the relative stability of monazite and huttonite at 300 900 c and 200 1000 mpa: metasomatism and...

American Mineralogist, Volume 92, pages 1652�1664, 2007

0003-004X/07/0010�1652$05.00/DOI: 10.2138/am.2007.2459 1652

INTRODUCTION

Huttonite (ThSiO4) is a relatively rare member of the monoclinic monazite group of minerals, which also include scarce cheralite [CaTh(PO4)2] and ubiquitous monazite [(Ce,La,Nd)PO4] (Förster 1998). Huttonite has a lower P-T, tetragonal dimorph, thorite, which is considerably more com-mon. At 1 bar, the experimental, unreversed phase transition between thorite and huttonite is located between 1210 and 1225 °C (Finch et al. 1964; Seydoux and Montel 1997). At higher pressures, huttonite has been shown experimentally to be stable over a range of temperatures normally encountered in the lower crust to upper mantle (Fig. 1; Dachille and Roy 1964; Seydoux and Montel 1997; Mazeina et al. 2005). It is obvious from Figure 1 that the location of the thorite-huttonite equilibrium in P-T space remains uncertain, whether via the half reversals of Seydoux and Montel (1997) or through transposed temperature drop calorimetry with its large error bars (Mazeina et al. 2005). This uncertainty is apparently due, at least in part, to the sluggish reaction kinetics of the thorite-huttonite transition, which appears to result from the formation of an extra Th-O bond in huttonite compared to thorite (Finch and Hanchar 2003). The P-T stability Þ eld for huttonite, as outlined in Figure 1, does not agree with the P-T conditions presumed present during its formation in its three conÞ rmed occurrences. These include (1) its type-locality,

Gillespie�s Beach, southern Westland, South Island, New Zealand (Pabst and Hutton 1951; Förster et al. 2000), where it occurs as minute grains in the beach sands; (2) granitic pegmatites from Bogatynia, Poland (Kucha 1980), where the huttonite component in the monazite approaches 66 mol%; and (3) nepheline syenites from Brevik, Norway (Meldrum et al. 1999). For example, the suggested source for type huttonite from Gillespie�s Beach, are the local Otago schists (Förster et al. 2000). These schists were metamorphosed under amphibolite-facies conditions (500 °C, 700 MPa; Yardley 1982) indicating that they experienced P-T conditions well outside the experimentally calibrated stability Þ eld for huttonite (Fig. 1). Similar P-T regimes hold for the other two huttonite occurrences as well. Speer (1982) suggested that admixtures of other elements, such as the (Y + REE), may extend the stability Þ eld of huttonite to lower temperatures, which could be the case for the huttonite-enriched monazite from the Bogatynia, Poland pegmatites. Although, in the case of huttonite from Gillespie�s Beach, the (Y + REE) contents are too low to account for a possible lowering of its P-T stability Þ eld (Förster et al. 2000). However, such a speculation would suggest that instead of direct high-temperature formation, end-member huttonite might be indirectly approached via substitution of the monazite by an increasing ThSiO4 fraction at considerably lower temperatures. This interpretation assumes complete or extensive miscibility between monazite and huttonite under the nominal metamorphic pressures and temperatures present in the mid to lower crust. Extensive miscibility has been documented both * E-mail: [email protected]

The relative stability of monazite and huttonite at 300�900 °C and 200�1000 MPa: Metasomatism and the propagation of metastable mineral phases

DANIEL E. HARLOV,1,* RICHARD WIRTH,1 AND CALLUM J. HETHERINGTON2

1GeoForschungsZentrum Potsdam, Telegrafenberg, D-14473 Potsdam, FR Germany2Department of Geosciences, University of Massachusetts, 611 North Pleasant Street, Amherst, Massachusetts 01003-9297, U.S.A.

ABSTRACT

Monazite is both partially replaced and overgrown by a ThSiO4 phase along grain rims in a series of experiments from 300 to 900 °C and 200 to 1000 MPa. All experiments consisted of 10 mg of 100�500 μm size, euhedral to subhedral crystals of a natural Th-free monazite-(Ce), 5 mg of Th(NO3)4·5H2O, 2.5 mg of SiO2, and 5 mg of H2O loaded into 3 mm wide, 1 or 1.3 cm long platinum capsules that were arc welded shut. Experimental conditions were: 300 °C at 200 and 500 MPa; 300, 400, 500, 600, and 700 °C at 500 MPa (cold seal hydrothermal autoclave); and 900 °C at 1000 MPa (Catz assembly; piston-cylinder press). Back-scattered electron (BSE) imaging, electron back-scattered diffraction (EBSD) analysis, and transmission electron microscopy (TEM) indicates that in the experiments from 500 to 900 °C, the ThSiO4 phase took the form of monoclinic huttonite implying that huttonite, associated with monazite, could exist metastably over a much greater P-T range than previously thought. TEM analysis of a foil cut perpendicular to the monazite-huttonite interface from the 600 °C, 500 MPa experiment using a focused ion beam (FIB) indicates that the huttonite as well as the interface between the huttonite and monazite is characterized by ß uid inclusions. High-resolution TEM analysis indicates that the huttonite-monazite interface is coherent. In the case of replacement of monazite by huttonite, ß uid-aided dissolution-reprecipitation is proposed as the most likely mechanism responsible.

Keywords: Monazite, huttonite, dissolution-reprecipitation, metastability, metasomatism, EBSD, TEM, pseudomorphism

HARLOV ET AL.: THE RELATIVE STABILITY OF MONAZITE AND HUTTONITE 1653

in natural monazite, which can contain up to 30 mol% of the huttonite component (Förster and Harlov 1999; Meldrum et al. 1999), as well as in experimentally derived monazite where inclu-sions and rim grains derived from the metasomatized Durango ß uorapatite contained up to 38 mol% of the huttonite component (Harlov and Förster 2003). This extensive miscibility between monazite and huttonite has also been documented experimentally by Seydoux-Guillaume et al. (2002b).

The goal of this study was to investigate replacement and overgrowth of monazite by huttonite from 300 to 900 °C and 200 to 1000 MPa via a series of simple experiments involving natural monazite and a Th-Si-rich solution. The results from these experiments are then documented and interpreted on the micrometer and nanometer scale both visually and chemically by utilizing backscattered electron (BSE) microscopy, electron back-scattered diffraction (EBSD), electron microprobe (EMP) analysis, and transmission electron microscopy (TEM).

EXPERIMENTAL AND ANALYTICAL METHODS

Monazite-ß uid experiments The monazite-(Ce) used in all of the experiments consisted of clear, optically

bright olive-green, inclusion-free, 100 to 500 μm size, euhedral to subhedral crystals from a late-stage ferroan dolomitic carbonatite located as part of the Kangankunde carbonatite complex in the Chilwa Alkaline Province, southern Malawi (Garson and Campbell-Smith 1965; Holt 1965; Garson 1966; Woolley 1991; Wall and Mariano 1996). The monazite-(Ce) contains Th, U, and HREE amounts below the EMP detection limit (representative EMP analysis in Table 1). It is strongly enriched in LREE, with a (La/Y)CN ratio of about 1500, (CN = chondrite normalized). It contains variable amounts of SrO averaging around 1.8 wt%. Such features are typical of carbonatite monazites (Wall and Mariano 1996). An unusual feature of these crystals is that they are sector zoned with the degree of LREE-enrichment and Sr varying between the sectors (Cressey et al. 1999). Here, variations in La2O3 (up to 6.0 wt%) and Nd2O3 (up to 3.9 wt%) can occur between sectors. In contrast, Ce2O3 is nearly constant.

