Per. Mineral. (2002), 71,2,111-125

PERIODICO di MINERALOGIAestablished in 1930

http://go. to/permin

An International Journal ofMINERALOGY, CRYSTALLOGRAPHY, GEOCHEMISTRY,ORE DEPOSITS, PETROLOGY, VOLCANOLOGY

and applied topics on Environment, Archaeometry and Cultural Heritage

Electrochemical and electrokinetic aspects of porphyry copper systems:experiments and field observations

PETER MOLLER and FRANK TIMM

GeoForschungsZentrum Potsdam, Telegrafenberg, D-14473 Potsdam

Submitted, March 2001 - Accepted, July 2002

ABSTRACT. - Experimental field and laboratorystudies of porphyry copper systems indicate thatrecent electrochemical processes occur in theenvironment of large, vertically extended,semiconducting sulphide bodies. Vertical differencesin oxygen fugacities cause specific electrochemicalreactions around the sulphide mineralization,indicated by the dipole system with a negative selfpotential at the earth's surface near the deposits. Asa result of electrochemical reactions, fluids withrelatively high pH are formed at the cathode (top),while low pH fluids are formed at the anode(bottom) of sulphide bodies. Laboratory experimentsshow that high pH fluids move from thesurroundings of the cathode towards the anode andbeing only partly neutralized by acidic t1uids formedat the latter. The remaining acidic solution movesupwards for reasons of compensation wrapping theremaining high pH fluid. This onion-shell-likearrangement of the basic and acidic fluids aroundsulphide bodies is considered to maintain thepotassic alteration zone and promotes sericitisationof the intermediate, quartz-sericite zone of porphyrycopper systems, respectively. These two alterationzones, largely magmatic and transitional in origin,spatially coincide by and large with the presence ofthe two types of electrochemically altered t1uids. Atdepth, sulphides are electrochemically oxidised to

* Corresponding author, E-mail: pemoe@gfz-potsdam.de

sulphates such as anhydrite. The released Cu ions(and other electropositive ions) are captured by theelectric field and do not dissipate far into the countryrocks.

RIASSUNTO. - Gli studi sperimentali dilaboratorio e di campagna sui sistemi di porphyrycopper mettono in evidenza processi elettrochimicinegli areali occupati da corpi sulfurei semiconduttoriabbastanza ampi ed estesi vertical mente. Ledifferenze verticali nelle fug acita d'ossigenocausano specifiche reazioni elettrochimiche attornoaIle mineralizzazioni di solfuri indicate da sistemidipolari con autopotenziali negativi nella superficieterrestre in prossimita di tali depositi.

In seguito a tali reazioni elettrochimiche, siformano t1uidi con pH relativamente alto nel catodo(superficie superiore) mentre in corrispondenzadell'anodo (zone profonde) di tali corpi sulfurei sigenerano fluidi a basso pH. Esperimenti dilaboratorio indicano che i fluidi con valore di pHelevato si muovono dalle zone circostanti il catodoverso l'anodo e sono solo parzialmente neutralizzatida quelli acidi generati da quest'ultimo. Le soluzioniacide rimanenti si spostano verso I' alto per processidi compensazione avvolgendo i restanti fluidi conpH elevato.

Questa disposizione di tipo cipollare dei fluidibasici ed acidi attorn o ai corpi sulfurei eresponsabile del mantenimento della zona dialterazione potassica e dell'innesco della

112 P. MOLLER and F. TIMM

sericitizzazione delle zone intermedie a quarzo esericite dei sistemi di porphyry copper.

Queste due zone di alterazione, inizialmentemagmatiche e transizionali, coincidono spazialmentecon la presenza dei due tipi di fluidi a1teratielettrochimicamente. In profondita i solfuri sonoelettrochimicamente ossidati a solfati tipo anidrite.G1i ioni rame rilasciati (cost come gli altri ionielettropositivi) sono catturati dal campo elettrico enon si disperdono nelle rocce circostanti.

KEY WORDS: Porphyry copper deposits, fluid flow,electric field, electrokinetics, geobattery,sericitisation.

INTRODUCTION

Porphyry copper deposits are p1ug1ike dikesand cylindrical plutons which may be 1 to 10km in diameter (Gustafson and Hunt, 1975;Meyer, 1981). The ore normally occurs in rockfractures and is commonly zoned around acentre (Lowell and Guilbert, 1970; Guilbertand Lowell, 1974) with a vertical extension ofseveral kilometres (Meyer, 1981). The epizona1environment is considered the hearth of theporphyry base metal deposit (Guilbert andPark, 1986). Exploited porphyry copperdeposits represent the largest man-made open(e.g., ChuquicamatalChi1e, Bingham/USA) andunderground mining operations (e.g., E1Teniente /Chi1e) in the world.

