We report rhenium–osmium (Re–Os) isotope data and age information for bornite, chalcocite, and chalcopyrite from the Spar Lake and Rock Creek copper–silver (Cu–Ag) sediment-hosted deposits in the Spar Lake district, Mesoproterozoic Belt–Purcell Basin, Montana, USA. The Re–Os geochronometers in chalcocite and chalcopyrite do not record plausible mineralization ages at the Spar Lake deposit, which is currently interpreted to have formed at ca. 1410 million years ago (Ma) from prior work. However, a Re–Os isochron date of bornite (253.2 ± 26.9 Ma, n = 6) at Rock Creek with an extremely high initial 187Os/188Os ratio of 19.1 ± 0.2 can be interpreted as resetting of Re–Os in bornite that originally recorded a Mesoproterozoic timing of Cu–Ag mineralization in the Spar Lake district. The new Re–Os age of bornite at Permian–Triassic boundary time is compatible with a resetting at the time of building of the Sonoma Foreland Basin that resulted from the loading of an accretionary prism (i.e., Golconda Allochthon) onto the North American Plate. Our ca. 253 Ma bornite age does not constrain whether Cu was introduced to Spar Lake at this time or simply redistributed, but our findings open the possibility that Cu–Ag mineralization in the Spar Lake district might have taken place not only in the Mesoproterozoic, but also in addition, at Permian–Triassic boundary time. Indeed, numerous Cu–Ag, vanadium–uranium (V–U) sandstone-hosted and black shale-hosted deposits of assumed Permian to Triassic age are documented in the Midcontinent copper belt from western Texas to southern Idaho, including (1) Cu–Ag sediment-hosted deposits in western Colorado, southeastern Utah, and northern Arizona and (2) the Montpelier and Lake Alice sandstone-hosted Cu–Ag deposits in southeastern Idaho and western Wyoming, respectively.

The Mesoproterozoic Belt–Purcell Basin (BPB) in the USA and Canada is an exceptional archive of Mesoproterozoic Earth (Lydon 2007), in addition to it being an important metallogenic province with significant endowments of copper (Cu), cobalt (Co), and zinc–lead–silver (Zn–Pb–Ag). Building on pre-existing ore genesis models, the direct radiometric dating of sulfide minerals using the rhenium–osmium (Re–Os) geochronometer has identified the key geodynamic settings during the evolution of the basin that governed the mineralizing processes for each of the sediment-hosted Cu–Co and fault-hosted Zn–Pb–Ag ore deposits of Mesoproterozoic age (Coeur d'Alene, Leach et al. 1998; Arkadakskiy et al. 2009; Idaho Cobalt Belt, Slack 2012; Saintilan et al. 2017a; Sheep Creek deposit, Saintilan et al. 2021). These geochronology studies have identified the key role of reactivated tectonic features in the Archean to Paleoproterozoic basement in the focusing of fluid flow for sediment-hosted mineralization (Saintilan et al. 2017a, 2021). In addition, the initial 187Os/188Os isotope ratio has constrained the source of metals in those ore deposits either derived from crustal Archean to Paleoproterozoic basement lithologies (Arkadakskiy et al. 2009; Saintilan et al. 2017a) or magmatic, mantle-derived sources at the time of extensional tectonics and enhanced thermal gradient in the BPB (Saintilan et al. 2021). Such studies, together with ongoing mineral exploration, are testament to the open potential of mineral resources of this metallogenically important sedimentary basin.

A Cu-mineralized district centered on the Spar Lake ore deposit (production 1981–2012: 39.2 Mt at 0.62% Cu and 48 g/t Ag) and two key Cu–Ag prospects (e.g., Rock Creek and Montanore with inferred resources of 88.5 Mt at 0.67% Cu and 52 g/t Ag, and, 101.8 Mt at 0.68% Cu and 56 g/t Ag, respectively; Hecla Mining Company, Mineral Reserves and Mineral Resources as of 31 December 2016, News Release on 22 February 2017) are the major known deposits of many Cu–Ag-mineralized occurrences in western Montana, USA (Fig. 1). The Spar Lake deposit has been interpreted to form across a mixing zone between trapped natural gas in the host siliciclastic rocks (ca. 1470 million years-old Revett Formation, Ravalli Group) and metal-bearing brines (Hayes et al. 2012), but it remains one major mineral deposit in the basin that lacks direct Re–Os ages of Cu–sulfides to bring temporal constraints on the geodynamic context of ore genesis. At the present state of knowledge, a Mesoproterozoic uranium–lead (U–Pb) SHRIMP age 1409 ± 8 million years ago (Ma) for hydrothermal xenotime interpreted to be associated with Cu–Ag mineralization is commonly accepted as constraining the timing of ore formation (Aleinikoff et al. 2012). Elsewhere in the BPB, Cu mineralization hosted by sandstone of the Revett Formation has been interpreted as ca. 1110–1000  Ma (Grenvillian) epigenetic Cu mineralization, 440 million years after deposition of the host strata, based on Re–Os dating of trace quantities of molybdenite associated with cupriferous sulfides (Perelló et al. 2021).

