Quick Search:    advanced search

GSW Home   GeoRef Home   My GSW Alerts   Contact GSW   About GSW   Journals List   Help

This Article Abstract Figures Only Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (3) Citing Articles via Google Scholar Google Scholar Articles by Wilkinson, J. J. Articles by Boyce, A. J. Search for Related Content GeoRef GeoRef Citation

Ore-Forming Processes in Irish-Type Carbonate-Hosted Zn-Pb Deposits: Evidence from Mineralogy, Chemistry, and Isotopic Composition of Sulfides at the Lisheen Mine

J. J. Wilkinson and S. L. Eyre^*

Department of Earth Science and Engineering, Royal School of Mines, Imperial College London, Exhibition Road, London SW7 2AZ, United Kingdom

A. J. Boyce

Scottish Universities Environmental Research Centre, East Kilbride, Glasgow G75 0QF, Scotland

Corresponding author: e-mail, j.wilkinson{at}imperial.ac.uk

	Abstract

Top Abstract Introduction Geologic Setting Geology of the Lisheen... Methodology Ore Mineralogy and Paragenesis Sulfur and Lead Isotopes Discussion Model for Mineralization Conclusions APPENDIX References

 
Lisheen is a strata-bound zinc-lead deposit formed during the Mississippian by replacement of hydrothermally dolomitized, grossly stratiform breccia bodies located near the base of the Carboniferous Waulsortian Limestone. It represents one of a number of carbonate-hosted massive sulfide ore deposits in the Irish ore field that, due to several unique features, have been classified as Irish type.

Disseminated pyrite occurs in preore dolomite and around the margins of preore dolomite clasts within dolomite breccias. Early fine-grained sphalerite-pyrite mineralization occurs as infill of intergranular dolomite porosity. Locally, massive to semimassive iron sulfide is observed, mainly comprising pyrite with lesser marcasite. A complex polymetallic sulfide assemblage typifies the main ore stage, dominated by fine-grained disseminated, massive or colloform sphalerite and galena, with minor pyrite, chalcopyrite, arsenopyrite, tennantite, nickel- and cobalt-bearing minerals. Silver occurs in solid solution in tennantite, galena, and sphalerite. Dolomite and barite dominate the gangue, with lesser calcite. Main-stage mineralization involved the progressive replacement of preexisting iron sulfides and the dolomite breccias, initially by replacement of the breccia matrix and ultimately by replacement of clasts. Coarse crystalline sphalerite and euhedral galena crystals are generally restricted to fracture-fill mineralization or vugs within main ore-stage assemblages where they occur with euhedral dolomite and calcite.

Barite intergrown with main ore-stage sulfides has ³⁴S values of 14.3 to 18.1 per mil, consistent with the derivation of sulfate from coeval Carboniferous seawater. The ³⁴S values for sulfides range from –44.1 to +11.8 per mil, with a mean value of –13.7 per mil, typical of the Irish ore deposits. The dominant low ³⁴S signature is considered to be the result of bacterial reduction of coeval seawater sulfate. Extremely low ³⁴S values, in the range of –38 to –44 per mil, are only observed in preore disseminated pyrite; such extreme fractionations are thought to be due to low bacterial sulfate reduction rates coupled with oxidative cycles in near-sea floor pore waters. Main ore-stage sulfides have ³⁴S values in the range of –4 to –18 per mil, with a mode of –10 per mil, consistent with a typical bacterial fractionation from coeval seawater sulfate. Isotopic equilibrium between cogenetic sulfides is not observed. The bacteriogenic sulfur component was probably transported from bacterial colonies fringing the ore system by low-temperature brines.

The ³⁴S values of late ore-stage sulfides mainly range from –20.2 to +12.0 per mil, with the majority having relatively high values (mean = –3.0 ± 8.5, 1) interpreted as being due to the presence of a hydrothermal sulfur component, leached from the lower Paleozoic basement. For galena and sphalerite there is a general increase in ³⁴S values with depth in the system, with time, and with proximity to east-west– and northwest-trending faults. These relationships suggest that input of hydrothermal sulfur from depth via fractures became increasingly important. Hydrothermal sulfur appears to be more important at Lisheen than the other major Irish deposits.

Galena lead isotope analyses gave average ²⁰⁶Pb/²⁰⁴Pb, ²⁰⁷Pb/²⁰⁴Pb, and ²⁰⁸Pb/²⁰⁴Pb values of 18.183, 15.594, and 38.080, respectively. These data do not correlate with ore-stage, galena texture or ³⁴S. The results are comparable to previous data from Lisheen and from Silvermines, 35 km to the west, implying a common lead source in the lower Paleozoic basement.

The textural, mineral, chemical, and isotopic evidence suggests that main-stage ore was precipitated as a consequence of rapid supersaturation, caused by fluid mixing within the permeable dolomite breccias. This process involved relatively high temperature (ca. 240°C), metal-bearing solutions derived from a basement-equilibrated fluid reservoir (carrying Zn, Pb, Fe, Cd) and shallow, saline (ca. 25 wt % NaCl equiv) formation waters rich in bacteriogenic H₂S. Minor metals (Cu, As, Ni, Co) are thought to have been stripped from the footwall Old Red Sandstone during hydrothermal alteration around fault conduits. The availability of abundant seawater sulfate, operation of open-system bacterial sulfate reduction, and episodic availability of free oxygen imply that ore formation cannot have occurred at significant depth below the paleosea floor. Cessation of mineralization was due to a cut-off of the sulfur-rich brine supply, possibly by deposition of impermeable hanging-wall sediments. This process of ore formation is consistent with evidence from the other economic Irish-type deposits in the ore field.

