REE-bearing minerals in Drava river sediments, Slovenia, 
and their potential origin
Minerali z redkimi zemljami v sedimentih reke Drave (Slovenija) ter njihov potencialni izvor
Aleš ŠOSTER1, Janez ZAVAŠNIK2, Mihael RAVNJAK3 & Uroš HERLEC1
1University of Ljubljana, Faculty of Natural Sciences and Engineering, Department of Geology  
Aškerčeva cesta 12, SI-1000 Ljubljana; e-mail: ales.soster@geo.ntf.uni-lj.si; uros.herlec@geo.ntf.uni-lj.si 
2Jožef Stefan Institute, Center for Electron Microscopy and Microanalysis,  Jamova cesta 39, SI-1000 Ljubljana; 
e-mail: janez.zavasnik@ijs.si
3Cesta na Roglo 21, SI-3214 Zreče; e-mail: mihael.ravnjak@gmail.com
Prejeto / Received 12. 10. 2017; Sprejeto / Accepted 6. 11. 2017; Objavljeno na spletu / Published online 22. 12. 2017
Key words: monazite, xenotime, microanalysis, REE, heavy minerals, Drava, Slovenija
Ključne besede: monacit, ksenotim, mikroanaliza, REE, težki minerali, Drava, Slovenia
Abstract
Monazite and xenotime are major ore minerals for rare earth elements (REE), Y, and actinides, a powerful 
tool for radiometric dating and indicators of the geological genesis of the parent rock. In this contribution, we 
present results of microanalysis of monazite and xenotime mineral grains found in heavy mineral sand fraction 
from Drava river. Investigated grains of monazite(-Ce) were found to be exhibiting varying concentrations of Th, 
indicating mixed hydrothermal (or greenschist-amphibolite grade) and igneous origin. Monazite was found to be 
isomorphically replaced by cheralite (CaTh(PO4)2) and subordinately by huttonite (ThSiO4), forming monazite-
cheralite and monazite-huttonite solid solution. Chemical composition and the degree of isomorphic replacement 
of monazite grains are indicating their source from rocks of Eclogite belt and Granatspitz massif in Tauern 
Window, Austria. On the other hand xenotime(-Y) shows consistent concentrations of Y while concentrations of 
associated heavy rare earth elements (HREE) vary, indicating the origin of mixed affinities.  
Izvleček 
Minerala monazit in ksenotim sta pomemben vir redkih zemelj, itrija in aktinidov. Lahko ju uporabimo tudi 
kot orodje za radiometrično datiranje (geokronologijo), ter kot indikator geneze matične kamnine. V prispevku 
predstavljamo rezultate mikroanalize zrn monacita in ksenotima zaznanih v frakciji težkih peskov reke Drave. 
Cerijev monacit ima spremenljive vsebnosti Th, kar kaže na njegov hidrotermalni izvor ali izvor iz metamorfnih 
(od faciesa zelenega skrilavca do amfibolitnega faciesa) in magmatskih kamnin. Rezultati mikroanalize monacita 
kažejo izomorfno nadomeščanje s ceralitom (CaTh(PO
4)2) in v manjši meri s huttonitom (ThSiO4), pri čemer tvorijo 
trdno raztopino. Kemijska sestava in stopnja izomorfnega nadomeščanja monacitnih zrn nakazujejo njihov izvor 
iz kamnin Eklogitnega pasu in masiva Granatspitz v Visokih Turah v Avstriji. V ksenotimu je najbolj zastopan 
element Y, ki ima v vseh zrnih enakomerne vsebnosti, medtem ko se vsebnosti težkih redkih zemelj (HREE) 
spreminjajo, kar kaže na izvor iz različnih geoloških okolij. 
© Author(s) 2017. CC Atribution 4.0 LicenseGEOLOGIJA 60/2, 257-266, Ljubljana 2017
https://doi.org/10.5474/geologija.2017.018
Introduction
Heavy mineral sand deposits or placers are 
waterborne deposits of minerals with high specific 
density, formed as a result of water flow dynamics, 
either by wave motion, long-term coastal currents 
or by unidirectional flow associated with rivers. 