The Spar Lake deposit has a remarkable silver endowment in the form of native silver, argentiferous bornite, acanthite, and its metamorphic paramorph argentite (Emmons et al. 1926; Hayes and Einaudi 1986). Regionally, high silver deportment is typically associated with the younger Midcontinent copper–silver–vanadium–uranium (Cu–Ag–V–U) belt of intepreted Permian–Triassic age, which stretches from western Texas, through Oklahoma, New Mexico, Colorado, and Utah but also extends into southeastern Idaho and western Wyoming (Fig. 1 and its inset). Such Cu–Ag–V–U mineralization is found in sandstone and black shale where chalcocite replaced organic debris, e.g., from wood logs to spores, prior to compaction and lithification (Fischer 1937; Love and Antweiler 1973; Ripley et al. 1980; Huyck and Chorey 1991; Fay 2000). In the absence of geochronology directly from ore minerals, one hypothesis could consider that the Spar Lake Cu district might lie in the northwest–southeast strike extension of this Midcontinent, Permian–Triassic, Kupferschiefer-like Cu–Ag–V–U belt (Fig. 1) despite its commonly held view of being Mesoproterozoic in origin. In this study, we investigate Cu–sulfides of the Spar Lake district with the Re–Os geochronometer, to test whether the large Cu district around the Spar Lake deposit (1) solely results from mineralizing processes in the Mesoproterozoic, (2) bears a metamorphic overprint on ore distribution, or (3) whether its significant endowment and size could result from additional mineralization at the time of Permian–Triassic Cu–Ag–V–U sediment-hosted mineralization in the Midcontinent Belt.

2.1 Petrography and preparation of monophasic sulfide mineral separates

Exploitation of the Troy mine at Spar Lake started in the 1970s. The mine was operated sporadically until 2012. Reclamation is still ongoing and access to samples from the Spar Lake and Rock Creek deposits (Figs. 2A and 2B) was granted at the core archive of the Hecla Mining Company. For sulfide Re–Os geochronology work, a subset of four samples with disseminated or vein-type Cu mineralization comprises one sample from the Spar Lake deposit (UA-20-SL) and three samples from the Rock Creek deposit (Fig. 2C; UA-17-SL, UA-22-SL, and UA-25-SL). The paragenetic association of the various sulfides was established based on macroscopic descriptions and microscopic petrographic observations of polished thin sections using transmitted and reflected light microscopy (Fig. 3). The paragenetic relationships permitted to constrain the workflow for optimum mineral separation of individual sulfides into monophasic mineral separates according to the protocol presented in Saintilan et al. (2020). This workflow using 70–200 mesh size fractions combines the stepwise use of a Frantz Isodynamic Separator and additional treatment by heavy liquid separation of the magnetic (M) and non-magnetic (NM) fractions obtained at a given current. Bornite was isolated in the M0.72 fractions (i.e., fraction of magnetic material at an applied current of 0.72 A) before purification from remaining gangue mineral by heavy liquid separation. Similarly, chalcopyrite and chalcocite mineral separates were produced by using 0.9 and 1.0 A current on the bulk fraction or on the NM0.72 fraction of sample UA-20-SL, and were collected in the M0.9 and M1.0 fractions, respectively. Those fractions underwent heavy liquid separation to remove remaining gangue minerals. Quality control of the final mineral separates was conducted according to the protocol by Saintilan et al. (2020). Aliquots of each magnetic fraction were embedded in epoxy and the mounts studied by reflected light microscopy and by scanning electron microscopy (SEM; Figs. 3I3L) using a Zeiss Sigma 300 field emission SEM operated in backscattered electron mode (beam conditions 15 kV). In addition, a Bruker energy-dispersive X-ray spectroscopy system with dual silicon-drift detectors, each having an area of 60 mm2 and a resolution of 123 eV, was used for single-spot analysis to further control the mineralogy of each mineral separate.

