Replacement of primary monazite by apatite-allanite-epidote

American Mineralogist, Volume 83, pages 248-258, 1998

Replacement of primary monazite by apatite-allanite-epidote coronas in an amphibotitefacies granite gneiss from the eastern Alps

Fnrrz FtNcnn, Icon BnosKA,x MAr,cor,vr P. Ronnnrs, mln ANonms Scrmnuaren

Institut fiir Mineralogie der Universitiit Salzburg, Hellbrunnerstrasse 34, 4-5020 Salzburg, Austria

Assrnlcr

Accessory monazite crystals in granites are commonly unstable during amphibolite fa-cies regional metamorphism and typically become mantled by newly formed apatite-allan-ite-epidote coronas. This distinct textural feature of altered monazite and its growth mech-anism were studied in detail using backscattered electron imaging in a sample ofmetagranite from the Tauern Window in the eastern Alps. It appears that the outer rims ofthe former monazites were replaced directly by an apatite ring with tiny thorite inter-growths in connection with Ca supply through metamorphic fluid. Around the apatite zone,a proximal allanite ring and a distal epidote ring developed. This concentric corona struc-ture, with the monazite core regularly preserved in the center, shows that the reactionkinetics were diffusion controlled and relatively slow.

Quantitative electron microprobe analyses suggest that the elements released from mon-azite breakdown (P, REE, Y, Th, U), were diluted and redistributed in the newly formedapatite, allanite, and epidote overgrowth rings and were unable to leave the corona. Thissupports the common hypothesis that these trace elements are highly immobile duringmetamorphism. Furthermore, microprobe data suggest that the preserved monazite coreslost little, possibly none of their radiogenic lead during metamorphism. Thus, metastablemonazite grains from orthogneisses appear to be very useful for constraining U-Th-Pbprotolith ages.

On the basis of these findings and a review of literature data, it seems that monazitestability in amphibolite facies metamorphic rocks depends strongly on lithologic compo-sition. while breaking down in granitoids, monazite may grow during prograde metamor-phism in other rocks such as metapelites.

INrnonucrroN

The LREE-phosphate monazite is widely reported asan accessory mineral in peraluminous granitoids andhigh-grade metamorphic gneisses. Despite its very minoroccuffence and generally small size, monazite is never-theless of outstanding geological importance. First, it isvery useful for geochronology, because it incorporateslarge quantities of Th and U (Parrish 1990; Montel et al.1996). Second, monazite crystals typically carry between40 and 80Vo of the LREE content of their host rock (Bea1996; Schitter 1997). Thus, monazite growth, fraction-ation, or destruction in the course of various geologicalprocesses has the potential to influence greatly the REEsystematics of rocks. Considering this, it is surprising thatthe stability relations of monazite have attracted relativelylittle interest. Although important data are available onmonazite solubility in felsic melts (Rapp and Watson1986; Montel 1993). However, very little is known aboutmonazite-forming or -destroying reactions during meta-morphism. For example, it is commonly believed that

* Present address: Geological Institute of the Slovak Academyof Sciences, Dubravska Cesta 9, 842 26 Bratislava, Slovakia.

0003-o04x/98/0304-0248$05.00 248

monazite is stable under medium- to high-grade meta-morphic conditions, but that monazite often reacts to al-lanite or apatite over the temperature range of low-gradeconditions (Overstreet 1967; Bons 1988; Ward et al.1991; Smith and Barreiro l99O; Lanzuotti and Hanson1996). In this paper, we document the partial replacementof accessory magmatic monazite by an apatite-allanite-epidote paragenesis, which formed when the host pera-luminous granite was metamorphosed under amphibolitefacies conditions. Using quantitative electron microprobeanalysis, we discuss the element redistribution that oc-curred in conjunction with monazite breakdown. Lastly,we discuss some potential implications of our findings forU-Th-Pb geochronology using monazite.

INvnsrrclrED RocK

The monazite replacement process was investigated us-ing a sample of S-type granite gneiss from the centraleastern Alps (Granatspitz massif, Tauern Window Fig. 1).The magmatic formation age of the rock is about 320-330 Ma (Cliff 1981). Amphibolite facies metamorphismin the Tauern Window occurred during the Tertiary inconnection with Alpine orogenesis. The exposed rocks

FINGER ET AL.: MONAZITE REPLACEMENT, AMPHIBOLITE FACIES GNEISS 249

TleLe 1. Chemical and modal composition of the Granatspitzgnerss

Trace Rare earthelements elemenls

wtTo oxides (ppm) (ppm)Modal

mineralogy

sio, 70 91Tio, o.24Al,o3 14.83FeO. 1 65MnO 0.03MgO 0.42CaO 1.28Na.O 3.01K,O 5 09P,Ou O.23L.O.t . 0 84

Total 98 53

Qtz 31Kfs 31Pl 28B r 5M s 3Acc 2

C r T L aRb 249 CeSr 81 NdY 1 1 S mZt 93 EuN b 5 T bBa 339 YbTh 16 Lu

24.15 0 822.3

5.U5

0.580 6 71 .48o21

520

\ o -l o,l 590

530

J

eX2

X

f

X Xx x X

X X

ffi P"-o.M""o.oi" .o"k"

fix_l zilledal-venediger| | and Tuxea granitoids

lT +l Gr"nut"pic s,"nit"

f-l ttleumorptric country rocksL------l ot lhe Variscan granitoids

Frcuno 1. Sample location and geological setting. (a) Posi-tion of the Tauern Window (TW) within the geological frame-work of central Europe. This tectonic window exposes the Pen-ninic unit, which is the lowermost structural zone of the easternAlps. For more detailed geological information see Frasl (1958)and Frank et al (1987) (b) Geological sketch map ofthe centralTauern Window. The Granatspitz granite massif is one of thenumerous metamorphosed Variscan plutons that occur as domesalong the central axis of the Tauern Window (Finger et al. 1993).The sample chosen for this study (arrow) was collected near theFelbertauern mountain road. at the Odalm. where several freshboulders of typical Granatspitz granite gneiss had been broughtdown by a local landslide. The degree of Alpine metamorphismreached about 550-600 'C and 5-6 kbar in this area (Frank etal. 1987). The temperatures and metamorphic isograds shown aretaken from Hoernes and Friedrichsen (1974\ and based on Oisotope thermometry.

form an E-W-oriented thermal and tectonic dome struc-ture (Fig. l). Pressure-temperature conditions of the Al-pine-age metamorphism in the Tauern Window rangefrom about 4-5 kbar and 400-500'C at its periphery toabout 6 kbar and 600'C in the center (Dachs 1986; Franket al. 1987).

The sample used for this study is highly representativeof the Granatspitz massif. It is a little deformed, medium-to coarse-grained two-mica gneiss with some green bio-

Notei FeO. - total iron as FeO. Mineral abbreviations after Kretz (1983)except Acc : accessory minerals

tite-muscovite pseudomorphs after cordierite (Frasl1967). Magmatic textures are widely preserved in therock. Alpine metamorphism caused a slight filling of theplagioclases with white mica and some epidote or clino-zoisite grains, exsolution of rutile (sagenite) and ilmenitein biotite, partial alteration of biotite to muscovite, andrecrystallization of small quartz crystals and some oli-goclase along grain boundaries. There is no textural ormineralogical evidence for any later major low-Z over-printing. The chemical and modal composition of the rockis given in Table l.

MrcnornxruREs oF MoNAzrrE, REPLACEMENT

The transformation of the accessory monzvite grains toapatite, allanite, and epidote can be best studied in back-scattered electron (BSE) images (Fig. 2). The crystals aremostly 20 to 50 pm in size, rarely up to 100 pm. Thetransformed grains are always characterized by a distinctcorona structure, with a preserved monazite core in thecenter and a mantle of apatite, allanite, and epidote asconcentric growth rings.

The monazite zone in the center of the pseudomorphsgenerally appears fairly homogeneous in the BSE image.In some cases, faint concentric compositional zonationmay be seen, which is interpreted as a magmatic growthfeature (see next section).

The apatite zone s;.rfiolunds the monazite zone and israrely more than a few micrometers wide. In most casesit appears spotty in the BSE image (Fig. 2). This spottytexture results mainly from the intergrowth of many smallthorite crystals, which produce bright spots in the BSEimage. In a few cases, tiny monazite islands were iden-tified within the apatite zone. These seem to testify thatthe apatite is directly replacing primary mon.vite. Also,because apatite formed adjacent to only the preservedmonazite core, it appears that the phosphorous requiredfor apatite growth came directly from the breakdown ofmonazite. The outer rim of the apatite zone is mostlyeuhedral to subhedral (Fig. 2), which suggests that it mayrepresent the former rim of the primary monazite grain.

, +- * 'I T

T

250 FINGER ET AL.: MONAZITE REPLACEMENT, AMPHIBOLITE FACIES GNEISS

GRAIN 7

@lrunz ffi4 l-lern lfiillfill]lflep

FrcunB 2. Representative BSE-images of partially replacedrelatively large, accessory monazites from the Granatspitz granitegneiss. The sketches beside each photograph are schematic illus-trations of the different mineral zones. Grain 3 has grown con-temporaneously with two small magmatic zircon crystals and ishosted by biotite (note the cleavage). Grain 7 is a single crystalobtained from a heavy mineral fraction and is embedded in ep-oxy resin. Numerals refer to microprobe point analyses listed inTable 2. Mineral abbreviations are those recommended bv Kretz( re83) .

These observations indicate that apatite was only able togrow within or at least very close to the former monazitelattice.

The allanite zone mantles the apatite zone and alwaysappears relatively bright in the BSE images (Fig. 2). Ingeneral, it has roughly the same width as the apatite zone.However the outer boundaries of the allanite zone aretypically anhedral, with spiky and lobate protrusions intothe epidote mantle. Thus, it appears that the allanite zoneis not strictly replacing earlier monazite, but developinglargely outside of the former monazite grain after the lat-ter had been partially replaced by apatite.

The epidote zone always forms the outer shell of thecorona. It is typically more voluminous than the allanitezone. The epidote apparently invades the minerals thathost the former monazite grains preferentially along theircleavage planes (Fig. 2, grain 3).

The extent to which monazite is replaced differs fromgrain to grain. In some cases, it appears that more than50 volvo of the original monazite fell victim to the trans-formation process. In other grains, the reaction productsform only a thin rim, where not more than lO volVo ofthe monazite was altered. Within some larger apatite crys-tals, armored monazite inclusions occur that lack evi-dence of transformation. The strongest alteration com-monly is found in monazites located along grainboundaries of the major rock-forming minerals or alongcleavage planes of biotite and plagioclase. This patternsuggests that the CaO, SiO, AlrO., and FeO required forapatite-allanite-epidote corona formation were suppliedby a fluid phase.

In general, it appears that where the apatite zone lssmall then the allanite and the epidote zones are alsosmall, and vice versa. These observations indicate somekind of stoichiometric control on the replacement reactionand clearly argues against a purely metasomatic process.The almost perfectly concentric arrangement of apatite,allanite, and epidote growth zones implies that the trans-formation process was governed primarily by the kineticsof the outward diffusion of elements derived from themonazite. This is discussed later in more detail.

ANlr,vt:tca,l RESULTS

The monazite zone

Eight relict monazite grains were analyzed. All hadfairly similar compositions. Examples of analyses fromthe two grains shown in Figure 2 are given in Table 2together with their mineral formulae calculated on thebasis of four O atoms. The monazite compositions are asolid solution consisting dominantly of the LREE-phos-phate end-member (monazite s.s.), with about 5-I07oeach of the huttonite (ThSiO.), brabantite [CaTh(PO,),]and xenotime (Y-HREE-PO.) components.

Generally, the highest La contents and LalSm ratioswere measured in the center of the monazite grains. Bothvalues tend to decrease slightly with distance from thecenter of the grain. Conversely, Y contents increase out-ward (Figs. 2 and 3). This kind of chemical zonation mostlikely reflects normal magmatic crystaL4iquid fraction-ation involving early monazite crystallization with suc-cessively greater depletion of the melt in those elementswith the highest monazite/melt Ko values (Wark and Mil-ler 1993). Also, the huttonite/brabantite ratio may behigher in the centers of the monazites; this is a commonlyobserved feature in magmatic monazites (I. Broska, un-published results).

The theoretical REE pattern for magmatic monazite hasbeen calculated from the whole-rock data (Table 1) usingthe K" values from Cocherie (1984) and Sawka (1988)and the simple relation Cr.. : C..*Ko. The calculatedpattern is similar to the monazite core composition (Fig.3), suggesting that monazite crystallized early in chemi-cal equilibrium with granite melt.

The apatite zone

In most microprobe analyses carried out in the apatitezone, significant amounts of ThO, were recorded (Table2), including domains that appear to be free of any brightthorite spots in the BSE image. Only in a few analyzedpoints could a Th-poor apatite composition be obtained(e.g., Table 2, analysis 6). Likewise, pure thorite com-positions could be obtained only in a very few of thebigger bright spots. Therefore, it seems that apatite andthorite are intergrown on an extremely fine scale, whichis beyond the spatial resolution of the microprobe.

On the basis of measurements obtained with an ex-panded 5 pm diameter beam, we integrated the averagecomposition of the thorite plus apatite mixture that formsthe apatite zone. The mean composition of this mixture

FINGER ET AL.: MONAZITE REPLACEMENT, AMPHIBOLITE FACIES GNEISS

TeeLe 2. Microprobe analyses and mineral formulae from the grains illustrated in Figure 2

251

Grain 7 Grain 3

1 2 3 4 5mon mon ap all ep

6 7 8 9 1 0ap all all ep epa P

1mon

2mon mon

+mon

D N

sio,La,O"Ce,O"PrrO"Ndro3SmrO.Yro.Tho,UO,Al,o3FeOCaOMgoFTotalo - F

Total

27.76 27 .710.69 0.87

1 3.05 13 3327.53 2857

3.29 3.211 1 . 3 6 1 1 1 2

1 .81 1 700.97 0.99I01 9 .290.24 0 36n d n d .n d n d .1 3 1 1 1 0n d n.d.n d n d .

97 04 98.240 00 0.00

97 04 98.24

0 962 0.9540 028 0.0360 197 0.2000 413 0.4250.049 0.0470 .166 0 .1610.026 0.0240.021 0 0210.084 0.0860.002 0 003n d n dn . d n d0.058 0.048n d n dn d n d .

24.28 28.65 37 02o.41 0.47 1 95

1 3.51 12.88 0.8929.00 25.56 1.58

3.58 3.81 0 101 1 5 4 1 1 . 9 3 0 8 82 .09 235 0232 .01 267 0247.78 8.84 3 920 3 6 0 5 9 0 0 3n d n d . n . d .n d n d . n . d1 18 1 49 50.17n.d. n.d. n.d.n d n d . 3 . 0 6

9974 99 24 100.060.00 0.00 1.29

99 74 99.24 9877

0 959 0 966 5.6260 016 0.019 0.3500 200 0.189 0 0590.426 0.373 0.1040 052 0.055 0 0070 165 0.170 0 0570.029 0.032 0 0140.043 0.057 0 0230.071 0 084 0.1600.002 0.005 0 001n d n d n d .n d n d . n d .0.051 0.064 I649n d n d . n . d .n d n d . 1 . 7 3 5

3 9 2 6 n d1.31 32.16u . / 5 5 5 /1 1 5 1 1 2 50 1 2 1 5 20.54 4.500 .10 0 730 1 6 0 1 6063 1 050 03 0.o7n.d. 18 33n.d- 10.82

52.29 11.79n.d 0 303.01 n.d.

99.36 94.271.27 0 00

98.09 94.27

5.A24 n.d.0.230 2.9590.049 0.1890074 0.3790.008 0.0510.034 0.1480 006 0.0230.015 0.0080.025 0 0220 001 0 000n d. 1.987n.d. 0.8329 816 1 162n.d- 0.0411.669 n.d.

n d. n.d. n.d.31 72 35 48 35.776.05 2.22 2 21

12 68 5.62 5 901 60 0.85 0 853.96 275 2.470.32 0 93 0.740 05 1 .05 1 .121 36 0.21 0.220.05 0 28 0.31

17.19 22 34 24.3810.92 745 7.7710.75 16 74 15.830.32 0 26 0.28n.d n.d. n.d

96.98 96.55 98.240 .00 000 000

96 98 96.55 98.24

n d. n.d n.d.2.993 3.024 2.9860.210 0.070 0.0680 438 0.175 0 1800 055 0.026 0 0260 134 0.084 0.0860 011 0.027 0 0210.002 0.048 0.0500.029 0004 00040 000 0 005 0.0061 .912 2 244 2.3980.861 0 559 0.5421 .086 1 .528 1 .416o.o44 0 033 0.034n.d n.d. n.d

28.53 29 56 37.110.72 0.57 2 53

1 1 . 6 9 1 0 5 5 0 4 826.7A 25 26 0.913 . 7 7 3 6 0 0 1 0

12.89 12.55 0.402.03 2 55 0.011.69 2 98 0.158.70 I29 6.85o 20 0 34 0.12n d. n.d. n-dn.d. n.d. n.d.1 1 0 1 . 4 7 4 8 . 9 9n d n.d. n-dn d . n .d 3 .10

98 10 98.70 100740.00 0.00 1 31

98 10 98.70 99 44

0 971 0.984 5 6210.029 0.022 0 4530.173 0.153 0 0310.394 0.364 0.0600.055 0.052 0 0070.185 0.176 0.0260.028 0.035 0.0010.036 0.062 0.0140.080 0.083 0.2790.002 0 003 0.005n.d. n.d. n.d.n d n.d. n.d.o.o47 0062 9391n.d. n.d. n.d.n d. n.d. 1 754

n.d. n d32.20 35 08

J I T Z . O Y

11 .34 6 521 41 0723 .73 312o.75 0.53o25 0 931 . 3 5 0 1 80.03 0 23

19.38 23.959 . 1 9 I 1 7

11.28 15 550.30 0.27n d n . d

96.97 97.930.00 0.00

96 97 97.93

n d. n.d2.968 2.9620.196 0.0840 383 0.201oo47 00220 123 0 094o 0 2 4 0 0 1 50.012 0 0420.028 0 0030.001 0 0042.105 2 3830.708 0 5761.114 1 .4070.041 0.034n d n . d

PSi

CePrNdSm

ThUAIFe

MgF

Note: Formula proportions of cations based on 4 O atoms (monazite); 3 O atoms (apatite), and 12 5 O atoms (allanite and epidote). Numbers in thesecond row refer to analysis points shown in Figure 2 n.d. - not detected

(Table 3) is broadly similar in all measured grains, and itshows that the estimated proportion of thorite in the ap-atite zone is approximately 7 wt%o.

It appears from the analytical results that the apatitelattice contains a relatively high amount of the lessingitecomponent [Cq(REE),(SiO,),(EOH)] (see analysis 6 inTable 2). Because the apatite zone is generally very nar-row, we cannot preclude the possibility that the measuredconcentrations of REE and Y are enhanced by X-raysexcited from neighboring monazite or allanite zones.However, we consider this potential contamination effectto be minor because first, when non-stoichiometric anal-yses were excluded the concentrations of REE and Y areroughly the same in narrower and broader sectors of theapatite zones. Second, no Al or Fe was detected in thestoichiometric apatite analyses, which suggests that noallanite was involved. Third, REE in the apatite appearto be broadly balanced by Si, following correction forthorite content, which would not be the case if the REEwere derived from monazite excitation.

The chondrite-normalized LREE patterns of the apatitezone and the monazite zone are subparallel, but the La.fYratios are significantly lower in the apatite zone (Fig. 3).Nevertheless, the LalY ratios are still fairly high for ap-atite. As an example, Figure 3 shows a hypothetical mag-

matic apatite pattern calculated with literature Ko values(Sawka et al. 1988) from the whole-rock composition ofthe sample. This pattern lies at much lower levels and itsshape is clearly flatter, which suggests that the apatitezone inherited its enriched and steeply fractionated pat-tern from the pre-existing monazite. For comparison,some of the normal accessory apatites of the sample wereanalyzed. Their LREE contents were always close to thedetection limit (Ce,O, is about O.O5-0.1Vo), which isroughly 5 to 10 times lower than the REE content re-corded in the apatite coronas (as predicted by the K"modeling).

The allanite zone

REE+Y+Th make up between 0.8-1 atoms per for-mula unit (apfu) in the allanite zone, indicating that onlya small amount of clinozoisite-epidote solid-solutioncomponent is present (Table 2). The Al contents are highcompared to normal magmatic allanite from granitoids(Petrik et al. 1995), which implies that not much ferrial-lanite component is present. Generally the allanite zonehas the same compositional range in all analyzed grains.However, microprobe profiles show that slight composi-tional zoning exists across the allanite rings in most cases

@ig. a), with decreasing LREE (La-Nd) and increasing

252

1 0 6

1 0 3

* Mnz core

X Mnz rim

-\

-e---

\

10'

L a C e P r N d S m E u G d T b D y Y

Frcunn 3. Chondrite-normalized REE variation diagram forselected light and middle REE, showing the REE chemistry ofthe whole-rock, the monazite and the minerals in the alterationcoronas using the averages from Thble 3 Heavy REE were notdetermined due to very low concentuations below or near to thedetection limit. However, it can be assumed that their behaviorapproximates that of Y (included on the diagram in the place ofHo). Normalization values used are those of Wakita et al. (1971)Filled symbols and crosses are measured concentrations. Opensymbols are theoretical equilibrium compositions of magmaticmonazite, apatite, and allanite calculated from the whole-rockcomposition using the Ko values of Cocherie (1984) and Sawka(1988) and the expression C""" : C*^*Ko.

Y and U outward. The REE variation is apparently bal-anced mainly by the substitution: lpgg:*(A2) +Fe'?*(M1) : Ca'*(A2) * 41:*(M1) (Dollase 1971).

LalNd, Nd/Sm, and LREE/Y ratios in the allanite zoneare on average somewhat higher than in the monazitezone. The chondrite-normalized REE pattern of the allan-ite zone is thus steeper than that of the monazite (Fig. 3),but flatter than that of magmatic allanites from granitoids(see data compilation in Petrik et al. 1995). Using pub-lished K" values (Cocherie 1984) and the whole-rockcomposition of the sample, we estimate that early mag-


1 0 "

1oo

matic allanite in equilibrium with the granitic magmawould have a much steeper LREE pattern (Fig. 3).

The epidote zone

Analyses from the epidote zone reveal little grain tograin compositional variation. Approximately 0.35-0.5REE apfu substitute for Ca in the A2 position. This meansthat the epidote zone contains a considerable allanite sol-id-solution component. Nevertheless, a clear composi-tional gap separates the epidote zone from the allanitezone (Fig. 4). This compositional gap, which may reflectsome kind of sohus behavior, is clearly seen in the BSElmages.

The Al content of the epidote zone is between 2.1 and2.4 apfu, indicating the presence of a clinozoisite com-ponent. From the REE+Y+Th content, which is around0.5 apfu, it appears that only about 50 wtVo of the totalFe in the epidote zone occurs as Fe3* (Petrik et al. 1995).

Like the allanite zone, slight compositional zoning oc-curs across the epidote zone in most cases, in which theLa and Ce contents decrease outward (Fig. a). This sys-tematic decrease in La and Ce is compensated for byincreasing Ca in the A2 position accompanied by increas-ing Al and decreasing Fe. Thus, the overall effect is anincreasing clinozoisite component rimward. Neodymium,Sm, X and U show the reverse behavior compared withLa and Ce and display a slight increase outward acrossthe epidote zone (Fig. 4). Figure 3 shows that the LREE-Y distribution in the epidote zone is much flatter than inthe allanite zone. Lanthanum contents in the epidote zoneare approximately half of those in the allanite zone. How-ever the amount of Sm in the epidote is higher than inthe allanite. Yttrium contents are about three times higherin the epidote zone than in the allanite zone and reachvalues almost equivalent to the monazite. Th is relativelylow in the epidote zone, whereas UO, (0.2-0.4 wt%o) isfairly high.

DrscussroN

The chemical budget of the alteration reaction

High field-strength fface elements, such as those re-quired for the formation of monazite (B LREE, Y, Th)are commonly considered to be highly immobile duringmetamorphism. Thus it was interesting to determinewhether, from the chemical point of view, the new cor-onas around the monazites may be viewed simply as a"diluted" monazite, fed with additional Ca, Fe, Al, andSi. To test this hypothesis, a model calculation was car-ried out. In a first step, the mean modal composition ofthe corona systems was estimated. Relative volumes ofthe apatite, the allanite, and the epidote zone were deter-mined by point counting several BSE images, which gavea mean of 24 vol%o apatite (including thorite inclusions),33 vol%o allanite, and 43 vol%o epidote. These volumescorrespond to 2l wt%o apatite, 38 wtVo allanite, and 41wt%o epidote, using average densities of 3.3 g/cm' (apa-tite), 4.2 g/cm3 (allanite), and 3.5 g/cm. (epidote).

Using these weight proportions and the chemical data

101

FINGER ET AL.: MONAZITE REPLACEMENT. AMPHIBOLITE FACIES GNEISS

Tlele 3. Average analyses of monazite cores, rims, and the mineral zones from the alteration corona

253

Mnz core( n : ) ) \

Mnz rim Ap zone(n : 18) (n: 121-

Aln zone( n : 1 3 )

Ep zone(n : 14 )

P.ousio,La,O.Ce,O"PrrO"Nd,o"Sm,O.Yro.Tho,UO,Al,o3FeOCaOMgo

Total

27.99 + O.750 5 0 + 0 3 0

13.24 + 0 9427 36 + 1.063 66 + 0.30

12 00 + 0.63211 + 0 .281 49 + 0.51823 + 1 .370.27 + 0.13

1.40 + 0 28

98 25

28.45 + 0.75o 43 + 0.20

1 2 6 6 + 1 0 82670 + 1 .513 6 7 + 0 1 3

11 .93 + 0 .502 42 + O.392.24 + O 477.85 + 1.500.42 + O 25

1 .44 + 0 .17

98.20

37 03 + 2.902 .10 + 1 .49u-ol - u 3t1.25 + 0.960.20 + 0 200.48 + 0.430 .11 + 0 .070 1 9 + 0 0 65.69 + 2.860 . 1 8 + 0 . 1 5

49 96 + 3.71

97 86

32.39 + 0.515.63 + 0 50

11 .30 + 0 911 .58 + 0 13433 + 0.29o.77 + 0.19o 42 + 0.141 48 + 0.200.05 + 0.03

1822 + 0 .9910.36 + 0.791 1 2 9 + 1 0 20.42 + O.O4

98.23

35.76 + 0.622 39 + 0 .195.72 + O.400 84 + 0.202.96 + O.210.85 + 0.311_19 + 0 .250.39 + 0.130.23 + 0.09

23.01 + 1 287.90 + 0.53

15.68 + 0 980.40 + 0.05

97.33- Composition of the apatite zone is an average of 12 integrated analyses from four grains using a beam diameter of 5 pm See text for further

details

given in Table 3, the mean chemical composition of thecorona was calculated, i.e., the chemical composition ofa mixture zone. The result, given in column 1 of Table4, shows that the non-monazite elements (SiO,, AlrO3,FeO, CaO, and trace MgO) make up approximately 75wtTo of the corona. The remaining monazite elements,given in column 2 of Table 4 (La, Ce, Pa Nd, Sm, Th,U, Y, P), have been normalized to a total of 97 wtVo(column 3 of Table 4). This total is reasonable for theseelements in monazite because small amounts of Si. Ca.Gd, and HREE are not considered in the calculation. Theresulting normalized monazite composition is strikinglysimilar to that measured at the rims of the preserved mon-azites (compare columns 3 and 4 in Table 4). Uraniumand Y contents are slightly higher and La and Ce contentsare slightly lower in column 3 compared to column 4.Indeed, it appears reasonable to assume that the primarymonazite rim, which had been replaced through reaction,was slightly higher in Y and U and lower in Ce and Lathan the present monazite rim, considering the zoning pat-terns recorded in the preserved monazites (Table 3). Theapproximately 25 wt%o theoretical monilzite in the coronasystem (column 2 of Table 4) matches well with the tex-tural observation that the apatite zone is directly replacinga former monazite rim, since the apatite zone also makesup about 25 wt%o of the corona system according to theresults of point counting.

Although the calculations certainly suffer from someuncertainties inherent to the analytical and modal aver-ages used, the modeling basically suggests closed systembehavior for the elements released through monazitebreakdown. However, although the overall budget ap-pears well balanced, it should be kept in mind that theprocess of monazite destruction leads to significant small-scale elemental redistribution between the mineral zones,effectively concentrating U, X HREE, and MREE, rela-tive to LREE toward the margin of the corona zone (Table3) .

This kind of internal element fractionation throuehout

the corona may have important consequences when rockswith altered monazite are exposed to penetrative leachingthrough fluids or to non-equilibrium partial melting pro-cesses. The enhanced mobilization of the U, Y, HREE,and MREE relative to the LREE could result if the outerzone is preferentially affected by such a process. Like-wise, strong trace and REE fractionation is feasible in thepossible case of selective instability of the allanite or ap-atlte zones.

Nature of the monazite breakdown and reaction

Textures show unequivocally that monazite was notstable during amphibolite facies regional metamorphismof the granitoid rock studied here. However, as mentionedpreviously, not all monazite grains were subjected to al-teration. Armoring of the monazite crystals by mineralswith low intracrystalline diffusion coefficients preventedbreakdown and corona development. This is well illus-trated in Figure 2 (grain 3) where no reaction rim hasdeveloped in the place where the monazite crystal surfaceis shielded by an epitactic zircon. Several mon.vites werealso observed that had survived intact as armored inclu-sions within apatite. Conversely, the breakdown of mon-azite was most pronounced in crystals located along grainboundaries or cleavage planes of major minerals. Theseobservations suggest that the basic requirement for acti-vating the breakdown process is the presence of a meta-morphic fluid phase which delivered Ca, Fe, Si, and Alto monazite grain boundaries. However, considering thedistinct concentric alteration textures, it appears that theactual reaction kinetics were controlled by element dif-fusion from the monazite into the corona minerals andvice versa. At the P-Z conditions and time scale of theregional metamorphic event, elemental diffusivities wereapparently too low for complete monazite consumption.Thus, the extent of corona development around monazitemay serve as an indicator of the time that a granitoid rockspent at elevated metamorphic P-Z conditions.

Microtextures suggest that the new apatite was able to

254

analysis point number

Frcunn 4. Zoning profile outward across the allanite and ep-idote zones of a representative grain. Note the two different Y-axis scales. The analysis points 1 to 7 are in a line and approx-imately 2 pm apatt, except for points 3 and 4 (about 4 pm apart).


Tasle 4, Results of mass balance calculations

D A

sio,La"O"CerO"Pr,O"Ndr03Sm,O.Yro"Tho,UO,Al,03FeOCaOMgo

Total

/ . v o

27 263246 8 60 9 82950 6 60.681 9 40 . 1 5

16 267 1 3

21 39032

97 78

I -VO

3.246.860.982.95u o o

0 6 81 .940 1 5

25 43

30 37

12.3726 183.74

1 1 2 42 5'l2617 4 00 5 8

97 00

24.450 4 3

12.6626.70

J O /

1 1 9 32.422.24

0.42

1 .44

98 20

/Vote.' Column 1 shows the composition of an apatite-allanite-epidotemixture using the estimated modal proportions of each mineral in the co-rona and their average compositions from Table 3. Column 2 is the com-position of the corona minus the "non-monazite" elemenls- As a simplifi-cation, Si and Ca have been trealed as non-monazite elements, althoughsmall amounts are oresent. The values in Column 2 are then recastto 97wt% in Column 3. Column 4 is the measured average composition of mon-azite rims from Table 3. Note that the close similarity between Columns 3and 4 indicates closed-system behavior.

grow only within, or at least very close to, the formermonazite lattice. Thus, it seems that the PO" tetrahedraof the monazite served directly as building blocks for thegowth of apatite. Because the REE have particularly lowdiffusion coefficients in accessory minerals (Liang andWatson 1995), it may be assumed that the LREE releasedduring monazite breakdown only diffused very slowlythrough the apatite zone outward into the allanite/epidotecorona. In any case, the reaction rate would decrease asthe apatite zone grew. This reduced reaction rate, coupledwith insufficient time at elevated temperature, probablyexplains why complete destruction of the monazite coreswas never attained.

Unlike the apatite front that advanced inward into themonazite, the allanite and the epidote ring formed as afringe outside of the original monazite grain. Protrusionsof allanite into the epidote zone suggest that allanite wassuccessively replacing earlier-formed epidote. Likewise,the epidote ring enlarged outward, commonly invadingthe host minerals preferentially along their cleavages(Fig. 2, grain 3), which served either as channels for themetamorphic fluids or as energetically favorable sites foraccessory mineral growth. The nucleation and growth ofthe high-LREE allanite ring adjacent to the apatite is gov-erned most likely by the particularly high LREE concen-ffation in this location. Decreasing availability of LREEaway from the apatite probably resulted in the formationof the LREE-poorer epidote ring only. This interpretationis supported by the fact that both the allanite and theepidote zone have similar internal chemical gradients inwhich the LREE decrease outward (Fig. 4). It is interest-ing that discrete allanite and epidote growth rings formedduring reaction, rather than one continuously zoned al-lanite-epidote solid solution. These rings may suggest amiscibility gap between these two phases. However, as

Ep ido te

*-+---xluoz=x--*-l(

23


TeeLe 5. Central European examples of amphibolite facies metagranitoids with Ap-Aln-Ep coronas around monazite

255

Protolith type Locality

Alpeiner granodioritic gneissSulztal granite gneiss

Winnebach migmatite gneiss

Granatspitz granite gneiss

Hochweissenf eld granite gneissZinken granite gneiss

Waldbach granite gneiss

Weitersfeld "pencil" gneiss

Bittesch gneissLiesnica gneiss

high-K l-type, medium-grained B1-granodioriteKfs phyric S-type Bt granite +Crd

granitic to granodioritic Crd-bearing S-typediatexite

medium to coarse-grained two-mica Crd-bearing S-type granite

Kfs phyric high-K l-type Bt granitefine-grained leucocratic S-type two mica

granitefine- to medium-grained leucocratic S-type

two mica granite

evolved l-type granite

evolved l-type granitemigmatite

central Otztal-Stubai crystalline basement(Austro-alpine unit)

central Tauern Window

Seckau-Bdsenstein Massif (Austro-alpine unit)

Raabalpen Massif (Austro-alpine unit)

Moravian Unit, eastern Bohemian Massil, Austria

Veporic unit, Slovenske Rudohorie, CarpathianMountains

Notej P- f estimates for all the rocks are in the range 500-600 "C and 4-7 kbar (Hoernes and Friedrichsen 1974; Frank et al 1987; Bernroider 1989;Hoinkes and Thoni 1993; Plasienka et al 1997; Schermaier et al. in oress)

yet no experimental data are available on the solid solu-tion relationship between epidote and allanite.

The bulk of Y released from the former monazite wasapparently able to move across the allanite zone into theepidote zone where it concentrated mainly near the outerrim of the corona. It appears that Y was pushed succes-sively rimward due to its poor fit in the allanite-epidotelattice sites, whereas the LREE were preferentiallytrapped in proximal sites. The HREE have probably be-haved in a similar fashion to Y in view of their similarionic radii. Figure 4 shows that the MREE, such as Sm,display behavior intermediate to Y and LREE. Uraniumwas redistributed similar to Y and was concentrated inthe epidote zone, whereas Th became trapped mainly inthe apatite zone as indicated by the presence of many tinythorite inclusions. Despite such localized mobility, mass-balance calculations reveal that none of the former mon-azite constituents was able to leave the corona system insignificant amounts. This finding supports the claim thatelements such as U, Th, Y, and the REE are highly im-mobile during metamorphism.

Whether the observed breakdown of monazite can bedescribed by a stoichiometric reaction is difficult to an-swer. It is likely that the Ca required to form apastite andepidote-allanite was provided by the breakdown of theanorthite component in plagioclase. The Fe needed forthe formation of allanite-epidote could be derived fromthe metamorphic breakdown of biotite, and the potassiumthus liberated may have been consumed in the formationof muscovite (together with Al remaining from the an-orthite breakdown). Howeve! in view of the low reactionrate and the few external atoms required, the necessaryamounts of Ca, Fe, Si, and Al could be supplied simplyby a fluid phase chemically buffered through other meta-morphic reactions and by the solubility equilibria of themajor mineral components of the rock.

Stability of monazite during regional metamorphism

The textures noted here do not appear to have beenpreviously described in the literature. However, the re-placement of primary monazite by apatite-allanite-epidotecoronas is common in many S-type and high-K I-typeamphibolite facies metagranitoids from the Alps, Carpa-thians and eastern Bohemian massif (Table 5). Thus,monazite seems to be unstable under normal amphibolitefacies conditions in granitoid lithologies. Previous workshows that, where newly formed monazite crystals werefound in metagranitoid rocks, metamorphic temperatureswere always in excess of 700 'C (e.9., Schenk and Todt1983; Friedl et al. 1994; Bingen et al. 1996; Biittner andKruhl 1997).

In contrast, it is well known that amphibolite faciesmetapelites commonly contain newly formed metamor-phic monazite, as demonstrated in many geochronologicalstudies worldwide (e.g., Parrish 1990 and referencestherein). Smith and Barreiro (1990) document a casewhere monazites grew in pelitic shists under progradeconditions near the staurolite-in isograd at conditions aslow as ca. 525'C and 3 kbar. Franz et al. (1996) reportmetamorphic monazite growth between 450 and 700'Cin low pressure-high temperature metapelites in northernBavaria. A particularly convincing situation occurs in theMoravian Unit in the eastern part of the Bohemian Mas-sif. Here, at metamorphic conditions of 500-550 'C and4-5 kbar (Bernroider 1989), new monazites grew in me-tapelitic paragneisses, whereas interlayed granitic ortho-gneisses contain primary monazite with Ap-Aln-Ep cor-onas (Finger et al. 1996 and unpublished data).

Implications for U-Th-Pb monazite geochronology

Monazite has been shown to be useful for dating mag-matic and high-temperature metamorphic events (Parrish1990). This usefulness results from the fact that monazite

256

Frcunn 5. Sketch showing grain 7 (Fig. 2) with the resultsof U-Th-Pb chemical dating (in Ma) included in the circles. Dataalso given in Table 5.

is highly retentive (i.e., it frequently exhibits closed-sys-tem behavior) and it has a high-closure temperature ofroughly 725 "C. The typically high concentrations of Uand Th in monazite also contribute to the reliability ofthe age dates. As a result, monazite very often provideshighly precise and concordant U-Th-Pb ages.

In the case of amphibolite facies metamorphic rocks,most published concordant monazite ages have been ob-tained from paragneisses, where monazite formed duringregional metamorphism. A few studies have attempted todate primary monazite relics from orthogneisses using U-Pb isotope systematics. In these cases, strongly discordantresults were obtained from multi-grain monazite frac-tions, which yielded Pb-Pb ages older than the metamor-phic event (Mougeot et al. 1997; Friedl 1997). Based onthe observations made in the present study, discordantresults may be attributed to one or both of the following:(l) The monazites themselves suffered little or no Pb loss,but the analyzed fractions contained monazite with sec-ondary coronas that contained a metamorphic U-Pb com-ponent. (2) The monazites themselves lost radiogenic Pbduring metamorphism, e.g., through diffusion aroundmisoriented domains on the sub-micrometer scale (Blacket al. 1984), but were not completely reset.

For the Granatspitz granite gneiss, it was instructive totest to what extent the monazite cores of the pseudo-morphs have preserved their primary lead contents. To dothis, high-precision electron microprobe measurementswere carried out for Th, U, and Pb in one of the monazitegrains (Fig. 5), using high beam currents and long count-ing times (for analytical procedure see appendix). Theresultant data permit calculation of the total Pb monazitemodel ages (Suzuki et al. 1991; Montel et al. 1996), as-suming that the measured Pb is entirely radiogenic. Thisassumption is reasonable, as common Pb is generally lessthan l%o of the total Pb in monazites having Paleozoicand older ages (e.g., Parrish 1990).

The results of the analyses, given in Table 6, suggestthat the preserved monazite cores did not experience ma-jor Pb loss during Alpine metamorphism. Pb contents inseven analysis points of the measured monazite grainyield Variscan model ages (Fig. 5). The weighted average


Tae|-e 6. Results of U-Th-Pb chemical model age dating grain 7'

Analysis Age

8 184 0.375 0.1327 066 0.306 0 1198 279 0.349 0.1417.535 0.472 0.1327180 0.405 0.1327 810 0.336 0.1356 588 0 465 0.1 16

m 7-6m 7-7m 7-8m 7-9m 7-10m 7 -11m 7 -13weighted averageapatite 1^^^+i l^ )

apatite 3

314 + 43330 + 50335 :r 43327 + 45347 + 49339 + 45321 + 50330 + 102 7 + 5 3J O r t z J

1 7 + 6 3

6.653 0.358 0 0092731 0158 00085 607 0.275 0 005

" See Figures 2 and 5 Model ages calculated using the method of Mon-tel et al (1996) given here in the appendix Also included are three anal-yses from the apatite zone of another grain.

age calculated for these analyses is 330 Ma, which is inperfect agreement with previous 320-330 Ma zircon andRb-Sr whole-rock ages (Cliff 1981). Of course, becausethe precision of the microprobe dating technique is lim-ited, slight Pb loss, corresponding to a few million yearstime, could remain undetected.

U-Th-Pb analyses were also carried out in the apatitezone of another grain, which was wide enough to allowanalysis with a 5 pm diameter beam. Here hardly any Pbwas detected, despite the high Th contents due to thethorite intergrowths in some cases (Table 6). These lowPb contents are compatible with the much younger meta-morphic age obtained from the apatite zones.

Thus U-Th-Pb microprobe data suggest that the meta-stable monazites from amphibolite facies orthogneissesmay well be suitable for revealing magmatic protolithages, particularly where ion microprobe or laser ablationICP-MS can be applied. Strong discordancies of multi-grain monazite fractions, combined with enhanced com-mon Pb contents as shown by Mougeot et al. (1997) andFriedl (1997), may be dominantly due to the presence ofsecondary alteration rims. When using conventional U-Pb multi-grain analysis, it is crucial that metamorphiccoronas be removed through abrasion, if possible, togeth-er with potential thin marginal lead loss zones of themonazite itself. We expect that such drastically abradedfractions will provide near concordant ages in manycases.

As previously mentioned, several monazites remainedunaffected by the transformation process, because theywere protected as inclusions in magmatic apatites. Fordating purposes, such crystals could be particularly use-ful. These may be recognized in heavy mineral grainmounts by their euhedral shape and high transparency.Conversely corona-bearing monazites always appeared asdull, rounded grains.

Acrnowr-BocMENTS

This paper resulted mainly from research work caried out in the frameof a Lise Meitner fellowship project (M00150-GEO) to Broska The Aus-trian FWF is thanked for financial support Parts of the fieldwork werekindly supported by the BMWF through grant OWP 39. Hans Steyrer isgratefully thanked for his invaluable drafting of some of the figures David

Wark and Antonio Lanzirotti provided us with thought-provoking reviewsthat improved the clarity and significance of this work. Last, but by nomeans least, we would like to express our thanks to Volker Hdck for hissuccessful efforts in obtatning the microprobe facility at SalzburgUniversity

RBrnnnNcrs crrED

Bea, F (1996) Residence of REE, I Th and U in granires and crustalprotoliths; Implications for the chemistry of crustal melts Joumal ofPetrology, 37,521 552

Benisek, A and Finger, E (1993) Factors controlling the development ofprism faces in granite zircons: A microprobe study Contributions toMineralogy and Petrology, ll4, Ml-451

Bernroider, M (1989) Zur Petrogenese prlikambrischer Metasedimenteund cadomischer magmatite im Moravikum Jahrbuch der GeologischeBundesanstalt, 132, 349-37 3.

Bingen, B., Demaiffe, D, and Hertogen, J (1996) Redistribution of rareearth elements, thorium, and uranium over accessory minerals in thecourse of amphibolite to granulite metamorphism: The role of apatiteand monazite in orthogneisses from southwestern Norway Geochimicaet Cosmochimica Acta, 60, 1341-1354.

Black, LP, Fitzgerald, JD., and Harley, SL. (1984) Pb isoropic compo-sition, colour, and microstructure of monazites from a polymetamorphicrock in Antarctica Contributions to Mineralogy and Petrology, 85, 141-148

Bons, AJ. (1988) Intracrystalline deformation and slaty cieavage devel-opment in very low-grade slates from the central Pyrenees GeologicaUltraiectina, 56

Biittneq S H and Kruhl, J (1997) The evolution of a late Variscan high-T/low-P region: The south-eastern margin of the Bohemian massif.Geologische Rundschau, 86, 21-38.

Clifl R A (1981) Pre-Alpine history of rhe Pennine zone in the ThuernWindow, Austria: U-Pb and Rb-Sr geochronology Contribution forMineralogy and Petrology, 71, 262-266

Cocherie, A. (1984) Interaction manteau-cro0te: son r0le dans la gendsed'associations plutoniques calco-alcalines, contraintes g6ochimiques(6lements en traces et isotopes du strontium et de I'oxygdne), 246 p.Ph D thesis, Universit6 de Rennes, Rennes

Dachs, E (1986) High-pressure mineral assemblages and theirbreakdown-products in metasediments south of the Grossvenediger, Tauern window,Austria Schweizer Mineralogisch Petrographische Mitteilungen, 66,145 - 161 .

Dollase, W.A. (1971) Refinement of the crystal structures of epidote, al-lanite and hancockite American Mineralogist, 56, 44'7-464

Exley, R A (1980) Microprobe studies of REE-rich accessory minerals:Implications for Skye granite petrogenesis and REE mobility in hydro-thermal systems Earth and Planetary Science Letters, 48, 97-1 10

Finger, E, Frasl, G., Haunschmid, B , Lettner, H , Schermaier, A , Schin-dlmayr, A O , Steyreq H P, and von Quadt, A (1993) The zentralgneiseof the Thuern Window (Eastern Alps): Insight into an Intra-Alpine Var-iscan batholith. ln E Neubauer and J von Raumer, Eds, The pre-Me-sozoic Geology in the Alps, p. 375 391 Springer Verlag, Berlin.

Finger, F, Benisek, A, Broska, L, Friedl, G, Haunschmid, B , Schermaier,A , Schindlmayer, A, Schitter, E, and Steyrer, H P (1996) Altersdatierenvon Monaziten mit der Elektronenmikrosonde-eine wichtige neue Meth-ode in den Geowissenschaften, p 118-122 6th Symposium Tektonik-Strukturgeologie-Kristallingeologie, Universitiit Salzburg, Abstracrs.

Frank, W., H0ck, V, and Milleq C (1987) Metamorphic and rectonichistory of the Central Tauem Window In H.W Fliigel and P Faupl,Eds., Geodynamics of the Eastern Alps, p 34-55. Deuticke, Wien

Franz, G, Andrehs, G., and Rhede, D. (1996) Crystal chemistry of mon-azite and xenotime from Saxothuringian-Moldanubian metapelites, NEBavaria, Germany European Journal of Mineralogy, 8, 1097-1 1 18.

Frasl, G (1958) Zur Seriengliederung der Schieferhiille in den MittlerenHohen Tauem Jahrbuch der Geologische Bundesanstalt, I01,323-4'72.

- (1967) Glimmerpseudomorphosen nach Cordierit im Zentralgneisdes Granatspitzkerns, Hohe Tauern. Mineralogische Mitteilungen Joan-neum. l ( l t . t l - l /

Friedl, G. OSSZ) U/PU Datierungen an Zirkonen und Monaziten aus Ges-

257

teinen vom 6sterreichischen Anteil der Bdhmischen Masse, 242 pPh.D. thesis, Universitiit Salzburg, Salzburg.

Friedl, G., von Quadt, A., and Finger, F. (1994) 340 Ma U/Pb-Monazitalteraus dem niedercisterreichischen Moldanubikum und ihre geologischeBedeutung Tena Nostra, 3(94), 43-46

Hoernes, S and Friedrichsen,H (.1974) Oxygen isotope studies on meta-morphic rocks of the Western Hohe Tauern area (Austria) SchweizerMineralogische und Petrographische Mitteilungen, 51, 769 788

Hoinkes, G and Thcini, M (1993) Evolution of the Otztal-Stubai, Scarl-Campo and Ulten Basement Units In F Neubauer and J. von Raumer,Eds, The pre-Mesozoic Geology in the Alps, p. 485-494 SpringerVerlag, Berlin

Kretz, R (1983) Symbols for rock-forming minerals American Mineral-ogist, 68, 217-219

Lanzirotti, A and Hanson, G N (1996) Geochronology and geochemistryof multiple generations of monazite from the Wepawaug Schist, Con-necticut, USA: Implications for monazite stability in metamorphicrocks Contributions to Mineralogy and Petrology, 125,332-340

Liang, Y. and Watson, E.B. (1995) A simple model for sector zoning inslowly grown crystals: Implications for growth rate and lattice diffu-sion, with emphasis on accessory minerals in crustal rocks. AmericanMineralogist, 80, 1 179-1 187

Montel, JM (1993) A model for monazite/melt equilibrium and appli-cation to the generation of granitic magmas. Chemical Geology, 110,127-146

Montel, J M., Foret, S , Veschambre, M., Nicollet, C., and Provost, A(1996) Electron microprobe dating of monazite. Chemical Geology,13r, 37 -53

Mougeot, R, Respaut, JP, Ledru, P, md Maignac.C (1997) U-Pb ge-ochronology on accessory minerals of the southern edge of the Velaygranite (French Massif Central) European Journal of Mineralogy, l,141-156

Overstreet, W.C (1967) The geological occurence of monazite. U.S. Geo-logical Survey Professional Paper, 530.

