www.geologicacarpathica.com
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
, AUGUST 2014, 65, 4, 257—271 doi: 10.2478/geoca-2014-0017
Distribution of elements among minerals of a single
(muscovite-) biotite granite sample – an optimal approach
and general implications
VOJTĚCH JANOUŠEK
1,2
, TOMÁŠ NAVRÁTIL
3
, JAKUB TRUBAČ
1,2
, LADISLAV STRNAD
4
,
FRANTIŠEK LAUFEK
1
and LUDĚK MINAŘÍK
3
1
Czech Geological Survey, Klárov 3/131, 118 21 Prague 1, Czech Republic;
vojtech.janousek@geology.cz; jakub.trubac@geology.cz; frantisek.laufek@geology.cz
2
Institute of Petrology and Structural Geology, Charles University in Prague, Albertov 6, 128 43 Prague 2, Czech Republic
3
Institute of Geology, Academy of Science, Rozvojová 269, 165 00 Prague 6, Czech Republic; navratil@gli.cas.cz; alex@gli.cas.cz
4
Laboratories of the Geological Institutes, Charles University in Prague, Albertov 6, 128 43 Prague 2, Czech Republic;
ladislav.strnad@natur.cuni.cz
(Manuscript received January 17 2014; accepted in revised form June 5, 2014)
Abstract: The petrography and mineral chemistry of the coarse-grained, weakly porphyritic (muscovite-) biotite Říčany
granite (Variscan Central Bohemian Plutonic Complex, Bohemian Massif) were studied in order to assess the distribu-
tion of major and trace elements among its minerals, with consequences for granite petrogenesis and availability of
geochemical species during supergene processes. It is demonstrated that chemistry-based approaches are the best suited
for modal analyses of granites, especially methods taking into account compositions of whole-rock samples as well as
their mineral constituents, such as constrained least-squares algorithm. They smooth out any local variations (mineral
zoning, presence of phenocrysts, schlieren…) and are robust in respect to the presence of phenocrysts or fabrics. The
study confirms the notion that the accessory phases play a key role in incorporation of many elements during crystalli-
zation of granitic magmas. Especially the REE seem of little value in petrogenetic modelling, unless the role of acces-
sories is properly assessed and saturation models for apatite, zircon, monazite±rutile carefully considered. At the same
time, the presence of several P-, Zr- and LREE-bearing phases may have some important consequences for saturation
thermometry of apatite, zircon and monazite.
Key words: modal analyses, trace-element residence, ICP-MS, Central Bohemian Plutonic Complex, Říčany granite.
Introduction
The distribution of chemical elements among minerals in a
single granite specimen represents a fundamental and in-
triguing problem, which, however, has attracted surprisingly
little attention in the current literature (e.g. Gromet & Silver
1983; Sawka 1988; Evans & Hanson 1993; Wark & Miller
1993; Bea 1996). Still, a good understanding of the net con-
tributions of individual phases to the whole-rock chemical
budget is a necessary prerequisite should any relevant petro-
genetic modelling be undertaken. Moreover, particular host
minerals show variable resistance to alteration and/or weath-
ering, thus controlling the degree to which the given element
could be mobilized into the environment.
For such studies, reliable and precise concentration data in
small sample volumes are crucial. The data must be sufficient to
minimize the effects of inhomogeneities in the analysed mate-
rial (for individual minerals it could be the presence of zoning,
inclusions, alteration zones…). The required precision is often
far beyond the limits of the electron microprobe microanalysis
(EMPA); the prohibitive cost effectively rules out the ion
probe as well. Fortunately the ICP-MS technique enables a
reliable determination of sub-ppm amounts of elements both in
the solution (dissolved whole-rock powders or monomineralic
separates) and in situ, using the laser-ablation (LA) apparatus.
In addition, an accurate estimate of the mineral modal pro-
portions has to be made. Theoretically this should pose no
problem, given that the modal analyses have provided a ba-
sis for classification of holocrystalline igneous rocks for over
eighty years (Niggli 1931; Johannsen 1932, 1937, 1938, 1939;
Streckeisen 1974; Le Maitre 2002).
The classical technique relies on counting of individual
mineral grains over a regular grid, either in a standard thin sec-
tion or a polished rock slab. More sophisticated alternatives
include computer-aided image analysis, powder X-ray diffrac-
tion (P-XRD) and chemistry-based mathematical approaches
(normative recalculations or statistical methods employing
the chemistry of the bulk rock and its mineral constituents).
The current paper initially focuses on methodological is-
sues connected with obtaining a modal analysis representa-
tive of a large, coarse-grained granite sample. The various
methods are compared and the most trustworthy approach/
modal data chosen (Appendix 1*). Then, using a combina-
tion of EMPA with (LA) ICP-MS data, an attempt is made to
identify the principal mineral hosts for individual elements,
evaluating the relative contributions of each of them to the
whole-rock budget. Finally we discuss the general implica-
tions for obtaining modal analyses of coarse-grained rocks,
modelling of igneous processes, obtaining reliable whole-
rock trace-element analyses and saturation thermometry.
* – Appendices 1—5 (only as a Supplement in the electronical
version; www.geologicacarpathica.com)
258
JANOUŠEK, NAVRÁTIL, TRUBAČ, STRNAD, LAUFEK and MINAŘÍK
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA, 2014, 65, 4, 257—271
Our case study concentrates on the Lower Carboniferous
(Mississippian) Říčany (muscovite-) biotite granite pluton
from the Central Bohemian Plutonic Complex (CBPC), Czech
Republic (Fig. 1a). This small body is relatively well known,
as it has been, over the years, a subject of numerous studies
concerned with its geology, composition and genesis (e.g.
Katzer 1888; Orlov 1933; Němec 1978; Palivcová et al. 1992;
Janoušek et al. 1997; Trubač et al. 2009). Moreover, even
more extensive research was done on element mobility during
the weathering, mass fluxes, cycling and balance of elements
within the model forest ecosystem (Minařík & Houdková
1986; Minařík & Kvídová 1986; Minařík et al. 1998, 2003;
Navrátil et al. 2002, 2004, 2007). For both types of studies,
detailed assessment of the elemental distribution among indi-
vidual mineral phases has clearly been long overdue.
Geological setting
The CBPC is one of the largest composite granitoid com-
plexes in the Central European Variscides (Fig. 1a). The in-
dividual petrographic types can be grouped into several
suites, based on petrography, age, whole-rock and mineral
geochemistry (Holub et al. 1997; Janoušek et al. 2000b; Žák
et al. (2014) and references therein) (Fig. 1b). The oldest
among them is the normal calc-alkaline Sázava suite
( ~ 355 Ma) forming much of the north-eastern CBPC. The
most common in the central-southern CBPC are ~ 346 Ma
old, K-rich calc-alkaline granodiorites to granites of the Blatná
suite with minor basic bodies. The (ultra-) potassic Čertovo
břemeno suite ( ~ 337 Ma) is formed by K-Mg-rich mela-
granites and melasyenites, the most basic types correspond-
Fig. 1. a – Location of the
Variscan Central Bohemian
Plutonic Complex (CBPC)
within the Bohemian Massif.
b – Sketch of the CBPC with
the main granitoid suites after
Janoušek et al. (2000b) (normal
calc-alkaline Sázava, high-K
calc-alkaline Blatná, peralumi-
nous Maršovice and (ultra-)
potassic Čertovo břemeno).
c – Geological outline of the
Říčany Pluton and the sur-
rounding units (after Janoušek
et al. 1997). The position of
the sampled locality, the work-
ing Žernovka quarry, is also
indicated.
259
DISTRIBUTION OF ELEMENTS AMONG MINERALS OF A SINGLE GRANITE SAMPLE
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA, 2014, 65, 4, 257—271
ing to the so-called durbachites. Strongly peraluminous Mu-Bt
granodiorites (often Crd-bearing) of the Maršovice suite are
regionally rather insignificant.
The Říčany Pluton (Fig. 1c), the main representative of the
namesake suite, is a late (post-tectonic), shallow-level gra-
nitic body that has intruded the boundary between low-grade
Upper Proterozoic to Lower Paleozoic metasediments of the
Teplá—Barrandian Unit and dominantly high-grade metased-
iments of the Moldanubian Unit. Its eastern margin is ob-
scured by Permo-Carboniferous sediments. The only modern
geochronological information available is the
40
Ar—
39
Ar bio-
tite age of 336 ± 3.5 Ma (unpublished data of H. Maluski, cited
in Janoušek et al. 1997).
The intrusion has a roughly elliptical outline (13
×9 km)
and is mainly made up of two distinct granite varieties. The
outer, ‘strongly porphyritic’ one contains more abundant
K-feldspar phenocrysts, whereas in the central, ‘weakly por-
phyritic’ facies, the phenocrysts are scarce (Katzer 1888)
(Fig. 1c). The granite encloses numerous large biotite-rich
mafic enclaves and less common metasedimentary xenoliths.
The Pluton is cut by many pegmatite and aplite dykes. The
central part has been intruded by several small bodies of
fine-grained, equigranular, two-mica Jevany leucogranite
and the southern exocontact is rimmed by the so-called Mar-
ginal aplite (Němec 1978).
Petrology and mineral chemistry
The studied rock comes from the central, ‘weakly porphy-
ritic’ facies. It is relatively fresh, with the plagioclase show-
ing limited argillitization along the cleavage planes (Fig. 2a).
The scarce small K-feldspar phenocrysts are however slightly
kaolinized and the biotite suffered incipient chloritization.
The average grain size of the groundmass is 1—5 mm and
rare phenocrysts may attain up to 3—5 cm. A completely un-
weathered sample of the granite was not available due to
nowadays only limited quarry activity and widespread sur-
face kaolinization of the pluton (Pivec 1969).
The K-feldspar phenocrysts are strongly perthitic and
many show pronounced cross-hatched twinning; Carlsbad
twins are also common. They contain numerous inclusions
Fig. 2. Photomicrographs of the studied weakly porphyritic (muscovite-) biotite Říčany granite Žer-1. a – Photomicrograph of the typical
magmatic texture at a rim of a small K-feldspar phenocryst with characteristic cross-hatched twinning. Crystals of the main rock-forming
minerals exhibit no effects of solid-state deformation and are only slightly altered (plagioclase cores). Crossed polars. b – Back-scattered
electron (BSE) image of micro-inclusions of magnetite in, and around, larger biotite flake. c – Typical BSE image of biotite, rich in apatite
inclusions of variable shape and size. d – BSE image of the zircon-like ABO
4
-type phase enclosed in plagioclase.
260
JANOUŠEK, NAVRÁTIL, TRUBAČ, STRNAD, LAUFEK and MINAŘÍK
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA, 2014, 65, 4, 257—271
of biotite, quartz and oligoclase (an ‘hourglass’ texture of
Pivec 1970).
Plagioclase occurs as subhedral prismatic crystals with
common albite law twin lamellae, on average 1.2—2.0 mm
across. Apart from those enclosed within the K-feldspar,
they are of uniform oligoclase composition (An
11—19
; Appen-
dix 2) and are usually chemically unzoned.
Biotite forms subhedral flakes on average 0.5—0.75 mm (up
to 1.2 mm) across, with a strong pleochroism: X – straw yel-
low, Y = Z – dark rusty-brown. Pleochroic haloes around
submicroscopic inclusions are very distinct and common.
Biotite is often associated with grains of magnetite (Fig. 2b).
Anhedral quartz grains with weak undulose extinction are
up to 2 mm across.
The biotite may also be overgrown by scarce primary mus-
covite (up to 0.2 mm across). Also the plagioclase encloses
some small (0.1—0.2 mm) flakes of muscovite, at least some
of which appear to be of primary magmatic origin.
According to our microscopic and EPMA study and previ-
ous work of Kodymová & Vejnar (1974), the granite con-
tains significant proportions of rutile (0.1 mm). Small
apatite (0.1—0.2 mm) prisms or needles are often enclosed by
biotite (Fig. 2c). Euhedral crystals of titanite (0.1 mm), brown
to reddish in colour, and euhedral dipyramidal crystals of
zircon (0.2 mm), pink to pale brown are less common acces-
sories. Monazite (0.2 mm) usually occurs in forms aggregates
of spherical grains. Magnetite (0.1 mm) occurs mostly in an-
hedral fragments, only rarely forming octahedra, up to 0.5 mm
across. Ilmenite (0.1 mm) forms black opaque grains of non-
metallic lustre, sometimes lamellated. Subangular hedral zir-
con-like ABO
4
-type phase (up to 0.1 mm across) represents a
newly encountered, rather rare accessory mineral. It occurs in
plagioclase, very often together with monazite (Fig. 2d).
Modal proportions
Petrographic approaches
In the present case, both the conventional point counting
and image analysis of the stained slab (Fig. 3a—b; see Appen-
dix 1 for full analytical details on all methods used) were
taken to approximate well the mineral proportions in the
studied sample’s matrix, as no sizeable K-feldspar pheno-
crysts are present in the hand specimen. The results do not
differ greatly (Table 1). The image analysis (Fig. 3c—e) indi-
cates somewhat higher amounts of biotite (9.0 vs. 7.3 vol. %:
image analysis vs. conventional point counting, plagioclase
(31.2 vs. 27.1 vol. %) and quartz (38.1 vs. 32.8 vol. %), at the
expense of the K-feldspar (21.8 vs. 25.0 vol. %). Unfortu-
nately the contents of kaolinite could not be determined by
the image analysis. This mineral largely resisted staining and
was thus counted, at least in part, as plagioclase.
