Monazite–allanite phase relations in metapelites

Chemical Geology 279 (2010) 55–62

Contents lists available at ScienceDirect

Chemical Geology j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / c h e m g e o

Research paper

Monazite–allanite phase relations in metapelites Frank S. Spear Department of Earth and Environmental Sciences, Rensselaer Polytechnic Institute, 110 8th Street, Troy, NY 12180, United States

a r t i c l e

i n f o

Article history: Received 21 May 2010 Received in revised form 28 September 2010 Accepted 1 October 2010 Editor: R.L. Rudnick Keywords: Monazite Allanite Phase relations

a b s t r a c t Calculations of the transition from allanite to monazite-bearing assemblages in typical pelitic bulk compositions have been made using thermodynamic data estimated from oxide sums and inferred from natural parageneses. Calculations in the CFASHPCe and MnNCKFMASHPCe systems place the allanite to monazite transition in the middle amphibolite facies (525–600 °C) for a bulk composition similar to Shaw's average pelite. The temperature of the transition is pressure dependent, and strongly dependent on the bulk rock CaO content, consistent with inferences from natural parageneses. The transition is also a function of the bulk Al2O3 content, although the calculated result is opposite to that inferred from natural samples. Comparison with published results in the La–Mg system suggest that the nature of the REE phosphate does not greatly influence the conditions of the transition. © 2010 Elsevier B.V. All rights reserved.

1. Introduction The quantitative relationship among pressure, temperature and the relative stabilities of allanite and monazite is crucial for the interpretation of the ages of these minerals. A number of studies have examined progressive metamorphic sequences, with different conclusions depending on the suites examined. In general, it has been concluded that in rocks of little or no calcium (and sufficient LREE and P), monazite is formed by at least the low greenschist facies and persists throughout metamorphism (e.g., Overstreet, 1967). In rocks of sufficient CaO and LREE, allanite forms at low grades and reacts out at conditions of around the staurolite and/or kyanite isograd (e.g., Smith and Barreiro, 1990; Kingsbury et al., 1993; Wing et al., 2003; Kohn and Malloy, 2004; Fitzsimons et al., 2005; Rasmussen et al., 2006; Janots et al., 2006, 2008; Tomkins and Pattison, 2007; Corrie and Kohn, 2008; see discussion in Spear and Pyle, 2002). Indeed, several studies (e.g., Smith and Barreiro, 1990; Wing et al., 2003; Janots et al., 2006, 2008; and the experimental study of Janots et al., 2007) have even reported the formation of lowgrade monazite (low greenschist facies) that gives rise to allanite and is eventually replaced by a new generation of monazite around the conditions of the staurolite or Al2SiO5 isograd. However, the relationships between these minerals are rather complex and the reactions by which the transition occurs, and quantification of the relationship between allanite stability and bulk rock composition (notably CaO content), have not been resolved in detail. This paper explores the relationship between allanite (CaCeFeAl2 (SiO4)3OH) and monazite (CePO4) during progressive metamorphism by examining both the chemographic relations and through thermo-

E-mail address: [email protected]. 0009-2541/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.chemgeo.2010.10.004

dynamic calculations. A similar set of calculations was undertaken by Janots et al. (2007) based on new thermodynamic data in the system SiO2–Al2O3–FeO–Fe2O3–MgO–CaO–Na2O–K2O–P2O5–La2O3–CO2–H2O and equilibrium assemblage diagrams were presented for a system containing LaPO4 (monazite) and CaLaMgAl2(SiO4)3OH (dissakisite). The results from this and the present study substantiate the inferences of earlier workers and make it possible, within the errors of the thermodynamic data, to calculate the monazite isograd in rocks of any pelitic bulk composition.

2. Chemographic relations The simplest chemical system in which to represent the chemographic relations between monazite and allanite is SiO2–Al2O3–FeO– CaO–H2O–P2O5–Ce2O3 (CFASHPCe). Schists in such a system will contain quartz and fluid (H2O) in excess. Additionally, phases such as chlorite, garnet, staurolite, chloritoid, pyrophyllite, kyanite, sillimanite, andalusite are typical. To simplify representation, this paper will first consider rocks in which an Al2SiO5 polymorph is in excess. Projection from quartz, H2O and Al2SiO5 leaves four species to represent chemographically. Phase relations involving garnet will be examined initially because, as will be shown below, phase relations between monazite and allanite depend to a first order on the bulk CaO content of the rock and the composition of the garnet that is stable at the P–T conditions of interest. In addition, metapelites also contain appreciable amounts of K, Mg, Na, and Mn. Muscovite projection removes K from consideration and plagioclase projection removes Na. Projection through an additional AFM phase (e.g., chlorite, biotite, staurolite, chloritoid, cordierite) removes the need to represent an MgO component. MnO stabilizes garnet to lower temperatures and has the effect of shifting the P–T