The principal mechanism in the formation of 
heavy mineral deposits is a natural separation of 
low and high-density minerals in a manner where 
low-density minerals are removed by flowing 
medium thus concentrating high-density minerals 
(roBB, 2004). Placer deposits are known as a 
significant source of many industrially important 
minerals, i.e., diamond, gold, tantalite-columbite, 
cassiterite, rutile, ilmenite, uraninite, zircon and 
rare earth element (REE)-bearing minerals (roBB, 
2004; rao & Misra, 2009). Principal ore minerals 
mined for REE are monazite, bastnaesite, and 
xenotime (voncken, 2016). Monazite is a phosphate 
mineral of light rare earth elements (LREE) 
with generalized formula CePO4, which besides 
Ce contains variable proportions of La, Nd, and 
Pr. Based on predominant LREE incorporated 
in monazite a suffix is given to denote the most 
258 Aleš ŠOSTER, Janez ZAVAŠNIK, Mihael RAVNJAK & Uroš HERLEC
abundant auxiliary element (i.e., monazite(-Ce)) 
(voncken, 2016). Monazite occurs as an accessory 
mineral in peraluminous granites, syenitic and 
granitic pegmatites, quartz veins, carbonatites, 
migmatites and paragneisses (Förster, 1998a; 
Breiter, 2016). Monazite is considered to be a 
major host of LREE and smaller fractions of Y, 
heavy rare earth elements (HREE), and actinides 
(Bea, 1996). Therefore it presents a powerful 
tool for monazite-based radiometric dating 
(harrison et al., 1995; Zhu & o‘nions, 1999; 
schandl & gorton, 2004; grand‘hoMMe et al., 
2016). Xenotime, in contrast to monazite, beside 
Y contains also a significant amount of HREE, 
among which most commonly occurring are Dy, 
Yb, Er, and Gd while Tb, Ho, Tm, and Lu are 
occurring less often (Förster, 1998b; Švecová et al., 
2016; voncken 2016). Depending on its geological 
genetic environment, xenotime is considered to be 
a primary source of HREE and actinides and it is 
a common accessory mineral in many non-basic 
igneous rocks, most granitic rocks and granitic 
pegmatites where it accounts for significant 
fraction of Y and HREE of bulk rock composition 
(Bea, 1996; Förster, 1998b), while it is also present 
in migmatites and high-grade metamorphic rocks 
(Förster, 1998b). Minerals such as monazite 
and xenotime are considered to be of economic 
importance especially when enriched in actinides 
(i.e., U and Th) for their use in nuclear industry, 
however they may pose an environmental hazard 
due to increased natural radiation (alaM et al., 
1999; vassas et al., 2006; rao & Misra, 2009). 
Regarding actinides, monazite is known to more 
often contain Th, while xenotime is more likely 
to contain U and also smaller amounts of Th (van 
eMDen et al., 1997; kiM et al., 2009; deer et al., 
2013). This research aims to present the results 
of microanalysis performed on REE-bearing 
minerals from heavy mineral fraction of Drava 
river and to make an attempt to elucidate their 
potential provenance.
Geographical and geological background
Our case study area is situated near village 
Zlatoličje in NE Slovenia, along Drava river (Fig. 
1a). River Drava has its source in Innichen (San 
Candido) in the Puster Valley of South Tyrol, 
Italy, and flows eastwards through East Tyrol and 
Carinthia regions in Austria into region Styria 
in Slovenia and further along Croatia-Hungary 
border, where it joins the Danube near the city 
of Osijek (tocKner et al., 2009). Along its path, 
it has a number of tributaries with their sources 
in Hohe Tauern, i.e., Isel, with its source beneath 
Grossvenediger (3674 m) joining Drava near 
Lienz and Möll with its source near Heiligenblutt 
below Großglockner (3798 m). Knowledge of 
drainage area and therefrom eroded rocks, of the 
Fig. 1. Location of the study area (a) with detailed geological setting (b) (modified after Žnidarčič & Mioč, 1987).
259REE-bearing minerals in Drava river sediments, Slovenia, and their potential origin
river and its tributaries combined with in-detail 
chemical analyses of minerals, both from parent 
rock and sediments, is crucial for determining 
the provenance of the grains downstream. The 
region of Hohe Tauern, in detail discussed by 
schiMD et al. (2013) and scharF et al. (2013), is 
considered to be the source of gold found in fluvial 
sediments of river Drava (BerManec et al., 2014), 
therefore broader area of Hohe Tauern could also 
be considered as a source of minerals in heavy 
mineral sands. 
In the area of our case study, fluvial sediments of 
river Drava are deposited onto Cenozoic basement 
mainly represented by Upper Miocene (Tortonian 
to Messinian) marlstones and sandstones 
paleogeographically a part of Central Paratethys 
and Pliocene sand, gravel and conglomerate with 
intermediate coal beds deposited in the Lake 
Pannon realm (Fig. 1b; Pleničar et al., 2009). The 
Quaternary period is represented by sediments 
of Drava reflecting changes in flow dynamics 
and migrating river channels, thus most common 
sediments are granulometrically heterogeneous 
gravel and sand of the river channel, floodplains 
sediments and organic clay of oxbow lake facies 
(Fig. 1b; Mioč & Žnidarčič, 1989; Pleničar et al., 
2009).
Materials and methods
Field sampling
Sampling was conducted in the river bank 
gravel, in the river channel and in the distal 
parts of river bars to ensure a high yield of the 
heavy mineral fraction. Three bulk samples of 
initially 2 kg each, were washed to remove any 
plant detritus and clay particles. Bulk samples 
underwent manual gravitational separation, 
which was achieved by panning to obtain a rela-
tively pure heavy mineral concentrate. Particles 
larger than 5 mm were removed manually. Heavy 
mineral concentrate was further processed by 
heavy liquid separation in which we used lithium 
meta-tungstate heavy fluid (LMT) with density 
[σ = 2.95 g/cm3] to further remove light miner-
al grains. The remaining mineral grains were, 
for the needs of optical and electron microscopy, 
vacuum-cast into two component epoxy resin. 