An idealised, tectonically undisturbedporphyry copper deposit consists of acylindrical ore shell of chalcopyrite with orwithout bornite within a zone of potassicalteration (Lowell and Guilbert, 1970; Titleyand Beane, 1981; Beane and Tit1ey, 1981;Guilbert and Park, 1986). This ore shell ispartly overlapped by a phyllic alteration zonecarrying abundant pyrite and lesserchalcopyrite (Fig. 1). Laterally, the alterationgrades outwards to a propy1itic zone (ch1orite­epidote-carbonate) and upwards into an«advanced argillic assemblage» (Meyer andHem1ey, 1967). The propy1itic zone mayrepresent either an early sheath around thepotassic zone prior to the sericitic alteration orthe result of weakening solutions from the

sericitic alteration. Much of the sericite-pyritezone may also be superimposed on the K­silicate (Meyer, 1981). The sericite zone maytaper out downward. It is generally acceptedthat the potassic, phyllic and argillic alterationreflects the interaction of host rocks withhydrothermal fluids, although some additionalsericite and/or clay may also be added duringsupergene alteration (Sillitoe and McKee,1996; Tit1ey and Marozas, 1995). Mostporphyry ore deposits have been once orrepeatedly tectonically overprinted.

The diverse alteration processes are poorlyunderstood (Titley, 1994) and much debated.The one-fluid concept of alteration is based onthe consideration of the thermodynamicstability of the K-bearing minerals.Sericitisation of K-fe1dspar necessitates either afluid with a decreasing activity ratio of K+/H+at constant temperature or a fluid decreasing intemperature at constant K+/H+ (Fig. 2) oranything between these extremes. According toBeane and Tit1ey (1981) changes inhydrothermal minerals are largely controlled bychemical effects and the phyllic assemblageshould reflect low Nat, Ca2+, and Mg2+ relativeto K+ in the fluid. Changes in H+ concentrationhas not been discussed in detail, but is widelyassumed.

The two-fluid model of alteration is based onhydrogen and oxygen isotope studiessuggesting a magmatic fluid that produced theK-fe1dspar-biotite assemblage (Sheppard et al.,1969, 1971). Post-mineralisation encroachmentof meteoritic fluids on the central intrusioncaused sericitic alteration. Initially, exogenicmore or less acidic fluids, however, wouldrender magmatic signatures due to water-rockinteraction when moving toward the centre ofthe intrusion. Henley and McNabb (1978)proposed that a low density magmatic vapourentered the core of the intrusion and condensedto a high-salinity liquid in the carapace of theintrusion. Along the margins of the plume,condensed fluids were probably entrained intocirculating ground water.

Beane and Tit1ey (1981) summarised theproperties of fluid inclusion studies as follows:

Electrochemical and electrokinetic aspects ofporphyry copper systems: experiments andfield observations 113

I

Ore and pyrite!Hematite Fe203

Goethite FeOOH OxidationJarosite KFe3(S04h(OH)6 zone

AlterationA----------- - - ---------

Supergene =-~---~-~-=-=-=-=-aueranon A..:g!..ll~ '.r.r:»:-

t- - - - -·- - - - - - - ­-.-----------. . .. .. .. .. .. .!....:;...:;..~.-; - - -

. . . . ", ,

I ••••••••••••••••••••••••••••••••

.. .. .. '

............ , .1

.. .. "' /

.". .

.............................., .. ~,. . . . . .........................., ................................

Fig. 1 - Idealised cross-section through the mineralization (right side) and alteration zones (left side) of a porphyry coppersystem redrawn after Lowell (1968) and Beane and Titley (1981). Propylitic zone: chlorite-epidote-carbonate; Phyllicalteration zone: quartz-sericite; Potassic alteration zone: K-feldspar, biotite; Argillic alteration: quartz-kaolinite-chlorite;Low grade ore with low total sulphide contents; ore shell: pyrite, chalcopyrite, bornite, molybdenite; pyrite shell: pyritewith little chalcopyrite and bornite.

400 and

.-.0 A0..........Q. muse K-sparE pyro~ 300

kaol

2001 2 3 4 5

logCK+rH+)

Fig. 2 Stability of K-silicate minerals as a function of the activity ratio of K+ and H+ at 500 bars and quartz saturation(after Beane and Titley, 1981). Starting from point A in the potassic zone, the arrows show that by decreasing temperatureand/or decreasing log activity of the K+/H+ ratio sericitisation is possible. (and: andalusite; pyro: pyrophyllite; kaol:kaolinite; muse: muscovite, K-spar: K-feldspar)

114 P. MOLLER and F. TIMM

(i) magmatic or early alteration fluids showhigh homogenisation temperature and salinity(>700°C; 30-60% NaCl-equiv.), (ii) halite­saturated fluids of intermediate temperatureindicate inclusions that homogenise at 250­600°C and often indicate boiling systems, and(iii) late stage fluids are characterised byhomogenisation temperatures of <450°C andsalinities <20% NaCl-equiv. Guilbert and Park(1986) suggest that ore deposition took placefrom mesothermal boiling fluids (250-500°C,rarely 600-700°C, and 30-60% NaCl-equiv.).

The host rocks of some porphyry copperdeposits (for example El Teniente and LosBronces, both in Chile) contain large amountsof anhydrite, particularly at depth (Gustafsonand Hunt, 1975; Camus, 1975; Warnaars,1985). 834S of anhydrite and sulphide differ by10 to 15%0 (Field et al., 1983; Ohmoto andRye, 1979). Field and Gustafson (1976)suggested that the early anhydrite-chalcopyrite­bornite assemblage with initial 834S of about0%0 is associated with the potassic alteration,whereas the later anhydrite-pyrite-chalcopyriteassemblage with enhanced 834S between +5and + 10%0 is related to the quartz-sericitealteration.