	Introduction

Top Abstract Introduction Geologic Setting Geology of the Lisheen... Methodology Ore Mineralogy and Paragenesis Sulfur and Lead Isotopes Discussion Model for Mineralization Conclusions APPENDIX References

 
The Lisheen deposit, situated in County Tipperary, Ireland (Fig. 1), is the most recent of six carbonate-hosted Zn-Pb deposits to have been discovered in the Irish ore field in the last 35 yr. Found in 1990 by a Chevron Mineral Corporation-Ivernia West joint venture, the deposit began production in late 1999. It contained a total premining resource (undiluted) of 16.7 million metric tons (Mt) of ore, grading 14.1 percent Zn and 2.4 percent Pb (weighted average calculated from Fusciardi et al., 2003), together with approximately 16 percent Fe and 26 g/t Ag (Elmes et al., 2000). The ore is hosted principally by carbonate breccias near the base of the Mississippian (Courceyan) Waulsortian Limestone. The deposit is the most significant of a number of mineralized occurrences in the Rathdowney trend of the southern Midlands (Fig. 1) that include the Galmoy deposit (Doyle et al., 1992), 10 km to the northeast. Lisheen, together with the deposits of Galmoy, Navan, Tynagh, and Silvermines, represents the deposit class known as "Irish-type" as redefined by Wilkinson (2003).

View larger version (55K): [in this window] [in a new window]   FIG. 1. Simplified geologic map of the Irish Midlands, showing location of major deposits.

	Geologic Setting

Top Abstract Introduction Geologic Setting Geology of the Lisheen... Methodology Ore Mineralogy and Paragenesis Sulfur and Lead Isotopes Discussion Model for Mineralization Conclusions APPENDIX References

 
The Irish deposits are hosted by a Lower Carboniferous marine transgressive sequence that conformably overlies a terrigenous clastic red-bed succession (Old Red Sandstone) of Devonian-Carboniferous age. This lies unconformably on a peneplained basement of Silurian graywackes, siltstones, shales, and volcanic rocks that were deformed and weakly metamorphosed in the Caledonian orogeny. Mineralization is hosted by a variety of carbonate facies, primarily shallow marine deposits of the Navan Group and the deeper water mud mounds of the Waulsortian Limestone (Hitzman and Large, 1986; Phillips and Sevastopulo, 1986; Andrew, 1993; Hitzman, 1995; Hitzman and Beaty, 1996).

At Lisheen, the lower part of the Carboniferous marine sequence includes carbonate-rich shales, siliciclastic sedimentary rocks, and argillaceous bioclastic limestones (Lower Limestone Shale and Ballysteen Limestone; Philcox, 1983). The deposit is hosted by the Courceyan to early Chadian Waulsortian Limestone, which comprises coalesced wackestone or packstone mud mounds flanked by argillaceous bioclastic limestone (Philcox, 1983; Phillips and Sevastopulo, 1986). Growth fault control of sedimentation in the Courceyan-Chadian (immediately prior to and during deposition of the Waulsortian Limestone) is indicated by thickness variations across structures (Boyce et al., 1983; Hitzman et al., 2002; Carboni et al., 2003; Fusciardi et al., 2003).

	Geology of the Lisheen Deposit

Top Abstract Introduction Geologic Setting Geology of the Lisheen... Methodology Ore Mineralogy and Paragenesis Sulfur and Lead Isotopes Discussion Model for Mineralization Conclusions APPENDIX References

 
Stratigraphy
The local stratigraphy in the Lisheen area is described by Hitzman et al. (2002) and is summarized in Figure 2. The main host to mineralization is the Waulsortian Limestone, dominantly comprising massive, pale-gray biomicrite, together with green-gray and gray argillites, wavy laminated and stylonodular micrites. The Waulsortian is commonly brecciated, particularly toward its base, and it is these breccias that host the bulk of the ore. Although there is still debate over the relative importance of sedimentary and hydrothermal controls on the genesis of Waulsortian breccias (Hitzman and Beaty, 1996; Lee, 2002; Lee and Wilkinson, 2002; Reed and Wallace, 2003; Wilkinson and Lee, 2003), their significance is clear as the primary host rocks to ore in the south and west Midlands. At Lisheen, synsedimentary breccias are present locally in the hanging wall of normal faults (Hitzman et al., 2002; Carboni et al., 2003; Fusciardi et al., 2003), but it is not clear whether these are always precursors for the hydrothermally overprinted and mineralized dolomite breccias as has been documented at Silvermines (Lee and Wilkinson, 2002).

View larger version (30K): [in this window] [in a new window]   FIG. 2. Summary stratigraphy of the Lisheen area (modified after Eyre, 1998).

View larger version (41K): [in this window] [in a new window]   FIG. 3. Simplified structural map of the Lisheen deposit, showing main ore zones and locations of drill holes sampled in the present study (Lisheen mine, 1998).

Regional dolomitization
Several phases of dolomitization occur in the Lisheen area, the origin of which are still debated (e.g., Wilkinson, 2003). The earliest is a fine- to medium-grained, subhedral to euhedral gray dolomite that preferentially replaces micritic components of the precursor Waulsortian Limestone (D_1a). This is partially recrystallized to and/or postdated by a coarse white planar to weakly nonplanar dolomite (D_1b) that preferentially replaces coarser crystalline components of the Waulsortian, such as early marine calcite cements in stromatactis cavities and bioclasts. Together, these phases form what has been described as the "regional dolomite" that occurs throughout much of the Waulsortian in southeast Ireland (Hitzman et al., 1998).

