of 26
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.

923 | Minerals | Geology



Publish on:

Views: 28 | Pages: 26

Extension: PDF | Download: 0

  0361-0128/07/3685/923-26 923 Introduction H YDROTHERMAL alteration records the effects of fluid-rock in-teraction, and these effects are expressed as compositionalchanges that can be recognized in the geochemistry of therocks. Where compositional changes define gradients that canbe related to processes that form ore deposits, they potentially provide vectors toward sites likely to contain precious metalmineralization. Zoned hydrothermal alteration surroundingore deposits is the mineralogical expression of temperatureand compositional gradients centered on those ore deposits.  Whole-rock geochemical studies have been widely appliedto volcanic-hosted massive sulfide (VHMS) deposits (Gem-mell and Large, 1992; Whitford and Ashley, 1992; Callahan,2001; Gemmell and Fulton, 2001; Large et al., 2001a-c) andalso orogenic gold deposits (Eilu et al., 1997), particularly toevaluate compositional changes associated with hydrothermalmineral zonation and identification of gradients that provide vectors toward mineralization. Hydrothermal mineral zona-tion develops around precious metal epithermal quartz ±cal-cite ±adularia ±illite deposits (e.g., Buchanan, 1981; Whiteet al., 1995; Simmons et al., 2005) and in analogous environ-ments in active geothermal systems (Browne and Ellis, 1970;Hedenquist and Henley, 1985; Henley, 1985; Simmons andBrowne, 2000); however, only a few studies of epithermal de-posits or modern geothermal systems exist that examine the whole-rock geochemistry of altered rocks (Madeisky, 1996;Sherlock, 1996; Simpson and Mauk, 2000; Simpson et al.,2001; MacIntosh and Simmons, 2002; Shikazono et al., 2002;John et al., 2003; Simpson et al., 2003; Leavitt and Arehart,2005; Gemmell, 2007; Mauk and Simpson, 2007).Here, we evaluate whole-rock geochemical data as a poten-tial tool for exploration of epithermal Au-Ag deposits charac-terized by K-bearing hydrothermal alteration of enclosing volcanic host rocks. We provide details of the procedures forcalculating mass transfer and evaluating associated alterationminerals for the El Peñón epithermal deposit, northernChile. The same techniques are applied to data for rocks sur-rounding epithermal deposits at Sleeper, Nevada, UnitedStates, and Mount Muro, Kalimantan, Indonesia, to demon-strate their general utility in evaluating altered volcanic rocksuites ranging from mafic to felsic composition. These tech-niques provide information about both geochemical and min-eralogical gradients, which define vectors toward epithermalAu-Ag mineralization.  Whole-Rock Geochemical Techniques for Evaluating Hydrothermal Alteration, Mass Changes, and Compositional Gradients Associated with Epithermal Au-Ag Mineralization I AN  W  ARREN , †, * S TUART F. S IMMONS , AND J EFFREY L. M AUK School of Geography, Geology and Environmental Science, University of Auckland, Auckland Mail Centre, Auckland 1142, New Zealand  Abstract  Volcanic rocks hosting quartz ±calcite ±adularia ±illite epithermal deposits undergo elemental masschanges associated with K metasomatism, K-H metasomatism, and H metasomatism that are developed pro-gressively upward and outward from the site of mineralization as hydrothermal fluid ascends, boils, and cools.Resultant hydrothermal mineral zonation shows increasingly K-rich mineral assemblages with proximity todeposits, thus patterns of K enrichment provide a vector toward those deposits. We describe whole-rock geo-chemical techniques for identifying these patterns and for evaluating attendant hydrothermal mineral zonation. Whole-rock geochemical anomalies are evaluated by calculating mass changes associated with hydrothermalalteration, using a modified version of Gresens’ (1967) equation to compare the composition of altered rocksto fresh-rock equivalents. Hydrothermally altered rocks most affected by K metasomatism will be character-ized by the largest K gains and, generally, Na and Ca losses. Mass changes associated with K metasomatism arealso evaluated graphically using plots of molar (2Ca + Na + K)/Al versus molar K/Al. Since molar values areused to construct the plot, compositions of altered rocks can be compared