Thus, there are several questions in particularconcerning epigenitic post-ore sericitisation,which occurred later than propylitisation andpotassic alteration and at low temperatures(200-350°C) and low fluid salinity. Althoughsome supergene sericite might be formed byacidic fluids seeping from the top of theweathered ore body, the larger amounts seemto be of hypogene origin. A direct consequenceof the former process would be a sericitic orargillic zone at the contact between theenriched blanket and the potassic alterationzone, which has never been reported as asignificant feature. Thus, the hypogene sericitedemands for an acidic fluid from depth.

CURRENT ELECTROCHEMICAL MODELS

In nature, any sulphide ore body forms ageobattery with the cathode at top and the

anode at depth (Sato and Mooney, 1960;Becker and Telford, 1965; Blain andBrotherton, 1975; Bolviken and Logn, 1975;Bolviken, 1978; 1979). Sivenas and Beales(1982a,b) defined two types of self-driving,galvanic cells:

- The oxygen concentration cell (OCC),which is considered to consist of an inert dipoleelectrode with difference of oxygenconcentrations in the cathodic and anodicdomain yielding the driving forces for thereactions; and

- the sulphide galvanic cell (SGC), in whichthe sulphides are dissolved at the anode,whereas, most likely, a reaction similar to thatof the OCC occurs at the cathode.

In the OCC the ore body is thought to splitinto two half cells and to act only as anelectronic conductor for the transfer ofelectrons from the anode to the cathode (Satoand Mooney 1960; Bolviken and Logn, 1975;Stoll et al., 1995; Bigalke, 1996). In the SGCmodel the dipole electrode takes part inelectrochemical reactions. The anodic reactionat depth is the galvanic corrosion of sulphidesdue to the aeration of the top of the ore body.The reaction of the dissolved oxygen ingroundwater might be considered as the drivingforce of the cell. In nature, both types ofelectrode settings might occur side by side(Sivenas and Beales, 1982a). In terms of theOCC and SGC model the dipole field resultsfrom electrochemical processes, which yield aflow of H+ ions in the fluid and electrons in thesulfide body, both heading from the anode tothe cathode. The influence of also generatedOH- ions at the cathode is not considered(Bolviken 1979; Bolviken and Logn, 1975;Levinson 1980).

Contrasting the OCC model, Corry (1985)summarised experiences that contradict somefundamental assumptions made by the abovecited authors. He finds that negative selfpotentials are bound to quartz-sericite-pyritealterations associated with sulphide ore bodies,which is in agreement with our measurementsin alteration zones of porphyry copper depositsin Chile (Timm and Moller, 2001).

Electrochemical and electrokinetic aspects ofporphyry copper systems: experiments andfield observations 115

Furthermore, the model cannot explain selfpotentials exceeding 1 Volt (Kruger and Lacy,1949), no relation to the water table isobserved, no positive pole could be detected sofar, and the stability of self potentials imposesthe requirement that sulphides are chemicallystable in an environment within the electricalregime of measurement.

In a recent paper (Timm and Moller, 2001)published measurements of the redox-, self­and total potentials in Chuquicamata and LosBronces mines, both in Chile. The local redoxpotential is interpreted in terms of local H+activity. The total potential is the sum of theself- and local redox potentials which are notalways changing equi-directional.

Therefore, the aim of this study is to showthat specific electrochemical reactions takeplace at both ends of the sulphide depositgenerating fluids with high and low pH valuesat the cathode and anode, respectively, andthereby establishing a dipole field. Thegenerated fluids of different pH values couldbe the cause of distinctly different alterationreactions such as those associated withporphyry copper deposits. It will be shown that(i) a constant dipole field is established, if theore body is embedded in a environment ofdecreasing oxygen fugacity, and (ii) thevarious alteration zones differ in H+ activity.The changes in H+ activity are studiedexperimentally and in field measurements.

EXPERIMENTAL STUDIES

All sulphides are semiconducting materialsand many of them (e.g. chalcopyrite and pyrite)show considerable electric conductivity. Forexample, between the isolated sulphide grainsdispersed in non-conductive rocks in the dryopen pit of Chuquicamata, Chile, the electricconductance is in the range of 10-3 S/m.

In order to study electrochemical processesassociated with sulphide-bearing bodies, twotypes of experiments have been designed. Thefirst type of experiments exan;ines chemical

features of the electric potential field inaqueous systems. These experiments areconducted in tabular reaction cells withheight:length:width of 20:30:2 em (Fig. 3a).Chemically pure quartz sand (E. Merck) wastreated with 0.3 to 1 M Na2S04 solutionscontaining a pH sensitive indicator, for at least2 days at a pH value of about 6.5, before it wasflushed into a cell hosting two separate Pt meshelectrodes (not covered with Pt black)surrounding a polyethylene rod. The Pt meshelectrodes are connected to a DC power supply.After 2-3 days, the potential is fixed at 1.5 Vwith the cathode at top and the anode atbottom. Due to polarisation and electrolysis ofwater, OH- and H+- ions are generated at theelectrodes and the solution turns blue (alkaline)at the cathode and red (acidic) at the anode(Figs. 3b-c). The registered electric current is inthe range of about 1 /lA.

