Alderton Fallick The Nature and Genesis of Gold-Silver-Tellurium ...

Economic Geology Vol. 95, 2000, pp. 495–516

The Nature and Genesis of Gold-Silver-Tellurium Mineralization in the Metaliferi Mountains of Western Romania DAVID H. M. ALDERTON† Department of Geology, Royal Holloway, London University, Egham, Surrey TW20 0EX, United Kingdom AND

ANTHONY E. FALLICK

Isotope Geosciences Unit, Scottish Universities Research and Reactor Centre, East Kilbride, Glasgow G75 0QF, Scotland, United Kingdom

Abstract Precious metal (Au, Ag) and base metal (Pb, Zn) deposits in the Metaliferi Mountains of western Romania occur in steeply dipping quartz-Ca/Mn carbonate veins, which are hosted by Miocene andesitic stocks and lava flows, and surrounding sedimentary rocks. The deposits consist predominantly of sulfides (pyrite, chalcopyrite, sphalerite, galena), sulfosalts of As and Sb, and a diverse range of Au-Ag tellurides. The igneous host rocks have undergone mild, pervasive propylitic alteration, whereas immediately adjacent to the veins the wall-rock alteration assemblages consist of quartz, sericite, K feldspar, calcite, and pyrite. Fluid inclusion, stable isotope, and thermodynamic data suggest that the majority of the mineralization and hydrothermal alteration in these deposits was caused by low-salinity (0–5 wt % NaCl equiv), medium-temperature (200°–300°C), near-neutral (pH = 5–6) fluids, which underwent occasional boiling. The fluid inclusion and stable isotope data support a model in which a metal-bearing, magmatic fluid was exsolved from a crystallizing calc-alkaline melt and ascended to higher levels in the crust, undergoing some isotopic exchange with surrounding sedimentary rocks but limited mixing with ground waters. Although the deposits in this part of the Romanian Carpathians exhibit many of the geologic characteristics of classic lowsulfidation, volcanic-hosted, Au-Ag, epithermal deposits, they seem to have formed from essentially magmatic waters, and there is little evidence for the incorporation of meteoric fluids into the hydrothermal system.

Introduction ROMANIA is particularly well endowed with metallic mineral deposits and the country has a very long history of mining. Gold has been mined in Romania since pre-Roman times, and for much of the time since then, this country has been the most important producer of precious metals in Europe. The richest concentration of precious metals has been located in the Metaliferi Mountains, in the southern part of the Apuseni Mountains, of western Romania, an area also known as the “gold quadrilateral” or “gold quadrangle” (Ianovici and Borcos, 1982; Udubasa et al., 1992; Figs. 1 and 2). Exact production figures are impossible to establish (see below) but according to some suggestions this region (approx 600 km2 in size) could have produced more than 1,000 metric tons (t) of gold and ranks as the richest epithermal, gold-producing region of equivalent size in the world (Ghitulescu and Socolescu, 1941; Bache, 1987; Mitchell, 1996; Foster, 1997). In addition, these gold deposits have achieved mineralogical fame because of the large array of rare and well-formed mineral species found in them. In particular, there are relatively large concentrations of tellurium and the mines are the type localities for several telluride minerals (see below). In spite of the abundance of mineral deposits in the region, their historical economic importance, and their antiquity, there remains a general lack of published information on their geologic (particularly geochemical) characteristics. It is certainly true that the mineralogical diversity of the deposits has in the past led to many detailed mineralogical studies, but relatively little is known of their genesis and evolution, and they are, in fact, rather poorly characterized in comparison to † Corresponding

author: email, [email protected]

0361-0128/00/3044/495-22 $6.00

gold deposits in other regions of the world. We have carried out a detailed, geochemical study of mineralization located in the southernmost part of this area. The results of such studies should improve our understanding of the nature and genesis of these important deposits and allow them to be compared to other precious metal deposits worldwide. Regional Geology The majority of metal deposits in the Romanian-Hungarian region are located in the Carpathian fold belt (Fig. 1). This arcuate, orogenic belt is part of a much larger belt which extends westward into the Alpides (in Austria and Switzerland) and southward, merging with the Dinarides and Hellenides (in Serbia and Bulgaria). These orogens developed during the Late Cretaceous and Tertiary, following closure of the Tethyan ocean, due to the collision of continental fragments of Gondwana with continental Europe and the related subduction of small, intervening, oceanic basins (Herz and Savu, 1974; Csontos et al., 1992; Mitchell, 1992, 1996; Horvath, 1993; Csontos, 1995; Linzer at al., 1998; Nemcok et al., 1998). In this region, ocean closure was accomplished by the northwesterly movement of the Adriatic plate toward the European plate and subduction of the Penninic ocean. Final continental collision commenced in the west of the region in the Eocene and resulted in the formation of the Alpides. A remnant basin (floored by oceanic and thinned continental crust) remained to the east in a preexisting embayment between the Moesian and European plates. Continued northerly plate movement in the Neogene caused the eastward escape of microplates into this embayment and the lateral extrusion of accretionary-crustal wedges in front of these microplates. Foredeep and associated accretionary wedge