In the Lisheen area, the entire Waulsortian Complex, as well as relatively nonargillaceous beds within the lowermost part of the overlying Crosspatrick Formation, are dolomitized (Shearley et al., 1996). Regional dolomite occurs on both the hanging-wall and footwall sides of the main Killoran and Derryville faults (Hitzman and Shearley, 1991, unpub. technical report for Chevron Mineral Corporation of Ireland, Intercompany Technical Group on Carbonates Seminar LN91000514). In the northern part of the Rathdowney trend and along the margin of the Leinster massif, dolomitization of this type has been observed locally to extend higher in the stratigraphy, into the Aghmacart Formation. However, dolomite has not been observed in the Arundian-age upper portion of the Aghmacart Formation (Sheridan, 1977). The Rathdowney trend essentially marks the northwest limit of pervasively dolomitized Waulsortian, with the Lisheen and Galmoy deposits located just to the southeast of the dolomite front (Hitzman and Beaty, 1996).

The most recent studies conclude that D_1a formed during early seawater diagenesis but later underwent neomorphic recrystallization (Wright et al., 1999, 2000; Gregg et al., 2001). D_1b is thought to have developed in response to the onset of extensive fault-controlled hydrothermal circulation in the late Courceyan-Chadian (Wilkinson, 2003). The regional dolomite plays a pivotal role in arguments concerning deposit genesis because it is considered to predate ore formation at Lisheen and Galmoy (Doyle et al., 1992; Shearley et al., 1995; Eyre, 1998; Hitzman et al., 2002; Wilkinson et al., 2003), thereby providing a maximum limit for the age of mineralization.

Hydrothermal dolomitization
A weakly developed, dark-gray dolomite of uncertain genesis forms the first localized alteration phase at Lisheen (Hitzman et al., 1992). The first well-developed dolomite of inferred hydrothermal origin (D₂) forms the matrix of the grossly stratiform, wedge-shaped breccia bodies ("black matrix breccias") that occur near the base of the Waulsortian Limestone (Fig. 4). The term "matrix" is used here in a general sense to mean the material surrounding breccia clasts, rather than implying a specific origin for this material since its origin is uncertain. At Lisheen, it is not clear whether the matrix dolomite is (1) merely a replacement of the precursor dolomitized limestone, (2) a true cement to breccia clasts, or (3) predominantly a replacement of an earlier (synsedimentary) breccia matrix, as has been shown to be the case in the Silvermines district (Lee and Wilkinson, 2002). It could have formed from a combination of these processes (Hitzman et al., 2002). Fluid inclusion data from the Tynagh basin have shown that the black dolomite commonly observed associated with mineralization in the Irish ore field is likely to be a characteristic product of fluid mixing (Wilkinson and Earls, 2000), probably forming ahead of sulfides in a prograding reaction front (Hitzman et al., 2002; Lee and Wilkinson, 2002; Wilkinson et al., 2003).

View larger version (81K): [in this window] [in a new window]   FIG. 4. North-south cross section through the Derryville zone, showing major lithological units and location of main ore lens (Lisheen mine, 1998).

Crackle breccias and stockworks cemented by medium- to coarse-grained white, planar or weakly nonplanar variably ferroan dolomite (D₃), commonly referred to as "white matrix breccias," form a broad halo to the D₂ dolomite breccias and are developed through much of the thickness of the Waulsortian. These are comparable to the "crystalline dolomite breccias" at Galmoy (Doyle et al., 1992). The relative timing of these breccias is uncertain as they tend to occur above and/or lateral to D₂ dolomite breccias (Hitzman et al., 2002), such that crosscutting relationships are rarely observed. Shearley et al. (1996) inferred a post-D₂ timing, but the local presence of both crystalline white dolomite in D₂ breccia clasts and black dolomite in D₃ dolomite breccias (also observed at Silvermines; Lee et al., 2001; Lee and Wilkinson, 2002) implies a broadly contemporaneous origin (also see Fusciardi et al., 2003). The textural differences are believed to result from the mode of formation; black dolomite precipitated rapidly during fluid mixing, whereas white dolomite crystallized from a single hydrothermal fluid (Earls and Wilkinson, 2000; Wilkinson, 2003).

The dolomite breccias are crosscut by generally steeply dipping veins of ferroan dolomite, dolomite (D₄), and calcite (C₄). These phases also fill residual vuggy porosity within earlier dolomite generations. The veins typically display symmetrical drusy crystal growth and may be composite. They also commonly contain sulfides.

A late-stage, distinctive pink dolomite (D₅) crosscuts and aggressively dissolves all previous dolomite and/or sulfide generations and is often coprecipitated with minor pyrite and/or chalcopyrite (Eyre, 1998; Hitzman et al., 2002; Fusciardi et al., 2003). This is similar to late pink dolomite observed elsewhere in the Midlands (e.g., Boast et al., 1981; Wilkinson, 2003). Sometimes, residual porosity in this dolomite is filled by a white, blocky calcite (C₆).

Mineralization
Mineralization at Lisheen occurs within 30 m of the base of the Waulsortian, forming single and locally multiple, generally stratiform, sulfide lenses separated by D₂ dolomite breccias (Fig. 4). The orebodies comprise texturally complex, polymetallic massive sulfide, displaying both gradational and sharp, planar or undulating contacts with the enclosing breccias. Sometimes, postore stylolites are developed along these contacts. Sulfides are also present in the Lisduff Oolite Member (Fig. 2), predominantly in the footwall but also in the hanging wall of the Killoran and Derryville faults (Eyre, 1998; Hitzman et al., 2002; Fusciardi et al., 2003). The Lisduff Oolite is much less extensively mineralized than the basal Waulsortian and only comprises a minor proportion of the total resource.