The cells allow a study of the dispersion ofOH- and H+ ions into the porous medium(about 45% porosity). After about 2 weeks, thetwo coloured fields visibly contact each otherand neutralisation of OH- and H+ ions takesplace (Fig. 3b). Again 2 weeks later, the uppercoloured fluid field in Fig. 3c is distortedindicating that the OH- bearing fluid generatedat the cathode is replaced by neutral solutionfrom the top as indicated by the arrows. TheOH- ions and anions of the dye characterisingthe fluid from the cathode slowly movesagainst the electric field which is only possible,if fresh solution moves from the surface of thesand package towards the electrode assemblage(Fig. 3d). Due to enhanced dispersion of H+ions the remaining acidic fluid starts to envelopthe basic zone. The final result is that the spacebetween the electrodes is mainly occupied bythe high pH fluid which is wrapped by a shellof the low pH one.

In experiments, where three vertical sets ofelectrodes are arranged in the same cell but atvarious distances from the surface, each setdevelops its own system of fluid movement(Fig. 3e) at least if the distance of theelectrodes in each assemblage is less than thedistance between the different systems.

116 P. M OLLER and F. TIM/VI

/ AnOde

/ c athode

24 .10 .97S

Fig. 3 - Photographic reprodu ctions of the vertical fro ntplate of a tabular, experimental cell (a) filled with sand and0.3 M Na2S04 solution at pH 6.5 containing a pH sensitivedye. The blue and red coloured labels in (b-d) indicate theposition of the Pt-rnesh electrodes, at which OH- and H+ions are generated and makes the solution blue and red,respectively. After four days electromigration of ions leadto radial distributi ons around the electrodes (b). Four weekslater, the radial distribut ion is deformed and, particularly,n~3 iVl"I~1~2S04 Is<5 iml o"r\'arptf'Q) ;Co;lltftn;ing ~tpi-fsei\s~tivedye. The blue and red coloured labels in (b-d) indicate theposition of the Pt-rnesh electrodes, at which OH- and H+ions are generated and makes the solution blue and red,respectively. After four days electrorn igration of ions leadto radial distributions around the electrodes (b). Four weekslater, the radial distribution is deform ed and, particularly,the alkaline solution shows that it is being replaced by afresh one from the top (c) . Afte r another six weeks, thealkaline solution is only present in the range between the 'e lec trodes (d). The time depen dent d istr ibutio n of thedefor med , co loured fie lds ind icate the de velopment ofindu ced movem en t of the alkaline solution across theelect rode arrangeme nt down ward s (arrows) , the ac idicsolution ascends at greater distance from the electrodes.Three assemblages of electrodes at different covers of sandare driven by the same DC voltage of 1.5 V. These threesystems do not interfere with each other and develop in thesame way as shown in b to c (e).

Electrochemical and electrokinetic aspects ofporphyry copper systems: experiments andfield observations 117

The second type of experiment simulates theaction of a geobattery (Fig. 4). The pyriteelectrodes are manufactured by attaching acopper wire to pyrite plates cut from the samelump of massive pyrite. This mounting isimmersed in epoxy resin in order to seal allfissures and pores. After hardening of the resin,the site opposite to the fixed Cu wire is cleanedof resin by grinding. All half cells of thesulphide electrodes and the Ag/AgCl referenceelectrode are filled with 3 M KCl solution andare linked by agar-filled electrolyte junctions.Flushing the cells with either air or nitrogen(the latter being produced from liquid nitrogen)turn them into cathode or anode, respectively.

After about one hour, constant potentials areregistered for both half cells relative to theAg/AgCl reference electrode. Coupling thesulphide electrodes by a variable load, theelectric current in each half-cell is changedfrom zero to maximum current densities, bywhich the potentials decrease and increases atthe cathode and anode, respectively. Whenplotting the absolute anodic and cathodiccurrent densities vs. the measured potentials

(relative to the Ag/AgCl electrode) Fig. 5 isobtained, in which the extrapolated horizontalsections represent the exchange currentdensities at the electrodes under equilibriumconditions (Bockris and Reddy, 1977).

A detailed inspection of the surface of theelectrodes after a long-time experiment at thecorrosion potential reveals that the cathode isstill fresh but the anode shows brownish to redspots indicating the formation of oxidised ironcompounds. This proves that oxidationprocesses occurred at the anode, although itwas in the N 2-flushed environment. Thefundamental electrode reactions could be:

Cathode: 2H20 + O2 + 4e --+40H- (1)

Anode: 2H20 --+40H- + 4e + O2 (2)

Fe2+ --+ Fe3+ + e (3)

S/ --+~ + 2e (4)

Any anodically produced 02 (reaction (2)) isimmediately consumed by oxidation of otherionic species like Fe+2. Reaction 3 representsthe direct oxidation of Fe(II). Evidence forreaction (4) could not be found.

Jf'\

ll<p cat ll<p an

air N2--------, A A ,--------,I U I I

I 0'> I--I « - --1/

I II - II 0'> II « ,I II II II II II II II pyrite pyrite I.. ..

Fig. 4 - Design of a self-driving cell arrangement containing sulphide electrodes under air and nitrogen coupled to aAg/Agel reference electrode. The load allows the variation of the resistance between ]06 and 0 Ohm.

Fig. 5 - Plot of potential vs. logarithm of the absolutecurrent densities induced by electrochemical reactions atthe surfaces of the sulphide electrodes. All data are readafter 30 min and can be considered to represent a transientequilibrium. io,cat and io,an represent the exchange currentdensities at the cathode and anode, respectively. Thecorrosion current density and corrosion potential aredefined by the intersection of both trends lines. The dottedline shows the influence of decreasing exchange currentdensities on the trend of the cathodic potentials and on thecorrosion potential (Bockris and Reddy, 1977).