Grinding and polishing were performed stepwise 
using silicon carbide 240 and 1200 grit for grind-
ing and 9 µm, 3 µm and 1 µm diamond paste for 
polishing to high gloss finish.
X-ray diffraction analysis
Identification and quantification of major 
minerals forming bulk fluvial sediment and 
heavy mineral sands were performed using X-ray 
powder diffraction analysis (XRD, PW 3830/40, 
PANalytical B.V., Almelo, the Netherlands). Pow-
dered samples were inserted into a holder with 
divergence slit for incident X-rays at 0.04° and 
0.2 mm wide slit for diffraction path. Samples 
were analyzed in the angular span between 3° - 
70° 2θ with step 3°2θ /60 s. Source of X-rays was 
copper anode with the generator set at 30 mA 
and 40 kV, producing radiation with λ=1.5406 
Å. Quantification of entities was achieved using 
Rietveld refinement (goodness of fit: 2.085, R-ex-
pected: 13.95, R-profile: 19.97), individual FWHM 
function and Pseudo Voigt function. Identifica-
tion of mineral structure was performed using 
ICSD Database FIZ Karlsruhe (Pan ICSD 2.x).
Optical microscopy and scanning electron 
microscopy
Identification of minerals, observation of mor-
phology and grain size were performed by Zeiss 
Imager Z1.m polarizing optical microscope, ad-
ditionally equipped with in-lens calibrated dig-
ital camera for accurate scaling. Scanning elec-
tron microscopy (SEM) was performed by using 
JEOL JSM-5800 with thermionic source under 
high vacuum (5x10-6 bar). Preliminary miner-
al identification was made using backscattered 
electron imaging (BEI), where relative contrast 
in the image is directly related to mean atomic 
number of the investigated sample. Semi-quan-
titative chemical analysis of the mineral grains 
was performed using Oxford Instruments ener-
gy-dispersive X-ray spectrometer (EDS, mod. 
6841, Oxford Instruments Ltd.). The operating 
conditions for EDS analyses were 20 kV, 10 mm 
working distance and 0° tilt. Beam current was 
adjusted to yield a dead time of 20-30 %, while 
the live time was set at 60s. Data reduction was 
performed using atomic number-absorption-flu-
orescence matrix correction (ZAF). Prior to SEM 
observation samples were coated with few nm of 
amorphous carbon to improve the conductivity of 
the sample and prevent charging.
Results
Bulk fluvial sediment is uniformly composed 
of quartz, calcite, dolomite, chlorite, mica, 
plagioclase and traces of kaolinite and amphibole 
group minerals (Fig. 2a) as determined by XRD 
analysis. Bulk sediment samples with an initial 
260 Aleš ŠOSTER, Janez ZAVAŠNIK, Mihael RAVNJAK & Uroš HERLEC
mass of 2 kg each, yielded 46.2 – 79.7 g of the 
heavy mineral fraction after final treatment with 
LMT. Heavy mineral sands are present in the 
fraction between 0.06 and 0.5 mm corresponding 
with very fine sand to coarse sand. Prevailing 
minerals in the heavy mineral fraction are garnets 
accounting for 92 wt%, followed by hematite 
(6.7 wt%) and rutile (0.7 wt%) (Fig. 2b). Besides 
aforementioned minerals also diopside, epidote, 
ilmenite, hematite, magnetite, gold, pyrite, 
sphalerite, barite, chalcopyrite, molybdenite, 
monazite and xenotime were identified using 
optical or scanning electron microscopy, which 
overall account for less than 0.1 wt% of heavy 
mineral fraction, which is below detection limit 
of XRD analysis (Fig. 2b).
Fig. 2. Column diagram of the mineral composition of (a) bulk fluvial sediment and (b) heavy mineral fraction.