F.S. Spear / Chemical Geology 279 (2010) 55–62

P + Quartz + H2O + Al2SiO5

Apatite

To Garnet

+ Quartz + H2O + Al2SiO5 + Apatite

P Clinozoisite

Monazite

Fe

Ce

Alm

rn et

conditions of garnet-bearing assemblages. The effects of MnO will be explored below. However, for the initial assessment of the chemography these components will be neglected. Projection from quartz, H2O, and Al2SiO5 in the system CFASHPCe leaves the components FeO–CaO–P2O5–Ce2O3 to be represented chemographically. Phases to be considered in this system include allanite, monazite, apatite, and garnet (Fig. 1). In this subsystem, monazite is very nearly a fixed composition, garnet is a solid solution between almandine and grossular, and allanite is a complete solid solution between end member allanite (CaCeFeAl2(SiO4)3OH) and clinozoisite (Ca2Al3(SiO4)3OH) (Gieré and Sorenson, 2004). Apatite (Ca5(PO4)3OH) displays minor solid solution towards a LREE end member (Ca4Ce(SiO4)(PO4)2OH) (e.g., Bingen et al., 1996). Other phases that are possible include rhabdophane (CePO4•H2O), which has been shown to have a highly restricted stability relative to monazite (Akers et al., 1993; see also Krenn and Finger, 2007, as a retrograde phase), and a Ce2O3 oxide phase, which has not been reported from natural occurrences. Given these considerations, there is a single four–phase volume in the tetrahedral phase diagram of Fig. 1 comprised of apatite+ garnet + allanite + monazite, which is flanked by the three-phase planes apatite + monazite + garnet, apatite + allanite + garnet, apatite + monazite + allanite, and monazite + allanite+ garnet. Of these, only apatite + monazite + garnet and apatite+ allanite + garnet are considered to be relevant for the metamorphism of pelitic schists because the LREE elements are always far less abundant than CaO+ Fe, eliminating garnet-absent assemblages, and P2O5 is typically in higher concentration than LREE, which means most assemblages are apatite-bearing. Inasmuch as apatite is nearly ubiquitous, further projection from apatite onto the plane FeO–PO5/2–CeO3/2 results in the chemography shown in Fig. 2. Note that pure allanite (CaCeFeAl2(SiO4)3OH) projects directly onto the plane FeO–PO5/2–CeO3/2 in this projection but the composition Ca20/13Ce6/13Fe6/13Al33/13Si plots at infinity and pure clinozoisite (Ca2Al3 (SiO4)3OH) plots on the P apex, but projected through infinity. Several inferences can be drawn about the relative observance of allanite versus monazite in metapelites from Figs. 1 and 2. Firstly, at a specified P and T, garnet + monazite will be observed in rocks low in Ca, whereas garnet + allanite will be observed in rocks of higher Ca (all assemblages with apatite). There should also be rocks with appropriate bulk composition to contain garnet + allanite + monazite. However, the LREE content of typical pelites is only a few tens to a few hundreds of ppm, so the assemblage garnet + allanite+ monazite will only occur in rocks of a very restricted range of bulk compositions.

Ga

56

Grs

Allanite To Clinozoisite To Clinozoisite Fig. 2. Schematic chemographic relations in the system CFASHPCe projected from quartz, H2O, Al2SiO5, and apatite onto the plane FeO–PO5/2–CeO3/2 showing the plotting positions for the assemblages garnet + allanite + monazite, garnet + allanite, and garnet + monazite (all + apatite).

What is most important to recognize in Figs. 1 and 2 is that the phase volume garnet + monazite + allanite (+ apatite) will shift position depending on the grossular content of garnet. As will be shown below, with increasing temperature, garnet in the assemblage garnet + allanite + monazite + apatite becomes more grossular-rich. Therefore, a rock that contains garnet + allanite at low grade will lose allanite and grow monazite at higher grade, as is observed in natural parageneses. Quantification of this continuous reaction is discussed below. Not so readily apparent in Fig. 1 is a field of two-phase assemblages garnet + apatite, which arises because of the incorporation (albeit limited) of Ce into apatite (Fig. 3). The field is irregularly shaped with the high-Ce surface being curved. Thus, it is possible at any T and P to have only garnet + apatite assemblages present. 3. Thermodynamic model

Monazite Fe Garnet

Ca Czoi - Aln

Ce Fig. 1. Schematic chemographic relations in the system CFASHPCe projected from quartz, H2O, and Al2SiO5 into the volume FeO–CaO–CeO3/2–PO5/2. The four-phase assemblage garnet + allanite + monazite + apatite is flanked by the three-phase assemblages garnet + allanite + apatite and garnet + monazite + apatite.