2.2 Rhenium–osmium radiogenic isotope geochemistry

For each analysis, between 40 and 240 mg of chalcopyrite, chalcocite, or bornite mineral separates were weighed and transferred into thick-walled borosilicate Carius tubes (Shirey and Walker 1995). Each sulfide aliquot was dissolved in inverse Aqua Regia (∼3 mL of 11N HCl and ∼6 mL 16N HNO3) with a known amount of 185Re + 190Os spike solution at 210 °C for 24 h (Laboratory of Rhenium-Osmium Isotope Geochemistry and Geochronology, Isotope Geochemistry and Cosmochemistry Group, Institute of Geochemistry and Petrology, ETH Zürich, and Canadian Centre for Isotopic Microanalysis, University of Alberta, Edmonton, Canada). The laboratory protocol used in the present work is described in full details in Selby and Creaser (2001), Selby et al. (2009), Hnatyshin et al. (2016), and Li et al. (2017). In brief, Os was isolated and purified from the inverse Aqua Regia solution by chloroform (CHCl3)–hydrobromic acid solvent extraction at room temperature, and, by microdistillation (Roy-Barman and Allègre 1995; Cohen and Waters 1996; Shen et al. 1996; Birck et al. 1997; Selby and Creaser 2001). The Re was isolated using an acetone–sodium hydroxide (acetone–NaOH) step (Matthews and Riley 1970; Bozkhov et al. 1985; Li et al. 2009), followed by HCl–HNO3-based anion chromatography (Morgan et al. 1991; Cumming et al. 2013). The Re and Os isotopic compositions were determined by negative thermal ionization mass spectrometry (N-TIMS) using a Thermo Scientific Triton mass spectrometer. Rhenium and Os were loaded onto outgassed Ni and Pt filaments, respectively. Rhenium was measured as ReO4 in static mode on Faraday collectors, whereas Os was measured as OsO3 in peak-hopping mode on a single electron multiplier (Creaser et al. 1991; Völkening et al. 1991). Measurement quality was monitored by repeated measurements of in-house Re (125 pg aliquot – 185Re/187Re = 0.59860 ± 0.00051, 2σ, n = 22) and Os (Durham Romil Os Standard, DROsS; Nowell et al. 2008; 50 pg aliquot – 187Os/188Os = 0.16092 ± 0.00069, 2σ, n = 12) standard solutions. Total procedural blank was 3.0 ± 0.1 pg Re, 30 ± 2 fg Os with a blank 187Os/188Os isotope composition using inverse Aqua Regia of 1.12 ± 0.29 (2σ, n = 3) at Institute of Geochemistry and Petrology, ETH Zürich. Total procedural blank was 2.3 ± 0.5 pg Re, 80 ± 50 fg Os with a blank 187Os/188Os isotope composition using inverse Aqua Regia of 0.80 ± 0.06 (2σ, n = 3) at the University of Alberta. The analytical uncertainties result from full error propagation of standard measurements, mass spectrometry analyses, and blanks.

3.1 Textural and paragenetic relationships between sulfides and gangue minerals in the host rock

Sample UA-20-SL comes from the Middle North orebody in the Troy mine at Spar Lake and is characterized by fine-grained sandstone comprising 2 cm long mottles made of medium-grained purple bornite with a halo of chalcopyrite (Fig. 2C). Bornite and chalcopyrite both occur as associated with coarse-grained hydrothermal quartz. Chalcopyrite does not replace bornite. Bands of microcrystalline “sericite”, i.e., hydrothermal white mica, occur in between pods of chalcopyrite or bornite accompanied by coarse-grained hydrothermal quartz (Figs. 3A and 3B). These sericite bands are younger than Cu–sulfides and hydrothermal quartz.