P2O5 La2O3 Ce2O3 Nd2O3 ThO2 P
* La* Ce* Nd* Th*
Mnz-1 27.44 23.56 35.16 10.07 3.76 0.953 0.356 0.527 0.147 0.035
Mnz-2 29.02 18.07 34.80 9.47 6.82 1.034 0.280 0.536 0 0.065
Mnz-3 30.36 19.08 33.52 9.34 6.44 1.016 0.278 0.485 0.132 0.058
Mnz-4 29.81 17.79 35.04 12.66 3.42 0.991 0.257 0.503 0.177 0.031
Mnz-5 32.05 18.01 33.42 11.72 3.92 1.032 0.252 0.465 0.159 0.034
Mnz-6 29.78 18.47 35.68 11.17 3.95 0.992 0.268 0.514 0.157 0.035
Mnz-7 29.04 19.50 34.96 11.51 3.71 0.977 0.285 0.508 0.163 0.034
Mnz-8 31.81 16.80 31.49 11.27 6.48 1.024 0.235 0.438 0.153 0.056
Mnz-9 28.32 17.61 33.29 11.50 7.82 0.966 0.262 0.491 0.165 0.072
Mnz-10 31.93 17.48 29.77 11.23 8.20 1.031 0.246 0.415 0.153 0.071
Mnz-11 30.27 18.62 33.95 10.82 4.47 0.997 0.267 0.483 0.150 0.040
Mnz-12 27.24 21.72 39.79 10.67 0 0.943 0.327 0.595 0.156 0
Mnz-13 29.24 18.31 34.16 13.09 4.32 0.983 0.268 0.496 0.186 0.039
Mnz-14 29.13 17.89 35.75 9.86 5.77 0.979 0.262 0.537 0.140 0.052
Mnz-15 28.46 19.36 36.54 11.72 3.12 0.968 0.287 0.496 0.168 0.029
Mnz-16 31.00 18.76 35.10 12.02 2.31 1.013 0.267 0.540 0.166 0.020
Mnz-17 29.26 17.02 37.33 12.29 2.77 0.980 0.248 0.488 0.173 0.025
Mnz-18 28.5 19.06 33.37 11.68 5.49 0.966 0.281 0.521 0.167 0.050
Mnz-19 28.92 18.68 35.29 10.94 4.95 0.968 0.278 0.522 0.158 0.046
Mnz-20 26.53 19.44 34.50 11.35 6.46 0.929 0.296 0.516 0.167 0.061
Mnz-21 31.26 14.42 36.59 12.53 4.43 1.020 0.205 0.452 0.172 0.039
Table 1. Relative concentrations of major oxides in monazite acquired by semi-quantitative EDS analysis given in wt%. Values 
marked by »*« represent formulae for 4 oxygen per formula unit (apfu).
261REE-bearing minerals in Drava river sediments, Slovenia, and their potential origin
Monazite
Monazite is present as individual liberated 
grains or included within other mineral grains 
(i. e. epidote, garnet (var. Andradite) or diopside) 
measuring from 5 to 200 µm. Individual grains of 
monazite are elongated to complex and angular 
to subrounded (Figs. 3a-c), while monazites 
included within other minerals are predominantly 
equant and circular, characteristic of mineral 
inclusions (Figs. 3d-f). Semi-quantitative EDS 
analysis revealed that monazite contains Ce2O3 
(28.49-33.97 wt%), La2O3 (12.29-20.09 wt%) and 
Nd2O3 (8.12-10.85 wt%) and smaller amounts of 
ThO2 (0-7.20 wt%). Major oxide composition of 
investigated monazites is given in Table 1. Higher 
ThO2 content was detected predominantly in 
individual liberated monazite grains, while in 
smaller grains included within aforementioned 
minerals, ThO2 content was generally lower. 
Xenotime
Grains of xenotime from heavy mineral 
fraction from Drava river are scarce and included 
within garnets (var. Spessartine) with individual 
grains measuring from 5 to 50 µm. Xenotime 
grains are elongated, sub-rounded or equant 
(Figs. 4a-c) which is characteristic of mineral 
inclusions. Semi-quantitative EDS analysis 
performed on xenotime grains showed that 
they contain Y2O3 (26.25-38.60 wt%) and Yb2O3 
(3.91-12.47 wt%) accompanied by smaller 
quantities of other HREE, i.e., Dy2O3 (2.52-
4.80 wt%) and Er2O3 (3.59-5.29 wt%). Major 
oxide composition of investigated xenotimes is 
presented in Table 2.
Fig. 3. SEM (BEI) images of monazite mineral grains (Mnz) in polished section of the heavy mineral fraction from Drava ri-
ver. Monazite is occurring as individual liberated grains (a-c) or as inclusions in other minerals, i.e., quartz and garnet (d-f).
P2O5 Y2O3 Dy2O3 Er2O3 Yb2O3 P
* Y* Dy* Er* Yb*
Xtm-1 41.37 49.02 5.50 4.10 0 1.106 0.726 0.056 0.041 0
Xtm-2 52.46 33.34 0 0 14.20 1.260 0.444 0 0 0.123
Xtm-3 52.68 41.10 5.64 3.81 6.78 1.164 0.621 0 0.039 0.067
Xtm-4 38.80 44.22 2.89 6.05 4.45 1.083 0.684 0.031 0.063 0.045
Xtm-5 39.93 47.04 4.20 4.35 4.48 1.092 0.713 0.044 0.044 0.044
Table 2. Relative concentrations of major oxides in xenotime acquired by semi-quantitative EDS analysis given in wt%. Values 
marked by »*« represent formulae for 4 oxygen per formula unit (apfu).
