.
(1989) Terra Abstracts, 1, 12.

Hitzman M.W. (1995) In: Anderson K. et al. (eds), Irish Carbonate-Hosted Pb-Zn
Deposits, SEG Guidebook 21, 3-23.
Menuge J.F. and Daly J.S. (1990) In: Gower C.F. et al. (eds) Geol. Assoc.
Canada Spec. Paper, 38, 41-52.

Mills H. et al. (1987) Nature, 327, 223-226.

O'Keeffe W. G. (1986) In: Andrew C. J. et al. (eds) Geology and genesis of

mineral deposits in Ireland . IAEG Spec. Publ., 617-624.

Peace W. M. and Wallace M. W. (2000) Geology, 2, 711-714.

Phillips W.E.A. (2001) In: Holland C.H. (ed.) The Geology of Ireland, 2 nd edition,
Academic Press, 192-196.

Russell M.J. (1986) In: Andrew C. J. et al. (eds) Geology and genesis of mineral
deposits in Ireland . IAEG Spec. Publ., 545-554.

Symons D.T.A. et al. (2002) Econ. Geol., 97, 997-1012.

Infiltration of a high-latitude magmatic arc by meteoric waters: hydrogen

and oxygen isotopic compositions of “unaltered” granitoids in the
Antarctic Peninsula

Willan, Robert C.R. 1 , Boyce, Adrian J. 2 & Fallick, Tony E. 2

1. Department of Geology and Petroleum Geology, University of Aberdeen ,

Scotland , AB34 3UE

2. Isotope Geosciences Unit, S.U.E.R.C., East Kilbride , Glasgow , Scotland ,

G75 0QF

Mineral deposits contain a record of interactions between the lithosphere and the
Earth's hydrosphere-biosphere-atmosphere. Over 25 years ago, mineral deposit
geologists recognised large-scale “flooding” of intrusive systems by meteoric
waters. Surface waters are now implicated in most crustal processes
(diagenesis, sedimentary basin development, deformation, regional
metamorphism, alteration of basement, magmagenesis and uplift). Huge
meteoric-hydrothermal systems (15000 km 2 ) have now been defined around
continental batholiths and it is postulated that the amount of fluid (H 2 O, CO 2 )
pumping in and out of the shallow crust (< 10-15 km) may be more than that
contained in the hydrosphere and atmosphere combined [5, 8]. Such fluid fluxes
have major implications for heat transfer and geochemical cycling, mineral
deposit formation, and past atmospheric compositions. Fossil hydrothermal
systems may contain a record of palaeoclimate [1, 2, 6, 7]. However, most
climate models assume an inert, dry, crust (“land”) and ignore volatile fluxes into
and out of the deep Earth. Given the close coupling of the solid Earth,
hydrosphere, biosphere and atmosphere in the Earth System, and the ancient
record preserved in hydrothermal rocks, the decline in funding and almost
complete lack of geoscience in recent Research Council planning are peverse,
especially given their new buzzphrase ‘Earth System Science'.

Much of the evidence for surface waters in the deep crust comes from the
inhabited tropical and temperate latitudes. Here, ‘normal' crustal rocks such as
granitoid plutons have relatively narrow stable isotopic compositions, ( 18 O
between +6 to +10 /+15 ‰, and D -40 to -80 ‰). Variations within these ranges
are primarily related to source (mantle versus crust), sedimentary contamination,
and volatile loss. High water:rock systems are relatively easily recognised, with
mineral compositions as low as -5 and -170 ‰ respectively. Recognition of low
water:rock systems is problematic because of domination of rock values and the
small contrast in D between local waters and average crust at low latitudes. To
complicate matters further, recent work on some porphyry copper deposits has
swung the meteoric-water pendulum the other way - alteration previously
ascribed to flooding by surface waters has been reassigned to primary magmatic
volatiles [4]. Another controversy is the origins of 18 O - and D-depleted primary
magmas. Are they derived from an unusual mantle source or by large-scale
assimilation of crust altered in pre-existing hydrothermal systems?

High latitudes have the greatest isotopic contrasts between average crust and
the hydrosphere and could best address the above problems, but have been
relatively unstudied. The Antarctic Peninsula contains a continental-margin
“Andean” magmatic arc formed between 230 and 10 Ma between 55° and 75°S.
The oldest plutons have an ‘S'-like geochemical character, related to
contributions of melt from Gondwana continental crust, whereas the youngest
were derived from more primitive mantle-like sources, related to magmatism in
thinner continental crust and accreted fore-arc. Although the arc formed at a high
latitude, the climate approached sub-tropical in the mid Cretaceous, and
Palaeocene–early Eocene, with cool, possibly glacial, intervals in the middle and
late Cretaceous. After continental breakup at 30 Ma, the climate deteriorated to
polar-glacial conditions.

We analysed granitoids of Permian to Eocene age along an east-west transect at

69-70°S, crossing from the Gondwana continental margin in the east, across the
magmatic arc and into the fore-arc region in the west. We used relatively fresh
samples that had previously been dated by U-Pb and Rb-Sr. Stable isotope
compositions lie well outside the normal magmatic range ( 18 O -7.3 to +10.6 ‰,
D -75 to -214 ‰). There is no correlation with source ( e Sr i or e Nd i ), in fact 50
Ma I-type plutons in the oceanic fore-arc have the highest 18 O values (+9.5 to
+10.6 ‰). Other felsic plutons have extraordinarily low 18 O in all minerals (<+5
‰) with quartz, normally resistant to low-temperature isotopic exchange, as low
as +3.9 ‰. Although, pervasive alteration is absent, and initial assessment of
mineral-mineral equilibria indicates isotopic equilibration at near-magmatic
temperatures (700 to 400°C), some biotite 18 O compositions (- 7 to +3 ‰) are
as low as the huge, intensely altered porphyry deposits at Butte and Climax.
Other mineral pairs give temperatures of 400-100°C, typical of hydrothermal
systems, and one third of mineral pairs are in gross disequilibrium.

All the plutons are D-depleted compared to normal mantle and crust. Most D data
suggest a broad decrease in D with time. This is unrelated to lithology and
source; low D values (-146 to –172 ‰) occurr in the fore-arc granites ( 18 O (+ 9
to +11 ‰). The lowest D (-209, -214 ‰) occurs in altered extension-related
gabbros (140 Ma) and arc granodiorite (120 Ma) in western Palmer Land.

There is a striking lack of correlation of D with 18 O. Some unaltered intrusions,

with ‘magmatic' 18 O and isotopic temperatures of >400° contain OH minerals
with very low D (-170 ‰) suggesting either a low-D source, or extreme loss of D
by degassing. By contrast, other ‘normal'- D intrusions (~-80 ‰) contain biotite
18 O as low as -7 ‰, isotopic temperatures of <400°, and discordant Rb-Sr ages,
indicating alteration by relatively D-enriched meteoric waters. Two 18 O- D
groups are present. A low-D group may have formed from (1) a low-D melt, (2)
by assimilation of low-D crustal rocks, (3) from highly degassed magmas, or (4)
by cryptic low water:rock ratio alteration by high-latitude meteoric waters. The
low- 18 O group, in isotopic disequilibrium but lacking obvious alteration,
indicates near-solidus equilibration with isotopically light meteoric waters at high
water:rock ratios. Such light values require buffering of the circulating waters by
18 O -depleted basement.

Three other recent studies of unmineralised , not obviously altered, high-

palaeolatitude plutons also indicate widespread alteration by light meteoric
waters [2, 3, 6]. Hence meteoric water flooding reported in large porphyry
systems elsewhere may be typical of the high-level intrusive environment, and
not related to the mineralising process. High-level plutons, and shallow
hydrothermal systems [9] in Antarctica could act as proxies for meteoric water
compositions, palaeolatitude and continental climate.
References

1] Blattner, P. & Williams, J.G. 1991. The Largs high-latitude oxygen isotope
anomaly ( New Zealand ) and climatic controls of oxygen isotopes in magmas.
Earth and Planetary Science Letters , 103 , 270-284.

2] Blattner, P., Grindley, G.W. & Adams, C.J. 1997. Low 18 O terranes tracking
Meosozic climates in the South Pacific. Geochimica et Cosmochimica Acta , 61 ,
569-576.

3] Dallai, L., Ghezzo, C. & Longinelli, A. 2001. Fossil hydrothermal systems

tracking Eocene climate change in Antarctica . Geology , 29 , 931-934.

4] Harris, A.C. & Golding , S.D. 2002. New evidence of magmatic-fluid - related
phyllic alteration: implications for the genesis of porphyry Cu deposits. Geology ,
30 , 335-338.

5] Kerrich, R. & Ludden, J. 2000. The role of fluids during formation and evolution
of the southern Superior Province lithosphere: an overview. Canadian Journal of
Earth Sciences , 37 , 135-164.

6] Nevle, R.J., Brandriss, R.E., Bird, D.K., McWilliams, M.O. & O'Neil, J.R. 1994.
Tertiary plutons monitor climate change in East Greenland . Geology , 22 , 775-
778.

7] Rumble, D., Giorgis, D., Ireland , T., Zhang, Z., Xu, H., Yui, T.F., Yang, J., Xu,
Z. & Liou, J.G. 2002. Low d 18 O zircons, U-Pb dating, and the age of the
Qinglongshan oxygen and hydrogen isotope anomaly near Donghai in Jiangsu
Province , China . Geochimica et Cosmochimica Acta , 66 , 2299-2306.

8] Touret, J.L.R. 2003. Fluids in the deep earth. Journal of Geochemical

Exploration , 78-79 , 659-663.

9] Willan, R.C.R. & Armstrong , D.C. 2002. Successive geothermal, volcanic-

hydrothermal and contact-metasomatic events in Cenozoic volcanic-arc basalts,
South Shetland Islands , Antarctica . Geological Magazine , 139 , 209-231.

Interpreting the chemical composition of fluid inclusions: The connection

to bulk water analyses

B.W.D. Yardley & D.A. Banks

School of Earth Sciences , University of Leeds , Leeds LS2 9JT , UK

[email protected]

Analyses of the cation and anion concentrations in fluid inclusions have become
relatively widespread in the literature, and greatly extend the range of crustal
environments for which the fluid chemistry is known. We have compiled a
database of both saline fluid inclusion analyses, ranging from low-grade
metamorphic to magmatic systems, and bulk fluid analyses from oilfield and
geothermal drilling.

Many chemical trends are similar for both data sets. For example, suites of high-
T (magmatic) fluid inclusion analyses (obtained by both single inclusion and bulk
crush-leach techniques) show similar relationships between transition metal
concentrations and chloride to oilfield brines, indicating similar speciation despite
very different temperatures and absolute metal levels (Figure 1).

Figure 1: Relationship between Cl for fluids from 2 contrasting settings. Circles,

oilfield brines, diamonds, magmatic fluid inclusions.

Transition metals vary in concentration systematically with temperature as well

as salinity, and these trends also are continuous from low-T oilfields through
metamorphic fluid inclusions, geothermal brines and magmatic fluid inclusions.
(Figure 2).
We conclude from these results firstly that the analytical techniques applied to
fluid inclusions today are sufficiently robust to yield reliable results at a useful
level for interpretation. However the consistent trends over such a wide range of
crustal conditions are themselves a demonstration of the relatively narrow range
of chemical environments in the crust. Common rock assemblages must limit f O
2 , pH and a H 2 S within rather narrow limits for salinity and temperature to
emerge as such dominant controls on metal levels in fluids.

Figure 2: Relationship between Zn/Cl 2 of oilfield and geothermal brines, and

fluid inclusions, plotted against measured or estimated temperature of the fluid
(trapping T for inclusions). Squares - shield brines, circles – oilfield brines,
triangles – metamorphic fluid inclusions, diamonds – magmatic fluid inclusions
and geothermal brines

References

AUDETAT, A., GUNTER, D. & HEIRRICH, C.A. (2000): Causes for large-scale
metal zonation around mineralised plutons: Fluid inclusion LA-ICP-MS evidence
from the Mole granite, Australia . Economic Geology , 95, 1563-1581.

BANKS, D.A., GUILIANI, G. YARDLEY, B.W.D. & CHEILLETZ, A. (2000):

Emerald mineralisation in Columbia : Fluid chemistry and the role of brine mixing.
Mineralium Deposita , 35, 699-713.
BOTTRELL, S.H. & YARDLEY, B.W.D. (1988): The composition of a primary
granite-derived ore fluid from SW England , determined by fluid inclusion
analysis. Geochimica et Cosmochimica Acta, 52, 585-588.

CARPENTER, A.B., TROUT, M.L. & PICKETT, E.E. (1974): Preliminary report
on the origin and chemical evolution of lead and zinc-rich oilfield brines in Central
Mississippi . Economic Geology , 69, 1191-1206.

