Sergey Ivanov's homepage

Home Up Gold in 2000 GSA-2000 Kumtor

Gold in 2000

Gold in 2000.pdf

 

Ore texture and stable isotope constraints on ore deposition mechanisms at the Kumtor lode gold deposit.

Sergey M. Ivanov1, Kevin M. Ansdell1, Dwayne L. Melrose2 

1Dept. Geological Sciences, Univ. Saskatchewan, Saskatoon, S7N 5E2, Canada 

2Kumtor Operating Company, 24 Ibraimova Street, Bishkek, Kyrgyzstan 7200312 

ABSTRACT: Mineralization at Kumtor gold deposit, Kyrgyzstan, is hosted in the Vendian carbonaceous phyllites and overprints the three deformational fabrics, which constrains its age to Late Paleozoic. Gold mineralization is related to intensive veining, stockworks and hydrothermal breccias. Native gold, Au and Au-Ag tellurides are intimately related with pyrite. Also, their textural association with altaite, magnetite, hematite, barite, strontianite, calcite, galena and sphalerite, and locally corroded pyrite, suggests that fluctuations in f(O2) was a significant factor in gold deposition. f(O2) fluctuations may also account for differences in the d34S signatures of Stage 2 vs. Stage 3 pyrites. CO2 effervescence during hydrothermal brecciation may have caused increase in f(O2), pH, depletion of fluids in 18O during Stage 3, and promoted carbonate deposition. Hydrolysis of carbonaceous matter from the hosting lithology might have initially produced and replenished lost CO2. Fluid inclusion analysis will be vital in constraining the nature of these processes.

INTRODUCTION

The Kumtor lode gold deposit, located in the eastern part of the Kyrgyz Republic (Central Asia), contains 288 t of gold at a grade of 3.57 g/t (Homeniuk, 2000). The objective of this paper is to summarize textural and preliminary stable isotope constraints on depositional mechanisms that may have been important during the formation of the deposit.

DEPOSIT GEOLOGY

The mineralization is hosted in Vendian low greenschist grade carbonaceous phyllites, and comprises veins, stockwork, hydrothermal breccias and associated pervasive alteration in the immediate hangingwall of the SW-NE trending moderately SE-dipping Kumtor Fault Zone (KFZ). The KFZ is a possible splay off the regional Nikolaev Lineament, which separates the Northern and Middle Tien-Shan, and consists of a series of anastomosing faults filled with clay- and carbonaceous matter-rich gouge. These faults separate blocks of friable to massive rocks, which include rock types that are exotic to both the hangingwall and footwall (Abeleira et al., 2000). In the vicinity of the deposit, the KFZ juxtaposes Cambro-Ordovician limestones in the footwall against Vendian phyllites in the hangingwall. These observations suggest a complex movement history, and significant amplitude of displacement. Three deformational fabrics are developed in the Vendian rocks (Abeleira et al., 2000): S1 is defined by phyllitic foliation which is subparallel to compositional banding (S0); S2 is a crenulation cleavage; and D3 is represented by kinkbands developed at cm to dm scale. These fabrics are generally orthogonal within the deposit, and are crosscut and overprinted by auriferous mineralization. Folds in Carboniferous clastic sedimentary rocks in the area are considered to have developed during D3, which provides a maximum age limit on mineralization. The main zone of mineralization lies to the SE of the KFZ, has a wishbone shape in plan view, strikes 035-055°, and dips 40-55° to the SE. The two arms join in the NE-part of the deposit where veining, stockwork and hydrothermal brecciation are intensively developed, and gold grade is highest. Common SW-NE and rarer NW-SE trending graphitic faults crosscut the orebody, and often separate intensively mineralized blocks from those lacking appreciable alteration, which suggest post-mineralization movement. The high grade ore is abruptly terminated to the NE by a series of these faults. Kinematic indicators (fault fabric, riedel shears, lineation) are relatively scarce and often yield contradictory evidence for the sense of movement along the graphitic faults, and perhaps reflect the sense of movement of late episodes, which postdate mineralization. Nevertheless, the majority of them suggest significant strike-slip movement, which is consistent with similar observations within the KFZ. To the SE, the orebody is also bounded by unmineralized rocks juxtaposed along non-planar graphitic faults oriented subparallel to KFZ, which have a predominantly vertical sense of movement. Ore grades vary from 1.2 g/t (the lowest exploitable grade) to over 70 g/t and are generally correlated to pyrite content. Ore shoots are confined to areas with high hydrothermal minerals/host rock ratio such as intensive stockwork, hydrothermal breccia, or banded carbonate-pyrite rocks formed in the late stages of mineralization. The veins within the stockwork seem to be controlled by the earlier orthogonal fabrics. In addition, randomly distributed hydrothermal breccia bodies, up to several meters in size, are situated within the stockwork. The transition into breccias is gradational: the wall rock/vein material ratio and the degree of disorientation of separate phyllite fragments along with the number of small clasts in the hydrothermal matrix increase as one progresses from stockwork into the hydrothermal breccia bodies.