Four main ore zones have been defined: Main, North, Derryville, and Bog (Fig. 3). The Main zone forms the largest orebody, followed by Derryville, North, and the Bog zone. Detail on metal zoning patterns in the four zones is available in Hitzman et al. (2002) and Fusciardi et al. (2003).

	Methodology

Top Abstract Introduction Geologic Setting Geology of the Lisheen... Methodology Ore Mineralogy and Paragenesis Sulfur and Lead Isotopes Discussion Model for Mineralization Conclusions APPENDIX References

 
Textural and petrographic relationships of sulfides were determined by the detailed study of more than 200 polished blocks and thin sections in reflected light and under the scanning electron microscope (SEM). This work was based only on drill core samples collected between 1994 and 1997, prior to underground exposures becoming available for study. Representative drill core samples covering the spectrum of ore stages, mineralization styles, and a range of lateral and vertical positions within the four main ore zones (Fig. 3) were collected in order to evaluate temporal and spatial variations in sulfide mineralogy and sulfur isotope composition. Quantitative analysis of sulfides using the microprobe was carried out mainly on the ore-stage mineralization to determine the temporal and spatial distribution of trace elements, particularly in relation to putative hydrothermal feeder structures. Analyses were performed using a Cameca SX50 wavelength dispersive electron microprobe at the Natural History Museum, London. A summary of results is presented in Table 1 and full results are available from the author on request.

Lead isotope analysis
Galena samples for lead isotope analysis were drilled from core specimens. Approximately 2 to 5µg of galena was dissolved in an excess of 6M HCl. These solutions were pipetted directly onto a single rhenium filament using a standard lead loading technique with phosphoric acid and silica gel. The ²⁰⁶Pb/²⁰⁴Pb, ²⁰⁷Pb/²⁰⁴Pb, and ²⁰⁸Pb/²⁰⁴Pb ratios were measured on a VG 54E mass spectrometer. The results were normalized to the NBS 981 international standard and corrected for instrumental mass fractionation (see Table 3).

	Ore Mineralogy and Paragenesis

Top Abstract Introduction Geologic Setting Geology of the Lisheen... Methodology Ore Mineralogy and Paragenesis Sulfur and Lead Isotopes Discussion Model for Mineralization Conclusions APPENDIX References

Based on the broad textural and mineralogical associations observed within the deposit, mineralization has been classified into four stages: (1) preore stage, (2) main ore stage, (3) late ore stage, and (4) postore stage (Fig. 5).

View larger version (16K): [in this window] [in a new window]   FIG. 5. Summary paragenesis for the Lisheen deposit (after Eyre, 1998). Width of bar corresponds to relative abundance; bars for sulfides are shown hatched for clarity.

View larger version (91K): [in this window] [in a new window]   FIG. 6. Core photographs and reflected light photomicrographs of preore-stage samples. (a). Preore-stage pyrite preferentially replacing the matrix of typical black dolomite matrix breccia. Sample LK262/153.6m, Derryville zone. (b). Pale-brown early sphalerite preferentially replacing matrix of dolomite breccia. Sample LK121-189.3m, Main zone. (c). Preore-stage colloform pyrite with minor carbonate and sphalerite inclusions overgrown by subhedral pyrite (py) and marcasite (ma). Sample LK420/190.7m, Derryville zone. (d). Fine-grained granular pyrite (py) pseudomorphous after barite laths. Timing of euhedral pyrite uncertain. Subhedral-anhedral sphalerite (sp) overgrows and replaces pyrite. Sample LK262/162.1m, Derryville zone. (e). Massive sulfide sample showing colloform banding. Preore-stage fine-grained pyrite-marcasite (py-ma) with sphalerite-rich outer zone is overgrown by pyrite-marcasite band, itself succeeded and partly replaced by ore-stage sphalerite (sp), dolomite (dol), and euhedral galena (gn). Sample LK394/164.3m, Derryville zone (Derryville fault). (f). Preore-stage massive pyrite (py) and marcasite (ma) that has undergone cataclasis and cementation by dolomite (dol) and sphalerite (sp). Sample LK262/159.1m, Derryville zone.

Marcasite occurs in colloform masses, elongate radiating laths intergrown with pyrite, and as coarse subhedral grains. Sphalerite, galena, and dolomite locally occur as inclusions within the banded iron sulfides, either in discrete layers or as isolated crystals along particular iron sulfide growth bands (Fig. 6e). Galena commonly occurs as small spheres within massive iron sulfide. Sulfides displaying radial coxcomb textures are generally uncommon but are well developed in the Lisduff Oolite Member adjacent to the main fault zone.

Cataclastic textures in pyrite and marcasite are common near to major faults in the Main and Derryville zones. There is infilling and partial replacement of pyrite along fractures by sphalerite and dolomite (Fig. 6f) and also by ore-stage galena.

