Within A Stratovolcano, E1 Queva, Northwest Argentina: Lead-Silver, Manganese, and Native Sulfur Mineralization
Within A Stratovolcano, E1 Queva, Northwest Argentina: Lead-Silver, Manganese, and Native Sulfur Mineralization
Within A Stratovolcano, E1 Queva, Northwest Argentina: Lead-Silver, Manganese, and Native Sulfur Mineralization
Abstract
relationshipsbetween mineralization,little-eroded
volcaniclandforms,thermal springs,and solfataras 22"-
can be worked out with a minimum of conjecture
and lead to a better appreciationof what is con-
sideredto be one of the most important situations
for ore depositionin the volcaniccycle. It is hoped
that the resulting model will provide a better geo- _...'"'.).
C3 i ,,,,/
metrical basisfor geochemicalstudiesof ore forma-
tion in the "epithermal"environment. I
I
•',•'•"
'FIG.
e/
I S.Antonio
e3ujuy
northwestArgentina,some53 km directlysouthwest
of the regionalcapitolof San Antonio de los Cobres
in an area that, thoughremote,is accessible by both
rail and road (Fig. 1). The mine and zonesof hy-
drothermalalterationin its vicinityare on the flanks
•de
los
Cobres e Salta
:::::::::::::::::::::::::::
•OCITOS
67øW.
I0 kin.
i
i:[•
[•
Andesites
•Less
Dffcites.
[
Dadtes&Rhyodacites.
J
),' Dip and Strike
•, ",,,. Road; Ra'llroad
• Pastos
GraJxles
groupMIOCENE-
red beds. PLIOCENE
• Contour in meters
• Granites SILURIAN
--• Lutites
&Areni•s ORDOVICIAN All--- -4A S[CnO UNœ
NATIVE SULFUR MINERALIZATION 1193
Fro. 2. Geologicalmap and sectionof the Nevados de Pastos Grandesregion, northwesternArgentina. Geology taken
from Vilela (1969) with stratigraphic ages modified after Plan NOA-1 Geo16gico-Minero,Salta.
1194 RICHARD H. $ILLITOE
FIG. 3. The little-eroded eastern side of the Nevados de Pastos Grandes stratovolcano viewed from the road
leadingto Santa Rosa de PastosGrandes(Fig. 2). The peak on the left is Cerro Azufre and the native sulfur
depositis located on the col to its right.
ward flaring of the alteration zone is absent. The rhyodaciteflows. In spiteof an examinationof •
alteration zone outcropsover a maximum vertical altered areas, no intrusive stockswere recognize,
interval of about 1,000 m, between 4,500 and 5,500 althoughthe presenceof dikes and sills is local
m above sea level. The effects of alteration render evident.
identificationof originalrocktypesextremelydifficult First impressions
suggestthat a large part of t!
in parts of the area, but they are thoughtto consist alteration zone has undergone argillic alteratio:
largely of quartz-bearingandesite and dacite to However, thin-sectionand X-ray study of select{
FIG. 4. The main alteration zone on the western flank of the Nevados de Pastos Grandes stratovolcano viewed
from the road north of Pocitos (Fig. 2). The peak in the centeris Cerro Gordo.
NATIVE SULFUR MINERALIZATION 1195
samples of heavilyalteredmaterialindicates
that the The lead-silver mineralization, described more
alterationis partly of the advancedargillic type fully below,is mostwidespread as a llhear belt in
(HemleyandJones,1964),termedsolfataric altera- the vicinity of the E1 Queva mine, not far abovethe
tion by earlierworkers(Burbank•!O50). The hy- lowestaltitudeat whichthealteration zofieoutcrops.