VEIN PARAGENESIS

Four stages of auriferous mineralization are recognized at Kumtor. The weakly auriferous quartz-carbonate-albite-chlorite-sericite-pyrite Stage 1 mineralization is defined by primarily pervasive alteration and rare small veins and veinlets. Locally, alteration is accompanied by removal of carbonaceous matter, causing the rocks to exhibit a ?bleached? appearance in outcrop, and recrystallization of sericite and chlorite. Stage 2 is defined by intensive veining, stockwork and hydrothermal breccia development. Pervasive alteration associated with this stage includes sericitization, chloritization, and silicification, whereas pyritization, carbonatization and feldspathization are less important. Veins and hydrothermal breccia infill consist of variable amounts of carbonates, quartz, pyrite, K-feldspar, sericite, and chlorite, minor amounts (normally 0-10%) of chalcopyrite, hematite, barite, and strontianite and accessory magnetite, scheelite, ferberite, rutile, cassiterite, sphalerite, galena, native gold, calaverite, petzite, sylvanite, altaite, melonite and tetrahedrite. However, a distinct paragenesis can often be identified in these veins. Potassium feldspar is early in the paragenesis and forms almost monomineralic veinlets and the rims of more complex veins. The later mineral growth in these veins consists of dolomite, quartz, K-feldspar, calcite, pyrite, hematite, barite and sometimes albite. The latest mineral assemblage that develops during Stage 2 consists of ferroan dolomite, relatively euhedral pyrite, albite, later potassium feldspar, barite, hematite, gold, calaverite, petzite and sylvanite. Veins, stockwork, hydrothermal breccia and associated pervasive alteration are most characteristic of Stage 3. The mineral assemblages lack potassium feldspar and the two main minerals that locally make up to ~100% of the vein and breccia infill are carbonates and pyrite. Among other major minerals are albite (0-25%), quartz (0-10%), sericite (0-15%), chlorite (0-15%), chalcopyrite (0-5%), barite (0-5), and hematite (0-20%). The accessories include scheelite, rutile, gold- and gold-silver tellurides, and native gold. Stage 3 mineral assemblages comprise the zones of highest grade ore. The Stage 4 carbonate-pyrite rocks are often strongly layered to massive, have a brecciated appearance, and form planar bodies elongated in NE-SW to W-E directions. They are associated with zones of intensive deformation defined by the development of layering, folding, and brecciation and boudinage of previously altered phyllites and hydrothermal rocks. The mineral composition corresponds in general to the previous stage (carbonates and pyrite, and minor albite and quartz).