Chemical methods
For sample Žer-1, the following mesonormative composi-
tion (‘Improved Granite Mesonorm’ of Mielke & Winkler
1979) is obtained (wt. %): Or 29.1, Pl (Ab + An) 36.8,
Table 1:
Estimates
of
modal
percentages
(vol.
%
and
wt.
%,
original
data
in
bold)
of
the
individual
minerals
in
the
Říč
any
granite
Žer-1.
1
–
Data
on
mineral
densities
used
in
recalculation
of
wt.
%
to
vol
.
%
(or
vice
versa
)
are
from
Robie
et
al.
(1967),
whole-rock
density
of
the
Říčan
y
granite
from
Hejtman
(1948).
Original
data
are
in
bold.
2
–
For explanation of the individual methods (point-counting, imag
e analysis, least-squares calculation and XRD), see text. For m
ore detailed outcome of the constrained least-squares method,
see
Table
3.
3
–
‘Improved Granite Mesonorm’ of
Mielke & Winkler (1979).
261
DISTRIBUTION OF ELEMENTS AMONG MINERALS OF A SINGLE GRANITE SAMPLE
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA, 2014, 65, 4, 257—271
Fig. 3. Photograph of the polished hand specimen of the sample
Žer-1, of natural appearance (a) and stained (b). The remaining im-
ages are separated colour components corresponding to the plagio-
clase (c) K-feldspar (d) and quartz with biotite (e). Box in figure (b)
indicates the area used for image analysis (c—e).
SiO
2
70.34
TiO
2
0.36
Al
2
O
3
14.64
Fe
2
O
3
0.62
FeO
0.94
MgO
1.14
MnO
0.024
CaO
1.36
SrO
0.046
BaO
0.126
Li
2
O
0.015
Na
2
O
3.71
K
2
O
5.55
P
2
O
5
0.145
F
0.153
CO
2
0.06
C
0.016
S tot
<0.005
H
2
O
+
0.51
H
2
O
-
<0.05
F(ekv)
–0.064
S(ekv)
–0.001
Total
99.69
A/CNK
1.00
K
2
O/Na
2
O
1.50
Table 2: Whole-rock major- and
minor-element analysis of the stud-
ied sample (wt. %).
A/CNK =
= m olar Al
2
O
3
/(CaO + Na
2
O + K
2
O)
uncorrected for apatite.
K
2
O/Na
2
O ratio is given by weight.
For whole-rock trace-element com-
position, see Appendix 4.
Qtz 25.2, Bt 5.8, Ap 0.34, Mgt 0.9, Ilm 0.34, Cal 0.14 and
Crn 0.54 (Table 1). The negligible proportion of normative
corundum is in line with the subaluminous nature of the
sample (A/CNK ~ 1.0 – Table 2). Normative calcite and co-
rundum were further not considered as they do not corre-
spond to mineral phases present in the rock.
The approximate proportions of the main rock-forming
minerals (wt. %) were also obtained by the constrained least-
squares (LSQ) method (Albarède 1995) implemented in the
GCDkit. The whole-rock chemical composition and typical
EMP analyses of the main rock-forming minerals served as
an input (Table 3). Phosphorus, not present in appreciable
amounts in any of the main phases but representing an essen-
tial structural component in two common accessories, apatite
and monazite, was disregarded. The low sum of squared re-
siduals (R
2
= 0.071) indicates an excellent fit. The propor-
tions of K-feldspar (29.4 wt. %) and plagioclase (36.6 wt. %)
are closely comparable to the mesonorm; but the quartz per-
Table 3: Constrained least-squares approximation to the modal composition (wt. %).
1
–
Shown are real electron microprobe analyses of the main rock-forming minerals (see Appendix 2).
2
–
The observed whole-rock concentrations (Table 2) and the best estimate by the constrained least-squares method, with the correspond-
ing differences (residuals).
3
–
The sum of squared residuals indicating a goodness of fit.
Whole rock
2
Plagioclase
1
Pl14
K-feldspar Kfs1
Biotite Bt14
Quartz Real
Estimated
Residual
Squared
residual
SiO
2
64.05
64.16
37.50
99.86
70.34
70.44
–0.10
0.010
TiO
2
0.00
0.00
3.33
0.00
0.36
0.31
0.05
0.002
Al
2
O
3
21.76
18.47
14.80
0.02
14.64
14.78
–0.14
0.019
FeOt
0.04 0.03 17.26
0.00
1.50
1.64
–0.14
0.018
MnO
0.05
0.03
0.24
0.00
0.02
0.05
–0.02
0.001
MgO
0.00 0.00 13.35
0.00
1.14
1.25
–0.11
0.011
CaO
3.87
0.01
0.04
0.00
1.36
1.42
–0.06
0.004
Na
2
O
9.42
0.96
0.14
0.00
3.71
3.75
–0.04
0.001
K
2
O
0.40
15.61
9.47
0.00
5.55
5.61
–0.06
0.004
wt. %
36.6 %
29.4 %
9.3 %
24.7 %
R sqr
0.071
3
262
JANOUŠEK, NAVRÁTIL, TRUBAČ, STRNAD, LAUFEK and MINAŘÍK
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA, 2014, 65, 4, 257—271
centage (24.7 wt. %) is somewhat lower (Table 1). The
amount of biotite, however, is nearly double (9.3 wt. %) but
much closer to estimates by point counting and image analy-
sis (if recast to wt.%).
Powder X-ray diffraction (P-XRD)
The results of Rietveld quantitative phase analysis (Ap-
pendix 1) are summarized in Table 1 and Fig. 4. Compared
Fig. 4. Observed (circles), calculated (solid line) and difference
Rietveld profiles for the studied sample. Markers of the peak posi-
tions (five rows of the vertical line segments) are for (from top to
bottom) quartz, orthoclase, biotite, kaolinite and plagioclase.
Fig. 5. Whole rock-normalized trace-element patterns for individual rock-forming (a) and accessory (b) minerals. Mineral abbreviations
after Kretz (1983). Grey field denotes the total variation in the dataset.
with other methods, the plagioclase content is significantly
underestimated (25.1 wt. %). The reason may be the com-
plex preferred orientation of plagioclase, which shows two
cleavage plane systems, namely [001] and [010], while only
one correction for preferred orientation (for [001] direction)
was applied in the Rietveld fit. The amount of K-feldspar
(30.4 wt. %) is in agreement with the results of constrained
LSQ (29.4 wt. %) and mesonorm (29.1 wt. %); while the
quartz content is slightly higher (33.8 wt. %). The estimate
of biotite (9.5 wt. %) is in line with other methods, except
for the mesonorm. Clearly, the correction for preferred orien-
tation in the [001] works sufficiently in this case. The con-
tent of kaolinite (1.2 wt. %) is significantly lower than that
obtained by point counting (7.4 wt. %).
Mineral chemistry
The compositions of individual minerals were analysed
by a combination of three methods (Appendix 1), EMPA
(see averages and typical analyses in Appendix 2), in situ
LA ICP-MS and ICP-MS analyses of dissolved monominer-
alic separates (Appendix 3, summarized in Appendix 4). La-
ser-ablation analyses were preferred for mineral separates in
which the presence of minor admixture of phases/inclusions
with contrasting chemistry was of particular concern (quartz,
plagioclase and biotite). On the other hand, wet analysis was
chosen for K-feldspar, minimizing the problem of the small-
scale heterogeneity (abundance of perthite lamellae). Fig. 5
shows selected trace-element patterns normalized by the
whole-rock (WR) contents and Fig. 6 illustrates the chon-
drite-normalized REE patterns.
263
DISTRIBUTION OF ELEMENTS AMONG MINERALS OF A SINGLE GRANITE SAMPLE
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA, 2014, 65, 4, 257—271
Fig. 6. Chondrite (Boynton 1984)-normalized REE spiderplots for the individual minerals. Zrn-l = zircon-like ABO
4
-type phase. Grey field
denotes the total variation in the dataset.
Feldspars and kaolinite
K-feldspar is characterized by roughly double enrichment
of Ba, Rb and Pb over the whole-rock abundances; also Sr is
slightly elevated (Fig. 5a). The plagioclase contains rather
high Sr and Be (see Navrátil et al. 2002; Navrátil 2003). On
the other hand, concentrations of transition metals and high
field strength elements (HFSE) in both feldspars are very
low. The kaolinite is stripped of much of the original K-feld-
spars’ large-ion lithophile (LILE) budget, and contains only
Cd, Pb and Ta in amounts exceeding those in the whole rock.
The plagioclase contains more than twice as much REE as
the K-feldspar (81.3 vs. 26.6 ppm). Still the patterns are of
similar shape, featuring a high degree of LREE over HREE
enrichment and sizeable positive Eu anomalies (Fig. 6a). The
feldspars share normalized Eu contents nearly identical to the
whole rock but the concentrations of the remaining REE are
significantly lower. The limited REE mobility in course of
weathering of feldspars resulted in significantly higher appar-
ent concentrations in the residual kaolinite (
ΣREE ~191 ppm).
Only Eu contents approach those of the host rock and thus the
Eu anomaly turns to a strongly negative one.
Biotite and muscovite
The biotite is classified as phlogopite with 13.7—17.2 wt. %
total Al
2
O
3
(Al
IV
1.10—1.18 and Al
VI
0.16—0.27 apfu) and is
relatively Mg-rich (Mg/(Fe
T
+ Mg = 0.57—0.59). The content
of Al
2
O
3
for muscovite is 31.3 wt. % total (Al
IV
1.44 and
Al
VI
3.44 apfu). Both micas are slightly enriched in Rb, Mn
(see Navrátil et al. 2007), Ni, Co, Zn, Nb and Ta. Elevated
contents of Cd, Th, U, Zr and Hf are also characteristic of
muscovite. The main differences lie in the total REE con-
tents (7.3 ppm for biotite vs. 3396 ppm for muscovite). The
muscovite REE pattern is steep, with a much higher degree
of LREE/HREE enrichment compared to biotite (Fig. 6a).
The Eu anomaly is negative in both cases.
Accessory minerals
Apatite: As shown by EMPA, the main minor constituents
in apatite are Mn, Fe and Cl. It is further characterized by an
enrichment of Mn, Cd and U (Fig. 5b). The Pb contents are
close to those in the whole rock. Apatite is rich in REE, and
the MREE in particular (
ΣREE=3962 ppm; Fig. 6b). A neg-
ative Eu anomaly is characteristic.
Zircon: This mineral is rather Hf poor (0.01—0.02 apfu) but
contains high concentrations of other HFSE, such as Nb and
Ta. Zircon is a major Cd, Zn, Ni and Co bearing-phase. High
concentrations of U with Th are typical, as well as high
ΣREE (2615 ppm) and parabolic, convex-up chondrite-nor-
malized pattern with elevated HREE contents (Fig. 6b).
While the Eu anomaly is negative, the Ce anomaly, typical
of most igneous zircons (Hoskin & Schaltegger 2003), is ab-
sent. This, together with a rather low degree of HREE en-
richment, might reflect the presence of inclusions (see fig. 1
in Hoskin & Schaltegger op. cit.) or an admixture of the zir-
con-like (Zrn-l) phase.
Monazite: Monazite shows large concentrations of P, Zr
and Hf. Enrichment in radioactive elements is characteristic,
with Th prevailing over U. Pb contents are also high. This
mineral is enriched in REE, and especially LREE, with Ce
264
JANOUŠEK, NAVRÁTIL, TRUBAČ, STRNAD, LAUFEK and MINAŘÍK
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA, 2014, 65, 4, 257—271
being the most abundant (
ΣREE=91,205 ppm; Fig. 6b). A
deep negative Eu anomaly is characteristic.
Rutile: Rutile (TiO
2
> 98 wt. %) is a major transition metal-
bearing phase (Ni, Co and Cd); it displays elevated U and Th
contents. High HFSE concentrations (Zr, Hf, Nb and Ta) are
also typical. It is a REE-rich mineral (
ΣREE=7358 ppm)
with a fair degree of LREE/HREE fractionation and a deep
negative Eu anomaly.
Ilmenite and magnetite: Both minerals are characterized
by high contents of metals (Mn and Cd, less enriched Pb, Zn,
Co and Ni) as well as HFSE. The U and Th contents are also
somewhat elevated (Fig. 5b). The minerals are enriched in
REE (
ΣREE for ilmenite is 2735 ppm, for magnetite
3818 ppm), and the LREE in particular (Fig. 6b). The Eu
anomaly is deeply negative in both cases.
Titanite: Titanite shows low Al (0.09—0.11 apfu) and Fe
(0.03—0.03 apfu) contents. It is enriched in HFSE (Zr, Hf, Th
and U, less so Nb and Ta). Elevated Cd contents are also
typical. On the other hand, the LILE (Ba, Rb and Sr) are de-
pleted compared to the whole rock (Fig. 5b). The
ΣREE is
high (21,314 ppm) and the REE pattern relatively steep
(Fig. 6b) with a pronounced negative Eu anomaly.