Samples UA-25-SL (Moran ore zone, C-mineralized horizon) and UA-17-SL (Chicago ore zone, A- and B-mineralized horizons) are representative of Ag-bearing Cu–sulfide mineralization in clean silicified white sandstone with disseminated spots of fine-grained chalcocite (sample UA-17-SL), or grey sandstone with fine-grained dark blue bornite (sample UA-25-SL) at Rock Creek (Fig. 2C). In sample UA-25-SL, bornite is associated with authigenic coarse-grained hydrothermal quartz and albite (Figs. 3C3G). A Cu–Ag–sulfide is intergrown with bornite or included in albite. In these albite- and quartz-bearing mineralized zones (Figs. 3D3G), bornite is spatially associated with manganese-rich (Mn-rich) calcite, and <100 µm wide, square-shaped pores filled with Mn-rich calcite rimmed by brown laths or rectangles of biotite with rutile inclusions (Fig. 3D). There are two types of occurrences of white micas: (1) grain-rimming, fine-grained lath of white mica and (2) bands of medium-grained, hydrothermal sericite 400–500 µm in width (Fig. 3C). These bands of sericite are not directly spatially associated with bornite. Conversely, in this sample, chalcocite is rare as local replacement of bornite (Fig. 3E). On the contrary, sample UA-17-SL comprises mostly chalcocite with rare inclusions of bornite (Fig. 3H). Bornite is also found (<1–2 vol.%) interstitial to detrital quartz and hydrothermal quartz overgrowths. Sample UA-22-SL, which comes from the Chicago ore zone at the contact between the C- and D-mineralized beds at Spar Lake, is characterized by silicified beige sandstone with coarse-grained chalcopyrite stringers (Fig. 2C).

3.2 Re–Os isotope data

Rhenium and total Os (Ostotal) concentrations in the analyzed sulfides are moderate and high, respectively (0.75–47.9 ng g–1 Re and 57–2540 pg g–1 Os; Table 1). Common Os contents in chalcopyrite (UA-22-SL) and chalcocite (UA-17-SL) at Spar Lake, estimated using abundances in 192Os, represent a negligible fraction of Ostotal (1–42 pg g–1 192Os). Neither the chalcopyrite data nor the chalcocite data are positively correlated in 187Os/188Os versus 187Re/188Os isochron space (Fig. 4—inset). In sample UA-20-SL at Spar Lake, preliminary full procedural Re–Os analysis of chalcopyrite using the 185Re + 190Os spike indicated that chalcopyrite is virtually bereft of common Os and the total Os content corresponds to radiogenic 187Os*. Thus, three additional chalcopyrite aliquots were processed by using a 185Re + 188Os + 190Os spike to constrain the total Re and 187Os* budget to calculate model ages for each aliquot. These model ages are irreproducible and in excess of the age of the host rocks. Similarly, model age estimates for chalcopyrite (UA-22-SL) and chalcocite (UA-17-SL) return impossibly old model ages largely in excess of the age of the host rocks (Table 1).

For bornite at Rock Creek (UA-25-SL), the moderately high 187Re/188Os values (329–876) are positively correlated with highly radiogenic 187Os/188Os ratios (20.6–22.7). The Re–Os data for bornite yield a Model 1 date of 253.2 ± 26.9 (27.2) Ma (bracketed uncertainty includes the 0.31% uncertainty in the decay constant of 187Re after Smoliar et al. (1996); n = 6; mean square of weighted deviates (MSWD) = 1.5; 2σ; initial 187Os/188Os ratio—Osi, 19.1 ± 0.2; Fig. 4). The MSWD of 1.5 indicates that the scatter between the data points is solely caused by the analytical uncertainty (Vermeesch 2018). The relatively high uncertainty of the isochron regression is related to the limit spread of 187Re/188Os values in between the various aliquots.