Main ore stage
A complex polymetallic sulfide assemblage characterizes the main ore stage, dominated by sphalerite and galena, with minor pyrite, chalcopyrite, arsenopyrite, tennantite, dolomite, and barite. Coprecipitation of some phases is suggested by the presence of mutual growth boundaries, although replacement of one sulfide by another is also common. No consistent paragenetic sequence can be defined and the assemblage appears to be the product of multiple overprinting and recrystallization events. Sulfides of the main ore stage often replace early iron sulfides (Fig. 7a) and overprint the dolomite breccias, preferentially replacing the D₂ dolomite matrix and ultimately also D_1a clasts (Fig. 7b). This sequence of progressive replacement can be clearly traced laterally into the ore from essentially pristine, unmineralized breccias. Apparently brecciated massive sulfide pseudomorphs may in fact represent preexisting dolomite breccia textures (Fig. 7b), although rebrecciation in situ during intense mineralization close to the main fault zones is also likely.

View larger version (98K): [in this window] [in a new window]   FIG. 7. Core photographs and reflected light photomicrographs of main ore-stage samples. (a). Massive fine-grained pyrite in gray Waulsortian micrite (m), crosscut and replaced by sphalerite (sp) and galena (gn) in white dolomite stockwork. Sample LK121/190.2m, Main zone. (b). Complex massive sulfide with pyrite (py) intergrown with and replaced by galena (gn) and pale-brown sphalerite (sp). Note apparent breccia texture (replaced dolomite breccia?). Sample LK262/159.85m, Derryville zone. (c). Colloform sphalerite containing thin bands of fine-grained pyrite and occasional botryoidal pyrite intergrown with dolomite and fine-grained, zinc-rich (?geopetal) carbonate. Sample LK262/159.85m, Derryville zone. (d). Pyrite (py) and sphalerite (sp) incorporated in dolomite (dol) euhedra along growth surfaces and overgrown by apparently homogeneous colloform sphalerite. Sample LK416/193.6m, Derryville zone. (e). Selective replacement of foraminifera (Endothyrid). Sphalerite (sp) preferentially replaces chamber walls with chambers filled by galena (gn) and carbonate gangue (c), the whole surrounded by pyrite (py). An unidentified Fe-As sulfide (u) with minor Ni, Co, and In forms a thin layer inside the microfossil. Sample LK137/134.9m, Derryville zone (sub-Waulsortian, footwall of Derryville fault). (f). Mixed ore-stage sulfide assemblage comprising intergrown euhedral pyrite (py) and galena (gn), together with sphalerite (sp), probably pseudomorphous after barite laths. Sample LK325/146.2m, Derryville zone. (g). Selective replacement of ooid by galena (gn), surrounded and partly replaced by sphalerite (sp) and minor pyrite (py). Galena replacement may postdate plastic deformation of ooid reflected by oval shape and curviplanar contacts that parallel distorted laminae (upper right). No replaced stylolites truncating laminae, or fractured, or spalled laminae are observed indicating replacement preceded mechanical compaction, or that mechanical compaction did not occur. Sample LK394/164.3m, Derryville zone (Derryville fault). (h). Symmetrical vug-fill sulfides comprising sphalerite (sp) intergrown with laminated and/or colloform pyrite (py), with both being replaced from the margins inward by subhedral galena (gn). Carbonate infilling in remaining space (c), with minor dissolution of sphalerite along contact. Sample LK420/196.0m, Derryville zone.

Colloform sphalerite displays marked compositional variation (Table 1). Individual samples may contain up to 40 distinct zones with significant variations in Cd, Fe, and As concentrations (Fig. 8a). Cu (up to 0.11 wt %) is also observed in some zones. The color of colloform sphalerite generally correlates with mineral chemistry. For example, in a sample from the North zone, gray-yellow bands contain up to 0.12 wt percent Fe (avg 0.06 wt % Fe), yellow sphalerite contains up to 1.95 wt percent Fe (avg 0.37 wt % Fe), and orange-brown layers contain up to 2.84 wt percent Fe (avg 0.63 wt % Fe). Other elements may also be present such as Ag (up to 0.24 wt %), Cd (up to 0.47 wt %), and As (up to 0.60 wt %). In two samples analyzed for Tl, highly variable concentrations of up to 0.54 wt percent were found. However, there are no statistically significant correlations between major or trace elements within any of the colloform sphalerite analyzed.

View larger version (9K): [in this window] [in a new window]   FIG. 8. Examples of variations in chemical composition of sphalerite as determined by microprobe analyses (supplementary information on detection limits and full data available from author on request). (a). Traverse across main ore-stage color-banded colloform sphalerite (sample 262/159.1m, Derryville zone). Note lack of correlation between element concentrations in individual bands and occasional detectable Cu. (b). Traverse across late ore-stage finely crystalline, red-amber color-zoned sphalerite (sample 471/210.4m, Main zone). This sample is unusual in that it shows elevated Cd and Fe and shows a rare correlation between Cd and Fe concentrations.

Replacement textures are common, such as the replacement of dolomite rhombs by sphalerite and/or galena. Selective replacement of microfossils by sphalerite and other sulfides is also observed locally (Fig. 7e), as has been reported at Navan (Anderson et al., 1998). Sphalerite bow-tie textures in pyrite and elongate laths of sphalerite in a mixed assemblage of pyrite and galena are probably pseudomorphs after early barite or gypsum (Fig. 7f). Sphalerite also occurs in small veinlets crosscutting and replacing pyrite (Fig. 7a) and galena.

Galena locally forms skeletal and dendritic growths, similar to those described from Navan (Anderson et al., 1998). However, ore-stage galena is predominantly coarsely crystalline or massive and replacive, as indicated by caries textures, where galena replaces D₂ dolomite, and by the selective replacement of microfossils, early marine cements, and ooids (Fig. 7g). Galena is particularly abundant in zones proximal to the Killoran and Derryville faults where it commonly contains inclusions of tennantite, especially in the Main zone. Galena inclusions are also found within tennantite. Boulangerite is commonly associated with galena.