drothermal minerals identified are chiefly dickite, Small podsof similar lead-silve/-mineralizationwere
kaolinite,quartz,alunite,and•pYritewith smaller also encounteredelsewhere in the large alteration
amountsof transparentgypsum(in veinletsup to zone,oneoutcrop being•ear theaccumulation of
20 cm across)and variousclaymineralsof dubious native sulfur on Cerro AzuYre. Stream-sediment
affiliation; pyrophyllitewas searchedfor but not geochemical samples collected:away fromknown
identified. The claysand alunitepreferentiallyre- mineralizationin the large alteration zone gave lead
placeplagioclase phenocrystsbut locallythe whole andzincvaluesashighas2,400and3,900ppm,re-
rock has been completelytransformed. Some of spectively(J. Mancini, unpub. rept.), thereby sub-
these minerals,especiallythe clays and gypsum, stantiatingthe presenceof scattered•ineralization.
locallymay be of supergene origin but, sincethey
commonlyoccurwith unoxidizedpyrite, a hypogene ß Lead-Silver Mineralization'
,
FiG. 5. Talus fragments of perlitic rhyolite coated and cemented by manganese oxides
(black), from the depositon the southwestflank of Cerro Azufre (Fig. 2).
porous,slag-like texture is commonin much of the surface replacementof galena. Postmine gypsur
gray, intensely silicified rock carrying fine-grained chalcanth.
ite, and iron sulfates coat the walls,
sulfides. The silicified rock may be devoid of workingsin the E1 Queva mine.
alumina-bearingmineralsor may containlocal dick-
Manganese Mineralization
ite, in part pseudomorphous after plagioclasepheno-
crysts. A single group of small quartz crystals Six separatesmall depositsof manganeseoxid.
intergrown with barite was found at one locality. are located in the Nevados de Pastos Grandes r
Barite occursas aggregatesof interlocking,white, gion, most of which have been worked in a-sin;
tabularcrystalsin vuggypocketsor as isolatedcrys- way, and other occurrencesprobably exist. All tl
tals embeddedin dickite, alunite, or chalcedony;in- observed manganeseoxide concentrationsare sit•
dividual crystalsattain 4 cm in length. The white ated around the peripheryof the stratovolcano ar
dickiteand cream,white, or yellow alunite commonly are not spatiallycoincidentwith the zonesof hydr•
infilledvuggyopenspace. thermal alteration (Fig. 2).
Pyrite and galenaare the principal sulfides. They Three distinct modes of occurrenceof mangane;
are typically fine grained and commonlyoccur as oxidesare apparent: 1. The manganeseoxides.
disseminationsin silicified rock. Microscopicex- four localitiescement talus, float material, or terra•
aminationof polishedsectionsof ore also revealsthe gravels that overlie volcanic.rock and in whk
presenceof complexintergrowthsof silver sulfosalts, Ordovicianphyllite,Quaternaryandesiteand daci
includingstephaniteand pyrargyrite,and minor chal- or perlitic rhyolite are the dominant fragrn/m!