Two lines of evidence inherent to samples UA-20-SL (Spar Lake) and UA-25-SL (Rock Creek) may suggest that bornite and chalcopyrite are part of the same mineralizing event without chalcopyrite proceeding from replacement of bornite: (1) systematic association of these Cu–sulfides with coarse-grained hydrothermal quartz, (2) a common feature of bands of “sericitization” that postdate and physically surround zones of bornite, hydrothermal quartz, and Mn-rich calcite mineralization on the one hand, and chalcopyrite–hydrothermal quartz mineralization on the other hand (Figs. 3A3C). Manganese-rich calcite is a characteristic by-product of oxidation of organic compounds by MnO2 during diagenesis, while sulfate is being reduced into hydrogen sulfide (H2S) during thermochemical sulfate reduction (Okita and Shanks III 1992). This process is advocated to explain the origin of H2S for Cu–sulfide precipitation in the Spar Lake deposit resulting from the interaction between metal-bearing brines and a sour gas trap in the Revett Formation paleoreservoir (Hayes et al. 2012). At Spar Lake, “sericitization”, which is compatible with acidic conditions in hydrothermal fluids, would be at disequilibrium with the widespread occurrence of bornite with subhedral to anhedral Mn-rich calcite (Figs. 3C3F). In fact, “sericitization” was described previously in the Spar Lake district and interpreted as being related to events postdating the ca. 1410  Ma epigenetic Cu–Ag mineralization related to sour gas traps at Spar Lake (Aleinikoff et al. 2012; Hayes et al. 2012). It was suggested that, in amongst possibilities for younger events, this “sericitization” could be related to fluid flow and mineralizing processes related to the giant hydrothermal cell in the Coeur d'Alene district, the closest vein of which is only 5 km away from Mapple Cliffs in the western Spar Lake district (Hayes et al. 2012 quoting pers. comm. by J. Mauk, 2003; J. Mauk, pers. comm. 2023, about the dimension of the footprint of hydrothermal alteration in the Coeur d'Alene district). An age at ca. 1293  Ma (arsenopyrite Re–Os model age: 1293 ± 9  Ma, 2σ, including uncertainty in the decay constant of 187Re; Arkadakskiy et al. 2009) has been proposed for the Coeur d'Alene mineralization. Therefore, this evidence collectively suggests that bornite and chalcopyrite are part of the same Mesoproterozoic mineralizing event. The partial replacement of bornite by chalcocite (sample UA-17-SL; Fig. 3H) at Rock Creek cannot be tied to any of these potential mineralizing events considering the available paragenetic relationships alone.

4.1 Preservation and disturbance of Re–Os systematics in Cu–sulfides

Our Re–Os study documents that the Re–Os geochronometers in chalcocite and chalcopyrite in the Rock Creek and Spar Lake deposits do not record plausible mineralization ages. The irreproducible model ages in excess of the age of the host rocks for chalcopyrite in sample UA-SL-20 (Fig. 3I) suggests either variable loss of Re or uptake of radiogenic Os lost from other minerals (e.g., Stein et al. 2003). The scatter of chalcopyrite and chalcocite data from samples UA-SL-22 (Fig. 3J) and UA-SL-17 (Fig. 3K), relative to a 1410  Ma reference isochron representing the interpreted mineralization age (Aleinikoff et al. 2012), also suggests loss of Re or uptake of radiogenic Os. Thus, for both chalcocite and chalcopyrite, the Re–Os system points to an event(s) of disturbance/open-system behavior at some time after formation. A likely candidate for the disturbance of the Re–Os systematics in chalcopyrite of assumed Mesoproterozoic age is the “sericitization” event that may be related to the 1293 -Myr-old Coeur d'Alene mineralization. This assumption could imply that chalcocite (and, by corollary, the partial replacement of bornite by chalcocite) also predates the timing of this hydrothermal event related to the neighboring Coeur d'Alene hydrothermal system. This work confirms the fact that the Re–Os geochronometer in chalcopyrite (1) is relatively prone to disturbance by post-mineralization tectono-thermal events and (2) might not record the timing of such disturbance (Stein et al. 2003; Saintilan et al. 2021). Additional work is required to determine whether the Re–Os geochronometer in chalcocite (e.g., Re–Os dating of chalcocite ± bornite at the Lince–Estefanía deposit, Chile; Tristá-Aguilera et al. 2006) is generally prone to disturbance, or whether the present study is an isolated case.

Conversely, the bornite Re–Os data from Rock Creek (UA-25-SL; Fig. 3L) display well-correlated behavior on a Re–Os diagram (Fig. 3) but yield a much younger age than the presumed Mesoproterozoic mineralization age. Importantly, the Re–Os data array intersects a reference 1410-Myr-old isochron (see inset in Fig. 4) but display a shallow slope indicative of the ca. 253  Ma isochron age they record. We discuss two potential interpretations of this ca. 253  Ma bornite age at Rock Creek:

  • (1) Assuming that the Rock Creek deposit was formed at the same time as the Spar Lake deposit (Hayes et al. 2012; ca. 1410  Ma), our Re–Os isochron date of bornite (253.2 ± 26.9  Ma) at Rock Creek with an extremely high Osi-bornite at 19.1 ± 0.2 is then interpreted to record resetting of the Re -Os system in bornite in connection with a geodynamic/thermal event that took place at Permian–Triassic boundary time. The extremely high Osi-bornite (i.e., common Os available in the environment of mineral precipitation) of the 253  Ma regression reflects (i) likely high Osi at time of mineralization at ca. 1410  Ma as a result of metal derivation from Paleoproterozoic to Archean basement rocks as identified in the Idaho Cobalt Belt (Osi-cobaltite of 4.7 ± 2.2 at ca. 1349 Ma, Saintilan et al. 2017a) and the Coeur d'Alene district (Osi-arsenopyrite of 0.9 ± 1.2 at ca. 1293  xMa; Arkadakskiy et al. 2009) coupled with (ii) the decay of 187Re and accumulation of radiogenic 187Os* in bornite between 1410 and 253 Ma until resetting of the Re–Os clock in bornite at the late Permian.

  • (2) Alternatively, Rock Creek bornite is Permian in age and may be a newly formed Cu mineralization at this time. In this case, common Os recording a high Osi-bornite would have been mobilized from an extremely radiogenic source that could either be the same Paleoproterozoic to Archean basement rocks as for the Mesoproterozoic Spar Lake deposit or from the Spar Lake deposit itself. In the following paragraph, these two possible age interpretations are discussed further.

The closure temperature of the Re–Os geochronometer in bornite is empirically estimated to be higher than 360 °C (Saintilan et al. 2023). This closure temperature is possibly as high as the closure temperature of the Re–Os geochronometer in pyrite at or above 400 °C (Brenan et al. 2000; Morelli et al. 2010; Saintilan et al. 2017b). To explore whether bornite mineralization could have undergone open system behavior of its Re–Os systematics after an assumed Mesoproterozoic timing of mineralization, we look at the history of temperature conditions that the host rocks in the Revett Formation experienced post-depositionally:

  • (1) During the first episode of the poorly known East Kootenay orogeny ca. 1350–1300 Ma, which coincided with the termination of the Belt–Purcell sedimentation, the Belt–Purcell Supergroup experienced compression triggering large-scale folds, regional burial metamorphism, and granitic intrusion (McMechan and Price 1982; Doughty and Chamberlain 1996). Gradual heating of Belt–Purcell sedimentary rocks up to the biotite isograd of greenschist burial metamorphism started at ca. 1370  Ma with intrusion of felsic plutons (Doughty and Chamberlain 1996). Thermal conditions for biotite production at greenschist–facies conditions are at about 400 °C (Ferry 1984). Maximum burial diagenesis and metamorphic temperatures at greenschist–facies conditions were approximately 425 °C at the Spar Lake deposit (Hayes 1990) and 375 °C in the Mount Fisher area (150 km to the northwest of Spar Lake in southern Alberta, Canada; McMechan and Price 1982). During burial metamorphism, chlorite consumption and dehydration–decarbonation are the intensive variables (Mather 1970; de Béthune 1976; Ferry 1984) that would have led to production of biotite, without penetrative deformation, in localized chlorite–chalcocite zones of the calcite-cemented sandstone paleoreservoir hosting the Spar Lake deposit (Fig. 3D; Hayes and Einaudi 1986; Hayes et al. 2012). These biotite-rich horizons bear argentite rather than acanthite (Hayes and Einaudi 1986). Argentite is the paramorph of acanthite that is stable above 179 °C (Emmons et al. 1926; Hayes and Einaudi 1986) and the ore-forming temperature conditions at ca. 170 °C (Hayes 1990). Therefore, a link between biotite formation and silver sulfide is attested in specific biotite-rich horizons. This association signals the redistribution/recrystallization of Ag mineralization in connection with the East Kootenay orogeny, up to ca. 40 million years after the accepted age of epigenetic Cu–Ag mineralization at Spar Lake ca. 1410 Ma. The effects of the East Kootenay orogeny on the Cu–sulfides studied here could include the disturbed Re–Os behavior observed in chalcopyrite and chalcocite, but for bornite, the Re–Os data show no record of thermal or geodynamic events related to the East Kootenay orogeny, only an event in Permian time.

  • (2) Orebody magnetite associated with Cu sulfides in the host Revett Formation at the Spar Lake and Rock Creek deposits has mineral chemistry typical of burial metamorphic magnetite. At Spar Lake, paragenetic relationships of magnetite with Cu–sulfide inclusions and oxygen isotope data on magnetite are permissive of the interpretation that magnetite formed through recrystallization of Cu–sulfides during burial metamorphism (Nadoll et al. 2012). Additional geochemical evidence using oxygen isotope in coexisting and probably re-equilibrated magnetite–calcite pairs in the host rock document burial temperature conditions at 437–511 °C (Nadoll et al. 2012). Yet, absolute time stamps for the thermal conditions of this burial metamorphism that drove these reactions with magnetite are lacking.

