The Sulphide Queen carbonatite, which contains the rare-earth orebody at Mountain Pass in California, is a moderately dipping tabular intrusion associated with 1.40 Ga alkaline plutons of similar size and orientation, as well as abundant carbonatite and alkaline dikes. The Mountain Pass carbonatite and alkaline rocks lie in a narrow belt of occurrences of alkaline rocks 130 km long. Both the carbonatite and the associated alkaline rocks are chemically and mineralogically unusual. The mafic members of the Mountain Pass alkaline rock suite belong to the ultrapotassic group, and are most akin to the lamproite subdivision of that group. These rocks are part of a unique ultrapotassic suite in that it ranges from typical mafic lithologies to silicic granite. The association with significant amounts of carbonatite is also unique. Chemical similarities suggest that the carbonatite, which has yielded slightly younger (1.38 Ga) ages, is related to the ultrapotassic rocks. It is not clear if the carbonatite and associated alkaline rocks came from a common magmatic parent, but they likely came from a common enriched source in the mantle.
- rare-earth mineralization
- ultrapotassic alkaline rocks
- enriched mantle
- Sulphide Queen
- Mountain Pass
Since its discovery in the early 1950s, the carbonatite orebody at Mountain Pass has been discussed widely in the geological literature. For many years, it was the world’s premier source of rare earths. Its unique nature has been noted by many researchers (e.g.,Olson et al. 1954, Heinrich 1966, Jones & Wyllie 1983, Mariano 1989, Castor & Nason 2003, Haxel 2005). Bastnäsite, the most abundant REE mineral in the carbonatite, is of primary igneous origin, a feature that is generally not shared with other carbonatites (Mariano 1989).
The U.S. Geological Survey Professional Paper of Olson et al.(1954) still stands as an excellent source of information on the deposit and its geological setting. Although a considerable body of knowledge was acquired through intense study of the orebody and the nearby geological setting by company geologists, mainly between 1979 and 1989, only minor amounts of data about the orebody have been published since 1954 because of company reticence to provide public information, a policy that has been relaxed in recent years.
The main objectives of this contribution are to present geological, geochemical and mineralogical data on the Mountain Pass carbonatites and alkaline rocks that were collected since 1954, and to offer some thoughts on the genesis of these rocks. I was the Chief Geologist at Mountain Pass between 1984 and 1989, and much of the information that follows was acquired during that period.
The Mountain Pass operation of Chevron Mining Inc. (originally under the name of Molycorp Inc.) produced rare earths since 1954, and from the 1960s to the mid-1990s was the world’s largest source of light rare-earth commodities. After 1998, sales of rare-earth materials from Mountain Pass declined substantially owing to environmental constraints and international competition (mostly from China). Mining and rareearth separations ceased by 2002; however, extraction of selected REE commodities from stockpiles resumed in 2007.
The Mountain Pass rare-earth orebody, which is in the Sulphide Queen carbonatite, has current reserves of more than 20 million tonnes of ore with an average grade of 8.9% rare-earth oxide (Castor & Nason 2004). The ore, which typically contains 10–15% bastnäsite- (Ce), is mostly composed of calcite, dolomite and barite, along with generally minor amounts of other minerals. All available data indicate that Ce is the dominant REE in primary bastnäsite and other REE minerals in carbonatite from Mountain Pass; therefore, the -(Ce) suffix will be omitted in most cases. However, few detailed data on REE content are available for some REE minerals that are considered to be secondary, such as sahamalite.
Unlike most other carbonatites, which are associated with sodic alkaline rocks, the carbonatite at Mountain Pass is associated with ultrapotassic intrusive rocks that occur in a narrow north-trending belt in southeastern California (Fig. 1). These rocks are part of an alkaline suite that ranges from mafic (shonkinite) through syenite to granite. The Mountain Pass carbonatite and alkaline rocks do not form the concentric, circular to ovoid masses that are characteristic of large alkaline silicate – carbonatite complexes worldwide. Instead, the Sulphide Queen carbonatite and the associated alkaline plutons constitute a group of crudely tabular or lensoid, moderately to shallowly west-dipping intrusions that trend north–northwest in a portion of the of the ultrapotassic rock belt 10 km long. Figure 2, a stacked series of cross sections produced from drill data, shows the geometry of the Sulphide Queen carbonatite and a large associated composite mass of alkaline rocks. For the purposes of this paper, the area containing carbonatites and alkaline plutons as originally mapped by Olson et al.(1954) is divided into the Mountain Pass area, which lies north of Interstate Highway 15, and the Mineral Hill area, which lies south of the highway (Figs. 1, 3). In addition to the large intrusive masses, abundant and generally steeply dipping dikes of carbonatite and alkaline rocks also are present in the Mountain Pass and Mineral Hill areas.
The composition of the Mountain Pass carbonatite is unusual, even among carbonatites. It is extremely enriched in some elements, such as the light rare-earth elements (LREE) and Ba, and depleted in others, such as Nb and P. Although they are not as strikingly unique, the associated alkaline rocks show the same pattern of chemical enrichment and depletion. In this paper, the unusual properties of the Mountain Pass carbonatite are attributed to derivation from the same source that produced the associated ultrapotassic magma or to derivation from the ultrapotassic magma itself.
On the basis of U–Th–Pb and 40Ar/39Ar dating, shonkinite at Mountain Pass was emplaced at 1410 ± 5 Ma, followed by syenite intrusion at 1403 ± 5 Ma (Dewitt et al. 1987). The granitic rocks were not dated, but are assumed to be close in time to 1400 Ma on the basis of mineralogical similarities with the syenite. Carbonatite emplacement closely followed the alkaline magmatism, but yielded Th–Pb dates of 1375 ± 5 Ma on monazite, indicating a significant gap between alkaline and carbonatite intrusion (Dewitt et al. 1987).
Sample Locations and Methods of Analysis
Analyzed samples from the Mountain Pass and Mineral Hill areas were taken from outcrops, the openpit mine, and drill holes. Samples from other areas are from outcrops only. Locations and brief descriptions are given in Table 1.
Whole-rock and some trace-element analyses of the alkaline igneous rocks (Table 1) were performed by fusion ICP–AES (inductively coupled plasma – atomic emission spectrometry) in 1994 (Acme Laboratories, Vancouver, Canada), except analyses for F, by colorimetry (Nevada Bureau of Mines and Geology Laboratory, Reno, Nevada). Some trace element analyses (* in Table 2) were performed by INAA (instrumental neutronactivation analysis) in 1989 by Bondar Clegg (now a division of Inchcape Inspection and Testing Services), Denver, Colorado. The electron-microprobe analyses of minerals in the alkaline igneous rocks (Table 3) were done at the Washington State University Geosciences Microprobe Laboratory. Whole-rock analyses of carbonatite samples (Table 4) were mostly performed by fusion ICP–AES in 1986 (Skyline Labs, Inc., Tucson, Arizona), but some whole-rock and trace-element data (* in Table 4) were obtained using fusion ICP–AES in 1994 (Acme Laboratories, Vancouver, Canada). Rareearth oxide contents (REO) were measured by XRF (X-ray fluorescence) in 1986 by the Molycorp Inc. Mountain Pass laboratory. Total REO and major oxides in bastnäsite concentrates (Table 5) were performed by XRF, and individual REO analyses by wet-chemical methods at the Molycorp Inc. Mountain Pass laboratory in 1987.
The Paleoproterozoic Rocks
Host rocks for the carbonatite and alkaline intrusions in most of the Mountain Pass belt are complexly folded, banded, granulite-facies gneiss and schist with variable amounts of quartz, microcline, biotite, garnet, hornblende, orthopyroxene, and sillimanite. Large masses of weakly foliated granitic rocks are also present in the Mountain Pass area and regionally (DeWitt et al. 1989, Wooden & Miller 1990, Miller & Wooden 1993). The earliest supracrustal rocks of Proterozoic age (1.9–2.3 Ga) were intruded by 1.72–1.76 Ga plutons, metamorphosed with synchronous plutonism between 1.70 and 1.71 Ga, and cut by postorogenic intrusions at 1.63–1.69 Ga (Wooden & Miller 1990).
Large areas of Middle Proterozoic granite were mapped to the east of Mountain Pass in Nevada by Volborth (1962) and Bingler & Bonham (1973). Kwok (1987) and Anderson & Bender (1989) studied Middle Proterozoic granites in southeastern California, southern Nevada, and western Arizona (Fig. 1), reporting 1.4 to 1.5 Ga ages for them. These rocks form the southwestern part of a wide northeast-trending zone containing Proterozoic anorogenic granitic rocks that transects North America, Greenland, and Scandinavia prior to the breakup of the Precambrian continent of Laurentia (Anderson 1983). This zone contains several Precambrian rare-earth deposits in addition to Mountain Pass (Castor 1993).
The Ultrapotassic Alkaline Rocks
The potassic character of the alkaline intrusions at Mountain Pass has been known for some time (Olson et al. 1954). More recently, their ultrapotassic nature has been noted (Castor & Gleason 1989, Haxel 2005). However, no comprehensive comparison of these rocks with classic ultrapotassic provinces has been made. Mountain Pass is located in a narrow belt 130 km long in southeastern California that is defined by occurrences of ultrapotassic rocks (Fig. 1) (Castor & Gleason 1989, Castor 1991). This belt lies near the southwestern end of the anorogenic granite zone noted above.
The presence of alkaline rocks similar to those at Mountain Pass in the Barrel Spring area of the Piute Mountains about 100 km to the south of Mountain Pass (Gleason 1988) was a key discovery. However, Molycorp geologists also recognized independently in the late 1980s that the belt extended outside the 13-kmlong area originally defined by Olson et al.(1954). I found minor occurrences of ultrapotassic dike rock to the north of Mountain Pass near Mesquite Lake and to the south as far as the Bobcat Hills in 1988 and 1989. Ultrapotassic intrusions in the northern part of this belt, including those in the Mountain Pass area, are unmetamorphosed rocks that have not been changed except by local hydrothermal alteration. In contrast, both the ultrapotassic intrusions and the host rocks in the south part of the belt (from the Bobcat Hills to Rattlesnake Canyon) were overprinted by Mesozoic greenschistgrade metamorphism.
Mafic (<55 wt.% SiO2) alkaline rocks in the Mountain Pass and Mineral Hill areas contain more than 3 wt.% K2O, 3 wt.% MgO, and have K2O:Na2O ratio > 2 (Table 2), satisfying the definition for ultrapotassic rocks of Foley et al.(1987). These data also satisfy many of the chemical and mineralogical requirements listed by Bergman (1987) and Scott-Smith (1996) for the lamproite class of ultrapotassic rocks. However, they differ somewhat in mineralogy from classic lamproitic occurrences, are significantly different in the range of silica content, and are generally higher in Al and lower in Ti contents. In addition, the Mountain Pass ultrapotassic rocks are unique in their association with significant amounts of carbonatite. Mitchell (2005) cited five occurrences of carbonatite associated with potassic rocks elsewhere in the world; however, these occurrences are either minor in terms of carbonatite volume, associated with rocks that are merely potassic, or associated with an alkaline series that includes nepheline-bearing rocks.
In the Mountain Pass and Mineral Hill areas, fine- to coarse-grained alkaline rocks (Figs. 4A–C, 5A, B, F) occur in plutons as much as 1.8 km long. By comparison, the dike rocks are predominantly fine-grained (Figs. 4D–F). Olson et al.(1954) divided the alkaline rocks into four types on the basis of mineralogy: shonkinite, syenite, quartz syenite, and granite. Shonkinite is dark syenite with more than 60% mafic minerals in the mode according to Le Maitre (2002). Although a potassic rock-type, shonkinite is not considered an ultrapotassic rock. The term is used here for mafic alkaline rock in the Mountain Pass area for continuity with previous work. Much of the rock originally mapped as shonkinite in the area is a dark-colored rock with less than 60% mafic minerals, or alkali feldspar melasyenite (Le Maitre 2002). Rocks termed syenites here are light in color, contain less than 30% mafic minerals, and are mostly alkali feldspar syenite (Le Maitre 2002).
Shonkinite and melasyenite comprise about 80% of the exposed alkaline rocks in the Mountain Pass – Mineral Hill area; syenite, quartz syenite, and granite are relatively minor. In general, the alkaline intrusions have cross-cutting relationships, indicating progression from older mafic rocks to younger silicic rocks (Olson et al. 1954). Shonkinite and melasyenite plutons are cut by syenite, quartz syenite, and granite dikes. The dikes commonly have fine-grained idiomorphic (Fig. 4D) to finely aplitic textures. Some of the most silicic consist of porphyry with a microcrystalline matrix (Fig. 4E). Late, fine-grained shonkinite and melasyenite (Fig. 5D) dikes are exceptions to the mafic-to-felsic intrusive sequence. Several such dikes cut granite and syenite (Olson et al. 1954), but in places show evidence of co-intrusion with syenite (Haxel 2005). Rare dikes of relatively coarse melasyenite that contain late granular carbonate (Fig. 5E) have been found in the Mineral Hill and New Trail Canyon areas. In places, complex breccias along borders of alkaline plutons contain several alkaline rock-types and indicate, by cross-cutting relationships, successive intrusion of more and more felsic rock-types.
In most parts of the belt of ultrapotassic rocks outside the Mountain Pass – Mineral Hill area, the alkaline rocks are mainly very fine-grained dikes of shonkinite or melasyenite (Fig. 4F) less than 5 m wide. However, the Barrel Spring area contains a large composite pluton of quartz syenite and porphyritic granitic nearly 10 km long (Gleason 1988), along with several masses of shonkinite and melasyenite (Fig. 5C).
Mineral Assemblages in the Alkaline Rocks
Shonkinite and melasyenite in the Mountain Pass – Mineral Hill area are dark green to mottled green and pink or purple rocks (Figs. 4A, B) that contain major amounts of phlogopite and pyroxene with variable amounts of K-feldspar, amphibole, and olivine (Table 1, Figs. 5A, B). Apatite and magnetite are accessory, whereas ilmenite, rutile, zircon, fluorite, barite, monazite, baddeleyite and thorite occur in trace amounts. Titanite is absent or rare. The late melasyenite dikes consist of a fine-grained gray rock with mica and pyroxene in a groundmass of K-feldspar (Fig. 5D). The mineral assemblages of mafic rocks are discussed in some detail here because of their similarity to those of lamproites.
Phlogopite in Mountain Pass – Mineral Hill shonkinite and melasyenite occurs mostly as 1- to 8-mm books with reddish brown to yellow pleochroism (Fig. 5A, B). Electron-microprobe analyses (Table 3) reveals Mg/ Fe values of 3.7 to 5.2, and TiO2 and BaO are more than 4% and 3%, respectively. Mica in the metamorphosed mafic rocks from the south part of the belt of ultrapotassic rocks is olive-brown to green (Fig. 5C) magnesian biotite relatively low in Ti and Ba (sample BoH, Table 3).
Pyroxene in Mountain Pass – Mineral Hill shonkinite and melasyenite has optical characteristics of diopside, and electron-microprobe analyses indicate it to be rich in Ca and Mg and low in Fe, Ti, and Na (Table 3). Such diopside is typical of lamproites (Mitchell & Bergman 1991), but the Mountain Pass diopside is somewhat higher in Al than that in many lamproites. Pyroxene is rare or lacking in the rocks from the southern part of the ultrapotassic belt.
Olivine is generally altered to fine talc or serpentine (or both), but unaltered olivine was reported by Olson et al.(1954) and later found during examination of drill samples by Molycorp geologists (Fig. 5B). Olivine, altered or not, is lacking in some shonkinite and melasyenite, and the most mafic Mountain Pass rock analyzed (sample CSH, Table 2, Fig. 5A) does not contain olivine. The olivine is magnesian (Table 3), which is typical of lamproites (Mitchell & Bergman 1991). No olivine or evidence of pre-existing olivine was found in rocks outside the Mountain Pass – Mineral Hill areas; greenschist-grade metamorphism probably destroyed any olivine present in shonkinite and melasyenite in the Barrel Spring and Rattlesnake Canyon areas to the south.
Sparse, late, primary amphibole in shonkinite and melasyenite has white to yellow or pale pinkish brown pleochroism (Fig. 5A) that is consistent with a potassium-rich richterite composition (Table 2). Magnesioriebeckite occurs in shonkinite from the large mass north of the Sulphide Queen carbonatite, but this is probably a later amphibole related to fenitization during intrusion of the carbonatite. Pale green to bluish green pleochroic amphibole occurs abundantly in the mafic rocks from the Bobcat Hills and Barrel Spring in the south part of the ultrapotassic belt (Table 1, Fig. 5C). It is actinolite, but enriched in Na and K (Table 2).
The K-feldspar in Mountain Pass – Mineral Hill ultrapotassic rocks is microcline. It is interstitial in the shonkinite and melasyenite, locally forming poikilitic masses >2 cm in diameter. It has compositions of Or92 to Or97, contains little or no Na, and has minor to significant amounts of Ba (Table 3), exhibiting a similar composition to sanidine in lamproites (Mitchell & Bergman 1991). Some Mountain Pass microcline has extremely high levels of Ba (nearly 3.0% BaO, Table 3) even if compared with K-feldspar in lamproitic rocks, which contain as much as 1.9% BaO (Wagner & Velde 1986). The most mafic shonkinite at Mountain Pass contains <5% microcline, which ranges from high-Ba to low-Ba, with enclaves of barite. Melasyenite from the Sulphide Queen area and the Tors area 1 km to the southeast contains cloudy grains of pseudoleucite, some with a distinctly trapezohedral shape (Fig. 5B), surrounded by clear, late high-Ba microcline. This pseudoleucite is now mainly low-Ba K-feldspar that contains patches of an unidentified K–Na aluminum silicate and small grains of Fe-rich biotite. It resembles the turbid pseudoleucite described in lamproite by Mitchell & Bergman (1991). The K-feldspar in the melasyenite dike-rock from the Bobcat Hills contains less Fe and lower (but detectable) Ba than that from Mountain Pass (Table 3).
The REE minerals identified during SEM examination in mafic rocks from the Mountain Pass and Mineral Hill areas include monazite, ancylite, allanite, and an unknown Th + Zr + REE + Ca silicate mineral (Table 2). Roeder et al.(1987) reported more than 2% REO in apatite from Mountain Pass shonkinite. In the southern part of the ultrapotassic belt, allanite is typically the only REE mineral (Table 2).
Magnetite with as much as 8% Cr2O3 generally comprises 1–2% of the shonkinite. Fluorite, generally a trace mineral, makes up as much as 1% of some samples of shonkinite. Baddeleyite, apparently the primary Zr phase in Mountain Pass shonkinite (Table 2) because of its partial replacement by zircon in one sample (Fig.»6A), was also found in minor amounts in the Mesquite Lake melasyenite. Barite is common in some samples of shonkinite and melasyenite (e.g., late shonkinite FSH, Table 2); however, in others, high Ba contents are mainly accounted for by high Ba in feldspar and phlogopite (e.g., TORSH and NT2, Table 3). Titanite is lacking or present in trace amounts in mafic ultrapotassic rocks from the Mountain Pass – Mineral Hill area, but is common in some samples from the southern part of the ultrapotassic belt (Table 2), where Ti may have been remobilized during metamorphism.
Most rocks mapped as syenite in the Mountain Pass – Mineral Hill area contain less than 25% mafic minerals. They are light-colored, fine- to coarse-grained rocks (Fig. 4C) that grade into quartz syenite. Syenite from the Tors area typically contains more than 65% perthitic K-feldspar. Albite is rare as separate grains. Accessory minerals are Mg–Na amphibole, phlogopite or magnesian biotite, and aegirine (sample TORSY, Table 2). Trace minerals are apatite, rutile, magnetite, ilmenite, pyrite, zircon, thorite, allanite, and monazite.
Fine-grained, leucocratic syenite (Fig. 4D) and quartz syenite mostly form dikes in the Mountain Pass area. Some are porphyritic (Fig. 4E). Mafic minerals in these rocks are generally highly altered. Quartz syenite porphyry dikes contain abundant perthitic microcline, sparse mica, and rare amphibole phenocrysts in a matrix of fine-grained K-feldspar, quartz, and sodic amphibole. The syenite contains zoned phenocrysts of mica with phlogopite in the core and magnesian annite in the rim. The K-feldspar contains no detectable Ba, but barite enclaves were identified by SEM–EDS. Titanite, a minor accessory in some samples of quartz syenite from the Mineral Hill area, contains inclusions of monazite and possible bastnäsite and synchysite. Zoned crystals of zircon up to 0.5 mm long are present in trace amounts.
Syenite and quartz syenite in the southern part of the ultrapotassic belt is mostly perthitic K-feldspar with minor quartz and accessory bitotite. Amphibole is lacking or occurs in minor amounts, and has the same pale green to pale bluish green pleochroism as that in the more mafic rocks there. Separate grains of plagioclase are rare or lacking. Trace minerals are allanite, zircon, thorite, and monazite (Table 2).
Most granite intrusions at Mountain Pass are aplite or porphyry dikes, but significant amounts of granite occur in the large alkaline rock mass north of the Sulphide Queen carbonatite, and a 300 m × 500 m pluton composed mostly of medium-grained granite occurs at Mineral Hill, about 5 km southeast of the Sulphide Queen carbonatite (Olson et al. 1954). Relatively unaltered samples of Mineral Hill granite consist mostly of perthitic microcline, quartz, and albite, with accessory aegirine, sodic amphibole and ilmenite, and with traces of biotite, fluorite, rutile, and thorite. Trace amounts of zircon occur in relatively coarse, zoned, euhedral crystals (Fig. 5G) that are similar to those in the syenite and quartz syenite.
Composition of the Alkaline Rocks
The mafic rocks in the Mountain Pass alkaline igneous suite have chemical features of ultrapotassic suites such as the lamproitic rocks in the Leucite Hills, Wyoming (Kuehner et al. 1981, Mirnejad & Bell 2006), the West Kimberley region, Australia (Wade & Prider 1940, Jaques et al. 1984, Nixon et al. 1984), southeastern Spain (Venturelli et al. 1984, Nixon et al. 1984), and Holsteinsborg, Greenland (Scott 1979). In addition to high potassium, ultrapotassic rocks have relatively low sodium, and the ratio K2O/(Na2O + K2O) generally does not increase with increasing silica (Fig. 7A) as it does for other igneous series. In this way, the Mountain Pass alkaline rocks are distinct from the relatively sodic alkaline rocks typically associated with carbonatite. With the exception of the rocks from southeastern Spain, most ultrapotassic suites do not include rocks with more than 60% SiO2. However, the Mountain Pass suite is distinctive among ultrapotassic suites because it includes granite with more than 75% SiO2 (Table 2). As in other utrapotassic suites, Al is low in the mafic alkaline rocks in the Mountain Pass belt, but is present in greater amounts in the syenites and granites (Table 2). The MgO contents of the mafic members of the Mountain Pass alkaline rocks, at 6 to 15%, are similar to levels of MgO in most other ultrapotassic suites, and Mountain Pass rocks have CaO and Fe oxide contents similar to those in other ultrapotassic suites; however, the level of TiO2 is relatively low.
Whereas BaO is a trace component in most igneous rocks, it exceeds 0.5% in most ultrapotassic rocks. Some rocks in the Mountain Pass utrapotassic belt have BaO as high as 1.5%, and mafic members of the suite have the highest amounts (Fig. 7B), which is unusual among igneous rock series, but not in ultrapotassic suites. In addition, the Mountain Pass mafic rocks have elevated concentrations of both compatible trace elements (e.g., Cr, Ni) and incompatible trace elements (e.g., Ce, Th, Zr) (Table 2, Figs. 7C–F), a characteristic of ultrapotassic rocks (Foley et al. 1987). Comparison of Ni with Mg number indicates that most of the mafic rocks of the Mountain Pass ultrapotassic suite were derived from primitive mantle-derived magmas (Fig. 8), a feature noted by Haxel (2005) for shonkinite dikes in the Mountain Pass area.
The orebody at Mountain Pass is unusual among large bodies of carbonatite because of its extreme enrichment in the LREE. Similarly, the ultrapotassic rocks of the Mountain Pass suite show LREE enrichment, even if compared to other ultrapotassic suites (Fig. 7D). The Th content of ultrapotassic rocks is very high for rocks with mafic compositions, and the Mountain Pass suite shows extreme Th enrichment, particularly for silica-rich members (Fig. 7E), some of which contain more than 200 ppm Th (Table 2).
Chondrite-normalized distributions of the REE for Mountain Pass alkaline rocks are characterized by steep, nearly linear, LREE-dominated curves with nonexistent or insignificant Eu anomalies (Fig. 9; Castor 1991). Such patterns of REE distribution are typical of ultrapotassic rocks (Foley et al. 1987).
Spidergram plots of incompatible elements normalized to primitive mantle compositions show the strong correlation between the ultrapotassic rocks of the Mountain Pass ultrapotassic belt and other ultrapotassic rocks. The mafic members of the Mountain Pass suite have a very similar overall pattern of trace-element enrichment to average lamproite (Fig. 10), but have relatively low Ta and Nb, and generally lower Hf and Ti. This pattern is consistent, regardless of whether the rocks are relatively unaltered specimens from Mountain Pass – Mineral Hill or altered or metamorphosed rocks from elsewhere in the ultrapotassic belt. Spidergram patterns for more silicic members of the ultrapotassic belt diverge more widely from the average lamproite, but generally show the same overall pattern of enrichment (Fig. 10).
The Mountain Pass Carbonatites
Carbonatite at Mountain Pass includes the large tabular Sulphide Queen carbonatite mass, which is as much as 150 m thick and was about 700 m by 150 m in plan at the surface prior to mining (Fig. 2) (Olson et al. 1954). The Mountain Pass and Mineral Hill areas also contain carbonatite dikes that range from a few millimeters to 3 m thick. Such dikes are abundant in the vicinity of the Sulphide Queen mass, and are much less common in the Mineral Hill and Mineral Springs areas to the south. The total length of the 130-km-long Mountain Pass belt of ultrapotassic rocks that includes carbonatite intrusions is about 15 km.
Although there is considerable mineralogical and chemical variability, carbonatite in the Sulphide Queen deposit consists mainly of bastnäsite – barite sövite (calcite carbonatite) and bastnäsite – barite – dolomite carbonatite (beforsite), or of rock that is intermediate between these two types (bastnäsite–barite dolomitic sövite). These rock types generally constitute ore (5% or more REO) and locally contain as much as 25% REO over 2 m drill-core intervals. On the basis of surface mapping of the carbonatite, early investigators noted that sövite was the dominant type (Olson et al. 1954). However, drill-core sampling, mine mapping, and petrological work by Molycorp geologists showed later that dolomitic carbonatites (beforsite and dolomitic sövite) are more abundant than sövite (Figs. 11, 12, 13). Unpublished petrographic work by R.L. Sherer and A.N. Mariano during the late 1970s and early 1980s was particularly instrumental in unravelling the complexity of the mineral assemblages in carbonatite at Mountain Pass.
In addition to the ore types, several other types of carbonatite occur in, and adjacent to, the orebody. These include parisite–barite sövite and monazite-bearing sövite, dolomitic sövite, and beforsite. The orebody is further complicated by the presence, particularly in the hanging wall, and at its northern and southern ends, of breccia containing variable amounts of carbonatite matrix and altered country-rock clasts. The distribution of carbonatite types in the Sulphide Queen carbonatite body is complex in detail. For the most part, dolomitebearing carbonatite lies above the sövite, although the latter occurs in many places as a thin border along the hanging wall of the carbonate mass. On the basis of detailed (1:600) geological mapping in the pit and drill-core logs acquired during the 1970s and 1980s, the carbonatite body consists of sheets of different types of carbonatite, with some intervening breccia and host rock that is generally arranged in sheet-like masses that parallel the carbonatite sheets (Figs. 11, 12, 13).
Dikes of carbonatite are common in the vicinity of the Sulphide Queen orebody and are present, but less common, in a larger area from 1 km northwest of the orebody to 10 km southeast of the orebody (Olson et al. 1954). Drilling by Molycorp has shown that carbonatite dikes occur very abundantly in an envelope of altered country-rock adjacent to the orebody. Notable among the carbonatite dikes is the Birthday vein, the original discovery site of bastnäsite-bearing carbonatite in the area. The Birthday vein is a moderately to steeply southdipping carbonatite dike as much as 3 m thick about 700 m north of the Sulphide Queen orebody. It occurs in a swarm of similarly oriented dikes about 300 m long. It contains coarse-grained carbonate, barite, bastnäsite, and quartz, and may be a carbonatite pegmatite. I have collected masses of pure bastnäsite as much as 5 cm thick from places along this dike swarm. A silicified carbonatite dike in the same area contains strontianite, galena, and the rare mineral cerite. Although some dikes in the Mountain Pass area contain bastnäsite or parisite and qualify as ore with as much as 10% REO, many have relatively low REO contents.
Mineral Assemblages of the Carbonatites
Except for the “white sövite,” a texturally distinct type discussed below, bastnäsite–barite sövite forms the basal part of the Sulphide Queen orebody, and most the carbonatite in the northern part of the pit. It also commonly forms a thin zone along the hanging wall of the carbonatite. In the thick, southern part of the orebody, sövite makes up less than half of the ore thickness. The sövitic ore contains early-formed bastnäsite, along with generally recrystallized (Fig. 14A), but locally single-crystal (Fig. 14B) phenocrysts of barite, in a groundmass of fine- to medium-grained calcite and barite. In some cases, bastnäsite prisms are aligned along the border of barite phenocrysts, suggesting crystallization after the displacement of barite during its growth. Where unaltered, the sövite is a pink to mottled white and reddish brown rock that typically contains about 65% calcite, 25% strontian barite, and 10% bastnäsite. However, relative amounts of these three phases vary considerably (Table 4), and alteration following primary crystallization produced more complex assemblages of minerals.
Bastnäsite in sövitic ore is typically coarse grained (Figs. 14A, 15A); on the basis of data from 126 samples collected on a single level during pit mapping, the average diameter of bastnäsite crystals is about 300 μm. Bastnäsite in the sövite generally forms hexagonal prisms, strongly elongate along the c crystallographic axis. As previously noted, it is bastnäsite-(Ce), as is the bastnäsite in other types of ore, which show little variation in REO distribution (Table 5). Parisite, which occurs locally in the sövite, commonly forms fan-like intergrowths with bastnäsite (Fig. 6B). For the most part, monazite occurs sparingly in the sövite, generally as small primary euhedra and patches of radially disposed secondary needles.
Much of the sövitic ore has been altered. Finegrained, anhedral quartz comprises as much as 60% of the rock locally (Table 4), generally as pervasive flooding or stockwork veining (Fig. 15A). The silicification was mainly at the expense of calcite, although partial replacement of barite and bastnäsite also took place. Silicification is common along the northwesttrending Celebration Fault (Fig. 11), where Olson et al.(1954) mapped a series of silicified lenses. Sövite with weak to advanced alteration to talc (Table 4) occurs in the northern part of the pit, mostly as a weakly sheared gray-green rock with brick-red barite augen and deformed prisms of bastnäsite. Allanite occurs locally in talc-altered sövite. The other alteration minerals, chlorite, phlogopite, and magnesioriebeckite, generally occur in xenoliths, but are locally in the carbonatite itself. Replacement of calcite and strontian barite by strontianite is common, and celestine occurs as bladed replacements of, and overgrowths on, strontian barite and as late veins. Iron hydroxide is locally abundant, particularly in silicified ore. Galena is locally common, generally in altered carbonatite, in places as abundant irregular late veinlets that are associated with late carbonate.
Beforsite (dolomitic carbonatite) was found only in the southwestern corner of the Sulphide Queen orebody at the surface (Olson et al. 1954) and was not significant during early mining, but in the 1980s, it became an important type of ore. A geological map of the 4670 level shows its extent in 1987 (Fig. 11). On lower levels, it was found to stretch further to the north along the hanging wall of the orebody. The beforsite typically overlies sövitic ore, and is separated from it by dolomitic sövite (Figs. 11–⇑13).
The beforsite contains ferroan dolomite as the major carbonate phase. The average mode is about 55% dolomite, 25% barite, 15% bastnäsite, and 5% calcite. It is a light gray to pale brown or pale pinkish brown rock that contains abundant gray, white, or pale red to pink phenocrysts of barite (Figs. 15B), commonly as single crystals rather than recrystallized aggregates. The dolomite, which mostly occurs as brownish gray to pale yellowish brown rhombs, is locally oxidized and dark brown. It crystallized after the formation of the barite phenocrysts. The dolomite rhombs are set in fine-grained, pale yellow to light pink or nearly white interstitial material that consists of bastnäsite, calcite, and barite. Apparently, the bastnäsite crystallized from late residual fluid in the beforsite, after barite and dolomite crystallization, as opposed to the case in sövite, in which the bastnäsite was comparatively early. The beforsite ore is commonly crudely banded (Fig. 15B). Barite-rich zones in the beforsite may represent cumulates along internal intrusive contacts. Along the south wall of the pit, the beforsite locally contains steeply dipping, braided, discontinuous bands of late bastnäsite + calcite. The texture likely formed by upward streaming of relatively late-stage REE- and Ca-rich fluids.
Bastnäsite in the beforsite ore is relatively fine grained (Fig. 14C), with an average diameter of crystals equal to 87 μm (based on data from 118 specimens). It occurs as stubby hexagonal prisms weakly elongate along the c axis. Monazite content of the beforsite ore is variable, but locally, it is as much as 5% and occurs mostly as irregular veinlets of “bone” monazite with a microcrystalline granular to radiating acicular texture. In places, this late monazite surrounds pockets of finegrained bastnäsite, and may have been deposited prior to the bastnäsite (Fig. 14D). Some parisite occurs as intergrowths with bastnäsite. Intergrowths of as many as four minerals, ranging from REE fluorocarbonate with little or no calcium, to calcium-rich rare-earth carbonate, have been found in beforsite using SEM–EDS (Fig. 6C), which may include all phases in the series bastnäsite – parisite – röntgenite – synchysite (Donnay & Donnay 1953) or unnamed phases such as those reported by Meng et al.(2002). Sahamalite (Jaffe et al. 1953), and some synchysite are considered to have originated as secondary phases in the beforsite.
The beforsite generally contains a little quartz as late interstitial grains that, with calcite, postdate the formation of bastnäsite. Locally, beforsite contains quartz as discrete crystals as much as 1.5 mm in diameter, some of which are dipyramidal euhedra, indicating a high-temperature and probable magmatic origin. A few irregular masses of greasy gray quartz to 30 cm across were noted in the beforsite during pit mapping, but quartz veining akin to that found in the sövite (Fig. 15A) was not found. Dark brown limonitic alteration occurs in places in the beforsite, particularly along faults and in brecciated zones; however, the beforsite is generally unaltered, and galena and late carbonate veins are rare. On the basis of the lack of alteration and its occurrence in the core of the orebody, intrusion of the beforsite likely postdated that of the sövite. Much of the sövite shows evidence of late Mg enrichment in the form of dolomitization or talc alteration, possibly as a result of alteration during the emplacement of beforsite. However, clear cross-cutting relationships showing beforsite intruding sövite were not observed.
Bastnäsite–barite dolomitic sövite
Dolomitic sövite ore occurs in a zone 30 to 60 m wide between the sövitic and beforsitic ore-types (Figs. 11, 12, 13). It contains both dolomite and calcite in variable amounts, and the carbonate minerals show evidence of secondary redistribution such as calcite veining (Fig. 15C) and dolomitization. The dolomite is present in various forms, ranging from rounded to sharply crystalline rhombs or fine anhedral mosaics. The dolomitic sövite is commonly limonitic, with “limonite” present as a dark brown pseudomorph after rounded to sharp rhombs of dolomite or as a spongy replacement of all carbonate. Barite phenocrysts in the dolomitic sövite are typically white to pink and recrystallized. Some dolomitic sövite contains coarse bastnäsite as in the sövite, and some has fine-grained, late, beforsitestyle bastnäsite. The bastnäsite in dolomitic sövite is commonly yellowish brown or brown, rather than in the paler colors seen in the other ore-types.
Dolomitic sövite ore has high contents of the strontianite component, particularly in rocks with coarse bastnäsite. It is also locally rich in fine-grained, anhedral quartz. It rarely contains talc. “Black ore”, a term used by Molycorp miners and mill personnel for dark brown, earthy, and commonly porous material with abundant veins of white calcite, is mainly restricted to the dolomitic sövite. It locally contains extremely high contents of bastnäsite owing to calcite and dolomite removal. During the 1980s, such ore was found to cause major problems in the mill, probably owing to the propensity of strontianite to report with bastnäsite in the flotation concentrate. Although REE minerals other than bastnäsite are common in dolomitic sövite, they mostly occur in minor amounts. Microcrystalline “bone” monazite is more abundant in the dolomitic sövite than primary monazite. Synchysite occurs as a partial replacement of bastnäsite, and secondary sahamalite and ancylite also have been identified. These secondary REE minerals generally occur with secondary calcite.
I consider the dolomitic sövite to mostly be a mixed rock-type, not a separate intrusive type. The dolomite is residual primary dolomite in rock that was originally beforsite, but was introduced secondarily in rock that was originally sövite. On the basis of bastnäsite grainsize (the most reliable way to differentiate beforsite and sövite ore in the absence of original carbonate mineralogy), some dolomitic sövite was originally sövite and some beforsite. A line drawn between carbonatite with coarse grains of bastnäsite (typical of the sövite), and fine grains of bastnäsite (characteristic of the beforsite), roughly bisects the dolomitic sövite zone. The dolomitic sövite may thus in part be dolomitized sövite and in part calcitized beforsite. It probably formed by redistribution of carbonate minerals during and after intrusion of the beforsite.
White bastnäsite–barite sövite
Relatively small amounts of white sövite occur in the southwestern part of the pit, where it forms a lensoidal mass as much as 25 m thick above the beforsitic ore (Fig. 13). It contains no dolomite, but it differs from the basal sövite in that it contains fine-grained late bastnäsite, single-crystal white barite phenocrysts, and “bone” monazite like the beforsite. This minor ore-type, which locally has very high contents of bastnäsite, may be the product of late-stage calcitization of beforsite by rising residual fluids responsible for late deposition of bastnäsite and calcite in the underlying beforsite.
Parisite is present in some of the ore as a minor mineral. However, parisite sövite with little or no bastnäsite was reported by R.L. Sherer (unpubl. data, 1979) as a thin, sheet-like intrusion in the south wall of the pit. It contains about 20% of flow-oriented, coarse greenish brown parisite, distinguished petrographically as hexagonal plates with the shortest dimension along the c axis. A similar intrusive body was intersected by drilling in the mid-1980s beneath the orebody (Fig. 13). Parisite also occurs abundantly as coarse greenish plates in carbonatite on the Windy claim about 20 km southeast of the Sulphide Queen orebody. The Windy Hill parisite has a relatively high proportion of Nd (A.N. Mariano, pers. commun., 2007), but SEM–EDX spectra indicate that it is parisite-(Ce). Because parisite contains a lower level of the REE than bastnäsite, it lowers the grade in the concentrate and is not considered an ore mineral; however, parisite occurs widely in the ore, and may form significant amounts of some concentrates.
Carbonatite that contains monazite in amounts that approach or exceed bastnäsite contents occurs in and adjacent to the Sulphide Queen orebody. In addition, monazite-rich sövite with little or no bastnäsite comprises many of the carbonatite dikes in the vicinity of the orebody. Monazite is not taken into solution during REE extraction at Mountain Pass, and the effect of adding rock with significant amounts of monazite to mill feed is therefore deleterious. However, because monazite carbonatite rarely contains more than 5% REO, it was effectively avoided, beginning in the mid- 1980s, by milling ore above that cutoff grade.
Although monazite beforsite occurs at Mountain Pass, nearly all of the monazite-rich carbonatite exposed in the pit is dolomitic sövite constituting a mass about 20 m across in the hanging wall (Fig. 11), and monaziterich sövite at the north and south ends of the orebody (Fig. 13). Olson et al.(1954) noted the presence of such rock as satellite bodies near the north end of the Sulphide Queen carbonatite, and as inclusions in “gray carbonate – pink barite rock.” In hand specimen, the monazite-rich carbonatite is generally fine-grained and equigranular (Fig. 15D) because large phenocrysts of barite are sparse or lacking. The monazite occurs predominantly as primary crystals, although “bone” monazite is present in some samples. Where present, bastnäsite occurs as sparse corroded grains, possibly xenocrysts, which are generally in coarse, early-formed grains similar to bastnäsite in the sövite.
Most of the monazite-rich carbonatite is associated with breccia, and it commonly contains small phlogopite- rich clasts. It occurs around breccia at the north and south ends of the pit (Fig. 13) and in the hanging wall of the orebody. It locally contains abundant xenoliths of the host gneiss.
Breccia with fenitized clasts of country rock and a carbonatite matrix comprises significant amounts of the Sulphide Queen carbonatite mass, and rarely forms ore. It occurs abundantly in the northern part of the pit and under the mill near the south end of the pit. In the hanging wall of the orebody, it ranges from a stockwork of randomly oriented to sheeted dikes in altered gneiss (Fig. 15E), through clast-supported breccia (Fig. 15D), to matrix-supported breccia that grades into monaziterich carbonatite. In places, it seems to have formed in situ by intrusive stoping and injection of carbonatitic magma into shattered country-rock. Breccia that occurs locally in sheet-like masses in the orebody may have been emplaced as intrusive breccia or it may have formed screens between separate intrusive sheets of carbonatitic magma. Clasts of country rock in carbonatite are invariably fenitized, giving some the appearance of pink to red syenite with a rim of dark phlogopite or chlorite (or both) (Fig. 14F). Breccia in the northern part of the pit is strongly altered to talc and chlorite in places, which renders clast identification difficult. In the footwall of the orebody, much of the breccia is composed of rounded clasts of gneiss, shonkinite, and syenite in a matrix of crushed rock with talc, chlorite, and magnesioriebeckite and little or no carbonatite. This breccia was likely formed prior to intrusion of carbonatitic magma.
Composition of the Carbonatites
Carbonatite compositions vary widely at Mountain Pass, as might be suspected given the mineralogical variability (Table 4). On the basis of data from hundreds of 30-m blast-hole analyses, average REO content of the Sulphide Queen carbonatite mass is about 7%; however, not all is ore, and carbonatite with REO below the cut-off grade is present, particularly in marginal parts of the orebody. At Mountain Pass, carbonatite typically has low levels of P2O5; the maximum reported in Table 3 is 1.7%, and most samples contain less than 0.5%. The Ba contents of Mountain Pass carbonatite are extreme, ranging between 1.7% and 33.1% BaO in Table 3. The Sr contents also are very high (0.4–14.2% SrO, Table 4), which is consistent with the presence of abundant celestine and strontianite. In addition, SEM–EDS spectra show that much of the barite in the Mountain Pass carbonatite contains minor to abundant Sr, and ranges from strontian barite to barian celestine. Few Ti data are given in Table 3 concerning TiO2 content, but the data on hand show low levels of TiO2.
Although individual samples may have a highly variable chemical composition, the sövite ore is typically rich in Ca, Ba, and Sr. It has a low level of Mg and generally has low Fe oxide contents except where altered. Carbonatite dikes in the Mountain Pass – Mineral Hill areas are generally sövitic, with REO contents of about 3%, but some reach ore grade. Dike sample HP–20B, taken from a sövite dike about 2 km southwest of the Sulphide Queen orebody, is ore grade (about 7.5% REO), but has a low BaO content, along with high Fe oxide, SiO2, and Al2O3 due to the presence of xenoliths (Table 4).
Beforsite ore typically has a relatively high MgO (Table 4), along with CaO:MgO of about 2:1, and relatively low levels of Sr and Pb. Such ore commonly has high REO content; 10-m blast-hole samples drilled in beforsite in the southwestern corner of the pit in the 1980s consistently contained more than 10% REO and, in some instances, more than 15% REO.
Despite its common brown color, which suggests locally intense conversion to iron oxyhydroxides, samples of dolomitic sövite in Table 3 are not particularly rich in Fe oxide. In comparison with the beforsite, the dolomitic sövite has high Sr (>5% SrO, Table 4).
Few trace-element data are provided here for Mountain Pass carbonatite. In part, this is due to difficulties in obtaining accurate conventional INAA and ICP–AES data for this unusual rock, with its extremely high REE contents. In Table 3, I give some trace-element data for three samples.
Carbonatite at Mountain Pass has been characterized as ferrocarbonatite on the basis of its high Ba and REE content (Le Bas 1987). Ferrocarbonatite, which is typified by high Fe and Mn contents (Woolley & Kempe 1989), occurs as late-stage intrusions in some carbonatite complexes and carries most of the economic wealth of carbonatites, including rare-earth deposits (Le Bas 1987). However, although some of the carbonate is relatively iron-rich, ferrocarbonatite is rare or not found at Mountain Pass. Instead, most of the carbonatite is chemically calciocarbonatite (sövite), magnesiocarbonatite (beforsite), or is intermediate between these two types (dolomitic sövite) (Fig. 16). At Mountain Pass, carbonatite with a ferrocarbonatite or near-ferrocarbonatite composition was altered (silicified) or contaminated by country rock.
On the basis of mapping and drill-hole data, the Sulphide Queen carbonatite is surrounded by an envelope of carbonatite dikes and associated fenitic alteration in the host metamorphic rocks that is more than 200 m thick in places. The fenitized rock generally contains brick-red to pink K-feldspar, dark Mg-rich mica and carbonate. Magnesioriebeckite commonly replaces earlier amphibole and pyroxene, and it occurs widely in small amounts along fractures. Chlorite and hematite are also widespread alteration-induced minerals. Fluorite is present in places, generally in veinlets. The alkaline intrusive rocks are locally affected by similar fenitization, which may generally be distinguished by the presence of red alkali feldspar, blue to green magnesioriebeckite, and hematite.
A zone of radioactive fenitized gneiss as much as 30 m wide in the Mountain Pass area was interpreted as a pre-Laramide shear zone (Olson et al. 1954). This zone extends for about 5 km southeast from the Sulphide Queen carbonatite. It is the locus of several REE prospects, including an occurrence of allanite-rich fenitized gneiss with as much as 9% REO, and occurrences of carbonatite dikes.