TEXTURAL SETTING OF GOLD

The ore grade is positively correlated to the pyrite content. Auriferous minerals are represented by native gold, calaverite, petzite, and sylvanite, and are generally intimately associated with pyrite. The textural relationships of the auriferous phases to pyrite and other vein minerals include: (1) Primary inclusions in pyrite; (2) Overgrowth films, crack fills in pyrite and other minerals, and secondary inclusions in pyrite. Primary inclusions in pyrite are represented by the tellurides, native gold, barite, calcite, magnetite, hematite, chalcopyrite, galena and sphalerite. Galena, sphalerite, and altaite occur exclusively as primary inclusions in pyrite in the Kumtor ores. Many pyrite grains are oscillatory zoned and exhibit growth bands typified by varying arsenic content (0-1.9 wt %). These bands highlight the growth patterns in the pyrite grains, and indicate that the pyrite grains were commonly corroded and recrystallized after initial growth. Texturally later native gold and gold-containing tellurides are typically associated with corroded pyrite grains. Locally, they occur along the contacts between earlier and later pyrite phases. The latest ("post-pyrite") auriferous tellurides and native gold appear to occur in association with barite, strontiobarite, strontionite, hematite, chalcopyrite, and scheelite. All these minerals overgrow, seal cracks in, and occur as secondary inclusions in pyrite. Hematite often forms very large inclusions in pyrite and contributes in places as much as 30% to the grain volume. Inclusions of magnetite in the pyrite matrix are usually partially replaced by hematite.

STABLE ISOTOPES

A stable isotope study has been initiated to assist in unravelling the fluid evolution of the Kumtor deposit. However, the complex textural relationships between hydrothermal minerals means that not all samples yield suitable pure mineral separates. d18O values for quartz from Stage 2 veins are higher and less variable (15.4+/-0.65 permil, n=14) than those for Stages 3 and 4 (11.4+/-1.8 permil, n=8). Stage 2 hematite d18O values are -0.5+/-1.6 permil (n=8). Quartz-hematite geothermometer, using fractionation factors of Zheng and Simon (1991), yielded temperatures of 320+/-45° C for Stage 2. The calculated d18Ofluid for Stage 2 is thus 9.5+/-1.5 permil, which falls in the range of d18O values typical of metamorphic or magmatic fluids (Sheppard, 1986). The lower d18O values for quartz from Stages 3 and 4 suggest changing d18O fluid and temperature conditions. Influx of another fluid with a lower d18O value may explain the decreasing d18O values of quartz with time. Another possible cause for changing d18O signatures of the fluids may be effervescence of CO2. d34S values for pyrite from Stage 2 (-0.6 to 1.6 permil, n=7) exhibit a tighter range than for Stages 3 and 4 (-1.1 to 3.1 permil, n=7), whereas diagenetic pyrites in the phyllites range from 1.0 to 14 permil (n=7). The tight range exhibited by the hydrothermal pyrite suggests that the sulfur in the deposit was derived externally, and not from the carbonaceous phyllites at the site of deposition.

DISCUSSION

Textural evidence suggests that the deposition of hematite, calaverite, and native gold is related to the episodes of pyrite corrosion. This relationship can be used to constrain the evolution of possible depositional conditions using the f(O2)-pH diagram calculated for conditions typical of natural ore deposition environments (Zhang and Spry, 1994, Cooke et al, 1996). Calculations indicate that the deposition of calaverite appears to be most easily triggered by: a) a drop in temperature since its stability field expands with decreasing temperature, and b) by an increase in oxygen fugacity. Changes in pH and total sulfur are not as significant in calaverite deposition. Evidence for fluctuations in f(O2) are provided by the replacement of magnetite by hematite and formation of hematite at the expense of pyrite (pyrite corrosion) at certain stages of the growth of pyrite grains. Such changes would allow calaverite to precipitate. The presence of native gold in association with hematite may indicate that the upper boundary of the calaverite stability field was crossed indicating that f(O2) values were as high as -31 log units at temperatures of approximately 300° C. Fluctuations in f(O2) may also account for slightly lighter and more variable d34S signatures of the Stage 3 and 4 pyrite in comparison to Stage 2 pyrite. McKibben and Eldridge (1990) detected extreme sulfur isotope zonation in pyrite grains related to boiling-induced increase in f(O2) and pH, using SHRIMP techniques. Such variations cannot be resolved using bulk techniques employed here, but the variability of d34Spyrite for Stages 3 and 4 may imply zonation in the sulfur isotope composition of pyrites. The presence of hydrothermal breccias in the orebody suggests that catastrophic fluid pressure drops and related boiling or effervescence may be important processes. Boiling of fluids has been suggested as a cause of increases in f(O2) and pH due to preferential loss into the vapour phase of H2, CH4, and H2S relative to SO2, SO3, and O2 (Drummond and Ohmoto, 1985). Boiling may also cause changes in the d18O value of hydrothermal fluids due to fractionation of 16O into vapour phase and decreasing temperature, although the extent of boiling has to be extreme (Truesdell and Nathenson, 1977, Larson and Taylor, 1987, Lynch et al, 1990, Shmulovich et al, 1999). The effects of fluid unmixing in the form of CO2 loss have been the subjects of a number of studies (Bottinga, 1969; Truesdell and Nathenson, 1977; Averkin, 1987; Lynch et al, 1990; Bowers, 1991). The enrichment of stockwork and hydrothermal breccias with carbonate minerals may be explained by reactions of the type:

Ca2+ + 2HCO3 2-  = CaCO3¯ + H2O + CO2­

which tend to proceed to the right if CO2 is removed. The CO2 loss will cause f(O2) to increase, and the fractionation of 18O into CO2-rich gaseous phase offers an explanation for the observed lighter d18O values for quartz from Stages 3 and 4. The hydrolysis of readily available carbonaceous matter at Kumtor might have initially produced and replenished lost CO2. Lynch et al (1990) calculated, by using the fractionation factor of Truesdell (1974), that it is necessary to convert 25% by mass of H2O from an original fluid to CO2 in order to obtain a 4permil depletion in d18Owater at temperature of about 300° C, the amount of water will be smaller at lower temperatures. The drop in d18O value of quartz of approximately 4 permil from Stage 2 to Stages 3 and 4 could be explained by this process, although it may also be related to the influx of isotopically light water.

CONCLUSIONS

1. Mineralization at Kumtor is related to intensive veining, stockwork and hydrothermal breccia development, is hosted in Vendian carbonaceous phyllites, and crosscuts the three deformational fabrics. The youngest fabric is considered to be Carboniferous in age.

2. Pervasive alteration and veins with crack-sealing and fibrous textures were typical at earlier, relatively weakly auriferous stages of mineralization, whereas the bulk of gold is related to stockwork and hydrothermal breccias, which are more typical of later mineralization stages.

3. The inception the stockwork development and hydrothermal brecciation was demarked by changes in mineralogy (the predominant quartz+K-feldspar in earlier stages vs carbonate+pyrite in later stages) and stable isotope signatures of precipitated minerals (18O-depleted quartz, and slightly lighter and more variable d34Spyrite values for later stages), which signifies changes in fluid chemistry and/or deposition conditions.

4. The textural setting of gold-bearing phases implies that increase in f(O2) was likely a significant factor in gold deposition.

5. These changes may be explained by efferevescence of CO2 during hydrothermal brecciation, which might have caused increase in f(O2), pH, and depletion of fluids in 18O during Stage 3 through 4, and promoted carbonate deposition. The hydrolysis of carbonaceous matter from the host lithology might have initially produced and replenished lost CO2.

6. However, the present data on textural relationships and stable isotopes is not sufficient in order to allow unambiguous constraint on the fluid origin and the physico-chemical processes acting during ore deposition. Obtaining additional information on composition and character of fluids, and stable isotope (dD, d13C) signatures of minerals and fluids are crucial in placing constraints on the origin of fluids and understanding of the physicochemical processes involved in ore deposition. These are the subjects of ongoing research.

ACKNOWLEDGMENTS

The study is funded through a Cameco-NSERC grant to KMA, and the analytical facilities at the University of Saskatchewan are partially funded by an NSERC Major Facilities Access Grant. Authors are grateful to E. Parviainen, P. Litvinov, A. Sharomov, I. Rekhin and A. Goncharov for logistical support during fieldwork and discussion of deposit geology. Discussions with D. Thomas and V. Sopuck were very fruitful and have helped to guide the research in a more efficient way. SMI would also like to express appreciation of help and patience of the personnel with the Mining Engineering Department at the Kumtor mine site during the three field seasons.

REFERENCES:

Abeleira, A., Ansdell, K.M., Thomas, D, Heaman, L. and D.L. Melrose. 2000. Geology and structural evolution of the Kumtor region, Tien Shan, Kyrgyzstan. (This volume).

Averkin, Yu.A., 1987. Component deposition from hydrothermal solutions on loss of CO2 (in Russian): Geokhimiya, v. 11, p. 1580-1586.