Galena generally does not contain detectable concentrations of trace elements (Table 1). Massive Waulsortian-hosted ore-stage galena locally contains elevated Te (up to 0.27 wt %). Ag concentrations are usually below detection with the exception of a few samples located proximal to faults that contain up to 0.19 wt percent Ag. The phase that contains the highest concentrations of Ag in the deposit is tennantite (see below).

Tennantite, chalcopyrite, arsenopyrite, and barite are most abundant near the Killoran and Derryville faults, especially within the footwall Lisduff Oolite Member. In this unit, tennantite commonly occurs as irregular inclusions in galena, chalcopyrite, and sphalerite and is characterized by up to 3.91 wt percent Zn, 5.63 wt percent Fe, and 0.32 wt percent Ag (Table 1). This contrasts with tennantite in galena from the Waulsortian, which contains up to 8.55 wt percent Zn, less Fe (<1.72 wt %), and less Ag (<0.24 wt %).

Short intervals of massive chalcopyrite are observed in drill core from the footwall Lisduff Oolite of the Main zone and this contains up to 0.73 wt percent As (avg 0.16 wt %). Minor chalcopyrite in the Bog zone Lisduff Oolite also contains As, up to 0.11 wt percent (avg 0.07 wt %). Minor disseminated arsenopyrite rhombs commonly occur in ore-stage galena, proximal to faults in the Main and Derryville zones.

Main ore-stage pyrite generally comprises fine- to coarse-grained euhedra intergrown with and locally interstitial to sphalerite and galena (e.g., Fig. 7f). Trace element concentrations in ore-stage pyrite are quite variable (Table 1). Arsenic concentrations range up to 5.21 wt percent and are significantly higher than in preore pyrite. In As-rich samples, As correlates with Cu with both showing a strong antipathetic relationship with Fe. Co and Ni are usually below detection, with the notable exception of pyrite hosted by the Lisduff Oolite Member in the footwall of the Killoran fault (up to 0.44 wt % Co and 1.92 wt % Ni). This pyrite also contains up to 3.34 wt percent As, which is strongly correlated with Ni content. Elevated levels of Co and Ni are also observed in the Ballysteen Limestone close to a secondary fault (F-291) in the hanging wall of the Derryville fault (Fig. 3).

Barite was probably deposited both prior to and during the main period of sulfide mineralization. Early sulfides, in particular pyrite and sphalerite, and ore-stage sulfides (Fig. 7f) both replace barite. Later barite, associated with complex mixed sulfide assemblages, crosscuts early pyrite-sphalerite assemblages.

Late ore stage
Coarse (1–7 mm) crystalline sphalerite and euhedral galena crystals are restricted to fracture-fill mineralization that crosscuts the main ore-stage assemblages and to vugs within main-stage ore. These phases commonly predate or were coprecipitated with D₄ dolomite and C₄ calcite veins and vug-filling cements (Fig. 7h). Sphalerite was generally the last sulfide to precipitate, commonly as rims on galena. Concentrically zoned pyrite grains locally occur within these sphalerite rims and late, euhedral, fracture-filling galena is sometimes observed.

Although texturally late, there is no evidence to suggest that these phases are unrelated to the main ore stage and are therefore regarded as representing waning hydrothermal activity, precipitated within residual porosity and tectonic fractures.

View larger version (31K): [in this window] [in a new window]   FIG. 9. Variation of cadmium and iron concentrations in sphalerite as a function of textural type and location. Note that the vast majority of samples are enriched in either cadmium or sphalerite, but not both, suggesting substitution of one or other element into the same lattice site. Darker brown sphalerite may be either Cd or Fe rich. The most Cd rich sphalerites are hosted by the Lisduff Oolite; Fe-rich sphalerites are generally from within the main fault zones. Most sphalerite in the main orebodies (within thick hatched line) contains variable Fe and low Cd.

Postore stage
Sulfides that occur in veins that clearly crosscut all previous stages are defined as postore stage. At Lisheen, these sulfides are commonly associated with pink saddle dolomite (D₅), which is associated with extensive dissolution of earlier carbonates and sulfides. This often leaves extensive vuggy porosity and abundant pink dolomite typically correlates with poor ore grades. Chalcopyrite and pyrite are the dominant sulfides, typically forming encrustations on saddle dolomite crystals together with minor sphalerite and galena. There is white blocky calcite (C₆) infilling of residual porosity in some samples.

	Sulfur and Lead Isotopes

Top Abstract Introduction Geologic Setting Geology of the Lisheen... Methodology Ore Mineralogy and Paragenesis Sulfur and Lead Isotopes Discussion Model for Mineralization Conclusions APPENDIX References

 
Sulfur Isotopes
A total of 119 analyses were carried out on 53 samples of sulfides (mainly pyrite, sphalerite, and galena) and barite from the four main ore zones and from all stages of the paragenesis (Table 2, Appendix).

The ³⁴S values show a very wide range from –44.1 to +11.8 per mil, with a mean value of –13.7 per mil (Fig. 10a). Extremely low ³⁴S values in the range of –38 to –44 per mil are only observed in preore-stage disseminated pyrite (Fig. 10b), and pyrite in general has lower values than the other sulfides, a feature typical of the Irish-type deposits (e.g., Boyce, 1990; Anderson et al., 1998; Fallick et al., 2001; Blakeman et al., 2002). Sphalerite and galena have similar ranges in isotopic composition.

