American Mineralogist, Volume 90, pages 122131, 2005

Relationships between SEM-cathodoluminescence response and trace-element composition

of hydrothermal vein quartz
MARIANNE R. LANDTWING AND THOMAS PETTKE*

Isotope Geochemistry and Mineral Resources, Swiss Federal Institute of Technology, ETH Zentrum NO, CH-8092 Zrich, Switzerland

ABSTRACT
Laser-ablation ICP-MS data of hydrothermal vein quartz with zonation in scanning electron micro-
scope cathodoluminescence reveal two groups of trace elements, one that co-varies in concentration
with luminosity and another that remains uniform throughout. Bright luminosity correlates with a high
total abundance of trace elements in early quartz, including Al (up to 410 g/g), Ti (up to 240 g/g),
K (up to 330 g/g), Li (up to 8 g/g), Na (up to 36 g/g), and Fe (up to 20 g/g). Up to 20 times
lower concentrations of these elements are associated with dark luminosity in late quartz. Concentra-
tions of P (~21 g/g), Ga (~0.3 g/g), Ge (~1.3 g/g), Sn (~1.5 g/g), Cu (~0.3 g/g), and Ag (~0.1
g/g) demonstrate no relationship with luminosity. Charge balance cannot be achieved for bright
luminescent quartz unless the presence of up to 6 g/g H+ (not analyzed by LA-ICPMS) or interstitial
Al3+ is invoked. Interestingly, the inferred ~6 g/g H+ remains constant for quartz containing more
than 10 mol total trace elements and might represent the solubility of H+ in hydrothermal quartz at
temperatures between 450 and 700 C.
LA-ICPMS results indicate that the fluid chemistry determines the quartz trace-element pattern,
which may serve as a monitor for the chemical environment from which quartz crystallized. Ratios of
Na/Al and Ti/Al are uniform, whereas K/Al evolves toward lower values with decreasing luminosity.
Combined evidence, including quartz vein textures, silica solubility data, and estimates of the tem-
perature of quartz growth, suggest that early quartz crystallized fast but late quartz grew more slowly
at 350 to 425 C. It is speculated that the higher the quartz growth rate is, the more trace elements
are incorporated into quartz, besides the temperature dependence known from the literature. Clearly,
growth rate, temperature and fluid chemistry are important parameters affecting luminosity; however,
their relative importance may vary from case to case.

INTRODUCTION Graham 1996; Wilkinson et al. 1999; Wilkinson and Johnston

1996). For certain applications, such as the provenance studies
Cathodoluminescence (CL) microscopy is an efficient method
of clastic sediments, CL-colors are essential (e.g., Zinkernagel
for visualizing microtextures in minerals that are invisible in
1978), and the combination of SEM-CL with conventional CL
conventional transmitted-light microscopy (e.g., Gtze et al.
(Herzog et al. 1970) may give the best results in such cases. For
2001; Pagel et al. 2000). Therefore, CL may aid in establishing
petrographic and paragenetic studies of magmatic and hydro-
the succession of paragenetic associations in minerals and rocks,
thermal quartz samples, however, SEM-CL provides excellent
such as growth zoning, inherited cores, healed microfractures,
petrographic resolution.
and dissolution or alteration features. Whereas CL has already
Variability in luminescence colors and in CL emission bands
become an essential tool in zircon U-Pb geochronology, extensive
can result from intrinsic (structural) defects, which include trans-
applications to other minerals, e.g., quartz, are only emerging.
lational errors, point defects, and inclusions of minerals, gas or
Scanning electron microscope (SEM)-CL offers several advan-
liquid (Gtze et al. 2001), as well as from different extrinsic
tages when compared to conventional CL, including: (1) a good
defects, i.e., the incorporation of lattice-bound trace elements.
spatial resolution (down to single micrometer scale) due to the
Quartz has a strong configuration of Si-O bonds and limited in-
small size of the electron beam; (2) the ability to detect small
terstitial space, allowing incorporation of only a limited amount
variations (g/g range) at low trace-element concentrations; and
of other elements into its structure. The concentration of such
(3) the capability of combining CL with back-scattered electron
substitutional or interstitial impurities generally does not exceed a
(BSE) or secondary electron (SE) mode, or energy dispersive
few hundred g/g and varies significantly in natural quartz (e.g.,
spectroscopy (EDS) microanalysis. The combination of routine
Bambauer 1961; Audtat and Gnther 1999; Flem et al. 2002;
CL observations with SEM seems to be especially favorable for
Monecke et al. 2002). The trace-element inventory of quartz
the interpretation of growth textures in quartz characterized by
depends in a complex way on many parameters, among which
generally low trace-element contents (e.g., Boiron et al. 1992;
the chemical composition of the quartz-forming environment,
Cathelineau et al. 1993; Parnell et al. 2000; Penniston-Dorland
temperature, pH, oxygen fugacity, and growth rate have been
2001; Redmond et al. 2001; Rusk and Reed 2002; Valley and
considered most important (e.g., Gtze et al. 2001; Pagel et al.
* E-mail: thomas.pettke@erdw.ethz.ch 2000). Because of the high sensitivity of CL, only a few analyti-