495

496

ALDERTON AND FALLICK

19°E 0

N

Brno

100

200

km

EUROPEAN PLATFORM 48°N Budapest Tertiary thrust sheets ('Outer Carpathians') Cretaceous thrust sheets and basement ('Inner Carpathians')

Apuseni Mountains

PANNONIAN BASIN

Area of Figure 2

Cretaceous-Palaeocene igneous rocks ('banatites') Neogene calc-alkaline igneous rocks Late Neogene-Quaternary alkaline igneous rocks Major thrusts

Belgrade

MOESIAN PLATE

National boundary

Bucharest

FIG. 1. Simplified geologic map of the Carpathian fold belt. The rectangle in the Apuseni Mountains shows the approximate position of the study area as detailed in Figure 2. Modified from Pécskay et al. (1995).

sediments (flysch and molasse), thrust fronts, and metamorphism all progressively migrated eastward in response to this plate movement. The closure of this small ocean was aided by passive subduction of oceanic crust to the southwest and west. Final closure of the oceanic basin and continental collision took place in the the east during the Miocene and resulted in the emplacement of several nappe complexes (Dallmeyer et al., 1996) over the foredeep sediments and on to the continental margin represented by the European platform (these are the outer Carpathians in Fig. 1). At the same time, back-arc extension, strike-slip faulting, and thermal subsidence in the west led to the formation of the Pannonian basin, an extensive area containing shallow marine and brackish water (Planderova et al., 1993). The development of the Carpathian fold belt and the Pannonian basin was accompanied by widespread igneous activity (Fig. 1). A suite of Late Cretaceous to early Eocene acidicintermediate intrusive and extrusive igneous rocks (“banatites”) contain Cu-Mo-Fe porphyry and skarn mineralization (Cioflica, 1989) and are thought to have formed during an early stage of subduction. However, Neogene volcanic and subvolcanic rocks are much more extensive (Szabo et al., 1992). Three main groups of Neogene igneous rocks have been recognized: (1) acidic tuffs and ignimbrites of early Miocene age, (2) calc-alkaline stratovolcanoes of mid-Miocene to Pliocene age, and (3) alkaline volcanic rocks of PlioPleistocene age. 0361-0128/98/000/000-00 $6.00

The calc-alkaline igneous rocks are by far the most abundant in the region, and in Romania it is these rocks that host much of the precious and base metal deposits. They mostly occur in a magmatic arc which follows the plate boundaries and the sites of former subduction (Mason et al., 1998; Seghedi et al., 1998). Ages of volcanic activity decrease toward the east and southeast, following the retreat (roll back) of subduction. The Apuseni Mountains consist of nappes of Paleozoic and older metamorphic rocks and a cover of Mesozoic rocks, which were stacked in their current position and metamorphosed prior to the Late Cretaceous (part of the inner Carpathians of Fig. 1; see Dallmeyer et al., 1999). Widespread Mesozoic basaltic pillow lavas are thought to indicate the former presence of the small oceanic basins (Radulescu and Sandulescu, 1973). Neogene, calc-alkaline, igneous rocks are particularly abundant in the Metaliferi Mountains. Here they appear to display an overall northwest-southeast alignment, probably because of underlying, basement, fracture control. These igneous rocks have a geochemical signature which is indicative of subduction (see below), but other evidence of subduction is less apparent (see Seghedi et al., 1998) and they appear to be immediately postcollisional in age. A partially subducted crustal slab (of Moesian plate) has recently been inferred from gravity modeling just to the south of the Apuseni Mountains (Szafian et al., 1997) and could represent the source material for these magmas (as advocated for the volcanism in the eastern Carpathians; see Mason et al., 1998).

496

METALIFERI MOUNTAINS, WESTERN ROMANIA: Au-Ag-Te MINERALIZATION

497

23°E

3 4

46°N

5

2

1

Certej Quaternary sediments Miocene stocks Miocene lavas Miocene sedimentary rocks Cretaceous sedimentary rocks Mesozoic basalts

10 km

Carboniferous metasedimentary rocks Fault

3

Hondol

1

Sacaramb

4

Coranda

2

Magura

5

Bocsia

MINES:

N

FIG. 2. Geologic map of the southern part of the Metaliferi Mountains, showing sample localities. Modified from Bordea and Borcos (1972), Bordea et al. (1978), Borcos et al. (1981), and Lupu et al. (1982).

Methods The volcanic rocks which host the precious metal deposits are reasonably well-exposed and a representative selection of the unaltered rock types could be collected fairly easily. Altered samples and mineralized veins were mostly collected in situ underground. Some of the deposits investigated during the present study have been exploited for over 200 years and many parts of the mines are now no longer accessible. However, access is still possible in some of the upper levels via old but maintained workings or via new exploration drives, and many of the important veins can still be examined and sampled. The locations of the mineral deposits studied are shown in Figure 2 and further details are given in the Appendix. Geochemical analyses of rocks were obtained by X-ray fluorescence spectrometry using a Philips PW1480 spectrometer. Major elements were determined on fused glass discs and trace elements were determined on pressed powder pellets, using the major element concentrations for calculation of matrix corrections. Total volatile content of the rocks (loss on ignition, LOI) was obtained by igniting the sample at 1,100°C and measuring the weight loss. Some problems were experienced with low analytical totals—particularly in the altered and mineralized rocks. These rocks contain large amounts of volatiles (H2O, CO2, and S) and it is thought that sulfur, in particular, may not be completely released during the ignition process. 0361-0128/98/000/000-00 $6.00

