Petrographic Comparison of Carbonate-hosted Hematite-Bearing Massive Replacement Ores of the Bingham, Superior, and Bisbee Districts

Introduction

The geology of carbonate-hosted ores in porphyry copper deposits, like their porphyry-hosted equivalents, reflect the geochemical environment in which they form. As with porphyry-hosted ores, carbonate-hosted ores within a porphyry district evolve with time toward more hydrolytic conditions. Einaudi (1982) reviewed the characteristics of porphyry copper skarns and described a spectrum of deposits based on the relative degree of development of early skarn (which correlated with potassic alteration in porphyry) and late "silica-pyrite" mineralization (the carbonate-hosted analogue of D-veins in feldspathic rocks) (footn 1). The principal factors governing the degree of development of massive replacement mineralization are still poorly understood. Einaudi suggests the ultimate control may be in the timing of groundwater influx, which might be "retarded due to deep emplacement, a prolonged period of intrusive activity, large size of intrusions, or relatively dry and impermeable wall rocks." These factors would enhance the development of the early skarn stage. Conversely, development of massive replacement bodies would be inhibited by these factors. Although the extent to which massive replacement mineralization develops relative to skarn within a district may depend on the system-wide variables listed above, the physical and geochemical processes that govern the formation of massive replacement ores, as reflected by their geological characteristics, appear to be fundamentally similar.

Bingham, Bisbee, and Superior, all porphyry-related districts (footn 2) that contain important carbonate-hosted ores, show some interesting similarities between their late-stage massive replacement ores. All three districts host massive pyrite-chalcopyrite bornite and hematite-chalcopyrite replacement ores. Samples from the deep North Ore Shoot drilling at Bingham (Utah), the Superior underground workings (Arizona), and the Dallas-Cole, Gardner, and Sacramento shafts (Bisbee, Arizona) were studied to characterize the mineralogy of the hematite-chalcopyrite ores. Descriptions of hematite-bearing ores of the Yauricocha district (Peru) from the literature were also noted, although samples from this deposit were not studied. As with other types of carbonate-hosted ores, the hematite-chalcopyrite ores in all three deposits show striking similarities in their grain size, relative proportions of opaque minerals, ore textures, metal grades, and overall zonal patterns, but differ in their non-opaque gangue mineralogy. The differences in their non-opaque mineralogy may simply reflect differences in their host rock compositions.

Preliminary review of three polished thin sections of a core sample from drill hole D363A (3215 footage mark) in the North Ore Shoot highlights some important similarities with the hematite-bearing carbonate-replacement ores of the Superior and Bisbee districts (Arizona) (table 1).

Host rocks

The host rocks in each district consist of either limestone or dolostone. The carbonate host rocks for most of the replacement ores at Bingham are silty/cherty limestones (Jordan/Highland Boy or Commercial/Yampa limestone (footn 3)). Ores of the Superior district occur primarily in clean, sparry (50-100 m grain size) dolostone of the lower Martin and lower Escabrosa Formations. Ores of the Bisbee district occur in the silty limestone of the Abrigo, the thinly-bedded lower Martin, and the more massive-bedded Escabrosa Formations. Because the ore-forming process for massive sulfide mineralization involves almost complete replacement of all components in the protolith by ore minerals, the dominant mineralogy of the ores is relatively independent of the type of carbonate rock that is replaced (in contrast to skarns which contain much greater amounts of minerals that contain locally-derived components such as Ca and Mg). Although the dominant mineralogy appears independent of protolith composition, the protolith does control two important additional aspects of replacement ores: mineralogy of minor, non-opaque minerals, and favorability of replacement. These controls are discussed below.