View larger version (12K): [in this window] [in a new window]   FIG. 10. Histograms of sulfur isotope compositions, subdivided according to (a) sulfide phase and (b) ore stage. Sphalerite and galena have very similar ranges and average values, whereas pyrite is much more variable and trends to much lower 34S values. Preore sulfides (mainly pyrite) are characterized by very low 34S values indicative of an extreme bacteriogenic fractionation. Main ore-stage values are intermediate due to mixing between bacteriogenic and hydrothermal sulfur sources coupled with less extreme bacteriogenic fractionation during sulfate reduction. Late ore-stage sulfides are dominated by hydrothermal sulfur with high 34S. Postore-stage sulfides have low 34S values, probably indicative of sulfur remobilized from early main ore-stage sulfides.

View larger version (20K): [in this window] [in a new window]   FIG. 11. Variations in sulfur isotope composition of (a) sphalerite, (b) galena, and (c) pyrite with depth and textural type. Outliers: BZ = Bog zone, DZ = Derryville zone, MZ = Main zone, NZ = North zone. Samples from sub-Waulsortian units are indicated. Sphalerite and galena show trends toward higher 34S values with depth in the main ore zones and in sub-Waulsortian sulfides consistent with input of an 34S-enriched sulfur component from depth. Pyrite does not show such a trend in the main ore zones, suggesting that much of it formed relatively early when the flux of hydrothermal sulfur from depth was minor or absent.

Sphalerite:

Sphalerite shows a broad correlation between sulfur isotope composition and sample depth. More negative ³⁴S values occur in shallower, Waulsortian-hosted sulfides, and higher ³⁴S values occur lower in the stratigraphy at the base of the Waulsortian and in the Ballysteen Limestone Formation, including the Lisduff Oolite Member (Fig. 11a). Variation in the sulfur isotope composition of sphalerite between the ore zones is also observed. Sphalerite in the Main zone is relatively homogeneous in sulfur isotope composition (Fig. 11a). ³⁴S values for sphalerite from the Derryville zone are slightly more negative and have a wider range, and sphalerite from the Bog zone has the widest range of isotopic compositions, albeit based on only four analyses (Fig. 11a).

Galena: ³⁴S values of galena (n = 34) fall within the range of –26.8 to +3.2 per mil (avg –11.7 ± 7.5 per mil, 1; Fig. 10a). Main ore-stage galena (n = 22), like sphalerite, has a narrow range of ³⁴S values with an average of –11.2 ± 3.3 per mil. Galena hosted by the Lisduff Oolite Member and intergrown with fine-grained sphalerite has the highest ³⁴S value of +3.2 per mil. The lowest ³⁴S value, –26.8 per mil, was obtained from late- to postore-stage, vein-hosted galena associated with tennantite at the base of Waulsortian-hosted massive sulfide in the Derryville zone. Galena shows a clearly defined increase in ³⁴S values with depth and increasing heterogeneity from the Main to Derryville to Bog zones (Fig. 11b), similar to sphalerite.

Pyrite: ³⁴S values in pyrite (n = 37) range from –44.1 to +11.8 per mil (avg –22.4 ± 13.3 per mil, 1; Fig. 10a). Disseminated preore-stage pyrite hosted by well-developed hydrothermal D₂ dolomite has the lowest ³⁴S values of –44.1 to –38.1 per mil. Main ore-stage pyrite (n = 24) has a restricted range of sulfur isotope compositions but is still more variable than sphalerite and galena and also has markedly lower values than the other sulfides, with a mean of –20.1 ± 7.1 per mil. In the Waulsortian, no correlation with depth is apparent (Fig. 11c). However, the highest ³⁴S value of +11.8 per mil is from relatively deep in the system, from a vein within the upper calcarenite member of the Ballysteen Limestone Formation. Pyrite from the Main zone has a more uniform isotopic composition than in the Derryville zone, similar to the other main sulfides. Only two data from one drill hole in the Bog zone were analyzed, and these are rather similar.

Chalcopyrite: Chalcopyrite samples, all from the Main zone and representing the main ore stage, have typical ore-stage ³⁴S values ranging from –6.6 to –8.3 per mil (n = 3; Fig. 10a).

Barite: Barite intergrown with main ore-stage sulfides in the Derryville zone (n = 2) has ³⁴S values of +18.1 and +17.2 per mil (repeat analyses on splits of a single sample, Fig. 10a). Barite associated with main ore-stage sulfides in the Main zone (n = 2) has lower values (although possibly sulfide contaminated) of +14.3 and +14.5 per mil.

Lead isotopes
A preliminary study of four galena samples was carried out to provide supplementary data on the lead isotope character of the deposit and to compare the results with existing data from the Irish ore field. The samples analyzed were from massive ore-stage galena and from vein-hosted, late- to postore-stage galena (Table 3). The results are compared to data from other deposits in the ore field in Figure 12.

View larger version (17K): [in this window] [in a new window]   FIG. 12. 207Pb/204Pb vs. 206Pb/204Pb plot, showing lead isotope compositions of galena from Lisheen (this study) with error bars, compared to deposit array (encompasses average values for all Irish deposits and prospects) and average Silvermines value (see Table 3). Lisheen data fall close to the Silvermines average, implying a common lead source from local lower Paleozoic rocks. Pb growth curve of Stacey and Kramers (1975) is marked in 250 Ma increments for reference.