Zircon-like ABO
4
-type phase: The results of the EMPA of
the zircon-like ABO
4
-type phase are summarized in Appen-
dix 5. Because of the relatively high Th contents, crystals are
metamict and altered, and yield low analytical totals. The
analysed phase appears to be a member of the zircon—thorite
solid solution series (0.27—0.33 apfu Th, 0.49—0.55 apfu Zr),
with substitution of Si by P (0.22—0.24 apfu) in the tetrahedral
position B. The phase is highly enriched in Fe (0.03—0.23 apfu)
and Ca (0.20—0.22 apfu. The Fe and Ca atoms, which proba-
bly enter the crystal structure during self-amorphization and
interactions with circulating fluids, could be positioned in
channels running parallel to the c axis of the crystal structure
(Geisler et al. 2002, 2003). Silica removal accompanied by a
hydration by post-magmatic, low-T fluids is also very probable.
Discussion
Modal analysis
Main rock-forming minerals
The choice of the most appropriate method for estimating
modal abundances of the main rock-forming minerals was
governed by comparison of the observed and calculated
model whole-rock contents of the major- and minor-element
oxides, as well as the most common among the LILE (Ba,
Rb and Sr, which are hosted in granites mostly by feldspars
and micas – Hanson 1978). Examining Fig. 7, it is clearly
seen that the constrained LSQ method provides the most accu-
rate estimate. As this approach does not give good assess-
ments of contents of minor phases, no muscovite was included
in the calculation and 0.2 wt. % of this mineral was added on
the basis of the point counting. The ‘best’ modal proportions
of the main rock-forming minerals are thus estimated to be:
0.365 Pl, 0.293 Kfs, 0.247 Qtz, 0.093 Bt and 0.002 Ms.
As for the rock classification in the QAP triangle (Fig. 8)
(here essentially the vol. % of Kfs, Pl and Qtz recast to 1),
the modal estimates by all methods yield comparable results
and the studied rock classifies consistently as monzogranite.
Still, the image analysis and, to some extent, point counting
tend to overestimate somewhat the Q proportion, whereas
the P-XRD approach results in a slightly lower 100*P/(A + P)
ratio of 47 % (other methods giving 55—59 %).
Fig. 7. Bar plots of calculated whole-rock contents of major- and minor-element oxides (a) and selected LILE (Rb, Sr and Ba) (b) calculated
from representative mineral analyses (Appendices 2 and 4) and modal abundances estimated on the basis of the P-XRD, granite mesonorm,
constrained least-squares (LSQ), image analysis of the stained thin slab and point counting of a standard thin section (Table 1). The best
agreement with the whole-rock contents (vertical orange lines) is achieved by the LSQ method.
265
DISTRIBUTION OF ELEMENTS AMONG MINERALS OF A SINGLE GRANITE SAMPLE
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA, 2014, 65, 4, 257—271
Fig. 8. QAP (quartz, K-feldspar, plagioclase) triangle of Streckeisen
(1974) comparing previous modal analyses of the Říčany granite
(Steinocher 1950, 1953; Palivcová 1965) with those obtained for
sample Žer-1 in course of the present study by various methods.
If necessary, the results were recast to vol. %.
Zr Nb La
Nd Eu Yb Y P
wt. %
Ap
2.875 1.166 1.431 0.661 0.473 0.314
0.314
Mnz
1.442 10.761 0.046
0.098 0.641 0.838 0.661 0.524 0.046
Zrn 0.032 0.116 2.459 2.882 3.780 0.550 1.240
0.032
Rt
0.327
0.081
0.655 0.634 3.543 0.691 1.391
0.081
Mgt
14.835 10.647 1.121
2.544 13.639 8.811 7.978
1.121
Ttn
1.443 9.563
0.343
0.260 0.498
0.340
0.255
0.255
Table 4: Maximum contents of selected accessory minerals (wt. %).
(ppm) Ap
1
Mnz Zrn Rt Real
2
Calc.
4
Zr
1.2 10086.7
453122.0 44550.4 145.5 145.5
Nb
0.1
35.3 3288.9 4675.2 3.8 4.6
La
424.5 26574.3
496.4 1861.8 12.2 13.5
Nd
1105.0 13148.3
446.9 2031.7 12.9 10.2
Eu
18.3
40.8 6.9 7.4
0.3
0.1
Yb
72.2 57.0 86.9 69.1
0.5 0.3
Y
1173.8 839.3 447.3 398.7 5.5 4.2
P
186228.0 111734.2
0.0
17.5 585.4 585.4
Estimated modal
proportions (wt. %)
3
0.290 %
0.040 %
0.023 %
0.082 %
Table 5: Modal percentages of the most abundant accessories (wt. %) estimated by the unconstrained least-squares method.
1
–
Real mineral analyses are summarized in the
Appendix 2.
2
–
The whole-rock composition not accounted for
by the main rock-forming minerals. See text for
discussion.
3
–
Estimated modal percentages obtained by the
least-squares approach.
4
–
Composition of a mixture (ppm) calculated using
these mineral proportions.
Accessory phases
After subtracting the elements hosted by the main rock-
forming minerals (not only hosted incorporated in the lattice,
Estimates are based on an assumption that the given
mineral contains the full whole-rock inventory of the
individual elements, after subtracting those hosted by
the main phases (0.365 Pl, 0.293 Kfs, 0.247 Qz, 0.093 Bt
and 0.02 Ms). Minima for each mineral are shown in
bold; these represent maximum possible contents of
each phase (last column).
but also contained in small inclusions, fluid inclusions or
sorbed at grain boundaries, interfacial films etc.), the rest of
the whole-rock chemical inventory has to be accounted for by
the accessories. A crude estimate of a maximum modal abun-
dance can be made using trace-element characteristics of the
individual minerals. Assuming that such a phase accommodates
the entire whole-rock inventory of the given element, these are
as follows (Table 4, elements in brackets being the most restric-
tive ones): apatite < 0.314 wt. % (P), monazite < 0.046 wt. %
(La), zircon < 0.032 wt. % (Zr), rutile < 0.081 wt. % (Nb),
magnetite < 1.121 wt. % (La) and titanite < 0.255 wt. % (Y).
For some minerals, these constraints should approach the
real modal abundances; the others contain no unique essen-
tial structural component, and instead share their trace-ele-
ment load with other mineral phase(s). For instance, the
maximal apatite content is likely to be overestimated, as part
of P resides in monazite. Likewise, the LREE hosted in min-
erals other than monazite can hardly be neglected.
Regarding the opaque phases, ilmenite and magnetite are
the possible carriers of the ferromagnetic properties in the
granite. It is worth stressing that the mean magnetic suscepti-
bility of the Říčany granite is low (13.13—105.3
×10
—6
[SI]:
Trubač et al. 2009) and thus the rock must contain only a
very limited amount of ferromagnetic component. Moreover,
in the constrained LSQ calculation, the whole-rock Fe bud-
get is exhausted by biotite and residual TiO
2
is as low as
0.05 wt. % (Table 3). According to petrographic investiga-
tion (Kodymová & Vejnar 1974), rutile seems to be the most
important Ti-bearing accessory. Ilmenite, magnetite and ti-
tanite were therefore not taken into further consideration, in
accordance with the microscopic and electron-microprobe
studies showing all three phases to be rather rare. The modal
abundance of the zircon-like ABO
4
phase remains unknown
as the complete trace-element data are not available.
In order to assess the abundances of the remaining acces-
sory minerals (Ap, Mnz, Zrn and Rt), the unconstrained LSQ
method was applied to the outstanding trace-element inven-
266
JANOUŠEK, NAVRÁTIL, TRUBAČ, STRNAD, LAUFEK and MINAŘÍK
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA, 2014, 65, 4, 257—271
tory. Trace elements which often represent their essen-
tial structural components (Zr, Nb, La, Nd, Eu, Yb, Y
and P) were selected. In the solution (Table 5), Ap
dominates (0.29 wt. %) over Rt (0.082 %), Mnz
(0.04 %) and Zrn (0.023 %). The results are feasible as
all obey the upper limits (Table 4) and the residuals for
all modelled elements are satisfactory.
Whole-rock geochemical budget
Total balance
The overall balance of individual elements in the
studied sample is illustrated by means of the balloon
plot (Jain & Warnes 2006). This diagram conveys im-
portant aspects of tabular data without obscuring the
exact numerical values (Fig. 9, see caption for explana-
tion of the principle). For the trace elements, the final
balance is also presented as an upper continental crust
(Taylor & McLennan 1995) normalized spiderplot
(Fig. 10). Here, both the real concentrations and com-
puted abundances match rather well. Only middle-
heavy REE and, most notably, Th and U are strongly
underestimated, suggesting the presence of an addi-
tional, U,Th, HREE-rich phase not accounted for in
our model. The obvious candidate is the zircon-like
ABO
4
mineral (Appendix 5; Fig. 6b), the role of which
could not be quantified as we do not have information
on its modal percentage and most of the trace-element
signature. Its mean concentrations of Th, U and Yb
nevertheless correspond to ~ 27,000, > 3500 and
> 350
× those found in the upper continental crust
(Taylor & McLennan 1995).
Fig. 9. Balloon plot (Jain & Warnes 2006) expressing the balance of indi-
vidual elements in the studied granite sample. The relative contributions
of individual minerals are expressed by the colour and size of circles
plotted. Each number was obtained by multiplication of the particular ele-
ment/oxide concentration within the given mineral and the best estimate
of its modal abundance. The goodness of fit could be assessed comparing
row totals with the real whole-rock analyses.
Fig. 10. Spiderplot of trace-element concentrations observed
in the whole-rock sample Žer-1 (filled squares) and calculated
(empty squares), normalized by abundances in the average
upper continental crust (Taylor & McLennan 1995). The cal-
culation was made for all the minerals present in Fig. 9.
267
DISTRIBUTION OF ELEMENTS AMONG MINERALS OF A SINGLE GRANITE SAMPLE
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA, 2014, 65, 4, 257—271
Residence in individual minerals
Returning to Fig. 9, one can examine the residence of par-
ticular elements in detail. Silica is hosted almost equally by
quartz (35 wt. % of its whole-rock abundance), plagioclase
(33 %), and somewhat less by K-feldspar (27 %). Bulk Ti,
Fe and Mg are held by biotite, whereby rutile contains a fifth
of the whole-rock TiO
2
. Much of the MnO is also accounted
for by biotite but almost a third is hosted by K-feldspar. As
expected, the main reservoirs for Al
2
O
3
are the feldspars
(Pl: 52 %, Kfs: 38 %). Surprisingly little Al, however, resides
in biotite (10 %), owing to its small modal abundance. Most
of the CaO and Na
2
O are bound to plagioclase ( ~ 90 %); only
10 % of the former is stored in apatite and 8 % of the latter in
K-feldspar. Part of the Na
2
O contents in K-feldspar is to be
ascribed to small plagioclase inclusions observed during pet-
rographic studies. For K
2
O, K-feldspar is clearly the key host
(81 %), biotite accounting for a mere 16 % of the whole-rock
budget. The main reservoir for P is apatite; monazite’s con-
tribution is small, close to 7 %.
Concerning the LILE, Rb is accumulated by K-feldspar
(55 %), biotite comes second (24 %) and plagioclase third
(20 %). The two feldspars represent almost equally impor-
tant Sr reservoirs (Pl: 57 %, Kfs: 42 %). The bulk of Ba is
concentrated in K-feldspar (83 %); both biotite (12 %) and
plagioclase (5 %) are rather unimportant. The overwhelming
majority of Be (79 %) is located in plagioclase; K-feldspar
and biotite share the rest. Lead is another element for which
the feldspars are the main hosts (Kfs: 61 %, Pl: 37 %).
The most important LREE reservoirs are monazite (41 % La)
and plagioclase (33 % La), while muscovite, K-feldspar,
rutile and apatite seem to be much less significant. The bulk of
Eu is accounted for by the feldspars (Kfs: 50 % and Pl: 38 %);
the apatite contribution represents a mere 8 %. On the other
hand, apatite is the most important host for the middle and
heavy REE as well as Y. Less significant HREE-bearing phase
is rutile. Surprisingly, zircon and monazite do not rank among
important sinks for HREE, owing to their scarcity.
Regarding the radioactive elements, both feldspars seem to
incorporate much U (Kfs: 35 and Pl: 25 %). Among accesso-
ries, rutile accounts for over 20 % of both Th and U. Almost
equally significant for Th are K-feldspar, monazite and rutile,
less so plagioclase (15 %) and muscovite (13 %). Minerals such
as zircon and apatite play only marginal roles in hosting the ra-
dioactive elements, unlike probably the newly recognized zir-
con-like ABO
4
phase, even though its role cannot be quantified.
Among the HFSE, the zircon contains only ~ 70 % of the
total Zr and Hf; with rutile accounting for the rest. About
60 % of the Nb and Ta budget are hosted by biotite. We can
only speculate that this may be in the form of submicroscopic
inclusions; for rutile remains approximately one fifth of the
whole-rock inventory.
General implications
Modal analysis
Petrographic techniques: In studies of granitic rocks, un-
stained thin sections are best suited for most textural obser-
vations, as well as description of optical properties and iden-
tification of mafic minerals. In many cases, however, it is
difficult to determine some of the colourless phases, particu-
larly if the feldspars are not twinned or their grains are small.
One way out is to use cold-stage cathodoluminescence (CL),
in which the (mostly) blue K-feldspars are easily distin-
guished from ochre—yellow plagioclases and almost non-lu-
minescent quartz (Marshall 1988; Janoušek et al. 2000a,
2004). CL also facilitates rapid determination and mapping
of some accessory phases, such as of apatite, monazite and
zircon. Nowadays largely neglected, staining turns K-feld-
spars yellow and plagioclase white, also facilitating point
counting (Gabriel & Cox 1929; Hollocher 2013).