Opaque petrography

The grain size of hematite-bearing ores is very similar in all three deposits, typically 300-600m, 200-1000m, and 500-1200m, for specular hematite laths, pyrite cubes, and chalcopyrite clots, respectively (figure 1). Approximately 75% of the rock consists of Cu-Fe-S-O minerals at Superior and Bisbee, but Bingham appears to be more sulfide-rich. The relative proportions of hematite and sulfides in massive replacement bodies varies bimodally at Superior and Bisbee. Most hematite-bearing ores in these two deposits contain less than 20-25% sulfide, whereas massive sulfide ores contain only traces of relict hematite. At first glance, this might suggest fundamental differences in the origin of massive sulfide and hematite-bearing ores. Consanguinity of the two ore types, however, is suggested by the ubiquitous shell of hematite-bearing ores around massive sulfide bodies of all scales at Superior, and the common correspondence between the two ore types at Bisbee and Yauricocha.

All of the samples show a progression in time (for a given sample) from hematite to pyrite to chalcopyrite formation. The tips of specular hematite blades are resorbed by pyrite and chalcopyrite (figure 2). At Bisbee, pyrite pseudomorphs thinly-bladed specular hematite. Robust (more equant) varieties of specular hematite, however, which are common at Superior and Bingham, appear to be unfavorable to pseudomorphic replacement. All hematite in the Bingham sample is primary specular hematite, with no evidence for earlier magnetite, as is the case in most Superior ores and some (the distal Dallas-Cole shaft) Bisbee ores. Samples of Bisbee ores from the Gardner and Sacramento shafts (near the center of the district) contain relict magnetite which is cross-cut by and pseudomorphed by specular hematite (footn 4). Specular hematite occurs as minor 10-20 m inclusions in the cores of many pyrite grains in all three deposits (footn 5) (figure 3). Chalcopyrite forms cross-cutting veins in pyrite grains and rims on pyrite grains (figure 4), possibly forming by replacement of pyrite rather than direct precipitation in open fractures. Bornite was not observed in the Bingham sample (but was reported in the core log of the massive sulfide intercept from footages 3134 - 3145). At Superior and Bisbee bornite occurs mostly in massive sulfide ores rather than in hematite-bearing ores. Sphalerite and galena are conspicuously absent from most hematite-bearing ores at Superior and were not detected in any of 10 samples from Bisbee or the three thin sections from the Bingham North Ore Shoot.

"Sieve-textured" pyrite (i.e. pyrite containing 5-15% equant, rounded 25-50 m holes) occurs in hematite-bearing ores of all three districts (figures 2,3, and 4). The holes most likely represent sites where inclusions within the pyrite have since been dissolved. Anhydrite, the product of calcite dissolution by relatively oxidized sulfate-bearing solutions occurs in all three districts as 40 m inclusions within pyrite grains (i.e. predates pyrite deposition). Anhydrite at Bingham contains trace inclusions of red hematite (figure 5). "Sieve-textured" pyrite at both Bingham and Superior contain 35-50 m inclusions of quartz with 2 m platy, high-birefringent mineral inclusions within the quartz (muscovite), suggesting early deposition of sulfides occurred at quartz saturation and below the muscovite-K-feldspar pH buffer (i.e. acidic conditions analogous to D-vein formation in feldspathic rocks).

Rhythmic layering is observed in all three deposits (figure 6). This texture is interpreted as indicating diffusional replacement processes from near-by fractures during metasomatism and is common near contacts between different ore types (e.g. massive sulfide and hematite-bearing ores) and between contacts between ore and wall rocks at Superior. Einaudi (pers. comm., 1997) suggests that rhythmic layering is evidence for direct replacement of carbonate by ore (as opposed to retrograde alteration of earlier-formed skarn). Experience at Superior suggests rhythmic layering is generally restricted to rocks within a meter or two of a contact, probably reflecting limits imposed by the efficiency of diffusional transport of components.

Non-opaque gangue petrography

Late carbonate veins cross-cut all ore mineral assemblages in all three deposits. As at Superior, the veins at Bingham consist of ankerite or siderite growing inward from the walls and late calcite as the center-line (figure 7). Siderite and ankerite locally form a narrow (1-10 cm) margin at the contact between massive hematite ores and dolomite at Superior. Siderite at Yauricocha is associated with hematite at contacts with limestone (Cerro de Pasco Corp., 1970, p. 119). I interpret ankerite/siderite to represent the coolest, final stages of iron metasomatism wherein the pH of the hydrothermal solutions is sufficiently high to stabilize carbonate.