0003-004X/05/0001122$05.00/DOI:10.2138/am.2005.1548 122
 LANDTWING AND PETTKE: LA-ICPMS TRACE ELEMENTS AND SEM-CL OF HYDROTHERMAL QUARTZ
cal methods are currently capable of determining the trace-ele-
ment abundance variations that may influence the CL response
in quartz, including proton-induced X-ray emission (PIXE),
secondary ion mass spectrometry (SIMS) and, most recently,
laser-ablation inductively coupled-plasma mass-spectrometry
(LA-ICPMS). For the most abundant elements, such as Al, Ti,
and possibly Fe, the electron microprobe (EMP) also can be
used in some cases.
Several studies have explored the relationships between CL
response and the trace-element inventory of quartz crystals (e.g.,
Gtze et al. 2001; Marshall 1988; Mller et al. 2000, 2002;
Penniston-Dorland 2001; Ramseyer and Mullis 1990; Stevens
Kalceff et al. 2000, and references therein). There is a general
consensus that luminosity variations in quartz relate to compo-
sitional zoning in H+, OH-, P5+, Ge4+, Ti4+, Al3+, Ga3+, Fe2+,3+, Li+,
Na+, and K+, yet the details of these dependencies have remained
largely unresolved. This uncertainty may be due in part to the
fact that not all trace elements present at concentrations of more
than a few hundred ng/g have been determined for the same ana-
lytical areas. Thus the correlation of CL responses with selected
trace-element concentrations as reported in the literature might
not be as straightforward, because the CL response may have
been significantly influenced by elements that were traditionally
not determined.
In the present study, we have used LA-ICPMS to quantify
the distribution of 12 trace elements in quartz (out of 25 ele-
ments tested), with the advantage of data acquisition for all
elements on each spot under identical conditions and low limits
of detection (LOD). The aim was to constrain the trace-element
distribution in hydrothermal vein quartz and its relation to the
SEM-CL response. Conventional CL has not produced any dis-
tinct variability for these samples from the Bingham porphyry
Cu-Au-Mo deposit (P. Redmond, personal communication 2000).
Quartz trace-element data are then related to the chemistry and
temperature of the ore-forming fluids at the Bingham porphyry
Cu-Au-Mo deposit, allowing a consideration of parameters that
are commonly believed to influence the incorporation of trace
elements into the quartz structure, and hence may directly relate
to the CL response of quartz.
GEOLOGICAL SETTING AND SAMPLES STUDIED
Hydrothermal quartz samples related to bornite-chalcopy-
rite-digenite mineralization were selected from stockwork veins
of the central, high-grade orebody of the Bingham porphyry
Cu-Au-Mo deposit. The deposit is centered on a series of late
Eocene porphyry dikes of quartz monzonitic, latitic, and quartz
latitic composition (e.g., Redmond et al. 2001), intruded into
older equigranular monzonite and a thick succession of Pennsyl-
vanian quartzites and silty limestones (Lanier et al. 1978). The
high-grade gold ore (>1 ppm Au) forms a cigar-shaped body
whereas the copper orebody, as defined by the 0.35 wt% Cu
cutoff grade, displays an inverted cup-shaped geometry (Phil-
lips et al. 1997).
More than a dozen hydrothermal quartz stockwork vein
samples containing bornite, digenite, and chalcopyrite were
investigated for their CL characteristics to identify the quartz
generation that crystallized during ore mineral precipitation.
The fluids forming these veins were in equilibrium with hydro-
123
thermal K-feldspar and biotite, as these minerals either occur
in the vein selvages or are abundant in the pervasively altered
host rock. SEM-CL studies of these samples revealed two types
of quartz (Fig. 1), implying a two-stage vein history, with early
quartz veining (Q1) formed in response to hydrofracturing of
the host rock, followed by reopening and void- and fracture-
filling quartz (Q2). Quartz Q1 is bright to gray luminescent,
forms the bulk of the veins, and consists of subhedral to an-
hedral quartz grains, some of which exhibit euhedral growth
zoning. Quartz Q2 is dark luminescent, closely associated with
Cu-Fe sulfide deposition, and occurs in a distinctive veinlet and
microfracture network, locally truncating Q1 quartz growth
zones, or as vug fillings interstitial to Q1 quartz grains. Tran-
sitions in CL response between Q1 and Q2 quartz are usually
abrupt. The microtextural and CL characteristics of Q1 and Q2
quartz are consistent for all quartz stockwork veins from the
center of the Bingham orebody and can be used to reconstruct
the fluid evolution of the central part of the Bingham deposit
(Landtwing et al. 2002; 2004).
Laser-ablation ICPMS analyses focused on one typical quartz
stockwork vein (sample D 211-19) originating from drill hole
number 211 at a depth of 5.8 meters (19 feet). This quartz vein
is hosted by the first mineralizing intrusion (quartz monzonite
porphyry), which shows intense potassic alteration. Secondary
K-feldspar forms the matrix of the host rock and replaces plagio-
clase phenocrysts and magmatic biotite sometimes too. Shreddy
mica with phlogopitic composition (Bowman et al. 1987) occurs
in hairline veinlets and replaces amphibole and also magmatic
biotite in part.
METHODS
SEM-CL
SEM-CL images were obtained from doubly polished 150 to 300 m thick
sections with a CamScan CS44LB instrument (Interessengemeinschaft fr Elek-
tronenmikroskopie, ETH Hnggerberg, Zrich). Back-scattered electron (BSE)
and CL images were taken immediately after each other under the same analyti-
cal conditions (acceleration voltage of 15 kV, beam-current density of 10 to 15
nA/mm, untilted sample adjusted in height to the lower edge of the ellipsoidal CL
mirror) using an EDAX Phoenix digital image acquisition system. The SEM is
equipped with a 4-quadrant semiconductor BSE detector (KE Developments Ltd.)
and with a motor-driven elliptical mirror focusing the CL signal onto a standard
photomultiplier detector. For the SEM analyses, the samples were coated with a
thin film of carbon to prevent charging during irradiation. This coating was removed
for LA-ICPMS analysis.
Normalization of CL spectra gray values
The SEM-CL apparatus employed here is not equipped to quantify the wave-
lengths and the intensities of the CL spectra but only records the CL signals as
gray scale images. To compare CL response at the location of 20 to 40 m LA pits,
the average gray value of these zones of interest was quantified using the program
Digital Color Meter from Apple Computer. The resulting values correspond to the
average gray value of thirteen 13 13 pixel squares on a 1024 768 pixel computer
screen and range between 0% (white) and 100% (black). The brightness and the
intensity of the CL signal strongly depends on a series of analytical variables,
including the thickness of the carbon coating, the adjustment of the sample relative
to the ellipsoidal CL mirror, the magnification used for image acquisition, filament
drift, and beam exposure time. Therefore, the measured gray values can be related to
each other directly only for a single CL image acquisition. The comparison of gray
values acquired at different imaging sites then requires an internal standardization
of the CL response. This normalization was done assuming that the average gray
value of the uniform Q2 quartz precipitated during the last stage of quartz deposi-
tion is identical for all CL images, namely 70.9% (Table 1).
This leads to the following relation:
124 LANDTWING AND PETTKE: LA-ICPMS TRACE ELEMENTS AND SEM-CL OF HYDROTHERMAL QUARTZ

A GVX,1, meas
Q2 qtz
+ GVX,2, meas
Q2 qtz
+ + GVX,n, meas
Q2 qtz
GV Xi , norm = GV Xi , meas u ref (1) AX = (3)
AX n
where GV refers to the normalized (subscript norm) or measured (subscript meas)
1, Q2 qtz 2, Q2 qtz n, Q2 qtz gray value of the superscripted measurement (1 = 1st, 2 = 2nd, , i = ist, , n =
GVref, meas
+ GVref, meas
+ + GVref, meas
Aref = (2) nth measurement), originating from the subscripted imaging area (X denotes the
n imaging area of interest, ref the reference imaging area, to which all other imaging
and areas were normalized). A corresponds to the arithmetic average of all (n) gray
values measured in Q2 quartz (superscript Q2 qtz) in the reference imaging area

FIGURE 1. Images of porphyry-Cu vein quartz samples D 40462.5 (a) and D 21119 (b, c) acquired by SEM-BSE (left) and SEM-CL (right).
The bright-luminescent hydrothermal quartz generation Q1, forming the stockwork veins, is overgrown or truncated by dark-luminescent Q2 quartz
associated with the bulk of Cu ore deposition. Abbreviations: qtz = quartz, bn = bornite, dg = digenite, and cpy = chalcopyrite.
 LANDTWING AND PETTKE: LA-ICPMS TRACE ELEMENTS AND SEM-CL OF HYDROTHERMAL QUARTZ 125