copyrite and sphalerite,the latter as inclusionsin Manganeseoxides are colloformand coat and o
galena. mentfragmentsin the alluvialandcolluvialmateria
Postglacial supergeneoxidation, in an acid en- (Fig. 5). 2. In the occurrence near Olacapatotl
vironmentabove the presentwater table, reachesa manganese oxidesoccuras veinletsand dissemin:
maximumdepth of about 30 m and has resultedin tions in a dacite flow; and 3. In the deposit ;
widespreadtransportedgoethiteand jarositederived Boratera Antuco the manganese oxides occur •
from pyrite, and anglesiteafter galena. Crusts of friable beds and lenses admixed with subordina
anglesiteenvelopmassesof galena in the surface goethitein a travertinehorizonthat constitutesa
zone. One sampleof beudantiteof probablesuper- old thermal-springapron. Both manganese oxid•
gene origin was also identified and minor amounts and travertine havebeenexploited. The manganes,
of covellite were visible in polished sectionsas a bearingtravertineis overlainby a mudflowdeposi
NATIVE SULFUR MINERALIZATION 1197
a secondtravertinehorizonwith tracesof manganese, TABLE 2. Partial Analysis of Water from Boratera Antuco
Spring (in ppm)
and finally a thin tuff at the presentsurface,the
wholesectionapproximately 20 m in thickness. Ca 14.2 HCOa 374.7
A selection
of manganeseoxidesfromthedeposits Mg 6.6 SO4 185.8
Na 646.0 C1 950.0
wasX-rayedand the resultsconfirmed
that crypto- K 84.0 NOa 9.0
melanewas the principalmineral and that some Cu 0.2
amorphousmaterial was also present. In all de- As 1.0 pH 6.9
posits,transparent,
botryoidalchalcedony
is present,
as a surfacecoatingto manganese
oxides. Goethite
is alsolocallypresent,particularlyat BorateraAn- borateswith the Boratera Antuco springand spring
tuco, and gypsum and yellowish-greenmontmoril- apron corroboratesa volcanic source for borates in
lonite occur in the depositnear Olacapato. At the salaresof the Puna, where they are exploitedat
Boratera Antuco, pale-green,fibrous calcite and a several localities. A partial analysisof the spring
borate,ulexite, fill vugs in the manganese oxides. water (Table 2) shows that it is not bicarbonate-
Samplesof travertinegave an X-ray patternfor rich as might be expected(White, 1967) but rela-
calcite. tively enrichedin Na and C1.
Five samplesof manganeseoxides from the Althoughtravertineaccompanies manganeseoxides
manganeseshowswere analyzedto determinetheir only at Boratera Antuco and at the Antuco show
copper,zinc, lead, silver, and arseniccontentswhich to the south, it is consideredthat all the manganese
are listedin Table 1. Althoughcopper,lead, and oxide deposits around the Nevados de Pastos
silverare consistently low, the zinc contentreaches Grandesstratovolcano are the productsof once-active
over0.6 percentin two samples from the depositon thermalsprings. At severallocalitiesthe manganese
the southwestside of Cerro Azufre, and arsenic oxide concentrations are now situated a few meters
exceeds1 percentin two samplesfrom Boratera abovethe presentvalleybottomsbut they outcropat
Antuco. Comparableheavy metal contentshave the ground surface,their original site of deposition.
beendescribedin manganeseoxidesassociated with They must thereforehave been depositedin post-
thermal-springapronsin the westernUnited States glacial times and after all volcanicactivity (except
(Hewett et al., 1963) andare probablyattributable for the depositionof the thin tuff at, and in the im-
to adsorbtionby the manganese oxidesof elements mediatevicinityof, BorateraAntuco) in the general
from solution. The highestzinc contentnotedby region had ceased. The main period of manganese
theseauthorsis, however,only0.3 percent.Hewett depositionhas been followedby only minor stream
et al. (1963) also recorded0.07 percentboron in incision. Consequently, a relativelyrecentperiod of
thermal-springaprons,which clearly correspondsmanganesedepositionis favored, which at Boratera
to the borates identified at Boratera Antuco. Antuco is still continuing,apparentlyat a reduced
At BorateraAntuco,a sluggishspringwith a rate.
wintertemperature of 7øC issuesonly severaltens The manganesedepositsare clearly similar to
of metersupslopefrom the old manganese-bearing others on the Puna of Argentina (Angelelli'et al.,
thermal-spring
apron. This apronclearlyrepresents 1970) and in northernmostChile (Ruiz et al., 1965;
previousactivityfrom the samespringsystem,at Cruzat Ossa, 1970). In the Chilean occurrences,
a slightlyhigherlevel prior to streamentrenchment. manganeseoxides occuras superficialaccumulations
Borates,recoveredcommercially, and minor manga- depositedfrom thermalspringssurroundingQuater-
neseandiron oxidesare, in fact,precipitated by the nary volcanoes. Hewett and Fleischer (1960) and
watersfrom the presentspring. The association of Hewett et al. (1963) have also describedmanganese
oxide deposits,commonlyaccompanied by travertine,
TABLE1. Partial Analysesof Mn Oxides,E1 QuevaArea at activeand inactivethermal springsin the western
(in ppm, exceptwherestated) United States.