The most striking difference between the hematite-bearing ores of the three districts is the silicate gangue mineralogy. Non-opaque gangue in the North Ore Shoot sample consists primarily of talc (verified by XRD - see figure 8 (big) or small)) possibly replacing fine-grained amphibole (figure 5). The Bingham sample contains only traces of quartz. The carbonate gangue clearly postdates sulfide deposition. Some of the more robust gangue mineral grains may be relict low-iron amphibole, but this identification will require supplementary electron microprobe work. The non-opaque mineralogy of the Superior hematite-bearing ores is dominated by quartz where non-opaque minerals occur (although the massive sulfide ores contain considerable amounts of muscovite, kaolinite, and zunyite). Much of the hematite-bearing ore at Superior consists exclusively of opaque minerals and approximately 5% porosity. At Bisbee, the non-opaque gangue in hematite-dominant ores consists primarily of quartz + calcite + significant (up to 5%) green chlorite tufts (figure 5). These differences in gangue mineralogy probably represent the nature of the protolith. Bisbee ores replaced argillaceous units which may have supplied a local source of aluminum (thus the chlorite). Superior ores replaced clean carbonate and so lacked a local aluminum source, precipitating only quartz. Massive replacement mineralization in the Bingham district was preceded by a well-developed skarn stage and/or Mg-metasomatism that formed local dolomite or calcite-talc assemblages. Non-opaque gangue minerals at Bingham may be replacement products after such skarn minerals, although the textures, especially the local rhythmic layering, suggest replacement of carbonate. Also, talc in the Bingham hematite-bearing ores locally appears to grow off of hematite laths, which might suggest a closer relationship (i.e. coprecipitation during direct replacement of carbonate). Coprecipitation of talc and hematite might reflect higher temperatures of formation than quartz-carbonate stability or lower Ca:Mg ratios in the protolith (i.e. is more analogous to magnesian skarns than calcic skarn environments).

Geochemical interpretation

If the observed non-opaque gangue minerals are in equilibrium with the opaque phases, they record some of the geochemical characteristics of the ore fluids responsible for massive replacement. Variations in the loga_SiO2,aq, loga_Mg++/(a_H+)², and loga_Al+++/(a_H+)³, possibly controlled in part by protolith composition, might be responsible for the differences observed. The phase diagrams shown in figure 9 are modified from those published in Bowers, et al. (1984). The temperature of 400ºC is a little high for these kinds of deposits, but not unreasonably so. Preliminary fluid inclusion work at Superior suggests temperatures of formation circa 300 - 350ºC (footn 6). The phase diagrams are drawn for the system Al₂O₃-MgO-SiO₂-CO₂-H₂O. The X_CO2 value of 0.01 was chosen based on the absence of evidence for high CO₂ fluids in fluid inclusions at Superior and because the ores are composed of a small number of phases, indicating the system was relatively open with respect to volatiles and aqueous components. The diagrams are truncated at the saturation limits (i.e. supersaturated conditions not shown). Each diagram is drawn using two of the components as variables and the third as an axis normal to the page. The phase stability fields are vertical projections of three-dimensional saturation surfaces looking down the third axis onto a plane defined by the two variable components. The mineral names labeled in each field are the minerals that would be stable under the conditions if the solution is saturated with respect to solids. The solution may be undersaturated with respect to solids anywhere on the diagram if the value of the activity ratio of the third axis (the one normal to the page) is too small (footn 7). Thus, the diagrams only identify which minerals would be stable for conditions that are saturated with respect to a solid phase. Each diagram shows what a typical contour for the third dimension would look like, with activity ratio increasing toward corundum, (figure 9a), quartz (figure 9b), and magnesite (figure 9c).