The thermometric behavior of the fluid inclusions was studied using doubly polished wafers (~ 200 µm thick) and a Linkam TH600 microscope heating-freezing stage. This was calibrated using synthetic H2O- and CO2-rich fluid inclusions in quartz; the accuracy of the resulting data is approximately ±0.2°C in the range –30° to +50°C, ±0.5°C in the range 50° to 200°C, and 1°C at temperatures above 200°C. Salinity estimates were determined from the last melting temperatures of ice, utilizing the equation given by Bodnar (1993). Mineral separation for stable isotope analysis was accomplished using a combination of handpicking under a binocular microscope, magnetic separation, elutriation in water, and flotation in sodium polytungstate. Purity of mineral separates was checked by X-ray diffraction and chemical analysis (ICPAES). Minerals were analyzed for their O and C isotope compositions in the stable isotope laboratories at Royal Holloway. Carbon dioxide was released from carbonates by reaction with phosphoric acid and the isotopic ratios were measured on a VG PRISM mass spectrometer. Quartz, mica, and barite were analyzed by automatic laser fluorination, using a modification of the technique outlined by Mattey and Macpherson (1993). The oxygen was released using BrF5, concentrated using silica gel at liquid nitrogen temperatures and analyzed directly on a VG Optima mass spectrometer. Analytical errors for δ13C and δ18O in carbonates are approximately ±0.1 per mil (2σ) and for δ18O in silicates approximately ±0.15 per mil (2σ). Micas and fluid inclusions were

497

498


analyzed for δD at the Scottish Universities Research and Reactor Centre, East Kilbride. Water was released at 1,250°C using induction heating and reduced to hydrogen for analysis using a uranium furnace. The δD values were determined using a VG SIRA 9 mass spectrometer. Analytical errors for δD are approximately ±5 per mil (2σ). The O and H isotope composition of fluids precipitating the various mineral phases was calculated from their isotopic composition using published experimental mineral-water isotopic fractionation relationships (see Table 1). Analyses for δ34S were obtained using the technique of Robinson and Kusakabe (1975; for sulfides) and that of Coleman and Moore (1978; for sulfates). The SO2 released by these methods was isotopically analyzed on a VG SIRA 11 dual inlet mass spectrometer; the overall precision is ±0.2 percent (1σ) for each method. The mineralogy of the veins and wall-rock alteration assemblages was determined using electron microscopy and Xray diffraction. More detailed characterization of the sheet silicates was established by glycolation and heat treatment. The presence of K feldspar in hand specimens and thin sections was confirmed by staining with sodium cobaltinitrite. Igneous Host Rocks The most common Neogene igneous rocks in the study area and the hosts to the mineralized veins are porphyritic andesites. These andesites contain abundant hornblende and plagioclase feldspar, together with varying amounts of biotite, quartz, and partially resorbed pyroxene. Most andesites are fairly fresh but propylitic alteration is also common (see below). Some differences in mineralogy and texture of the andesites can be recognized in the field and several varieties are distinguished in the geologic maps of the area. In the area studied here, there are two main andesite types: the Barza type (found at Magura and Hondol), which contains plagioclase feldspar, amphibole, and quartz, and the Sacaramb type (found at Sacaramb and Bocsia), which contains plagioclase feldspar, amphibole, quartz, biotite, and pyroxene. However, detailed mapping in the vicinity of the Sacaramb mine has shown that each andesite type can exhibit distinct mineralogical variations (Fig. 3b). In the study area the andesites occur as intrusive stocks and necks and as extrusive pyroclastics and lava flows. There is some attempt to distinguish intrusive and extrusive rocks in the published geologic maps (see also Fig. 2) but in practice this discrimination can sometimes be difficult, and published interpretations differ. This difficulty arises in part because the environment was probably transitional between intrusive and extrusive, where subvolcanic vents erupted onto the land surface as lava flows. The igneous rocks were intruded through and into Carboniferous schist (the local basement), Mesozoic sedimentary (sandstone, limestone) and volcanic rocks (basalt and pyroclastic), and Neogene sedimentary rocks (sandstone and shale; Figs. 2 and 3b). The chemical composition of the unaltered igneous rocks is typical for low K andesites found at destructive plate margins (Table 2). The low Nb content, high contents for some incompatible elements (e.g., Ba, Sr, Th), and relatively high initial 87Sr/86Sr ratios are compatible with a subduction affinity, probably at a continental margin (Wilson, 1989). 0361-0128/98/000/000-00 $6.00