4 FeS2 + 10 H20 + 15 O2 ==8 S04-2 + 4 FeOOH + 16 H+ (4)

Eh =1.23 + 0.015 log (/02) - 0.059 pH (5)

it. Into each clot a polyethylene ring with astopper is pressed. The stopper has a holethrough which the electrode is inserted. Bothelectrodes are slowly pressed through the clayonto the surface of the rock. The electrolytecontact is established and maintained by thewet clay. The clay-rock redox potential at theinterface (from hereon termed clay potential) isvaried by ionic interaction with the underlyingrock. The readings started with values muchhigher than the final ones and decreased by 20to 100 mV with time. Because the redoxpotential was measured when the changes ofreadings were 1 mV/min, it follows that thelong-time redox potential would be lower thanthe registered one. The observed, slowapproach to a final value might be due to ionexchange between rock and clay, reaction ofoxygen with sulphide minerals at the rock'ssurface (eqA), and/or changes of the oxygenfugacity in the wet bentonite (eq. 5):

-4

P. MOLLER and F. TIMM

100K /1

loqi, an

1 M\

00

I

-80

-120

<I

118

40loqi, cat

t,

U0)

«0-0)

«-+-.: ,(1) ,~ corrosion potential\~'-""

> -40 10K' /'

E

MEASUREMENTS OF ELECTRIC POTENTIALS

IN THE FIELD

Since the reported experiments indicateconsiderable differences of redox potentialsbetween the basic and acidic solutions, theexperimental results on redox potentials will becompared with field measurements of redoxpotentials between the zones of potassic andsericitic alteration in the dry open pit of theChuquicamata mine, Chile.

In-situ redox potentials are measured onrock surfaces between a Pt electrode and anAg/AgCl reference electrode (both fromMettler Toledo) placed at distances of 15 to 30em (Timm and Moller, 2001). The rock surfaceis cleaned of dust, sprayed with water, and twoseparated clots of wet, commercial bentonitewith a pH value of 7.7 at 20°C are pasted onto

Across bench 01 near the bottom of the verydry open pit of Chuquicamata, Chile, thepotassic zone shows lower Eh values than thequartz-sericite zone (Fig. 7). Assumingconstant local oxygen partial pressure duringthe measurements of the clay potentials andneglecting other redox active couples, eq.(5)can be used to convert measured clay potentialsinto pH units (Fig. 8). The pH axis is calibratedby the measured pH value of the clay of 7.7and the theoretical slope of 0.059 mV/pH atroom temperature. In both alteration zones thevalues scatter around specific means, yielding apotential difference of about 60 mV, whichcorresponds to about one pH unit at roomtemperature. Thus, it is qualitatively confirmedthat the quartz-sericite zone shows higher Ehvalues than the potassic zone, which is in goodagreement with the pH difference visualised inour described experiments (Fig. 3).

Electrochemical and electrokinetic aspects ofporphyry copper systems: experiments andjield observations 119

DISCUSSION

The porphyry copper system as a geobattery

In any galvanic cell the maximum potentialdifference at equilibrium is lowered undercurrent flow, i.e. when reactions take place atboth electrodes (Fig. 5). Depending on thespecific conductivity of the ore body, theestablished potential difference is always lessthan the maximum value. In porphyry coppersystems with their low electric conductivity, theexchange of electrons between the electrodes ishampered. In terms of Fig. 5 the working pointof the geobattery might be well in the range ofthe horizontal segments of the log/if vs. <pcurves, i.e. they work near to equilibriumconditions. As shown schematically in Fig. 5,the actual potentials of the electrodes undercurrent depend on the exchange current

densities which in turn depend on structure andcomposition of the electrodes and the solution­solid interface (Bockris and Reddy, 1977).Lowering the cathodic potential, e.g. underconditions of reduced exchange current densityio,cat (dotted lines in Fig. 5) has a positive effecton electrochemical reactions taking place at theenhanced negatively biased cathode. Undersuch conditions the electrochemical reductionof O2 at the negatively charged surface isfavoured by the lowered activation energy Ell asshown schematically in Fig. 8 (Bockris andReddy, 1977).

In order to discuss electrochemical reactionsand electrokinetic effects it is necessary tosketch an electric circuit that simulates typicalelectrical conditions of a porphyry coppersystem (Fig. 9), based on the variousexperimentally results. Different from those

Chuquicamata, profile C

60Falla Americana

Abundant K-feldspar,little sericite

40

20 Potassic alteration;K-feldspar,+/- sericite

• t- - - - - - - - - - - - - _II!!I - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - .

II

II

Phyllic alteration;quartz/sericite

"iii"-E2oa.~ -20U

-40

-60

-80

50

II

~-~~"7~" ·Abundant sericite, IIlittle K-feldspar

150

Molybdenite veins

200 250 300 350

Distance (m)

400

Fig. 6 Plot of measured clay potentials across the lithologic boundary of the potassic and quartz-sericite zones at thelowermost bench in the open pit of the Chuquicamata mine. In the dry open pit of Chuquicamata the sericite zonequalitatively show the expected higher potential than the potassic zone. Dashed lines indicate the arithmetic means of theclay potentials of both zones.