In conclusion, it is likely that burial metamorphism at the biotite isograd during the East Kootenay orogeny led to redistribution/recrystallization of Ag mineralization in biotite-rich horizons of the host Revett Formation. Biotite-grade greenschist–facies metamorphism related to the ca. 1350–1300  Ma East Kootenay Orogeny and “sericitization” linked to acidic hydrothermal fluid flow responsible for the neighboring Coeur d'Alene mineralization at ca. 1293  Ma may have caused disturbance of the Re–Os systematics in chalcopyrite and chalcocite. Yet, neither this localized metamorphic overprint at temperatures of ca. 375–425 °C nor the “sericitization” event are recorded by the Re -Os systematics of bornite at Rock Creek in an assumed pre-existing Mesoproterozoic ore. Therefore, it is likely that the Re–Os 253  Ma isochron age of bornite at Rock Creek reflects the resetting of the Re–Os clock in bornite above its closure temperature (currently empirically constrained as being possibly higher than 425 °C) during a subsequent, younger burial/thermal event.

4.2 Resetting of the bornite Re–Os geochronometer at the Permian–Triassic boundary during the Sonoma Orogeny

We propose that our interpreted resetting of the bornite Re–Os chronometer at Permian–Triassic boundary time relates to the formation of the Sonoma Foreland Basin (SFB—Fig. 1) from the Permian–Triassic boundary (ca. 252.0 Ma) until the end-Smithian (Early Triassic; ca. 250.7 Ma; Dickinson 2006, 2013; Blakey 2008; Ingersoll 2008). The SFB resulted from the loading of an accretionary prism (i.e., Golconda Allochthon) onto the North American Plate (Caravaca et al. 2018 and references therein; Balgord et al. 2021). The lithospheric heterogeneity of the basement steered differential sedimentary accumulation in the SFB. Accumulation of Paleozoic marine sedimentary series reached about 6 km in the northern part of the SFB (e.g., present-day central Idaho and northern Utah), whereas marine and terrigenous sediments reached at most ca. 650 m in the southern part of the SFB (e.g., southwestern Utah; Peterson et al. 1977; Yonkee and Weil 2015; Rowley et al. 2005; Caravaca et al. 2018). In this geometry, the role of the Paleoproterozoic to Mesoproterozoic Farmington Terrane (FT, Fig. 1) was crucial in controlling differential sedimentary accumulation in the SFB. During the Sonoma Orogeny, the FT behaved as a “thermally attenuated weak lithosphere” due to its lesser rigidity in the patchwork of surrounding Archean to Paleoproterozoic terranes in the northwestern USA (Caravaca et al. 2018). The extension of the SFB in northern Idaho remains undocumented (Caravaca et al. 2018; Balgord et al. 2021, their fig. 18A). Yet, given the position of the Spar Lake district in the direct strike extension of the FT in the basement of the SFB, it is possible that a narrow but deep extension of the SFB existed where the Spar Lake district was positioned in the basement (Fig. 1). In such a narrow foreland with a high-subsidence rate (e.g., 500 m/Myr documented from the Permian-Triassic boundary to the end of Induan-start of Olenekian: ca. 1.3 Myr interval; Caravaca et al. 2018), the Spar Lake district could be plausibly positioned in the basement to this basin close to the wedge-top to foredeep transition (Fig. 1). In this geodynamic setting, Mesozoic thickening is attested and caused resetting with partial reheating with argon loss in regionally distributed micas and hornblende (e.g., 214 Ma age of muscovite and biotite in the southern Purcell Anticlinorium; McMechan and Price 1982; Rioseco et al. 2022). Thus, thermal conditions would have reached, if not exceeded, the closure temperature of the Re–Os geochronometer in bornite (>425° C?). Then, upon cooling, the Re–Os clock in bornite recorded a resetting in the late Permian ca. 253 Ma, close to the Permian–Triassic boundary.

4.3 Bornite in the Spar Lake district: solely Mesoproterozoic mineralization or additional contribution at the time of formation of the Midcontinent copper belt?