to the compositions of primary andsecondary K-, Na-, Ca-, and Al-bearing minerals that are located in the same compositional space, allowingidentification of important hydrothermal minerals (e.g., adularia, illite, smectite) and alteration processes thatare reflected in trends from fresh-rock compositions toward the compositions of hydrothermal minerals. Theintensity of K metasomatism, encompassing both K gains and Na and Ca losses, can be represented by theslope of the line between an altered rock composition and the srcin (i.e., the molar K/(2Ca + Na + K) value). Determinations of mass changes in altered rocks surrounding selected epithermal deposits demonstrate thepredominance of K metasomatism proximal to, and commonly increasing in intensity toward, mineralized veins. Comparisons of K mass changes to trace element concentrations indicate that the area affected by Kmetasomatism is more extensive (100s to 1,000s of meters) than that containing anomalous concentrations of precious metals, base metals, and pathfinder elements (10s to 100s of meters); therefore, whole-rock geo-chemical techniques potentially extend the area over which geochemical targeting may be effective. Data fromthis study show that intensity of K metasomatism (molar K/(2Ca + Na + K) values) and concentrations of pre-cious metals and pathfinder elements increase toward ore and are greatest proximal to ore, so that pathfinderelement and whole-rock geochemical anomalies are complementary. † Corresponding author: e-mail, ian.warren@meridiangold.com*Current Address: Meridian Gold Company, 9670 Gateway Drive, Suite200, Reno, Nevada 89521. ©2007 Society of Economic Geologists, Inc. Economic Geology,  v. 102, pp. 923–948  Mass Transfer and Hydrothermal Alteration in Epithermal Environments Giggenbach (1984, 1988) described the effects of masstransfer associated with alteration of volcanic rocks of felsiccomposition in a subaerial hydrothermal system (Fig. 1).Potassium metasomatism, accompanied by silicification anddissolution of Ca-, Mg-, and Na-bearing phases occurs in up-flow zones where liquid ascends, boils, and cools. In order tomaintain equilibrium between hydrothermal fluid and wallrock, K + is removed from the liquid as it ascends and cools,forming K-feldspar, K-mica (pure end member of illite) andK-rich clays (illite and interlayered illite-smectite). Their sta-bilities are sensitive to pH, and exsolution of CO 2 due to boil-ing, in particular, causes precipitation of adularia (Browneand Ellis, 1970; Simmons and Browne, 2000). Quartz solubil-ity decreases as temperature decreases, so that K metasoma-tism should be accompanied by silica deposition (Giggen-bach, 1984; Fournier, 1985). Outward from the upflow zone,decreasing temperature, sub-boiling conditions, and steam-heated waters lead to increasing CO 2 fugacity and acid gen-eration. There, H metasomatism favors cation leaching to thefluid producing Al-rich mica and clay assemblages and fixingof Ca into calcite with CO 2 from the fluid (Simmons andChristenson, 1994). At the periphery of the hydrothermal system, where mete-oric fluid recharges the system, the opposite trends are ob-served, with descending and heating fluid producing Mg andCa metasomatism. As fluid descends and temperature in-creases, dissolution of early formed K-bearing phases adds K + to the fluid, and Mg 2+ and Ca 2+ are added to altered rock,forming calcite and chlorite; albite is conserved or forms in as-sociation with weak Na metasomatism. Away from the ascend-ing hydrothermal plume, rock-dominated alteration is charac-terized by minimal mass transfer of rock components, andsrcinal minerals are converted to stable hydrothermal miner-als that incorporate variable additions of CO 2 , S, and H 2 O.These effects explain the zoned patterns of hydrothermalalteration found around precious metal quartz ±calcite ±adularia ±illite epithermal orebodies (Fig. 1). With increas-ing distance (1–>50 m) from gold-bearing veins, adulariagives way to illite as the main K-bearing mineral accompany-ing quartz. Upward and outward, shallow and peripheral clay-carbonate alteration comprises mixed-layer illite-smectite,smectite, pyrite, and carbonate, while deep and regionalpropylitic alteration includes quartz, adularia, albite, illite,chlorite, calcite, pyrite, and