- © 2015 by the Mineralogical Society of America
Apatite is a superb mineral by which to investigate the nature of fluids that have passed through and altered a rock (metasomatic processes). Its ubiquity allows it to act as a reservoir for P, F, Cl, OH, CO2, and the rare earth elements. It is also a powerful thermochronometer and can be chemically altered by aqueous brines (NaCl–KCl–CaCl2–H2O), pure H2O, and aqueous fluids containing CO2, HCl, H2SO4, and/or F. Thus, apatite is the perfect tracker of metasomatic fluids, providing information on the timing and duration of metasomatism, the temperature of the fluids, and the composition of the fluids, all of which can feed back into the history of the host rock itself.
- magnetite–apatite ores
- lithospheric mantle
What is Metasomatism? What is Apatite? How are the Two Related?
“Metasomatism” is the word used to describe a metamorphic process in which the chemical composition of a rock is altered in a pervasive manner by a fluid or fluids, normally with an aqueous component, which moves along grain boundaries or cracks in the rock. Metasomatism, therefore, involves the introduction and/or removal of chemical components (mass transfer) as a result of the interaction of the rock with these fluids. Fluid-aided mass transfer and subsequent mineral re-equilibration are the two defining features of metasomatism. Taking into account geological time scales, the fluid volume and flow rate are not limiting factors in metasomatism. However, the fluids must be able to move along grain boundaries and/or cracks in the rock and be chemically reactive with the minerals that they encounter such that mass transfer is promoted. The passage of fluids through a rock can be deduced from up to five lines of evidence: (1) altered mineral composition; (2) partial to total re-equilibration of mineral phases; (3) reaction textures along mineral grain boundaries; (4) formation of mineral inclusions; (5) trails of fluid inclusions through the rock's minerals.
The calcium phosphate apatites [Ca5(PO4)3(F,Cl,OH)] are one of the most common accessory minerals (actually, a mineral group) in sedimentary, metamorphic, and igneous rocks. Apatite is highly susceptible to various fluid-induced (metasomatic) chemical and textural changes over a wide range of pressures and temperatures, from the Earth's surface to the lithospheric mantle, making it an ideal mineral for “fingerprinting” metasomatic processes. Importantly, apatite can serve as the principal host for the rare earth elements (REEs) in most rocks. REEs are incorporated into or removed from the apatite structure via the coupled substitution reactions (Pan and Fleet 2002; Hughes and Rakovan 2015 this issue): and The amount of REEs in apatite is linked to whether apatite has undergone metasomatism. For example, REEs can be metasomatically removed from apatite to form other REE-bearing minerals, such as monazite [(Ce,La,Nd,LREE) PO4] and xenotime [(Y,HREE)PO4], which are commonly found as inclusions in apatite.
A TOOL FOR TRACKING FLUID INTERACTION IN IGNEOUS AND METAMORPHIC ROCKS
Partitioning of F, Cl, and OH into Apatite
Studies of natural apatite (Pan and Fleet 2002), as well as experimental studies (Schettler et al. 2011), have shown that F, Cl, H2O (as OH), as well as CO3, can substitute extensively, if not totally, for each other on the halogen site (the anion column; see Hughes and Rakovan 2015 this issue). The proportions of F, Cl, OH, and CO3 within apatite will depend on three factors (Zhu and Sverjensky 1991): (1) the melt and/or fluid composition; (2) the presence of other F- and Cl-bearing minerals, such as biotite, muscovite, amphibole, or scapolite; and (3) the P–T conditions. Of the four major anions, F heavily partitions into apatite in metamorphic rocks and most quartz-bearing igneous rocks (where fluorapatite serves as the major sink for F), whereas Cl, OH, and CO3 tend to be found only in minor amounts. The preference of F for apatite is reflected in the F, Cl, OH, and CO3 partitioning data between apatite and fluids (Fig. 1). This behavior, coupled with the degree of partitioning of F and Cl between apatite and co-existing biotite (presumed to be in equilibrium) has allowed an apatite–biotite F–Cl geothermometer to be developed (Zhu and Sverjensky 1992; Sallet 2000). Chlorine, OH, and/or CO3 only become major anions in the anion column when apatite forms in mafic and ultramafic igneous rocks (quartz absent), in lithospheric mantle rocks, or when apatite is a biomaterial (Boudreau et al. 1995; O'Reilly and Griffin 2000; Krause et al. 2013; Rakovan and Pasteris 2015 this issue).
