Anda di halaman 1dari 15

American Mineralogist, Volume 86, pages 652666, 2001

0003-004X/01/0506652$05.00 652
INTRODUCTION
The cathodoluminescence of quartz has been used to study
a variety of geologic problems. Sedimentary petrologists have
used cathodoluminescence (CL) of quartz to understand frac-
turing, pressure solution, and cementation of sandstones (Sippel
1968; Zinkernagel 1978), and to trace the provenance of quartz
grains in sandstones (Zinkernagel 1978; Matter and Ramseyer
1985; Owen and Carozzi 1986). The implementation of rou-
tine observation of CL with the scanning electron microscope
(scanned CL with the SEM) has expanded the applications
of the technique. Scanned CL of quartz as imaged by the SEM
displays spectacular textures on the scale of tens to hundreds
of micrometers that cannot be seen using plane-light
photomicroscopy or conventional emulsion-film images of CL
obtained with optical microscopy (see Fig. 1). Scanned CL has
the advantages of having relatively stable operating conditions,
small scales of observation, and the ability to detect very faint
levels of CL. The microscopic textures of quartz revealed us-
ing scanned CL have led to new interpretations of diagenetic
processes and provenance in sandstones (Hogg et al. 1992;
Milliken 1994a, 1994b; Dickinson and Milliken 1995; Demars
et al. 1996; Seyedolali et al. 1997). Other fields of geology
have also benefited from this new technique. Microfractures in
quartz observed using scanned CL have been correlated with
oxygen isotopic measurements to determine the fractionation
of isotopes during crystal growth (Onasch and Vennemann
* Current address: Department of Earth and Planetary Sciences,
Johns Hopkins University, Baltimore, Maryland 21218, U.S.A.
E-mail: sarahpd@jhu.edu
Illumination of vein quartz textures in a porphyry copper ore deposit using scanned
cathodoluminescence: Grasberg Igneous Complex, Irian Jaya, Indonesia
SARAH C. PENNISTON-DORLAND*
Department of Geological Sciences, The University of Texas at Austin, Austin, Texas 78712, U.S.A.
ABSTRACT
Vein quartz from the Grasberg Igneous Complex in Irian Jaya, Indonesia, displays a wide range
of features when observed using scanned cathodoluminescence imaging on the electron microscope.
Examination of seventeen samples revealed concentric growth zoning, quartz-filled microfractures,
dark luminescence near sulfides, changes in crystal size and orientation from vein edges to vein
centers, turbid growth zoning, and the truncation of concentric growth zoning. Individual features
observed along two large photomosaic traverses from vein margins to vein centers coupled with
trace element analyses from the electron microprobe allow determination of a detailed history of
fracture infilling. The two traverses indicate a simple, two-stage history of crystal growth into an
open fracture followed by mechanical closing of the fracture with resulting microfractures. Electron
microprobe trace-element analyses show that the Fe content of quartz reaches values of up to 4900
ppm in dark-luminescent regions near sulfide crystals and decreases smoothly to ~200 ppm along
traverses away from the sulfides. The solution to the diffusion equation coupled with existing ex-
perimental data about Fe diffusion in SiO
2
indicate the duration of the diffusion was perhaps on the
order of thousands of years.
1995) and the flow of hydrothermal fluids in an igneous body
(Valley and Graham 1996). Growth and dissolution textures in
quartz phenocrysts revealed by scanned CL help igneous pe-
trologists to decipher magmatic processes (Watt et al. 1997).
Boiron et al. (1992) used scanned CL to characterize the se-
quence of events within Au-bearing quartz veins from the
French Massif Central and the Cassiar Mountains in British
Columbia. They correlated microfractures with analyses of fluid
inclusions to trace the evolution of the vein-forming fluids.
This paper presents a new application of scanned CL: the
study of vein quartz in a porphyry Cu ore deposit. Fracturing is
an important process in porphyry Cu ore systems. Fluid flow
through fractures results in mineral precipitation and vein-as-
sociated alteration that together account for much, if not all, of
the hypogene ore in porphyry Cu ore deposits (Beane and Titley
1981; Titley 1982). To fully understand the mechanisms of vein
formation and the deposition of ore minerals, it is essential to
examine textures of vein minerals. Typical vein mineralogy in
porphyry Cu ore deposits includes quartz, sulfides, and oxides.
Sulfides (e.g., pyrite and chalcopyrite) and oxides (e.g., hema-
tite and magnetite) yield much information about mineral
parageneses when examined using optical microscopy, but they
provide limited textural information about vein-forming pro-
cesses. Quartz also typically yields limited textural informa-
tion using conventional optical microscopy. With scanned CL,
however, vein quartz from the Grasberg porphyry Cu ore de-
posit reveals numerous textural features (Penniston-Dorland
1997a, 1997b). This paper describes the features imaged using
scanned CL on quartz-sulfide veins from the Grasberg Igneous
Complex, a porphyry Cu ore deposit in the Gunung Bijih
(Ertsberg) District of Irian Jaya, Indonesia (Van Nort et al. 1991;
PENNISTON-DORLAND: SCANNED CL VEIN QUARTZ TEXTURES 653
MacDonald and Arnold 1994; Mealey 1996). The application
of this technique to veins is an important development in the
study of porphyry Cu ore deposits, because scanned CL, in
conjunction with data for fluid inclusions and trace elements,
can be used to better understand the combination of fluid char-
acteristics and vein-forming processes necessary to concentrate
ore minerals to form a deposit of economic value. The scanned
CL textures of quartz found in the veins from the Grasbery are
present in at least one other porphyry Cu ore deposit as well.
Preliminary work by Seyedolali et al. (1998) discovered, using
scanned CL, that vein quartz from the porphyry Cu deposit at
Butte, Montana, has similar cryptic textures that are only re-
vealed using scanned CL.
GEOLOGIC SETTING
The Grasberg Igneous Complex (GIC) is the second largest
of several igneous bodies located in the Gunung Bijih (Ertsberg)
District in the Central Range of Irian Jaya, Indonesia (Fig. 2;
McMahon 1994a, 1994b). Sedimentary rocks, ranging from
mid-Jurassic to Miocene in age, host the Pliocene intrusions of
the Ertsberg District. Wall rocks consist of sandstones, mud-
stones, and minor limestones of the Jurassic-Cretaceous
Kembelangan Group, which are overlain by limestones and
minor quartzose sandstones of the Tertiary New Guinea Lime-
stone Group (Quarles van Ufford 1996). Folding and faulting
in the Central Range have developed since the Middle Miocene.
Most fold axes trend northwest (Fig. 2) and include the Yellow
FIGURE 1. Textures observed using scanned CL compared to other methods, sample EK1-17. The edge of the vein is at the bottom of the
photo, the center of the vein at the top. (a) Plane-polarized light conventional film photograph. Black semicircle in upper right corner of image
is ink dot. (b) Conventional film photographs collected with camera mounted on CL luminoscope. Black semicircle in upper right corner of
image is ink dot. The three light-blue lines on the image are tracks left by electron microprobe traverses. (c) Photomosaic of scanned CL images
collected on the SEM. Bright spots on all scanned CL images in all figures are pieces of corundum remaining from the thin section polishing
process. Thin white lines on all scanned CL images are scratches remaining from the thin section polishing process.
PENNISTON-DORLAND: SCANNED CL VEIN QUARTZ TEXTURES 654
FIGURE 2. Geologic map of the Ertsberg Mining District.
Abbreviations are: G: Grasberg, E: Ertsberg, YVS: Yellow Valley
Syncline.
Valley syncline, into which the GIC was emplaced. A few dip-
slip faults parallel the fold axes. These and a set of NE-trend-
ing strike-slip tear faults are contractional structures that
resulted from the island arc-continent collision associated with
convergence of the Pacific and Australian plates (Hamilton
1979). At least two NW-trending thrust faults were reactivated
as strike-slip faults at about 4 Ma, probably due to a change in
direction of plate motion. The GIC is pervasively cut by strike-
slip faults that post-date veining (Sapiie 1998).