Obtaining a statistically valid modal analysis by point
counting of a standard thin section can be troublesome, espe-
cially for coarse-grained or porphyritic rocks (Chayes &
Fairbairn 1951; Chayes 1954, 1965; Hutchison 1974). In or-
der to maximize the counted area in these cases, it was pro-
posed to use rock slabs or point count directly on the outcrop
(e.g. Hutchison 1974) – but such methods are cumbersome
and seldom used. Clearly, the situation becomes even more
complicated when the rock possesses a fabric, as one deals
essentially with a 2D section of a three-dimensional aniso-
tropic body.
In our case, the approach using the stained polished slab is
superior, as it represents an area larger than the standard sec-
tion (ca. 119 vs. 6 cm
2
). The current work confirms that the
tedious point-counting of the stained sample can be success-
fully replaced by computer-aided image analysis. However,
this method also brings about some new problems, for exam-
ple, some degree of ambiguity caused by alteration of the
feldspar.
P-XRD: The Rietveld method has been widely used for
quantitative phase analysis of various geological materials
and industrial products (Madsen & Scarlett 2009 and refer-
ences therein). This full-pattern profile fitting method has
several advantages over other diffraction approaches to
quantitative phase analysis using at most a few of the stron-
gest reflections from each phase in the mixture (Bish & Post
1989). Nevertheless, the application of the Rietveld method
to geological materials still poses significant problems. The
most pronounced difficulties are micro-absorption effects,
complex preferred orientation, various chemical substitu-
tions in minerals and also the presence of structural defects.
Our estimate of modal composition obtained by P-XRD is
probably biased by the defective structure of kaolinite,
which is affected by planar disorder. Considering the com-
plexity of the diffraction pattern, the large number of reflec-
tions and presence of significant preferred orientation, the
results from P-XRD are reasonable, in other words they are
comparable to other approaches. The advantage of the P-XRD
method over standard petrographic techniques is that it rep-
resents, like the chemical approaches, the whole volume of
the homogenized sample.
Chemical methods: For the sample Žer-1, the bulk-rock
analysis represents ~ 30 kg, or over 11,000 cm
3
of the rock,
and thus should be robust to the presence of fabrics, pheno-
crysts and local inhomogeneities such as schlieren. Major-
element or norm-based schemes (Streckeisen & Le Maitre
268
JANOUŠEK, NAVRÁTIL, TRUBAČ, STRNAD, LAUFEK and MINAŘÍK
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA, 2014, 65, 4, 257—271
1979; De La Roche et al. 1980; Debon & Le Fort 1988) have
been preferred by many authors for classification purposes.
However, some normative recalculations involve minerals
not encountered in common felsic plutonites and thus their
use is not straightforward (e.g. CIPW and Catanorm – Hut-
chison 1975). For this reason, Mielke & Winkler (1979) de-
veloped their ‘Improved Mesonorm’, designed specifically
for granitic rocks.
With the spread of personal computers, mathematical
methods became popular. They attempt to find the best esti-
mate of modal proportions based on major- and minor-ele-
ment analyses of the large whole-rock sample and its typical
mineral constituents. As usually the number of chemical
components exceeds that of the mineral phases, there is no
unique solution and the algorithms rely on the least-squares or
linear programming approaches (e.g. Wright & Doherty 1970;
Le Maitre 1981; Laube et al. 1996; Paktunc 1998; Janoušek et
al. 2006). Potential pitfalls could lie in the choice of typical
mineral compositions and/or the weathering/alteration ef-
fects. The methods are also rather insensitive to the presence
of minor/or accessory phases, whose contents have to be as-
sessed separately.
Here the results obtained by mesonormative and LSQ ap-
proaches are closely comparable for felsic minerals (Fig. 8).
The amount of biotite, however, is nearly half (5.8 vs.
9.3 wt. %, respectively), and much lower also than estimates
by the remaining methods. The reason probably lies in the
fact that the mesonorm also estimates the proportion of mag-
netite and ilmenite, whereby in the LSQ calculation the sole
iron-bearing phase is biotite. The normative contents of the
opaque phases seem overestimated by the mesonormative al-
gorithm, especially that of magnetite (0.9 %).
Taken together, several of the methods employed here to
estimate the modal proportions of the main rock-forming
minerals have yielded comparable results. It is necessary to
stress that the current case is a particularly favourable one, as
no significant zoning of mineral phases, not even of plagio-
clase, has been observed in the Říčany Pluton. Moreover,
K-feldspar phenocrysts in the studied sample are small and
sparse. Still, the constrained LSQ method (Albarède 1995)
was found superior to all other approaches. The obtained
percentages of the main rock-forming minerals compare well
with the modal analyses determined previously by conven-
tional point counting (Steinocher 1950, 1953; Palivcová
1965) (Fig. 8).
Petrogenetic modelling
Collectively, the four accessories, apatite, monazite, zircon
and rutile host more than half of the total LREE and over
three quarters of the HREE + Y, Zr and Hf of the studied gran-
ite sample. The cases of Th, U, Ti, Ni, Co, Nb and Ta are sim-
ilar, albeit not so extreme. This confirms the notion that the
accessories are often crucial in controlling the behaviour of
many elements in granitic systems (e.g. Mittlefehldt & Miller
1983; Miller & Mittlefehldt 1984; Bea 1996). Particularly
the REE are of little value in petrogenetic modelling of frac-
tional crystallization of the main rock-forming minerals, un-
less the role of accessories is properly assessed.
Determination of the whole-rock trace-element abundances
Some 10—30 % of the whole-rock HREE, U, Th, Nb, Ta, Ti,
Cd, Co and Ni, and even more of Zr and Hf, are contained in
the resistant accessories, zircon and rutile. Thus the pressure
bomb or sample fusion are absolutely essential in sample de-
composition, if these elements are to be determined quantita-
tively (e.g. Potts 1987; García de Madinabeitia et al. 2008 and
references therein). Using a combined HF, HCl and HNO
3
at-
tack is clearly inadequate and will induce low total REE con-
tents and artificially increased degrees of LREE/HREE
fractionation. Alas such cases are by no means rare – for in-
stance, the acid decomposition casts serious doubts on some
of the results by Minařík et al. (1998).
Saturation thermometry
Our study provides some interesting implications for the
accessory phase saturation thermometry commonly used in
the granite studies (see Janoušek 2006 and Anderson et al.
2008 for review). As in Říčany phosphorus is hosted nearly
exclusively by apatite (86 % of the whole-rock contents), the
saturation thermometry involving this mineral should apply
be applicable. Indeed the temperature calculated for the un-
corrected whole-rock P
2
O
5
content and that based on apatite-
hosted P
2
O
5
are both mutually closely comparable (946 vs.
931 °C according to the model of Harrison & Watson 1984).
The fact that they are unrealistically high for a granitic mag-
ma can be related to the sensitivity of the algorithm to exact
determinations of P
2
O
5
contents, as the isotherms converge
rapidly for increasingly acidic compositions (see fig. 3a in
Janoušek 2006).
Zircon incorporates a mere 62 % of the whole-rock Zr
budget (Fig. 9), the rest being hosted by other phases, most
importantly the early crystallizing rutile. The zircon satura-
tion thermometry (Watson & Harrison 1983) thus yields a
liquidus temperature estimate (784 °C) ca. 40 °C higher than
that calculated for the corrected Zr concentration.
The case of monazite is more complex, though. The model
of Montel (1993) includes additional parameters apart from
the temperature, major-element and LREE contents of the
magma. One is the fraction of REE phosphates in the mona-
zite but this can be directly measured (the average for Říčany
monazite is 89.5 mol %). More difficult to constrain are the
water contents of the magma. Fortunately, the effect is not
great (738 °C is obtained for 3 wt. % H
2
O; 761—724 °C for
1—5 wt. % H
2
O).
Plagioclase represents an important sink for LREE, with
monazite containing some 21—41 % of the whole-rock in-
ventory. This brings problems for the LREE-based monazite
thermometry. While the uncorrected LREE content yields
738 °C, the corrected content gives barely 668 °C, that is
about 70 °C less (for 3 wt. % H
2
O in the magma melt). Such
a corrected temperature seems to be too low for a granitic
magma. However, one should take into account the petrolog-
ical evidence indicating that the monazite saturation level
was probably reached early, prior to the onset of the plagio-
clase crystallization. Thus using the uncorrected temperature
seems more justifiable here.
269
DISTRIBUTION OF ELEMENTS AMONG MINERALS OF A SINGLE GRANITE SAMPLE
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA, 2014, 65, 4, 257—271
Conclusions
The study of the petrography and mineral chemistry of a
single large sample of coarse-grained, weakly porphyritic
(muscovite-) biotite granite from the Říčany intrusion
(Czech Republic) has yielded the following conclusions:
1. The methods taking into account the whole-rock com-
position as well as the true mineral chemistries (linear pro-
gramming or constrained least squares) are particularly
suitable for modal analyses of granitic rocks. The modes
then represent a large volume of sample and thus (i) smooth
out any local variations, such as small-scale crystal accumula-
tion, (ii) account for the presence of phenocrysts, and (iii) are
insensitive to shaped-preferred orientation/fabric.
2. Accessory phases control the behaviour of many trace el-
ements in differentiation of felsic granitic systems. Clearly the
REE are of little value in petrogenetic modelling of the main
rock-forming minerals fractionation, unless the role of acces-
sories is properly assessed and existing saturation models for
apatite, zircon, monazite ± rutile are carefully considered.
The fact that many of the essential structural components
(P, Zr, LREE) used in apatite, zircon and monazite saturation
thermometry are incorporated into other minerals may lead
to significant overestimation of the liquidus temperatures. In
the present case, the saturation temperatures for zircon
would be overestimated as ca. 40 °C due to significant con-
tents of Zr in the early crystallized rutile. However, the cor-
rection for monazite (—70 °C) probably should not be applied
as most of the extra LREE are hosted by plagioclase, a rela-
tively late mineral.
In the studied sample, over 80 % of the whole-rock Zr and
Hf and ca. 10—30 % of HREE, U, Th, Nb, Ta, Ti, Cd, Co and
Ni are contained in resistant accessory phases Zrn and Rt.
Thus the pressure vessel or sample fusion – and not merely
a combined acid attack – are absolutely essential in sample
decomposition if these elements are to be determined quanti-
tatively.
Acknowledgments: This manuscript benefited from insight-
ful comments by I. Broska, an anonymous reviewer and han-
dling editor I. Petrík, as well as from careful editing by E.
Petríková. The authors are indebted to Z. Korbelová for assis-
tance at the electron microprobe and J.K. Novák (both Geo-
logical Institute, Czech Academy of Sciences, Prague – GLI)
who carried out the point counting analysis. V. Sedláček
(GLI) did the mineral separations, P. Hasalová (Czech Geo-
logical Survey) helped with staining of feldspars. Janoušek
and Trubač acknowledge the support by the Grant Agency of
the Czech Republic (GAČR, No. P210/11/1168) and Czech
Ministry of Education, Youth and Sports (Project LK 11202);
Minařík and Navrátil were supported from institutional
Project No. RVO67985831. This study is a part of the Ph.D.
research of Jakub Trubač.
References
Albarède F. 1995: Introduction to geochemical modeling. Cam-
bridge University Press, Cambridge, 1—543.
Anderson J.L., Barth A.P., Wooden J.L. & Mazdab F. 2008: Ther-
mometers and thermobarometers in granitic systems. In: Putirka
K.D. & Tepley III F.J. (Eds.): Minerals, inclusions and volca-
nic processes. Mineral. Soc. Amer.; Geochem. Soc. Rev. Miner.
Geochem., Washington 69, 121—142.
Bea F. 1996: Residence of REE, Y, Th and U in granites and crustal
protoliths; implications for the chemistry of crustal melts. J.
Petrology 37, 521—552.
Bish D.L. & Post J.E. 1989: Modern powder diffraction. Mineral.
Soc. Amer. Rev. Miner., Washington 20, 1—369.
Boynton W.V. 1984: Cosmochemistry of the rare earth elements:
meteorite studies. In: Henderson P. (Ed.): Rare earth element
geochemistry. Elsevier, Amsterdam, 63—114.
Chayes F. 1954: The theory of thin-section analysis. J. Geol. 62,
92—101.
Chayes F. 1965: Reliabilty of point counting results. Amer. J. Sci.
263, 719—724.
Chayes F. & Fairbairn H.W. 1951: A test of the precision of thin-sec-
tion analysis by point counter. Amer. Mineralogist 36, 704—712.
Debon F. & Le Fort P. 1988: A cationic classification of common
plutonic rocks and their magmatic associations: principles,
method, applications. Bull. Minéral. 111, 493—510.
De La Roche H., Leterrier J., Grandclaude P. & Marchal M. 1980:
A classification of volcanic and plutonic rocks using R
1
R
2
-dia-
gram and major element analyses – its relationships with cur-
rent nomenclature. Chem. Geol. 29, 183—210.
Evans O.C. & Hanson G.N. 1993: Accessory-mineral fractionation
of rare-earth element (REE) abundances in granitoid rocks.
Chem. Geol. 110, 69—93.