A Recorder of Fluid Flow in the Lithospheric Mantle
O'Reilly and Griffin (2000) have demonstrated that apatite is widespread in the Phanerozoic lithospheric mantle. In a study of apatite-bearing lithospheric mantle xenoliths from a series of volcanic fields located in Australia, Alaska, Germany, and France, they further demonstrated that the apatite could be divided into two geochemically distinct groups on the basis of their F, Cl, and OH content and on the presence or absence of structural CO3, Sr, and trace elements (e.g. U, Th, and REE). The first apatite group had high F (1.2–2.6 wt%) and low Cl (0.1–0.5 wt%), with no detectable CO3. These apatites have a composition consistent with high-pressure crystallization from magmas whose composition ranged from silicate to carbonate. The second group of apatites had high amounts of Cl (1.5–2.5 wt%) and CO3 (0.7–1.7 wt%), and low amounts of F (0.2–0.3 wt%). Because of their different composition, these latter apatites recorded evidence of metasomatism of the first apatite group by CO3- and Cl-rich fluids derived from a primitive mantle source region.
Chlorapatite as a Fluid Tracer in Mafic- to Ultramafic Rocks
Chlorapatite [Ca5(PO4)3Cl] is a common mineral in layered mafic intrusions (e.g. Boudreau and McCallum 1990; Boudreau et al. 1995) and in veins and dikes associated with Cl-endmember scapolite (marialite) such as at the Ødegårdens Verk in Norway (Harlov et al. 2002b). In both cases, chlorapatite acts as a tracer for localized metasomatic processes. Chlorapatites from the Stillwater Complex in the USA (Boudreau and McCallum 1990) and the Ødegårdens Verk possess numerous inclusions of monazite and/or xenotime in areas of the chlorapatite that have been altered by fluids to carbonated chlorhydroxylapatite [Ca10(PO4,CO3)6(Cl,OH,F)2] (Stillwater Complex) or hydroxyl-fluorchlor apatite [Ca5(PO4)3(F,Cl,OH)] (Ødegårdens Verk; Fig. 2A,B). In both locations, the metasomatized areas in the chlorapatite are also heavily depleted in REEs, which directly have contributed to the formation of monazite and xenotime.
In contrast, fluorapatite from an ultramafic, nepheline-bearing clinopyroxenite—located in mafic–ult ramafic, Uralian–Alaskan-type complexes from Kytlym and Nizhny Tagil (Ural Mountains, Russian Federation)—has been metasomatized to chlorapatite (Krause et al. 2013). This includes both matrix grains and apatite inclusions in altered areas of the clinopyroxene phenocrysts. The fluid responsible was a CaCl2-enriched brine derived from the alteration of the matrix Ca-bearing plagioclase by NaCl brines.
Fingerprinting Metasomatism During Metamorphism
Fluorapatite is a common accessory mineral in most metamorphic rocks and generally has been inherited from the protolith. In granulite-facies rocks, monazite and/or xenotime inclusions in fluorapatite are common (Fig. 3A–D, Fig. 4A) (e.g. Hansen and Harlov 2007). However, these inclusions can also be found in fluorapatites from amphibolite-facies and lower-grade rocks under certain conditions (Pan et al. 1993). Inclusion growth has been directly linked to metasomatism by aqueous fluids containing H2O, CO2, and (K,Na) Cl during and after the peak of metamorphism (Harlov and Förster 2003; Hansen and Harlov 2007). These fluids could originate either from outside the immediate rock system (be introduced externally) or be the result of internal high-grade mineral reactions. Evidence for this is seen via transmission electron microscopy (TEM) of apatite TEM foils (cf. Harlov et al. 2005) containing monazite inclusions (Fig. 4B). Here voids, presumably once fluid filled, are seen at specific points along the monazite–apatite interface. Monazite and xenotime inclusions can also form in fluorapatite from H2O-rich metapelites that have experienced partial melting during granulite-facies metamorphism (Harlov et al. 2007), apparently due to alkali-bearing fluids expelled during crystallization of the alkali–SiO2-rich melt.