262 Aleš ŠOSTER, Janez ZAVAŠNIK, Mihael RAVNJAK & Uroš HERLEC
Discussion
Heavy mineral fractions of fluvial sediments 
from Drava river are predominantly composed 
of garnets, hematite, rutile and other minerals 
in traces (e.g., gold, monazite, xenotime, etc.) 
that account for less than 0.1 wt% of analyzed 
samples. Heavy mineral fraction is in size order 
of 0.06-0.5 mm corresponding with very fine to 
coarse sand and accounts for 2.3-4 wt% of bulk 
fluvial sediment.
Investigated monazites from Drava river show 
variable concentrations of REE and Th. Among 
REE, La, Ce, and Nd were identified in all ana-
lyzed samples, and their relationship can be 
summarized in the following manner based on 
their abundance in monazite: Ce > La > Nd. REE 
plotted in the ternary diagram show that major-
ity of the analyzed grains are shifted towards 
Ce-rich corner thus indicating cerium monazite 
– monazite(-Ce) variety (voncKen, 2016) (Fig. 5a).
Monazite shows variable content of ThO2 (0-
7.20 wt%) of which highest concentrations were 
measured in larger isolated monazite grains. 
Detrital monazite texture, mode of occurrence 
and its geochemistry have been found to point 
toward its genetic environment, therefore, en-
abling separation of igneous monazite from its 
hydrothermal counterpart (Burnotte et al., 1989; 
schanDl et al., 1994; wanG et al., 1994). schanDl 
& gorton (2004) demonstrated that hydrother-
mal monazite can be distinguished from igneous 
monazite by combining petrographic analyses 
with advanced electron microscopy techniques, 
performed on monazite mineral grains, and con-
cluded that hydrothermal monazite (or green-
schist-amphibolite metamorphic facies mona-
zite) generally contains lower concentrations of 
ThO2 (< 1 wt%), compared to igneous monazite, 
in which concentrations of ThO2 were found to be 
significantly higher (3 – >5 wt%). Based on these 
facts it can be concluded that monazite in heavy 
mineral fraction from Drava river originates 
from two or more genetically different sources, 
hydrothermal (< 1 wt% ThO2) and igneous (3 – 
>7.20 wt% ThO2) source. Furthermore, monazite 
grains with highest ThO2 concentrations are rel-
atively larger and show high mineral liberation 
factor, indicating longer transport distance than 
their lower ThO2 content counterparts.
In order to delineate possible intercrystal-
line substitution and intercrystalline partition-
ing behaviour of REE, Y, Th and U in monazite, 
composition of detrital monazites from Drava 
river was plotted in the diagram 4×(P+Y+REE) 
versus 4×(Si+Th+U), proposed by harlov et al. 
(2008), which presents the degree of cheralite 
(CaTh(PO4)2) or huttonite (ThSiO4) isomorphic 
substitution in monazite (Fig. 5b). Majority of the 
investigated detrital monazites from Drava riv-
er show slight or significant monazite-cheralite 
isomorphic substitution; some studied specimens 
even plotted above monazite-cheralite substitu-
tion line indicating the transition of monazite 
to cheralite. Only a small number of analyzed 
monazites plot near monazite-huttonite isomor-
phic substitution line. Observations using BEI 
image contrast did not show any huttonite or 
cheralite exsolutions within monazite, thus in-
dicating they rather form solid solutions with 
monazite. Diagram of LREE versus U+Th+Pb 
(Fig. 5c) shows one primary field and another 
rather discrete. Fields are divided from each oth-
er by a virtual horizontal line at 0.84 LREE apfu 
(atoms per formula unit). Monazites in the pri-
mary field have a relatively high concentration of 
REE but somewhat lower concentrations of Th as 
their counterpart below 0.84 LREE apfu.
Xenotime grains show relatively consistent 
Y and variable HREE concentrations and lack 
presence of any actinides (i.e., U and Th). Among 
Fig. 6. Comparison of chemical composition (a) and the degree of isomorphic replacement in monazite (b) between studied 
Drava river monazites and monazites from Tauern Window and Pohorje Mts.
263REE-bearing minerals in Drava river sediments, Slovenia, and their potential origin
HREE, Dy, Er, and Yb were identified in ana-
lyzed samples. The quantitative relationship be-
tween Y and HREE in xenotime can be summa-
rized in the following manner: Y >> Yb > Dy ≥ Er. 
Due to the high relative abundance of Y (26.25-
38.60 wt%) and low relative abundance of HREE 
(∑HREE: 8.39-14.19 wt%) investigated xenotime 
grains can be attributed to its Y-rich variety – 
xenotime(-Y).
The HREE+U versus Y diagram of apfu nor-
malized values for xenotime (Fig. 5d) shows very 
scattered results, exhibiting relatively consist-
ent concentrations of Y, while concentrations of 
HREE vary extensively. Lack of grouping and 
scattered results point towards the origin of 
xenotime from variable geological backgrounds 
(Guastoni et al., 2016).