TABLE 1. LA-ICPMS element concentration data for hydrothermal quartz generations from a stockwork vein from the center of the Bingham
porphyry Cu-Au-Mo deposit, Utah
LA-ICPMS Quartz Gray Gray Li Na Al SiO2 P K Ti Fe Cu Ga Ge Ag Sn Total
Short type value value oxides
number measured normalized* (g/g) (g/g) (g/g) (wt%) (g/g) (g/g) (g/g) (g/g) (g/g) (g/g) (g/g) (g/g) (g/g) (wt%)

analyzed 7 23 27 29 31 39 49 57 65 71 73 107 118

isotope
Sample D 211-19, area K
db09e08 Q1 43.8 46.7 5.0 20 386 100 30 233 214 16 0.16 0.39 2.5 <0.08 1.6 100.148
db09e09 Q1 53.4 57.0 4.8 17 404 100 20 265 237 8.6 0.25 0.24 1.5 0.07 1.7 100.156
db09e10 Q1 57.6 61.5 4.8 13 312 100 16 187 179 8.9 1.0 0.26 0.78 <0.06 1.9 100.118
db09e11 Q1 58.8 62.8 4.6 14 309 100 23 191 177 17 <0.11 0.24 2.1 <0.05 1.8 100.120
db09e12 Q1 57.3 61.2 6.3 6.3 181 100 15 52 101 3.6 0.16 0.24 0.86 0.05 1.6 100.063
db09e13 Q1 60.5 64.6 5.0 2.9 138 100 28 27 78 7.6 0.25 0.20 1.4 <0.09 2.1 100.051
db09e14 Q1 56.5 60.3 5.2 2.4 152 100 22 30 86 6.8 <0.13 0.20 1.3 <0.08 2.0 100.054
db09e15 Q1 41.9 44.7 5.1 36 94 100 18 16 184 8.5 0.43 0.24 1.5 <0.07 1.2 100.061
db09e16 Q1 35.4 37.7 4.3 6.8 265 100 28 105 151 4.9 <0.12 0.35 1.2 <0.07 1.3 100.097
db09e17 Q1 43.5 46.4 5.4 17 306 100 20 157 180 <4.1 <0.15 0.31 0.93 0.08 <0.69 100.115
db09e18 Q1 56.1 59.8 7.4 5.4 112 100 21 22 77 <4.3 <0.10 0.35 <0.63 <0.08 1.6 100.044
db09f03 Q1 58.1 62.0 5.4 3.7 109 100 24 17 78 <4.1 <0.13 0.15 0.90 0.10 1.9 100.043
db09e03 Q2 68.1 72.7 2.9 3.5 50 100 25 2.6 12 <3.7 0.24 0.31 1.1 0.13 0.92 100.019
db09e04 Q2 65.8 70.2 2.6 1.0 48 100 25 1.6 39 <3.8 0.13 0.22 0.68 <0.08 1.3 100.023
db09e05 Q2 65.4 69.8 3.0 0.6 45 100 25 <1.1 27 <4.4 0.17 0.18 2.1 <0.08 1.2 100.020
db09e06 Q2 65.9 70.4 2.4 <0.5 52 100 20 1.6 27 <3.9 <0.13 0.28 1.4 <0.08 1.0 100.020
db09e07 Q2 66.7 71.2 2.8 1.4 45 100 23 <1.0 23 <3.9 0.15 0.25 2.3 <0.07 1.7 100.019
Average
in Q2 qtz Ak = 66.4 70.9

Sample D 211-19, area B

db10a04 Q1 53.4 54.2 5.5 8.9 167 100 17 54 116 20 0.40 0.30 1.1 <0.05 1.9 100.064
db10a05 Q1 50.8 51.5 4.5 6.7 222 100 21 81 123 <5.3 <0.14 0.28 0.94 <0.09 1.6 100.079
db10a06 Q1 48.0 48.7 5.7 15 361 100 20 201 198 13 <0.14 0.35 0.75 <0.08 1.4 100.133
db10a07 Q1 41.3 42.0 3.7 23 408 100 19 330 220 11 0.16 0.26 0.96 <0.08 1.6 100.162
db10a08 Q1 49.8 50.5 5.8 6.3 266 100 24 87 148 6.9 <0.14 0.14 <0.68 <0.08 1.5 100.093
db10a09 Q1 53.2 54.0 3.8 7.9 255 100 21 129 122 10 <0.19 0.23 <0.70 <0.07 1.5 100.091
db10a10 Q1 47.3 48.0 2.7 19 363 100 24 259 194 9.7 <0.14 0.21 0.81 <0.08 1.3 100.141
db10a11 Q1 44.0 44.7 4.0 12 312 100 23 180 166 11 <0.14 0.31 0.97 <0.09 1.7 100.117
db10a03 Q2 69.6 70.6 2.5 1.3 64 100 31 4.4 19 <11 1.2 0.31 1.8 <0.19 <1.6 100.024
--- Q2 70.5 71.6
--- Q2 69.4 70.4
Average
in Q2 qtz Ab = 69.8 70.9

Sample D 211-19, area A

db10a12 Q1 55.9 55.9 7.6 15 214 100 22 42 119 15 53 0.42 1.6 0.23 1.0 100.081
db10a13 Q1 55.0 55.0 7.4 5.7 249 100 24 68 143 9.4 <0.30 0.35 1.8 <0.06 1.4 100.087
db10a14 Q1 50.2 50.2 6.2 13 261 100 20 157 211 5.9 <0.17 0.27 1.6 <0.06 1.5 100.130
db10a15 Q1 61.8 61.8 4.8 3.6 147 100 22 37 79 <5.3 0.19 0.33 0.95 <0.09 1.3 100.052
db10a16 Q1 52.5 52.5 5.2 4.2 163 100 21 63 154 <5.6 <0.11 0.29 <0.54 <0.08 1.4 100.071
db10a17 Q1 52.9 52.9 5.2 5.8 166 100 20 59 154 7.6 <0.15 0.62 <0.72 <0.08 1.3 100.071
db10a18 Q1 54.8 54.8 4.2 4.2 142 100 22 56 126 <5.5 <0.20 0.25 <0.70 <0.09 1.3 100.061
db10b03 Q1 50.4 50.4 4.2 5.1 129 100 21 33 133 <5.9 <0.16 0.17 1.0 <0.08 1.5 100.057
db10b04 Q1 49.5 49.5 7.8 9.6 296 100 21 89 169 <4.9 <0.16 0.26 1.2 <0.09 2.3 100.103
db10b09 Q1 58.9 58.9 5.0 12 270 100 21 145 124 12 <0.17 0.37 <0.66 <0.08 2.1 100.097
db10b05 Q2 71.9 71.9 1.8 3.1 22 100 20 4.2 13 <5.2 <0.15 0.34 2.5 <0.09 1.5 100.013
--- Q2 70.3 70.3
--- Q2 74.5 74.5
--- Q2 66.7 66.7
Average
in Q2 qtz Aa = 70.9 70.9

Sample D 211-19, area G

db09f05 Q1 61.6 59.9 3.9 1.3 63 100 16 2.2 65 <3.6 0.60 0.19 0.98 0.13 1.2 100.028
db09f06 Q1 59.8 58.2 6.1 4.6 196 100 14 58 130 <3.2 <0.12 0.21 0.85 0.16 0.79 100.071
db09f08 Q1 55.5 54.0 6.7 4.7 221 100 15 78 144 4.1 0.30 0.26 2.0 <0.07 1.3 100.081
db09f09 Q1 56.9 55.3 9.0 4.9 228 100 17 50 144 6.4 0.26 0.18 0.48 <0.08 1.5 100.080
db09f04 Q2 72.9 70.9 2.4 3.6 45 100 14 4.4 37 <3.4 0.87 0.33 0.96 <0.07 1.6 100.020
db09f07 Q2 73.2 71.2 2.1 2.5 37 100 19 <2.4 27 <7.9 2.4 0.36 1.7 <0.21 2.1 100.017
--- Q2 72.9 70.9
--- Q2 72.6 70.5
Average
in Q2 qtz Ag = 72.9 70.9
Note: < 0.11 For concentrations below the limit of detection the LOD is given as <value.
* CL gray values were normalized to average Q2 quartz gray value (Aa = Aref ) as described in the text.
SiO2 = 100% used as internal standard for quantication.
Element concentrations <LOD were not considered for the calculation of average values.
126 LANDTWING AND PETTKE: LA-ICPMS TRACE ELEMENTS AND SEM-CL OF HYDROTHERMAL QUARTZ