epidote (e.g., Simmons et al.,2005). As orebodies are approached, changes in the stability of the dominant K-bearing mineral from smectite to mixed-layer illite-smectite to illite to adularia correspond to gradi-ents of increasing K. Evaluation of Mass Transfer in Hydrothermally Altered Rocks Mass transfer studies of hydrothermally altered rocks eval-uate the chemical gains and losses produced by conversion of primary minerals to secondary minerals (Leitch and Lentz,1994). Direct comparisons between the compositions of al-tered rocks and equivalent fresh rocks are not possible if al-teration is accompanied by significant mass or volumechanges (Gresens, 1967; Babcock, 1973; Grant, 1986;MacLean and Kranidiotis, 1987; Appleyard, 1991; MacLean,1990; MacLean and Barrett, 1993). To overcome this obsta-cle, Gresens’ (1967) approach assumes that one or severalelements are immobile during hydrothermal alteration. Be-tween rocks from the same unit, changes in the concentra-tions of immobile elements reflect volume changes that,once accounted for, allow the altered-rock geochemistry tobe compared to that of an equivalent fresh rock. A numberof workers have since used this approach to evaluate thecompositional changes associated with hydrothermal alter-ation (e.g., Grant, 1986; Appleyard, 1991; MacLean andKranidiotis, 1987; MacLean, 1990; MacLean and Barrett,1993). The method of MacLean and Kranidiotis (1987),MacLean (1990), and MacLean and Barrett (1993) is funda-mentally the same as Gresens’ (1967) and uses immobile el-ement ratios just as in the graphical isochon method of Grant(1986). Calculation of mass change does not require density measurements; however, without these measurements, vol-ume changes can only be estimated using immobile elementratios. The fundamental equation of Gresens (1967) used to cal-culate mass transfer is: ∆  X = a {[F  v  ( S B  /S A ) ×  X  B ] – X  A },(1) where ∆  X = mass change for component X, a = starting mass,arbitrarily 100 g for concentrations in wt percent or 1 metric924 WARREN ET AL. 0361-0128/98/000/000-00 $6.00 924 clay carbonate pyrite propylitic quartz illite adularia pyrite propylitic quartz chalcedony adularia carbonates pyrite, Au-Ag, Ag-Pb-Zn lattice textures, crustiform-colloform banding  P 0-10 m        P P 50-100 m      5      0    -     1     0      0     m F IG .1. Schematic illustration of the scale of alteration and zonation of hydrothermal minerals surrounding epithermalAu-Ag deposits formed from near-neutral pH and reduced solutions (Simmons et al., 2005). Large-scale alteration zonationis shown on the left, and alteration zonation proximal to orebodies is shown on the right.  ton (t) for concentrations in parts per million, F  v  (vol factor)= ratio of volume of the altered rock to the fresh rock, S B  /S A = the ratio of the specific gravity of the altered rock to thefresh rock, X  B = concentration of component X in the alteredrock, and X  A = concentration of component X in the freshrock.Calculated mass change values will be in wt percent (g/100g) or parts per million (g/t), depending on the units of X  A,B ,after a is moved to the left side of the equation. In order to solve equation (1), ∆  X or F  v  must be known. F  v  can be determined by setting ∆  X = 0, the solution if an ele-ment is immobile. An average F  v   value can be calculatedusing multiple immobile elements, and the F  v   values calcu-lated for each immobile element should ideally show no vari-ation or only slight variations. Immobility can be tested withbinary plots of immobile element concentrations that willform a straight line, ideally through the srcin, for each ge-netically related suite of rocks. The straight line reflects con-stant ratios between the immobile elements; values are offsettoward or away from the srcin relative to the fresh-rock valuedue to mass gain or loss, respectively (MacLean and Kranidi-otis, 1987; MacLean, 1990; MacLean and Barrett, 1993).After determining immobile elements, F  v  is solved for by set-ting ∆  X = 0 and rearranging: X  A = F  v  (S B  /S A ) ×  X  B ,(2) X  A  / X  B = F  v  (S B  /S A ),(3)andF  v  = X  A S A  /X  B S B ,(4) where the X terms refer to the concentration of the selectedimmobile element(s) in the fresh (A) and