1 { l Fumaroles
0
Solar with halite '"• '<:'"'
:--:_•__-_
-:•:_-_-•_-_-_-r•:
Subvolcanicbasement--'
.... ------• + + . ----•----•--:--•--••--•_--:
......
-z .... zz l•_+++
-F +4- i---------
..... -
1 z s K•. ?_/•orphyry
.... ' copper
stock
.:.;-;.:;:..'.:.
Advanced
argillicalteration
andsilic/h•ation.• Nativesulfurdeposit.
III Precious and base metal veins and pipes.
e Thermal springs, manganese deposits and travertine aprons.
Ft. 6. Diagram showingthe interrelationshipsbetween hydrothermal alteration and precious and
base-metal,manganeseoxide and native sulfur mineralization associatedwith a calc-alkaline stratavol~
cano,as interpretedfrom the Nevadosde Pastas Grandesregion, northwesternArgentina. The subjac-
ent porphyry copper-bearingstock may completethis scenario (Sillitoe, 1973).
floor of a U-shapedvalley cut 400 to 500 m below had largely ceasedprior to alteration-mineralization
the surroundingdeposittonalslope of the volcanic and only the thin bed of tuff at Antuco and possibly
edifice. Therefore .sulfide depositiontook place someandesitcflowscappingthe highestgroundout-
beneatha cover of only 400 to 500 m of volcanic sidethe alterationzonesseemto postdatealteration.
rocks, and subordinate sulfide mineralization oc- Sincealterationand sulfidemineralizationare largely
curred even closer to the surface. The sulfide min- preglacialand the peripheralmanganeseoxide ac-
eralizationis an integral part of the main hydro- cumulations are demonstrably postglacial,a time lag
thermal alterationzone, as evidencedby the presence of perhaps1 or 2 m.y. musthaveexistedbetweenthe
of dickitc and alunite in both altered volcanics and, two events,the secondof which is still weakly con-
intergrownwith lead-silversulfides. The alteration tinuing.
and mineralizationappear to have been emplaced It is believedthat the mineralizingfluidsresponsi-
beneathand marginal to the vent regionsof the vol- ble for generation of lead-silver ore and advanced
canic conesat a level beneaththat characterizedby argillic alteration precipitatednative sulfur as they
native sulfur deposition. The Cerro Azufre sulfur approachedthe volcanicvent regions,where they
depositwas probablyformedcloseto a vent on the were presumablyrepresentedby fumarolesand sol-
Cerro Azufre cone. The sulfide mineralization is fataras. A gaseousphasewas apparentlygenerated
evidentlytransitionalupward to native sulfur min- in the surficial zone and presumablycarried traces
eralization,sinceminor native sulfur accompanies
the of metals (White and Waring, 1963). Judging.by
main lead-silver ore and minor lead-silver mineraliza- the presenceof advancedargillic alteration,min-
tion occurs near .the native sulfur deposit and eralizingfluidsat the levelsobservedpossessed a low
both native sulfur and lead-silverdepositsare en- pH and were capableof extremehydrolyticleaching
velopedby the samealterationtype. of bases(Hemley and Jones,1964; Hemley et al.,
The manganeseoxide accumulationsare located 1969). The silicifiednodes,characterizedby almost
peripherallywith respectto alterationand sulfide total leachingof both basesand alumina, could be
mineralization in the center of the edifice, were consideredas the conduits through which most of