The timing of resetting at ca. 253  Ma of the Re–Os geochronometer in bornite mineralization of assumed original Mesoproterozoic age at Rock Creek in the Spar Lake district also opens the intriguing possibility of a connection with the numerous Cu–Ag, Cu–U–V sandstone-hosted and black shale-hosted deposits of assumed Permian to Triassic age that are documented in the Midcontinent copper belt from western Texas to southern Idaho through Oklahoma, Kansas, New Mexico, Colorado, and Utah (Fig. 1 and its inset map; Fischer 1937; Love and Antweiler 1973). The Cu–Ag and V–U sediment-hosted deposits in western Colorado, southeastern Utah, and northern Arizona are hosted in Triassic rocks that would have been positioned in a forebulge setting of the SFB from the Permian–Triassic boundary to the Lower Triassic (Fig. 1). The Montpelier and Lake Alice sandstone-hosted Cu–Ag deposits in southeastern Idaho and western Wyoming, respectively, are hosted in Triassic rocks deposited in the SFB (Fig. 1; Fischer 1937; Love and Antweiler 1973). Thus, we speculate that

  • (1) there is an open potential for Cu–Ag sediment-hosted deposits where Permian to Triassic sedimentary rocks might be identified in a northwest–southeast-striking band stretching from southeastern Idaho–western Wyoming to northern Idaho; and

  • (2) although textural evidence and Re–Os isotope geochemistry data favor the interpretation of isotopic resetting of Mesoproterozoic bornite mineralization at ca. 253  Ma at Rock Creek, we cannot rule out the possibility that Cu mineralization in the Spar Lake district formed, at least in part, in the Permian to Triassic as recorded by Re–Os in bornite. If our current understanding of the bornite Re–Os geochronometer is that it requires temperatures in the excess of 400–425 °C to reset, then this process should also fully reset regional 40Ar/39Ar mica ages, but this is not observed. In the southern Purcell Anticlinorium, 40Ar/39Ar mica ages range between ca. 1318  Ma and ca. 265  Ma (McMechan and Price 1982; Rioseco et al. 2022). Therefore, further regional geochronology in the Spar Lake district and the Midcontinent copper belt is critically needed to constrain the potential timing(s) of mineralization and the extent of hydrothermal fluid flow leading to the interpreted Permian to Triassic Cu–Ag sediment-hosted mineralization in this transcontinental metallogenic province. In either case, the Permian Re–Os age of bornite is testament to the effect of geological events—either tectonic or metallogenic—not previously known from Montana.

This work was supported by an Early Postdoc.Mobility Grant (P2GEP2_162075) and an Ambizione Fellowship (PZ00P2_180133) of the Swiss National Science Foundation awarded to NJS. RAC acknowledges support from a NSERC Discovery Grant, Canada. NJS and RAC are indebted to Steve Petroni (former General Manager, Exploration for Hecla Mining Company (retired)) for overall scientific and logistical support to the study, James J. Myers, senior geologist at Hecla Mining Company, for the selection of drill hole sample intervals, and Dustin Howell, geotechnician, for assistance with sample collection. Dr. Dean McDonald, senior vice president, exploration, Keith Blair, chief resource geologist, and Brandi Rollins, GIS technician, Hecla Mining Company, are acknowledged for providing mineral resource information and the geological cross-section for the Rock Creek and Montanore prospect. Permission to publish this work was given by Jonathan Moore, Exploration Manager, Hecla Northwest US Exploration. We acknowledge the pertinent comments and suggestions by the Associate Editor and two anonymous reviewers. We thank Editor-in-Chief Sally Pehrsson for a smooth and efficient overall editorial handling.

All data are available in the figures and table of this paper.

Conceptualization: NJS, RAC

Data curation: NJS, RAC

Formal analysis: NJS, RAC

Funding acquisition: NJS, RAC

Investigation: NJS

Methodology: NJS, RAC

Project administration: NJS

Validation: NJS, RAC

Visualization: NJS

Writing – original draft: NJS

Writing – review & editing: NJS, RAC

This work was supported by an Early Postdoc.Mobility Grant (P2GEP2_162075) and an Ambizione Fellowship (PZ00P2_180133) of the Swiss National Science Foundation awarded to NJS. RAC acknowledges support from a NSERC Discovery Grant, Canada.

This work is licensed under a Creative Commons Attribution 4.0 International License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author(s) and source are credited.