120 P. MOLLER and F. TIMM

Profile Chuquicamata C

60

pH

~pH potassic zone = 0.66

pH(bentonite) = 7.720

40

>.§.~

~ 0 +--~---T~---'----'----'----'---="'C--'--------------T+-------,---------~-,----,---+--,------,'0Q.

~U

-20

-40

~pH quartz/sericite zone =0.86----'

-60

Fig. 7 - Interpretation of clay potentials in terms of pH values that are established at the clay-rock interface, assuming thatno other redox process is acting than assumed in eq. (5). The pH axis is calibrated by the pH of the wet clay, which yieldpH 7.7 at 20°C. The slope of the calibration line is assumed to be 0.059 mV/pH. The resultant pH at the clay-rock interfacein the potassic zone is alkaline, but neutral to weakly alkaline in the quartz-sericite zone.

being already published (Sato and Mooney,1960; Becker and Telford, 1965; Blain andBrotherton, 1975; Bolviken and Logn, 1975;Bolviken, 1978, 1979; Sivenas and Beales,1982a,b) additionally OH- ions are involved.

For three reasons the electrochemical processat the cathode might be the rate controllingstep: (i) low solubility of oxygen ingroundwater, (ii) consumption of oxygen in theweathering zone of the deposit, and (iii) theanode-reacting domain being much larger thanthe cathodic domain. Some of the releasedCu(I) at depth might move in the electric fieldtowards the cathode, where it iselectrochemically or chemically precipitated inthe «enriched blanket» together with theinfiltrating dis sol ved copper from theweathering zone at the very top of the orebody. The electric field concentrates Cu fromboth sources in the cathodic domain.

Given an electric field, fluid flow is inducedin porous media by electrochemical effects asshown by the reported experiments. Fluid flowis caused by two effects:

• The generated OH- and H+ ions at thecathode and anode, respectively, (electrolysis)have to be charge-compensated for reasons ofelectroneutrality of the system which isachieved by migration of hydrated ions andcounterions moving in opposite directions(electro-migration), and

• the walls of all pores are electricallycharged (mostly negatively) with thecorresponding countercharge in the adjacentfluid, which may give rise to an unipolarcurrent in capillaries (electro-osmosis).

By friction mainly the cations with theirvoluminous solvation shells drag along a watercolumn in tiny pores. The superposition ofelectro-migration and electro-osmosis seems to

Electrochemical and electrokinetic aspects ofporphyry copper systems: experiments

M....·..OII

M·..··..O

observations 121

Distance of oxygen from the surface of the electrode

Fig. 8 - Schematic representation of the Morse curves (after Bockris and Reddy, 1977) which describe the change inpotentials as a function of the distance between the reactants showing the two assumed reactions steps (i) the dissociationof oxygen bonds at the surface and (ii) the formation of metal-oxygen compounds MO at the electrode surface, where Mrepresents any reaction centre. The overlap of the two curves define the chemical activation energy Eact. Because therespective curve for (ii) can be lowered by a negative overvoltage h of the surface of the reactant M, the electrochemicalactivation energy Ell for this process is much less than Eact in case of the purely chemical reaction. At a negatively biasedsurface (cathode) the reaction of chemisorbed oxygen with water in order to form OH- ions is energetically favoured.

yield a net fluid flow from the top to thebottom at least at the low potential differencesas present in natural systems. A morequantitative discussion of the electrokineticeffects will be given elsewhere. The dispersionof H+ exceeds by far that of the slower OH­ions. For this reason, OH- ions dominate H+ions in the space between the electrodes. Thiscontrols the confined distribution of the basicfluid in Figs. 3c and d.

Chemical changes of fluids and mineralogicalalterations

The chemical changes in the ambient fluidsof the electrodic domains (Fig. 9) may inducealteration of the country rocks and in the post-

ore sedimentary cover. Chemical anomalies inpost-ore cover have already been suggested byBolviken (1978; 1979) who discussed«secondary redox potentials» at the earth /ssurface that might be applied to exploration.

The electrochemically generated alkalinefluid prevents the oxydation of the sulphidebody in two ways, (i) acidic solutionsinfiltrating from the zone with intensiveweathering are neutralised, and (ii) the residualoxygen of the infiltrating fluids areelectrochemically consumed by reaction (l).As may be derived from the experiments, thealkaline fluid tends to move across the orebody downwards and partly neutralises theacidic fluid formed at depth. The remaining,

122

O2 - rich

transitionzone

O2 - poor

P. MOLLER and F. TIMM

Oxidation ofsulphides

<DCoN

.S:!:Q..c0-:::J(j)

Fig. 9 - Schematic sketch of a porphyry system linking an oxygen rich and poor zone. The sulphides are part of the centralbody (within the bold line boundary) and act as an electron conductor. By electrochemical reactions alkaline and acidict1uids are generated at the stippled and dashed surface areas. The given electrochemical reactions represent fundamentalones but others (eqs. (3) and (4» could also playa role. The alkaline t1uid neutralises the acidic t1uids from the oxidationzone infiltrating the sulphide zone. The residual alkalinity stabilises the potassic zone or reacts with the acidic t1uidsascending from depth. At the bottom oxidation of sulphides occur according to eqs. (3) and (4) which might lead toformation of anhydrite.

upwards moving low pH fluid generated at theanode wraps the downwards moving,remaining high pH fluid. Across the contactzone of both fluids neutralisation occurs.