precipitatedby thermal-springwaters from which the fluids ascended. Sulfideswere precipitateddur-
they scavenged metals,and are accompanied at An- ing this base-leaching
process. The fine grain size
tuco by travertineaprons. The minor manganese of the sulfidesand the aphaniticnature of the silica
oxide crusts in the alteration zone provide a link suggesta rapid dumpingof materialfrom the trans-
betweenthe centralsulfideand peripheraloxidemin- portingfluidsprobablydue to an abruptdecreasein
eralization. temperatureas the surfacewas approached,a mecha-
These mineralizationevents represent the final nism that would also accountfor barite deposition
phasesin the activity of the Nevadosde Pastas (Holland, 1967). Althoughthe mineralizingfluids
Grandes stratavolcano,correspondingto what is probablyderivedat leasta part of their metalsand
generallyassignedto the fumarolicstage. Lava sulfur from a magmaticsource,they were likely to
emissionleadingto construction cones have beencomposed
of the volcanic largely of heated,low-salinity
NA TI VE SULFUR MINER.4LIZ.4 TION 1199
meteoricwaters (cf., Schmitt, 1950a, b; White, Fleischer(1960) and Hewett et al. (1963). Hence
1955). Stable isotopestudiesof Miocene,"epi- it is temptingto considerthat porphyrydepositsare
thermal",preciousmetaldepositsof volcanicaffilia- commonlyfringed and cappedby zones of "epi-
tion in Nevada have recently confirmedthe pre- thermal" preciousand base metal veins (Sillitoe,
ponderance waterin the orefluid (Tay- 1973), whichin somecasesincludeor passoutward
of meteoric
lor, 1973). In the caseof the similarCaudalosa to manganese veinsin depthandmanganiferousther-
Peru, however,mag- mal-springapronsat surface.
silverdepositin south-central
matic water has been shown to be an important Bicarbonate-type springs,associatedwith traver-
process(Sawkinsand Rye, tine aprons,commonlyissuein closeproximity to
factorin the metallization
1974). "epithermal"basemetal vein depositsas at Creede,
The marginalmanganese mineralizationwas ap- Colorado,althoughthey are not directly responsible
parentlyproducedduring the debouchment at sur- for ore deposition(White, 1967). This may be due
face of somewhatdifferent fluids. Thermal-spring to displacement of the springorificesduringerosion,
waters,of the sort believedto have formed manga- so that their originalmarginalpositions,exemplified
neseoxide deposits,are normallyof the Na-Ca- at E1 Queva and in northernmostChile (Cruzat
HCOa type accordingto White (1967). He has Ossa, 1970), are lost.
also suggested that separationof a more fugitive, The observationthat E1 Queva sulfide ore was
CO2-richvaporphasefrom higherdensity,metal- formed 400 to 500 m beneath surface and that minor
bearingbrinesand its subsequent admixturewith mineralization occurred at even shallower depths
the meteoriccircul'ation
couldexplainsuchbicarbon- confirmsthe suggestionsof earlier workers that the
atespringwaters.In theNevadosde PastosGrandes "epithermal"mineralizingenvironmentcan be as
region,however,the presentBorateraAntucospring shallowas 300 (Lindgren, 1933; Burbank,1950) to
water is more a dilute Na-C1 type, but this water 1,000 m (Hewett, 1964).