(Aref) or in the imaging area X (Ax), respectively. The normalized CL response of Depth (m)
of laser Pit 30
Q1 quartz varies between 37.7% and 64.6% (Table 1). 20
A relative color scale is used here to describe the different gray scales in SEM- 10
CL pictures: (a) white to light gray luminescent quartz generations are described as 0 Number of
0 50 100 150 laser pulses
CL-bright, (b) light gray to dark gray luminescent quartz generations as CL-gray; Sample D211-19
10a 10b 10c 10d
and (c) dark gray to non-luminescent quartz generations as CL-dark.

Intensity (counts / sec)

LA-ICPMS
The LA-ICPMS system at ETH Zrich consists of a prototype 193 nm ArF
Excimer laser system (Lambda Physik, Germany; Gnther et al. 1997) combined
with an ELAN 6100 ICPMS instrument (Perkin Elmer, Canada). Special homog-
enization optics allow creation of a laterally homogeneous laser energy distribution
across the laser pit permitting adjustment of the beam size from 15 to 120 m.
High laser energy density on the sample surface (25 J/cm2) was favorable for
the controlled ablation of difficult quartz samples such as those from Bingham.
Control of the ablation progress is improved by direct observation of both the
sample behavior during the ablation process and real-time transient signal recording
on the computer screen. The sample aerosol is transported by He-Ar carrier gas
into the ICP. Operating conditions of the ICPMS were similar to those reported
in Pettke et al. (2004).
Analyses were generally done with a 30 m beam diameter. This size reflects
a compromise between high spatial resolution (i.e., small pit sizes) and suffi-
ciently low LODs (3 criterion; Longerich et al. 1996) needed to determine the
abundances of all trace elements of interest. A few analyses of Q2 quartz were
measured with a 20 m beam, and CL-homogeneous areas of Q1 quartz in some
cases allowed use of a 40 m beam. Each element was measured about 6 times
per second for its respective isotope. Thorough surface cleaning with acetone,
alcohol, and warm aqua regia (75% HClconc and 25% HNO3 conc), and with a clean-
ing laser shot before analysis, did not eliminate all surface contamination from
the polishing procedure (Fig. 2; see below). Therefore, signals from the first five
to ten analytical sweeps (corresponding to < 2 s) were discarded. The signal of
all measurements started immediately after this initial cleaning period and was
always 5.2 seconds (31 sweeps). This signal interval corresponds to an analytical
volume of 3500 to 7100 m3, or 9.4 to 18.7 ng of sample material for a 30 m
diameter pit of 510 m depth.
Absolute quantification of LA-ICPMS transient signals (Fig. 2) was carried
out by integration of signal intensities for all elements, correction for background
contributions, and calibration of intensity ratios with NBS 610 glass from NIST as
the external standard. These apparent element concentrations were transformed into
true values by internal standardization to 100 wt% SiO2 for the quartz (Longerich
et al. 1996), resulting in a total of 100.013100.162 wt% oxides.
Uncertainties on element concentrations from single analytical spots are about
3 to 5% at the two standard deviation level, mainly resulting from the flicker of
the plasma ion source. This uncertainty increases when concentrations approach
their respective LODs (recall that LODs are vastly different for different elements;
Heinrich et al. 2003, their Fig. 12); hence, concentration data for Fe, Cu, Ge, and Ag
have an estimated uncertainty of 2030%. Similarly, accuracy is limited chiefly by
the analytical precision; the approach of data quantification chosen here is accurate
and the interested reader is referred to Heinrich et al. (2003) and references cited
therein for a detailed discussion.
RESULTS AND DISCUSSION
Trace-element inventory of hydrothermal quartz and cor-
relation with CL-response
Twenty-five trace elements were measured by LA-ICPMS
on a large sample suite of Bingham quartz to determine which
elements are present in significant concentrations. Based on
these reconnaissance data, a shortened analytical routine was
developed for Li, Na, Al, P, K, Ti, Fe, Cu, Ga, Ge, Ag, and Sn,
to improve their quantification. Results are reported in Table
1. Of these, Cu and Ag were commonly below their respective
LODs of around several tens to one hundred ng/g. The elements
Mg, V, Mn, Zn, Rb, Sr, Y, Mo, Cs, Ba, W, Pb, and Bi were also
tested, but they were below their respective LODs on the order
of ten to a few hundred ng/g. Five profiles on four hydrothermal
quartz crystals from the sample D 211-19 form the basis of our
start ablation end ablation
106
surface
contamination
background
signal
23 Na
104 39 K
29 Si 27 Al
103 31 P
7 Li
100
0 50 60 70 80 90
Time (sec)
FIGURE 2. Transient LA-ICPMS signal of quartz analyzed with a
30 m laser ablation pit (shot db10d05; Tables 1 and 3). This particular
shot shows an unusually strong variation in trace-element concentrations
with depth. Therefore, the signal was integrated for four depth intervals
(marked at the top by 10a to 10d, respectively). Of these, only the first
interval can be compared with the CL response as the penetration depth
of CL is only up to several micrometers. Note the conspicuous signal
for Na and K at the beginning of ablation, suggesting that surface
contamination arising from the polishing procedure cannot be cleaned
chemically (see text).
observations (Figs. 3 and 4).
Concentrations of Al, K, Ti, and Na in quartz vary consider-
ably with variable CL intensities (Table 1), revealing a gener-
ally negative correlation between trace-element concentrations
and CL gray scale values (Fig. 4). Highest concentrations were
obtained for the bright luminescent Q1 quartz. The lowest
concentrations of these elements were observed in the dark
luminescent Q2 quartz next to sulfides, increasing smoothly
with brightening of the CL response. Abundances of Fe and
Li are also significantly different between the two quartz types
and again lower in Q2 quartz. Noteworthy is the observation
that the Fe content of Q2 quartz associated with the deposition
of sulfides at Bingham is below the LOD of ~4 g/g, which is
in sharp contrast to concentrations of up to ~5000 g/g Fe ob-
tained on stockwork quartz around sulfides from the Grasberg
porphyry Cu-Au deposit (Penniston-Dorland 2001). Copper,
Sn, Ga, and P abundances do not vary with the CL response,
and Ge is heterogeneous over the whole range of CL gray scale
values. Sporadically high concentrations of Cu, which correlate
with elevated Fe contents (e.g., sample D 211-19, shot number
db10a12; Table 1), are interpreted as micro-inclusions of Cu-Fe
sulfides, hence not considered further.
Comparison of quartz trace-element ratios reveal that only
K/Al evolves significantly from Q1 quartz to Q2 quartz, namely
from ~0.3 to ~0.06 (molar). Ratios of Li/Al tend to increase from
~0.1 to ~0.2 (molar). This result indicates that the activity of K
might have decreased from Q1 to Q2 quartz growth whereas
LANDTWING AND PETTKE: LA-ICPMS TRACE ELEMENTS AND SEM-CL OF HYDROTHERMAL QUARTZ 127