Under natural conditions the basic fluid

stabilises K-feldspar and biotite, the maincomponents of the earlier formed potassic zone,whereas the acidic fluid favours sericitisation ofK-feldspar and mica of the earlier alteration.The released K+ ions, like the Cu+ ions, are

Electrochemical and electrokinetic aspects ofporphyry copper systems: experiments andfield observations 123

captured by the electric field and are directedtowards the cathodic domain, where they maybe consumed by secondary biotitisation andformation of bornite in the potassic zone,respecti vely. Thus, the two pronouncedalteration zones coincide with the presence ofthe two types of fluids resulting from essentiallythe same meteoric fluid but altered byelectrochemical reactions at the ore body.

In the anodic domain considerable massoxidation must occur for which the increasingcontents of anhydrite with depth of porphyrysystems may give evidence (Guilbert andLowell, 1974; Gustafson and Hunt, 1975). Itmay be argued that most of the oxydationproducts are formed in magmatic or«transitional» processes as defined by Burnhamand Ohmoto (1980), however, according totheir reported diagram anhydrite is only inequilibrium with magnetite at temperaturesbelow 300°C and 1 kbar. At highertemperatures only S02 is formed and arguingthat this S02 forms the sulphate still leaves thequestion unanswered from where the necessaryoxygen is derived? The describedelectrochemical model also yields a simpleexplanation for the presence of oxidationproducts such as anhydrite at depth. Accordingto this idea, some of the anhydrite of thequartz-sericite zone might be epigenetic.

CONCLUSIONS

Assuming that the experimental studiessimulate typical features of a natural self­driving, galvanic cell as represented byporphyry copper systems (Fig. 9) it follows:

- electric fields are established due to thepresence chemical gradients of, e.g., oxygen inthe earth's crust. In case of vertically extendedporphyry copper deposits this leads to cathodicand anodic domains at the top and bottom,respectively,

- basic and acidic fluids are formed byelectrochemical reactions across cathodic andanodic domains of the ore body, respectively,and

- fluids move along the ore zone from top tobottom.

The interaction of the fluids leads tochemical and mineralogical alteration of thehost rocks. Although most of the hydrothermalalteration is magmatic to post-magmatic(<<transitional» after Burnham andOhmoto, 1980) in origin, the anticipatedelectrochemical reactions yield obviouslysimilar results. Since the latter only came intoaction after considerable erosion of theoverburden of the porphyry systems, i.e.meteoric water with dissolved oxygen hadaccess to the ore body, the electrochemicallyinduced alteration is later than the magmaticones. The electrochemically generated fluids,basic at top and acidic at depth, tend toconserve the earlier established hydrothermalalterations. In the more alkaline environmentK-feldspar and biotite are stabilised, whereasthe H+ ions are consumed by sercitisation ofresidual K-feldspar and biotite of the quartz­sericite zone and the adjacent potassicalteration yielding secondary or late sericite.Thus, the alteration is the result of both earlymagmatic and late electrochemical processes,the contribution of which have still to bestudied in detail. It cannot be just by chancethat the two pronounced alteration zones ofporphyry systems are met by fluids thatcorrespond with the mineralogical compositionof the former. It rather turns out that theporphyry system as a whole is a passive, self­organising system with the remarkabletendency for conservatism of its alterationzones and metal content.

ACKNOWLEDGMENTS

The authors are indebted to Dr. G. Behn,CODELCO, Santiago/Chile and Freraut, J. Rojasand A. Faunes (CODELCO, DivisionChuquicamata) for making it possible to do thedescribed field work in the mine of Chuquicamata.The electrochemical laboratory experiments havebeen conducted by B. Richert. The text has beenimproved by critical reviews obtained from Dr. B.Bolviken and Dr. P. Furness.

124 P. MOLLER and F. TIMM

REFERENCES

BEANE R.E. and TITLEY S.R. (1981) Porphyrycopper deposits: Part II. Hydrothermal alterationand mineralization. Eeon. Geol., 75th AnniversaryVol. 235-269.

BECKER A. and TELFORD W.D. (1965)Spontaneous polarisation studies. Geophys.Prospecting 13, 173-188.

BIGALKE 1. (1996) - Theo retisch.e un.dexp erimentelle Untersuchungen zumGeobatteriemodell. PhD thesis Univ. Frankfurt;Verlag Mainz, Wissensehaftsverlag Aachen, 151p.

BLAIN C.P. and BROTHERTON R.L. (1975) - Self­potentials in relation to oxidation of nickelsulphide bodies. Trans. Inst. Mining Metall., 84,BI23-1n

BOCKRIS J. O'M. and REDDY K.N. (1977) - ModernElectrochemistry. 3rd ed. Plenum Press, NewYork, 1432p

BOLVIKEN B. (1978) - The redox potential field ofthe earth. In L.H. Ahrens (ed): Origin anddistribution of the elements. Pergamon Press, NewYork, 649-665.

BOLVIKEN B. (1979) - The redox potential field ofthe earth. In L.H. Ahrens Origin and distributionof elements. Pergamon Press, Oxford, New York,649-665.

BOLVIKEN B., and LOGN O. (1975) - Anelectrochemical model for element distributionaround sulphide bodies. In LL.Elliot and W.K.Fleeher (eds): Geochemical Exploration. ElsevierAmsterdam, 631-648.