may well be distinctfrom that whichprecipitated the It is now widely acceptedthat important "epi-
maintravertine-manganese oxideaccumulation; it is thermal" mineral depositsdevelopbeneathvolcanic
probablylittle more than meteoricwater that has calderaswith ores emplacedduring final phasesof
percolatedthroughthe volcanicpile. the caldera cycle in ring and radial fracturesgen-
erated during cauldroncollapse(Smith and Bailey,
Overview 1968; Albers and Kleinhampl, 1970; Lipman and
Steven,1970; Stevenet al., 1974). It is clear from
It was recentlysuggested (Sillitoe, 1973) that
the evidenceat E1 Queva, however,that the "epi-
high-level,advancedargillic alterationand sulfide
thermal" environmentcan also exist in the upper-
deposition like that in the E1 Quevaarea can take
most parts of a stratovolcanoat which a calderahas
placein volcanic superstructures aboveconsolidating,
cogeneticmagmabodiesin which porphyrycopper not formed. The rarity of suchdepositsin uncol-
deposits are generated(Fig. 6). If this is so, then lapsededificesis readily explicablein terms of their
the hydrothermalphenomenarepresentedat E1 susceptibilityto erosion. As in the caseof mineral-
Quevaare probablyproducedby fluidsderivedfrom ized calderas,however,ore depositionconcludesthe
the grossdilutionof high-density magmaticbrines volcanicactivity and is localizedby zonesof struc-
of thetypethatgiveriseto potassium silicatealtera- tural weakness. It would seemthereforethat "epi-
thermal"ore depositionin the upperparts of a vol-
tion, with suprajacent, convectively-circulating
groundwaters(Sheppardet al., 1971). The possi- canicedificeis dependenton processes accompanying
the consolidationof residualmagmachambersin the
bilitytherefore existsthattheE1 Quevaalterationis
transitional downward to sericitic alteration in the roots of the edifice or in the underlying basement
upper parts of a porphyrycopper-bearing stock (Fig. 6) and is not controlledexclusivelyby the
peculiaritiesof calderadevelopment. It is interest-
(Sillitoe, 1973; fig. 1).
In this regardit is instructivethat on the propyli- ing to notethat few examplesof Quaternarycalderas
tic or little-alteredfringes of certain porphyry cop- in the central Andes have been recognized.
per-typedeposits,
withinand beyondthe lead-zinc- Acknowledgments
silverzone (Jerome,1966), as at Bisbee,Arizona,
and Butte, Montana (Meyer et al., 1968), veins I am most grateful to St. J. Mancini, then of Plan
carryingmanganese carbonatesand silicatesoccur NOA-1 Geo16gico-Minero,Salta, Argentina for an
that couldbe considered
as deeperequivalentsof the introductionto the geology of the E1 Queva area
peripheralmanganese oxide accumulations in the and to other geologists,especiallyDr. C. Lurgo, of
Nevadosde PastosGrandesregion. As notedabove, the sameorganizationfor usefuldiscussions.Chemi-
a geneticconnection betweenmanganiferous veins cal analyseswere carried out in the laboratoryof
andhot-spring apronswasadvanced by Hewettand Plan NOA-1, Salta under the supervisionof Ing.
1200 RICHARD H. SILLITOE
Steven, T. A., Luedke, R. G., and Lipman, P. W., 1974, -- 1969,Descripci6ngeo16gicade la Hoja 6c, San Antonio
Relations of mineralization to calderas in the San Juan de los Cobres, Provincias de Salta y Jujuy: Buenos
volcanic field, southwestern Colorado: U.S. Geol. Sur- Aires, Direc. Nac. Geologia Mineria Bol. 110, 66 p.
vey Jour. Research,v. 2, p. 405-409. White, D. E., 1955, Thermal springs and epithermal ore
Taylor, H. P. Jr., 1973, O•8/Ox6evidencefor meteoric-hydro- deposits: Eco•r. GEOL.,50th anniv. vol., pt. 1, p. 99-154.
thermal alteration and ore deposition in the Tonopah, -- 1967, Mercury and base-metal depositswith associated
Comstock Lode and Goldfield mining districts, Nevada: thermal and mineral waters, in Barnes, H. L. ed., Geo-
Eco•r. Gv.oL., v. 68, p. 747-764. chemistry of hydrothermal ore deposits: New York, Holt,
Turner, J. C. M., 1970, The Andes of northwesternArgen- Rinehart and Winston, Inc., p. 575-631.
tina: Geol. Rundschau,v. 59, p. 1028-1063. and Waring, G. A., 1963, Data of geochemistry,6th
Vilela, C. R., 1953, Los perlodos eruptivos en la Puna de ed., chap. K: Volcanic emanations: U.S. Geol. Survey
Atacama: Assoc. Geol. Argentina Rev., v. 7, p. 5-36. Prof. Paper 440-K, p. K1-K29.