Grasberg is Dutch for grass mountain, named after the
original grassy knob that rose to 4200 m elevation before min-
ing began. The GIC has a surface area of roughly 3 km
2
and
consists of intrusive and extrusive igneous rocks, with the in-
trusive units ranging in composition from quartz monzonite
and monzonite to diorite (MacDonald and Arnold 1994;
McMahon 1994c). The GIC hosts a porphyry Cu ore deposit
that has the third largest Cu reserves and largest Au reserves in
the world (Mealey 1996; Steve van Nort, personal communi-
cation, 2000). The GIC is young; McDowell et al. (1996) de-
termined K-Ar ages on magmatic biotite from the GIC that
range from 2.8 to 3.2 Ma. It was emplaced at relatively shal-
low depths (<2 km; Weiland and Cloos 1996), crosscutting a
syncline in the surrounding sedimentary units of the New
Guinea Limestone Group.
The GIC is divided into three main phases (Fig. 3). The
earliest produced the Dalam Igneous Complex (DIC), which
consists of a fragmental unit, a volcanic unit, and andesite flows.
The equivalent igneous rock at depth is an intrusive diorite lo-
cated below 3500 m elevation (MacDonald and Arnold 1994).
The second phase produced the Main Grasberg Intrusion (MGI).
The MGI is an equigranular quartz monzodiorite plug that is
pervasively altered and locally the primary host of quartz-mag-
netite and younger chalcopyrite vein systems, in which most
of the highest-grade Cu and Au ore is found. The latest phase
FIGURE 3. Geologic map of the Grasberg Igneous Complex
showing sample locations. The grid is from PT Freeport Indonesia.
The grid numbers are in meters and are centered on the original Ertsberg
deposit, which has the coordinates 20000 N, 20000 E (see McMahon
1994c, p.174).
PENNISTON-DORLAND: SCANNED CL VEIN QUARTZ TEXTURES 655
produced the Late Kali Intrusion (LKI), a wedge-shaped dike
that extends southeast from near the center of the complex and
crosscuts the MGI and the DIC. The LKI is a quartz
monzodiorite, which has a higher percentage of phenocrysts
(primarily feldspars, hornblende, and biotite) that are gener-
ally larger than those of the MGI. At depths below 3800 m
elevation, vein-rich parts of the LKI are ore-grade material.
Veins in the GIC range in size from 1 mm to 10 cm wide,
but most are less than 1 cm wide. They are generally straight to
slightly curved; some veins branch and others form a three-
dimensional interconnected network (referred to as
stockwork). Crosscutting veins are common, especially in
the center of the GIC. Veins in the GIC dip steeply and have
preferred orientations consistent with emplacement into a pull-
apart region of a NW-trending, left-lateral strike-slip fault sys-
tem (Sapiie 1998). Quartz (Qtz) is commonly oriented
perpendicular to the vein wall. Other important vein minerals
include magnetite (Mag), chalcopyrite (Ccp), and pyrite (Py).
Biotite (Bt) is found in veins post-dating the LKI. Mineral ab-
breviations are from Kretz (1983).
Veins in the GIC are divided into two distinct stages
(Penniston-Dorland 1997b). Stage 1, the major ore-forming
event, created at least five generations of cross-cutting veins in
the MGI and DIC. Stage 2 created at least five generations of
cross-cutting veins observed in the LKI and likely extending
into the DIC and MGI. The generalized sequence of veins in
Table 1 is a summary based on field, hand sample, and thin
section observations of materials found in pit levels as deep as
3840 m elevation, the deepest level of mining as of 1995, when
sampling was done for this study (Penniston-Dorland 1997b).
In the center of the high-grade ore zone, it is abundantly
clear that an early quartz-magnetite stockwork system is cross-
cut by a chalcopyrite stockwork that becomes more pyrite-rich
away from the core. Veins from stages 1c through 1e were re-
sponsible for much of the Cu ore deposition in the Grasberg.
SAMPLES
Seventeen quartz-sulfide veins from the Grasberg Igneous
Complex were selected for study. Samples from all three phases
of igneous activity near the center of the GIC were examined.
Most of the samples were collected from regions or traverses
of float samples from along bench faces within the open-pit
mine (Fig. 3). Collecting complete vein samples in place was
difficult because they almost always broke apart along the vein
when separated from the outcrop. Therefore, most samples were
float from the base of the bench from which they were blasted
loose. Dates are indicated for the different sample locations in
Figure 3 as mining activity gradually expands exposures avail-
able in the open pit mine to deeper levels. The veins selected
represent different stages in the history of the complex. Three
of the veins are from the quartz-magnetite-chalcopyrite
stockwork zone within the MGI (MGI2-3, MGI2-21, and MGI6-
1). These are from the highest-grade region of Cu mineraliza-
tion. Nine were collected from the MGI and the DIC outside
the central stockwork zone, but are most likely distal expres-
sions of the vein-forming events in the high-grade stockwork
zone (D1-6, D1-15, D1-27, D2-5, D2-25, D7-5, D7-10, RDC1-
18, and RDC1-28). Five of the veins collected in the LKI are
from stage 2 of veining (EK1-17, EK1-18, LK2-1, LK2-5, and
LK2-8), and appear to postdate the bulk of high-grade Cu min-
eralization. Only one of these veins (EK1-18) contains chal-
copyrite.
All the CL features imaged (except for turbid regions and
changes in crystal orientations) are found throughout the GIC
(see Table 3). The most striking difference observed in veins
from different regions in the GIC was in the apparent bright-
ness of cathodoluminescence. Two of the veins from the LKI
luminesced very brightly, whereas none of the veins from the
DIC matched the luminescence intensity of those LKI veins.
These brightly luminescent veins also produced the sharpest
photographic images and are used to illustrate many of the CL
features in the following sections.
METHODS
Scanned CL images were produced using an Oxford Instru-
ments photomultiplier-based CL detector (Model CL 302) in-
stalled on a JEOL T330A scanning electron microscope at the
University of Texas at Austin. The Hamamatsu multi-alkali
photomultiplier has relatively uniform sensitivity across the
visible spectrum. Sample currents near 90% of maximum were
used (the T330A does not provide a numerical readout of sample
current). An accelerating voltage of 10 kV was used. The sample
is a carbon-coated, polished thin section, which is placed ap-
proximately one millimeter beneath a parabolic mirror that
collects the light emitted by the sample. Carbon coat thickness
is approximately 25 nm on all samples.
Scanned CL is a computer-assisted method of recording the
CL response of a sample to bombardment by an electron beam
that rasters across its surface. The light emitted in response to
the electron beam is captured by the photomultiplier device.
Scanned CL imaging of minerals on the SEM, therefore, pro-
duces panchromatic black and white images that capture the
relative intensities of CL emissions. The intensity of the CL
displayed on the SEM monitor and the images produced de-
pends upon several factors, including the intensity and wave-
length of the light received by the detector, the thickness of the
carbon coat, and the degree of sample heating. Although the
images do not provide quantitative information about either
light intensity or wavelength, they show a wealth of textural
features pertaining to the growth and fracture-healing history
of the quartz crystals. The scale of these features is small enough
that using the electron microprobe greatly enhances the amount
of detail observed compared to using a cold-cathode CL de-
TABLE 1. Generalized sequence of veins
Stage 1 veins Post-DIC/MGI igneous events Pre-LKI intrusion
a-b Mag Qtz and Qtz Mag (at least 2 generations)
c-d Qtz Mag Ccp Py (at least 2 generations)
e Ccp/Py
Stage 2 veins Post-LKI intrusion
a Bt Qtz
b Qtz Bt Py Mag Ccp
c Bt Qtz
d Qtz Bt Py Ccp Mag
e Py Qtz Bt
Notes: Mag = magnetite; Qtz = quartz; Ccp = Chalcopyr; Py = pytite; Bt =
biotite.