altered (B) rocks. When the calculated F  v   value of equation (4) is substitutedback into equation (1) to solve for ∆  X of mobile elements, theinverse of the density ratio (S A  /S B , eq 4) required to solveequation (1) is implicit in the F  v  term. Density measure-ments, therefore, cancel out, and the mass transfer calcula-tion procedure is essentially that of MacLean and Kranidiotis(1987), MacLean (1990), and MacLean and Barrett (1993)and is mathematically equivalent to the isochon method of Grant (1986): ∆  X = [(X  Ai  / X  Bi ) ×  X  B ] – X  A ,(5) where ∆  X = mass change for mobile component X in g/100 gor g/t (the denominator, 100 g or 1 t, is a from eq (1) that ismoved to the left side of the equation), X  Ai  / X  Bi = the ratio of the immobile element concentration of the fresh rock to thealtered rock (or the average of several immobile element ra-tios), X  B = concentration of mobile component X in the al-tered rock, and X  A = concentration of mobile component X inthe fresh rock. Molar element ratios: mass transfer and associated alteration mineralogy In addition to calculations of absolute mass changes usingthe modified Gresens’ (1967) equation (eq 5), mass transfereffects can be evaluated graphically and related to associ-ated hydrothermal alteration minerals using molar elementratios calculated from whole-rock geochemical data, whichisfundamentally the same as the Pearce element ratio tech-nique of Stanley and Madiesky (1994). The effects of volumechanges are eliminated by comparing ratios with the same de-nominator, and the expression of geochemical analyses asmolar values allows comparison to mineral stoichiometries(Stanley and Madeisky, 1994; Madeisky, 1996). Plots of (2Ca+ Na + K)/Al versus K/Al molar ratios provide a graphicalmeans for evaluating the degree of K metasomatism, K, Ca,and Na depletion, and Ca metasomatism affecting alteredrocks (Fig. 2; Madeisky, 1996). Where CO 2 (C) analyses areavailable, the effects of Ca metasomatism (calcite), which canpartly mask the effects of K metasomatism, can be removedby using the molar ratio (2(Ca-C) + Na + K)/Al (Madeisky,1996) and assuming all analyzed CO 2 (C) to be contained incalcite. With the use of molar values, the compositions of al-tered rocks can be compared to the composition of hy-drothermal minerals, allowing geochemical trends to be re-lated to alteration mineralogy so that mineral zonation can beevaluated. The degree of K metasomatism can be determinedfrom the slope of the line between the srcin and each datumpoint by dividing the K/Al molar ratio by the (2Ca + Na +K)/Al molar ratio (Stanley and Madeisky, 1994; Madeisky,1996). If Al is immobile during hydrothermal alteration, themagnitude of displacement from a fresh-rock composition isproportional to the actual K, Na, and Ca transferred duringhydrothermal alteration (Stanley and Madeisky, 1994). Thisplot not only shows the important mass transfer processes andattendant alteration mineralogy associated with ascending,boiling, and cooling neutral pH and reduced hydrothermalfluid (Giggenbach, 1984), it provides the means to identify geochemical and mineralogical gradients centered on ep-ithermal orebodies. Mass transfer processes and associated alteration that canbe recognized using the (2Ca + Na + K)/Al versus K/Al molarratio plot are illustrated in Figure 2. Rocks containing only plagioclase (albite-anorthite end members and solid solu-tions) will plot at a (2Ca + Na + K)/Al value of 1 along the x-axis, rocks containing only K-feldspar and/or biotite will plotdirectly above this point at a K/Al value of 1, and kaoliniteand chlorite, Al-bearing phases that contain no K, Na, or Ca, will plot at the srcin. The proportion of Al contained in chlo-rite, rather than the Fe or Mg content, determines the de-gree to which its presence causes data points to shift towardthe srcin. Phases that contain no K, Ca, Na, and Al techni-cally plot at the srcin but their occurrence does not affect where samples are located in this compositional space (e.g.,the amount of pyrite has no affect on where the data willplot). The line extending from 1,0 to 1,1 represents the rangeof feldspar compositions and mixtures with biotite. Fresh fel-sic rocks plot on or near this line as determined by their K/Almolar