All experiments consisted of 10 mg of monazite-(Ce) crystals, 5 mg of Th(NO3)4·5H2O, 2.5 mg of SiO2, and 5 mg of H2O loaded into 3 mm wide, 1 or 1.3 cm long platinum capsules. The amount of Th(NO3)4·5H2O and SiO2 added was calculated on a molar basis to be the equivalent of replacing roughly 1/3 of the monazite with pure ThSiO4. After loading, the platinum capsule was pinched

and then arc-welded shut using an argon torch while the capsule was partially immersed in an ice water bath. Three sets of experiments were performed (300 °C at 200 MPa; 300, 400, 500, 600, and 700 °C at 500 MPa; and 900 °C at 1000 MPa) (Table 1). The experiment at 900 °C, 1000 MPa (AM54) was done using CaF2 assemblies corrected for friction (Harlov and Milke 2002) in a two-piston-cylinder apparatus as described by Johannes et al. (1971) and Johannes (1973). The platinum capsule was placed vertically within the assembly such that the NiCr thermocouple tip was halfway up along the side of the capsule without touching it. Thermal gradients are estimated to be within 20 °C. Uncertainty in the pressure is estimated to be ±25 MPa (Harlov and Milke 2002). The experiment was left up for 8 days and then quenched by turning off the current and allowing the H2O-cooled jacket to cool the vessel. This resulted in temperatures of less than 50 °C being reached within 15 s of quench. Experiments at 300 °C, 200 and 500 MPa (AM55 and AM72); 400 °C, 500 MPa (AM68); 500 °C, 500 MPa (AM69); 600 °C, 500 MPa (AM52 and AM70); and 700 °C, 500 MPa (AM71) (Table 1) were done using a standard cold-seal autoclave in conjunction with the hydrothermal apparatus. The internal thermocouple was placed such that the tip was half way up along the platinum capsule. After the run, the autoclave was quenched using compressed air. Temperatures of 100 °C were generally reached within 1 min. After each experiment, the platinum capsule was cleaned, weighed, and punctured. The ß uid in the capsule was tested using pH paper and found to have a pH somewhere between 1 and 2. The punctured capsule was then dried at 105 °C for 6�12 h and weighed again to determine the ß uid content. The monazite grains were removed, mounted in epoxy grain mounts, and then polished.

Mineral analysisMicroscopic investigation was done using back-scattered electron (BSE)

imaging. BSE pictures were made on a Zeiss DSM 962 digital scanning electron microscope with 20 kV acceleration voltages.

Electron microprobe (EMP) analysis of the monazite and ThSiO4 rims were made using the CAMECA SX50 and SX100 electron microprobes at the Geo-ForschungsZentrum Potsdam. EMP operating conditions, analysis technique, and standards are described in Harlov and Förster (2002). These included a 20 kV accelerating potential, 50 nA beam current, 1 μm diameter electron beam. Elements analyzed for (including the speciÞ c spectral lines used) were P (Kα), Si (Kα), Al (Kα), Th (Mα), U (Mβ), Y (Lα), La (Lα), Ce (Lα), Pr (Lβ), Nd (Lβ), Sm (Lβ), Gd (Lβ), Tb (Lβ), Dy (Lβ), Ho (Lβ), Er (Lβ), Yb (Lα), Lu (Lα), Ca (Kα), Sr (Kα), Fe (Kα), and Pb (Mβ). Primary standards included pure metals for Th and U, vanadinite for Pb, synthetic REE phosphates prepared by Jarosewich and Boatner (1991), and natural minerals, such as the Durango ß uorapatite, and synthetic oxides for the other elements. The analytical errors for the (Y + REE) depend on the absolute abundances of each element. Relative errors are estimated to be <1% at the >10 wt% level, 5−10% at the 1 wt% level, 10−20% at the 0.2�1 wt% level, and 20−40% at the <0.1 wt% level. For concentrations below 0.1 wt%, the analytical precision for the actinides is much higher, i.e., approximately 10%. Detection limits were approximately 200−300 ppm for all elements monitored. To ensure that the relative differences observed are correct within EMP error, both silicate and phosphate analyses per sample were obtained during one single measurement session including multiple checks of the calibration. Mean EMP analyses of the original monazite-(Ce) and the ThSiO4 rims are given in Table 2. An EMP traverse, in 3 μm increments across the monazite-huttonite interface in experiment AM52 (600 °C, 500 MPa), was made for Th and Si (oxide wt%) using the same EMP conditions as described above.

Electron back scattered diffraction (EBSD) analysisOverviews with regard to the theory behind EBSD and its application are given

in Loretto (1994) and Prior et al. (1999). In essence, back-scattered electrons from a divergent source, incident on crystal planes (in a tilted sample) at the Bragg angle, are diffracted into a pair of cones to form Kikuchi bands imaged on a phosphor screen (e.g., Nishikawa and Kikuchi 1928). The band orientation is a function of atom identity, atom position within the lattice, and overall crystal symmetry. Assynthetic Kikuchi bands for best-Þ t modeling are dependent on the assumed crystal structure (Wright and Adams 1992; Adams et al. 1993).

EBSD analysis of the ThSiO4 phases in experiments AM55, AM68, AM69, AM70, AM52, AM71, AM72, and AM54 were carried out on a Zeiss DSM 960 SEM located at Amherst College, Amherst, Massachusetts. This SEM is equipped with a tungsten Þ lament and secondary, forescatter, and backscatter electron detectors for examination of samples attached to a 70° inclined viewing stage and aligned perpendicular to the phosphorus screen of the EBSD detector. Integrated with the SEM is a range of analytical software including a Channel 5 EBSD controller

0

400

800

1200

1600

2000

2400

2800

3200

3600

300 500 700 900 1100 1300

Temperature (°C)

P (

MP

a) Huttonite

Thorite

Seydoux and Montel (1997)

Mazeina et al. (2005)

FIGURE 1. Diagram showing the experimentally determined stability Þ elds for huttonite and thorite (Seydoux and Montel 1997) and calculated stability huttonite and thorite stability Þ eld based on the calorimetric data of Mazeina et al. (2005).


and the entire suite of HKL Channel 5 EBSD data acquisition and processing software for live-time capture of EBSD patterns, pattern solution, and orientation contrast/phase mapping.

Before being placed in the SEM, all samples were prepared via chemical-mechanical means to achieve the highly polished ß at surface required for EBSD analysis (Fynn and Powell 1979). This included Þ rst polishing with 0.35 μm alu-mina powder and then polishing for at least 2 h with colloidal silica. The samples were then coated with a 40�60 Å thick layer of carbon. SEM operating conditions included an accelerating voltage of 20 kV and a beam current of ~80 μA focused to a spot of ~2 to 3 μm diameter. The working distance was between 20 and 25 mm. Stage and image control processes were controlled via PGT Spirit Software. Diffraction pattern collection, evaluation, and indexing were completed using HKL Flamenco software linked to MSA�s American Mineralogist Crystal Structure Database (www.minsocam.org/MSA/Crystal_Database.html).

Transmission electron microscopy (TEM)Examples of ThSiO4 rims surrounding the experimentally reacted monazite-

(Ce), suitable for TEM investigation, were Þ rst selected using BSE and SE photos. Specimen preparation was accomplished by using focused ion beam (FIB) milling (Wirth 2004). FIB preparation was conducted under ultrahigh-vacuum conditions in an oil-free vacuum system in a FEI FIB200 instrument at the GeoForschun-gsZentrum Potsdam. TEM-ready foils of approximately 15 × 10 × 0.150 μm representing cross sections across the monazite-ThSiO4 interface, were cut directly from the reacted monazite grain in the epoxy grain mount by means of a 30 kV Ga ion beam. The TEM foil was protected from abrasion by the Ga ion beam via a 1 μm thick Pt layer deposited using a high-purity organic Pt gas (C9H16Pt, 99.9%), which decomposes in the Ga ion beam. Once cut, each TEM foil was placed on

TABLE 1. Experimental results

Experiment P (MPa) T (°C) Time (d) Monazite (mg) Th(NO3)4·5H2O (mg) SiO2 (mg) H2O* (mg) Remarks

AM55 200 300 71 10.13 4.83 2.64 7.40 ThSiO4 as thorite formed as relatively large masses of small grains (<1 μm) both surrounding a small subset of the monazite grains as well as occuring as masses independent of the monazite grains. AM72 500 300 20 10.43 5.41 3.19 7.91 ThSiO4 as thorite formed as large masses of small grains (<1 μm) of thorite both surrounding a small subset of the monazite grains as well as occuring independent of the monazite grains. A few small blebs of tightly packed microcrystalline ThSiO4 masses are seen growing along monazite grain rims. AM68 500 400 24 10.26 5.64 2.71 8.14 ThSiO4 as huttonite formed as very thin (1–5 μm) rims along a small subset of monazite grain rims as well as monazite grain cracks. The edges of these rims tend to have a granular look as though composed of numerous small (<1 μm) crystals. AM69 500 500 24 10.41 4.41 1.93 7.35 ThSiO4 as huttonite formed as thin to moderate (5–10 μm) continuous rims along a small subset of monazite grain rims. AM52 500 600 28 10.17 5.22 2.09 8.75 ThSiO4 as huttonite formed as thin to moderate (5–50 μm) continuous rims along a small subset of monazite grain rims. Some monazite grains entirely replaced by ThSiO4. AM70 500 600 24 10.41 5.49 2.57 7.50 ThSiO4 as huttonite formed as thin to moderate (5–50 μm) continuous rims along a small subset of monazite grain rims. Some monazite grains entirely replaced by ThSiO4.AM71 500 700 9 10.57 5.25 2.60 8.38 ThSiO4 as huttonite formed as thin to moderate (5–50 μm) continuous rims along a small subset of monazite grain rims. AM54 1000 900 8 10.77 4.88 2.60 8.07 ThSiO4 as huttonite formed as thin to moderate (5–50 μm) continuous rims along a small subset of monazite grain rims.