Early paleontological and stratigraphic considerations have indicated that there were three distinct phases of volcanism in the area and that the most widespread (second) phase (and the one most closely associated with the precious metal mineralization in the area) was essentially of Sarmatian to Pannonian (middle Miocene) age (Ghitulescu and Socolescu, 1941; Cioflica et al., 1973; Borcos et al., 1986). More recent K-Ar and Rb-Sr dating studies (Lemne et al., 1983; Pécksay et al., 1995; Rosu et al., 1997; Alderton et al., 1998) have confirmed these conclusions: the timing of igneous activity in the southern Metaliferi Mountains spans the period from 7.4 to 14.7 Ma, the volcanic rocks in the immediate study area having ages of ~14 Ma. Mineral Deposits Base and precious metal deposits have been exploited at numerous localities in the Metaliferi Mountains since preRoman times. However, the earliest workings were probably from placers and similar shallow deposits and it was only after the Roman occupation that significant underground exploitation occurred (Ghitulescu and Socolescu, 1941; Ianovici and Borcos, 1982). Production data are difficult to obtain because of this long history of mining and because of the recent secrecy attached to such information. Production of gold probably never exceeded a few tons per annum, but it was regular, and Ghitulescu and Socolescu (1941) estimated that more than 1,000 t of gold had been extracted in the Metaliferi Mountains by 1941. Mitchell (1996) estimates that the area has produced about 1,600 t Au, although the source of this figure is not given. Some of the gold deposits in the Metaliferi Mountains also contain substantial amounts of silver and base metals (Cu, Pb, Zn), although overall these metals have been of lesser economic importance. There are several Neogene porphyry copper deposits in this region, although most are relatively small and marginally economic. There is extensive literature on the mineral deposits of the area, although little of it is recent and much is descriptive in nature. Most of the earlier studies were published in German (e.g., von Hingenau, 1857; von Cotta and von Fellenberg, 1862; Semper, 1900; and references listed in Beck, 1905), French (e.g., Ghitulescu and Socolescu, 1941) or, more recently, Romanian (Ianovici et al., 1976; and references cited in Ianovici and Borcos, 1982). Little information is available in English but Ianovici and Borcos (1982) provide a good introduction to the deposits (see also Emmons, 1937). The summaries presented below are based on personal observations and information presented by Udubasa et al. (1992). For this study, samples were collected from five deposits in the southern part of the district: Magura, Baiga-Hondol, Coranda, Bocsia, and Sacaramb (see Fig. 2, and Appendix). The majority of the samples were obtained from the largest deposit, Sacaramb, but samples from the other deposits were included in this study for comparative purposes, particularly as they are currently the focus of further exploration and evaluation. Sacaramb The deposit of Sacaramb (formerly Nagyag or Szekeremb) was discovered in 1740 and has been worked continuously from 1745 to the present day. At Sacaramb, more than 230

498

0361-0128/98/000/000-00 $6.00

499

V V V V V V/A

94-38 95-15 95-18 95-21 95-22 95-24

15.4

15.3 16.7 16.7 15.5 15.0 14.1

16.5 17.7

14.0 10.6* 14.7 14.3 15.3

16.5 14.3

14.5

11.7 12.1

40

35 30 40 40

13.3 13.9 11.0 12.6 (12.4)

11.2

9.2

11.8 13.0 7.7 11.2

–58

–57

–54

–65 –70 –51 –70

–57, –75

–60, –64 –66

δD mica (‰SMOW)

δD fluid inclusion (‰SMOW)

–51

–76

–34

–74 (barite)

–60

30

13.2 (11.4)

11.9

11.7 13.1

δ18O mica (‰SMOW) (corrected)

16.3 15.9

25 35

% quartz with mica

–50

12.8 14.1

δ18O mica (‰SMOW)

16.3 16.0

10.1*

13.4, 13.8 13.8, 16.0 15.5

δ18O quartz (‰SMOW)

13.5, 13.3 14.8

21.3

14.4, 12.6

16.4

14.0 14.8

–6.7, –6.9 –3.8

–7.5

–3.6, –3.3

–6.8

–6.1 –5.9

14.7, 14.7 –7.6, –6.9, –7.0

–6.3 –7.2

16.2, 16.0 14.7

–5.4, –5.4

16.5, 16.5

δ13C calcite (‰PDB)

–12.3 14.8, 15.4 –5.6, –5.5

δ18O calcite (‰SMOW)

6.4 7.8 7.8 6.6 6.1 5.2

7.6 8.8

5.8 5.4 6.4

5.1

7.6 5.4

5.6

7.4 7.0

7.4 7.1

–6.1, –6.5 4.7 4.9 6.6

δ18O fluid (‰SMOW) (from quartz)

4.2

6.9 8.0 2.7 6.2

6.9

6.8 8.1

δ18O fluid (‰SMOW) (from mica)

–33

(–51)

(–76)

(–34)

–32 (–74)

–29

–30, –48 (–60) –40 –45 –26 –45

(–50)

–37 –41

δD fluid (‰SMOW)

231

Comments

δ18O barite = 3.4 δ18O fluid = –11.4 Poor precision on δD

301 (27)

(269) (375)

216

(458)

Temperature (°C)

Notes: Rock type: A = altered rock, F = fresh rock, V = vein, V/A = veins in altered rock Mica values in parentheses indicate samples separated by handpicking; * = samples probably a mixture of magmatic and hydrothermal quartz (see text) Fluid compositions calculated for a temperature of 250°C, except for barite, which is 100°C; δD fluid values in brackets are from fluid inclusion analyses Mineral-water fractionation equations for determination of fluid compositions and temperatures taken from Matsuhisa et al. (1979) (δ18O quartz), Sheppard and Gilg (1996) (δ18O and δD mica), and Friedman and O’Neil (1977) (δ18O calcite and barite); Temperatures in parentheses derived from calcite-muscovite fractionation; others from quartz-muscovite fractionation

Mean

V V A V A V V A V V V

F A V V V V A A V V V A A V A A V/A V/A

Sacaramb 94-1 94-5 94-6 94-7 94-8 94-9 94-10 94-11 94-12 94-13 94-15 94-16 94-19 94-21 95-2 95-5 95-8 95-12

Other 94-22 94-24 94-25 94-26 94-27 94-28 94-29 94-30 94-31 94-35 94-37

Rock type

Sample no.