PENNISTON-DORLAND: SCANNED CL VEIN QUARTZ TEXTURES 656
vice attached to an optical microscope. The luminescence of
some of the samples is faint enough that using the photomulti-
plier detector allows for better characterization and imaging of
features than capturing photographic images on a cold-cath-
ode CL device. Bright spots on all scanned CL figures are pieces
of corundum remaining from the thin section polishing pro-
cess. Thin white lines on all scanned CL images are scratches
remaining from the thin section polishing process.
Several veins were also studied with a cold-cathode-ray elec-
tron-source microscope-mounted luminoscope at the Depart-
ment of Mineral Sciences, National Museum of Natural History
in Washington, D.C., to examine the colors and light intensi-
ties captured on conventional color emulsion film compared to
the panchromatic images produced by the SEM. Each sample
was examined visually through the optical microscope and
photographed. Regions that luminesce less brightly on this CL
device appear less bright on the SEM image. In addition, red
CL detected with the luminoscope generally correlates with
darker CL regions of the SEM CL images, whereas purple and
blue regions correlate with brighter CL regions of SEM CL
images.
The causes of different colors of CL in quartz have not yet
been established, but are ascribed to a combination of struc-
tural defects and trace-element impurities (Marshall 1988;
Ramseyer et al. 1988). Early studies correlated CL colors with
rock-forming environments. Quartz formed in high-tempera-
ture/fast-cooling environments was observed to luminesce blue-
violet, whereas quartz from low-temperature/slow-cooling
environments was observed to luminescesce rusty or dull red
(Smith and Stenstrom 1965; Zinkernagel 1978; Dengler and
Sprunt 1977; Sprunt et al. 1978). More recent research has cor-
related CL colors with trace-element impurities in quartz. For
example, Sprunt (1978, 1981) correlated a high Ti/Fe ratio with
blue CL and a low Ti/Fe ratio with red CL.
To correlate CL brightness with the trace elements Al, Fe,
and Ti, trace-element compositions of quartz were measured
using wavelength dispersive spectrometers on a JEOL
Superprobe at the University of Texas at Austin. Data correc-
tions were performed using ZAF software. The conditions used
for trace-element analysis of quartz were an accelerating volt-
age of 15 kV, a beam current of 25 nA measured on brass, a
minimum beam size (about 1 m), and a 90 s maximum count-
ing time. Traverses were analyzed using 5 m distances be-
tween points. The detection limits were equal to or below 0.02
wt% Fe for the three traverses run at 15 keV and below 0.06
wt% Fe for the traverse run at 10 keV (e.g., Le Maitre 1982).
The error for each measurement was calculated using the dQant
program of Geller Micronalytical which follows the reason-
ing of Le Maitre (1982). Relative accuracy of the analyses,
monitored using comparisons between measured and published
compositions of standards, is within 0.5 wt% for FeO in K412
glass (reported concentration of FeO = 9.96 wt%).
CATHODOLUMINESCENCE AND ELECTRON MICRO-
PROBE RESULTS
The observed CL features have been grouped into six main
categories: concentric growth zoning, changes in crystal size
and orientation from vein edges to centers, turbid growth zon-
ing, dark luminescence near sulfides, quartz-filled microfrac-
tures, and the truncation of concentric growth zoning.
Concentric growth zoning, dark luminescence near sulfides,
truncation of concentric growth zoning, and many of the quartz-
filled microfractures observed were only detectable using
scanned CL and not using optical microscopy. The changes in
crystal sizes and orientations are observable using optical mi-
croscopy, but are more striking using scanned CL. The turbid
growth zoning is not directly observable using optical micros-
copy, however, observing the population of fluid inclusions at
the base of regions of turbid growth zoning hints at the com-
plexity observed using scanned CL in these regions. These fea-
tures are described in the following sections, with representative
images taken primarily from the brightly luminescing veins of
the Late Kali Intrusion. Table 2 lists luminescent features ob-
served throughout the GIC.
TABLE 2. Cathodoluminescence results
Igneous Sample Sulfide Rhythmic Irregular Dark near Changes in Turbid Truncated Brightness
Unit minerals banding dark bands sulfide crystal orientation regions banding
Dalam D1-6 Py Yes * N.D. N.D. Faint
Igneous D1-15 Py, Ccp N.D. N.D. Very Faint
Complex D1-27 Py, Ccp N.D. N.D. Very Faint
D2-5 Py, Ccp Yes Yes Yes N.D. No Yes Moderate
D2-25 Py Yes Yes N.D. N.D. No No Faint
D7-5 Py, Ccp Yes N.D. N.D. Faint
D7-10 Py, Ccp N.D. N.D. Very Faint
Main MGI2-3 Py, Ccp Yes Yes Yes N.D. No Yes Moderate
Grasberg MGI2-21 Py, Ccp Yes Yes N.D. N.D. No Yes Faint to Bright
Intrusion MGI6-1 Py, Ccp Yes Yes Yes N.D. No Yes Faint to Mod.
RDC1-18 Py, Ccp Yes Yes N.D. N.D. No Yes Moderate
RDC1-28 Py, Ccp Yes Yes Yes N.D. No Yes Moderate
Late EK1-17 Py Yes Yes Yes Yes Yes Yes Bright
Kali EK1-18 Py, Ccp Yes Yes Yes Yes Yes Yes Bright
Intrusion LK2-1 Py N.D. N.D. Very Faint
LK2-5 Py Yes Yes Yes N.D. No Yes Moderate
LK2-8 Py Yes Yes No N.D. No Yes Moderate
Notes: Py = pyrite, Ccp = chalcopyrite, N.D. = no data.
* = not observed due to faintness of image.
PENNISTON-DORLAND: SCANNED CL VEIN QUARTZ TEXTURES 657
Individual features observed in seventeen samples, the two
photomosaic traverses from vein edges to vein centers, and the
trace-element analyses lead to an interpretation of a detailed
history of fracture infilling for the two LKI veins. The chro-
nology inferred from the photomosaics consists of a simple,
two-stage vein history: early crystal growth into an open frac-
ture followed by closing of the fracture with distortion of im-
pinging crystals. Concentric growth zoning, changes in crystal
sizes and orientations from vein edges to centers, turbid growth
zoning, and dark luminescence near sulfides are interpreted to
have formed during early crystal growth into an open fracture.
Quartz-filled microfractures and truncation of concentric
growth zones were formed during and after the mechanical clos-
ing of the vein, along with continued formation of dark lumi-
nescence near sulfides.
Concentric growth zoning
Well-defined bands of light and dark luminescence that vary
in width up to about 30 micrometers reveal euhedral crystal
shapes in the majority of quartz crystals (Fig. 4). Typically, the
crystal shapes defined by the concentric growth zones point
toward the center of the vein, clearly indicating growth into an
open extension fracture. Growth zone orientations commonly
vary within a single thin section of a vein. Concentric growth
zoning varies in both intensity and wavelength of CL. Colors
of zoning range from blue to red along with intermediate shades
of purple, which are likely caused by combinations of red and
blue. The origin of the growth zones is uncertain because the
activator of the CL is unknown. The phenomenon may result
from cyclic trace element incorporation during the precipita-
tion of quartz. If the quartz precipitated from a dynamic fluid
with oscillating composition or precipitation rate or pulsation
in fluid pressure or temperature, the result would be concentric
zones in quartz with different trace element concentrations. No
correlations were found, however, between variations in the
amounts of Fe, Al, or Ti in quartz and concentric growth zones.
The concentric zoning may be controlled by other elements, crys-
tal defects, oxidation state of Fe, or compositional variation in Fe,
Al, or Ti below the resolution of the electron microprobe.
Changing crystal sizes and orientations
In the photomosaic traverses, the orientation and size of
quartz crystals change from the vein wall to the center (Figs. 5
and 6). Along the vein margins, quartz crystals are smaller and
display random patterns of concentric growth zone orientations.