The hematite-bearing ores of the Bingham and Bisbee districts both contain gangue indicating the solutions were saturated with respect to a solid phase of the system Al₂O₃-MgO-SiO₂-CO₂-H₂O and so the conditions of their formation can be plotted readily within the mineral stability fields, although their actual/precise locations within these fields requires additional information (see below). The hematite-bearing ores of the Superior commonly lack non-opaque gangue and so may have been undersaturated with respect to minerals of the system Al₂O₃-MgO-SiO₂-CO₂-H₂O, making estimation of the fluid composition much more speculative. The ores of the Bingham district clearly occur relatively near the porphyry center, whereas the Bisbee ores occur in a more distal location, and the porphyry at Superior remains undiscovered, suggesting it is even more distal. The geochemical positions of the three districts on the three sets of activity diagrams in figure 8 are internally consistent.

The relative geochemical conditions of the three deposits shown on figure 9b (loga_Mg++/(a_H+)² - loga_Al+++/(a_H+)³ are additionally based on the assumption that the protolith composition played some role in affecting the composition of the fluid. Superior and Bingham are plotted at low loga_Al+++/(a_H+)³ relative to Bisbee because the host rocks there are less argillaceous than Bisbee. The deposits are plotted in order of increasing loga_Mg++/(a_H+)² from Bingham and Bisbee to Superior to reflect the relative dolomite content in the protolith (As stated above, the dolomite host rocks in the Superior district are almost 100% dolomite and the Bingham carbonates are described by Atkinson and Einaudi (1978) as silty but neither argillaceous nor dolomitic). The solid symbols used for Bingham and Bisbee indicate the solution was saturated with respect the mineral in whose stability field they plot. Superior is plotted with an open symbol to indicate undersaturation with respect to solids. The relative loga_SiO2,aq was chosen to reflect the relative availability of silica from the wall rocks. Bingham, being hosted by a few carbonate beds within a thick quartzite section, plots near quartz saturation with the highest loga_SiO2,aq. Although Bisbee contains quartz gangue, the P-T assumptions for which the diagrams are drawn make coprecipitation of quartz and chlorite impossible without a_SiO2,aq supersaturation. Coprecipitation of quartz and chlorite is possible at lower P-T conditions, though.

The consistency between activity diagram plots supports, but does not conclusively demonstrate, that the protolith composition may have played a major role in determining the non-opaque gangue assemblages.

Other observations made in the Bingham core shed

While scanning the core in the Bingham core shed, I noticed several other types of ore that are similar to those of the Superior district, and some that are strikingly different.

"Unconsolidated ore" (Bingham mine terminology), consisting of loose, granular, massive pyrite > chalcopyrite that does not hold together well, is similar to massive sulfide zones at Superior which were brecciated by post-mineral faults. Some exploratory drilling at Superior encountered similar "rock", too, that contained 10% talc interstitial to the pyrite (chalcopyrite was present in only trace quantities in these intercepts). I interpret these drill intercepts at Superior to represent pyrite-rich varieties of early, low-temperature magnesian skarn, which pre-date main-stage copper-gold mineralization and which are essentially barren of copper in the positions within the hydrothermal system where the core was drilled (i.e. they are distal and are structurally controlled). Base metal assays demonstrated that much of this kind of loose pyrite-talc rock contained elevated zinc, common to skarn, but not to main-stage mineralization at Superior.

Magnetite, which is fairly common in the massive ores of the North Ore Shoot, is absent in most orebodies at Superior. Magnetite at Superior is restricted to early skarn mineralization, either as rhythmically-banded intergrowths with talc (low temperature magnesian skarn in dolostones), or as a minor constituent of amphibole-bearing skarns (which are generally altered to talc-calcite). Where magnetite does occur in massive ores, it is restricted to portions of the A-bed (lower Martin formation) mantos, associated with talc and sphalerite, which I interpret to represent overprinting of the early (footn 8) skarn by later massive "main-stage" mineralization.