* The H2O content is the total amount of H2O during the experiment and, therefore, includes the H2O contributed by the Th(NO3)4·5H2O.

TABLE 2. Mean monazite-(Ce) and ThSiO4 rim compositions (wt%)

Monazite-(Ce)* AM68 AM69 AM52 AM70 AM71 AM54

T (°C) 400 500 600 600 700 900P (MPa) 500 500 500 500 500 1000Time (days) 24 24 28 24 9 8No. analyses 24 2 3 17 4 8 2P2O5 29.41 0.68 0.20 0.18 0.04 0.06 SiO2 0.10 17.79 18.19 18.35 18.51 18.55 18.71ThO2 0.03 79.72 80.62 81.16 81.56 81.61 81.67La2O3 19.73 0.37 0.17 0.13 0.09 Ce2O3 35.21 0.91 0.28 0.25 0.03 0.23 Pr2O3 3.04 0.07 0.03 0.03 0.03 Nd2O3 9.61 0.37 0.11 0.09 0.03 0.10 Sm2O3 0.61 0.06Gd2O3 0.19CaO 0.21SrO 1.31 Total 99.45 99.95 99.61 100.19 100.16 100.68 100.38P 0.987 0.031 0.009 0.008 0.002 0.003 Si 0.004 0.957 0.984 0.988 0.996 0.993 1.003Th 0.976 0.993 0.994 0.999 0.994 0.997La 0.288 0.007 0.003 0.003 0.002 Ce 0.511 0.018 0.006 0.005 0.001 0.005 Pr 0.044 0.001 0.001 0.001 0.001 Nd 0.136 0.007 0.002 0.002 0.001 0.002 Sm 0.008 0.001Gd 0.002Ca 0.009Sr 0.030

Note: blank = below EMP detection limit.* Mean analysis value for natural monazite from southern Malawi used in the experiments, which averages over the sector zoning.


a perforated carbon Þ lm atop a copper grid. Carbon coating to prevent charging in the TEM was not applied.

TEM was carried out in a Philips CM200 instrument operated at 200 kV and equipped with a LaB6 electron source. Electron energy loss spectra (EELS) were acquired with a Gatan imaging Þ lter. Chemi-cal composition was measured by using an EDX spectrometer. Analyses were corrected for absorption and ß uorescence.

RESULTS

Experimental resultsIn all experiments, a ThSiO4

phase (or phases) formed in asso-ciation with the monazite (Table 1; Fig. 2). In the 300 °C experiments at 200 MPa (AM55) and 500 MPa (AM72), the ThSiO4 phase took the form of a Þ ne-grained, polycrystalline mass made up of randomly oriented crystals less than 1 μm in size partially en-closing a subset of the monazite grains in addition to growing along apparent cleavage planes or cracks in the monazite (Fig. 2a). In the 400 to 700 °C experiments at 500 MPa and in the 900 °C ex-periment at 1000 MPa, a ThSiO4 phase partially replaced and/or overgrew the monazite along the grain rims. This replacement oc-curred for 10�20% of the monazite grains (Table 1; Figs. 2b�2h). BSE imaging and EMP traverses of these experiments indicate that the chemical boundary between the two phases is compositionally sharp on the micrometer scale (Fig. 3). In the 600 °C, 500 MPa experiments (AM52 and AM70), a few of the smaller monazite grains were totally replaced by this ThSiO4 phase as a pseudomorph while retaining the apparent mono-clinic symmetry of the original monazite grain (Fig. 4). Partial replacement and/or overgrowth of monazite by a ThSiO4 phase along the grain rim is relatively more common in the 400, 500, 600, and 700 °C experiments at 500 MPa (Figs. 2b�2f) and less com-mon in the experiment at 900 °C and 1000 MPa (AM54) (Fig. 2g). In the 900 °C experiments, such overgrowths are rare and, when

FIGURE 2. BSE photos of experimentally metasomatized monazite at 300 °C, 200 MPa, AM55 (a); 400 °C; 500 MPa, AM68 (b); close-up of a multi-crystalline ThSiO4 rim on monazite from experiment AM68 (c); 500 °C, 500 MPa, AM69 (d); 600 °C, 500 MPa, AM52 (e); 700 °C, 500 MPa, AM71 (f); 900 °C, 1000 MPa, AM54 (g); and 600 °C, 500 MPa, AM52 (h). The white areas are ThSiO4 and the dark gray areas are monazite. Small 1�10 μm size voids are circled in (e�g). The location and direction of the EMP traverse in (e) (cf. Fig. 3) is indicated by black arrow. See text for further explanation.


they do occur, are more limited in scope. Instead, in AM54, the majority of the Th and SiO2 have reacted to form small (5�10 μm), isolated, euhedral, tetragonal grains of thorite. In general, for each of the other experiments at lower temperatures, a por-tion of the Th4+ and SiO4

4� in solution crystallized out as discrete crystals of thorite. In all cases, these crystals were conÞ rmed to be thorite from their tetragonal symmetry (where obvious), EMP analysis, EBSD patterns (see below), and, more indirectly, from the P-T constraints under which these experiments were performed (Table 1; Fig. 1).

Small voids on the 1�10 μm scale appear to have formed in the monazite interior as well as along the ThSiO4-monazite interface in a subset of the monazite grains from the 600�900 °C experiments (Fig. 2). Originally, such voids may have been ß uid inclusions and/or possibly could have formed from dam-aged areas in the monazite due to preferential dissolution during the experiment. These are especially evident in Figures 2e, 2f, and 2g. In some metasomatized monazite grains (e.g., AM52 at 600 °C and 500 MPa), the ThSiO4 phase appears to have grown along pre-experimental cracks (or possible cleavage planes?) in the monazite, such that isolated islands of monazite have

resulted that are surrounded by the ThSiO4 phase (e.g., Fig. 2h). Subsequent ThSiO4-free cracks in the monazite appear to have formed during quenching and/or during the mounting and polishing process because they are continuous from the monazite into the ThSiO4 phase. In the 300 °C experiments, the Þ ne-grained granular mass of ThSiO4 crystals in AM55 and AM72 (Fig. 2a) were not conducive to good EMP analyses, although electron diffraction X-ray (EDX) spectra did conÞ rm that their composition was a ThSiO4 phase. Comparing the EMP analyses of the ThSiO4 phases from AM68, AM69, AM52, AM70, AM71, and AM54 indicates that within analytical uncertainty, AM54, at 900 °C and 1000 MPa, is pure ThSiO4 as compared to the lower temperature experiments in which the ThSiO4 phase still contains apparent traces of P, La, Ce, Pr, Nd, and Sm, whose general abundances appears to show a vague dependence with respect to temperature (cf. Table 2).

Electron back-scattered diffraction (EBSD) investigation of the experiments

Five to six monazite grains with well-deÞ ned ThSiO4 rims were examined from experiments AM69 (500 °C; 500 MPa), AM52 and AM70 (600 °C; 500 MPa), AM71 (700 °C; 500 MPa), as well as AM54 (900 °C; 1000 MPa) (Figs. 2d�2h). The central portions of each grain, i.e., the original monazite, gave strong EBSD patterns with 7 or 8 identiÞ able Kikuchi bands which, when indexed by the software using MSA�s American Mineralogist Crystal Stucture Database, indicated a monoclinic pattern utilizing a minimum of 6 bands (Fig. 5). The match to

FIGURE 3. EMP traverse across the huttonite-monazite interface for the traverse shown in Figure 2e. The dotted line indicates the approximate boundary between the huttonite rim and original monazite core. The arrow indicates the direction of the EMP traverse. See text for further explanation.

TABLE 3. Huttonite and monazite unit cell parameters

Huttonite Monoclinic; P21/nThis study Taylor and Ewing (1978)ao = 6.78 Å ao = 6.8 Åbo = 6.97 Å bo = 6.96 Åco = 6.50 Å co = 6.54 Å β = 104.92° β = 104.99°

Monazite Monoclinic; P21/nThis study Ni et al. (1995)ao = 6.81 Å ao = 6.79bo = 7.04 Å bo = 7.01co = 6.50 Å co = 6.46β = 103.54° β = 103.63°

FIGURE 4. BSE photos of experimentally metasomatized monazite at 600 °C showing examples of a monazite grains that have been nearly totally replaced by huttonite (a) and a monazite grain completely replaced by huttonite (b) with small inclusions of ThO2.


a standard monazite pattern indicated a mean angular deviation (MAD) value of <0.8 in which the lower the MAD value, the greater the statistical certainty of the match. A minimum of three points was measured for each ThSiO4 rim (Fig. 5d) as well as for several grains either nearly or totally replaced by ThSiO4 (Fig. 4). Only measurements with 6 or more well-deÞ ned Kikuchi bands were used. Each of these measurements indicated a MAD value <1.25 using the unreacted monazite as a proxy reference standard. As a consequence, coupled with EMP data, each of the ThSiO4 rims in these experiments was positively identiÞ ed to have a monoclinic structure and, therefore, to be huttonite as opposed to tetragonal thorite. The slightly higher MAD values are to be expected when monoclinic huttonite is matched to the index of a stoichiometric ideal monazite. This is because of the slight reduction in the unit-cell lattice parameters due to the incorporation of Th and Si (Table 3). In all Þ ve experiments, the collected Kikuchi band diffraction patterns show near perfect crystallographic continuity between the monazite core and the huttonite rim (compare Figs. 5a and 5b). The very small devia-tion observed is presumed to be caused by a slight shift in the unit-cell parameters (Table 3).

For experiments AM68 (400 °C, 500 MPa), AM72 (300 °C, 500 MPa), and AM55 (300 °C, 200 MPa) (Figs. 2a�2c), the original monazite also gave strong EBSD patterns that resulted

in a good match to the monazite standard. However, obtaining distinct diffraction patterns from the ThSiO4 regions proved problematic. For experiment AM68, there were only 5 grains with ThSiO4 rims wide enough to ensure conÞ dence that EBSD was restricted to the ThSiO4 zone with no contribution from the neighboring monazite. For each of these rims, the EBSD pattern was either very poor or undetectable. Subsequently, any attempt to index the 2�3 Kikuchi bands actually discerned was not possible because the results would be statistically meaning-less. The most likely explanation in this case is that the limited EBSD patterns observed in AM68 indicate that the ThSiO4 rims consist of a micro-crystalline or even a nano-crystalline aggregate of randomly distributed grains or could be mosaic crystals, i.e., single crystals with slightly misoriented segments. High-mag-niÞ cation BSE photographs of these rims show some evidence for this micro-crystallinity (e.g., examine the thin, apparent multicrystalline, ThSiO4 rim in Fig. 2c). Such randomness would make it difÞ cult to generate good EBSD patterns.

For experiments AM72 and AM55, it was not possible to obtain EBSD patterns from any of the ThSiO4 grain aggregates (Fig. 2a). This was despite the fact that each of these samples was polished multiple times to reduce topographic effects and, as such, to ensure that any absence of an EBSD pattern was not a surface effect. The identity of this independent ThSiO4 phase as

FIGURE 5. Kikuchi band images of the monazite core (a) and the huttonite rim (b) along with the indexed Kikuchi band pattern (c) corresponding to analysis points a and b on the BSE image of a monazite grain with a well-developed huttonite rim (d) from experiment AM52 (600 °C, 500 MPa).


thorite was not directly determined but rather inferred because it formed independently of the monazite within the thorite stability Þ eld as deÞ ned in Figure 1.

TEM investigation of experiment AM52Three TEM foils were extracted using FIB from a variety

of locations relative to the ThSiO4 rims on monazite in experi-ment AM52 (600 °C; 500 MPa) (Figs. 6�10). These included TEM foil 1 that was cut perpendicular to the ThSiO4-monazite interface (Figs. 6 and 7), TEM foil 2 cut in monazite in the im-mediate vicinity of a ThSiO4 rim such that it included a small portion of the ThSiO4 area (Figs. 8b and 9a), as well as TEM foil 3 cut directly in a ThSiO4 rim parallel to the grain edge (Figs. 8c and 9b). Electron diffraction analysis of the ThSiO4 in these foils indicated that it has a monoclinic symmetry (huttonite) as opposed to a tetragonal one (thorite) thus conÞ rming the EBSD analysis.

In TEM foil 1, the interface between the huttonite and the monazite is quite sharp and is highlighted by numerous, small voids. These voids are interpreted as ß uid inclusions that were opened during the FIB milling process (Figs. 7a and 7b). Whereas the monazite appears to be inclusion-free, the huttonite contains both these ß uid inclusions as well as dark inclusions of ThO2. In many cases, the dark inclusions are associated with a partial void (Fig. 7b). In addition, the ß uid inclusion density appears to show a vague increase as the huttonite-monazite interface is approached. In TEM foil 2 (Figs. 8a and 8b), the immediate vicinity of the sharp huttonite-monazite interface is characterized by huttonite, with associated ß uid inclusions, growing along fractures and/or possible cleavage planes in the monazite (Fig. 9a). In some places, the huttonite veins appear to completely surround the monazite essentially isolating it from the rest of the crystal. Similarly in TEM foil 3 (Figs. 8a and 8c), huttonite is seen growing in a crack in the monazite in the vicinity of the monazite-huttonite interface (Fig. 9b). Also, a monazite remnant is seen deep within the huttonite region.

High-resolution TEM imaging indicates that at the interface, the lattice fringes from the huttonite and monazite appear to be continuous with little or no mismatch implying that the interface between the huttonite and the monazite is coherent (Fig. 10). This is conÞ rmed by the electron diffraction pattern taken over the interface (see inset in Fig. 10).

FIGURE 6. BSE photos of experimentally metasomatised monazite with a huttonite rim (600 °C, 500 MPa, AM52) showing the location of the FIB cut. The exact location of the foil is between the two spots marked with an �x� in (b).

FIGURE 7. Full extent of the FIB cut TEM foil (cf. Fig. 6) is shown in (a) with a close up of the interface shown in (b). Large, elongate circles reß ect holes in the carbon substrate upon which the foil rests. The dark band across the top of a is a portion of the 1 μm thick Pt layer. Note the sharp interface between the monazite and huttonite, the numerous ß uid inclusions (now bright empty voids) in the huttonite, and dark inclusions, commonly associated with the ß uid inclusions. The dark inclusions represent high concentrations of more electron-absorbing Th, in the form of ThO2, which were included during the growth of the huttonite rim.


DISCUSSION

Nucleation and growth of the huttonite rimsThere is no experimental evidence that solid-state volume

diffusion of Th and Si cations in monazite could account for the breadth and depth of the huttonite rims documented in Figures 2 and 6�9, the sharp compositional interface between the monazite and huttonite (e.g., Figs. 2, 3, and 7a), the apparent replacement of monazite by huttonite in these rims, or even the replacement of whole monazite grains by huttonite (Fig. 4) (see discussion in Teufel and Heinrich 1997; Crowley and Ghent 1999; Cher-niak 2000; Foster et al. 2002; Seydoux-Guillaume et al. 2002a; Cherniak et al. 2004; Krenn and Finger 2004; Gardés et al. 2006). For example, it has already been demonstrated that the rate of U and Th diffusion in orthophosphates, and monazite in particular, is negligible at temperatures up to 750 °C, even during hydrothermal experiments at low pressures (<500 MPa) (Teufel and Heinrich 1997; Seydoux-Guillaume et al. 2002a).

Formation of the huttonite rims must have taken place by two possible mechanisms, both of which would explain the lack of a mismatch between the huttonite and monazite lattice fringes (Fig. 10). Mechanism 1 involves nucleation and growth of hut-tonite on those monazite crystal faces with the most favorable crystallographic orientation. Such a scenario is facilitated by the strong similarity of the monazite and huttonite lattice param-eters (Table 3), thus favoring epitactic growth of huttonite on a monazite substrate. Once the Þ rst atomic layer is laid down, the huttonite subsequently grows outward as an overgrowth with no replacement of the monazite grain. The presence of any porosity

FIGURE 8. BSE photos of experimentally metasomatized monazite with a huttonite rim (600 °C, 500 MPa, AM52) showing the locations of two FIB cuts (a). The location of each of the TEM foils is between the two spots marked with an �x� in b and c.

in the huttonite rim could then be explained as being due to the incorporation of ß uid droplets by an outwardly moving huttonite-ß uid interface. Mechanism 2 involves initial crystallization of huttonite on the monazite crystal face, but then growing inward via dissolution of the monazite with replacement by huttonite. Such a process must be, by deÞ nition, a coupled one because the rates of both the dissolution of the old mineral phase and the reprecipitation of the new mineral phase have to be equal so as to maintain contact between the reactants, products, and ß uids, allowing for the transportation of material across the reaction front (Putnis 2002). In such a process, the monazite is being replaced either partially or totally by a huttonite pseudomorph (e.g., Figs. 2h and 4).

As a process, dissolution-reprecipitation is deÞ ned as the means by which a mineral phase is replaced by either an altered version of the same mineral phase or by a new mineral phase under the prevailing P-T-X conditions (see deÞ nition and discus-sion in Putnis 2002 and Putnis et al. 2007; see also O�Neil 1977; Yanagisawa et al. 1999; Rendón-Angeles et al. 2000a, 2000b; Tomaschek et al. 2003; Putnis and Metzger 2004; Labotka et al. 2004; Geisler et al. 2005; Harlov et al. 2005). The dissolution-re-precipitation process operates essentially as a ß uid-aided chemi-cal reaction, involving a Gibbs free energy. The reactive volume is characterized by a pervasive ß uid-Þ lled porosity throughout the metasomatized region and a sharp compositional boundary between the newly formed and original mineral phase. This ß uid-Þ lled porosity allows for rapid ß uid-aided mass transfer to and from the reaction front at a rate approximately 10 orders of magnitude higher than simple volume diffusion through the crystal lattice (see discussion in Harlov et al. 2005). Supply of the huttonite-monazite interface with Th and Si from the sur-rounding solution and transport of P and (Y + REE) from the interface into solution could have occurred via the following general reaction:

(Ce, La, Nd, Pr)PO4 + Th(NO3)4 + H4SiO4 (1) Monazite-(Ce) (in solution) (in solution)

= ThSiO4 + (Ce, La, Nd, Pr)3+ + PO43� + 4 HNO3.

Huttonite rim (in solution) (in solution) (in solution) on Monazite

The fate of any excess PO43� and (Y + REE)3+ in solution most


likely plates out on unreacted monazite grains. Reaction 1 does not take into account the fact that a certain proportion of the Th4+ and SiO4

4� in solution appear to have crystallized out as grains of thorite. ConÞ rmation that dissolution-reprecipitation did occur, in conjunction with reaction 1, is seen both in the monazite grains partly replaced by huttonite (e.g., Figs. 2h and 9b), monazite grains nearly or totally replaced by huttonite (Fig. 4), as well as in the acidity of the solution at the end of the experiment. Ad-ditional conÞ rmation of reaction 1 is also seen in the relatively high solubility of Th in nitric acid (Schmidt et al. 2007).

More importantly, both mechanisms 1 and 2 occurs in a P-T region, i.e., 500, 600, and 700 °C at 500 MPa, and 900 °C at 1000 MPa, where independent thermochemical data suggests that huttonite should be thermodynamically unstable relative to thorite (Fig. 1). The persistence of huttonite as an apparently stable phase during the experiment (and in nature) could possibly be due to the sluggish transformation of huttonite to thorite at such low temperatures (Finch and Hanchar 2003). Whatever the case, the Gibbs free energy driving reaction 1 must be negative over this P-T range. Determination of this Gibbs free energy as

well as how it changes with changing P and T would be more difÞ cult because the data needed, i.e., the activities of each probable species in solution coupled with internally consistent standard-state thermodynamic data for monazite, huttonite, and thorite, are, at present, either not determined or not known with any accuracy over the P-T range of the experiments.

Mass transfer by diffusion through a ß uid to and from the huttonite-monazite interface must be accomplished through a ß uid-Þ lled interconnected porosity within the newly formed huttonite. Evidence for this porosity is seen in the abundant ß uid inclusions in the huttonite, especially at the monazite-huttonite interface (e.g., Figs. 7a and 9a). Similar dissolution-reprecipita-tion associated porosities or evidence that such porosities existed previously have been documented in metasomatized ß uorapatite (Harlov et al. 2005, their Figs. 9 and 12), in feldspar (Putnis et al. 2007), in tourmaline (Henry et al. 2002), as well as in a series of other metasomatized minerals (see review and discussion in Putnis 2002). Presumably once the dissolution-reprecipitation process ceased, such that the huttonite rim ceased to grow, the rim recrystallized, thereby destroying the interconnected porosity and leaving behind isolated ß uid inclusions (e.g., Fig. 7). Another possibility is that recrystallization of the huttonite rim and the subsequent destruction of this porosity halted the dissolution-reprecipitation process. Only a series of timed experiments, ranging from hours to days, that were carefully analyzed using a combination of BSE, EBSD, and EMP analysis culminated by a subsequent TEM investigation of foils sampled perpendicular to the monazite-huttonite interface using FIB, would be able to address this question.

In general, formation and maintenance of this nanoporosity

FIGURE 9. Section of the TEM foil from the FIB cut in Figure 8b (a). Note the veins of huttonite, with accompanying ß uid inclusions (bright holes), in the monazite. All other background elongate holes, lines, etc. come from the perforated carbon Þ lm on which the TEM foil rests. Section of the TEM foil from the FIB cut in Figure 8c (b). There are also remnant islands of monazite (circled area) whereas at the grain boundary, replacement of monazite by huttonite Þ rst occurs along cracks and veins in to the monazite.

FIGURE 10. High-resolution TEM image of the huttonite-monazite grain boundary interface. The approximate location of the interface is denoted by the black diagonal line. The electron diffraction pattern in the lower right-hand corner is taken from the whole of the image in the Þ gure and shows the overall coherency of the monazite and huttonite lattices.


can be explained by several mechanisms. These include differ-ences in the molar volumes between the two phases as well as differences in the solubility of the two phases in the common solution. In the Þ rst case, if the molar volume for the precipitated phase is smaller than the molar volume of the dissolving phase, a porosity will result in the precipitated phase (e.g., chlorapatite → ß uor-hydroxylapatite; Harlov et al. 2002; their Fig. 6c). In the present case, the difference in the molar volume between monazite (44.98 cm3 for CePO4; Smyth and McCormick 1995) and huttonite (44.58 cm3; Mazeina et al. 2005) is approximately 0.40 cm3 or a decrease in volume of 0.88%. Such a decrease is probably not enough to account for the total porosity required. Differences in the solubility of monazite and huttonite in the acidic solution present in the capsule must also have played a role. To impose porosity on the precipitated huttonite phase, more moles of monazite must have been dissolved than moles of huttonite precipitated such that reaction 1 was a molar deÞ cit reaction (see discussion in Putnis 2002). Such a reaction, coupled with a reduction in the molar volume from monazite to huttonite, would have been the mechanism behind the development of an interconnected porosity in the huttonite rims. If the opposite had occurred, i.e., the number of moles of monazite dissolved was less than the moles of huttonite precipitated, in combination with the small difference between the monazite and huttonite

molar volumes, then any sort of porosity forming in the newly precipitated huttonite would have been limited. Instead, the initial dissolution-reprecipitation process should have resulted in a solid, porosity-free layer of huttonite replacing perhaps the Þ rst few tens of nanometers of the monazite rim. This would have then armored the monazite against further dissolution. Any further expansion of the huttonite rim would have to occur solely as the result of direct precipitation of ThSiO4 from the solution resulting in overgrowth of the monazite by huttonite.

Support for the interpretation that the monazite grains have been partially replaced by huttonite is seen in Figures 2h, 4, and 9. For example in Figure 2h the huttonite is seen replacing the monazite in a manner that resulted in the subsequent broadening of cracks and/or possible cleavage planes in the monazite. In such a process, regions of monazite are thus isolated and subsequently replaced by huttonite, again as a ß uid-mediated process via an interconnected porosity in the huttonite. This form of replace-ment apparently can leave behind small, isolated remnants of the original monazite within the huttonite (e.g., Fig. 9b). Such isolated remnants would explain the traces of P and (Y + REE) seen in the EMP analysis of huttonite in the lower-temperature experiments (Table 2). At higher temperatures, e.g., AM54 at 900 °C, however, reaction kinetics are apparently fast enough such that the monazite has been totally replaced (Table 2). The

FIGURE 11. BSE photographs of monazite grains in a selection of granulite-facies metabasite samples from the Val Strona area of the Ivrea-Verbano zone, northern Italy (Harlov and Förster 2002) (a and b). Sample IV94-15 (a) comes from a location close to the clinopyroxene-rich transition zone between the granulite- and amphibolite-facies zones along the traverse. Sample IZ96-181 (b) comes from a metabasite layer approximately in the middle of the granulite-facies zone. Examples of two monazite grains from the clinopyroxene-rich transition region between granulite- and amphibolite-facies rocks along a traverse of lower Archean crust located in Tamil Nadu, southern India (Hansen and Harlov 2007) (c and d). In all the BSE pictures, the brighter the area the more enriched the monazite is in the ThSiO4-CaTh(PO4)2 or the huttonite-cheralite component.


paucity of huttonite rims replacing monazite at 900 °C and 1000 MPa or complete lack at 300 °C, coupled with the abundance of granular ThSiO4 as thorite, would imply that at these temperature and pressure extremes, reaction kinetics favored the formation of thorite directly from the available Th4+ and SiO4

4� in solution as opposed to the replacement (and overgrowth) of monazite by huttonite. At 300 °C, the reaction kinetics were probably too slow to allow the huttonite to crystallize out using the monazite as a substrate thereby allowing stable thorite to eventually form. At 900 °C and 1000 MPa, reactions kinetics were so quick that a large fraction of the Th4+ and SiO4

4� in solution were used up in the nucleation and formation of thorite crystals, almost immediately before the remainder had a chance to attack the monazite grains and subsequently form the few scattered huttonite rims that are seen. The indeterminate result at 400 °C and 500 MPa is more difÞ cult to explain. It suggests that the ThSiO4 rims that did form did so as overgrowths along the monazite grain rim in the form of a tightly packed microcrystalline mass, possibly of thorite or perhaps of both huttonite and thorite. Close examination of Figure 2c would seem to partially conÞ rm this idea. Whatever the case, the results of these experiments would suggest that there is a certain P-T region, apparently around 500�700 °C and 500 MPa, in which, aided by monazite as a substrate, growth of huttonite is optimized compared to the independent nucleation and growth of thorite.

Monazite-huttonite relationships in nature In nature, monazite grains commonly exhibit complex

textures, primarily due to partial enrichment or depletion in both ThSiO4 (as huttonite) and CaTh(PO4)2 (as cheralite). In monazite from igneous rocks, these textures take the form of magmatic-induced zoning (e.g., Harlov and Förster 2002, their Fig. 2f; Broska et al. 2000). In high-grade metamorphic rocks, monazite grains tend to have complex ThSiO4-CaTh(PO4)2 textures, completely unrelated to magmatic zoning (e.g., Watt and Harley 1993; Watt 1995; Fitzsimons et al. 1997; Zhu et al. 1997; Bingen and van Breemen 1998; Förster and Harlov 1999; Terry et al. 2000; Cocherie et al. 2005; Dahl et al. 2005; Gon-calves et al. 2005; Kohn et al. 2005; Pyle et al. 2005; Finger and Krenn 2007). These textures are generally interpreted as being due to the partial metasomatism of the monazite grain during some pre-, peak-, or post-peak metamorphic ß uid-related event (e.g., Bingen and van Breemen 1998; Harlov and Förster 2002). They generally take the form of curved lobate-like ThSiO4-CaTh(PO4)2-enriched or -depleted regions with sharp, cusp-like, compositional boundaries between the metasomatized region and the original monazite such as those seen in Figure 11. In many cases, these regions appear to have intergrown with each other as well as partially replace apparently older regions suggesting multiple events during the growth history of the monazite grain. The results of this study support the conclusions of several of the above workers that such textures are most likely the result of dis-solution-reprecipitation processes initiated and controlled by the surrounding pore ß uids. Whereas ß uids in the experiments, from this study, were characterized by a low pH, in the form of nitric acid, the pore ß uids responsible for these textures in nature are most likely characterized by a high pH because experimentally increasing the pH of a ß uid in general tends to increase the solu-

bility of Th (Oelkers and Poitrasson 2002; Schmidt et al. 2007). If ThSiO4-CaTh(PO4)2-enrichment or depletion in the monazite were primarily diffusion controlled, then the compositional boundary between the Th-enriched or Th-depleted region and the original monazite would be diffuse as opposed to being sharp such as that seen in each of the examples in Figure 11 as well as the examples referenced above. Dating these variably enriched regions in the monazite could then allow for the possible dating of various ß uid-related metamorphic events during the P-T his-tory of the rock (e.g., Williams et al. 2006).

The results of this study could explain why type huttonite apparently comes from rocks that were metamorphosed at tem-peratures and pressures at which thorite would be expected to be the stable ThSiO4 phase (Förster et al. 2000). Namely, type huttonite grains were originally monazite grains with a probable nominal ThSiO4-CaTh(PO4)2 component, which, at some point in their history, were metasomatized in a highly reactive, high-pH ß uid excessively enriched in Th with stochiometrically equivalent amounts of available silica. The result of this metasomatism is that the subsequent huttonite grains are essentially pseudomorphs of the original monazite grains. It also implies that natural occurrences of huttonite mistaken for thorite are probably much more com-mon than previously realized and that greater care must be taken in distinguishing these huttonite grains from thorite using EBSD or Raman spectroscopy or a combination of both techniques. This would be especially true for ThSiO4 grains associated with mona-zite. Why huttonite grains formed in this manner might persist over geologic time periods under P-T conditions more favorable to thorite could simply be due to the very sluggish transformation of huttonite to thorite (Finch and Hanchar 2003) under these rather moderate P-T conditions, e.g., 500 °C and 700 MPa in the case of the Otago schists for type huttonite (Yardley 1982). Such an explanation might hold in general for most of the natural occur-rences of huttonite found in the crust.

Further experimental work, building on this study, which involved replication of ThSiO4-CaTh(PO4)2-enriched regions in natural monazite, similar to the textures documented in Figure 11, could give additional insights into the chemistry and nature of the ß uid or partial melts responsible for their formation and, from a more global perspective, responsible for the metaso-matism of the host rock. Such experiments would also help to answer questions concerning Th mobility in ß uids and partial melts during metamorphism as well as metasomatically induced Th depletion or enrichment in other minerals such as zircon and ß uorapatite. Ultimately, these experiments could be expanded to include U mobility as well. A basic conclusion from this study is that the growth and propagation of a metastable mineral phase in a rock fabric does not necessarily reß ect the P-T conditions under which the rock formed or the subsequent P-T path it has experienced. Rather the textures and chemistry of an appar-ently metastable mineral must be carefully documented, both on a micrometer and nanometer scale, relative to all the other minerals in the rock before a chemical history, with or without ß uids or melts, may be inferred. In this respect, whether on the nanometer or kilometer scale, ß uid-rock interaction, mineral metastability, and mineral equilibria (or disequilibria) all come together to present a reality that may or may not reß ect the true chemical history of the rock.


ACKNOWLEDGMENTSWe thank Dieter Rhede and Oona Appelt for support with the electron micro-

probe. Helga Kemnitz is acknowledged for assistance with the SEM. Joe Pyle and Peter Crowley are thanked for their assistance in the initial phases of the EBSD measurement process. Reviews by Joe Pyle, Anne-Magali Seydoux-Guillaume, and Darrell Henry helped to improve the content and clarity of the paper.

REFERENCES CITEDAdams, B.L., Wright, S.L., and Kunze, K. (1993) Orientation imaging micros-

copy: The emergence of a new microscopy. Metallurgical Transactions, 24A, 819�831.

Bingen, B. and van Breemen, O. (1998) U-Pb monazite ages in amphibolite- to granulite-facies orthogneiss reß ect hydrous mineral breakdown reactions: Sveconorwegian Province of SW Norway. Contributions to Mineralogy and Petrology, 132, 336�353.

Broska, I., Petrík, I., and Williams, C.T. (2000) Coexisting monazite and allanite in peraluminous granitoids of the Tribecz Mountains, western Carpathians. American Mineralogist, 85, 22�32.

Cherniak, D.J. (2000) Rare earth element diffusion in apatite. Geochimica et Cosmochimica Acta, 64, 3871�3885.

Cherniak, D.J., Watson, E.B., Grove, M., and Harrison, T.M. (2004) Pb diffusion in monazite; a combined RBS/SIMS study. Geochimica et Cosmochimica Acta, 68, 829�840.

Cocherie, A., Be Mezeme, E., Legendre, O., Fanning, C.M., Faure, M., and Rossi, P. (2005) Electron-microprobe dating as a tool for determining the closure of Th-U-Pb systems in migmatitic monazites. American Mineralogist, 90, 697�618.

Cressey, G., Wall., F., and Cressey, B.A. (1999) Differential REE uptake by sector growth of monazite. Mineralogical Magazine, 63, 813�828.

Crowley, J.J. and Ghent, E.D. (1999) An electron microprobe study of the U-Th-Pb systematics of metamorphosed monazite; the role of Pb diffusion versus overgrowth and recrystallisation. Chemical Geology, 157, 285�302.

Dachille, F. and Roy, R. (1964) Effectiveness of shearing stresses in accelerat-ing solid phase reactions at low temperatures and high pressures. Journal of Geology, 72, 243�47.

Dahl, P.S., Hamilton, M.A., Jercinovic, M.J., Terry, M.P., Williams, M.L., and Frei, R. (2005) Comparative isotopic and chemical geochronometry of monazite, with implications for U-Th-Pb dating by electron microprobe: An example from metamorphic rocks of the eastern Wyoming Craton (U.S.A.). American Mineralogist, 90, 619�638.

Finch, C.B., Harris, L.A., and Clark, G.W. (1964) The thorite-huttonite phase transformation as determined by growth of synthetic thorite and huttonite single crystals. American Mineralogist, 49, 782�785.

Finch, R.J. and Hanchar, J.M. (2003) Structure and chemistry of zircon and zir-con-group minerals. In J.M. Hanchar and P.W.O. Hoskin, Eds., Zircon, 53, p. 1�25. Reviews in Mineralogy and Geochemistry, Mineralogical Society of America, Chantilly, Virginia.

Finger, F. and Krenn, E. (2007) Three metamorphic monazite generations in a high-pressure rock from the Bohemian Massif and the potentially important role of apatite in stimulating polyphase monazite growth along a PT loop. Lithos, 95, 103�115.

Fitzsimons, I.C.W., Kinny, P.D., and Harley, S.L. (1997) Two stages of zircon and monazite growth in anatectic leucogneiss: SHRIMP constraints on the duration and intensity of Pan-African metamorphism in Prydz Bay, East Antarctica. Terra Nova, 9, 47�51.

Förster, H.-J. (1998) The chemical composition of REE�Y�Th�U-rich accessory minerals from peraluminous granites of the Erzgebirge-Fichtelgebirge region, Germany. Part I: The monazite-(Ce)�brabantite solid solution series. American Mineralogist, 83, 259�272.

Förster, H.-J. and Harlov, D.E. (1999) Monazite-(Ce)�huttonite solid solutions in granulite-facies metabasites from the Ivrea-Verbano Zone, Italy. Mineralogical Magazine, 63, 587�594.

Förster, H.-J., Harlov, D.E., and Milke, R. (2000) Composition and Th�U�total Pb ages of huttonite and thorite from Gillespie�s Beach, South Island, New Zealand. Canadian Mineralogist, 38, 675�684.

Foster, G., Gibson, H.D., Parrish, R., Horstwood, M., Fraser, J., and Tindle, A. (2002) Textural, chemical, and isotopic insights into the nature and behavior of metamorphic monazite. In F. Poitrasson, U. Schaltegger, and J.M. Hanchar, Eds., Chemistry and physics of accessory minerals: Crystallization, transforma-tion, and geochronological. Chemical Geology, 191, 183�207.

Fynn, G.W. and Powell, W.J.A. (1979) The cutting and polishing of electro-optic materials, 216 p. Adams Hilger, London.

Gardés, E., Jaoul, O., Montel, J.-M., Seydoux-Guillaume, A.-M., and Wirth, R. (2006) Pb diffusion in onazite: An experimental study of Pb2+ + Th4+ ↔ 2 Nd3+ interdiffusion. Geochimica et Cosmochimica Acta, 70, 2325�2336.

Garson, M.S. (1966) Carbonatites in Malawi. In O.F. Tuttle and J. Gittins, Eds., Carbonatites, p. 33�71. Wiley, London.

Garson, M.S. and Campbell-Smith, W. (1965) Carbonatite and agglomeratic vents

in the Western Shire Valley. Geological Survey of Malawi, Memoir no. 3, Government Press, Zomba, Malawi.

Geisler, T., Pöml, P., Stephan, T., Janssen, A., and Putnis, A. (2005) Experimental observation of an interface-controlled pseudomorphic replacement reaction in a natural crystalline pyrochlore. American Mineralogist, 90, 1683�1687.

Goncalves, P., Williams, M.L., and Jercinovic, M.J. (2005) Electron-microprobe age mapping of monazites. American Mineralogist, 90, 578�585.

Hansen, E.C. and Harlov, D.E. (2007) Whole rock, phosphate, and silicate com-positional trends across an amphibolite- to granulite-facies transition, Tamil Nadu, India. Journal of Petrology, in press.

Harlov, D.E. and Förster, H.-J. (2002) High-grade ß uid metasomatism on both a local and regional scale: The Seward Peninsula, Alaska and the Val Strona di Omegna, Ivrea-Verbano zone, northern Italy. Part II: Phosphate mineral chemistry. Journal of Petrology, 43, 801�824.

�� (2003) Fluid-induced nucleation of REE-phosphate minerals in apatite: Nature and experiment. Part II. Fluorapatite. American Mineralogist, 88, 1209�1229.

Harlov, D.E. and Milke, R. (2002) Stability of corundum + quartz relative to kyanite and sillimanite at high temperature and pressure. American Mineralo-gist, 87, 424�432.

Harlov, D.E., Förster, H-J., and Nijland, T.G. (2002) Fluid induced nucleation of (Y + REE)-phosphate minerals in apatite: nature and experiment Part I. chlorapatite. American Mineralogist, 87, 245�261.

Harlov, D.E., Wirth, R., and Förster, H.-J. (2005) An experimental study of dis-solution- reprecipitation in ß uorapatite: Fluid inÞ litration and the formation of monazite. Contributions to Mineralogy and Petrology, 150, 268�286.

Henry, D.J., Dutrow, B.L., and Selverstone, J. (2002) Compositional polarity in replacement tourmaline�An example from the Tauern Window, Eastern Alps. Geological Materials Research, 4(2), 23 p.

Holt, D.N. (1965) The Kangankunde Hill rare earth prospect. Results of an eco-nomic investigation. Malawi Ministry of Natural Resources, 130 p. Geological Survey Department, Bulletin 20, Government Press, Zomba.

Jarosewich, E. and Boatner, L.A. (1991) Rare-earth element reference samples for electron microprobe analysis. Geostandards Newsletter, 15, 397�399.

Johannes, W. ( 1973) Eine vereinfachte Piston-Zylinder-Apparatur hoher Genauig-keit. Neues Jahrbuch für Mineralogie, 1973, 337�351.

Johannes, W., Bell, P.M., Mao, H.K., Boettcher, A.L., Chipman, D.W., Hays, J.F., Newton, R.C., and Seifert, F. (1971) An interlaboratory comparison of piston-cylinder pressure calibration using the albite-breakdown reaction. Contributions to Mineralogy and Petrology, 32, 24�38.

Kohn, M.J., Wieland, M.S., Parkinson, C.D., and Upreti, B.N. (2005) Five genera-tions of monazite in Langtang gneisses: Implications for chronology of the Hi-malayan metamorphic core. Journal of Metamorphic Geology, 23, 399�406.

Krenn, E. and Finger, F. (2004) Metamorphic formation of Sr-apatite and Sr-bear-ing monazite in a high-pressure rock from the Bohemian Massif. American Mineralogist, 89, 1323�1329.

Kucha, H. (1980) Continuity in the monazite-huttonite series. Mineralogical Magazine, 43, 1031�1034.

Labotka, T.C., Cole, D.R., Fayek, M., Riciputi, L.R., and Stadermann, F.J. (2004) Coupled cation and oxygen-isotope exchange between alkali feldspar and aqueous chloride solution. American Mineralogist, 89, 1822�1825.

Loretto, M.H. (1994) Electron Beam analysis of Materials (second edition), 272 p. Chapman and Hall, London.

Mazeina, L., Ushakov, S.V., Navrotsky, A., and Boatner, L.A. (2005) Formation enthalpy of ThSiO4 and enthalpy of the thorite to huttonite transition. Geo-chemica et Cosmochimica Acta, 69, 4675�4683.

Meldrum, A., Boatner, L.A., Zinkle, S.J., Wang, S.-X., Wang, L.-M., and Ewing, R.C. (1999) Effects of dose rate and temperature on the crystalline-to-me-tamict transformation in the ABO4 orthosilicates. Canadian Mineralogist, 37, 207�221.

Ni, Y., Hughes, J.M., and Mariano, A.N. (1995) Crystal chemistry of the monazite and xenotime structures. American Mineralogist, 80, 21�26.

Nishikawa, S. and Kikuchi, S. (1928) Diffraction of cathode rays by mica. Nature, 121, 1019�1020.

Oelkers, E.H. and Poitrasson, F. (2002) An experimental study of the dissolution stoichiometry and rates of a natural monazite as a function of temperature from 50 to 230 °C and pH from 1.5 to 10. Chemical Geology, 191, 73�87.

O�Neil J.R. (1977) Stable isotopes in mineralogy. Physics of Chemistry and Minerals, 2, 105�123.

Pabst, A. and Hutton, C.O. (1951) Huttonite, a new monoclinic thorium silicate, with an account on its occurrence, analysis, and properties. American Min-eralogist, 36, 60�69.

Prior, D.J., Boyle, A.P., Brenker, F., Cheadle, M.C., Day, A., Lopez, G., Peruzzo, L., Potts, G.J., Reddy, S., Spiess, R., Timms, N.E., Trimby, P., Wheeler, J., and Zetterstrom, L. (1999) The application of electron backscatter diffraction and orientation contrast imaging in the SEM to textural problems in rocks. American Mineralogist, 84, 1741�1759.

Putnis, A. (2002) Mineral replacement reactions: from macroscopic observations to microscopic mechanisms. Mineralogical Magazine, 66, 689�708.


Putnis C.V. and Metzger, K. (2004) A mechanism of mineral replacement: Isotope tracing in the model system KCl-KBr-H2O. Geochimica et Cosmochimica Acta, 68, 2839�2848.

Putnis, A., Hinrichs, R., Putnis, C.V., Golla-Schindler, U., and Collins, L.G. (2007) Hematite in porous red-clouded feldspars: evidence of large-scale crustal ß uid-rock interaction. Lithos, 95, 10�18.

Pyle, J.M., Spear, F.S., Wark, D.A., Daniel, C.G., and Storm, L.C. (2005) Con-tributions to precision and accuracy of monazite microprobe ages. American Mineralogist, 90, 547�577.

Rendón-Angeles, J.C., Yanagisawa, K., Ishizawa, N., and Oishi, S. (2000a) Effect of metal ions of chlorapatite on the topotaxial replacement by hydroxylapa-tite under hydrothermal conditions. Journal of Solid State Chemistry, 154, 569�578.

�� (2000b) Topotaxial conversion of chlorapatite and hydroxylapatite to ß uor-apatite by hydrothermal ion exchange. Chemical Materials, 12, 2143�2150.

Schmidt, C., Rickers, K., Bilderback, D.H., and Huang, R. (2007) In situ synchro-tron-radiation XRF study of REE phosphate dissolution in aqueous ß uids to 800 °C. Lithos, 95, 87�102.

Seydoux, A.-M. and Montel, J.-M. (1997) Experimental determination of the thorite-huttonite phase transition. Terra Nova 9, Abstract Supplements No. 1, 421.

Seydoux-Guillaume, A.-M., Paquette, J.L., Wiedenbeck, M., Montel, J.-M., and Heinrich, W. (2002a) Experimental resetting of the U-Th-Pb systems in monazite. In F. Poitrasson, U. Schaltegger, and J.M. Hanchar, Eds., Chem-istry and Physics of accessory minerals: Crystallization, transformation, and geochronological. Chemical Geology, 191, 165�181.

Seydoux-Guillaume, A.-M., Wirth, R., Heinrich, W., and Montel, J.-M. (2002b) Experimental determination of thorium partitioning between monazite and xenotime using analytical electron microscopy and X-ray diffraction Rietveld analysis. European Journal of Mineralogy, 14, 869�878.

Smyth, J.R. and McCormick, T.C. (1995) Crystallographic data for minerals. Mineral physics and crystallography: A handbook of physical constants. AGU Reference Shelf 2, 17 p.

Speer, J.A. (1982) The actinide orthosilicates. In P.H. Ribbe, Ed., Orthosilicates, 5, 113�135. Reviews in Mineralogy, Mineralogical Society of America, Chantilly, Virginia.

Taylor, M. and Ewing, R.C. (1978) The crystal structure of the ThSiO4 polymorphs: Huttonite and thorite. Acta Crystallographica Section B, 34, 1074�1079.

Terry, M.P., Robinson, P., Hamilton, M.A., and Jercinovic, M.J. (2000) Monazite geochronology of UHP and HP metamorphism, deformation, and exhuma-tion, Nordøyane, Western Gneiss Region, Norway. American Mineralogist, 85, 1651�1664.

Teufel, S. and Heinrich, W. (1997) Partial resetting of the U-Pb isotope system

in monazite through hydrothermal experiments: An SEM and U-Pb isotope study. Chemical Geology, 137, 273�281.

Tomaschek, F., Kennedy, A.K., Villa, I.M., Lagos, M., and Ballhaus, C. (2003) Zir-cons from Syros, Cyclades, Greece-Recrystallization and mobilization of zircon during high-pressure metamorphism. Journal of Petrology, 44, 1977�2002.

Wall, F. and Mariano, A.N. (1996) Rare earth minerals in carbonatites: A discussion centered on the Kangankunde Carbonatite, Malawi. In A.P. Jones, F. Wall, and C.T. Williams, Eds., Rare earth minerals: Chemistry, origin and ore deposits, 7, p. 193�225. Mineralogical Society Series, Chapman and Hall, London.

Watt, G.R. (1995) High-thorium monazite-(Ce) formed during disequilibrium melt-ing of metapelites under granulite-facies conditions. Mineralogical Magazine, 59, 735�743.

Watt, G.R. and Harley, S.L. (1993) Accessory phase controls on the geochemistry of crustal melts and restites produced during water-undersaturated partial melting. Contributions to Mineralogy and Petrology, 114, 550�566.

Williams, M.L., Jercinovic, M.J., Gonclaves, P., and Mahan, K. (2006) Format and philosophy for collecting, compiling, and reporting microprobe ages. Chemical Geology, 225, 1�15.

Wirth, R. (2004) Focused ion beam (FIB): A novel technology for advanced ap-plication of micro- and nanoanlysis in geosciences and applied mineralogy. European Journal of Mineralogy, 15, 863�875.

Woolley, A.R. (1991) The Chilwa alkaline igneous province of Malawi: A review. In A.B. Kampunzu and R.T. Lubala, Eds., Magmatism in extensional struc-tural settings, The Phanerozoic African plate, p. 377�409. Springer-Verlag, Berlin.

Wright, S.I. and Adams, B.L.A. (1992) Automatic analysis of electron backscatter diffraction patterns. Metallurgical Transactions, 23A, 759-767.

Yanagisawa, K., Rendon-Angeles, J.C., Ishizawa, N., and Oishi, S. (1999) Topot-axial replacement of chlorapatite by hydroxylapatite during hydrothermal ion exchange. American Mineralogist, 84, 1861�1869.

Yardley, B.W.D. (1982) The early metamorphic history of the Haast schists and related rocks of New Zealand. Contributions to Mineralogy and Petrology, 81, 317�327.

Zhu, X.K., O�Nions, R.K., Belshaw, N.S., and Gibb, A.J. (1997) SigniÞ cance of in situ SIMS chronometry of zoned monazite from the Lewisian granulites, northwest Scotland. Chemical Geology, 135, 35�53.

MANUSCRIPT RECEIVED SEPTEMBER 21, 2006MANUSCRIPT ACCEPTED JUNE 5, 2007MANUSCRIPT HANDLED BY DARRELL HENRY

the relative stability of monazite and huttonite at 300 900 c and 200 1000 mpa: metasomatism and...

Documents