Magnetite–Apatite Ore Deposits: Magmatism and Metasomatism
Magnetite–apatite ore deposits occur worldwide and represent highly evolved bodies, generally associated with volcanism, in which the apatite records varying degrees of fluid–rock interaction starting from shortly after crystallization and continuing down to ambient P–T conditions. Notable examples include the Kiirunavaara and Grängesberg ore deposits in Sweden (Harlov et al. 2002a; Jonsson et al. 2010); the Mineville and Pea Ridge ore deposits in the USA (Sidder et al. 1993; Lupulescu and Pyle 2008); and a series of magnetite–apatite ore deposits located in the Bafq region of Iran (Daliran et al. 2010).
In each of these ore deposits, monazite and/or xenotime inclusions are commonly found in apatite that has experienced fluid-induced depletion of REE + Na + Si + Cl (Fig. 5A–E). Once formed, these inclusions were later reworked via subsequent metasomatic and deformational events, which resulted in: (1) Ostwald ripening where the larger monazite xenotime grains have grown larger by consuming the smaller grains, thereby reducing the total number of inclusions (Fig. 5A,B,E); (2) intergrowth of the monazite and xenotime grains with magnetite (Fig. 5E); and (3) later-stage fluid-aided formation of allanite via the reaction of monazite and xenotime with the surrounding silicate minerals (Fig. 5D).
REPLICATING THE FLUID–MINERAL RELATIONSHIP BETWEEN APATITE, MONAZITE, AND XENOTIME IN THE LABORATORY
The hypothesis that fluids can induce monazite and xenotime inclusions to form in apatite in nature under a wide range of P–T conditions is supported experimentally (Fig. 6A–E, Fig. 7A–C) (Harlov et al. 2002b, 2005; Harlov and Förster 2003). The formation and growth of monazite and xenotime as inclusions or as rim grains are the result of coupled dissolution–reprecipitation processes during the metasomatic alteration of apatite (Putnis 2009). Both inclusions, and rim grains, can form over a wide P–T range, starting at near-surface pressures and 100 °C (Boudreau and McCallum 1990; Harlov et al. 2005). Whether nucleation will occur is highly dependent on the level of reactivity between the fluid and the apatite, and the amount of REEs available.
Regions of the apatite affected by coupled dissolution–reprecipitation are characterized by crystallographic continuity between the reacted and unreacted apatite (Fig. 6C) and a pervasive, interconnected, three dimensional micro- and nano-porosity in the reacted regions (Fig. 6B,D, Fig. 7B) that allows fluids to infiltrate the apatite structure (Harlov et al. 2005; Putnis 2009). The presence of an interconnected pore structure greatly increases the rate of mass transfer, giving rise to the rapid (hours–days) nucleation and growth of monazite and/or xenotime inclusions.
During metasomatic alteration of the apatite, Na and/or Si are preferentially removed out of the apatite–fluid system compared to the REEs. Once sufficient concentrations of REEs have been reached, nucleation and growth of monazite and/or xenotime inclusions (Fig. 6E, Fig. 7B) will occur via the following two general mass transfer reactions (cf. reactions 1 and 2): and Because the monazite and/or xenotime are derived directly from the host apatite, their composition will reflect the amount of (Y + REE + Th + U) available in the host apatite. This is seen in the generally low amounts of Th and U in the monazite and xenotime, which reflect the trace amounts of Th and U generally found in apatite (e.g. Hansen and Harlov 2007), although some exceptions exist (e.g. Harlov and Förster 2003).
The micro- and nano-pores within the interconnected pore system provide nucleation sites for the formation of monazite and xenotime inclusions (Fig. 6B,E). While these mineral inclusions can be euhedral to subhedral or occur in clumps that have grown in the fluid-filled voids (Fig. 6B,E), the monazite inclusions can also be highly elongated parallel to the apatite c axis along the monazite b axis, suggesting an epitaxial relationship (Fig. 7A–C; see also Figs. 3C,D and 4A) (Pan et al. 1993; Pan and Fleet 2002). In some experiments, monazite inclusions in fluorapatite have been made to undergo Ostwald ripening by deliberately prolonging the experiment for 2 to 3 times the normal length (Harlov et al. 2005). This implies that even after formation, fluids are still affecting inclusion morphology and growth.
Fluids found to induce the formation of monazite and/or xenotime inclusions in apatite, include pure H2O, H2O–CO2 fluids, and aqueous solutions with KCl, H2SO4, and HCl (Harlov et al. 2002a, 2005; Harlov and Förster 2003). However, Na in NaCl brines discourages the growth of monazite and/or xenotime inclusions and rim grains. Aqueous Na prevents Na from leaving the apatite structure such that charge balance is maintained (see reactions 1 and 3), therefore inhibiting REE from forming monazite and xenotime. Exceptions to this observation are seen in solubility experiments involving a REE-bearing fluorapatite in NaCl brines at 700–900 °C and 700–2000 MPa (Antignano and Manning 2008). Here, small crystals of monazite formed on the surface of the fluorapatite during its partial, incongruent dissolution in the brine. This behavior is also seen in the incongruent dissolution of fluorapatite in peraluminous granitic melts (Wolf and London 1995).
APATITE, RECORDING THE FINGERPRINT OF METASOMATISM
As has been deduced from nature (e.g. Hansen and Harlov 2007) and from experiments (e.g. Harlov et al. 2005), the F, Cl, OH, and CO3 content in apatite, coupled with the possible presence of monazite and xenotime inclusions, can serve as valuable “fingerprints” for recording metasomatic events in rocks. This “fingerprint” helps constrain the composition of the infiltrating fluids responsible both for the metasomatism of the apatite and for the rock as a whole. In this respect, apatite serves as a valuable recorder of metasomatic events—helping to demystify an otherwise often mysterious process—that can occur over a wide range of temperatures and fluid compositions. Just as importantly, the presence of coexisting biotite, along with monazite, and xenotime inclusions in the apatite, can give some indication of the temperature of a metasomatic overprint via the biotite–apatite (Zhu and Sverjensky 1992; Sallet 2000) and monazite–xenotime (Gratz and Heinrich 1998) geothermometers. Lastly, metasomatic events can be timed using monazite–xenotime U–Pb geochronology (Williams et al. 2007) and/or apatite U–Pb, (U–Th)/He, and Lu–Hf thermochronology (see Chew and Spikings 2015 this issue).
The articles in this issue of Elements all point to the amazing versatility of this common accessory mineral. The fact that apatite can be chemically altered by metamorphic and igneous fluids and so give information on the composition of these fluids; the fact that re-equilibration temperatures can be estimated; and the fact that re-equilibration and accompanying mass transfer can be dated over a broad temperature range, demonstrate the incredible amount of information that may be gleaned with regard to changing P–T–X conditions in the host rock from this seemingly minor and yet ubiquitous mineral.
Collaborations and discussions with various colleagues over the years and technical support, under the direction of Wilhelm Heinrich, at the GeoForschungsZentrum have all contributed to this article. Reviews by Yuanming Pan, Andrew Putnis, and Alan Treiman along with colleagues and students at Miami University (USA), Oxford, Ohio and the Indian Institute of Science, Bangalore (India) were of great help in the revision of the manuscript.