Overall, the ore in the Bingham core was less porous than typical Superior ore possibly due to less post-depositional leaching of soluble minerals (e.g. anhydrite) by late fluids.

The metal grades of the intercepts reported on the Bingham core log are similar to those at Superior. Most hematite-bearing ore contains 2% copper, locally up to 6% Cu (compared to 3 - 6% copper at Superior). Massive sulfide intercepts containing bornite may exceed 10% copper, but most massive sulfide in the core yield assays in the range of 2.5 - 4% cu (typical Superior sulfide ore contains 4 - 6% Cu, with bornite-bearing ore in the 10 - 15% Cu). Gold grades are very similar, generally 0.03 opt Au, but the Bingham core has several intercepts in the 0.1x opt Au range, which is relatively rare at Superior.

The mineralogical zonation of the Bingham intercept (summarized in the figure 10 (big) or (small) ) is strikingly similar to that documented at Superior (figure 11a - by me) and reported at Bisbee (figure 11b - Bryant, 1964, Schwartz and Park, 1932) and Yauricocha ( figure 11c (big) or (small) - Cerro de Pasco Corp. 1970). I have drawn the mineral zones as roughly concordant as suggested by the Utah Copper core log graphic sketches. Mineral zonation at Superior, however, tends to parallel discordant feeder veins rather than bedding, and is only concordant where massive sulfide bodies form mantos within the larger hematite body (figure 11a). Hematite at Superior, Bisbee, and Yauricocha forms a concentric border around massive sulfide bodies, probably due to pH gradients formed by reaction of ore fluid with carbonate rocks. The massive sulfide body in the Bingham core may be cutting an early massive magnetite skarn body which formed along the upper and lower contacts of the limestone bed and so wouldn't necessarily develop a concentric sheath of hematite (figure 10b - my preferred interpretation). The presence of relict garnetite and the predominance of talc gangue supports this hypothesis. Specular-hematite ores near the middle of the bed may have directly replaced carbonate, allowing the formation of rhythmic layering characteristic of such a process. A massive sulfide body fringed by specular hematite ore is suggested by Boutwell (1905) at the Highland Boy mine of the Bingham district (footn 9). Specular hematite directly replacing calcite in the Highland Boy mine is documented in plate 31A (p. 192) and plate 37 (p. 208) of Boutwell's monograph. Unlike the ores at Superior, the Highland Boy massive sulfide ores are reported to have directly replaced carbonate with no intervening hematite-bearing zone (p. 193, 199). The Bingham core log records a central zone of low-grade massive sulfide, which also occurs in the orebodies at Bisbee and Superior. Experience at Superior demonstrates that individual feeder veins have central zones of high-grade bornite ore that grade laterally into lower-grade massive sulfides. The "barren pyrite zone" that occurs in the Superior C-bed orebody along one of the major feeders of the replacement body consists of massive pyrite-quartz (+berthierine?) and appears to post-date the copper-bearing stage, occupying a central location because it overprints an earlier high-grade core (figure 11a).

As at Superior and Bisbee, the copper-bearing and zinc-lead-bearing ores of the Highland Boy mine were mineralogically and temporally distinct (Boutwell, 1905, p. 200). At Superior, some (most?) zinc-bearing ores are related to the early skarn stage which was locally overprinted by main-stage copper-gold mineralization. The uppermost ores in the Magma Vein contain Zn-Pb, but their relationship to other types of mineralization are unknown.