	Discussion

Top Abstract Introduction Geologic Setting Geology of the Lisheen... Methodology Ore Mineralogy and Paragenesis Sulfur and Lead Isotopes Discussion Model for Mineralization Conclusions APPENDIX References

Since there is good evidence that mineralization is localized by the Lisheen fault array, the timing of fault movement provides some constraints on the timing of mineralization. We consider that it is highly unlikely for extensive fluid flow to exploit these potential fluid conduits except during periods of active tectonism, since any fault-related permeability would be rapidly sealed by hydrothermal precipitates in times of quiescence. Sediment thickness variations imply that fault-controlled sedimentation in the area initiated immediately prior to Waulsortian deposition, significantly earlier than previously thought (Carboni et al., 2003; Fusciardi et al., 2003). Thus, mineralization could have begun as early as the Courceyan (cf. Lee and Wilkinson, 2002). The frequent observation of cataclastically deformed preore sulfides in the Main and Derryville zones overprinted by main ore-stage mineralization (e.g., Fig. 6f) indicates that ore formation did not entirely postdate fault movements, as suggested by Sevastopulo and Redmond (1999), Hitzman et al. (2002), and Carboni et al. (2003). Synsedimentary faulting continued during deposition of the Crosspatrick Formation, suggesting that mineralization could have persisted into, or initiated in, the Chadian (Fusciardi et al., 2003).

The fact that ore-stage sulfides at Lisheen replace early marine cements and regional dolomite indicates that the main phase of mineralization must have occurred subsea-floor during or subsequent to early diagenesis of the Waulsortian Limestone. However, it is probable that ore formation occurred close in time and space to the oxic and/or suboxic transition in the connate porewaters (Wilkinson et al., 2003). Furthermore, the sulfur isotope data indicate that this must have occurred at a depth where the ore-forming system remained open to seawater sulfate. These features, coupled with the evidence for near-sea floor and exhalative mineralization in other Irish-type deposits (Wilkinson et al., 2003), suggest that mineralization probably occurred within a few tens of meters, and at most a few hundred meters below the sea floor, much sooner during burial than suggested by Hitzman et al. (2002).

Source of lead
The limited new lead isotope data from Lisheen show no obvious correlation between ore stage or galena texture and lead isotope composition and there is no indication of a covariance between lead isotopes and ³⁴S values in galenas (Table 3), as has been observed in some Mississippi Valley-type deposits (Sverjensky, 1981). This lead isotope homogeneity is consistent with results from the other Irish deposits and suggests a well-homogenized crustal reservoir. The data are comparable with those reported for Lisheen by Hitzman et al. (1992) and are similar to results from Silvermines, 35 km to the west (O’Keeffe, 1986; Fig. 12). The similarity between Lisheen and Silvermines implies a common lead source; the match between deposit galenas and estimated composition of the lower Paleozoic sequence in the southwest Midlands at the time of mineralization (Dixon et al., 1990) implicates Silurian graywackes, siltstones, and grits of the Hollyford Formation. The lead isotope data therefore support fluid-flow models involving deep basement circulation and are inconsistent with fluid flow through and derivation of lead from the Old Red Sandstone, as summarized by Everett et al. (2003).

Distribution and source of other metals
Petrographic and mineral chemical studies show that Cu-, As-, Co-, and Ni-bearing phases are abundant close to the major faults and in sub-Waulsortian mineralization, particularly within the Lisduff Oolite (also see Fusciardi et al., 2003). Similarly, these elements are enriched (and covary) in other main ore-stage phases (sphalerite, pyrite, chalcopyrite) in proximity to the major structures. The co-occurrence of these elements in a range of phases, both as major and minor constituents, suggests that they were enriched in the hydrothermal fluids active during the main ore stage. The clear spatial association with major east-west faults implies that these structures were important conduits for fluids transporting these elements from depth. Intense bleaching of the Old Red Sandstone is observed close to the main fault zones at Lisheen (Mallon, 1997; Everett, 1999; Hitzman et al., 2002; Everett et al., 2003), and results of acid leaching experiments on unaltered Old Red Sandstone samples show that Ni and Co are readily liberated (C. Everett, pers. commun., 2000). Therefore, it is suggested that Cu, As, Co, and Ni were stripped during early, localized alteration of the sandstone footwall around fault conduits and redeposited in ore-stage minerals at higher levels by buoyant hydrothermal fluids.

Trace element substitutions in sphalerite appear to have been controlled by different processes. Fe is generally enriched in vein sphalerite from close to faults, particularly in the Derryville and Bog zones and in the sub-Waulsortian stratigraphy. There is no obvious correlation with fluid inclusion homogenization temperatures (Eyre, 1998), suggesting that temperature is not a primary control on Fe content. Given the probability of high sulfur fugacity in the upper parts of the system during ore formation (see below), it is suggested that the Fe content of sphalerite is controlled by sulfur fugacity in an inverse relationship (cf. Czamanske, 1974).

Cd concentrations in sphalerite are generally decoupled from Fe and show no obvious systematic pattern in space or time. A weak correlation with sulfur isotope composition (Fig. 13) suggests that Cd may be cotransported with isotopically heavy sulfur (derived from depth; see below) and is favorably partitioned into the first sphalerite precipitated in fluid conduits. There is no evidence for liberation of Cd during alteration of the Old Red Sandstone, therefore it is suggested that Cd is sourced, with Pb, from the lower Paleozoic basement.

View larger version (8K): [in this window] [in a new window]   FIG. 13. Range of cadmium concentrations in sphalerite samples (determined by microprobe analysis) plotted against sulfur isotope composition. A weak positive correlation exists, suggesting that cadmium is associated with the hydrothermal sulfur component, both being derived from the lower Paleozoic basement.