CONNOLLY , C.A. , WALTER, L.M., BAADSGAARD, H. & LONGSTAFFE, F.

(1990): Origin and evolution of formation waters, Alberta Basin , Western Canada
Sedimentary Basin . I. Chemistry. Applied Geochemistry , 5, 375-395.

FRITZ, P. & FRAPE, S.K. (Eds.) (1987): Saline water and gases in crystalline
rocks. Geological Association of Canada Special Paper 33.

KAMENETSKY, V.S., van ACHTERBERGH, E., RYAN, C.G., NAUMOV, V.B.,

MENAGH, T.P. & DAVIDSON, P. (2002): Extreme chemical heterogeneity of
granite-derived hydrothermal fluids: an example from inclusions in a single
crystal of miarolitic quartz. Geology, 30, 459-462.

LAND, L.S., McPHERSON, G.L. & MACK, L.E. (1988): The geochemistry of
saline formation waters, Miocene offshore Louisiana . Gulf Coast Association of
Geologists Transactions, 38, 503-511.

McCAIG, A.M., TRITLLA, J. & BANKS, D.A. (2000): Fluid mixing and recycling
during Pyrenean thrusting: evidence from Fluid inclusion halogen ratios.
Geochimica et Cosmochimica Acta , 64, 3395-3412.

MEERE, P.A. & BANKS, D.A. (1997): Upper crustal fluid migration: an example
from the Variscides of SW Ireland. Journal of the Geological Society, London ,
154, 975-985.

MUNZ, I.A., YARDLEY, B.W.D., BANKS, D.A. & WAYNE, D. (1995): Deep
penetration of sedimentary fluids in basement rocks from southern Norway :
evidence from hydrocarbon and brine inclusions in quartz veins. Geochimica et
Cosmochimica Acta, 59, 239-254.

SMITH, M.P., BANKS, D.A., YARDLEY, B.W.D. & BOYCE, A. (1996): Fluid
inclusion and stable isotope constraints on the genesis of the Cligga Head Sn-W
deposit, SW England . European Journal of Mineralogy, 8, 961-974.

WILLIAMS, A.E. & McKIBBEN, M.A. (1989): A brine interface in the Salton Sea
geothermal system, California : fluid geochemical and isotopic characteristics.
Geochimica et Cosmochimica Acta, 53, 1905-1920.
 Mantle Derived Fluids in a Granite Associated Sn-W Deposit: He and Ar
Isotopes from Panasqueira

Pete Burnard 1 and Dave Polya 2

1 Centre de Recherches Pétrographiques et Géochimiques (CRPG-CNRS), B.P.

20, 54501 Vandoeuvre-lès- Nancy , France . e-mail : [email protected]

2 Department of Earth Sciences, University of Manchester , Manchester , United

Kingdom , M13 9PL e-mail: [email protected]

Hydrothermal fluids associated with ore generation commonly contain 3 He

derived from the Earth's mantle (e.g. Simmons et al, 1987; Stuart et al., 1995; Hu
et al., 1998; Burnard et al., 1999; Kendrick et al., 2001; Ballentine et al., 2002;
Kendrick et al., 2002; Shail et al., 2003). The association between ore formation
and 3 He is not accidental: like 3 He, the heat required to fuel an ore-forming
hydrothermal cell originates in the mantle. As a result, a basic ingredient for ore
genesis - heat - is accompanied by 3 He, usually supplied by the mantle in fairly
constant proportions (Lupton et al., 1989). Dilution of 3 He and heat bearing
“magmatic” fluids by cooler fluids is common in many ore forming systems
because the physico-chemical changes during cooling of the magmatic fluids
leads to precipitation of metalliferous ores. Therefore the mechanisms and
timescales of mixing between magmatic and “dilute” fluids are of considerable
interest to economic geologists.

The distribution of He and Ar isotopes in the Earth mean that they naturally
complement each other during mixing between broadly “magmatic” and “surface”
derived fluids. Helium (especially 3 He) is practically absent in surface derived
fluids. As a consequence, the He isotopic composition of a low temperature,
groundwater type fluid will change dramatically on addition of traces of mantle-
derived “magmatic” fluid. Conversely, 36 Ar concentrations are extremely low in
mantle-derived magmas and fluids (Burnard et al., 1997), therefore the 40 Ar/ 36
Ar ratio of a predominantly magmatic fluid is sensitive to trace additions of low
temperature marine or meteoric fluids. Analyses of He and Ar isotopes from the
same sample covers mixing over the complete range of magmatic-dominated to
low-temperature-dominated hydrothermal fluids.