Gabriel A. & Cox E.P. 1929: A staining method for the quantitative
determination of certain rock minerals. Amer. Mineralogist 14,
290—292.
García de Madinabeitia S., Sánchez Lorda M.E. & Gil Ibarguchi J.I.
2008: Simultaneous determination of major to ultratrace ele-
ments in geological samples by fusion-dissolution and induc-
tively coupled plasma mass spectrometry techniques. Anal.
Chim. Acta 625, 117—130.
Geisler T., Pidgeon R.T., van Bronswijk W. & Kurtz R. 2002:
Transport of uranium, thorium, and lead in metamict zircon
under low-temperature hydrothermal conditions. Chem. Geol.
191, 141—154.
Geisler T., Pidgeon R.T., Kurtz R., van Bronswijk W. & Schleicher
H. 2003: Experimental hydrothermal alteration of partially
metamict zircon. Amer. Mineralogist 88, 1496—1513.
Gromet L.P. & Silver L.T. 1983: Rare earth element distribution
among minerals in a granodiorite and their petrogenetic impli-
cations. Geochim. Cosmochim. Acta 47, 925—939.
Hanson G.N. 1978: The application of trace elements to the petro-
genesis of igneous rocks of granitic composition. Earth Planet.
Sci. Lett. 38, 26—43.
Harrison T.M. & Watson E.B. 1984: The behavior of apatite during
crustal anatexis: equilibrium and kinetic considerations.
Geochim. Cosmochim. Acta 48, 1467—1477.
Hejtman B. 1948: Directory of quarries in Czechoslovakia, No. 26,
Český Brod. [Soupis lomů ČSR, č. 26, okres Český Brod.]
SGÚ ČSR, Praha, 1—71 (in Czech).
Hollocher K. 2013: Staining feldspars in thin section. Accessed on
23 December 2013 at http://minerva.union.edu/hollochk/
c_petrology/staining_feldspars.htm
Holub F.V., Machart J. & Manová M. 1997: The Central Bohemian
Plutonic Complex: geology, chemical composition and genetic in-
terpretation. Sbor. Geol. Věd, ložisk. Geol. Mineral. 31, 27—50.
Hoskin P.W.O. & Schaltegger U. 2003: The composition of zircon
and igneous and metamorphic petrogenesis. In: Hanchar J.M.
& Hoskin P.W.O. (Eds.): Zircon. Mineral. Soc. Amer.;
Geochem. Soc. Rev. Miner. Geochem., Washington 53, 27—62.
270
JANOUŠEK, NAVRÁTIL, TRUBAČ, STRNAD, LAUFEK and MINAŘÍK
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA, 2014, 65, 4, 257—271
Hutchison C.S. 1974: Laboratory handbook of petrographic tech-
niques. John Wiley & Sons, New York, 1—527.
Hutchison C.S. 1975: The norm, its variations, their calculation and
relationships. Schweiz. Mineral. Petrogr. Mitt. 55, 243—256.
Jain N. & Warnes G.R. 2006: Ballon plot. R News 6, 35—38.
Janoušek V. 2006: Saturnin, R language script for application of ac-
cessory-mineral saturation models in igneous geochemistry.
Geol. Carpathica 57, 131—142.
Janoušek V., Farrow C.M. & Erban V. 2006: Interpretation of
whole-rock geochemical data in igneous geochemistry: intro-
ducing Geochemical Data Toolkit (GCDkit). J. Petrology 47,
1255—1259.
Janoušek V., Rogers G., Bowes D.R. & Vaňková V. 1997: Cryptic
trace-element variation as an indicator of reverse zoning in a
granitic pluton: the Říčany granite, Czech Republic. J. Geol.
Soc., London 154, 807—815.
Janoušek V., Bowes D.R., Braithwaite C.J.R. & Rogers G. 2000a:
Microstructural and mineralogical evidence for limited involve-
ment of magma mixing in the petrogenesis of a Hercynian
high-K calc-alkaline intrusion: the Kozárovice granodiorite,
Central Bohemian Pluton, Czech Republic. Trans. Roy. Soc.
Edinb., Earth Sci. 91, 15—26.
Janoušek V., Bowes D.R., Rogers G., Farrow C.M. & Jelínek E.
2000b: Modelling diverse processes in the petrogenesis of a
composite batholith: the Central Bohemian Pluton, Central Eu-
ropean Hercynides. J. Petrology 41, 511—543.
Janoušek V., Braithwaite C.J.R., Bowes D.R. & Gerdes A. 2004:
Magma-mixing in the genesis of Hercynian calc-alkaline gran-
itoids: an integrated petrographic and geochemical study of the
Sázava intrusion, Central Bohemian Pluton, Czech Republic.
Lithos 78, 67—99.
Johannsen A. 1932, 1937, 1938, 1939: A descriptive petrography of
the igneous rocks. University of Chicago Press.
Katzer F. 1888: Geologische Beschreibung der Umgebung von
Říčan. Jb. Geol. Reichsanst. 38, 355—417.
Kodymová A. & Vejnar Z. 1974: Accessoric heavy minerals in plu-
tonic rocks of the Central Bohemian Pluton. [Akcesorické těžké
minerály v hlubinných horninách středočeského plutonu.] Sbor.
Geol. Věd, ložisk. Geol. Mineral. 16, 89—128 (in Czech).
Kretz R. 1983: Symbols for rock-forming minerals. Amer. Mineral-
ogist 68, 277—279.
Laube N., Hergarten S. & Neugebauer H.J. 1996: MODUSCALC
– a computer program to calculate mode from a geochemical
rock analysis. Comput. and Geosci. 22, 631—637.
Le Maitre R.W. 1981: GENMIX – a generalized petrological mix-
ing model program. Comput. and Geosci. 7, 229—247.
Le Maitre R.W. 2002: Igneous rocks: a classification and glossary
of terms: recommendations of the International Union of Geo-
logical Sciences, subcommission on the systematics of igneous
rocks. Cambridge University Press, 1—236.
Madsen I.C. & Scarlett N.V.Y. 2009: Quantitative phase analysis.
In: Dinnebier R.E. & Bilinge S.J.J. (Eds.): Powder diffraction:
Theory and practise. RCS Publishing, Cambridge, 298—331.
Marshall D.J. 1988: Cathodoluminescence of geological materials.
Unwin Hyman, Boston, 1—145.
Mielke P. & Winkler H.G.F. 1979: Eine bessere Berechnung der
Mesonorm für granitische Gesteine. Neu. Jb. Mineral., Mh.,
471—480.
Miller C.F. & Mittlefehldt D.W. 1984: Extreme fractionation in fel-
sic magma chambers; a product of liquid-state diffusion or
fractional crystallization? Earth Planet. Sci. Lett. 68, 151—158.
Minařík L. & Houdková Z. 1986: Element distribution during the
weathering of granitic rocks and formation of soils in the area
of the massif of Říčany. [Distribuce prvků při zvětrávání hornin
a tvorbě půd v oblasti říčanského masívu.] Acta Montana 74,
59—78 (in Czech, with English summary).
Minařík L. & Kvídová O. 1986: Fractionation of rare earth elements
during weathering of rocks. [Frakcionace vzácných zemin při
zvětrávání hornin.] Acta Montana 72, 63—74 (in Czech, with
English summary).
Minařík L., Žigová A., Bendl J., Skřivan P. & Š astný M. 1998: The
behaviour of rare-earth elements and Y during the rock weath-
ering and soil formation in the Říčany granite massif, Central
Bohemia. Sci. Total Environ. 215, 101—111.
Minařík L., Skřivan P., Novák J.K., Fottová D. & Navrátil T. 2003:
Distribution, cycling and impact of selected inorganic contami-
nants in ecosystem of the Lesní potok catchment, the Czech
Republic. Ekologia, Bratislava 22, 305—322.
Mittlefehldt D.W. & Miller C.F. 1983: Geochemistry of the Sweet-
water Wash Pluton, California; implications for “anomalous”
trace element behavior during differentiation of felsic magmas.
Geochim. Cosmochim. Acta 47, 109—124.
Montel J.M. 1993: A model for monazite/melt equilibrium and appli-
cation to the generation of granitic magmas. Chem. Geol. 110,
127—146.
Navrátil T. 2003: Biogeochemistry of the II.A group elements in a
forested catchment. Unpublished Ph.D. Thesis, Charles Uni-
versity in Prague, 1—113.
Navrátil T., Skřivan P., Minařík L. & Žigová A. 2002: Beryllium
geochemistry in the Lesní Potok Catchment (Czech Republic),
7 years of systematic study. Aquat. Geochem. 8, 121—133.
Navrátil T., Vach M., Skřivan P., Mihaljevič M. & Dobešová I.
2004: Deposition and fate of lead in a forested catchment,
Lesní potok, Central Czech Republic. Water Air Soil Poll., Fo-
cus 4, 619—630.
Navrátil T., Shanley J.B., Skřivan P., Krám P., Mihaljevič M. &
Drahota P. 2007: Manganese biogeochemistry in a Central
Czech Republic catchment. Water Air Soil Poll. 186, 149—165.
Němec D. 1978: Genesis of aplite in the Říčany massif, central Bo-
hemia. Neu. Jb. Mineral., Abh. 132, 322—339.
Niggli P. 1931: Die quantitative mineralogische Klassifikation der
Eruptivgesteine. Schweiz. Mineral. Petrogr. Mitt. 11, 296—364.
Orlov A. 1933: Contribution to the petrography of the Central Bo-
hemian Granite Massif (the Říčany-Benešov-Milevsko-Písek
region). [Příspěvek k petrografii středočeského žulového masívu
(Říčansko—Benešovsko—Milevsko-Písecko).] Věst. St. Geol. Úst.
Čs. Repl. 9, 135—144 (in Czech).
Paktunc A.D. 1998: MODAN: an interactive computer program for
estimating mineral quantities based on bulk composition. Com-
put. and Geosci. 24, 425—431.
Palivcová M. 1965: The Central Bohemian Pluton – a petrographic
review and an attempt at a new genetic interpretation. Krystal-
inikum 3, 99—131.
Palivcová M., Waldhausrová J., Ledvinková V. & Fatková J. 1992:
Říčany granite (Central Bohemian Pluton) and its ocelli- and
ovoids-bearing mafic enclaves. Krystalinikum 21, 33—66.
Pivec E. 1970: On the origin of phenocrysts of potassium feldspars
in some granitic rocks of the Central Bohemian Pluton. Acta
Univ. Carol, Geol. 1970, 11—25.
Pivec E. 1969: Residues of surface kaolinization in granite of Říčany.
[Relikty povrchové kaolinizace v říčanské žule.] Čas. Mineral.
Geol. 14, 61—67 (in Czech, with English summary).
Potts P.J. 1987: A handbook of silicate rock analysis. Blackie & Son
Ltd., Glasgow and London, 1—622.
Robie R.A., Bethke P.M. & Beardsley K.M. 1967: Selected X-ray
crystallographic data, molar volumes, and densities of minerals
and related substances. U.S. Geol. Surv. Bull. 1248, Washington,
1—87.
Sawka W.N. 1988: REE and trace element variations in accessory
minerals and hornblende from the strongly zoned McMurry
Meadows Pluton, California. Trans. Roy. Soc. Edinb., Earth
Sci. 79, 157—168.
271
DISTRIBUTION OF ELEMENTS AMONG MINERALS OF A SINGLE GRANITE SAMPLE
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA, 2014, 65, 4, 257—271
Steinocher V. 1950: The position of some plutonic and dike rocks
of the plutonic mass of Central Bohemia in P. Niggli’s quanti-
tative mineralogical and chemical system. Part I. Sbor. St.
Geol. Úst. Čs. Rep.,Odd. Geol. 17, 721—764.
Steinocher V. 1953: The position of some plutonic and dyke rocks
of the Pluton of Central Bohemia in P. Niggli’s quantitative
mineralogical and chemical system. Part II. Sbor. Ústř. Úst.
Geol., Odd. Geol. 20, 241—288.
Streckeisen A. 1974: Classification and nomenclature of plutonic
rocks. Geol. Rdsch. 63, 773—786.
Streckeisen A. & Le Maitre R.W. 1979: A chemical approximation
to the modal QAPF classification of the igneous rocks. Neu.
Jb. Mineral., Abh. 136, 169—206.
Taylor S.R. & McLennan S.M. 1995: The geochemical evolution of
the continental crust. Rev. Geophys. 33, 241—265.
Trubač J., Žák J., Chlupáčová M. & Janoušek V. 2009: Magnetic
fabric of the Říčany granite, Bohemian Massif: a record of he-
lical magma flow? J. Volcanol. Geotherm. Res. 181, 25—34.
Wark D.A. & Miller C.F. 1993: Accessory mineral behavior during
differentiation of a granite suite: monazite, xenotime and zir-
con in the Sweetwater Wash pluton, southeastern California,
U.S.A. Chem. Geol. 110, 49—67.
Watson E.B. & Harrison T.M. 1983: Zircon saturation revisited:
temperature and composition effects in a variety of crustal
magma types. Earth Planet. Sci. Lett. 64, 295—304.
Wright T.L. & Doherty P.C. 1970: A linear programming and least
squares computer method for solving petrologic mixing prob-
lems. Geol. Soc. Amer. Bull. 81, 1995—2008.