Quantification of the allanite–monazite phase relations requires thermodynamic data, which are not available for all phases of interest. In this study, data for the silicates are taken from the compilations of Berman (1988), with modifications described by Pattison et al. (1999), Spear and Kohn (1996), Spear and Wark (2009), and Spear and Pyle (2010). Values of entropy, volume and heat capacity for monazite and apatite have been taken from the compilations of Robie et al. (1978) and Robie and Hemingway (1995) or from oxide sums from data therein. Enthalpy of OH apatite is from Robie and Hemingway (1995) and enthalpy of monazite is from Ushakov et al. (2001). Values of S, V and Cp for allanite were computed from tabulated values in Berman (1988) and Robie et al. (1978) from the relation allanite = clinozoisite +

57

v4 v3

Ce

Heat capacity expression is Cp = a + b/(T0.5) + c/T2 + c/T3 + f*T + g*T2. Volume expression follows Berman (1988).

v2 v1

0 0 0 − 3.4670 × 10− 5 0 b

T = 530 C, P = 6000 bars. System: SiO2–Al2O3–FeO–CaO–H2O–PO4–Ce2O3.

Monazite

Apatite

a

Monazite Allanite

Allanite

1.0 1.0 1.0 0.8 0.2 1.0 0.76 0.24 0.989 0.011

OH-Apt CeOH-Apt Aln Czoi Mnz

SiO2 H2O A12SiO5 Fe3Al2Si3O12 Ca3A12Si3O12 CePO4 CaCeFeA12(SiO4)3OH Ca2A13(SiO4)3OH Ca5(PO4)3(OH) Ca5(PO4)3(F)

Component

Quartz Fluid Kyanite Garnet

Apatite

Mole fraction

Phase

Formula

Table. 2 Thermodynamic data used in this study.

Table 1 Reference assemblage and P–T conditions for calculation of enthalpies of allanite and Ce–OH–apatite. Phase

0 0 − 2.41 × 10− 6 0 0

g b a S

390.40 411.32 358.135 287.1 133.06

H

− 6738500 − 6612111 − 6303179 − 6894970 − 1967800

Formula

Ca5(PO4)3 Ca4Ce(SiO4)(PO4)2OH CaCeFeAl2(SiO4)3OH Ca2Al3(SiO4)3OH CePO4

V

Heat capacitya

c

d

f

1/3almandine − 1/3 grossular + CeO3/2–AlO3/2. Enthalpy values for allanite and Ce-apatite were computed from an assumed reference assemblage at 530 °C, 6 kbar (Table 1) and mixing was assumed to be ideal between allanite and apatite end members. This approach yields enthalpies that carry uncertainties directly related to the accuracy of the reference assemblage, but has the advantage of incorporating uncertainties in the activity models into the values (avoiding extrapolation from end-member tabulations). It is functionally equivalent to the applications of differential thermodynamics (the Gibbs method) described by Spear (1993) and is exact at the conditions of the reference assemblage. Data derived in this way are listed in Table 2. In the system CFASHPCe the assemblage garnet + allanite + monazite + apatite (+ quartz + Al2SiO5 + H2O) is divariant and P–T space can be contoured for the mole fraction of grossular (Xgrs) for this assemblage (Fig. 4a). The grossular content of garnet in this assemblage increases with increasing temperature, as mentioned above, resulting in the shift in the phase compatibility diagram with temperature as shown in Fig. 4b. Fig. 4 is sufficient to provide a qualitative understanding of the relative occurrences of monazite-bearing versus allanite-bearing assemblages. Typical bulk compositions are low in Ce2O3 (and other LREE) and will plot near garnet, so at low temperature the possible assemblages are Grt + Mnz + Apt, Grt + Mnz + Aln + Apt, or Grt + Aln + Apt. Apatite+ Grt 2-phase assemblages are also possible (cf. Fig. 3). With increasing T, garnet in the assemblage Grt + Mnz + Aln + Apt becomes increasingly Ca-rich, the result being that monazite may appear in Grt+ Aln + Apt assemblages or allanite may react out of Grt+ Mnz+ Aln + Apt assemblages. Because of the very low LREE

0.118600 0.1186 0.14494 0 0.123

Fig. 3. Schematic chemographic relations in the system CFASHPCe projected from quartz, H2O, and Al2SiO5 into the volume FeO–CaO–CeO3/2–PO5/2 showing the stability field for the 2-phase assemblage garnet + apatite.

0 0 0 124858000 0

Ca

− 12700000 − 12700000 − 5353060 − 2380530 − 852650

Grs

18.11 18.11 − 745.808 − 6509.28 0

t

Volume termsb

r ne

387.8 387.8 403.277 749.17 82.85

Brth

Ga

0 0 0 0 0

Apt

15.96 15.97 14.25 13.67 4.52

Fe Alm

0 0 0 − 5.516 × 10− 7 0

P + Quartz + H2O + Al2SiO5

0 0 0 − 1.2879 × 10− 12 0

F.S. Spear / Chemical Geology 279 (2010) 55–62

58

F.S. Spear / Chemical Geology 279 (2010) 55–62

a

b 12 + Quartz + H2O + Al2SiO5 + Apatite

10

Fe

XGrs

400 500 600 700

rne Ga

0.30

0.26

0.28

0.22

0.24

0.16

0.18

0.14

4

0.20

P kbar

t

8

6

e

nit

2

0 350

6 kbar

a All 450

550

650

750

850

Monazite

Ca

Ce

T (C) Fig. 4. (a) P–T diagram contoured for XGrs for the assemblage garnet+ allanite+ monazite + apatite + quartz + Al2SiO5 + H2O in the CFASHPCe system. (b) Pseudoternary FeO–PO5/2– CeO3/2 plot (projected from quartz, H2O, Al2SiO5, and apatite) showing the shift in garnet composition with temperature (P= 6 kbar) in the assemblage garnet+ allanite+ monazite.

contents of typical pelites, the stable co-existence of allanite and monazite will occur over only a very small temperature interval. With increasing temperature, allanite-bearing assemblages should become increasingly restricted and most rocks should contain monazite, as is observed in natural parageneses (e.g., Smith and Barreiro, 1990; Wing et al., 2003; Kohn and Malloy, 2004; Corrie and Kohn, 2008).

(CaO = 4.34 wt.%; Fig. 6b), depicts the allanite-to-monazite transition at around 480 °C at 2 kbar and 750 °C at 10 kbar. It has been suggested by Wing et al. (2003) that the bulk rock Al2O3 content also affects the temperature of the allanite-to-monazite transition so a series of EADs with Al2O3 contents of 0.75, 1.25 and 1.5 times that of Shaw's average pelite were constructed (Fig. 7). The transition from allanite to monazite is clearly a function of Al2O3

4. Equilibrium assemblage diagrams

10

+ Quartz + H2O + Al2SiO5 + Garnet + Apatite

Monazite

it e +

P kbar

8

n Al l a

6

Allanite

zit e Mona

An equilibrium assemblage diagram (EAD, also referred to as a “pseudosection”) for the system CFASHPCe is presented in Fig. 5 that delineates the P–T fields for assemblages that contain allanite, monazite, or both. The bulk composition has elemental proportions that are similar to that of Shaw's average pelite (Shaw, 1956; also used by Symmes and Ferry, 1992) and is listed in Table 3. The assemblage garnet + allanite + apatite is stable up to around 700 °C at a pressure around 5 kbar. Monazite then joins the assemblage and allanite reacts out within a few degrees leaving the assemblage garnet + monazite + apatite. The temperature width of the allanite–to-monazite transition depends on the bulk rock Ce content. Dotted lines in Fig. 5 show the increase in width of the allanite + monazite field with twice the amount of Ce2O3. Most importantly, the temperature of the transition is a first-order function of the bulk rock CaO content. The dashed lines show the location of the transition with half the CaO content of the original bulk composition. A more realistic bulk composition must include K2O, Na2O, MgO, and MnO, with consideration of the additional phases muscovite, biotite, and plagioclase. Fig. 6a shows an EAD for a bulk composition similar to Shaw's average pelite (Shaw, 1956). Only the phases quartz, fluid, muscovite, biotite, chlorite, garnet, Al2SiO5, apatite, allanite and monazite were considered, so equilibria involving staurolite (which would occupy a large central part of the diagram), cordierite (which would appear at low pressures, and melt (which would appear at high temperature) are not considered. This was done to simplify the calculations so as not to obscure the important accessory phase relations and does not strongly affect the conclusions. The allanite-tomonazite transition has a positive P–T slope and, depending on the pressure, occurs at approximately 400–550 °C in a rock with 2.17 wt.% CaO. A second EAD, constructed with twice the CaO content

12

4

2

0 350

450

550

650

750

850

T (C) Fig. 5. P–T equilibrium assemblage diagram (EAD) for the system CFASHPCe with the bulk composition similar to Shaw's average pelite (Shaw, 1956; CaO = 2.17 wt.%; Ce2O3 = 0.05 wt. %; see Table 3 for complete bulk compositions). Dotted lines show allanite-to-monazite transition with Ce2O3 = 0.1 wt.% (twice the value of the solid lines). Dashed lines show location of allanite-to-monazite transition with Ca = half the value of the solid lines (1.08 wt.% CaO).

F.S. Spear / Chemical Geology 279 (2010) 55–62

59

Table 3 Bulk compositions used in calculation of equilibrium assemblage diagrams. CFASHPCe

SiO2 A12O3 MgO FeO MnO CaO Na2O K2O H2O P2O5 Ce2O3 a

MnNCKFMASHPCe

MnNCKFMASHPCe

Fig. 5 solid

Fig. 5 dashed

Fig. 5 dotted

Fig. 6a

Fig. 6b

Fig. 7a

Fig. 7b

Fig. 7c

59.77 16.57 – 5.88 – 2.17 – –

59.77 16.57 – 5.88 – 1.08 – –

59.77 16.57 – 5.88 – 2.17 – –

59.77 16.57 2.62 5.88 0.07 2.17 1.73 3.53

59.77 16.57 2.62 5.88 0.07 4.34 1.73 3.53

59.77 12.43 2.62 5.88 0.07 2.17 1.73 3.53

59.77 20.71 2.62 588 0.07 2.17 1.73 333

59.77 24.86 2.62 5.88 0.07 2.17 1.73 3.53

a

a

a

a

a

a

a

a

0.30 0.05

0.30 0.05

0.30 0.10

0.30 0.05

0.30 0.05

0.30 0.05

0.30 0.05

0.30 0.05

Sufficient H2O is added to ensure the assemblage is H2O-saturated at all times.

content with the temperature of the transition decreasing (at 5 kbar) from 550 °C to 475 °C to 460 °C to 355 °C with Al2O3 contents of 12.43, 16.57, 20.71, and 24.85 wt.%, respectively (see Table 3 for bulk compositions). 5. Discussion The calculations presented in this study are consistent with most inferences about the allanite-to-monazite transition in metapelitic bulk compositions. Smith and Barreiro (1990) and later Wing et al. (2003), Kohn and Malloy (2004), and Corrie and Kohn (2008), all report the major appearance of monazite at P–T conditions near the staurolite and/or Al2SiO5 isograd. Kohn and Malloy (2004) did not believe that allanite was the precursor but further study on the same rocks reported by Corrie and Kohn (2008) infer that either allanite must have been the precursor to the monazite, or there is some as yet unidentified source of Ce in the rocks (for example grain boundaries). The staurolite and/or Al2SiO5 isograd in most metapelites occurs at

a

525–575 °C, which is the same range as is observed for the calculated transition for a bulk composition of Shaw's average pelite. The first order dependence of the transition on the bulk CaO content, noted by Kingsbury et al. (1993), Wing et al. (2003), and reviewed by Spear and Pyle (2002) is clear by comparing diagrams such as Fig. 6a and b. Dependence of the transition of allanite-to-monazite on the bulk rock Al2O3 content is apparent in Fig. 7 as was suggested by Wing et al. (2003). However, the dependence revealed by the calculations in this paper are the opposite as was predicted by Wing et al. (2003), in which the transition was shown to have a negative slope on a plot of CaO vs. Al2O3 in rocks of similar grade (Wing et al., 2003, Figs. 7 and 8). This result can be inferred by considering the relationship shown in Fig. 7 in which increasing the Al2O3 content lowers the temperature of the allanite-to-monazite transition. All three diagrams in Fig. 7 are calculated with bulk compositions with the same CaO contents. If one were to “adjust” the transition to the same P–T conditions (e.g., the staurolite isograd at around 550 °C), this thought exercise would require adding the most CaO to the rock with the highest Al2O3

b 12

10

12 + Qtz + H2O + Ms + Bt + Apt

Monazite + Grt Allanite + Chl + Grt

+ Qtz + H2O + Ms + Bt + Apt

10

Monazite + Chl + Grt Monazite + Grt + Ky

Allanite + Chl + Grt

Allanite + Grt

8

Monazite + Grt + Sil

P kbar

8

P kbar

Monazite + Grt

6

6 Allanite Monazite

Allanite + Chl

Monazite + Sil

4

4

Monazite + Chl Monazite + And

2

0 350

450

550

2

650

T (C)

750

850

Allanite + Chl

0 350

450

550

650

750

850

T (C)

Fig. 6. (a) P–T equilibrium assemblage diagram for the system MnNCKFMASHPCe showing the P–T conditions for assemblages stable in a bulk composition representative of Shaw's average pelite (Shaw, 1956; see Table 3 for composition). Only the phases indicated on the diagram were considered in the calculations. (b) EAD for a bulk composition similar to Shaw's average pelite with twice the CaO content. The maximum thermal stability of allanite is strongly dependent on the bulk rock CaO content.

60

a

c

b 12

12

Monazite + Chl + Grt + Ky Allanite + Chl + Grt

8

Allanite + Bt

Monazite + Bt

6

6

Allanite + Bt + Chl + Grt

10 Monazite + Chl + Grt

0 350

450

Allanite + Chl

Monazite + Bt + Ky

650

T (C)

750

850

0

350

Monazite + Chl + Grt + Ky

Monazite + Bt + Grt + Sil

6

Monazite + Chl + Ky Monazite + Bt + Sil

4

Monazite + Chl + And

Monazite + Bt + Chl + And

2

2

550

Monazite + Bt + Grt + Ky

Monazite + Bt + Chl + Grt

4

Bt

Allanite + Chl + Grt + Ky

Allanite + Chl + Ky

Monazite + Bt + Chl

2

+ Qtz + H2O + Ms + Apt

8

Allanite + Bt + Chl

4

Monazite + Bt + Grt + Ky

P kbar

P kbar

8

Allanite + Bt + Grt

+ Qtz + H2O + Ms + Apt

F.S. Spear / Chemical Geology 279 (2010) 55–62

Allanite + Bt + Chl + Grt

10

P kbar

10

+ Qtz + H2O + Ms + Apt

12

450

550

650

T (C)

750

850

0 350

450

550

650

750

850

T (C)

Fig. 7. P–T EADs at three different bulk Al2O3 contents (see Table 3). (a) 0.75 times Shaw's average pelite (Al2O3 = 12.43 wt.%); (b) 1.25 times Shaw's average pelite (Al2O3 = 20.71 wt.%); (c) 1.5 times Shaw's average pelite (Al2O3 = 24.86 wt. %). The allanite-to-monazite transition has a generally positive P–T slope and the temperature decreases with increasing Al2O3 content.

F.S. Spear / Chemical Geology 279 (2010) 55–62

6

5 l O3 1.0xA 2

l 2O

xA

1.0

3

3 l 2O

xA

2 O3 xAl 2

.75

10 kba r

CaO Wt. %

4

r

ba

0k

1 3:

5 kbar

1.5

1

0

bar

15 k

61

intermediate P–T conditions. Nevertheless, low-grade growths of presumably metamorphic monazite, which at higher grades gives way to allanite, have been reported from natural parageneses (e.g., Smith and Barreiro, 1990; Wing et al., 2003; Janots et al., 2006, 2008). The significance of these observations is not clear, however. It is possible that the low-grade overgrowths represent a coarsening of detrital monazite, although in some examples the compositions of the overgrowths is distinctly different from the detrital cores (e.g., Janots et al., 2008). It is also possible that the monazite overgrowths form metastably relative to florencite, the latter having been reported as a precursor to allanite in high-pressure metapelites from Morocco (Janots et al., 2006). It should further be noted that not all studies of low-grade metapelites have reported low-grade metamorphic monazite as a precursor to allanite (e.g., Kingsbury et al., 1993; Kohn and Malloy, 2004; Tomkins and Pattison, 2007; Corrie and Kohn, 2008) so the significance of the reported occurrences must remain uncertain. 6. Conclusions

300

400

500

600

700

800

T ˚C Fig. 8. Plot of the temperature of the allanite-to-monazite transition against the bulk rock CaO content (wt.%). The temperature increases with CaO content, but the slope is a function of the Al2O3 content and the temperature is a function of the pressure. Data from this study are shown in thick lines labeled with Al2O3 contents. Curves from Janots et al. (2007) are shown in light lines labeled only with pressure.

content. Therefore, these calculations reveal that a plot of the CaO vs. Al2O3 of rocks undergoing the transition at the same metamorphic grade should have a positive slope, rather than a negative one as suggested in the previous work. A very similar set of calculations was performed on allanite and monazite phase relations by Janots et al. (2007), although their study was done using a La–monazite (LaPO4) and Mg–allanite (dissakisite: CaLaMgAl2(SiO4)3OH). Although the chemical system studied by Janots et al. (2007) and the present work differ considerably, the upper thermal stability of allanite from the two studies are reassuringly similar in that both show a positive P–T dependence and a maximum thermal stability at around the staurolite isograd. Janots et al. (2007) also show a diagram that quantifies the allanite-to-monazite transition over a range of CaO contents with results similar to those found here. To better compare the results of these two studies, the temperature of the allanite-to-monazite transition was calculated as a function of bulk rock CaO content for bulk compositions with different Al2O3 contents, as well as at two pressures (5 and 10 kbar), and the results of Janots et al. (2007) were redrawn on the same scale. In truth, the scales may be slightly different because their original plot of Ca in mole % did not specify (a) whether the composition axis was mole % Ca or CaO, and (b) whether the normalization was on a hydrous or anhydrous basis. Nevertheless, the results are quite similar, especially considering that they are representative of different chemical systems. In detail, the results of the two studies show different inflections, but overall the similarity suggests that the influence of different REE in the theoretical system (Ce vs. La) or in the natural systems (La, Ce, Sm, Nd, etc) will not greatly affect the calculated stability of these phases. A significant difference between the results of the present study and that of Janots et al. (2007) is the presence of a low-grade stability field for metamorphic monazite. Such behavior was not encountered in the present study (there was no low temperature stability field encountered for monazite in any of the bulk compositions examined) and the reason for the discrepancy is not clear. Whether this is due to the different chemical systems studied or fundamental differences in the thermodynamic data employed is unclear, but multiple stability fields for a phase in a fixed bulk composition are not common in metamorphic systems, which typically do not show the most hydrated assemblages at

The results presented here should not be applied uncritically. Rather, they should serve as a template for further calculations on specific bulk compositions of interest. Further refinement of the thermodynamic data of accessory phases, especially allanite, monazite, and apatite are critical requirements for future improvements. In addition, it is essential that a more thorough assessment of the relative concentrations of the various trace elements in all silicates be made. Even though the concentrations may be low, it is possible that some phases under some conditions contain enough of a particular trace element to affect the stability of an accessory phase in a major way. A prime example is Y in garnet, which affects xenotime and monazite stability (e.g., Spear and Pyle, 2010), or Zr in a melt phase (e.g., Kelsey et al., 2008), which affects zircon stability. It is also possible that other, as yet unappreciated phases such as REE oxides, hydroxides, or phosphates are important reactants. The significance of rhabdophane has been downgraded (e.g., Akers et al., 1993), and searches for REE oxide phases using X-ray mapping have yielded negative results (e.g., Kohn and Malloy, 2004). Calculations with a phase of the composition Ce2O3 were made during the course of the present study, but the stability field was at temperatures too high to be of importance during metamorphism. However, it is not clear that the thermodynamic data for this compound are of sufficient quality to permit ruling out Ce2O3 or similar oxides as a potential source for LREE. Furthermore, it is possible that a significant part of the mass balance of accessory phases occurs between the phases and grain boundary reservoirs, as has been suggested by Corrie and Kohn (2008). Such a possibility is currently beyond our ability to model theoretically, but new instrumentation such as analytical high resolution TEMs or atom probes may permit grain boundary compositions to be assessed in natural samples in the near future. Acknowledgments This work was supported by grants from the National Science Foundation EAR-106738, EAR-337413, and EAR-409622 and the Edward P. Hamilton Distinguished Chair for Science Education at Rensselaer Polytechnic Institute. Constructive reviews by D. Harlov and an anonymous reviewer served to clarify the text and highlight additional studies of importance. References Akers, W.T., Grove, M., Harrison, T., Ryerson, F.J., 1993. The instability of rhabdophane and its unimportance in monazite paragenesis. Chemical Geology 110, 169–176. Berman, R.G., 1988. Internally-consistent thermodynamic data for minerals in the system Na2O–K2O–CaO–MgO–FeO–Fe2O3–Al2O3–SiO2–TiO2–H2O–CO2. Journal of Petrology 29, 445–522.

62

F.S. Spear / Chemical Geology 279 (2010) 55–62

Bingen, B., Demaiffe, D., Hertogen, J., 1996. Redistribution of rare earth elements, thorium, and uranium over accessory minerals in the course of amphibolite to granulite facies metamorphism: the role of apatite and monazite in orthogneisses from southwestern Norway. Geochimica et Cosmochimica Acta 60, 1341–1354. Corrie, S.L., Kohn, M.J., 2008. Trace-element distributions in silicates during prograde metamorphic reactions: implications for monazite formation. Journal of Metamorphic Geology 26, 451–464. Fitzsimons, I.C.W., Kinny, P.D., Wetherley, S., Hollingsworth, D.A., 2005. Bulk chemical control on metamorphic monazite growth in pelitic schists and implications for U–Pb age data. Journal of Metamorphic Geology 23, 261–277. Gieré, R., Sorenson, S.S., 2004. Allanite and other REE-rich epidote-group minerals. In: Liebscher, A., Franz, G. (Eds.), Epidotes. Reviews in Mineralogy and Geochemistry. Mineralogical Society of America, Washington, D. C., pp. 431–493. Janots, E., Negro, F., Brunet, F., Goffé, B., Engi, M., Bouybaouène, M.L., 2006. Evolution of the REE mineralogy in HP-LT metapelites of the Sebtide complex, Rif, Morocco: monazite stability and geochronology. Lithos 87, 214–234. Janots, E., Brunet, F., Goffè', B., Poinssot, C., Burchard, M., Cemiĉ, L., 2007. Thermochemistry of monazite-(La) and dissakisite-(La): implications for monazite and allanite stability in metapelites. Contributions to Mineralogy and Petrology 154, 1–14. Janots, E., Engi, M., Berger, J., Allaz, J., Schwarz, O., Spandler, C., 2008. Prograde metamorphic sequence of REE minerals in pelitic rocks of the Central Alps: implications for allanite–monazite–xenotime phase relations from 250 to 610 °C. Journal of Metamorphic Geology 26, 509–526. Kelsey, D.E., Clark, C., Hand, M., 2008. Thermobarometric modeling of zircon and monazite growth in melt-bearing systems: examples using model metapelitic and metapsammitic granulites. Journal of Metamorphic Geology 26, 199–212. Kingsbury, J.A., Miller, C.F., Wooden, J.L., Harrison, T.M., 1993. Monazite paragenesis and U–Pb systematics in rocks of the eastern Mojave Desert, California, U.S.A.: implications for thermochronometry. Chemical Geology 110, 147–168. Kohn, M.J., Malloy, M.A., 2004. Formation of monazite via prograde reactions among common silicates: implications for age determinations. Geochimica et Cosmochimica Acta 68, 101–113. Krenn, E., Finger, F., 2007. Formation of monazite and rhabdophane at the expense of allanite during Alpine low temperature retrogression of metapelitic basement rocks from Crete, Greece: Microprobe data and geochronological implications. Lithos 95, 130–147. Overstreet, W.C., 1967. The geologic occurrence of Monazite. 530, U.S. Geological Survey Professional Paper 530. Pattison, D.R.M., Spear, F.S., Cheney, J.T., 1999. Polymetamorphic origin of muscovite + cordierite + staurolite + biotite assemblages: Implications for the metapelitic petrogenetic grids and for P–T paths. Journal of Metamorphic Geology 17, 685–703.

Rasmussen, B., Muhling, J.R., Fletcher, I.R., Wingate, T.D., 2006. In situ SHRIMP U–Pb dating of monazite integrated with petrology and textures: does bulk composition control whether monazite forms in low-Ca pelitic rocks during amphibolite facies metamorphism? Geochimica et Cosmochimica Acta 70, 3040–3058. Robie, R.A., Hemingway, B.S., 1995. Thermodynamic properties of minerals and related substances at 298.15 K and 1 bar (105 pascals) pressure and at higher temperatures. U. S. Geological Survey. Robie, R.A., Hemingway, B.S., Fisher, J.R., 1978. Thermodynamic properties of minerals and related substances at 298.15 K and 1 bar (105 pascals) pressure and at higher temperatures. United States Geological Survey Bulletin 1452, 456. Shaw, D.M., 1956. Geochemistry of pelitic rocks. Part III: major elements and general geochemistry. Geological Society of America Bulletin 67, 919–934. Smith, H.A., Barreiro, B., 1990. Monazite U–Pb dating of staurolite grade metamorphism in pelitic schists. Contributions to Mineralogy and Petrology 105, 602–615. Spear, F.S., 1993. Metamorphic Phase Equilibria and Pressure–Temperature–Time Paths. Mineralogical Society of America Monographs. Mineralogical Society of America, Washington, D. C.. 799 pp. Spear, F.S., Kohn, M.J., 1996. Trace element zoning in garnet as a monitor of crustal melting. Geology 24, 1099–1102. Spear, F.S., Pyle, J.M., 2002. Apatite, monazite, and xenotime in metamorphic rocks. In: Kohn, M.J., Rakovan, J., Hughes, J.M. (Eds.), Phosphates: Geochemical, Geobiological, and Materials Importance. Reviews in Mineralogy and Geochemistry. Mineralogical Society of America, Washington, D. C., pp. 293–335. Spear, F.S., Pyle, J.M., 2010. Theoretical modeling of monazite growth in a low-Ca metapelite. Chemical Geology 266, 218–230. Spear, F.S., Wark, D.A., 2009. Cathodoluminescence imaging and titanium thermometry in metamorphic quartz. Journal of Metamorphic Geology 27, 187–205. Symmes, G.H., Ferry, J.M., 1992. The effect of whole-rock MnO content on the stability of garnet in pelitic schists during metamorphism. Journal of Metamorphic Geology 10, 221–237. Tomkins, H.S., Pattison, D.R.M., 2007. Accessory phase petrogenesis in relation to major phase assemblages in pelites from the Nelson contact aureole, southern British Columbia. Journal of Metamorphic Geology 25, 401–421. Ushakov, S.V., Helean, K.B., Navrotsky, A., Boatner, L.A., 2001. Thermochemistry of rareearth orthophosphates. Journal of Materials Research 16, 2623–2633. Wing, B.A., Ferry, J.M., Harrison, M., 2003. Prograde destruction and formation of monazite and allanite during contact and regional metamorphism of pelites: petrology and geochronology. Contributions to Mineralogy and Petrology 145, 228–250.