TABLE 1. Oxygen, Hydrogen, and Carbon Isotope Analyses of Minerals and Fluids from Mineral Deposits in the Metaliferi Mountains, Romania


499

500


a N

80

1 Location of section shown in Fig. 3b

60

2

90

70 90

13

2 Main vein groups: 1 Longin 2 Antelongin 3 Nepomuc 4 Magdalena 5 Karthause 6 Carolina 7 Ignatiu 8 Magdalena parallel 9 Erzbau 10 Weiss 11 Clara 12 Margareta 13 Anastasia

7

65

60 80

5

3 3

90

6

60

8 11

75

90

90

4

9

60

10

12

75

200m

100m

W

E

Main levels and elevation scale (above sea-level)

b

850m

Maria (784m)

3

1

Bernat IPEG (Daniel)

2

Ferdinand (637m) 550m

7

Carol (494m)

2

66

5 400m

2 Miocene andesite: qtz+hb+bi+px qz+hb+bi bi lavas/pyroclastics Miocene sandstone Cretaceous siltstone Carboniferous schist Mineralized vein

2

Nicolae 4

8 250m

?

100m

Fault

Main vein groups: 1 Longin 2 Antelongin 3 Nepomuc 4 Magdalena 5 Karthause 6 Carolina 7 Ignatiu 8 Magdalena Parallel

FIG. 3. a. Simplified map of the main groups of mineralized veins in the intermediate levels at Sacaramb (projected to surface). The host rock consists mostly of andesite. Stippled area shows the relatively barren, principal andesite stock of Ghitulescu and Socolescu (1941). Some representative vein dips are also shown. Modified from mine plans and Ghitulescu and Socolescu (1941). b. Simplified geologic cross section through the Sacaramb deposit. For location of section see (a). Modified from mine plans and Udubasa et al. (1992). Abbreviations for andesite varieties: bi = biotite, hb = hornblende, px = pyroxene, qtz = quartz. 0361-0128/98/000/000-00 $6.00

500

0361-0128/98/000/000-00 $6.00

58.46 0.70 17.18 5.99 0.16 2.43 7.27 3.13 1.86 0.35 2.10 99.63

501 22 799 8 169 20 62 96

17 802 9 141 17 63 104

2

19 756 9 152 19 63 104

1,506 51 28 46 17 28 11 21 9 19 79

61.15 0.54 17.23 5.27 0.16 2.49 5.16 3.53 2.29 0.25 1.23 99.29

94-4 Fresh

0.704606 0.704526 0.704667

1,356 53 19 57 17 28 10 22 10 18 47

59.01 0.59 17.02 5.72 0.11 2.80 6.35 3.29 1.76 0.21 1.18 98.03

94-3 Fresh

1,441 51 24 46 17 27 11 21 11 21 52

61.70 0.55 17.19 5.23 0.14 2.46 6.05 3.49 1.85 0.21 0.84 99.70

94-2 Fresh

Fe measured as Fe2O3 LOI = loss on ignition at 1,100°C 3 Major elements only 4 Approximate values 5 87 Sr/86Sr values from Alderton et al. (1998)

Sr 5

1 Total

87/86

Parts per million Ba 1,143 Ce 71 Cr 23 Cu 33 Ga 18 La 37 Nb 11 Nd 29 Ni 9 Pb 22 Rb 45 S4 Sc 21 Sr 909 Th 10 V 147 Y 29 Zn 78 Zr 146