Žák J., Verner K., Janoušek V., Holub F.V., Kachlík V., Finger F.,
Hajná J., Tomek F., Vondrovic L. & Trubač J. (2014): A plate-
kinematic model for the assembly of the Bohemian Massif
constrained by structural relationships around granitoid plu-
tons. In: Schulmann K., Martínez Catalán J.R., Lardeaux J.M.,
Janoušek V. & Oggiano G. (Eds.): The Variscan orogeny: Ex-
tent, timescale and the formation of the European Crust. Geol.
Soc. London, Spec. Publ. 405, 169—196. Doi: 10.1144/SP405.9
www.geologicacarpathica.com
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
, AUGUST 2014, 65, 4
Electronic supplement
Analytical techniques
Sampling
The whole-rock sample Žer-1 ( ~ 30 kg) was collected in
the Žernovka quarry, ca. 850 m NNW of the namesake vil-
lage off the pluton’s center (Fig. 1c, GPS 50°0’22.64” N;
14°44’56.13” E). Its bulk was crushed by a steel jaw crusher,
homogenized and ground in an agate mill at the Laboratories
of the Institute of Geology, Czech Academy of Sciences,
v.v.i. (CAS).
Mineral separation
The crushed material ( < 500 µm) was floated on a Wilfley
shaking table. The heavy fraction was filtered on glass frit,
rinsed with ethanol and air dried. The light fraction was
dried at 105 °C and was used for separation of the main
rock-forming minerals. The fraction between 63 and 250 µm
thereof served for magnetic separation of biotite. The biotite
separate was finally cleaned in acetylene tetrabromide with
density of 2.54 kg . m
—3
.
The quartz separate was obtained after biotite separation by
flotation using ANP1 (amino nitro paraffin) agent. K-feldspar
was separated in bromoform solution (2.59 kg . m
—3
). The
material with density greater than 2.54 kg . m
—3
was plagio-
clase. Following the separation of quartz and orthoclase,
muscovite separation was achieved in bromoform with den-
sity 2.74—2.84 kg . m
—3
. Kaolinite was isolated by centrifuga-
tion of the fine fraction ( < 25 µm) in bromoform.
The dried heavy fraction from the shaking table was sieved
on a 315 µm nylon sieve. The fine fraction was used for sepa-
ration of heavy minerals – zircon and monazite. Apatite was
recovered from the heavy fraction by centrifugation in methyl-
eniodide with density 3.18—3.25 kg . m
—3
. Magnetite was ob-
tained from the dark part of the heavy fraction by magnetic
separation with the intensity of magnetic field set to 0.2 A.
Electron microprobe and BSE imaging
The analyses of the major rock-forming and selected acces-
sory minerals were done with a fully automated CAMECA
SX-100 electron microprobe, employing
Φ(ρz) correction
procedure (Merlet 1992) at the CAS. All analyses were per-
formed at an acceleration voltage of 15 kV but with different
beam currents and spot sizes chosen according to the mineral
type. Thus 20—40 nA and a spot size of 10 µm were employed
for analyses of apatite, titanite, rutile, monazite, magnetite, il-
menite and zircon—thorite solid solution. For feldspars, micas
and quartz, the beam current was 10—15 nA and spot size
2 µm. Minor-element interferences have been checked rou-
tinely and corrected for by measuring the corresponding stan-
dards. All the mineral abbreviations are after Kretz (1983).
Whole-rock major- and minor-element analyses
The major-element whole-rock analysis of the whole-rock
sample was undertaken by wet chemistry in the laboratories of
the Czech Geological Survey, Prague – Barrandov. Further
analytical details are given in Dempírová (2010); the relative
2
σ uncertainties were better than 1 % (SiO
2
), 2 % (FeO),
5 % (Al
2
O
3
, K
2
O and Na
2
O), 7 % (TiO
2
, MnO, CaO), 6 %
(MgO) and 10 % (Fe
2
O
3
, P
2
O
5
). Interpretation and plotting
of the whole-rock geochemical data was done by the R-lan-
guage package GCDkit (Janoušek et al. 2006).
Trace-element analyses
Solution ICP-MS analyses
The trace-element analyses of mineral separates and
whole-rock samples were carried out after modified total di-
gestion in mineral acids (HF + HClO
4
) and borate fusion
(Na
2
CO
3
+ Na
2
B
4
O
7
) in Pt crucibles followed by solution
nebulization ICP-MS PQ3 VG Elemental at Charles Univer-
sity in Prague. All the chemicals involved were reagent
grade (Merck, Germany) and the acids were double distilled.
Deionized water from a Millipore system (Milli-Q Academic,
USA) was used for all dilutions.
The measured data were processed on-line using VG Plas-
maLab software, applying corrections for instrumental drift.
The analytical precision for all the elements analysed ranged
from 0.5 to 5 % relative. The accuracy of this analytical
method was checked using the G-2 and BCR-2 reference ma-
terials (USGS, USA). Trace-element ICP-MS analyses fol-
lowed the methods of Strnad et al. (2005).
In-situ laser ablation analyses
In-situ trace-element analyses were performed on a qua-
drupole-based ICP-MS Thermo Fisher X-Series II (Charles
University) coupled to a NewWave UP 213 laser microprobe
(NewWave Research; USA) operating at output wavelength
of 213 nm. All samples were prepared as polished thin sec-
tions. The data were acquired in the time-resolved and peak
jumping mode with one point measured per mass peak and
processed off-line by Thermo Fisher PlasmaLab software
2.5.11 321. The raster pattern was linear, approximately
JANOUŠEK et al.: Distribution of elements among minerals of a single (muscovite-) biotite granite sample – an opti-
mal approach and general implications
Appendix 1
ii
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA, 2014, 65, 4
JANOUŠEK et al.: Distribution of elements among minerals of a single granite sample
80
×400 µm. External calibration of the laser-ablation analy-
ses was done using standards NIST 610 and 612 (Pearce et al.
1997). For internal standardization
29
Si concentrations based
on electron microprobe measurements were applied. Data re-
duction included correction for the gas blank, the internal
standard and a calibration check; the data were processed
off-line in a MS Excel spreadsheet-based program. For de-
tails on analytical protocol and correction strategy see Strnad
et al. (2005).
Modal analyses
Point counting
Point counting was done on a thin section 3
×2 cm using a
standard optical microscope. The total number of points
counted was 4890.
Image analysis of stained rock slab
The rock sample was stained according to the method of
Gabriel & Cox (1929). The polished sample surface was ex-
posed to fumes of HF for about 15 min., then treated with a
concentrated solution of sodium cobaltinitrite for another
10 min., rinsed and dried. This reaction forms a yellow coat-
ing on K-feldspar and white on plagioclase; quartz remains
unaffected.
The image analysis was performed on a scanned stained
polished rock slab (ca. 17
×7 cm). The abundance of mineral
phases was estimated on the basis of pixel colour analysis by
Quick Photo Micro 2.2 software. Dark—black areas were at-
tributed to biotite, grey and greyish glassy-like colour to
quartz, white to plagioclase and yellow to K-feldspar.
Powder X-ray diffraction
A powder sample was prepared by grinding in an agate mill
and, subsequently, in an agate mortar. The powder X-ray dif-
fraction pattern was collected in a conventional Bragg-Bren-
tano geometry on the Philips X’Pert diffractometer equipped
with graphite secondary monochromator. The CuK
α
radiation
was used. To minimize the background, the sample was
placed on a flat low-background silicon wafer. Data were ac-
quired in the angular range 3—90° 2
Θ, with a step interval of
0.02° and a step-counting time of 9 s. A divergence slit of 0.5°
and a receiving slit of 0.1 mm were used.
The quantitative phase analysis of the sample was per-
formed using the Rietveld method (Young 2000). Refinement
was done by minimizing the sum of the weighted squared dif-
ferences between observed and calculated intensities at every
2
Θ step in a powder diffraction pattern (Bish & Post 1993).
Quartz, orthoclase, plagioclase (albite), biotite and kaolinite
were detected in the P-XRD pattern. The Rietveld refinement
was performed using the FullProf program (Rodríguez-Car-
vajal 2006). The structure models used in the refinement were
as follows: quartz (Le Page & Donnay 1976), orthoclase
(Prince et al. 1973), albite (Ferguson et al. 1958), biotite
(Brigatti & Davoli 1990) and kaolinite (Bish & Von Dreele
1989). The structural model of oligoclase instead of albite was
also tested in Rietveld refinements, however the differences
in the profile agreement factors between both Rietveld fits
were insignificant. The differences in the estimated modal
composition were below 0.1 wt. %. The background was de-
termined by linear interpolation between consecutive break-
points in the powder pattern. The pseudo-Voigt function was
employed to generate the line shape of the diffraction peaks.
The refinement involved the scale factor for each phase,
unit-cell parameters for each phase except albite, peak-width
parameters for each phase, two asymmetry parameters for
biotite and 2
Θ zero error. Atomic coordinates, overall isotro-
pic displacement factors, and site occupancy parameters
were fixed during the refinement for all phases. The unit-cell
parameters of albite were also fixed, since their refinement
caused divergence of the fit. The March-Dollase correction
for preferred orientation was applied. The [001] direction
was used for biotite, kaolinite, feldspar and plagioclase. The
refinement converged to the values of the profile agreement
factors R
p
= 8.9 % and R
wp
= 11.4 %.
Principles of the least-squares calculations
The least-squares method is employed to solve an overde-
termined set of linear algebraic equations, which occur
where there are more independent equations than variables
(Bryan et al. 1969). Given a matrix A and a vector y, we
want to know the vector x that fulfils.
y = Ax
1
Our estimate of the vector x should also be chosen so that the
computed and real elements of the vector y differ as little as
possible.
y’= Ax’
2
The squares of these differences are commonly minimized
(so-called least-squares method: Bryan et al. 1969; Albarède
1995):
R
2
= y’—y
2
3
R
2
= min
4
and the sum of squared residuals is taken as the measure of
the goodness of fit. Albarède (1995) discussed in a detail all
the necessary mathematical apparatus that is behind the solu-
tion. Moreover, Janoušek & Moyen (in print) provide exam-
ples of various types of calculations (including normative
calculations using real mineral compositions by the least-
squares method) and their implementation into the freeware
R language.
In geochemistry we very often examine variables that sum
up, for instance, to a unity or 100 %. The least-squares method
generally does not produce normalized solutions, in the form
of vectors whose components would sum up to a given num-
ber. We can define the Lagrange multiplier
λ as (Albarède
(1995):
J
A
A
J
x
J
T
T
T
1
)
(
1
−
−
=
λ
5
where x stands for an ordinary least-squares solution and J is
a vector, all elements of which are equal to 1. The con-
strained least-squares solution is than given by:
6
x
^
= x +
λ(A
T
A)
—1
J
iii
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA, 2014, 65, 4
Electronic supplement – Appendices 1—5
References
Albarède F. 1995: Introduction to geochemical modeling. Cam-
bridge University Press, Cambridge, 1—543.
Bish D.L. & Post J.E. 1993: Quantitative mineralogical analysis
using the Rietveld full-pattern fitting method. Amer. Mineralo-
gist 78, 932—940.
Bish D.L. & Von Dreele R.B. 1989: Rietveld refinement of non-
hydrogen atomic positions in kaolinite. Clays and Clay Miner.
37, 4, 289—296.
Brigatti M.F. & Davoli P. 1990: Crystal-structure refinements of
1 M plutonic biotites. Amer. Mineralogist 75, 305—313.
Bryan W.B., Finger L.W. & Chayes F. 1969: Estimating propor-
tions in petrographic mixing equations by least-squares approxi-
mation. Science 163, 926—927.
Dempírová L. 2010: Evaluation of SiO
2
, Na
2
O, MgO, K
2
O determi-
nations in silicate samples by z-score obtained from nineteen
interlaboratory tests. [Zhodnocení stanovení SiO
2
, Na
2
O, MgO
a K
2
O v silikátových vzorcích pomocí z-skóre získaných z de-
vatenácti mezilaboratorních porovnávání.] Zpr. Geol. Výzk.
v roce 2009, 323—326 (in Czech with English summary).
Ferguson R.B., Traill R.J. & Taylor W.H. 1958: The crystal struc-
tures of low-temperature and high-temperature albites. Acta
Crystallogr. 11, 331—348.
Gabriel A. & Cox E.P. 1929: A staining method for the quantitative
determination of certain rock minerals. Amer. Mineralogist 14,
290—292.
Janoušek V. & Moyen J.F. (in print): Mass balance modelling of
magmatic processes in GCDkit. In: Kumar S. & Singh R.N.
(Eds.): Modelling of magmatic and allied processes. Soc.
Earth Sci., Ser. 83, Springer, Berlin, 225—238.
Doi: 10.1007/978-3-319-06471-0_11
Janoušek V., Farrow C.M. & Erban V. 2006: Interpretation of whole-
rock geochemical data in igneous geochemistry: introducing
Geochemical Data Toolkit (GCDkit). J. Petrology 47, 1255—1259.
Kretz R. 1983: Symbols for rock-forming minerals. Amer. Mineralo-
gist 68, 277—279.
Le Page Y. & Donnay G. 1976: Refinement of the crystal structure
of low-quartz. Acta Crystallogr., Sect. B 32, 2456—2459.
Merlet C. 1992: Accurate description of surface ionization in elec-
tron probe microanalysis: an improved formulation. X-RAY
Spectrom. 21, 229—238.