FIGURE 3. LA-ICPMS trace

element concentration profiles
of quartz (qtz) of imaging area A
(left) and area B (right) of sample
D 211-19. Certain trace element
abundances (Al, Ti, Na, K)
correlate negatively with the
CL gray value, whereas others
(e.g., Sn and P) display uniform
concentrations across in CL zones.
Images a1 and b1 show the mixed
signal (SE and CL) of these two
quartz crystals. The contrast
between the sulfides bornite (bn;
white) and chalcopyrite (cpy; light
gray) on the BSE images (a2, b2)
is digitally enhanced. Images b3a
and b3b are two transmitted light
images of area B before (b3a)
and after (b3b) the LA-ICPMS
analyses. Symbols are measured
concentrations; dashes represent
limits of detection (LOD).

W FIGURE 4. Trace-element
concentrations (g/g) plotted
against the CL gray values (0%
= white, 100% = black) for
the hydrothermal vein quartz
sample D 211-19. The CL gray
value correlates negatively with
the concentrations of Al, K, Ti,
Na, Li, and Fe, while uniform
concentrations were found for P,
Sn, Ga and Ge across variable
luminosity.
128 LANDTWING AND PETTKE: LA-ICPMS TRACE ELEMENTS AND SEM-CL OF HYDROTHERMAL QUARTZ

that of the other elements might have remained largely uniform. 15000
Ratios of Na/Al and Ti/Al remain uniform, despite the fact that
these element concentrations vary by more than a factor of ten Al, Ga versus

Sum of Li + Na + P + K (nmol)
Li, Na, P, K
between Q1 and Q2 quartz. H+
Only a few ions such as Al3+, Ga3+, Fe3+, Ge4+, Ti4+, and P5+ 10000
appear suitable for the substitution of Si4+ in the quartz structure.
B
Some of these elements require charge compensation by ad-
1:1 line
ditional ions (H+, Li+, Na+, K+, Cu+, and Ag+) in interstitial sites
5000
related to structural channels (Gtze et al. 2001). Molar data
(Table 2) show that Al is by far the most prominent substitutional
ion that is charge-balanced dominantly by K (Si4+ 3+ +
1 Al1 K1 ) and, A Q1
to a lesser extent, by Na+ (Si4+ 1 Al 3+
1 Na+
1 ). Lithium+
(Si 4+
1 Al 3+ +
1 Li 1) 0
Q2
does not appear to be important in Q1 quartz in contrast to results 0 5000 10000 15000
obtained on hydrothermal quartz from the Alps (e.g., Perny et al. Sum of Al + Ga (nmol)
1992; Ramseyer et al. 1988; Ramseyer and Mullis 1990). Charge FIGURE 5. Plot of the sum of molar concentrations (nmol) of
balance of Al3+ with P5+ (Si4+ 3+ 5+
2 Al1 P1 ; xenotime-type substitution) univalent plus pentavalent cations vs. trivalent cations to check for
does not exhibit a systematic correlation. The sum of the trivalent charge balance assuming the substitution mechanisms Si4+ 1 [(Al, Ga)
3+