ratio, and fresh intermediate to mafic rocks plot pro-gressively to the right of this line in proportion to theamounts of Ca-bearing mafic minerals they contain. In Fig-ure 2, data from volcanic rocks of the Taupo volcanic zone(Graham et al., 1995) are shown to illustrate the area of theplot typical for fresh rocks; decreasing values of molar K/Aland increasing values of molar (2Ca + Na + K)/Al reflect therange of igneous rock compositions from basalt through torhyolite. The line of slope 1 that extends from the srcin to1,1, represents rocks that contain no Ca or Na. Along this MASS CHANGES AND COMPOSITIONAL GRADIENTS OF ALTERATION ASSOCIATED WITH EPITHERMAL Au-AgMINERALIZATION 925 0361-0128/98/000/000-00 $6.00 925  line, the dominant K-bearing phase, or proportion of phases,can be assessed based upon the K/Al molar ratio; K/Al is 1 forK-feldspar (adularia) and/or biotite, 0.33 for K-mica, and<0.33 for illite. The molar K/Al value decreases with increas-ing amounts of interlayered smectite, and samples with molarK/Al <0.2 to 0.33 that contain smectite will plot away from theline of slope 1 in proportion to the amount of Ca and Na con-tained in smectite. Samples with molar K/Al values <0.2 to0.33 that plot along the line of slope 1 likely contain kaoliniteand/or Al-bearing chlorite. The minerals plotted in Figure 2can be related to each other with generalized equations de-scribing the alteration and mass transfer associated with Kmetasomatism:NaAlSi 3 O 8 + K + → KAlSi 3 O 8 + Na + ,(6)albite K-feldsparNa 2 CaAl 4 Si 8 O 24 + 2H + + 2K + → KAlSi 3 O 8 + plagioclaseK-feldsparKAl 3 Si 3 O 10 (OH) 2 + 2SiO 2 + 2Na + + Ca 2+ ,(7)K-mica quartz3KAlSi 3 O 8 + 2H + → KAl 3 Si 3 O 10 (OH) 2 + 6SiO 2 + 2K + ,(8)K-feldspar K-mica quartzand2KAl 3 Si 3 O 10 (OH) 2 + 3H 2 O + 2H + → K-mica3Al 2 Si 2 O 5 (OH) 4 + 2K + .(9)kaoliniteIf Al is assumed to be immobile, rocks affected by K meta-somatism plot above and to the left of fresh-rock composi-tions, with molar K/Al values equal to or greater than thestarting composition, reflecting K addition accompanied by loss of Ca and Na (eqs 6 and 7). Conversion of K-feldspar toK-mica (illite), mixed-layer illite-smectite, and smectite low-ers molar K/Al values of the rock (eq 8), but this does notnecessarily mean lesser effects of K metasomatism becausethose effects include the degree of Ca and Na loss relatingto the destruction of primary plagioclase and its replace-ment by K-bearing phases. Displacement toward the srcinreflects removal of K, Na, and Ca associated with the for-mation of kaolinite and/or Al-bearing chlorite. Displace-ment to the right of the starting composition reflects Caand/or Na addition; in the near-neutral pH epithermal envi-ronment under consideration such a displacement mostlikely reflects Ca addition in the form of calcite because Nametasomatism can only contribute minor Na to the rock(e.g., Giggenbach, 1984, 1988). Weak Na metasomatism oc-curs in deep and peripheral parts of hydrothermal systems(Giggenbach, 1984; Simmons and Browne, 2000); strongerNa metasomatism may occur near intrusions where de-scending liquid, especially saline liquid, moves up steeptemperature gradients (Dilles and Einaudi, 1992; Sorensonet al., 1998).926 WARREN ET AL. 0361-0128/98/000/000-00 $6.00 926 RhyoliteDaciteAndesiteBasalt    K g   a   i  n ,  N  a ,  C  a l  o  s  s    K ,  N  a ,  C  a g   a   i  n N  a , C a  g a i  n K  l  o s s   K, N a, C a l o s s      K     /      A     l     (      m    o      l    a     r     )   0.2 0.4 0.6 0.8 1 1.2 1.4 2Ca+Na+K/Al (molar) KaoliniteChlorite Ca,Na-smectite AlbitePlagioclase K-micaK-feldspar BiotiteIlliteInterlayeredillite-smectite      P     l    a     g       i    o     c      l    a     s     e     +     A     l     b      i     t     e     +     K   -     f    e      l     d     s     p      a     r   C  a ,  N  a l  o  s   t F IG .2. Molar element ratio plot of (2Ca + Na + K)/Al vs. K/Al for typical fresh volcanic rocks. Mass transfer processesare shown with arrows that vector toward associated alteration minerals. The portion of the plot occupied by typical freshrocks is illustrated with data from fresh volcanic rocks of the Taupo volcanic zone, New Zealand (Graham et al., 1995).
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks