Geological Quarterly, 2015, 59 (2): 316-324 DOI: http://dx.doi.org/10.7306/gq.1223
The stability of xenotime in high Ca and Ca-Na systems, under experimental conditions of 250-350°C and 200-400 MPa: the implications for fluid-mediated low-temperature processes in granitic rocks
Bartosz BUDZYŃ1' 2' * and Gabriela A. KOZUB-BUDZYŃ3
1 Polish Academy of Sciences, Institute of Geological Sciences, Research Centre in Kraków, Senacka 1,31-002 Kraków, Poland
2 Jagiellonian University, Institute of Geological Sciences, Oleandry 2a, 30-063 Kraków, Poland
3 AGH University of Science and Technology, Faculty of Geology, Geophysics and Environmental Protection, al. A. Mickie wicza 30, 30-059 Kraków, Po l and
Budzyń, B., Kozub-Budzyń, G. A., 2015. The stability of xenotime in high Ca and Ca-Na systems, under experimental condi tions of 250-350°C and 200-400 MPa: the implications forfluid-mediated low-temperature processes in granitic rocks. Geo logical Quarterly, 59 (2): 316-324, doi: 10.7306/gq.1223
The stability of xenotime was tested by experiments in the presence of a silicate mineral assemblage and two different fluids, 2M Ca(OH)2 or Na2Si2O5 + H2O, under P-T conditions of 200-400 MPa and 250-350°C. The xenotime was stable in runs with 2M Ca(OH)2, replicating the low-temperature metasomatic alterations of granitic rocks, except in experiment at 350°C and 400 MPa, where some (Y,REE)-rich fluorapatite formed. Experi ments with Na2Si2O5 + H2O resulted in significant xenotime al teration and parti al replacement by an unknown (Y,HREE)-rich sili cate, and in the formati on of minor amounts of (Y,REE)-rich fluorapatite. The latter indicate preferential partitioning of Y and REE into silicates over phosphates during low-temperature, metasomatic processes in a high Na-Ca system, similar to peralkaline granitic rocks.
Key words: xenotime, fluorapatite, yttrium silicate, rare earth elements, experimental petrology.
INTRODUCTION Y-rich apatite and Y-rich epidote recognized in metamorphosed granitic rocks (Broska et al., 2005) or by fluorapatite and hingganite-(Y), documented in Skoddefjellet pegmatite from Xenotime, (Y,HREE)PO4, is an accessory mineral of gra Svalbard (Majka et al., 2011). The P-T conditions of xenotime nitic rocks, pegmatites, low- to high-grade metamorphic rocks, alterations are not tightly constrained, and experimental data and migmatites (Forster, 1998). Xeno-ime is also present as are necessary to fully understand the stabil i ty of xenotime as a detrital mineral in sedimentary rocks or as an authigenic phase, function of pressure-temperature conditions, accompanying commonly formed on zircon in siliciclastic rocks (Fletcher et al., mineral assemblage, and fluid composition. 2000; Rasmussen et al., 2004; Rasmussen, 2005). Because of Previous experimental works on xenotime focused on rela its U and Th con-ents, xeno-ime has applica-ions in iso-opic tively simple systems. The stabil ity of xenotime in the presence U-Pb dating using mass spectrometry (Rasmussen etal., 2004; of common metamorphic and igneous fluids (H2 O, NaCl and Rasmussen, 2005) and in chemical U-Th-total Pb dating using KCl brines, CaF2 + H2O, 1M and 2M HCl, 1M and 2M H2SO4, electron microprobe (Cocherie and Legendre, 2007; Hethering- 1M NaOH, and Na2Si2O5 + H2O) was tested at 500 MPa and ton et al., 2008; Suzuki and Kato, 2008). The application of 600°C, and 1000 MPa and 900°C, documenting the dissolution microanalytical methods, such as sensitive high-resolution ion and etchi ng of xenotime crystal faces, the formation of porous microprobe (SHRIMP) or elec-ron microprobe, to date xeno textures or the growth of small xenotime grains in some experi - time provides age data in textural context to constrain the age of ments - but no internal compositional alterations (Hetherington particular igneous, metamorphic or diagenetic processes. Al et al., 2010). The experiments at the same conditions of though xenotime is a relatively stable mineral, fluid-mediated al 500 MPa and 600°C, and 1000 MPa and 900°C, with start i ng teration may lead to xenotime breakdown and replacement by compositions of xenotime + SiO 2 + Al2O3 + ThSiO4 + Na2Si2 O5 + H2O, resulted in compositional alteration of xenotime and en richment in ThSiO4 along rims (Harlov and Wirth, 2012), par tially replicating the Th enrichment of xenotime in granitic pegmatites from the Hidra anorthosite, Norway (Hetherington * Corresponding author, e-mail: [email protected] and Harlov, 2008). Xenotime has also been tested for its solu Received: December 3, 2014; accepted: January 8, 2015; first bility in H2O and H2 O-NaCl fluids at 1 GPa and 800°C, showi ng published online: February 25, 2015 higher solubility of YPO 4 than CePO4 in pure H2 O to X NaCl = 0.27 The stability of xenotime in high Ca and Ca-Na systems, under experimental conditions of 250-350°C and 200-400 MPa: 317
and lower sol u bil ity with in creas ing NaCl con cen tra tion The experiments were conducted at the Deutsche (Tropper et al., 2Q11). The experiments replicating systems of GeoForschungsZentrum (Potsdam, Germany) usi ng a stan natural rocks, involvi ng xenotime + albite + K-feldspar + biotite dard cold-seal, 6 mm bore, René metal au toclaves with H2O + muscovite + SiO2 + CaF2 with a variety of fluids, 2M KOH, 2M as the pressure medium. Four gently flattened capsules, two NaOH, 2M Ca(OH)2 or Na2Si2O5 + H2O, ran under conditions of for xenot ime experi ments with 2M Ca(OH)2 and Na2Si2O5 + 45Qoc and 590 MPa, decreasing to 54Q MPa over 16 days, re H2O fluids, and two for “twin” experiments on monazite sulted in xenotime alteration in all runs, with the formation of (Budzyń et al., 2013, 2015), were placed in each autoclave. Y-rich britholite and fluorapatite (Budzyń and Harlov, 2Q11). Three sets of experiments were run under P-T conditions and There is a lack of pubi ished experi menial works regard i ng the with a duration of: (1) 2QQ MPa, 25QoC, 4Q days; (2) 2QQ MPa, stability of xenotime in the presence of silicates and flu i ds repli 35QoC, 40 days; and (3) 4QQ MPa, 35QoC, 2Q days (Table 1). cating conditions of low-temperature metamorphism or hydro Durations of the experiments were chosen based on previous thermal, post-magmatic alterations in granitic rocks. experiments on the stabilities of monazite and xenotime This study experimentally explores the stability of xenotime (Hetheri ngton et al., 2Q1Q; Budzyń and Harlov, 2Q11; Budzyń under conditions of 200-400 MPa and 25Q-35QoC, in the pres et al., 2Q11; Harlov et al., 2Q11). Pressures and temperatures ence of silicate minerals assemblages and fluid. Experiments were stable during the runs. At the end of the experiments, the with 2M Ca(OH)2 al ka line fluid rep li cate con di tions of the autoclaves were cooled usi ng compressed air, reachi ng tem- low-temperature metamorphism of granitic rocks. The second peraiures of <1QQoc within <1 min. After the runs, the cap set of experi ments, with Na2Si2O5 + H2O alkali fluid, shows Y sules were weighed, opened, and dried in a 1Q5oc oven. The and REE partitioning between phosphates and silicates, during experi menial products were mounted in epoxy and pol ished the low-temperature metamorphic overprint of granites or dur for elec tron microprobe analyses. The second port ion was ing fluid-mediated post-magmatic processes in peralkaline gra sprinkled on the SEM mount with adhesive carbon tape for nitic rocks. back-scattered electron (BSE) imaging.
ANALYTICAL METHODS EXPERIMENTAL AND ANALYTICAL METHODS The primary observations and analyses of the start i ng min EXPERIMENTAL METHODS erals and experimental products were performed using a Hitachi S-47QQ field emission scanning elect ron microscope The xenot ime used for the experi ments is a port ion of the (SEM) equipped with an energy-dispersive spectrometer (EDS) crysial from pegmatite from the North-West Frontier Provi nce at the Institute of Geological Sciences, Jagiellonian University (NWFP), Pakistan, also used in previous experimental works (Kraków, Poiand). by Hetherington et al. (2010), Budzyń and Harlov (2Q11), The chemical compositions of xenotime, (Y,REE)-rich Harlov and Wirth (2 Q1 2 ) and Budzyń et al. (2Q14). The other fluorapatite, and unnamed (Y,HREE)-rich sil icate were deter natural minerals used include gem-quality albite (Ab1QQ; mined us i ng a Cameca SX 100 electron microprobe equipped Roznava, Slovakia), labradorite (An6QAb27Kfs3; Chihuahua, with four-wavelength-dispersive spectrometers (WDS) at the Mexico), sanidine (Eifel region, Germany), muscovite (pegma Department of Special Laboratories, Laboratory of Electron tite, Siedlimowice, SW Poland), biotite (gneiss, Sikkim Microanalysis, Geological Institute of Dionÿz Stur (Bratislava, Himalaya, India), and garnet (Gore Mountain, NY, USA). The Slovak Republic). The xenotime was analyzed under conditions chosen mineral compositions and used weight proportions (Ta of 15 kV accelerating voltage, 18Q nA beam current, and 3 pm ble 1) roughly repl icate that of granitic rock. Relat ively high beam size, focused on the grain mount and coated with ca. 25 amounts of xenotime (ca. 5 mg of xenoii me in ca. 34.6 to nm carbon film. The natural and synihetic standards, and the 4Q.Q mg total charge of capsule) were added to guarantee ob corresponding spectral lines used for standardization were as servation of xenotime in specimens with experimental products, follows: apatite (P Ka ), PbCO3 (Pb Mß ), ThO2 (Th Ma ), UO2 because the material from each capsule was planned to be split (U Mß ), YPO4 (Y La ), LaPO4 (La La ), CePO4 (Ce La ), beiween 3-4 portions. CaF2 (Suprapure, Merck) was used in PrPO4(Pr Lß ), NdPO4 (Nd La ), SmPO4 (Sm La ), EuPO4 experi ments in excess as a source of Ca and F to increase re (Eu Lß ), GdPO4 (Gd La ), TbPO4 (Tb La ), DyPO4 (Dy Lß ), action rates, and to form fluorapatite. Synthetic SiO2 was used HoPO4 (Ho Lß ), ErPO4 (Er Lß ), TmPO4 (Tm La ), YbPO4 instead of natural quartz to increase reaction rates. Garnet was (Yb La ), LuPO4 (Lu Lß ), fayalite (Fe Ka ), barite (S Ka ), added to test the fluid-mediated partitioning of Y between xeno- wollastonite (Ca Ka , Si Ka ), SrTiO3 (Sr La ), Al2O3 (Al Ka ), and time and garnet. GaAs (As La ). The counti ng times (peak/background, in sec.) Crushed and sieved minerals to a fraction of 5Q-25Q pm were as follows: P 10/5, Pb 300/150, Th 60/30, U 30/20, Y 1Q/5, were washed in ethanol in an ultrasonic bath, and foreign or al La 10/5, Ce 10/5, Pr 30/15, Nd 10/5, Sm 10/5, Eu 30/15, Gd tered mineral grains were picked by hand under a binocular mi 20/10, Tb 20/10, Dy 30/15, Ho 60/30, Er 60/30, Tm 30/15, Yb croscope. The flu ids used included 2M Ca(OH)2 and Na2Si2O5 30/15, Lu 100/50, Fe 20/20, S 10/10, Ca 20/10, Sr 20/10, Al + H2O, as they were the most aggressive fluids in previous ex 10/10, Si 10/10, and As 20/20. (Y,REE)-rich fluorapatite and un periments on monazite (Budzyń et al., 2Q11) and xenotime named (Y,HREE)-rich sil icate were analysed usi ng two condi (Budzyń and Harlov, 2Q11). The minerals and reagents were tions of: (1) 15 kV, 20 nA for F (30/15 sec), Si (10/5), Na (10/5), weighed and mixed together while dry. The mixed sol i ds were Al (10/5), Mg (10/5) P (10/5), Ca (10/5), K (10/5), Cl (10/5), Fe loaded into 3 mm wide, 15 mm long Au capsules, into which (10/5), Mn (10/5), Ti (10/5); fol i owed by (2) 15 kV, 80 nA for Y fluid was added using a syringe, which were then pinched shut. (30/15), Sr (60/30), Pb (30/15), Ce (40/20), La (40/20), Nd Then the capsules were arc-welded shut us i ng a Lampert PUK (30/15), Pr (50/25), Sm (30/15), Eu (60/30), Gd (40/20), Tb US argon plasma torch. Prior to runs, the capsules were (20/10), Dy (60/30), Th (30/15), U (40/20), and 1-5 pm beam checked for leaks by weigh i ng, dry i ng in a 1Q5oc oven over size, depend i ng on the size of the grain ana iysed. night, and then weighi ng again. Chemical analyses of silicates and compositional WDS X-ray maps of xenotime from experi ments were performed us- 1 Broz uzń n Gbil A Kozub-Budzyń A. Gabriela and Budzyń Bartosz 318 Table 1
Experimental conditions, starting materials (mg), and remarks
Experi T P Du ration 2M Total Solids Real M ineral Xtm Ab Lbr Kfs Bt Ms Grt SiO2 CaF2 Na2Si2O5 H2O Remarks ment [oC] [MPa] (days) Ca(OH)2 charge added charge products Starting minerals are not X12C-G4 250 200 40 5.14 - 4.01 3.20 4.00 2.11 4.02 4.23 3.07 0.75 - 5.22 35.75 29.80 35.02 Wo altered. Wollastonite formed. Minerals not altered, X12C-G5 350 200 40 5.06 - 3.91 3.05 4.18 1.85 4.05 3.95 3.08 0.80 - 5.17 35.10 29.41 34.58 - new phases not pro duced. Delicate crystals of Y-bearing fluorapatite X12C-15 350 400 20 5.44 3.96 2.90 3.90 2.13 4.30 3.85 2.87 0.80 5.13 35.28 29.54 34.67 YAp - - formed on the xenotime surface. Del i cate crys tals of (Y,REE)-rich fluorapatite formed on the xenotime surface. K-feldspar is YAp, YSi, partially replaced by al- X12N-G4 250 200 40 4.99 4.06 3.02 3.85 1.90 4.33 4.00 2.79 4.72 5.95 39.61 33.30 39.25 - - Amph bite. Crystals of un named (Y,HREE)-rich silicate formed. Delicate needle-like crystals of amphibole are present. Crystals of (Y,REE)-rich fluorapatite formed on the xenotime surface. K-feldspar is partially re YAp, YSi, placed by albite. High X12N-G5 350 200 40 5.04 4.02 3.13 3.97 1.93 3.93 3.86 2.91 5.02 5.43 39.24 33.21 38.64 - - Amph amounts of unnamed (Y,HREE)-rich silicate formed. Delicate nee dle-like crys tals of am phibole are present. Crystals of (Y,REE)-rich fluorapatite formed on the xenotime surface. K-feldspar is partially re placed by albite. Large YAp, YSi, X12N-15 350 400 20 5.34 3.90 3.42 3.75 2.06 4.52 4.06 3.25 4.92 5.44 40.66 34.55 39.99 crystals (reaching 200 - - Amph pm across) of unnamed (Y,HREE)-rich silicate formed. Delicate nee dle-like crys tals of am phibole are present.
“Total charge” and “solids added” represent original weights of materials; “real charge” represents lower values related to loss of solids during charging the capsule The stability of xenotime in high Ca and Ca-Na systems, under experimental conditions of 250-350°C and 200-400 MPa: 319
Fig. 1. Experimental products from runs with 2M Ca(OH)2 fluid (BSE images)
A - xenotime, K-feldspar, biotite, and muscovite showing no signs ofalterations in run at 2GG MPa and 250°C; B - wollastonite formed in a run at 200 MPa and 250oC; C - unaltered silicates in products from runs at 400 MPa and 350oC; D - delicate, needle-like crystals of Y-bearing fluorapatite, formed on the xenotime surface in an experiment run at400 MPa and 350°C; mineral names abbreviations: B t- bi o tite, Grt - garnet, Kfs - K-feldspar, Ms - mus co vite, Qz - quartz, Xtm - xenotime, YAp - Y-beari ng fluorapatite ing a JEOL SuperProbe JXA-8230 electron microprobe ces 1-3 *). Wollastonite is the only phase formed in the lowest equipped with five wavelength-dispersive spectrometers in the P-T run at 200 MPa and 250°C (Fig. 1B). Delicate, needle-like Laboratory of Critical Elements AGH-KGHM, at the Faculty of crysials of Y-rich fluorapatite (identification based on the EDS Geology, Geophysics and Environmental Protection, AGH Uni analysis) are present on the xenotime surface in products of the versity of Science and Technology (Kraków, Poland). The 400 MPa and 350°C run (Fig. 1D). analyses were performed us l ng 15 kV accelerati ng voltage, 20 nA beam current, and 5 pm beam size for feldspars, 3 pm for EXPERIMENTS WITH XENOTIME AND NA 2 SI2 O 5 + H2 O FLUID micas, and a fo cused beam for garnet. The counti ng times on peak/background (in sec.) were 20/10 for Si, and 10/5 for the All experiments with Na2Si2O5 + H2O resulted in xenotime other elements in feldspars and micas; 10/5 for all elements ex atieration and in the formation of new phases (Fig. 2). The cept 30/15 for Y in garnet. Compositional WDS X-ray maps xenotime shows partial disso i ution and some etchi ng on the were collected using conditions of 15 kV, 100 nA, 100 ms dwell surface (Fig. 2B, F, K, L). The xenotime in experimental prod time, 0.33 pm step size and 0.3 pm beam size. ucts shows no compositional variat ions compared to the start ing NWFP xenotime (Table 2). Delicate crystals of (Y,REE)-rich fluorapatite are present on the xenotime surface (Fig. 2D-G) or RESULTS form individual, pristine grains (Fig. 2O) in all experimental P-T ranges. (Y,REE)-rich fluorapatite shows high variations of REE EXPERIMENTS WITH XENOTIME AND 2M CA(OH)2 FLUID content (Appendix 4), i.e. 21.97-22.16 wt.% (Y+REE)2O3 in X12N-05 (200 MPa, 350°C) vs. 6.42-12.85 wt.% (Y+REE)2O2 Xenotime and the other start i ng minerals in the experi men in X12N-15 (400 MPa, 350°C). These correspond to Ca and P tal products from runs with 2M Ca(OH) 2 are not altered regard variations of 32.86-37.81 wt.% CaO and 27.71-28.16 wt.% ing textures and composition (Fig. 1A, C, Table 2 and Appendi P2O5 (X12N-05), and 41.55-50.61 wt.% CaO and
* Supplementary data associated with this article can be found, in the online version, at doi: 10.7306/gq.1223 320 Bartosz Budzyń and Gabriela A. Kozub-Budzyń
10 0 36.01-37.90 wt.% P2O5 (X12N-15). The (Y, REE)-rich B CO h- co co 0 Ö co 0 co fluorapatite is characterized by a high fluorine content of 1— 0 0 5 0 05 101.27 100.19 100.51 3.41-7.48 wt.% in X12N-05 and 3.26-6.12 wt.% in X12N-15.
CO CNJ CNJ CNJ CNJ The small size of (Y,REE)-rich fluorapatite from the X12N-04 0 O 0 0 0 0 0 0 0 0 CO Ö Ö d 0 0 0 d 0 0 (200 MPa, 250°C) run prevented electron microprobe analyses. V V V v v Beside (Y,REE)-rich fluorapatite, an unnamed CO 0 LO CNJ CO CO CO LO CNJ h- 0 5 (Y,HREE)-rich sil icate formed in all runs with Na2 Si2 O5 + H2 O. CO CM 0 5 CNJ N. 0 5 0 5 N. ticed in Si content (Appendix 4 ), i.e. 67.05-73.68 wt.% SiO2 0 '3 - co 10 CM h- 0 LO CO CNJ CO h O co co 0 (200 MPa, 250°C), 48.62-49.37 wt.% SiO2 (200 MPa, 350°C) LU 0 d d d d d d d d d and 58.86-66.67 wt.% SiO2 (400 MPa, 350°C), and Y concen CO 0 CO CO 0 5 LO CO co 0 10 trations 11.17-12.98 wt.% Y 2 O3 (200 MPa, 250°C), CM CO 0 CNJ 0 CNJ 0 CO 0 co 0 co 0 co 0 O 0 d d d d d d 20.20-22.00 wt.% Y2 O3 (200 MPa, 350°C), and X 11.08-12.09 wt.% Y 2 O3 (400 MPa, 350°C). The elevated Ca CO 0 0 CO 10 CNJ CO 10 0 5 0 5 0 5 CNJ CNJ co 0 5 CM 0 LO h h CO h - N. LO LO co con tent of 1.70-2.42 wt.% CaO was noticed at 200 MPa and Q CO 0 id d id d id d |d d ld d co d 350°C, vs. lower 0.08-0.37 wt.% CaO at 200 MPa, 250°C, and CO 0.21-0.54 wt.% CaO at 400 MPa, 350°C. The unnamed CNJ CNJ 0 '3 CO 0 5 O 10 CO co CO CM 0 0 0 05 O 05 0 0 0 (Y,HREE)-rich silicate has relatively low concentrations of Th -Q 0 d d O d d O d d 1— and U, i.e. <0.03-0.11 wt.% ThO 2 , 0.04-1.49 wt.% UO2 CO (200 MPa, 250°C), <0.03 wt.% ThO 2 and <0.04 wt.% UO2 0 0 CNJ 10 CO CO O CO h- CNJ CNJ CNJ 0 5 CM CO CO CO CO LO CO LO CNJ co co CNJ "O (200 MPa, 350°C) and <0.03 wt.% ThO 2 , 0.39-2.11 wt.% UO2 CO 0 CO d CO d CO d cd d cd d cd d O (400 MPa, 350°C). There were no fluorine or chlorine substi CO 0 CO LO CNJ LO '3 - LO CO 0 co 0 5 0 tutes in the formed unnamed (Y,HREE)-rich silicate. CM '3 - 0 LO 0 LO 0 LO 0 LO 0 '3 0 LO 0 3 Ö 0 d d d d d d In all runs, K-feldspar was partially replaced by albite due to LU Ö 0 0 0 0 0 the albitisation process (Fig. 2J, M and Appendix 1). Delicate, CO 0 CO CO CO 10 CO CO LO CNJ 10 co 1 0 co needle-like crystals of Na, Fe, Mg, and Al-rich amphibole (iden CM CO 0 CO 0 CO 0 CO 0 CD 0 co 0 co 0 £ Ö 0 Ö d 0 d 0 d O d 0 d 0 d tification based on EDS analyses) formed in all runs (Fig. 2B, CO O). The small size of amphibole prevented electron microprobe CO 0 'sf CO CO CO CO 0 5 h- LO LO h- co h- CNJ CM 0 0 O 0 0 0 0 analyzes. The starting albite, garnet, and micas were not alt "O X Ö 0 Ö d 0 d 0 d O d 0 d 0 d tered (Ap pen di ces 1- 3 ). CO CNJ CNJ CNJ CNJ 0 CO CNJ CO CO CO LO 10 10 CNI 0 0 0 0 0 0 0 d d d d d d CL Ö 0 Ö 0 0 O 0 0 DISCUSSION CO CO CNJ 0 5 CO LO CNJ '3 - co '3 co 0 '3 - R CO CO 0 5 0 CO LO LO co 05 CNJ CM Ö 0 Ö d 0 5 05 d O d 0 d cd d The experiments show significant differences in xenotime >“ '3 - '3 - CO CO '3 - co stability, depending on fluid composition. The experiments with CM CO 0 N. 0 N. CO CO co 0 'si- 0 5 LO 2M Ca(OH)2 were prom isi ng regarding the alteration of xeno O 0 0 0 0 0 0 0 =) 0 0 Ö d 0 d 0 d 0 d c i d 0 d time and the formation of secondary fluorapatite. Previous ex periments, under conditions of 450°C and 590 MPa decreasing CM N. CO N. O 0 5 0 5 CO CO LO 0 CNJ co co to 540 MPa over 16 days, and with starting materials of xeno _c 0 0 0 0 0 0 0 0 0 0 0 1— Ö 0 Ö d 0 d 0 d 0 d 0 d 0 d time * albite * K-feldspar * biotite * muscovite * SiO2 * CaF2 * 2M Ca(OH)2, resulted in partial dissolution of the xenotime grain CM h - CO N. CNJ 0 5 '3 - CO '3 - CNJ co N. 0 co O C\1 CNJ CNJ 0 CNJ O CNJ 0 CNJ CNJ 0 CNJ 0 surfaces, and in the formation of Y-rich britholite and CO Ö 0 Ö d 0 d 0 d 0 d 0 d 0 d fluorapatite (Budzyń and Harlov, 2011). Furthermore, recent in CO 0 CNJ CO CO CNJ CNJ LO h - CO '3 - co c o experiments constraining the stability of xenotime, using the 0 LO CNJ CO CNJ '3 - CO CO CNJ co co CNJ CD CNJ CM LO 0 LO d LO d d d LO d LO d d same starting materials as used in this work, reported xenotime CL CO CO CO CO co co co alteration and the formation of (Y,HREE)-rich fluorapatite or CO 10 h- CO 0 5 co 0 5 C CO (Y,HREE)-rich britholite in the presence of 2M Ca(OH)2 fluid in the wide P-T range of 200-1000 MPa and 450-750°C, as well as the formation of (Y,HREE)-rich epidote at 800-1000 MPa 0 0 0 0 0 0 '3 - '3 - CNJ '3 - '3 - CNJ and 650°C (Budzyń et al., 2014). Here, the experi ments show (days) Duration that xeno time remains stable in the presence of alka I ine fluid CD £ with high-Ca activity under experimental P-T conditions of 200 0 0 0 0 0 0 0 MPa and 250-350°C, representing the lowest P-T conditions of p 0 0 0 0 0 0 c CNJ CNJ '3 - CNJ CNJ '3 - [MPa] CD the metamorphism of granitic rock. However, the presence of X 05 (Y,HREE)-rich fluorapatite indicates that limited alteration of c 0 0 0 0 0 0 KO ■ c 10 10 10 10 10 10 xenotime may occur under conditions of 400 MPa and 350°C. B CNJ CO CO CNJ co co CO In nature, progressive metamorphism may result in the '3 10 10 '3 10 10 gradual presence of xenotime. In lower greenschist to upper 0 0 0 0 Ó Ó Ó Ź Ź Ź amphibolite facies metasedimentary rocks from the CNJ CNJ CNJ CNJ CNJ CNJ Paleoproterozoic Mount Barren Group (southwestern Austrat Sample X X X X X X lia), detrital xenotime is stable be I ow 450°C, disappears duri ng The stability of xenotime in high Ca and Ca-Na systems, under experimental conditions of 250-350°C and 200-400 MPa: 321 Fig. 2. Experimental products from runs with Na2Si2O5 + H2O fluid (BSE images) A - unaltered minerals from an experiment at 200 MPa and 250°C; B - xenotime grains showing partial dissolution on the surface and delicate crystals of amphibole from runs at 200 MPa and 250°C; C - the unnamed (Y,HREE)-rich silicate replacing xenotime (200 MPa, 250°C); D-G - small crystals of (Y,REE)-rich fluorapatite formed on the xenotime surface in runs at 200 MPa and 250°C (D), and 200 MPa and 350°C (E-G); H, I - grains of the unnamed (Y,HREE)-rich silicate formed during an experiment at 200 MPa and 350°C; J - grain of K-feldspar with pseudomorphic replacement by albite along the rim due to an albitisation process; K - etched surface of the xenotime (400 MPa, 350°C); L - grains of the unnamed (Y,HREE)-rich silicate filling cracks of the xenotime (400 MPa, 350°C); M - xenotime partially replaced by the unnamed (Y,HREE)-rich silicate with the (Y,REE)-rich fluorapatite formed, and K-feldspar partially replaced by albite due to albitisation; N - large grains of the unnamed (Y,HREE)-rich silicate from 400 MPa and 350°C run; O - the (Y,REE)-rich fluorapatite and needle-like crystals of amphibole formed at 400 MPa and 350°C conditions; mineral names abbreviations: Ab - albite, Amph - amphibole, YSi - (Y,HREE)-rich silicate; other explanations as in Figure 1 322 Bartosz Budzyń and Gabriela A. Kozub-Budzyń Fig. 3. BSE image and compositional WDS X-ray maps of the xenotime (Xtm) partially replaced by the unnamed (Y,HREE)-rich silicate (YSi) under conditions of 400 MPa and 350°C Note that false enrichment of Y in grain boundaries of the xenotime and enrichment of Na and Ca in grain boundaries of the albite and CaF2, respectively, are related to relief of grains in a polished grain mount; explanations as in Figure 2 the progressive metamorphism of pelites under ported in S-type gran ites affected by greenschist- to amphibo- mid-greenschist to amphibo I ite facies, and is present again un lite facies metamorphism (Broska et al., 2005). The formation of der conditions of 650°C and ca. 8 kbar (Rasmussen et al., secondary Y-rich apatite and Y-rich epidote rims around xeno 2011). The growth of metamorphic xenotime was also docut time were documented in metamorphosed S-type granges, mented on detrital xenotime under temperature conditions of however, in low-Ca granites, the alteration of xenotime resulted 450°C in quartz-muscovite-chlorite schist (Rasmussen et al., in the forma tion of Y-enriched epidote, but not apatite (Broska 2011). The overgrowths were interpreted as resulting from et al., 2005). Our experimental setup had the bulk CaO content compositional alterations or from the replacement of detrital significantly increased in relation to that of the natural rocks due xenotime by metamorphic xenotime, due to a fluid-mediated to labradorite, and high amounts of CaF2 and 2M Ca(OH)2 were cou pled dis so lu tion-reprecipitation pro cess (Ras mus sen et al., used in excess to increase reaction rates. Despite the high 2011). The gradual presence of xenotime was also reported in availabili ty of Ca, no epidote formed and limi ted format ion of Alpine metapelites. The temperature conditions of xenotime (Y,REE)-rich fluorapatite occurred only at 400 MPa and 350°C. breakdown and the formation of secondary HREE-rich epidote The experimental results are consistent with observations in and apatite were constrained to 450-528°C, with the disap nature, showi ng that xenotime remains stable in a high-Ca sys pearance of xenotime and co-existing monazite replaced by al- tem under conditions of 200-400 MPa and 250-350°C, with lim lanite or epidote in the biotite zone (Janots et al., 2008). The re ited alteration at 400 MPa and 350°C. versed reactions, i.e. altanite breakdown and the formation of There are limited expert mental data on the stability of xeno monazite and xenotime, occurred at temperatures from 560 to time in the presence of the alkali fluid used in this work. Previ 610°C, depend I ng on the bulk CaO content and Ca/Na ratios ous experiments, utilizing a simple system of xenotime and (Janots et al., 2008). According to these works, xenotime is sta Na2Si2O5 + H2O fluid, performed under high-grade conditions of ble under temperature conditions below 450°C. The xenotime 1000 MPa and 900°C, documented the disso l ution of xenotime alterations re lated to fluid-mediated overprint were also re grain edges (Hetheri ngton et al., 2010). Another study used a The stability of xenotime in high Ca and Ca-Na systems, under experimental conditions of 250-350°C and 200-400 MPa: 323 mix of xenotime + ThO 2 + SiO2 + Al2O3 and Na2Si2O5 + H2O The volume diffusion of Pb in xenotime is extremely slow and fluid, resulting in ThSiO4 en richment along rims of xenotime experimentally constrained as slower than in zircon or monazite grains under conditions of 500 MPa and 600°C, and 1000 MPa (Cherniak, 2006). The Pb loss in 40 pm grain at 800°C is esti and 900°C (Harlov and Wirth, 2012). Because there was no mated to ca. 5% of total Pb over 100 Ma (Cherniak, 2006). Simi source of other elements present in the natural rock system, the larly, the closure temperature limiting the volume diffusion of Pb alteration mechanisms and formation of secondary phases in monazite is restricted to 800-900°C (Cherniak et al., 2004; were limited. In our study, the xenotime reacted in the presence Gardes et al., 2006). However, the fluid-aided al terations may of the alkali fluid and silicate minerals assemblage, roughly rep cause the diffusion of REE, Th, U, and Pb in monazite, resulting licating granitic rock composition. Fluid-aided alterations re in Th-U-Pb age disturbance (Teufel and Heinrich, 1997; sulted in dissoi ution of the xenotime surface (Fig. 2B, F, K), Seydoux-Guillaume et al., 2002; Harlov et al., 2011; Budzyń et without affecting the composition of internal domains, similarly al., 2013, 2015) or a resetting of the Th-U-Pb ages (Williams et to the previous experiments from Hetherington et al. (2010). al., 2011). The mechanism of alterations leading to However, dissolved xenotime supplied P, Y, and HREE to form compositional mod i fi ca tion in monazite is related to a (Y,REE)-rich fluorapatite, while lab radorite and CaF2 were a fluid-aided coupled dissolution-reprecipitation process. Al source of the remain i ng Ca and F. It has to be noted that though concentrations of U, Th, and Pb in the NWFP xenotime (Y,REE)-rich fluorapatite is not the main phase accumu I ati ng were low, there were no notable changes after the experiments component rel eased from the xenot ime. The alkali fluid with (Table 2). There are no modifications of Y and REE concentra high F activity promoted the mobility of Y and REE, incorpo tions that may indicate alteration of the xenotime via a rated in the unnamed (Y,HREE)-rich silicate that crystallised ei fluid-aided coupled dissolution-reprecipitation process, as doc ther as individual grains to grain aggregates (Fig. 2H, I, N) or umented in the experimental studies on monazite, including re formed subhedral to anhedral grains replac ing the xenot ime cent experi ments usi ng the same P-T condit ions and durat ion (Figs. 2C, L, M and 3). In the latter case, the unnamed (Budzyń et al., 2013, 2015). Taki ng these into account, as well (Y,HREE)-rich sil icate, present along with the xenotime rims or as the excess of CaF2 and fluid used to increase reaction rates, inside xenotime grains, probably formed due to fluid penetration it is un likely that a longer duration of experi ments could signifi - along cracks in the xenotime grains (Figs. 2L, M and 3). The al cantly affect the stabil i ty of the xenotime. Summarizi ng, the ex teration mechanism is interpreted as a pseudomorphic replace periments indicate that although alkali fluid mediates dissolution ment, due to a fluid-mediated coupled dissol ution-repre- and partial replacement of the xenotime by secondary phases, cipitation process (cf. Putnis, 2002, 2009; Harlov et al., 2011). the structure of xenotime prevents fluid infiltration from leadi ng There is also a distinct pattern of Y and REE substitution in the to compositional alteration of its internal domains. (Y,HREE)-rich silicate phase, showing similar concentrations of Y and rE e in runs at 2 0 0 MPa and 250°C to 400 MPa and 350°C, and signifi cant enrichment in Y and REE with an iso- CONCLUSIONS baric (200 MPa) temperature increase (250 to 350°C) from 11.08-12.98 to 20.20-21.27 wt.% Y 2O3, and from 5.28-7.24 to The experiments regarding the stability of xenotime, to 12.85-13.57 wt.% REE2O3. The composition of the unnamed gether with corresponding experiments on monazite from (Y,HREE)-rich sil icates is apparently variable and strongly de Budzyń et al. (2013, 2015), provide important data on the parti pends on the P-T conditions. tioning of Y, REE, Th, and U between phosphates and sil icates An analogue of the (Y,HREE)-rich sil icate phase in natural during low-temperature metasomatic processes. examples is difficult to find. Gerenite-(Y), 1. Xenotime is stable in the presence of alka i ine fluid with (Ca,Na)2(Y,REE)3Si6O 1 8 2H2O, a rare mineral recognized in high Ca activ ity, showi ng lim i ted alteration related to the forma mag matic aplite-pegmatites from the Strange Lake peralkaline tion of Y-rich fluorapatite under P-T conditions of 400 MPa and granitic pluton in northeastern Canada (Jambor et al., 1998), 350°C. con tains most of the cations as in the experimental products, 2. The experiments indicate that the formation of however, its composition is very different. An unnamed (Y,HREE)-rich silicates and (Y,REE)-rich fluorapatite can be re (Y,HREE)-bearing hydrated silicate associated with lated to low-temperature alteration of xenotime, mediated by al turkestanite, (Th,REE)(Ca,Na)2K1-x^x[Si8O20]n H 2O, was de kali fluid with high Na activ ity. A large amount of the unnamed scribed in peralkaline gran i tes of the Morro Redondo Complex (Y,HREE)-rich silicates formed in the experiments indicates a in Brazil (Vilalva and Vlach, 2010). Both phases are re I ated to preferential partitioning of Y and HREE into silicates, rather precip itation duri ng the post-magmatic stage, in the presence than phosphates, during low-temperature metasomatic pro of HFSE-rich fluids at temperature conditions of ca. 450°C cesses. (Vilalva and Vlach, 2010). The formation of a phase with a simi 3. The xenotime internal domains are not affected by lar composition to turkestanite was experimentally achieved compositional alteration in high Ca and Na-Ca systems, sugt during the alteration of monazite in the presence ofa silicate as gesting that xenotime can serve as a stable geochronometer in semblage and Na2Si2O5 + H2O at 450-600 MPa and metasomatically altered rocks of granitic composition. 450-500°C (Budzyń et al., 2011). The lower P-T experi ments with monazite, sil icates, and Na2Si2O5 + H2O fluid documented Acknowledgements. The analyses ofthe starting minerals the formation of a phase with a composition similar to steacyite, were financially supported by the National Science Centre re Th(Na,Ca)2 K1-x^x[Si8O20], replacing monazite at 200-400 MPa search grant number 2011/01/D/ST10/04588. This work was fi and 250-350°C (Budzyń et al., 2013, 2015). The formation of nancially supported by the Institute of Geological Sciences, Pol fluorapatite incorporating Y and REEs released from expert ish Academy of Sciences as the research project “EXP”. B. mentally-altered xenotime is limited, most likely due to the high Budzyń thanks W. Heinrich and D.E. Harlov for the access to availabili ty of Si in the bulk system, promoti ng the formation of the experimental laboratories of the Deutsche the unnamed (Y,HREE)-rich sil icate in this study, or a Th- and GeoForschungsZentrum, Potsdam, Germany. P. Konecnÿ is LREE-bearing silicate phase in monazite experiments (Budzyń thanked for his assistance with electron microprobe analyses. et al., 2013, 2015). The reviews from I. Broska and R. Kryza, and editorial handling Xenotime, similarly to monazite, serves as a Th-U-Pb by T. Peryt are greatly acknowledged. M. Jastrzębski is thanked geochronometer, characterized by a high closure temperature. for his comments on an early version of this article. 324 Bartosz Budzyń and Gabriela A. Kozub-Budzyń REFERENCES Broska, I., Will iams, C.T., Janak, M., Nagy, G., 2005. Alteration position, chronology, and a protocol for electron microprobe and breakdown of xenotime-(Y) and monazite-(Ce) in granitic microanalysis. Chemical Geology, 254: 133-147. rocks of the Western Carpathians, Slovakia. Lithos, 82: 71-83. Hetherington, C.J., Harlov, D.E., Budzyń, B., 2010. Experimental Budzyń, B., Harlov, D.E., 2011. The experimental alteration of initiation of dissolution-reprecipitation reactions in monazite and xenotime in the presence of fluids and aluminosilicate minerals. xenotime: the role of fluid composition. Mineralogy and Petrol Mineralia Slovaca, 42: 173-174. ogy, 99: 165-184. Budzyń, B., Harlov, D.E., Williams, M.L., Jercinovic, M.J., 2011. Jambor, J.L., Roberts, A.C., Grice, J.D., Birkett, T.C., Groat, L.A., Experimental determination of stability relations between Zajac, S., 1998. Gerenite-(Y), (Ca,Na)2(Y,REE)3Si6O18-2H2O, a monazite, fluorapatite, allanite, and REE-epidote as a function new mineral species, and an associated Y-bearing of pressure, temperature, and fluid composition. American Min gadolinite-group mineral, from the Strange Lake peralkaline eralogist, 96: 1547-1567. complex, Quebec-Labrador. Canadian Mineralogist, 36: Budzyń, B., Harlov, D.E., Konećny, P., 2013. Experimental, 7 93-800. fluid-aided, low temperature mobilization of Y+REE and Janots, E., Engi, M., Berger, A., Allaz, J., Schwarz, J.-O., actinides between (Y+REE)-bearing phosphates and silicates. Spandler, C., 2008. Prograde metamorphic sequence of REE Mineralogia - Special Papers, 41: 34. minerals in pelitic rocks of the Central Alps: implications for alla- Budzyń, B., Harlov, D.E., Majka, J., Kozub, G.A., 2014. Experi nite-monazite-xenotime phase re I at ions from 250 to 610°C. mental constraints on the monazite-fluorapatite-allanite and Journal of Metamorphic Geology, 26: 509-526. xenotime-(Y,HREE)-rich fluorapatite-(Y,HREE)-rich epidote Majka, J., Prsek, J., Budzyń, B., Bacik, P., Barker, A., Łodziński, phase relations as a function of pressure, temperature, and Ca M., 2011. Fluorapatite-hingganite-(Y) co ronas as products of vs. Na activity in the fluid. Geophysical Research Abstracts, 16: fluid induced xenotime-(Y) breakdown in the Skoddefjellet peg EGU2014-8583. matite (Svalbard). Mineralogical Magazine, 75: 159-167. Budzyń, B., Konećny, P., Kozub-Budzyń, G.A., 2015. Stability of Putnis, A., 2002. Mineral replacement reactions: from macroscopic monazite and disturbance of Th-U-Pb system under experimen observations to microscopic mechanisms. Mineralogical Maga tal conditions of 250-350°C and 200-400 MPa. Annales zine, 66: 689-708. Societatis Geologorum Poloniae, 85: 405-425. Putnis, A., 2009. Mineral replacement reactions. Reviews in Miner Cherniak, D.J., 2006. Pb and rare earth elements diffusion in xeno alogy and Geochemistry, 70: 87-124. time. Lithos, 88: 1-14. Rasmussen, B., 2005. Radiometric dating of sedimentary rocks: Cherniak, D.J., Watson, E.B., Grove, M., Harrison, T.M., 2004. Pb the application of diagenetic xenotime geochronology. diffusion in monazite: a combined RBS/SIMS study. Earth-Science Reviews, 68: 197-243. Geochimica et Cosmochimica Acta, 68: 829-840. Rasmussen, B., Fletcher, I.R., Bengston, S., McNaughton, N.J., Cocherie, A., Legendre, A., 2007. Potential minerals for determin 2004. SHRIMP U-Pb dating of diagenetic xenotime in the ing U-Th-Pb chem i cal age us i ng electron microprobe. Lithos, Stirling Range Formation, Western Austra I ia: 1.8 bill ion year 93: 288-309. minimum age for the Stirling biota. Precambrian Research, 133: Fletcher, I.R., Rasmussen, B., McNaughton, N.J., 2000. SHRIMP 3 2 9-337. U-Pb geochronology of authigenic xenotime and its potential for Rasmussen, B., Fletcher, I.R., Muhling, J.R., 2011. Response of dating sedimentary basins. Australian Journal of Earth Sci xenotime to prograde metamorphism. Contributions to Mineral ences, 47: 84 5 -859. ogy and Petrology, 162: 1259-1277. Förster, H.J., 1998. The chemical composition of REE-Y-Th-U-rich Seydoux-Guillaume, A.M., Paquette, J.L., Wiedenbeck, M., accessory minerals in peraluminous granites of the Montel, J.M., Heinrich, W., 2002. Experimental resetting of the Erzgebirge-Fichtelgebirge region, Germany. Part II: Xenotime. U-Th-Pb systems in monazite. Chemical Geology, 191: American Mineralogist, 83: 1302-1315. 165-181. Gardes, E., Jaoul, O., Montel, J., Seydoux-Guillaume, A.M., Suzuki, K., Kato, T., 2008. CHIME dati ng of monazite, xenotime, Wirth, R., 2006. Pb diffusion in monazite: an experimental study zircon and polycrase: Protocol, pitfalls and chemical criterion of of Pb2* * Th4* - 2Nd3* interdiffusion. Geochimica et Cosmochi possibly discordant age data. Gondwana Research, 14: mica Acta, 70: 2325-2336. 569-586. Harlov, D.E., Wirth, R., 2012. Experimental incorporation of Th into Teufel, S., Heinrich, W., 1997. Partial resetting of the U-Pb isotope xenotime at middle to lower crustal P-T util iz i ng alkali-beari ng system in monazite through hydrothermal experiments: an SEM fluids. American Mineralogist, 97: 641-652. and U-Pb isotope study. Chem i cal Geology, 137: 273-281. Harlov, D.E., Wirth, R., Hetherington, C.J., 2011. Fluid-mediated Tropper, P., Manning, C.E., Harlov, D.E., 2011. Solubility of CePO4 partial alteration in monazite: the role of coupled dissolu monazite and YPO4 xenotime in H2O and H2O-N aC l at 800 °C tion-reprecipitation in element redistribution and mass transfer. and 1 GPa: Implications for REE and Y transport during Contributions to Mineralogy and Petrology, 162: 329-348. high-grade metamorphism. Chemical Geology, 282: 58-66. Hetherington, C.J., Harlov, D.E., 2008. Metasomatic thorite and Vilalva, F.C.J., Vlach, S.R.F., 2010. Major- and trace-element com uraninite inclusion in xenotime and monazite from granitic position of REE-rich turkestanite from peralkaline granites of the pegmatites, Hidra anorthosite massif, southwestern Norway: Morro Redondo Complex, Graciosa Province, south Brasil. Min Mechanics and fluid chemistry. American Mineralogist, 93: eralogical Magazine, 74: 645-658. 8 06-820. Williams, M.L., Jercinovic, M.J., Harlov, D.E., Budzyń, B., Hetherington, C.J., Jercinovic, M.J., Williams, M.L., Mahan, K., Hetherington, C.J., 2011. Resetting monazite ages during 2008. Understanding geologic processes with xenotime: com fluid-related alteration. Chemical Geology, 283: 218-225.