elements Al and Ga is not charge-balanced by the sum of Li, Na, (K, Li, Na)+]+1 or a xenotime-type substitution Si4+ 3+ 5+
2 [(Al, Ga) P ]+1. An
P, and K, especially for Q1 quartz (Fig. 5). This result suggests apparent excess of trivalent cations is observed that can be compensated
that either H+ (not possible to measure with LA-ICPMS) is impor- for by invoking the presence of H+ or Al3+ incorporation into interstitial
tant for charge balance in hydrothermally precipitated quartz, or sites. As shown in Table 2, the incorporation of Al, K, and Ti dominates
that interstitial Al3+ may be present according to the substitution by far the trace-element inventory of these quartz samples.
reaction (Si4+ 3+
3 Al +4). Assuming that charge neutrality is effected
by incorporation of H+, the concentration of H increases from 0 TABLE 2. Average molar concentrations of trace elements in Q1 and
to ~6 g/g as the sum of Al3+ and Ga3+ increases up to ~10 000 Q2 quartz
Q1 quartz Q2 quartz
nmol. (Fig. 5; trend A). When this sum exceeds ~10 000 nmol, the
Lowest Highest Average Lowest Highest Average
inferred concentration of H+ remains constant at approximately 6 value* value conc. value* value* conc.
g/g (Fig. 5; trend B). Accordingly, the molar (Al + Ga)/H ratios (nmol) (nmol) (nmol) (nmol) (nmol) (nmol)
increase from ~1.6 to as high as 3.3 for (Al + Ga) concentrations Li 353 1297 760 219 427 338
Na 57 1554 421 <22 154 98
exceeding 10 000 nmol. These results compare well with those Al 1675 15121 8486 833 2365 1552
of Perny et al. (1992) who reported an average Al/H atomic ratio Si internal standard internal standard
of ~2.5 for hydrothermal quartz. P 437 960 666 598 1012 710
K 58 8443 2606 <26 113 86
Possible causes of trace element variations Ti 765 4940 2959 261 819 489
Fe <57 360 175 <67 <201
Variability in luminescence as observed relate to changes in Cu < 1.6 6.7 4.0 <2.4 3.8 2.7
Ga 2.0 8.9 4.0 2.6 5.2 4.1
quartz trace-element contents that could result from changes in Ge <6.7 34.6 16.7 9.4 33.8 23.7
fluid chemistry, temperature, and/or pH during quartz crystal- Ag 0.43 2.16 1.10 <0.61 1.24
lization, variable crystallization rate, or from a combination Sn <5.8 19.1 13.1 7.8 17.4 12.0
thereof. Our evaluation of possible causes for variable trace- * For concentrations below the limit of detection the LOD is given as < value.
High concentrations (interpreted as accidental micro-inclusions), not in-
element contents in quartz will provide qualitative arguments on cluded.
the importance of these parameters and is based on the extensive Element concentrations < LOD were not considered for the calculation of
average values.
study of vein petrography and the fluid chemistry by Landtwing
(2004). A quantitative assessment is chiefly hampered by the
unavailability of fluid chemical data and rigorous temperature
constraints for Q1 quartz growth. (Landtwing 2004; Landtwing et al. 2002). The maximum tem-
Quartz Q1 crystallized during stockwork veining, which is perature is constrained by the solidus of wet granites, 700750 C
commonly associated with hydraulic fracturing of the host rock. at ~100 MPa (Johannes and Holtz 1996). Hydrothermal alteration
Quartz solubility, at least in pure water, is a strong function of associated with stockwork veining is of potassic type, suggesting
pressure (Shock et al. 1989). Rapid quartz precipitation can high K+ activity in the hydrothermal fluid. Taken together, these
therefore be expected to occur in response to significant silica data indicate rapid quartz Q1 crystallization between ~450 and
oversaturation due to a drop in fluid pressure from lithostatic to 750 C at pressures below ~100 MPa.
hydrostatic associated with hydraulic fracturing (Rusk and Reed The fluid chemistry, formation temperature and pressure of
2002). Microthermometric total homogenization temperatures of Q2 quartz, in contrast, is well constrained (Landtwing 2004;
early CO2-bearing, intermediate-density fluid inclusions are be- Landtwing et al. 2002). Quartz Q2 crystallized in the narrow
tween 350 and 420 C. These measurements provide a minimum temperature interval from ~350425 C at pressures of ~14 to 21
temperature for entrapment of a one-phase fluid. The maximum MPa, together with the bulk of Cu-Fe sulfide precipitation. This
homogenization temperatures of brine inclusions in Q1 quartz are temperature interval corresponds to the retrograde quartz solu-
>600 C, indicating high temperatures of Q1 quartz formation bility region (as summarized by Fournier 1999), causing partial
 LANDTWING AND PETTKE: LA-ICPMS TRACE ELEMENTS AND SEM-CL OF HYDROTHERMAL QUARTZ
dissolution of Q1 quartz with later crack healing and void filling
by precipitation of Q2 quartz upon further cooling (Landtwing
2004). During Q2 quartz crystallization, the bulk salinity shows
a subtle trend toward lower values, from 44.2 to 39.7 wt% NaCl
equivalent, while concentrations of major cations do not exhibit
any resolvable compositional trend, ranging between 914 wt-%
Na, 610.5 wt-% K, 0.010.12 wt% Mg, 0.351.0 wt% Ca and with aging of the hydrothermal system, thus reducing the cool-
3.58.5 wt% Fe, respectively. Aluminum and Ti could not be
quantified in fluid inclusions due to their predominance in the
quartz host. Comparison of quartz trace element ratios reveal
that only K/Al ratios significantly evolve from Q1 quartz to Q2
quartz, namely from ~0.38 to ~0.05, ratios of Na/Al and Ti/Al
remain uniform. Recall that the fluid from which both quartz
types crystallized was likely sourced in the same magma chamber
(Landtwing 2004). Together, these lines of evidence suggest that
the composition of the Q1 quartz-forming fluid was similar to
that measured for Q2 quartz crystallization.
Our data and results from hydrothermal quartz of the Swiss
Alps suggest that fluid chemistry determines which alkali ions
are incorporated to compensate the charge of Al3+. Potassium is
the dominant monovalent cation in Q1 quartz (directly relating
to the inferred high activity of K during potassic alteration),
whereas Li appears to be important for the Alpine quartz [no
data were reported for K, however, by Perny et al. (1992)].
Based on these considerations, we conclude that the trace-ele-
ment pattern (i.e., trace element ratios) of quartz reflects the
chemical composition of (or better, the element activity ratios
in) the hydrothermal fluid.
Activity of H+, hence pH of the fluid, can be expected to play
a major role in charge balancing trivalent impurities, hence may
significantly affect their amount incorporated into quartz. Mass-
balance considerations (Ulrich and Heinrich 2002) indicate that
interaction of boiling solutions with wall-rock silicates to form
K-feldspar, quartz, and biotite as typical minerals of the potassic
alteration assemblage tends to buffer solution pH. This buffering
might account for the inferred constant concentration of H+ in
(Al + Ga)-rich Q1 quartz [Fig. 5, trend B, (Al + Ga) > 10 000
g/g]. Alternatively, the constant amount of ~6 g/g of inferred
H+ might correspond to saturation of the silica structure by H+
at temperatures between 450 and 700 C. Based on the available
data, these two explanations remain unresolved, however.
The next question to be addressed is: Why do the abundances
of some trace elements (e.g., Na, Al, Ti) vary by more than a
factor of ten while their interelement ratios remain essentially
uniform? Temperature of quartz crystallization is known from
experiment to affect the amount of Al incorporated into quartz,
having led to the controversial postulation that the Al content
of quartz may serve as a geothermometer (Dennen et al. 1970).
In our case, the sum of trace elements in quartz increases with
increasing crystallization temperature, but a rigorous test of
whether temperature alone can account for the variability in
trace-element contents remains impossible. Changes in quartz
crystallization rate are expected to have occurred during Q1 and
Q2 quartz deposition, induced by pressure fluctuations between
lithostatic and hot hydrostatic (Shock et al. 1989) and possibly by
variable fluid cooling rates. Vein textures indicate that Q1 quartz
crystallized during hydrofracturing, hence is likely to have been
deposited fast from a fluid oversaturated with respect to silica.
129
In contrast to this scenario, Q2 quartz crystallization probably
occurred at temperatures just below those of retrograde quartz
solubility (compare Fournier 1999), i.e., below ~425 C as in-
dicated by microthermometry (Landtwing 2004). Here, silica
solubility curves are flatter than at temperatures above ~500 C.
Moreover, temperature gradients will become less pronounced
ing rate for the fluid percolating through the fracture network.
Taken together, these arguments suggest that Q1 quartz grew
much faster than did Q2 quartz. From this difference, we specu-
late that the quartz crystallization rate may also prove to be an
important parameter controlling the amount of trace elements
incorporated into quartz: the faster quartz grows, the more trace
elements it incorporates.
Advantages of combined SEM-CL and LA-ICPMS
measurements
Laser-ablation ICPMS is a highly sensitive, fast, and cost-ef-
ficient analytical method. The 193 nm ArF laser with a homog-
enized energy distribution used here is particularly well-suited
for controlled ablation of quartz samples as the absorbance of
the 193 nm light by quartz is almost quantitative. As LA-ICPMS
systems are only equipped with a petrographic microscope, SEM-
CL imaging has to be done independently to map the sample,
identify analytical sites and suitable pit sizes.
LODs in the single ng/g range can be achieved for many heavy
(> mass 80) elements with a 120 m pit (i.e., low spatial resolution)
but growth zones of uniform CL response commonly are much
narrower (Fig. 4). Clearly, the apparent amplitude of trace-element
variations within one profile is directly related to the spatial reso-
lution of the analytical technique. If variability in CL response is
on a scale that is smaller than about 10 m, LODs achieved with
a 10 m pit are likely to be insufficient to significantly detect all
analytes of interest. The selection of laser pit size thus strongly
depends on the aims of the study, but good spatial resolution will
always compete with low limits of detection.
Caution has to be exercised when using LA-ICPMS data to
unravel chemical causes for variable CL response, because CL
images represent the response of a thin surface layer, and the
extension of the CL pattern with depth remains unknown. The
pear shaped penetration depth of the primary electron or ion
beam used in EMP or SIMS is up to a few micrometers only
(depending on the tuning of the machine and the total analysis
time), whereas LA-ICPMS samples a near-cylindrical volume of
material. Our laser system achieves a very good depth resolution
(refer to Fig. 2 and Table 3, which show a clear depth-related
zonation for Al, K, and Ti), thanks to the homogeneous energy
distribution of our system across the entire pan-shaped pit (e.g.,
Heinrich et al. 2003; Horn et al. 2001). Compositional variability
with depth can now be quantified by integrating signal sections
corresponding to depth intervals (analyses 10a, b, c, d; Fig. 2
and Table 3). These data suggest that the laser shot db10b05
(Fig. 2) started in a Q2 quartz zone as derived from SEM-CL
and actually penetrated into Q1 quartz after ~12 s of laser firing
(corresponding to ~20 m depth).
The ablation rate is variable primarily as a function of laser
energy for any given target material. Ensuring representative
sampling (Pettke et al. 2000) by analyzing every mass of in-
130 LANDTWING AND PETTKE: LA-ICPMS TRACE ELEMENTS AND SEM-CL OF HYDROTHERMAL QUARTZ