Potential origin of monazite(-Ce) in Drava river 
fluvial sediments
Potential source of monazite(-Ce) from Drava 
river was estimated by comparing its chemical 
composition and the degree of monazite-cheralite 
and monazite-huttonite isomorphic substitutions, 
with monazite(-Ce) from three localities; two from 
Tauern Window (FinGer et al., 1998; hoscheK, 
2016) and one from Pohorje Mts. (uher et al., 
2014). In the work of FinGer et al. (1998), samples 
of monazite were acquired from Granatspitz 
massif in Tauern Window, from S-type granite 
gneiss metamorphosed to amphibolite facies. 
Monazites, presented by hoscheK (2016), 
originate from Eclogite belt in Tauern Window, 
which is a thrust sheet between Venediger 
nappe and overlying Glockner nappe, consisting 
predominantly of Mesozoic metasediments and 
Fig. 5. Ternary diagram of Ce-La-Nd for investigated monazite (a) and diagrams of: 4×(P+Y+REE) versus 4×(Si+Th+U) apfu and 
(b) LREE versus Pb+Th+U apfu for investigated monazite (c) and HREE+U versus Y apfu normalized values of investigated 
xenotime grains (modified after HARLOV et al. (2008)) (d).
264 Aleš ŠOSTER, Janez ZAVAŠNIK, Mihael RAVNJAK  & Uroš HERLEC
metabasites that were metamorphosed during 
Alpine plate collision. Monazites, presented by 
uher et al. (2014), were acquired from granitic 
pegmatites emplaced in UHP rocks in Pohorje 
Mts, north-eastern Slovenia. Comparison of 
chemical composition of monazites from our 
case study with that of monazites from localities 
mentioned above are presented in the diagram 
Pb+Th+U versus LREE (Fig. 6a). The diagram 
shows three affinity fields marked in Roman 
numerals (I, II, III). The chemical composition 
of Drava monazites significantly corresponds 
with monazites originating from Eclogite belt 
and Granatspitz massif in Tauern Window. 
Affinity field (i), containing the majority of Drava 
monazites, corresponds greatly with monazites 
from Eclogite belt studied by hoscheK (2016), 
while monazites in affinity field (ii) correspond 
significantly with monazites from Granatspizt 
massif (FinGer et al., 1998). Affinity fields (i) and (ii) 
are divided by 0.84 LREE apfu virtual horizontal 
line as established above based solely on Drava 
monazites. Monazites from Pohorje Mts. (uher 
et al., 2014) on the other hand form a separate 
group (iii) below 0.70 LREE apfu which does 
not contain any of the investigated Drava river 
monazites. The degree of monazite-cheralite-
huttonite isomorphic replacement is shown in 
4×(Si+Th+U) versus 4×(P+Y+REE) (Fig. 6b). The 
diagram shows similarity in monazite-cheralite 
isomorphic replacements in studied Drava river 
monazites and monazites from Eclogite belt, 
while the degree of replacement in monazites from 
Granatspitz massif and Pohorje vary extensively, 
with only some points exhibiting a nearly similar 
degree of isomorphic substitution. Based on 
observed similarity in chemical composition 
and degree of monazite-cheralite isomorphic 
replacement, it can be concluded that majority of 
monazites, found in Drava fluvial sediments, are 
derived from localities in Tauern Window, among 
which rocks of Eclogite belt seem to be their most 
likely source.
Conclusions
1) Heavy minerals account for 2.3-4 wt% of bulk 
fluvial sediment of Drava river. Heavy mine-
rals are present in a grain size fraction be-
tween 0.04 and 0.5 mm and predominantly 
consist of garnet group minerals, hematite, 
rutile and other minerals that account for less 
than 0.1 wt%. 
2) Monazite is classified as monazite(-Ce) based 
on prevalent content of Ce compared to La and 
Nd. The variable content of Th in monazite 
suggests it originates from at least two diffe-
rent parent rocks or geological settings. The 
first group is igneous monazite where relative 
concentrations of ThO2 are 3 – >7.20 wt% and 
second, hydrothermal or greenschist-amphi-
bolite metamorphic grade monazite, with con-
centrations of ThO2 < 1 wt%. Monazite was 
found to be forming monazite-cheralite and 
subordinate monazite-huttonite isomorphic 
replacements. No phase separation was obser-
ved using BEI, indicating solid solution relati-
onship rather than mineral exsolution.
Fig. 6. Comparison of chemical composition (a) and the degree of isomorphic replacement in monazite (b) between studied 
Drava river monazites and monazites from Tauern Window and Pohorje Mts.
265REE-bearing minerals in Drava river sediments, Slovenia, and their potential origin
3) Based on relatively high abundance of Y com-
pared to ∑HREE, xenotime grains from he-
avy mineral fraction from Drava river can be 
classified as xenotime(-Y), which most pro-
bably originates from various sources or geo-
logical backgrounds.