Pearce N.J.G., Perkins W.T., Westgate J.A., Gorton M.P., Jackson
S.E., Neal C.R. & Chenery S.P. 1997: A compilation of new
and published major and trace element data for NIST SRM 610
and NIST SRM 612 glass reference materials. Geostand.
Newsl. 21, 115—144.
Prince E., Donnay G. & Martin R.F. 1973: Neutron diffraction
refinement of an ordered orthoclase structure. Amer. Mineralo-
gist 58, 500—507.
Rodríguez-Carvajal J. 2006: Full Prof .2k Rietveld profile matching
& integrated intensities refinement of X-ray and/or neutron data
(powder and/or single-crystal). Laboratoire Léon Brillouin,
Centre d’Etudes de Saclay, Gif-sur-Yvette Cedex, France.
Strnad L., Mihaljevič M. & Šebek O. 2005: Laser ablation and solu-
tion ICP-MS determination of rare earth elements in USGS
BIR-1G, BHVO-2G and BCR-2G glass reference materials.
Geost. Geoanal. Res. 29, 303—314.
Young R.A. 2000: The Rietveld method. Oxford University Press,
Oxford, 1—312.
iv
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA, 2014, 65, 4
JANOUŠEK et al.: Distribution of elements among minerals of a single granite sample
Appendix 2
Electron microprobe analyses.
a – Average electron-microprobe data used in mass-balance calculations (wt. %).
SiO
2
TiO
2
Al
2
O
3
FeOt MnO MgO CaO Na
2
O K
2
O P
2
O
5
WR
70.34
0.36
14.64
1.5
0.024
1.14
1.36
3.71
5.55
0.15
Qtz
99.86
n.d.
0.02
n.d.
n.d. n.d. n.d. n.d. n.d. n.d.
Or
64.16
n.d.
18.47
0.03
0.034
n.d.
0.01
0.96
15.61
0.01
Pl
62.97
n.d.
20.07
n.d.
n.d.
0.01
3.95
9.2
0.44
0.02
Bt
37.5
3.33
14.8
17.26
0.24
13.35
0.04
0.14
9.47
n.d.
Ms
49.38 0.27
31.26 3.72 0.05 0.98 0.12 0.27 9.79 n.d.
Kln
45.59 0.91
39.84 0.02 0.01 n.d. n.d. n.d. n.d. 0.03
Ilm
0.01
50.32
n.d.
42.71
3.091
n.d.
n.d.
n.d.
n.d.
0.01
Ttn
30.49
35.77
2.79
1.11
0.092
n.d.
27.58
n.d.
n.d.
0.01
Ap
0.04
0.01
n.d.
0.03
0.161
n.d.
55.98
n.d.
n.d.
42.68
Mgt
n.d.
n.d.
n.d.
n.d.
0.233
n.d. n.d. n.d. n.d. n.d.
Mnz
2.42
n.d.
n.d.
0.04
n.d.
n.d.
0.37
n.d.
n.d.
25.61
Zrn
31.1
n.d.
0.29
0.44
0.006
n.d.
0.14
n.d.
n.d.
n.d.
Rt
0.28
97.35
n.d.
1.59
0.031
n.d.
0.08
n.d.
n.d.
n.d.
n.d. — not detected.
b – Typical analyses for the main and accessory minerals (wt. % and apfu).
Feldspars
Biotite
Magnetite
ID Plg15T
Plg19T
16T
ID
Bt14
ID
Mgt1
SiO
2
63.75
66.44
64.26
SiO
2
37.50
SiO
2
0.03
TiO
2
n.d.
0.01
n.d.
TiO
2
3.33
TiO
2
0.01
Al
2
O
3
22.00
20.33
18.29
Al
2
O
3
14.80
Al
2
O
3
0.04
FeOt
n.d.
n.d.
0.02
FeOt
16.95
FeOt
95.55
MnO
0.02
n.d.
n.d.
MnO
0.24
MnO
1.17
MgO
0.009
n.d.
0.01
MgO
13.35
MgO
n.d.
CaO
3.59
1.72
0.03
CaO
0.04
CaO
0.01
Na
2
O
9.72
10.84
1.30
Na
2
O
0.14
Na
2
O n.d.
K
2
O
0.38
0.21
15.48
K
2
O
9.47
K
2
O n.d.
P
2
O
5
0.013 0.009 0.008
H
2
O+
2.44
P
2
O
5
n.d.
Total
99.46
99.61
99.39
F
1.66
ThO
2
n.d.
Si apfu
2.834
2.930
2.987
Total
99.92
U
2
O
3
n.d.
Ti
0.000
0.000
0.000 Si
apfu
2.812 Y
2
O
3
n.d.
Al
1.153
1.057
1.002 Al(IV)
1.188 TR
2
O
3
n.d.
Fe
0.000 0.000 0.001
Σ
4.000
Total
96.81
Mn
0.001
0.000
0.000 Al(VI)
0.121 Si
apfu
0.004
Mg
0.001
0.000
0.001 Ti
0.188 Ti
0.001
Ca
0.171
0.081
0.002 Fe
1.063 Al
0.005
Na
0.838
0.927
0.117 Mn
0.015 Fe
2+
0.964
K
0.021
0.012
0.918 Mg
1.492 Fe
3+
1.995
P
0.000 0.000 0.000
Σ 2.879
Mn
0.037
O
8.000
8.000
8.000 Ca
0.003 Mg
0.000
An mol. % 16.6
8.0
0.1
Na
0.020
Ca
0.000
Ab
81.3
90.9
8.5
K
0.906
Na
0.000
Or
2.1
1.2
91.4
Σ 0.930
K 0.000
F 0.394
P 0.000
OH* 1.606
Th 0.000
XMg
0.58
U
0.000
XFe
0.42
Y
0.000
n.d. — not detected, n.a. — not analysed.
v
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA, 2014, 65, 4
Electronic supplement – Appendices 1—5
Zircon
Monazite
Rutile
Titanite
ID Zrn5
Zrn7
Zrn10
Mnz8
Mnz9
Rt1
Ttn4
SiO
2
32.08
31.10
32.27
1.89
2.42
0.51
30.30
TiO
2
0.02
n.d.
n.d.
0.01
0.00
98.82
36.32
Al
2
O
3
0.06
0.29
n.d.
n.d.
n.d.
0.01
2.44
P
2
O
5
n.d.
n.d.
n.d.
26.61
25.61
0.00
0.02
FeO
0.14
0.44
0.05
0.00
0.04
0.46
n.a.
Fe
2
O
3
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
1.08
MnO
n.d.
0.01
n.d.
0.12
n.d.
n.d.
0.11
CaO
0.02
0.14
n.d.
0.41
0.37
0.10
27.23
Y
2
O
3
n.d.
0.39
0.11
0.55
0.40
0.06
0.36
ZrO
2
63.28
61.21
62.89
n.a.
n.a.
n.a.
n.a.
HfO
2
1.87
1.89
1.30
n.a.
n.a.
n.a.
n.a.
Ce
2
O
3
n.d.
n.d.
0.04
32.74
32.78
0.14
0.50
La
2
O
3
0.13
0.07
n.d.
15.07
14.51
n.d.
0.07
Nd
2
O
3
n.d.
n.d.
0.08
9.35
8.66
n.d.
0.60
Pr
2
O
3
0.12
0.10
n.d.
2.87
2.86
0.04
0.23
Sm
2
O
3
0.00
0.09
0.04
1.07
1.18
0.00
0.02
Eu
2
O
3
0.01
0.02
0.01
n.d.
n.d.
0.05
n.d.
Gd
2
O
3
0.02
0.04
0.01
0.48
0.37
n.d.
0.15
Tb
2
O
3
n.d.
n.d.
n.d.
0.08
0.03
n.d.
0.08
Dy
2
O
3
0.01
0.01
0.04
0.04
n.d.
0.01
0.08
Ho
2
O
3
n.d.
0.01
0.03
0.01
0.01
0.04
0.02
Er
2
O
3
0.00
0.02
0.00
0.05
0.02
0.00
0.03
Tm
2
O
3
n.d.
n.d.
n.d.
0.08
0.09
0.02
0.02
Yb
2
O
3
n.d.
n.d.
0.04
n.d.
0.03
0.02
0.01
Lu
2
O
3
n.d.
n.d.
n.d.
n.d.
n.d.
0.00
n.d.
ThO
2
0.13
0.17
0.03
8.18
10.42
0.02
0.02
U
2
O
3
0.18
0.20
0.60
0.50
0.37
0.03
0.01
Total
98.05
96.19
97.54
100.110 100.144 100.322 99.66
Si apfu
1.007 0.998 1.015 0.077 0.099 0.007 1.001
Ti
0.000 0.000 0.000 0.000 0.000 0.988 0.903
Al
0.002 0.011 0.000 0.000 0.000 0.000 0.095
P
0.000 0.000 0.000 0.912 0.887 0.000 0.001
Fe
2+
0.004 0.012 0.001 0.000 0.001 0.005 n.a.
Fe
3+
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
0.027
Mn
0.000 0.000 0.000 0.004 0.000 0.000 0.003
Ca
0.001 0.005 0.000 0.018 0.016 0.001 0.964
Y
0.000 0.007 0.002 0.012 0.009 0.000 0.006
Zr
0.968 0.958 0.965 n.a. n.a. n.a. n.a.
Hf
0.017 0.017 0.012 n.a. n.a. n.a. n.a.
Ce
0.000 0.000 0.000 0.485 0.491 0.001 0.006
La
0.001 0.001 0.000 0.225 0.219 0.000 0.001
Nd
0.000 0.000 0.001 0.135 0.127 0.000 0.007
Pr
0.001 0.001 0.000 0.042 0.043 0.000 0.003
Sm
0.000 0.001 0.000 0.015 0.017 0.000 0.000
Eu
0.000 0.000 0.000 0.000 0.000 0.000 0.000
Gd
0.000 0.000 0.000 0.006 0.005 0.000 0.002
Tb
0.000 0.000 0.000 0.001 0.000 0.000 0.001
Dy
0.000 0.000 0.000 0.000 0.000 0.000 0.001
Ho
0.000 0.000 0.000 0.000 0.000 0.000 0.000
Er
0.000 0.000 0.000 0.001 0.000 0.000 0.000
Tm
0.000 0.000 0.000 0.001 0.001 0.000 0.000
Yb
0.000 0.000 0.000 0.000 0.000 0.000 0.000
Lu
0.000 0.000 0.000 0.000 0.000 0.000 0.000
Th
0.001 0.001 0.000 0.075 0.097 0.000 0.000
U
0.001 0.001 0.004 0.005 0.003 0.000 0.000
A
0.995 1.005 0.988 1.020 1.024
B
1.009 1.009 1.015 0.988 0.986
Sum cat.
2.004 2.014 2.003 2.015 2.016 1.004 3.021
vi
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA, 2014, 65, 4
JANOUŠEK et al.: Distribution of elements among minerals of a single granite sample
Appendix 3: Laser-ablation ICP-MS data for the main rock forming minerals (ppm).
LA ICP-MS
Quartz
DL (LA)
ppm
q1 q2 q3 q4 q1 q2 q3 q4 q1 Avg
SD
0.01
Be
0.08 0.11
0.17 0.03 0.00 0.05 0.00 0.00
0.06
0.06
0.35
Mn
0.12 0.12 0.16 0.10 0.09 0.21 0.22 0.20 0.20
0.16
0.05
0.095
Co
0.01 0.01 0.00 0.01 0.01 0.01 0.03 0.00 0.01
0.01
0.01
1.15
Ni
1.97 1.77 0.94 1.27 2.29 2.17 3.02 2.93 3.12
2.16
0.77
0.48
Zn
0.10 0.22 0.26 0.19 0.11 0.64 0.69 0.69 0.65
0.39
0.26
0.26
Rb
0.03 0.05 0.05 0.05 0.35 0.08 0.07 0.05 0.09
0.09
0.10
0.085
Sr
0.10 0.16 0.09 0.06 0.12 0.13 0.07 0.13 0.12
0.11
0.03
0.045
Y
0.00 0.00 0.01 0.05 0.00 0.00 0.00 0.00 0.00
0.01
0.02
0.25
Zr
0.05 0.03 1.31
0.09 0.14 0.06 0.04 0.03
0.22
0.44
0.05
Nb 0.01
0.00
0.00
0.00
0.04
0.01
0.01
0.01
0.23
Cd
0.16 0.08 0.02 0.01 0.11 0.05 0.01 0.01 0.03
0.05
0.05
0.085
Ba
0.19 0.28 0.12 0.13 0.80 0.39 0.36 0.38 0.38
0.34
0.20
0.020
La 0.001 0.009 0.003 0.007 0.001 0.01 0.00 0.00 0.00
0.004
0.00
0.028
Ce 0.008 0.013 0.007 0.021 0.011 0.03 0.00 0.01 0.00
0.012
0.01
0.008
Pr 0.000 0.000 0.000 0.003 0.000 0.00 0.00 0.00 0.00
0.001
0.00
0.083
Nd
0.000 0.000 0.007 0.004 0.003 0.01 0.00 0.00 0.00
0.003
0.00
0.097
Sm
0.000 0.000 0.000 0.000 0.000 0.00 0.00 0.00 0.00
0.000
0.00
0.008
Eu 0.000 0.000 0.000 0.000 0.000 0.00 0.00 0.00 0.00
0.000
0.00
0.032
Gd
0.000 0.000 0.000 0.000 0.000 0.00 0.00 0.00 0.00
0.000
0.00
0.009
Tb 0.000 0.000 0.000 0.000 0.000 0.00 0.00 0.00 0.00
0.000
0.00
0.030
Dy
0.000 0.000 0.000 0.001 0.000 0.00 0.00 0.00 0.00
0.000
0.00
0.004
Ho
0.000 0.000 0.001 0.001 0.000 0.00 0.00 0.00 0.00
0.000
0.00
0.012
Er 0.000 0.000 0.003 0.005 0.000 0.00 0.00 0.00 0.00
0.001
0.00
0.002
Tm
0.000 0.000 0.000 0.001 0.000 0.00 0.00 0.00 0.00
0.000
0.00
0.022
Yb
0.000 0.000 0.000 0.002 0.000 0.00 0.00 0.00 0.00
0.000
0.00
0.003
Lu 0.000 0.000 0.000 0.000 0.000 0.00 0.00 0.00 0.00
0.000
0.00
0.03
Hf
0.00 0.00 0.08
0.00 0.01 0.00 0.00 0.00
0.01
0.03
0.05
Ta 0.01
0.01 0.01
0.01
0.02 0.02 0.02
0.01
0.01
0.21
Pb
0.08 0.07 0.07 0.08 0.11 0.15 0.13 0.12 0.11
0.10
0.03
0.02
Th
0.66 1.01 0.11 0.04 0.35 0.97 0.31
1.13
0.57
0.43
0.011
U
0.01 0.01 0.04 0.11 0.01 0.02 0.02 0.02 0.02
0.03
0.03
DL = detection limit.