TABLE 3. Integration intervals with progressive depth of LA-ICPMS Acta, 56, 175185.
shot db10b05, sample D 211-19, area A (compare Figure 2 Bowman, J.R., Parry, W.T., Kropp, W.P., and Kruer, S.A. (1987) Chemical and
and Table 1) isotopic evolution of hydrothermal solutions at Bingham, Utah. Economic
Geology, 82, 395428.
Integration interval 10a 10b 10c 10d Cathelineau, M., Boiron, M.C., Essarraj, S., Dubessy, J., Lespinasse, M., and
Depth below ~2.510 m ~8.516 m ~14.522 m ~20.528 m Poty, B. (1993) Fluid pressure variations in relation to multistage deforma-
surface tion and uplifta fluid inclusion study of Au quartz veins. European Journal
Quartz Q2 Q2 Q2 and Q1? Q1 of Mineralogy, 5, 107121.
generation Dennen, W.H., Blackburn, W.H., and Quesada, A. (1970) Aluminum in quartz as a
Li(g/g) 1.8 1.5 2.6 2.8 geothermometer. Contributions to Mineralogy and Petrology, 27, 332342.
Na(g/g) 3.1 2.6 2.6 1.7 Flem, B., Larsen, R.B., Grimstvedt, A., and Mansfeld, J. (2002) In situ analysis of
Al(g/g) 22 30 53 91 trace elements in quartz by using laser ablation inductively coupled plasma
SiO2 (wt-%) 100 100 100 100 mass spectrometry. Chemical Geology, 182, 237247.
P(g/g) 20 19 23 25 Fournier, R.O. (1999) Hydrothermal processes related to movement of fluid from
K(g/g) 4.2 3.9 11 21 plastic into brittle rock in the magmatic-epithermal environment. Economic
Ti(g/g) 13 21 37 69 Geology, 94, 11931211.
Fe(g/g) <5.2 <6.4 <7.1 <7.9 Gtze, J., Plotze, M., and Habermann, D. (2001) Origin, spectral characteristics and
Cu(g/g) <0.15 <0.18 <0.20 <0.23 practical applications of the cathodolumimescence (CL) of quartza review.
Ga(g/g) 0.34 0.19 0.28 0.16 Mineralogy and Petrology, 71, 225250.
Ge(g/g) 2.5 2.0 1.2 <0.82 Guillong, M. and Gnther, D. (2001) Quasi non-destructive laser ablation-
Ag(g/g) <0.09 <0.10 <0.12 <0.13 inductively coupled plasma-mass spectrometry fingerprinting of sapphires.
Sn(g/g) 1.5 1.3 1.8 <1.3 Spectrochimica Acta Part B-Atomic Spectroscopy, 56, 12191231.
Total element 100.013 100.015 100.024 100.038 Heinrich, C.A., Pettke, T., Halter, W.E., Aigner, M., Audtat, A., Gnther, D.,
oxides (wt%) Hattendorf, B., Bleiner, D., Guillong, M., and Horn, I. (2003) Quantitative
multi-element analysis of minerals, fluid and melt inclusions by laser-ablation
inductively-coupled-plasma-mass-spectrometry. Geochimica et Cosmochimica
Acta, 67, 34733497.
Herzog, L.F., Marshall, D.J., and Babione, R.F. (1970) The luminoscope; a new
terest sufficiently often, and knowing the ablation rate (~1.5 instrument for studying the electron-stimulated luminescence of terrestrial,
m/s in this case, slightly decreasing with depth), the depth of extra-terrestrial and synthetic materials under the microscope. In Space Appli-
ablation can be controlled well, and the integration interval of cations of Solid State Luminescent Phenomena, 70, 7998. Special Publication
- Pennsylvania State University, Materials Research Laboratory, University
the transient signal can be set accordingly. Minimization of the Park, Pennsylvania.
thickness of the ablated layer of sample could be further pushed Horn, I., Guillong, M., and Gnther, D. (2001) Wavelength dependant ablation rates
but would result in a slightly lower analytical precision and in for metals and silicate glasses using homogenized laser beam profilesimplica-
tions for LA-ICP-MS. Applied Surface Science, 182, 91102.
significantly poorer LODs, with the consequence that some of Johannes, W. and Holtz, F. (1996)Petrogenesis and experimental petrology of
the trace-element concentrations would fall below their LOD granitic rocks, 335 p. Springer, Berlin.
Landtwing, M.R. (2004) Fluid evolution and ore mineral precipitation at the Bing-
(Guillong and Gnther 2001). ham porphyry Cu-Au-Mo deposit, Utah, deduced from cathodoluminescence
An intriguing result, not only observed in this study, is that a imaging and LA-ICPMS microanalysis of fluid inclusions. Ph.D thesis, ETH
thin surface layer of polished quartz, possibly <1 m in thickness, Zurich, 260 p.
Landtwing, M.R., Heinrich, C.A., Halter, W.E., Pettke, T., Redmond, P.B., and
shows significant contamination (especially evident for Na and Einaudi, M.T. (2002) Fluid evolution at the Bingham Cu-Au-Mo-Ag porphyry
K) that could not be removed by washing the sample in warm deposit. Geochimica et Cosmochimica Acta, 66, 430430.
aqua regia in an ultrasonic bath before LA-ICPMS analysis (Fig. Lanier, G., John, E.C., Swensen, A.J., Reid, J., Bard, C.E., Caddey, S.W., and
Wilson, J.C. (1978) General geology of Bingham Mine, Bingham Canyon,
2). This finding suggests that some elements measured by an in- Utah. Economic Geology, 73, 12281241.
situ surface method (e.g., EMP) may actually show element Longerich, H.P., Jackson, S.E., and Gnther, D. (1996) Laser ablation inductively
coupled plasma mass spectrometric transient signal data acquisition and ana-
abundances that could suffer from serious contamination arising lyte concentration calculation. Journal of Analytical Atomic Spectrometry,
from the polishing procedure. Such surface contamination can be 11, 899904.
avoided by setting the LA-ICPMS integration intervals accord- Marshall, D.J. (1988) Cathodoluminescence of Geological Materials, 146 p. Unwin
Hyman Ltd, Boston, Massachusetts.
ingly, promising to obtain the real trace-element concentrations Monecke, T., Kempe, U., and Gtze, J. (2002) Genetic significance of the trace
incorporated into the quartz lattice. element content in metamorphic and hydrothermal quartz: a reconnaissance
study. Earth and Planetary Science Letters, 202, 709724.
ACKNOWLEDGMENTS Mller, A., Seltmann, R., and Behr, H.J. (2000) Application of cathodoluminescence
to magmatic quartz in a tin granitecase study from the Schellerhau Granite
This work would not have been possible without the logistical support of Ken- Complex, Eastern Erzgebirge, Germany. Mineralium Deposita, 35, 169189.