Conclusions

• The similarity of the opaque mineralogy, mineral zoning, ore textures, and metal grades of the Superior, Bingham, Bisbee, and Yauricocha carbonate-hosted massive hematite-bearing copper ores suggests similarities in the ore forming processes associated with late-stage ore fluids in porphyry-related deposits.
• The geochemical control on the mineral zonation in these deposits probably reflects pH gradients between oxidized, acidic ore fluids and carbonate wall rocks (Friehauf, 1996).
• The differences in non-opaque gangue mineralogy may be due to differences in wall rock composition and the distance from porphyry, both reflected in differences in the degree of saturation of the ore fluid with respect to minerals in the system Al₂O₃-MgO-SiO₂-CO₂-H₂O.
• The preservation of anhydrite as inclusions within pyrite grains suggests that these ores may have contained significant anhydrite during early stages of their formation which has since been removed, either by later hydrothermal solutions or by deeply-circulating post-mineralization fluids. The paucity of anhydrite in small (< meter scale) replacement bodies which represent incipient ore body formation suggests that any anhydrite-leaching event either post-dates mineralization or occurred early in the formation of a given replacement body.

Implications for Bingham

Carbonate-hosted massive replacement copper deposits in the other porphyry districts discussed above each account for between 15 and 40 million tons of high grade copper-gold ore. The discovery of similar ores at Bingham with similar copper grades and higher-than-normal gold grades is very encouraging from the perspective of exploration in the greater Bingham area.

Implications for Superior

Similarities of Superior massive carbonate-hosted ores with those of major porphyry copper districts like Bingham supports the model that ores of the Superior district are related to a porphyry copper center. Although porphyry-hosted ore in the district is currently undiscovered, this study should encourage proponents of further exploration in the district, in spite of the cessation of mining in July, 1996 due to exhaustion of ore reserves.

Acknowledgments

This study was funded by Magma Copper Co. (BHP Copper) and NSF Grant EAR 9418301. I'd like to thankfully acknowledge Marco Einaudi, Guillermo Pareja, Patrick Redmond, and Esra Inan for reading early drafts of this manuscript. I'd also like to thank Scott Manske, Alex Paul, Matt Knight, Eric Seedorff (BHP Copper), Geoff Balantyne, Charles Phillips (Kennecott Copper), Mark Reed (University of Oregon), and Mark Barton (University of Arizona) for moral support and guidance.

References

Atkinson, W.W., Jr., and Einaudi, M.T., 1978, Skarn formation and mineralization in the contact aureole at Carr Fork, Bingham, Utah: Econ. Geol., v. 73, p. 1326-1365.

Boutwell, J.M., 1905, Economic geology of the Bingham Mining district, Utah: USGS Prof. Paper 38, 413 p.

Bowers, T.S., Jackson, K.J., , and Helgeson, H.C., 1984, Equilibrium Activity Diagrams for Coexisting Minerals and Aqueous Solutions at Pressures and Temperatures to 5 kb and 600ºC, Springer-Verlag, 397 pp.

Bryant, D.G., 1964, Intrusive breccias and associated ore of the Warren (Bisbee) mining district, Cochise county, Arizona: unpub. Ph.D. dissert., Stanford University, 149 p.

Departamento de Geología de la Cerro de Pasco Corporation, 1970, Geología de los Yacimientos Minerales Operados por la Cerro de Pasco Corporation: Cong. Latinoamericano Geología, 1^st, Lima-Perú, 155 p. (p. 110-129).

Einaudi, M.T., 1982, Description of skarns associated with porphyry copper plutons - southwestern North America, in Titley, S.R., ed., Advances in Geology of the Porphyry Copper Deposits - Southwestern North America, Univ. Ariz. Press, p. 139-184.

Friehauf, K.C., 1996, Reaction paths of high-sulfidation state copper-gold ore fluids in carbonate rocks - a case study at the Superior porphyry-related deposit, AZ: Geol. Soc. America Ann. Meeting, Programs-and-Abstracts, October 30, 1996 ( Magmatic-Hydrothermal-Epithermal Transition and Associated Alteration Mineralization Theme session), vol. 28, no. 7, p. A-403.

Peterson, U., 1965, Regional Geology and major ore deposits of central Perú: Econ. Geol., v. 60, p. 407-474.

Schwartz, G.M., and Park, C.F., Jr., 1932, A microscopic study of ores from the Campbell mine, Bisbee, Arizona: Econ. Geol., vol. 27, p. 39-51.