Panish, R.R. (1990) U-Pb dating of monazite and its application to geo-logical problems Canadian Journal of Eafth Sciences, 27, 1431-1450.

Petrik, I, Broska, I, Lipka, J, and Siman, P (1995) Granitoid allanite-(Ce): Substitution relations, redox conditions and REE distributions (onan example of I-type granitoids, Western Carpathians, Slovakia). Geo-logica Carpathica, 46, 79-94

Plasienka, D, Janak, M., Frey, M., and Schmidt, S.T (1997) Structuraland metamorphic evolution of a cretaceous core complex: The Veporicsuperunit of the central Westem Carpathians (Slovakia). Terra Nova, 9,Abstract Supplement, 1, 158.

Rapp, R P and Watson, E B. (1986) Monazite solubility and dissolutionkrnetics: Implications for thorium and light rare earth element chemistryof felsic magmas Contributions to Mineralogy and Petrology, 94, 304-316.

Sawka, WN (1988) REE and trace element variations in the accessoryminerals and hornblende from the strongly zoned McMunay MeadowsPluton, California Tiansactions of the Royal Society of Edinburgh, 79,157- 168

Schenk, V and Todt, W. (1983) U-Pb-datierungen anZirkon und Monazitder Granulite im Moldanubikum Niederristerreichs (Waldviertet). Fort-schritte der Mineralogie, 61, 190-191

Schermaier, A., Haunschmid, 8., and Finger, F (1997) Distribution ofVariscan I- and S-type granites in the Eastern Alps: A possible clue tounravel pre-Alpine basement structures. Tectonophysics, 272,315 333

Schitter, F (1997) Spurenelementkonzentration in den gesteinsbildendenMineralien des Gebhartser Diorits und des Eisgamer Granits, bestimmtmittels der Insfiumentellen Neutronenaktivierungsanalyse, 75 p Diplo-marbeit thesis, Universitit Salzburg, Salzburg

Smith, H.A and Barreiro, B (1990) Monazite U-Pb dating of staurolitegrade metamorphism in pelitic schists. Contributions to Mineralogy andPetrology, 105, 602-615

Steiger, R.H and Jaiger, E (19'7'7) Subcommission on geochronology: Con-vention on the use of decay constants in geo- and cosmochronologyEuth and Planetary Science Letters, 36,359-362

Suzuki, K, Adachi, M., and Thnaka, T (1991) Middle Precambrian prov-enence of Jurassic sandstones in the Miro Terrane, central Japan: Th-


258 FINGER ET AL.: MONAZITE REPLACEMENT. AMPHIBOLITE FACIES GNEISS

U-total Pb evidence from an electron microprobe monazite study. Sed-imentary Geology, 75, l4l-147

von Quadt, A and Finger, E (1991) Geochronologische Untersuchungenim iisterreichischen Teil des Stidbdhmischen Batholiths: U-Pb Dati-erungen an Zirkonen, Monaziten und Xenotimen des Weinsberger Gran-its. European Journal of Mineralogy, 3, Beiheft 1, 281

Wakita, H , Rey, P, and Schmitt, R.A. (1971) Abundances of 14 rare-earthelements and 12 other trace elements in Apollo 12 samples: five igneousand one breccia rocks and four soils Proceedings of the 2nd LunarScience Conference, l3l9 1329.

Ward, CD, McArthur, J.M, and Walsh, J.N. (1991) Rare earth elementbehaviour during evolution and alteration of the Dartmoor granite, SWEngland. Journal of Petrology, 33, 785-815.

Wark, D.A and Miller, CF (1993) Accessory mineral behaviour duringdifferentiation of a granite suite: Monazite, xenotime and zircon in theSweetwater Wash pluton, southeastern California, USA. Chemical Ge-ology, 110,49-67

MaNuscnrpr REcETvED FBsnuany 19, 1997MeNuscprpr ACCEpTED NoveNrsen 1. 1997

AppnNmx: Er,ncrnoN MrcRopRoBE ANALyTTCALTECHNIQUES AND CHEMICAL MODEL.AGE

CALCULATIONS

Measurements were carried out on a Jeol JX 8600 in-strument equipped with thee wavelength-dispersive spec-trometers. Except for the determination of chemical mon-azite ages (see below), operating conditions were 15 kVand 30 nA, usually with a maximally focused beam ofabout I pm diameter. All elements were counted for l0s at the peak and for 2 x 3 s at the background positions.Under these conditions, detection limits of about 0.03-0.I wtVo could be obtained. ZAF corrections were rou-tinely applied. X-ray lines were chosen according to thesuggestions of Exley (1980). Commercially available syn-thetic glass standards were used for the REE. Standardsused for the other trace elements are discussed by Benisekand Finger (1993).

Electron microprobe analysis of very narrow mineralzones in the coronas is particularly problematic, becauseX-rays may excite not only elements from the phase ofinterest, but also those of adjacent minerals that fall with-in the excitation volume of the electron beam. This canpotentially lead to analyses that represent mixtures of twophases, rather than just that of the desired phase. To min-imize this risk, the analysis points were set at least 2 pmaway from the margins of the phases. Analyses were con-

sidered valid only when their stoichiometry was "per-fect." Analyses of apatite obtained 2 pm away from thecontact with allanite lack any Fe and Al, which indicatesthat any external compositional influence is mostly neg-ligible over this distance.

For the determination of total lead monazite modelages, which require a particularly precise measurement ofPb in the 0.1-0,2 wtEo range, the probe curent was in-creased to 250 nA at 15 kV. The beam size was enlargedto 5-10 pm diameter to avoid damage to the sample. Forthe analysis of Pb, counting times were 100 s (peak) and2 x 50 s (background). Counting times for the Th and Uwere also extended to 30 s (2 x 15 s) and 50 s (2 x 25s), respectively. Additionally, the elements La, Ce, Pr, Nd,Sm, P, and Y were determined with 10 s (2 x 3 s) count-ing times to enable a reasonable ZAF conection. A smallY interference on the PbMa line was eliminated by mea-suring a Pb-free yttrium standard and routinely colrectingthe monazite analyses by linear extrapolation (Montel etal. 1996).

For every measured point, the model age (r) was cal-culated by solving the following equation (see Montel etal. 1996):

Pb : Thl232[exp(tr'3'r) - 1]208 + (U1238.04)0.9928

x [exp()t'?38r) - 11206 + (U1238.M)0.0072

x [exp(],'?3sr) - ll20'7

where U, Th, and Pb concentrations are in parts per mil-lion, and \232, tr23s, and tr235 are the radioactive decay con-stants of 232Th, 238U, and 23sU, respectively (Steiger andJager 1977).The 2o errors for the single model ages werecalculated by propagating the Th, U, and Pb errors asresulting from the counting statistics of the microprobe(Montel et al. 1996). These were typically + 0.04-0.05wt%o for the Th, t 0.025-0.030 wtTo for the U, and -r

0.018-0.020 wt%o for the Pb, at 250 nA, 15 kV (all 2oerrors). To control the quality of the analyses, monaziteswith precisely known concordant U-Pb ages were mea-sured together with the unknown sample. A weighted av-erage of 324 * 18 Ma was obtained from the standardmonazites, which is in perfect agreement with the con-ventional U-Pb age of 318 + 4 Ma (von Quadt and Finger1991).

Documents

Replacement of primary monazite by apatite-allanite-epidote