necott Utah Copper, and we would especially like to thank E. Harrison, T. Smith, Mller, A., Lennox, P., and Trzebski, R. (2002) Cathodoluminescence and micro-
and G. Austin. Special thanks go to C. Heinrich who initiated the project, and to structural evidence for crystallisation and deformation processes of granites
P. Redmond, M. Einaudi, and E. Inan who contributed in the field and established in the Eastern Lachlan Fold Belt (SE Australia). Contributions to Mineralogy
most of the geologic context. Discussion with T. Driesner helped to clarify several and Petrology, 143, 510524.
aspects. K. Kunze and the Interessengemeinschaft Elektronenmikroskopie, ETH Pagel, M., Barbin, V., Blanc, P., and Ohnenstetter, D. (2000) Cathodoluminescence
Hnggerberg is thanked for assistance with the SEM-CL. Very thoughtful reviews in Geosciences, 514 p. Springer, Berlin, Heidelberg, Germany.
by P. Heaney and J. Wilkinson greatly helped to balance the conclusions reached Parnell, J., Earls, G., Wilkinson, J.J., Hutton, D.H.W., Boyce, A.J., Fallick, A.E.,
in this manuscript. This project was supported by ETH Project Grant 0-20663-99 Ellam, R.M., Gleeson, S.A., Moles, N.R., Carey, P.F., and Legg, I. (2000)
to C. Heinrich. Regional fluid flow and gold mineralization in the Dalradian of the Sperrin
Mountains, northern Ireland. Economic Geology, 95, 13891416.
REFERENCES CITED Penniston-Dorland, S.C. (2001) Illumination of vein quartz textures in a porphyry
copper ore deposit using scanned cathodoluminescence: Grasberg Igneous
Audtat, A. and Gnther, D. (1999) Mobility and H2O loss from fluid inclusions in
Complex, Irian Jaya, Indonesia. American Mineralogist, 86, 652666.
natural quartz crystals. Contributions to Mineralogy and Petrology, 137, 114.
Perny, B., Eberhardt, P., Ramseyer, K., Mullis, J., and Pankrath, R. (1992) Microdis-
Bambauer, H.U. (1961) Spurenelementgehalte und Farbzentren in Quarzen
tribution of Al, Li, and Na in alpha-quartzpossible causes and correlation with
aus Zerrklften der Schweizer Alpen. Schweizerische Mineralogische und
short-lived cathodoluminescence. American Mineralogist, 77, 534544.
Petrographische Mitteilungen, 41, 335369.
Pettke, T., Heinrich, C.A., Ciocan, A.C., and Gnther, D. (2000) Quadrupole mass
Boiron, M.C., Essarraj, S., Sellier, E., Cathelineau, M., Lespinasse, M., and Poty, B.
spectrometry and optical emission spectroscopy: detection capabilities and
(1992) Identification of fluid inclusions in relation to their host microstructural
representative sampling of short transient signals from laser-ablation. Journal
domains in quartz by cathodoluminescence. Geochimica et Cosmochimica
of Analytical Atomic Spectrometry, 15, 11491155.
 LANDTWING AND PETTKE: LA-ICPMS TRACE ELEMENTS AND SEM-CL OF HYDROTHERMAL QUARTZ
Pettke, T., Halter, W.E., Webster, J.D., Aigner-Torres, A., and Heinrich, C.A.
(2004) Accurate quantification of melt inclusion chemistry by LA-ICPMS: A
comparison with EMP and SIMS and advantages and possible limitations of
these methods. Lithos, in press.
Phillips, C.H., Smith, T.W., and Harrison, E.D. (1997) Alteration, metal zoning,
and ore controls in the Bingham Canyon porphyry copper deposits, Utah.
In D.A. John, and G.H. Ballantyne, Eds., Geology and Ore Deposits of the
Oquirrh and Wasatch Mountains, Utah, 29, p. 133145. Guidebook Series of
the Society of Economic Geologists, Littleton, CO.
Ramseyer, K. and Mullis, J. (1990) Factors influencing short-lived blue cathodolu-
minescence of alpha-quartz. American Mineralogist, 75, 791800.
Ramseyer, K., Baumann, J., Matter, A., and Mullis, J. (1988) Cathodoluminescence
colors of alpha-quartz. Mineralogical Magazine, 52, 669677.
Redmond, P.B., Landtwing, M.R., and Einaudi, M.T. (2001) Cycles of porphyry
dike emplacement, veining, alteration and mineralization in the Bingham
porphyry Cu-Au-Mo deposit, Utah. In A. Piestrzynski, et al., Ed., Mineral
Deposits at the Beginning of the 21th Century. Society for geology applied
to Mineral Deposits (SGA), Joint 6th biennial SGA-SEG meeting, Krakow,
Poland. p. 473476.
Rusk, B. and Reed, M. (2002) Scanning electron microscope-cathodoluminescence
analysis of quartz reveals complex growth histories in veins from the Butte
porphyry copper deposit, Montana. Geology, 30, 727730.
Shock, E.L., Helgeson, H.C., and Sverjensky, D.A. (1989) Calculation of the ther-
modynamic and transport-properties of aqueous species at high-pressures and
temperaturesstandard partial molal properties of inorganic neutral species.
131
Geochimica et Cosmochimica Acta, 53, 21572183.
Stevens Kalceff, M.A., Phillips, M.R., Moon, A.R., and Kalceff, W. (2000) Cath-
odoluminescence microcharacterisation of silicon dioxide polymorphs. In M.
Pagel, V. Barbin, P. Blanc, and D. Ohnenstetter, Eds., Cathodoluminescence
in Geosciences, 514 p. Springer, Berlin, Germany.
Ulrich, T. and Heinrich, C.A. (2002) Geology and alteration geochemistry of the
porphyry Cu-Au deposit at Bajo de la Alumbrera, Argentina (v. 96, p. 1719,
2001). Economic Geology, 97, 18631888.
Valley, J.W. and Graham, C.M. (1996) Ion microprobe analysis of oxygen isotope
ratios in quartz from Skye granite: healed micro-cracks, fluid flow, and hydrother-
mal exchange. Contributions to Mineralogy and Petrology, 124, 225234.
Wilkinson, J.J. and Johnston, J.D. (1996) Pressure fluctuations, phase separa-
tion, and gold precipitation during seismic fracture propagation. Geology,
24, 395398.
Wilkinson, J.J., Boyce, A.J., Earls, G., and Fallick, A.E. (1999) Gold remobiliza-
tion by low-temperature brines: evidence from the Curraghinalt gold deposit,
Northern Ireland. Economic Geology, 94, 289296.
Zinkernagel, U. (1978) Cathodoluminescence of quartz and its application to sand-
stone petrology, 69 p. E. Schweizerbartsche Verlagsbuchhandlung (Naegele
u. Obermiller), Stuttgart.
MANUSCRIPT RECEIVED NOVEMBER 15, 2003
MANUSCRIPT ACCEPTED SEPTEMBER 8, 2004
MANUSCRIPT HANDLED BY ALAN ANDERSON