Wt percent SiO2 TiO2 Al2O3 Fe2O 31 MnO MgO CaO Na2O K2O P2O5 LOI 2 Total 3

94-1 Fresh

922 0 183 18 66 111

1,453 33 13 87 15 13 10 19 5 11 48 160

58.22 0.60 16.95 5.27 0.13 2.38 5.96 3.98 1.85 0.27 4.31 99.92

94-39 Fresh

14 299 12 122 21 50 148

1,436 77 26 20 18 42 13 32 9 35 162

58.73 0.56 15.19 3.15 0.25 0.96 6.33 0.07 4.11 0.24 7.22 96.80

94-5 Altered

Sacaramb

799 70 20 46 15 38 11 29 10 33 136 20,000 12 71 10 102 13 154 128

70.09 0.47 12.45 3.80 0.20 0.61 0.30 0.00 3.26 0.19 7.71 99.10

94-20 Altered

0.706793 0.706440

1,637 78 26 111 20 43 14 30 13 46 202 20,000 13 94 12 124 21 1,883 148

61.38 0.60 16.26 4.89 1.74 0.85 0.68 0.05 5.40 0.25 6.66 98.75

94-10 Altered

42 7 154 20 93 149

1,565 73 25 59 20 44 12 32 11 88 228 2,220

65.28 0.61 17.09 3.80 1.38 1.10 0.41 0.02 5.62 0.25 4.73 100.29

95-8 Altered

227 9 115 20 64 130

1,341 55 26 20 16 37 11 25 8 34 206 2,460

57.59 0.53 14.25 4.12 0.50 1.14 7.35 0.06 4.82 0.21 6.72 97.29

95-12 Altered

13 29 5 139 13 1,260 95

616 37 5 6 23 19 8 17 8 368 253

60.15 0.61 18.42 4.49 0.16 0.86 1.26 0.03 5.07 0.17 7.71 98.91

94-23 Altered

0.705316 0.710541

18 533 5 141 21 66 104

506 33 6 5 18 16 8 16 6 14 40

58.54 0.59 18.36 5.16 0.13 2.37 6.53 3.37 1.24 0.19 2.85 99.34

94-25 Fresh

12 68 5 135 12 621 93

452 30 4 3 21 15 8 12 6 282 165

64.44 0.57 17.62 4.20 0.03 0.48 0.45 0.57 3.96 0.16 7.36 99.83

94-27 Altered

Magura

64.20 0.47 14.03 6.24 0.04 0.47 0.60 0.14 3.70 0.16 6.45 96.49

94-30 Altered

10,935 19 5 43 19 9 7 13 13 1,197 141 30,000 15 455 3 145 13 3,478 66

TABLE 2. Chemical Analyses of Representative Fresh and Altered Igneous Rocks from the Metaliferi Mountains, Romania

160 1 111 18 50 95

1,460 15 14 14 18 10 9 16 6 50 354 3,560

53.95 0.50 17.03 5.23 0.23 1.58 2.31 0.08 13.14 0.16 3.73 97.94

95-16 Altered

0.705143 0.712023

600 0 137 19 63 91

512 31 14 24 18 4 6 8 8 14 43 80

57.56 0.62 18.33 5.78 0.12 1.73 7.96 3.82 1.48 0.16 2.59 100.15

95-17 Fresh

Hondol


501

502


steeply dipping, mineralized veins occur within an area of ~1 km2 and to a depth of about 600 m below surface. There are reported to be more than 300 km of underground workings and during its lifetime about 85 t of combined Au + Ag have probably been extracted (Udubasa et al., 1992). The dominant trend of the veins is northwest-southeast, but some also trend northeast-southwest and in plan the veins take on a general rhomboid form, around a barren andesite core (Fig. 3a). Most of the veins are located within a series of andesite necks and associated lava flows, although at greater depths some also penetrate the underlying sedimentary rocks (Fig. 3b). The length of the important, individual veins is typically about 500 to 600 m, with a thickness varying from a few millimeters to 2 m (avg, 0.3 m). The veins reduce in number with depth and in addition their gold content decreases and the content of Pb and Zn increases. During the early history of the mine the gold grades were around 80 to 100 g/t, but more recently they have been closer to 2 to 3 g/t. Silver has always been abundant and Au/Ag ratios for the ores have varied between 1:1 and 1:10 (see also Anderson, 1944). Most of the veins have a vuggy texture and minerals occur as well-developed crystals; however, evidence of later shearing and displacement along some of the veins is also observed (Fig. 4a). Irregular hydrothermal breccias are common (Fig. 4b) as are thinner, ore-bearing veins containing clay or brecciated wallrock fragments (glauch veins). Wall-rock alteration assem-

blages are well-developed (Fig. 4a) and contain sericite, K feldspar, quartz, pyrite, and calcite. Although the mineralogy is exceedingly complex, a gangue of quartz, Ca-Mn carbonates, and barite is predominant. The ore minerals consist of sulfides (typically pyrite, alabandite, galena, sphalerite, chalcopyrite, stibnite, realgar), As-Sb sulfosalts (bournonite, boulangerite, tetrahedrite), and a wide variety of tellurides of Au, Ag, Pb, Sb, and Hg (Udubasa et al., 1992). The mine is the type locality for the minerals petzite (Ag3AuTe2), krennerite (AuTe2), stutzite(Ag5-xTe3), nagyagite (Pb5Au(Te,Sb)4 S5-8), muthmannite ((Ag,Au)Te), and krautite (MnHAsO4.H2O). Gold in its native form is rare and most of the gold ore consists of nagyagite and sylvanite. One notable feature of the deposit is that groups of veins with similar orientations and distinctive suites of ore and gangue minerals can be identified. In addition, paragenetic studies of individual mineralized veins have highlighted their mineralogical complexity (Giusca, 1935, 1936), and it must be inferred that individual veins were formed from several, discrete, mineralizing events. The goldand silver-bearing phases of mineralization are associated with galena and sphalerite, while sulfosalts and chalcopyrite make up a later phase. Carbonates tend to be late-stage minerals. Magura The Magura deposit displays similar overall geologic characteristics to those at Sacaramb, but the geology of the deposit

a

b

FIG. 4. a. Sheared vein, Weiss group, Sacaramb. Darker, central portion of the vein (v) consists of comminuted quartz and sulfides. White, wall-rock alteration selvage (w) has a rather sharp junction with unaltered andesite (a). Hammer for scale. b. Hydrothermal breccia containing angular and subangular andesite fragments, Weiss vein group, Sacaramb. Hammer for scale. 0361-0128/98/000/000-00 $6.00