From analyses of He and Ar isotopes and abundances in arsenopyrite and

wolframite from the Panasqueira W-Cu(Ag)-Sn deposit (Portugal), we show that
mantle-derived volatiles were extremely important in the genesis of these granite
associated ores.

The fluids trapped in Panasqueira ore minerals differ from most sulfides from
continental hydrothermal deposits analyzed to date (e.g. Stuart et al., 1995; Hu et
al., 1998; Burnard et al., 1999; Ballentine et al., 2002; Kendrick et al., 2002)   in
that they are characterized by comparatively high and constant 3 He/ 4 He
(excepting the wolframite sample) The He and Ar compositions of many
hydrothermal ore-forming fluids typically reflect mixing between two fluids, one
with high 40 Ar/ 36 Ar , high 3 He/ 4 He and the other with low 40 Ar/ 36 Ar and 3
He/ 4 He , (Stuart et al., 1995; Hu et al., 1998; Burnard et al., 1999; Ballentine et
al., 2002;), and it is likely that mixing accounts for the range in 40 Ar/ 36 Ar
observed in the Panasqueira arsenopyrite. The high 40 Ar/ 36 Ar , high 3 He/ 4
He fluid most likely was derived from a magmatic source while the low 3 He/ 4
He, low 40 Ar/ 36 Ar fluid was undoubtedly a surface derived (meteoric or
marine) fluid; the low temperature fluid probably had 3 He/ 4 He ratios lower than
true air saturated water (ASW) because meteoric or marine fluids will entrain
radiogenic 4 He (but not 40 Ar) from the crust during residence in shallow
aquifers. The lack of variation in 3 He/ 4 He suggests that the He concentration
of the low 3 He/ 4 He endmember is very low. However, it is not possible to
exclude the possibility that the range in measured 40 Ar/ 36 Ar in these samples
is due to variable degrees of atmospheric contamination of a single
homogeneous fluid trapped in all the arsenopyrite samples.

Thus, helium trapped in arsenopyrite and wolframite from the Panasqueira

hydrothermal deposit in central Portugal is overwhelmingly - » 80% - derived
from the mantle. This is surprising as the W-Cu(Ag)-Sn vein deposits are
spatially and temporally associated with an S-type granite. It is difficult to
reconcile 3 He/ 4 He ratios > 5Ra (where Ra is the atmospheric 3 He/ 4 He ratio)
with a granite produced by crustal anatexis, therefore it seems likely that mantle-
derived fluids unrelated to the granite were the source of the magmatic
component in the ore fluids.

The mantle appears to transport 3 He and heat to the surface in relatively

constant proportions ( 3 He/Q = 0.1-1 x10 -12 cm 3 STP J -1 ). However, the   3
He/heat ratios in the Panasqueira fluids (1-10 x10 -12 cm 3 STP J -1 , calculated
using the 3 He/ 36 Ar ratio) are considerably higher. Therefore, it does not seem
likely that gradual cooling of the pluton supplied heat to the hydrothermal system
(as this would result in low 3 He/Q ratios). Therefore, both He and heat in this
collision-related, granite-associated hydrothermal deposit were supplied from the
mantle.

References

Ballentine C. J., Burgess R., and Marty B. (2002) Tracing fluid origin, transport
and interaction in the crust. Reviews in Mineralogy and Geochemistry 47 , 539-
614.

Burnard P. G., Graham D. W., and Turner G. (1997) Vesicle-specific noble gas
analyses of "popping rock": implications for primordial noble gases in the Earth.
Science 276 , 568 - 571.
Burnard P. G., Hu R., Turner G., and Bi X. (1999) Mantle, crustal and
atmospheric noble gases in Ailaoshan gold deposits, Yunnan Province, China.
Geochim. Cosmochim. Acta 63 (10), 1595-1604.

Hu R., Burnard P. G., Turner G., and Xianwu B. (1998) Helium and argon isotope
systematics in fluid inclusions of Machangqing copper deposit in West Yunnan
Province , China . Chem. Geol. 146 (1-2), 55-63.

Kendrick M. A., Burgess R., Leach D., and Pattrick R. A. D. (2002) Hydrothermal
fluid origins in Mississippi valley-type ore districts; combined noble gas (He, Ar,
Kr) and halogen (Cl, Br, I) analysis of fluid inclusions from the Illinois-Kentucky
fluorspar district, Viburnum Trend and Tri-State districts, Midcontinent United
States. Econ. Geol. 97 (3), 453-469.

Kendrick M. A., Burgess R., Pattrick R. A. D., and Turner G. (2001) Fluid
inclusion noble gas and halogen evidence on the origin of Cu-porphyry
mineralising fluids. Geochim. Cosmochim. Acta 65 (16), 2651-2668.

Lupton J. E., Baker E. T., and Massoth G. J. (1989) Variable 3He/heat ratios in
submarine hydrothermal systems: evidence from two plumes over the Juan de
Fuca ridge. Nature 337 , 161 - 164.

Shail, R.K., Stuart, F.M., Wilkinson, J.J. and Boyce, A.J. (2003) The role of Post-
Variscan tectonic and mantle melting in the generation of the lower Permian
granites and the giant W-As-Sn-Cu-Zn-Pb orefied of SW England . Applied Earth
Science (Trans. Inst. Min. Metall. B) , 112, B127-128 [abstract]

Simmons S. F., Sawkins F. J., and Schlutter D. J. (1987) Mantle - derived helium
in two Peruvian ore deposits. Nature 329 , 429 - 432.

Stuart F. M., Burnard P. G., Taylor R. P., and Turner G. (1995) Resolving mantle
and crustal contributions to ancient hydrothermal fluids: He - Ar isotopes in fluid
inclusions from Dae Hwa W-Mo mineralisation. S. Korea . Geochim. Cosmochim.
Acta 59 , 4663 - 4673.

Spectacular silver segregation on the surface of sulfides  

E. L. Hobson 1 , P. Wincott 1 , D. J. Vaughan 1 , R. A. D Pattrick 1

1 Department of Earth Sciences and Williamson Research Centre for Molecular

Environmental Science, University of Manchester , Oxford Road, UK , M13 9PL.

Email: [email protected]
The study of mineral surfaces is critical to understanding the behaviour of
minerals during processes used in metal extraction, especially froth flotation (and
leaching). Of particular interest to sulfide mineral flotation is their behaviour in
alkaline solutions. The major aim of this study is to examine the effect of
solutions containing Ca(OH) 2 on the surface chemistry of the complex sulfide
minerals of the tetrahedrite Cu 12 Sb 4 S 13 -tennantite Cu 12 As 4 S 13 group.
Of specific interest are the changes in the amount and speciation of silver at the
mineral surface following treatment with Ca(OH) 2 . Natural samples of tennantite
(Cu 10   Zn 2 As 4 S 13 ) from Casapalca , Peru where analysed using the X-ray
photoelectron spectroscopy (XPS). Analysis of the fresh surface of tennantite
sample showed that copper, silver, zinc, lead, arsenic, antimony and sulfur are all
present at the surface as sulfide species. Following treatment in a saturated
Ca(OH) 2 solution the surface chemistry of the tennantite changed dramatically.
The most notable change was the decrease in concentrations of all elements
except oxygen and silver, the latter being increased to at least 4 monolayer
coverage. It is clear from XPS analysis that silver at the surface of tennantite is
not a metallic species but is consistent with a silver sulfide species. The
mechanisms for the migration of silver from the bulk to the surface, and the
environment of the silver at the surface of the minerals are critical to
understanding the process of silver enrichment at the surfaces of tetrahedrites –
tennantites following alkaline leaching. A potential driving force for the
mechanism of migration of silver through the bulk to the surface is the leaching of
copper resulting in an unstable metal-depleted surface; copper concentrations
were observed in the Ca(OH) 2 solution. This could act as an ‘effective'
geochemical gradient causing the movement of silver cations through the bulk to
stabilise the surface. The enormous enrichment silver at the surface subsequent
to migration results in the formation of a layer of a new phase on the surface of
the mineral. Diffusion through sulfides is known to occur through a variety of
pathways, for example fractures, structural planes, and via vacancies, and all
these mechanisms are likely to have contributed to the migration of silver from
the bulk of the tennantite to the surface.

|| MDSG 2004 home page || Scientific Programme ||