Toward the interiors of the veins, larger quartz crystals display
concentric growth zones that indicate growth toward the cen-
ter into an open cavity. In both veins, the first set of growth
zones in the larger crystals end in regions with abundant fluid
inclusions. In the sample illustrated in Figure 5, the zone rich
in fluid inclusions is followed by a zone of turbid growth (see
next section), which is followed by continued concentric growth
zoning that outlines large euhedral crystal shapes, with con-
centric growth zones that point toward the vein center. In the
sample illustrated in Figure 6, the region of abundant fluid in-
clusions is mantled by smaller quartz crystals that display ran-
domly oriented concentric growth zones, similar to the crystals
near the vein margin. Still farther along the traverse, these crys-
tals become larger, and growth zones again point toward the
vein center.
The changes in quartz crystal size and orientation away from
the vein wall are most likely due to competition among grow-
FIGURE 4. Concentric growth zones in vein quartz. (a) Scanned CL photomosaic of sample EK1-17. Concentric growth zones show euhedral
crystal shapes with quartz c-axes that point toward the top of the image, which is the center of the vein. (b) Scanned CL image of sample MGI2-
21. Concentric growth zones with quartz c-axis pointing out of page, which is toward the center of the vein.
PENNISTON-DORLAND: SCANNED CL VEIN QUARTZ TEXTURES 658
ing crystals as they precipitated from the fluid in the open frac-
ture. Crystals were initially oriented in various directions due
to random nucleation along the walls of the fracture. The small,
randomly oriented crystals competed for space and nutrients
as they grew. The fastest growth direction in quartz is along
the c-axis on the basal plane (Van Praagh 1947; Hale 1948;
Van Praagh 1949; Thomas et al. 1949). Small crystals oriented
with the c-axis roughly perpendicular to the vein margin would
grow fastest, overtaking crystals oriented in other directions.
This explains the transition to larger crystals with c-axes aligned
toward the center of the vein.
A change in the nature of the fluid or interruption in the
growth of crystals is recorded by the region of abundant fluid
inclusions and/or turbid growth zoning (see next section). Af-
ter the horizon of abundant fluid inclusions formed, growth
competition may have started over again, beginning with a sec-
ond sequence of small, randomly oriented crystals followed by
larger ones that grew toward the center of the vein (Fig. 6).
Alternatively, after the interruption, crystal growth could con-
tinue along the original orientation if enough of that crystal
lattice was undisturbed, as in Figure 5.
Turbid growth zoning
The two veins examined in photomosaic traverses have re-
gions in which the regular pattern of straight, evenly concen-
tric growth zones is disturbed. These regions display both dark
and light CL, but the patterns are complex. An abundance of
fluid inclusions is present at the base of these regions in both
samples (Fig. 7). The fluid inclusions could have formed by
several different mechanisms. First, they could result from the
FIGURE 5. Changing crystal sizes and orientations, sample EK1-
18. The wall of the vein is at the bottom of the photomosaic, the center
of the vein at the top. Crystals are smaller near the edge of the vein
and larger toward the vein center. A turbid region near the top of the
image is succeeded by concentric growth zones with the same
orientation as the bands below the turbid region. See text for discussion.
FIGURE 6. Changing crystal sizes and orientations, sample EK1-
17. The wall of the vein is at the bottom of the photomosaic, the center
of the vein at the top. Crystals are small near the vein edge and larger
toward the middle of the photo, at which point there are fluid inclusions
and a turbid region. Above the turbid region, the crystals are small and
are succeeded by larger ones near the top of the image. See text for
discussion.
PENNISTON-DORLAND: SCANNED CL VEIN QUARTZ TEXTURES 659
dissolution of quartz, which would create an irregular crystal
surface on which to trap fluid, creating fluid inclusions when
precipitation began again (Roedder 1984). Alternatively, fluid
turbulence may have been responsible for the abundance of
fluid inclusions in these zones. Joshi and Paul (1974) corre-
lated inclusion formation in KDP crystals with the presence of
a dynamic fluid. Janssen-van Rosmalen and Bennema (1977)
correlated the formation of inclusions in KDP crystals with
eddies in the fluid. A turbulent fluid would also be likely to
create the complex turbid zoning observed. Finally, the abun-
dance of fluid inclusions may reflect rapid precipitation of
quartz. Quartz that is precipitated rapidly, due to a drop in fluid
pressure or temperature, may result in irregular growth zones
(Roedder 1984).
Dark luminescence near sulfides
The Grasberg quartz-sulfide veins typically have earlier-
precipitated quartz on the margins of the vein and later-pre-
cipitated sulfides (primarily pyrite) in the center. Vein quartz
near sulfide crystals invariably shows darker CL than surround-
ing quartz (Fig. 8). In some examples, the tips of quartz crys-
tals projecting into the center of the vein toward sulfide crystals
have darker CL than the other parts of the crystal (Fig. 8a).
Dark-luminescent regions range up to 140 micrometers in width
(Fig. 8b) and are red in color.
The Fe content of quartz was measured in six traverses from
three samples that began at the edge of a quartz crystal closest
to the sulfide-bearing center and ended at the vein margin. The
highest content of Fe (up to 4900 ppm or 0.49 wt%) is found in
the darker-luminescent regions next to sulfides, with the Fe
content decreasing smoothly to ~200 ppm away from the sul-
fide (Fig. 9). Concentrations of Al and Ti have no consistent
correlation with the darker-luminescent regions. The correla-
tion of high Fe content of quartz to red luminescence agrees
with the results from Sprunt (1978, 1981), who correlated a
low Ti/Fe ratio to red luminescence.
The dark-luminescent regions are much wider than could
be produced by interaction of the electron beam with nearby
Fe-rich sulfides. Two traverses were repeated using a lower
accelerating voltage (10 kV compared to 15 kV for the first
traverses) to verify that the Fe detected was not due to excita-
tion of FeK X-rays in the nearby sulfides by secondary fluo-
rescence. The Fe content measured in these traverses was within
analytical error of that measured at 15 kV, indicating that the
detected Fe in the quartz crystals was not the result of second-
ary fluorescence.
Some of the dark-luminescent regions cut across growth
faces and grain boundaries (Fig. 8), indicating that the Fe-rich
regions postdate original crystal growth. The smoothly decreas-
ing Fe concentration profiles and the cross-cutting relationships
suggest that the regions are the result of diffusion of Fe into
the quartz crystals after the precipitation of quartz. Evidence
in one quartz crystal indicates that some diffusion occurred after
FIGURE 7. Turbid growth zoning, sample EK1-18. Chevron-shaped
concentric growth zones are shown at the bottom of the image.
Abundant fluid inclusions are seen as irregular dark spots in the chevron
shape above the white band. The region of turbid growth zoning is
directly above the region of fluid inclusions. A system of unfilled,
irregular fractures cuts from the upper rigth to the left of the image.
FIGURE 8. Dark luminescence near sulfides, sample EK1-17. (a)
Scanned CL photomosaic. Dark-luminescent regions are seen at the
tips of quartz crystals next to sulfide mineral grains (black region near
top of image). Right-hand large crystal is fractured (concentric growth
zones are tilted to the left in the upper part of the crystal relative to the
bottom part of the crystal). A wedge-shaped region of dark-
luminescence, seen on the right-hand side of the image emanates from
this fracture. It has a slightly higher Fe content compared to the nearby
brighter luminescing parts of the crystal. (b) Scanned CL photomosaic.
Dark-luminescent regions appear on either side of the sulfide minerals
(black band in lower part of image).
FIGURE 9. Scanned CL images, with electron microprobe traverses
marked on each image. The Fe contents (weight percent) of quartz
smoothly decrease across each dark-luminescent region from sulfide
grain contacts into the quartz crystal interiors. (a) Sample EK1-17
Traverse 1 (b) Sample EK1-17 Traverse 3 (c) Sample EK1-18 Traverse
2 (d) Sample D2-5 Traverse 2. All data collected at 15 kV. Missing
data points are bad microprobe analyses for which total weight percent
did not sum to 100%.
L
PENNISTON-DORLAND: SCANNED CL VEIN QUARTZ TEXTURES 660
PENNISTON-DORLAND: SCANNED CL VEIN QUARTZ TEXTURES 661
fracturing of the quartz (Fig. 8a).
The profiles of the measured Fe content of quartz may be
fit to a solution of the diffusion equation. The solution for one-
dimensional, semi-infinite diffusion is (Crank 1975):
C C C C erf
x
Dt
x i i
= +

( )
0
2

(1)
in which C
x
is the measured Fe content at distance x from the
crystal edge, C
i
is the constant concentration of Fe at the sur-
face of the crystal, C
0
is the baseline Fe content in the interior
of the quartz crystal, D is the diffusion coefficient for Fe in
quartz, and t is the duration of the diffusion of Fe into the quartz
crystal.
A reduced chi-square minimization algorithm was used to
obtain the best fit to the solution to Equation 1 for four profiles
(Bevington and Robinson 1992). Each fit, along with the re-
sulting values for the product Dt and the reduced chi-square
value,

2
, are shown in Figure 10. The reduced chi-square value
(

2
) consists of a sum of the squares of the deviation of the
data from the fitting function, divided by the squares of the
uncertainties at each point. This sum is then divided by the
number of degrees of freedom. An ideal value for

2
would be
unity (

2
= 1), however, small values of

2
can also represent
larger uncertainties in measurement [this is the case in sample
EK1-17, Traverse 3 (

2
= 0.84), which had a higher uncertainty
due to different microprobe operating conditions]. The highest
value (

2
= 4.21) occurred in the traverse (sample EK1-17,
Traverse 1) with the highest initial Fe concentration. If the first
data point is ignored, the fit shown in the figure results in a

2
= 2.7. Ignoring the first point in these traverses is justified be-
cause the first analysis is less than 5 m from the edge of the
quartz crystal, and may measure some Fe produced by interac-
tions between the beam and nearby sulfides.
Several studies have investigated the diffusion coefficient
of Fe in SiO
2
and silicate glasses (Stock and Lehmann 1977;
Khler and Frischat 1978; Atkinson and Gardiner 1981;
Kononchuk et al. 1998). Knowing the diffusion coefficient, D,
for Fe in quartz allows the calculation of the duration of the
diffusion event, t, from the Dt values obtained for each fit. Dif-
ferent coefficients were obtained experimentally and varied
depending on both the form of SiO
2
and the mechanism for the
diffusion of Fe. In two studies, the exchange mechanism re-
sponsible for Fe
+3
substitution was:
2 Si
+4
+ 2O
2
2Fe
+3
+ O
2
+ O-vacancy.
For this mechanism, the rate-limiting factor for the diffu-
sion of Fe was the diffusion of O vacancies. The diffusion
FIGURE 10. Solution to the diffusion equation fit to measured Fe contents. (a) Sample EK1-17 Traverse 1 (b) Sample EK1-17 Traverse 3 (c)
Sample EK1-18 Traverse 2 (d) Sample D2-5 Traverse 2. Data collected at 15 kV except sample EK1-17 Traverse 3, which was at 10 kV. Error
bars shown are 2, and

2
s the reduced chi-square fit parameter.
PENNISTON-DORLAND: SCANNED CL VEIN QUARTZ TEXTURES 662
coeffient was similar in these two studies (at ~600
o
C, D 1 to
710
20
cm
2
/s) despite different forms of SiO
2
in each study [syn-
thetic glass, Atkinson and Gardiner 1981; thermally grown SiO
2
layer in bonded-and-etched-back (BESOI) samples, Kononchuk
et al. 1998]. A value of 600
o
C is within the range of tempera-
tures expected in the GIC veins (400700
o
C in quartz
stockwork, Kyle and Bodnar 1995). The duration of the diffu-
sion event using this range of values for D is between 1.48
million and 18.2 million years (Table 3). The latter value is
physically impossible, as it is older than the age of pluton. In
other studies, the diffusion of Fe
+3
was measured independent
of O-vacancy migration. Diffusion coefficients measured in
these studies are larger [D 10
12
cm
2
/s in amethyst quartz,
Stock and Lehmann 1977; D 210
14
cm
2
/s in an SiO
2
layer
prepared by separation-by-implantation-of-oxygen (SIMOX)
technology, Kononchuk et al. 1998]. These larger diffusion
coefficients correspond to a shorter calculated duration of the
diffusion event (0.1 to 11.84 years). Diffusion coefficients at
lower temperatures (370440
o
C) have been investigated for
the diffusion of Fe in Na
2
O-Al
2
O
3
-Fe
2
O
3
-SiO
2
glass. These val-
ues fall within the range of the other reported diffusion coeffi-
cients (Table 4). These different diffusion coefficients span eight
orders of magnitude, which is a wide range of possible values
for D. The mechanism for Fe substitution in quartz in these
veins is unknown, so it is unclear which of these studies should
apply to the measured profiles. It is possible that none of these
studies is appropriate. It has been observed that diffusion coef-
ficients for O isotopes in quartz in wet hydrothermal experi-
ments are four to seven orders of magnitude larger than in dry
experiments (Joesten 1991). It is likely that the presence of
H
2
O also affects the diffusion of Fe by affecting either the rate
of the diffusion of O vacancies or the rate of diffusion of H
+
(which can also be responsible for balancing the charge defi-
ciency between Si
+4
and Fe
+3
). Therefore, intermediate values
for D (perhaps D 10
16
cm
2
/s) and t (correspondingly 1000
years) may be more appropriate.
A better-constrained value for the diffusion coefficient of
Fe in quartz under hydrothermal conditions is necessary to more
accurately constrain the duration of the diffusion event. The
significance of the duration of the diffusion event is that it would
represent the amount of time following crystallization of quartz
before the vein had cooled significantly enough for diffusion
of Fe to cease. This knowledge would contribute to the esti-
mates available for the duration of hydrothermal events. Re-
cent radiometric dating estimates of the duration of individual
alteration events using
40
Ar/
39
Ar methods range from ~10 000
to 500 000 years (Henry et al. 1997; Marsh et al. 1997). Nu-
merical modeling estimates range from < 10 000 to 800 000
years (Cathles et al. 1997; Marsh et al. 1997). One estimate
based on the exsolution of chalcopyrite and pyrrhotite from
sphalerite is 210 000 years (Mizuta and Scott 1997).
Even though the values for t are not well-constrained, the
values for the product Dt provide useful information for com-
paring the duration of diffusion in different veins. The Dt val-
ues lie within a narrow range from 3.2110
-10
to 7.4410
-10
m
2
,
less than one order of magnitude. Two different generations of
veining events may be represented by these samples. Sample
D2-5 is likely a distal vein related to the earlier quartz-magne-
tite-chalcopyrite stockwork, whereas the other samples (EK1-
17 and EK1-18) are from a distinctly later veining event
unrelated to the quartz-magnetite-chalcopyrite stockwork. The
similarity in time scales, as represented by similar Dt values,
indicates that the thermal conditions for the Cu ore-forming
vein events may not have been greatly different from those
during the non-ore-forming events.
Quartz-filled microfractures
In samples from every part of the GIC, irregular, dark-lu-
minescent bands clearly cut across and postdate the formation
TABLE 3. Calculated diffusion times for each profile using different diffusion coefficients of iron
Time (years)
Study 1 Study 2 Study 3 Study 4 Study 5 Study 6
Diffusion coefficients 1 10
12
cm
2
/s 4.3 10
14
cm
2
/s 2 10
14
cm
2
/s 1.5 10
15
cm
2
/s 6.9 10
20
cm
2
/s 1.3 10
20
cm
2
/s
EK1-18 Traverse 2 0.10 2.36 5.11 66.2 1 480 000 7 860 000
EK1-17 Traverse 3 0.19 4.29 9.28 120.2 2 690 000 14 300 000
EK1-17 Traverse 1 0.19 4.43 9.57 123.9 2 770 000 14 700 000
D2-5 Traverse 2 0.24 5.48 11.84 153.4 3 430 000 18 200 000
Notes: Studies defined as followed:
1 = Stock and Lehmann (1977) at 650 C.
2 = Khler and Frischat (1978)Glass III.2 at 440 C.
3 = Kononchuk et al. (1998)SIMOX at 900 C
4 = Khler and Frischat (1978)Glass IV.2 at 380 C.
5 = Kononchuk et al. (1998)BESOI at 600 C.
6 = Atkinson and Gardiner (1981) at 590 C.
FIGURE 11. Microscopic fracture, sample MGI6-1. Scanned CL
image shows an irregular darker band that cuts across across concentric
growth zones and grain boundaries in the lower left corner of the image.
PENNISTON-DORLAND: SCANNED CL VEIN QUARTZ TEXTURES 663
of the host crystals (Figs. 11 and 12). These bands comprise
only a minor volume of the quartz veins. Plane-polarized light
images from an optical microscope (Fig. 12b) show that lines
of inclusions, which commonly indicate healed fractures, are
uncommon. When viewed under crossed polarizers, these bands
consist of quartz in the same optical orientation as the surround-
ing crystal. Secondary-electron images confirm that the bands
are not boundaries between two different quartz crystals. These
bands usually emit red CL and are interpreted to be filled
microfractures. This type of feature has been observed in both
igneous rocks (Sprunt and Nur 1979; Seyedolali et al. 1997)
and sandstones (Onasch 1990; Milliken 1994a; Dickinson and
Milliken 1995; Laubach 1997).
There are two possible explanations for the origin of mi-
croscopic fracturing. The first is that it occurred during me-
chanical closing of the vein, when protruding quartz crystals
impinged upon one another. When the fluid in the fracture drains
or when fluid pressure inside a vein-forming fracture becomes
less than that required to hold the fracture open, the fracture
closes. Because some quartz-saturated fluid is still present, new
quartz is precipitated within the microfractures. Two observa-
tions directly support the hypothesis that some fracturing was
the result of mechanical closing of the open vein cavity. In
sample D2-25, the pointed termination of a large quartz crystal
on one side of the vein is truncated against the smaller quartz
crystals on the other side of the vein. The large truncated crys-
tal has many subgrains within it. When examined with scanned
CL, it becomes clear that these subgrains are separated by
quartz-filled microfractures. In sample EK1-17, a quartz crys-
tal near the center of the vein has been fractured, the top part
has rotated with respect to the bottom part, and there is a wedge-
shaped region of dark luminescence emanating from the frac-
ture (Fig. 8a). No correlation was found between Al or Ti
contents and microfractures. As discussed in the previous sec-
tion, however, microfractures that contain sulfides may be sur-
rounded by a dark-luminescent zone rich in Fe.
A second explanation for microscopic fracturing are local
movements associated with the late strike-slip faulting. In many
places, veins are reactivated as minor fault zones (Sapiie 1998).
This process probably accounts for some of the microfractures
observed near vein margins, in particular those that appear to
be micro-extension fractures (Figs. 11 and 12). Both vein clo-
sure and tectonic movements likely contribute to the
microfracturing observed in the GIC.
Truncation of concentric growth zones.
Some quartz displays truncation of concentric growth zones
by regions of homogeneous luminescence near grain bound-
aries (Fig. 13). These regions are most prominent in quartz crys-
tals near vein margins, but occur throughout the vein, including
in crystals near the center of the vein. In addition, most quartz
crystals studied show dark CL along grain boundaries. These
features were observed in all samples, except the faintly lumi-
nescent veins of the Dalam Igneous Complex (see Table 2).
There are several possible ways to create truncated growth
zones. Uniform luminescence may have formed by local dis-
solution and reprecipitation of quartz during closure of the
mineralized fracture or later stresses. Such pressure solution
could explain the indented shapes of some of these areas (Fig.
13a). A front of dissolution and reprecipitation might have
moved from the edge of the crystal inward. This mechanism
could explain why none of the crystals appear to be distorted,
and is believed to operate in isotope and cation exchange in
feldspars (ONeil and Taylor 1967). Diffusion into crystals from
grain boundaries is another possible explanation. No correla-
tion was found among the concentrations of Al, Fe, or Ti and
the truncated growth zones.
Vein history
In the two veins for which large photomosaic traverses were
created, a history of events can be constructed using the tex-
tures and cross-cutting relationships observed. Figure 14 illus-
trates a likely sequence of events in the history of a
quartz-sulfide vein as interpreted from the CL observations:
FIGURE 12. Microscopic fracture, sample EK1-17. (a) Scanned CL image. Irregular dark-luminescent band cuts the middle of photo. (b)
Plane-polarized photomicroscope image of same region. Irregular band is not visible in this photo.
PENNISTON-DORLAND: SCANNED CL VEIN QUARTZ TEXTURES 664
(1) The fracture opens. (2) Fluids pass through the widening
fracture and precipitate quartz in random orientations (Figs. 5
and 6). (3) Changes in fluid composition, pressure, tempera-
ture, precipitation rate, and/or flow rate create concentric growth
zones (Fig. 4). (4) Quartz crystals with c-axes oriented toward
the center of the fracture overtake crystals that have other ori-
entations (Figs. 5 and 6). (5) Rapid precipitation of quartz or
turbulence in the fluid creates a zone of inclusions and turbid
zoning (Fig. 7). (6) Quartz precipitation continues and eventu-
ally ceases. Iron from remaining fluid diffuses into quartz crys-
tals from the center of the vein (Fig. 8). (7) The vein closes,
fracturing quartz crystals where they impinge on each other
(Fig. 8a). Concentric growth zones are truncated by pressure
solution and reprecipitation or by diffusion (Fig. 13). Iron con-
tinues to diffuse into quartz nearest the vein center. (8) Precipi-
tation of sulfide minerals fills the interstices of the vein. Iron
may continue to diffuse into quartz. (9) Additional fracturing
occurs during later tectonic movements (Figs. 11 and 12).
This history was developed to explain the two photomosaic
traverses created on two young veins from the LKI. It is plau-
sible that similar histories apply to the veins from other vein-
ing events in the GIC, as similar features were found in nearly
all the veins examined (Table 3). The reason these youngest
veins were selected for creating photomosaic traverses is their
bright luminescence. It is uncertain why these younger veins
have a distinctly brighter luminescence. It is possible that the
different brightness is due to different trace-element abundances
in the later fluids. It is also a possibility that older veins were
subjected to a longer period of high temperatures, and that dif-
fusion or other crystal-chemical processes muted the lumines-
cence.
SUMMARY REMARKS
Vein quartz from the Grasberg Igneous Complex displays
spectacular textures on very small scales when observed using
scanned CL. Individual features observed along photomosaic
FIGURE 13. Truncation of concentric growth zones. Crystals display concentric growth zoning in centers, but not at edges. (a) Scanned CL
image, sample EK1-17. (b) Scanned CL image, sample RDC1-28.
FIGURE 14.
Schematic diagram of
sequence of events
within a single vein.
Black arrows indicate
fluid flow direction.
Black dots represent
fluid inclusions, light gray represents quartz with elevated Fe
concentrations, dark gray represents sulfide crystals. See text for
discussion.
PENNISTON-DORLAND: SCANNED CL VEIN QUARTZ TEXTURES 665
traverses from vein margins to vein centers coupled with trace-
element analyses from the electron microprobe allow the inter-
pretation of a detailed history of fracture opening and infilling
to be made for individual veins. The profiles of smoothly de-
creasing Fe contents of quartz crystals near sulfide mineral
grains fit the solution to the diffusion equation well, providing
the potential for evaluating the duration of the diffusion event
if the appropriate diffusion coefficient for Fe in quartz were
known. Values for the product Dt for different veins are useful
for comparing the relative time scales of different stages of
veining. The values of Dt for the veins analyzed are similar, all
falling within one order of magnitude, suggesting similar ther-
mal histories for veins from different stages of the history of
the GIC.
Scanned CL reveals numerous textures of quartz crystals
pertinent to the history of vein formation in porphyry Cu ore
deposits that were previously unnoticed. The different features
observed may be used along with other information, such as
fluid inclusion data and electron microprobe analyses, to bet-
ter understand the changing chemistry of ore-forming fluids
and the tectonic processes which enable fluid movement, as
well as potentially constraining the duration of hydrothermal
events. The advantages of this technique include small scales
of observation, the relatively stable operating conditions of the
SEM, and the ability to detect faint levels of CL emitted. The
main disadvantage of the images presented here is the lack of
CL color information. This deficiency can be corrected using
monochromator attachments to CL devices, which allow for
the collection of color information using scanned CL. Scanned
cathodoluminescence, therefore, in conjunction with other tech-
niques, is essential for greater understanding of the processes
that contributed to the formation of porphyry Cu deposits of
such a large magnitude as the Grasberg Igneous Complex.
ACKNOWLEDGMENTS
Thanks to Mark Cloos, John Ferry, Kitty Milliken, and Boswell Wing for
comments and suggestions made during the writing of this paper. Constructive
reviews by Virginia Sisson and Sorena Sorensen were helpful in refining this
paper. Thanks to Kitty Milliken, Fangqiong Lu, and Sorena Sorensen for assis-
tance with SEM CL, the electron microprobe, and CL photography with a
luminoscope. Thanks to Jim Bob Moffett, Steve Van Nort, Dave Potter, Kris
Hefton, Art Ono, Dave Mayes, Jay Pennington, and the other geologists and
staff of Freeport, Indonesia, Inc. for making this research possible. The research
was supported by a grant from Freeport McMoRan, Inc. Ertsberg Project Con-
tribution No. 17. This material is based upon work supported under a National
Science Foundation Graduate Research Fellowship.
REFERENCES CITED
Atkinson, A. and Gardner, J.W. (1981) The diffusion of Fe
3+
in amorphous SiO
2
and
the protective properties of SiO
2
layers: Corrosion Science, 21, 4958.
Beane, R.E. and Titley, S.R. (1981) Porphyry copper deposits, Part II. Hydrother-
mal alteration and mineralization. In Skinner, B.J., Ed., Economic Geology
Seventy-Fifth Anniversary, p. 235269. The Economic Geology Publishing
Company, Lancaster, Pennsylvania.
Bevington, P.R. and Robinson, D.K. (1992) Data Reduction and Error Analysis for
the Physical Sciences, 2 ed., 328 p. McGraw-Hill, Boston.
Boiron, M.C., Essarraj, S., Sellier, E., Cathelineau, M., Lespinasse, M., and Poty, B.
(1992) Identification of fluid inclusions in relation to their host microstructural
domains in quartz by cathodoluminescence. Geochimica et Cosmochimica Acta,
56, 175185.
Cathles, L.M., Erendi, A.H.J., and Barrie, T. (1997) How long can a hydrothermal
system be sustained by a single intrusive event? Economic Geology and the
Bulletin of the Society of Economic Geologists, 92, 766771.
Crank, J. (1975) The Mathematics of Diffusion, 414 p. Clarendon Press, Oxford.
Demars, C., Pagel, M., Deloule, E., and Blanc, P. (1996) Cathodoluminescence of
quartz from sandstones: Interpretation of the UV range by determination of
trace element distributions and fluid-inclusion P-T-X properties in authigenic
quartz. American Mineralogist, 81, 891901.
Dengler, L.A. and Sprunt, E.S. (1977) Cathodoluminescence of some quartzites
from the Bergell Alps. Eos, 58, 1239.
Dickinson, W.W. and Milliken, K.L. (1995) The diagenetic role of brittle deforma-
tion in compaction and pressure solution, Etjo Sandstone, Namibia. The Jour-
nal of Geology, 103, 339347.
Hale, D.R. (1948) The laboratory growing of quartz. Science, 107, 393394.
Hamilton, W. (1979) Tectonics of the Indonesian region, 345 p. U.S. Geological
Survey Professional Paper 1078.
Henry, C.D., Elson, H.B., McIntosh, W.C., Heizler, M.T., and Castor, S.B. (1997)
Brief duration of hydrothermal activity at Round Mountain, Nevada, determined
from
40
Ar/
39
Ar geochronology. Economic Geology and the Bulletin of the Soci-
ety of Economic Geologists, 92, 807826.
Hogg, A.J.C., Sellier, E., and Jourdan, A.J. (1992) Cathodoluminescence of quartz
cements in Brent Group sandstones, Alwyn South, UK North Sea. In A.C.
Morton, R.S. Haszeldine, M.R. Giles, and Brown, S., Eds., Geology of the
Brent Group, p. 421440. Geological Society Special Publication 61, Geologi-
cal Society of London.
Janssen-van Rosmalen, R. and Bennema, P. (1977) The role of hydrodynamics and
supersaturation in the formation of liquid inclusions in KDP. Journal of Crystal
Growth, 42, 224227.
Joesten, R. (1991) Grain-boundary diffusion kinetics in silicate and oxide minerals,
in Ganguly, J., ed., Diffusion, Atomic Ordering, and Mass Transport: Selected
Topics in Geochemistry, p. 345395. Springer-Verlag, New York.
Joshi, M.S. and Paul, B.K. (1974) Effect of supersaturation and fluid shear on the
habit and homogeneity of potassium dihydrogen phosphate crystals. Journal of
Crystal Growth, 22, 321327.
Khler, W. and Frischat, G.H. (1978) Iron and sodium diffusion in silicate glasses.
Physics and Chemistry of Glasses, 19, 103107.
Kononchuk, O., Korablev, K.G., Yarykin, N., and Rozgonyi, G.A. (1998) Diffusion
of iron in the silicon dioxide layer of silicon-on-insulator structures. Applied
Physics Letters, 73, 12061208.
Kretz, R. (1983) Symbols for rock-forming minerals. American Mineralogist, 68,
277279.
Kyle, J.R. and Bodnar, R.J. (1995) Fluid inclusion studies of porphyry-skarn ore
formation, Gunung Bijih (Ertsberg) District, Irian Jaya, Indonesia. Geological
Society of America Abstracts with Programs, 27, p. A65.
Laubach, S.E. (1997) A method to detect natural fracture strike in sandstone: American
Association of Petroleum Geologists Bulletin, 81, 604623.
Le Maitre, R.W. (1982) Numerical Petrology: Amsterdam, Elsevier Scientific Pub-
lishing Company, 281 p.
MacDonald, G.D. and Arnold, L.C. (1994) Geological and geochemical zoning of
the Grasberg Igneous Complex, Irian Jaya, Indonesia. Journal of Geochemical
Exploration, 50, 143178.
Marsh, T.M., Einaudi, M.T., and McWilliams, M. (1997)
40
Ar/
39
Ar geochronology
of Cu-Au and Au-Ag mineralization in the Potrerillos District, Chile. Economic
Geology and the Bulletin of the Society of Economic Geologists, 92, 784806.
Marshall, D. J. (1988) Cathodoluminescence of geological materials, 146 p., Unwin
Hyman, Boston.
Matter, A. and Ramseyer, K. (1985) Cathodoluminescence microscopy as a tool for
provenance studies of sandstones. In G.C. Zuffa, Ed., Provenance of arenites,
p. 191211. D. Reidel, Dordrecht.
McDowell, F., McMahon, T.P., Warren, P.Q., and Cloos, M. (1996) Pliocene Cu-Au
bearing igneous intrusions of the Gunung Bijih (Ertsberg) District, Irian Jaya,
Indonesia: K-Ar geochronology. The Journal of Geology, 104, 327340.
McMahon, T.P. (1994a) Pliocene intrusions in the Gunung Bijih (Ertsberg) Mining
District, Irian Jaya, Indonesia: petrography and mineral chemistry. International
Geology Review, 36, 820849.
(1994b) Pliocene intrusions in the Gunung Bijih (Ertsberg) Mining District,
Irian Jaya, Indonesia: major- and trace-element chemistry. International Geol-
ogy Review, 36, 925946.
(1994c) Pliocene Intrusions in the Ertsberg (Gunung Bijih) Mining District,
Irian Jaya, Indonesia: Petrography, geochemistry, and tectonic setting, 298 p.
Ph.D. dissertation. The University of Texas at Austin.
Mealey, G.A. (1996) Grasberg, 384 p. Freeport-McMoran Copper & Gold Inc.,
New Orleans, Louisianna.
Milliken, K.L. (1994a) The widespread occurrence of healed microfractures in
siliciclastic rocks: Evidence from scanned cathodoluminescence imaging. In
P.P. Nelson and S.E. Laubach, Eds., Proceedings of the 1st North American
Rock Mechanics Symposium, p. 825832. The University of Texas at Austin.
(1994b) Cathodoluminescent textures and the origin of quartz silt in Oli-
gocene mudrocks, South Texas. Journal of Sedimentary Research, A64, 567
571.
Mizuta, T. and Scott, S. (1997) Kinetics of iron depletion near pyrrhotite and chal-
copyrite inclusions in sphalerite: The sphalerite speedometer. Economic Geol-
ogy and the Bulletin of the Society of Economic Geologists, 92, 772783.
ONeil, J.R. and Taylor Jr., H.P. (1967) The oxygen isotope and cation exchange
PENNISTON-DORLAND: SCANNED CL VEIN QUARTZ TEXTURES 666
chemistry of feldspars. American Mineralogist, 52, 14141437.
Onasch, C.M. (1990) Microfractures and their role in deformation of a quartz arenite
from the central Appalachian foreland: Journal of Structural Geology, 12, 883
894.
Onasch, C.M. and Vennemann, T.W. (1995) Disequilibrium partitioning of oxygen
isotopes associated with sector zoning in quartz. Geology, 23, 11031106.
Owen, M.R. and Carozzi, A.V. (1986) Southern provenance of upper Jackfork Sand-
stone, southern Ouachita Mountains: Cathodoluminescence petrology. Geologi-
cal Society of America Bulletin, 97, 110115.
Penniston-Dorland, S.C. (1997a) Scanned cathodoluminescence of quartz and in-
terpretation of textures in quartz-sulfide veins in the Grasberg Igneous Com-
plex, a porphyry copper ore deposit. Geological Society of America, Abstracts
with Programs, 29, A61.
(1997b) Veins and Alteration Envelopes in the Grasberg Igneous Complex,
Gunung Bijih (Ertsberg) District, Irian Jaya, Indonesia, 402 p. M.S. thesis. The
University of Texas at Austin.
Quarles van Ufford, A.I. (1996) Stratigraphy, Structural Geology, and Tectonics of a
Young Forearc-Continent Collision, Western Central Range, Irian Jaya (West-
ern New Guinea), Indonesia, 420 p. Ph.D. dissertation. The University of Texas
at Austin.
Ramseyer, K., Baumann, J., Matter, A., and Mullis, J. (1988) Cathodoluminescence
colours of alpha-quartz. Mineralogical Magazine, 52, 669667.
Roedder, E. (1984) Fluid Inclusions, 646 p. Reviews in Mineralogy, Mineralogical
Society of America, Washington, D.C.
Sapiie, B. (1998) Strike-Slip Faulting, Breccia Formation and Porphyry Cu-Au Min-
eralization in the Gunung Bijih (Ertsberg) Mining District, Irian Jaya, Indone-
sia, 304 p. Ph.D. dissertation. The University of Texas at Austin.
Seyedolali, A., Krinsley, D. H., Boggs, Jr., S., OHara, P. F., Dypvik, H., and Goles,
G. G. (1997) Provenance interpretation of quartz by scanning electron micro-
scope-cathodoluminescence fabric analysis. Geology, 25, 787790.
Seyedolali, A., Reed, M., Rusk, B., and Goles, G.G. (1998), Dissolution and healing
fabrics of vein quartz in the Butte, Montana porphyry copper deposit revealed
by cathodoluminescence. Geological Society of America Abstracts with Pro-
grams, 30, p. A370.
Sippel, R.F. (1968) Sandstone petrology, evidence from luminescence petrography.
Journal of Sedimentary Petrology, 38, 530554.
Smith, J.V. and Stenstrom, R.C. (1965) Electron-excited luminescence as a petro-
logic tool. Journal of Geology, 73, 627635.
Sprunt E.S. (1978) Effects of impurities on quartz cathodoluminescence. Eos, 59,
1216.
(1981) Causes of quartz cathodoluminescence colors. Scanning Electron
Microscopy, 1, 525535.
Sprunt, E.S. and Nur, A. (1979) Microcracking and healing in granites: New evi-
dence from cathodoluminescence. Science, 205, 495497.
Sprunt, E.S., Dengler, L.A., and Sloan, D. (1978) Effects of metamorphism on quartz
cathodoluminescence. Geology, 6, 305308.
Stock, H.D. and Lehmann, G. (1977) Phenomena associated with diffusion of triva-
lent iron in amethyst quartz: The Journal of Physics and Chemistry of Solids,
38, 243246.
Thomas, L.A., Wooster, N., and Wooster, W.A. (1949) The hydrothermal synthesis
of quartz: Discussions of the Faraday Society: Crystal Growth, 341345.
Titley, S.R. (1982) The style and progress of mineralization and alteration in por-
phyry copper systems, American Southwest. In S.R. Titley, Ed., Advances in
Geology of the Porphyry Copper Deposits, p. 93116. The University of Ari-
zona Press, Tucson.
Valley, J.W. and Graham, C.M. (1996) Ion microprobe analysis of oxygen isotope
ratios in quartz from Skye granite: healed microcracks, fluid flow, and hydro-
thermal exchange. Contributions to Mineralogy and Petrology, 124, 225234.
Van Nort, S.D., Atwood, G.W., Collinson, T.B., Flint, D.C., and Potter, D.R. (1991)
Geology and mineralization of the Grasberg porphyry copper-gold deposit, Irian
Jaya, Indonesia. Mining Engineering, 43, 300303.
Van Praagh, G. (1947) Synthetic quartz crystals. Geological Magazine, 84, 98100.
(1949) The hydrothermal crystallization of vitreosil at constant tempera-
ture: Discussions of the Faraday Society. Crystal Growth, 338341.
Watt, G.R., Wright, P., Galloway, S., and McLean, C. (1997) Cathodoluminescence
and trace element zoning in quartz phenocrysts and xenocrysts: Geochimica et
Cosmochimica Acta, 61, 43374348.
Weiland, R.J. and Cloos, M. (1996) Pliocene-Pleistocene asymmetric unroofing of
the Irian fold belt, Irian Jaya, Indonesia: Apatite fission-track thermochronology.
Geological Society of America Bulletin, 108, 14381449.
Zinkernagel, U. (1978) Cathodoluminescence of quartz and its application to sand-
stone petrology. Contributions to Sedimentology, 8, 69 p.
MANUSCRIPT RECEIVED DECEMBER 6, 1999
MANUSCRIPT ACCEPTED JANUARY 23, 2001
MANUSCRIPT HANDLED BY WALTER E. TRZCIENSKI, JR.

Anda mungkin juga menyukai