502


is less well-documented. The mineralization is hosted by andesite and contained in six vein groups, each consisting of steeply dipping, north-south-trending veins and breccias. This deposit has never been as rich as Sacaramb; the veins are thin and gold distribution is patchy (although rich in places; Ghitulescu and Socolescu, 1941). The veins are made up of quartz (±calcite, barite); the ore minerals include abundant pyrite and base metal sulfides (chalcopyrite, sphalerite, galena) and minor, gold, silver, Ag-As-Sb sulfosalts, marcasite, arsenopyrite, and tellurides (krennerite, petzite, sylvanite, tellurantimony, altaite, montbrayite; Orlandea and Velciov, 1996). These minerals occur as delicate, euhedral crystals within voids in the veins. At least two stages of mineralization have been identified in this deposit, an Au-Ag stage and a PbZn (±Au, Ag) stage. Wall-rock alteration of the andesites has often been intense and has resulted in the development of sericite, K feldspar, quartz, and calcite (Fig. 5b). A small amount of exploration and exploitation is currently being carried out in the north and northeast of the deposit, in the Maciesu area. Baiga-Hondol These small deposits contain narrow veins of quartz and calcite (±barite) hosted by an andesite body. The main ore minerals include pyrite, sphalerite, galena, chalcopyrite, tetrahedrite, realgar, bournonite, stibnite, and native gold. Euhedral crystals indicate open-space deposition in the veins. Wall-rock alteration of the andesites is characterized by K feldspar- and sericite-rich assemblages. At higher levels the veins penetrate Cretaceous sediments and are worked in a low-grade Pb-Zn-Au (1.1g Au/t) open-pit operation at Coranda (Andrew, 1993). Bocsia This is a breccia-hosted, Pb-Zn orebody, which occurs in a small andesite body. The mineralogy is predominated by sphalerite, tetrahedrite, galena, pyrite, bournonite, and chalcopyrite. The gangue consists of barite, quartz, and calcite.

Quartz pseudomorphs after bladed calcite exhibit similar textures to those thought to be derived from boiling processes (Simmons and Christenson, 1994; see Fig. 6). The ore minerals occur as euhedral crystals in voids within the breccia and clearly grew after brecciation and the proposed phase of boiling. Wall-Rock Alteration The igneous rocks, which are hosts to the mineral deposits, have undergone mild but widespread pervasive propylitic alteration. Ferromagnesian minerals in the andesites are most affected and they have been variably altered to chlorite, calcite, epidote, and magnetite; feldspars have been partially altered to sericite. Chemical changes in these altered rocks are small, except for a marked addition of volatiles (LOI). This style of alteration is most apparent in zones which extend for at least several tens of meters around the mineralized veins and Ghitulescu and Socolescu (1941) note that mineral veins tend to be more abundant in areas of propylitized igneous host. Immediately adjacent to the mineralized veins the hydrothermal alteration is typically intense (e.g., Figs. 4a and 5b). Alteration selvages are commonly up to 1 to 2 m in width and consist of varying proportions of quartz, sericite, K feldspar, calcite, and pyrite. An XRD study of the mica has identified an interlayered smectite component (up to 15%). In most respects, the patterns of hydrothermal alteration conform to those of the adularia-sericite style of epithermal, precious metal mineralization, although the original classification was not developed for tellurium-rich deposits (Heald et al., 1987). No kaolinite or alunite has been identified, although geologic maps indicate that deposits in the area contain these minerals (Bordea and Borcos, 1972; Borcos et al., 1981). Chemical analyses of altered rocks can be readily explained by the secondary mineral phases present (Table 2). Marked enrichments in K, Rb, and volatiles (LOI) and marked depletions in Na, Mg, and Sr, point to the replacement of plagioclase, feldspar, and amphibole by K feldspar and K mica.

a

b

FIG. 5. Thin section photographs of fresh and altered andesites (transmitted light, crossed polars). a. Fresh andesite from Sacaramb (94-2), showing phenocrysts of quartz, plagioclase feldspar, biotite, and amphibole in a finer grained matrix. b. Altered feldspar-rich andesite from Magura (94-27) showing feldspar phenocrysts and groundmass replaced by fine-grained sericite, quartz, pyrite, and K feldspar. Scale bar = 0.5 mm. 0361-0128/98/000/000-00 $6.00

503

503

504


FIG. 6. Quartz after bladed calcite, Bocsia mine. Scale bar = 1 cm.