K-feldspar
k1 k2 k3 k4 Avg
SD
Be
0.58 0.85 2.15 1.77 1.34
0.74
Mn
2.15 2.44 1.33 1.72 1.91
0.49
Co
0.04 0.07 0.03 0.01 0.04
0.02
Ni
0.60 0.65 0.54 0.19 0.49
0.21
Zn
1.28 1.18 0.97 0.29 0.93
0.45
Rb
323.07 328.94 342.18 337.47 332.91
8.55
Sr
529.10 458.50 390.70 325.38 425.92
87.67
Y
0.07 0.56 0.14 0.10 0.21
0.23
Zr
2.05 5.52 5.23 1.16 3.49
2.21
Nb
0.10 0.13 0.04 0.05 0.08
0.04
Cd
0.06 0.04 0.02 0.04 0.04
0.02
Ba
4104.54 3186.51 1552.54 1134.56 2494.54 1391.38
La
1.76 1.77 1.58 1.49 1.65
0.14
Ce
1.41 2.07 2.61 1.26 1.84
0.62
Pr
0.07 0.17 0.12 0.12 0.12
0.04
Nd
0.14 0.72 0.46 0.44 0.44
0.24
Sm
0.02 0.17 0.08 0.06 0.08
0.06
Eu
1.21 1.07 0.93 0.75 0.99
0.19
Gd
0.02 0.14 0.05 0.04 0.06
0.05
Tb
0.00 0.02 0.00 0.00 0.01
0.01
Dy
0.01 0.09 0.02 0.02 0.03
0.04
Ho
0.00 0.01 0.00 0.00 0.00
0.01
Er
0.00 0.04 0.02 0.00 0.02
0.02
Tm
0.00 0.01 0.00 0.00 0.00
0.00
Yb
0.00 0.03 0.02 0.01 0.01
0.01
Lu
0.00 0.01 0.00 0.00 0.00
0.00
Hf
0.09 0.13 0.14 0.03 0.10
0.05
Ta
0.04 0.04 0.03 0.03 0.04
0.01
Pb
121.45 107.57 121.25 115.96 116.56
6.51
Th
0.14 0.16 0.23 0.07 0.15
0.07
U
0.10 0.28 0.25 0.83
0.36
0.32
vii
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA, 2014, 65, 4
Electronic supplement – Appendices 1—5
Plagioclase
Excluded from calculation
p1 p2 Avg
SD p3 p4 p5
Be
19.66 23.66 21.66
2.83 17.25 18.36 22.67
Mn
77.23 33.80 55.52
30.71 47.62 40.88 56.09
Co
2.29 0.92 1.61
0.97 1.28 0.77 1.54
Ni
2.90 1.35 2.12
1.10 3.29 2.22 2.33
Zn
18.12 7.98 13.05
7.17 10.68 6.73 12.87
Rb
268.20 46.29 157.25
156.91 154.30 101.40 90.50
Sr
499.30 500.80 500.05
1.06 524.90 606.90 512.70
Y
0.13 0.16 0.15
0.02 0.66 0.57 5.25
Zr
16.53 3.90 10.21
8.93 14.43 6.36
293.40
Nb
9.51 1.60 5.55
5.59 0.24 1.77 1.47
Cd
0.07 0.04 0.06
0.02 0.10 0.03 0.11
Ba
213.10 57.97 135.54
109.69 104.40 111.10 72.53
La
20.43 27.00 23.72
4.65
213.50
326.80
54.60 inclusions of Mnz, Ap?
Ce
36.29 40.69 38.49
3.11
283.30
356.50
129.40
Pr
3.94 5.47 4.71
1.09
40.92
51.38
10.52
Nd
11.22 13.24 12.23
1.43
125.50
159.00
34.09
Sm
0.57 0.88 0.73
0.22
6.80
8.76
3.22
Eu
0.60 0.84 0.72
0.17
1.12
1.32
1.05
Gd
0.59 0.57 0.58
0.01
4.45
5.88
2.33
Tb
0.02 0.03 0.024
0.01 0.18 0.25 0.22
Dy
0.03 0.05 0.037
0.01 0.27 0.28 1.05
Ho
0.00 0.00 0.004
0.00 0.03 0.03 0.21
Er
0.03 0.03 0.027
0.00 0.17 0.25 0.61
Tm
0.00 0.00 0.001
0.00 0.01 0.01 0.10
Yb
0.01 0.00 0.007
0.01 0.03 0.03 0.80
Lu
0.00 0.00 0.001
0.00 0.00 0.00 0.12
Hf
0.95 0.20 0.58
0.53 0.63 0.19 7.68
Ta
0.09 0.02 0.05
0.05 0.02 0.18 0.03
Pb
62.66 58.04 60.35
3.27 78.75 78.32 76.86
Th
4.90 1.34 3.12
2.52 14.98 8.14 21.04
U
1.97 1.35
1.66
0.44
10.26 10.11 8.87
Biotite
B1 B2 B3 Avg
SD
Be
4.481 5.162 4.468 4.70
0.40
Mn
1357.579 1380.579 1358.579 1365.58
13.00
Co
24.955 24.765 23.835 24.52
0.60
Ni
109.12 119.72 117.52 115.45
5.59
Zn
259.247 263.847 256.847 259.98
3.56
Rb
828.107 756.107 633.507 739.24
98.39
Sr
1.395 1.844 1.951 1.73
0.30
Y
0.053 4.416 4.036 2.84
2.42
Zr
0.915 2.243 0.567 1.24
0.88
Nb
91.419 95.239
124.999 103.89
18.38
Cd
0.079 0.096 0.005 0.06
0.05
Ba
1114.197 1186.197 1458.197 1252.86
181.43
La
0.031 0.801 0.901 0.58
0.48
Ce
0.03 2.372 2.756 1.72
1.48
Pr
0.001 0.623 0.568 0.40
0.34
Nd
0.003 3.189 3.232 2.14
1.85
Sm
0.001 1.139 1.017 0.72
0.62
Eu
0.054 0.103 0.132 0.10
0.04
Gd
0 0.915 0.918 0.61
0.53
Tb
0 0.139 0.124 0.09
0.08
Dy
0.002 0.733 0.695 0.48
0.41
Ho
0.001 0.134 0.141 0.09
0.08
Er
0.004 0.281 0.282 0.19
0.16
Tm
0 0.037 0.034 0.02
0.02
Yb
0.001 0.243 0.186 0.14
0.13
Lu
0.001 0.028 0.025 0.02
0.01
Hf
0.115 0.141 0.028 0.09
0.06
Ta
6.912 9.562
14.066 10.18
3.62
Pb
2.832 3.554 3.959 3.45
0.57
Th
0.076 0.709 0.534 0.44
0.33
U
0.027 0.133 0.086
0.08
0.05
viii
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA, 2014, 65, 4
JANOUŠEK et al.: Distribution of elements among minerals of a single granite sample
Appendix 4
1
–
Mineral
separates
dissolved
and
analyzed
in
solution
by
ICP-MS
technique.
See
text
Appendix
1
for
analytical
details.
2
–
Averaged
single-spot
analyses
as
determined
by
the
laser-ablati
on
ICP-MS.
3 –
Electron
microprobe;
too
high
to
be
determined
by
the
ICP-MS.
In
grey
are
shown
analyses
not
taken
into
consideration.
See
text
for
discussion.
Trace-element composition of the
whole-rock, individual rock-fo
rming and accessory minerals determined
by the ICP-MS technique
(ppm).
ix
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA, 2014, 65, 4
Electronic supplement – Appendices 1—5
Appendix 5
Typical electron-microprobe analyses of the zircon-like
ABO
4
-type phase (wt. % and apfu).
1 2 3 4 5 6
SiO
2
16.055 14.810 14.704 15.398 14.739 16.300
TiO
2
0.090 n.d. 0.087 n.d. 0.058 0.034
Al
2
O
3
0.667 0.733 0.650 0.642 0.878 0.704
FeO
1.220 2.241 4.426 0.732 6.781 1.002
MnO
0.026 0.132 0.186 0.057 0.439 0.058
MgO
0.101 0.202 0.174 0.186 0.236 0.128
CaO
4.765 4.875 4.612 4.795 4.574 4.641
K
2
O
0.137 0.091 0.098 0.086 0.095 0.218
P
2
O
5
6.582 6.699 6.684 6.888 6.353 6.407
SO
2
0.149 0.241 0.202 0.188 0.198 0.172
Sc
2
O
3
0.021 0.014 0.030 0.003 0.012 0.009
As
2
O
3
0.479 0.332 0.285 0.264 0.362 0.483
ZrO
2
27.758 24.869 23.731 26.561 25.640 27.874
HfO
2
0.570 0.569 0.596 0.617 0.571 0.692
Nb
2
O
5
n.d. 0.014
0.003
0.017
n.d. n.d.
SnO
0.010 n.d. 0.029 n.d. 0.036 0.005
PbO
0.214 0.265 0.278 0.296 0.192 0.208
UO
2
1.278 1.106 1.179 1.208 1.240 1.033
ThO
2
31.947 32.353 31.751 35.250 29.028 31.265
Ce
2
O
3
3.236 2.377 2.104 2.238 2.058 3.378
SmO
0.122 0.172 0.011 0.061 0.213 0.206
Gd
2
O
3
n.d. n.d. 0.169
0.210
n.d. 0.041
Dy
2
O
3
0.170 0.119 0.042 0.067 0.371 0.181
Er
2
O
3
n.d. n.d. n.d. n.d. n.d. n.d.
Yb
2
O
3
0.077 0.099 n.d. n.d. n.d. n.d.
Y
2
O
3
0.298 0.215 0.207 0.137 0.219 0.287
F
0.606 0.358 0.345 0.651 0.163 0.681
Total
96.578 92.886 92.583 96.552 94.456 96.007
Si
0.647 0.624 0.622 0.631 0.606 0.658
Ti
0.003 0.000 0.003 0.000 0.002 0.001
Al
0.032 0.036 0.032 0.031 0.043 0.034
Fe
0.041 0.079 0.156 0.025 0.233 0.034
Mn
0.001 0.005 0.007 0.002 0.015 0.002
Mg
0.006 0.013 0.011 0.011 0.014 0.008
Ca
0.206 0.220 0.209 0.210 0.201 0.201
K
0.007 0.005 0.005 0.004 0.005 0.011
P
0.225 0.239 0.239 0.239 0.221 0.219
S
0.006 0.010 0.008 0.007 0.008 0.007
Sc
0.001 0.001 0.001 0.000 0.000 0.000
As
0.012 0.008 0.007 0.007 0.009 0.012
Zr
0.546 0.511 0.489 0.530 0.514 0.549
Hf
0.007 0.007 0.007 0.007 0.007 0.008
Nb
0.000 0.000 0.000 0.000 0.000 0.000
Sn
0.000 0.000 0.001 0.000 0.001 0.000
Pb
0.002 0.003 0.003 0.003 0.002 0.002
U
0.011 0.010 0.011 0.011 0.011 0.009
Th
0.293 0.310 0.305 0.329 0.271 0.287
Ce
0.048 0.037 0.033 0.034 0.031 0.050
Sm
0.002 0.003 0.000 0.001 0.003 0.003
Gd
0.000 0.000 0.002 0.003 0.000 0.001
Dy
0.002 0.002 0.001 0.001 0.005 0.002
Er
0.000 0.000 0.000 0.000 0.000 0.000
Yb
0.001 0.001 0.000 0.000 0.000 0.000
Y
0.006 0.005 0.005 0.003 0.005 0.006
F
0.077 0.048 0.046 0.084 0.021 0.087
site A
0.921 0.917 0.909 0.914 0.886 0.929
site B
1.183 1.210 1.249 1.175 1.321 1.175
The EPMA results are recalculated on the basis of 4O apfu.
n.d. – not detected.
Site B includes Si, Al, P, As and S, the site A comprises all other
elemets (except F).