Bottinga, Y., 1969. Calculated fractionation factors for carbon and hydrogen isotope exchange in the system calcite-carbon dioxide-graphite-methane-hydrogen-water vapor. Geochimica et Cosmochimica Acta., v. 33, p. 49-64.

Bowers, T.S., 1991. The deposition of gold and other metals: pressure-induced fluid immiscibility and associated stable isotope signatures: Geochimica et Cosmochimica Acta, v. 55, p. 2417-2434.

Cooke, D.R., McPhail, D.C., and Bloom, M.S., 1996. Epithermal gold mineralization, Acupan, Baguio district, Phyllipines: geology, mineralization, alteration, and the thermochemical environment of ore deposition: Economic Geology, v. 91, pp. 243-272.

Drummond, S.E., and Ohmoto, H., 1985. Chemical evoluption and mineral deposition in boiling hydrothermal systems: Economic Geology, v. 80, p. 126-147.

Homeniuk, L. 2000. Kumtor gold project, Kyrgyz Republic. CIM Bulletin. March 2000. Larson, P.B., and Taylor, H.P.Jr., 1987. Solfataric alteration in the San Juan Mountains, Colorado: oxygen isotope variations in a boiling hydrothermal environment., Economic Geology, v. 82, pp. 1019-1036.

Lynch, J.V.G., Longstaffe, F.J. and Nesbitt, B.E., 1990. Stable isotope and fluid inclusion indications of large-scale hydrothermal paleoflow, boiling, and fluid mixing in the Keno Hill Ag-Pb-Zn district, Yukon Territory, Canada: Geochemica et Cosmochimica Acta, v. 54, pp. 1045-1059.

McKibben, M, and Eldrige, C. S., 1990. Radical sulfur isotope zonation of pyrite accompying boiling and epithermal gold deposition: a SHRIMP study of the Valles Caldera, New Mexico: Economic Geology, v. 85, pp. 1917-1925

Ohmoto, H. and Lasaga, A., 1982. Kinetics of reactions between aqueous sulfates and sulfides in hydrothermal systems: Geochimica et Cosmochimica Acta, v. 46, p. 1727-1745. Ohmoto, H., 1972. Systematics of sulfur and carbon isotopes in hydrothermal ore deposits. Economic Geology, v. 67, p. 551-578.

Ohmoto, H., and Goldhaber, M.B., 1997. Sulphur and carbon isotopes. In: Barnes, H.L. (Ed.) Geochemistry of Hydrothermal ore Deposits.

Ohmoto, H., and Lasaga, A.C., 1982. Kinetics of reactions between aqueous sulfates and sulfides in hydrothermal systems. Geochimica et Cosmochimica Acta, v. 46, p. 1727-1745.

Sheppard, S.M.F., 1986. Characterization and isotopic variations in natural waters. In: Stable isotopes in high temperature geological processes.

Valley, J.W, Taylor, H.P.Jr., and O'Neil, J.R., (editors). Mineralogical Society of America. Reviews in mineralogy v. 16.

Shmulovich, K.I., Landwehr, D., Simon, K., and Heinrich, W., 1999. Stable isotope fractionation between liquid and vapour in water-salt systems up to 600° C. Chemical Geology, v. 157, pp. 343-354.

Truesdell, A., and Nathenson, M., 1977. The effects of subsurface boiling and dilution on the isotopic composition of Yellowstone thermal waters. Journal of Geophysical Research. V. 82, p.3694-3704.

Truesdell, A.H., 1974. Oxygen isotope activities and consentrations in aqueous salt solutions at elevated temperatures: consequences for isotope geochemistry. Earth and Planetary Science Letters, v. 23, p. 387-396.

Zhang, X., and Spry, P.G., 1994. Calculated stability of aqueous tellurium species, calaverite, and hessite at elevated temperatures: Economic Geology, v. 89, p. 1152-1166.

Zheng, Y.-F., & Simon, K., 1991. Oxygen isotope fractionation in hematite and magnetite: A theoretical calculation and application to geothermometry of metamorphic iron-formations. European Journal of Mineralogy. 3; 5, Pages 877-886