Minor additions of light rare earth elements (La, Ce) are also noted. Variable enrichments in Cu, Pb, Zn, Fe, and S point to the presence of sulfides, whereas the abundance of Ba indicates the presence of barite. Variable amounts of Ca (and to a lesser extent Mg and Mn) are related to the presence of secondary Ca (Mg-Mn) carbonate. Rb-Sr isotope studies on alteration assemblages at Sacaramb (using whole rock and sericite) have produced ages of 13.7 (±0.7) Ma (Nepomuc 3 vein) and 14.8 (±0.4) Ma (Karthause vein; Alderton et al., 1998). Because of the widespread propylitic alteration it is difficult to obtain an accurate age for the exact volcanic rock units which host the mineralized veins. However, these ages of hydrothermal alteration are within error of the age determined for the unaltered volcanic rocks of the Sarcau stock at Sacaramb (14.0 ± 0.9 Ma, using biotite and whole rock) and also comparable to the ages of 13.6 ± 1.0 Ma (Sarcau) and 13.6 ± 1.7 Ma (Zuckerhut), derived from K-Ar dating of biotite by Lemne et al. (1983). Preliminary data, therefore, indicate that volcanism and the main phase of hydrothermal alteration at Sacaramb were almost coeval (within 400°C), but these sulfur-bearing minerals are clearly not cogenetic and were deposited in different stages and from fluids with slightly different compositions and/or at slightly different temperatures. The results of the S isotope study would, however, support the suggestion that pyrite is an early mineral which is associated with the wall-rock alteration, whereas the galena and sphalerite are distinctly later and were precipitated under similar conditions. The barite is clearly of a late paragenesis and formed under different conditions and from different fluids—possibly after the influx of low δ18O meteoric water into the hydrothermal system raised the fO2 and triggered a change from sulfide to sulfate precipitation.

TABLE 4. δ34S Values for Sulfides and Sulfates from the Metaliferi Mountains, Romania Sample no.

Mineral

92-3 94-6 94-6 94-14 94-18 94-22 94-22 94-23 94-26 94-26 94-31 94-32 94-32 94-38 94-38 94-38

Sphalerite Sphalerite Pyrite Pyrite Realgar Sphalerite Galena Pyrite Barite Sphalerite Barite Galena Sphalerite Galena Sphalerite Pyrite

δ34S (‰ CDT) +1.2 –0.5 –0.3 +0.2 +0.9 +1.5 +0.1 +2.6 +15.7 +1.2 +18.1 –1.5 +0.4 –0.7 +1.1 +1.2

Temperature (°C)

369–434

316–376 333–394

Note: Temperatures based on the isotopic fractionation between galena and sphalerite, using the fractionation relationships of Grootenboer and Schwarcz (1969), Kajiwara and Krouse (1971), Czamanske and Rye (1974), and Ohmoto and Rye (1979)

fractionation factors of Ohmoto and Rye, 1979). In the low fO 2 regions of the stability diagram the δ34S for the total fluid will approximate that for H2S, i.e., the fluid composition will be close to zero per mil. In such a situation the sulfur in the fluid could be interpreted as having a magmatic source, just as is traditionally invoked for a large number of epithermal vein systems (Henley, 1991) or other vein deposits associated with igneous rocks (Field and Fifarek, 1985). Unfortunately the stability field in the diagram also allows for higher fO 2 conditions and here the δ34S composition of the fluids could be much higher, by up to about 15 per mil. This would indicate other sources of sulfide sulfur, such as evaporite-bearing sediments. No constraints on temperature or fO 2 /pH conditions exist for the barite, as it was deposited after the sulfide mineralization. The presence of monophase fluid inclusions suggests a low temperature of formation, probably less than 100°C. At such temperatures and in the likely fluid conditions, the fractionation of δ34S between fluid and mineral could be in the large range of zero to >30 per mil and thus the δ34S values for the fluid could be anywhere in a large range (–15 to +15‰). Sulfur isotope geothermometry has been attempted using minerals thought to have coprecipitated. The results are mostly not geologically reasonable considering the fluid inclusion-based temperature estimates in the range 200° to 300°C. There is some variation in published mineral-mineral fractionation relations, but for the mineral pair most likely to be cogenetic (sphalerite and galena) the temperatures are mostly in the range 300° to 400 °C (Table 4). These elevated isotopic temperatures for the galena-sphalerite pair are similar to those reported in numerous other studies (Ohmoto, 1986), where the differences could be explained by the variation in δ34S of the H2S during ore deposition (e.g., Casadevall and Ohmoto, 1977), a variation in fO 2 conditions, and/or the marked depletion in S content of the fluids during metal 0361-0128/98/000/000-00 $6.00

Carbon isotopes Carbon isotope (δ13C) values for calcite and manganoan calcite vary from –2.3 to –3.3 per mil relative to PDB (Table 1). The lowest value (–12.3‰) occurs in the relatively fresh host andesite from Sacaramb. This rock has undergone mild propylitic alteration and the C isotope value of the calcite is quite distinct from the range for vein and alteration carbonates (–3.3 to –7.6). This would support the proposal that the pervasive propylitic alteration was an earlier event, caused by fluids which had a different origin and were unrelated to those causing the vein mineralization. Under the conditions of hydrothermal activity indicated in Figure 10, the carbon in the fluid would be mostly present as CO2 dissolved in water and as H2CO3. Figure 10 indicates that at these proposed conditions the difference between the C isotope composition of the carbonates and that of the associated fluid would be small (