4) Chemical composition and degree of isomorp-
hic replacement in Drava river monazite corre-
spond significantly with monazites from Tauern 
Window, indicating the possible origin of mona-
zite from Eclogite belt and subordinately from 
Granatspitz massif, while Pohorje Mts. can be 
disregarded as a potential source of monazite.
Acknowledgements
The authors acknowledge the financial support 
from the Slovenian Research Agency (research core 
funding contract No. 1000-17-0510; MR No. 39235) 
and Program group Geochemical and Structure pro-
cesses (P1-0195). SEM/EDS analyses were performed 
at Centre for Electron Microscopy and Microanalysis 
(CEMM), Jožef Stefan Institute, Ljubljana, Slovenia. 
Authors would like to thank two anonymous review-
ers for their detailed and constructive comments and 
suggestions improving the quality of the manuscript.
References
alaM, n. M., choWdurY, M. i., kaMal, M., ghose, 
s., islaM, n. M., MustaFa, n. M., Miah, M. M. 
n. & ansarY, M. M. 1999: The 226Ra, 232Th and 
40K activities in beach sand minerals and be-
ach soils of Cox‘s bazar, Bangladesh. Journal 
of Environmental Radioactivity, 46/2: 243-
250, doi:10.1016/S0265-931X(98)00143-X.
Bea, F. 1996: Residence of REE, Y, Th and U 
in granites and crustal protoliths; impli-
cations for the chemistry of crustal melts. 
Journal of Petrology, 37: 521-552, doi:10.1093/
petrology/37.3.521.
BerManec, v., PalinkaŠ, l. ŠouFek, M. & ZeBec, 
v. 2014: Gold in the Drava and Mura ri-
vers – Geological genesis and mineralogical 
analysis. Podravina, 13/25: 7-18.
Breiter, k. 2016: Monazite and zircon as major 
carriers of Th, U and Y in peraluminous gra-
nites: Examples from the Bohemian Massif. 
Mineralogy and Petrology, 110/6: 767-785, 
doi:10.1007/s00710-016-0448-0.
Burnotte, e., Pirard e. & Michel, g. 1989: 
Genesis of gray monazites: Evidence from the 
Paleozoic of Belgium Economic Geology, 84/5: 
1417-1429, doi:10.2113/gsecongeo.84.5.1417.
deer, W. a., hoWie, r. a. & ZussMan, j. 2013: 
An introduction to the rock-forming mine-
rals, 3rd edition. The Mineralogical Society, 
London: 498 p.
Finger, F., Broska, i., roBerts, M. P. & scherMaier, 
a. 1998: Replacement of primary monazite by 
apatite-allanite-epidote coronas in an amp-
hibolite facies granite gneiss from the eastern 
Alps. Am. Mineral., 83: 248-258, doi:10.2138/
am-1998-3-408.
Förster, h. j. 1998a: The chemical composition 
of REE-Y-Th-U-rich accessory minerals in 
peraluminous granites of the Erzgebirge-
Fichtelgebirge region, Germany, Part I: The 
monazite-(Ce)-brabantite solid solution se-
ries. American Mineralogist, 83: 259-272, 
doi:10.2138/am-1998-3-409. 
Förster, h. j. 1998b: The chemical composition 
of REE-Y-Th-U-rich accessory minerals in 
peraluminous granites of the Erzgebirge-
Fichtelgebirge region, Germany. Part II: 
Xenotime. American Mineralogist, 83: 1302-
1315, doi:10.2138/am-1998-11-1219.
grand‘hoMMe, a., janots, e., Bosse, v., seYdoux-
guillauMe, a. M. & de ascenção guedes, r. 
2016: Interpretation of U-Th-Pb in-situ ages 
of hydrothermal monazite-(Ce) and xeno-
time-(Y): evidence from a large-scale regi-
onal study in clefts from the western alps. 
Minereralogy and Petrology, 110/6: 787-807, 
doi:10.1007/s00710-016-0451-5.
Guastoni, a., Pozzi, G., secco, l., schiazza, M., 
Pennacchioni, g., Fioretti, a. M. & nestola, 
F. 2016: Monazite-(Ce) and Xenotime-(Y) 
from an LCT, NYF Tertiary pegmatite field: 
Evidence from regional study in the Central 
Alps (Italy and Switzerland). The Canadian 
Mineralogist, 54: 863-877, doi:10.3749/
canmin.1600021.
harrison, t. M., Mckeegan, k. d. & le Fort, 
P. 1995: Detection of inherited monazite 
in the Manaslu leucogranite by 208Pb/232Th 
ion microprobe dating: Crystallization 
age and tectonic implications. Earth and 
Planetary Science Letters, 133: 271-282, 
doi:10.1016/0012-821X(95)00091-P.
harlov, d. e., ProcháZka, v., Förster, h. j. & 
Matejka, d. 2008: Origin of monazite-xeno-
time-zircon-fluorapatite assemblages in pe-
raluminous Melechov granite massif, Czech 
Republic. Mineralogy and Petrology, 52/2: 
191-219, doi:10.1007/s00710-008-0003-8.
hoschek, g. 2016: Phase relations of the REE 
minerals florencite, allanite, and monazi-
te in quartzitic garnet–kyanite schist of the 
266 Aleš ŠOSTER, Janez ZAVAŠNIK, Mihael RAVNJAK & Uroš HERLEC
Eclogite Zone, Tauern Window, Austria. 
European Journal of Mineralogy, 28: 735-750, 
doi:10.1127/ejm/2016/0028-2549.
kiM, Y., Yi, k. & cho, M. 2009: Parageneses and 
Th-U distributions among allanite, monazite, 
and xenotime in Barrovian-type metapeli-
tes, Imjingang belt, Central Korea. American 
Mineralogist, 94: 430-438, doi:10.2138/
am.2009.2769.
Mioč, P. & Žnidarčič, M. 1989: Basic Geological 
map of SFRY, Interpretation of map sheet 
Maribor and Leibnitz. Federal Geological 
Survey Beograd, Beograd: 69 p.
Pleničar, M., ogorelec, B. & novak, M. 2009: The 
Geology of Slovenia. Geological Survey of 
Slovenia, Ljubljana: 612 p.
rao, n. s. & Misra, s. 2009: Sources of Monazite 
sand in Southern Orissa beach placer, 
Eastern India. Journal of the Geological 
Society of India, 74: 357-362, doi:10.1007/
s12594-009-0140-7.
roBB, l. 2004: Introduction to Ore-Forming pro-
cesses. Wiley-Blackwell, Oxford: 373 p.
schandl, e. s., gorton, M.P. & davis, d. v. 1994: 
Albitization at 1700 ± 2 Ma in the Sudbury 
- Wanapitei Lake area, Ontario: implicati-
ons for deep-seated alkalic magmatism in 
the Southern province. Canadian Journal of 
Earth Sciences, 31/3: 597-607, doi:10.1139/
e94-052.
schandl, e. s. & gorton, M. P. 2004: A textural 
and geochemical guide to the identification of 
hydrothermal monazite: Criteria for selection 
of samples for dating epigenetic hydrother-
mal ore deposits. Economic Geology, 99/5: 
1027-1035, doi:10.2113/gsecongeo.99.5.1027.
švecová, e., čoPjaková, r., losos, z., šKoDa, r., 
nasdala, l. & cícha, j. 2016: Multistage evo-
lution of xenotime(-Y) from Písek pegmati-
tes, Czech Republic: An electron probe mi-
cro-analysis and Raman spectroscopy study. 
Mineralogy and Petrology, 110/6: 747-765, 
doi:10.1007/s00710-016-0442-6.
tockner, k., uehlinger, u. & roBinson, c. t. 2009: 
Rivers of Europe. Academic press, London: 
728 p.
uher, P., janák, M., konečný, P. & vraBec, M. 2014: 
Rare-element granitic pegmatite of Miocene 
age emplaced in UHP rocks from Visole, 
Pohorje Mountains (Eastern Alps, Slovenia): 
accessory minerals, monazite, and uraninite 
chemical dating. Geologica Carpathica, 65/2: 
131-146, doi:10.2478/geoca-2014-0009.
van eMden, B., thornBer, M. r., grahaM, j. & 
lincoln, j. l. 1997: The incorporation of acti-
nides in monazite and xenotime from placer 
deposits in Western Australia. The Canadian 
Mineralogist, 35/1: 95-104.
vassas, c., Pourcelot, l., vella, c., carPena, j., 
PuPin, j. P., Bouisset, P. & guillot, l. 2006: 
Mechanisms of enrichment of natural radio-
activity along beaches of Camargue, France. 
Journal of Environmental Radioactivity, 91/3: 
146-159, doi:10.1016/j.jenvrad.2006.09.002.
voncken, j. h. l. 2016: The Ore Minerals and 
Major Ore Deposits of the Rare Earths. In: 
voncKen, j. (ed.): The Rare Earth Elements, 
Springer International Publishing, 15-52.
Wang, j., tatsuMoto, M., li, x., PreMo, W.r. & 
chao, e. c. t. 1994: A precise 232Th- 208Pb chro-
nology of fine-grained monazite age of Bayan 
Obo REE-Fe-Nb ore deposit, China. Geochim. 
Geochimica et Cosmochimica Acta, 58/15: 
3155-3169, doi:10.1016/0016-7037(94)90043-4.
Zhu, x. k. & o’nions, r. k. 1999: Monazite 
chemical composition: some implications for 
monazite geochronology. Contributions to 
Mineralogy and Petrology, 137/4: 351-363, 
doi:10.1007/s004100050555.
Žnidarčič, M. & Mioč, P. 1987: Basic geological 
map of SFRY. L 33-56, Maribor and Leibnitz 
1:100.000. Federal Geological Survey, 
Beograd.