BURNHAM C.W. and OHMOTO H. (1980) - Late­stage processes of felsic magmatism. Min. Geol.Spec. Issue, 8, 1-11.

CAMUS F. (1975) - Geology of the EI Tenienteorebody with emphasis on wall rock alteration.Eeon. Geol., 70, 1341-1372.

CORRY C.E. (1985) - Spontaneous polarisationassociated with porphyry sulfide mineralisation.Geophysics, 50, 1020-1034.

FIELD e.W. and GUSTAFSON L.B. (1976) - Sulphurisotopes in the porphyry copper deposit at EISalvador, Chile. Eeon. Geol., 71, 1533-1548.

FIELD e.W. RYE RO., DYMOND 1.R, WHELAN J.F.,and SENECHAL R.G. (1983) - Metalliferoussediments of the East Pacific. In W.C. Shanks(ed): Cameron volume on unconventional mineraldeposits. Amer. Inst. Min. Metall. & Petrol. Eng.,New York, 133-156.

GUILBERT J.M. and LOWELL J.D. (1974) ­Variations in zoning patterns in porphyry copperdeposits. CIM BulL, 73, 99-109.

GUILBERT J.M. and PARK e.P. (1986) - The geologyofore deposits. Freeman, New York, 985p.

GUSTAFSON L.B. and HUNT 1.P. (1975) - Theporphyry copper deposit at EI Salvador, Chile.Eeon. Geol., 70, 857-912.

HENLEY RW. and McNABB A. (1978) - Magmaticvapor plumes and ground-water interpretation inporphyry copper emplacement: Eeon. Geol., 73, 1­20.

KRUGER F.e. and LACY W.e. (1949) - Geologicalexplanation of geopphysical anomalies near Cerrode Pasco, Peru. Eeon. Geol., 44, 485-491.

LEVINSON A.A. (1980) - Introduction toExploration Geochemistry. Applied PublishingLtd. Wilmette, 924p.

LOWELL 1.D. (1968) - Geology of the Kalamazooorebody, San Manuel district, Arizona. Eeon.Geol., 63, 645-654.

LOWELL J.D. and GUILBERT J.M. (1970) - Verticaland lateral alteration-mineralization zoning inporphyry ore deposits. Eeon. Geol., 65, 373-408.

MEYER C. (1981) - Ore-forming processes ingeologic history. Eeon. Geol., 75th AnniversaryVol.,6-41.

MEYER C. and HEMLEY 1.1. (1967) - Wall rockalteration. In H.L. Barnes(ed): Geochemistry ofhydrothermal ore deposits. New York, Holt,Rinehart, and Winston, 166-235.

OHMOTO H. and RYE R.O. (1979) - Isotopes ofsulphur and carbon. In H.L. Barnes (ed):Geochemistry of hydrothermal ore deposits. 2nd

ed. Wiley and Sons, New York, 509-567.SATO M. and MOONEY H.M. (1960) - The

electrochemical mechanism of self-potential.Geophysics, 25, 226-249.

SHEPARD S.M.F., NIELSEN R.L. and TAYLOR H.P.(1969) - Oxygen and hydrogen isotope ratios ofclay minerals from porphyry copper deposits.Eeon. Geol., 64: 755-777.

SHEPARD S.M., NIELSEN R.L. and TAYLOR H.P.(1971) - Oxygen and hydrogen isotope ratios inminerals from porphyry copper deposits. Eeon.Geol., 66, 515-542.

SILLITOE R.H. and McKEE E.H. (1996) - Age ofsupergene oxidation and enrichment in theChilean porphyry copper province. Eeon. Geol.,91,164-179.

SIVENAS P. and BEALES F.W. (1982a) - Naturalgeobatteries associated with sulphide ore, I.Theoretical studies. J. Geoehem. Explor., 17, 123­143.

SIVENAS P. and BEALES F.W. (1982b) - Naturalgeobatteries associated with sulphide ore deposits,II. Field studies at the Viburnum Trend, SoutheastMissouri, USA. 1. Geoehem. Explor., 17, 145-160.

STOLL 1., BIGALKE J., and GRABNER E.W. (1995) ­Electrochemical modelling of self-potentialanomalies. Surv. Geophys., 16, 107-120.

Electrochemical and electrokinetic aspects ofporphyry copper systems: experiments andfield observations 125

TIMM F. and MOLLER P. (2001) - The relationbetween electric and redox potentials: evidencefrom laboratory and field measurements. 1.Geochern. Exp1or.,72,115-128.

TITLEY S.R. (1994) - Evolutionary habits ofhydrothermal and supergene alteration in intrusion­centered ore systems, southwestern North America.In D.R. Lentz (ed): Alteration and alterationprocesses associated with ore-forming systems. Shortcourse notes, VoU1, Geol. Ass. Can., 237-260.

TITLEY S.R. and BEANE R.E. (1981) - Porphyry

copper deposits. Part I: geologic settings,petrology, and tectogenesis. Econ. Geol., 75thAnniversary Vol., 214-235.

TITLEY S.R. and MARozAs D.L. (1995) - Processesand products of supergene copper enrichment. InF.W. Pierce and J.G. Bo llm (eds): Porphyrycopper deposits of the American Cordillera.Arizona Geol. Soc., 20, 156-168.

WARNAARS F.W. (1985) Porphyry copper andtourmaline breccias at Los Bronces-Rio Blanco,Chile. Econ. Geol., 80, 1544-1565.