Genesis of the Mountain Pass Ultrapotassic Rocks and Carbonatite
On the basis of their mineralogy and chemical composition, the mafic alkaline rocks at Mountain Pass and the more widespread occurrences of similar rocks in the Mountain Pass ultrapotassic belt fall into the Group I (Foley et al. 1987) or lamproitic (Bergman 1987) subclass of ultrapotassic igneous rocks. Haxel (2005) referred to late shonkinite dikes, mainly sampled in the Mineral Hill area, as “lamproite-like”. Because of the chemical and mineralogical similarities between the probable primary mafic members of the Mountain Pass suite and other ultrapotassic rocks, I infer that the origin of the Mountain Pass alkaline rocks is similar to that proposed for ultrapotassic rocks elsewhere.
Recent hypotheses about the origin of ultrapotassic rocks involve partial melting or metasomatism of the mantle (or both) (Sahama 1974, Foley et al. 1987, Bergman 1987). Assimilation of crustal material by mantle-derived magma is not now considered a major factor in the genesis of ultrapotassic rocks (Foley et al. 1987). Ultramafic xenoliths in some ultrapotassic rocks have been cited as direct evidence of mantle derivation (Foley et al. 1987). Other evidence for mantle derivation includes calculated pressures and temperatures of crystallization, and the presence of diamond in some occurrences of ultrapotassic rocks (Bergman 1987).
In order to explain the unusual mix of chemical factors in ultrapotassic rocks that suggest a depleted mantle source (high Mg, Cr and Ni contents; high Mg numbers) along with high contents of elements indicating enrichment (high REE, Ba, K, Th and Zr contents), a metasomatized source has been proposed (Nixon et al. 1984, Jaques et al. 1984). Mantle contamination by subducted crustal rocks has also been advocated (Venturelli et al. 1984). Asthenospheric mantle that should be present under thick continental crust is considered by many to be the most likely source. Mirnejad & Bell (2006) evoked two-stage mantle metasomatism to explain chemical and isotopic evidence in the Leucite Hills lamproites, in Wyoming: subductionrelated metasomatism of previously depleted Archean mantle, and later metasomatism related to hotspot or plume upwelling in the mantle, or to subduction that added K, volatiles, and heat to the source area.
The presence of more felsic rocks and the association of significant amounts of carbonatite at Mountain Pass are unusual when compared with other ultrapotassic series. The relatively high level of silica in felsic members of the Mountain Pass ultrapotassic suite may be due to crystallization of leucite (now pseudoleucite) and extended fractionation in a plutonic environment. Slow crystallization in this environment offers possibilities for crystal–liquid fractionation that could lead to more evolved rocks by removal of leucite and mafic phases forming shonkinite–melasyenite along with liquid with excess silica. Gleason (1988) considered shonkinite in the Barrel Spring area to be a cumulate of crystals formed from an intermediate magma, whereas Crow (1984) proposed that the shonkinite at Mountain Pass represents parental mantle-derived magma that gave rise to offspring syenitic magmas. Haxel (2005) pointed to fine-grained shonkinite dikes at Mountain Pass as a potential parental magma, but the data presented here show that rocks in large mafic alkaline masses may also represent a parental alkaline magma.
Plutonic textures exhibited by relatively large masses of ultrapotassic rock in the Mountain Pass area suggest relatively deep emplacement. Experimental work on Leucite Hills, Wyoming, rocks shows that leucite is unstable above 1 kbar P(H2O) (Barton & Hamilton 1978), and the occurrence of pseudoleucite in melasyenite with a plutonic texture from the Mountain Pass area indicates emplacement at depths near the maximum for the initial stability of leucite (2 to 3 km).
The Mountain Pass ultrapotassic suite shows a general progression in intrusive sequence from shonkinite to granite that could have resulted from intrusion of mafic magma into the crust followed by more-or-less in situ differentiation to more felsic magmas. However, late melasyenite dikes that locally cross-cut felsic rocks indicate renewed intrusion of mafic magma. On the basis of detailed geological mapping and evidence of the intrusive sequence, Watson et al.(1974) proposed that the Mountain Pass intrusions came from more than one differentiating source. Petrogenetic modeling based on REE contents led Crow (1984) to propose a separate origin for granite than for shonkinite and syenite at Mountain Pass. However, relatively high contents of elements such as Cr, Ni, and Zn in the Mountain Pass granites (Table 2) suggest that they are closely related to the ultrapotassic rocks.
The elevated silica contents of its felsic members and its association with carbonatite may be due to the more thoroughly evolved nature of the Mountain Pass suite relative to other ultrapotassic suites. Separation of alkaline silicate and carbonatitic magmas by liquid immiscibility at late stages of crystallization of the alkaline magma, as suggested by Kogarko et al.(1979), could account for this association. Compelling evidence for liquid immiscibility, such as that found in kimberlite (Wyllie 1978), has not been found in Mountain Pass rocks, but the presence of the New Trail Canyon dike rock (Fig. 5E) suggests that hybrid carbonatite–silicate rocks may be present.
I believe that the crystallization of mafic alkaline rocks at Mountain Pass from a magma with high Ca, Ba, F, and LREE contents is a compelling argument for the separation of a Mountain-Pass-type carbonatiterelated fluid from such a magma. Although carbonatite at Mountain Pass shares some chemical characteristics with other carbonatites, it is highly enriched in certain elements (Ba and REE) and low in others (particularly P, Nb, and Fe). Not only is the Mountain Pass carbonatite highly enriched in elements that are also inordinately abundant in the associated ultrapotassic rocks, but the geometry of Mountain Pass carbonatite intrusions is similar to that of the associated ultrapotassic plutons, both occurring as moderately dipping, sill-like intrusions and as steeply dipping dikes. If the origin of the carbonatite is related to processes that also gave rise to ultrapotassic magmatism, a parent magma similar to that which typically produces carbonatites in nephelinitic complexes (Jones & Wylllie 1983) is not necessary.
The isotopic ages of DeWitt et al.(1987) indicate that carbonatite emplacement took place some 25 Ma after shonkinite and syenite emplacement at Mountain Pass. According to E. DeWitt (pers. commun., 2004), the carbonatite age is suitably precise to establish this age difference, despite the relatively small difference relative to the ancient age of these rocks and problems with abundant common lead. It is hard to believe that two such spatially related and chemically similar unique intrusive types as the Mountain Pass carbonatite and ultrapotassic rocks are not also genetically related.
The relationship between the Mountain Pass ultrapotassic rocks and coeval anorogenic granitic rocks in the southwestern U.S. is unclear. The ultrapotassic suite has some chemical affinities with the anorogenic granites, particularly those at Gold Butte, Nevada, and the Hualapai Mountains, Arizona (Fig. 1), which have high levels of K, Ba, F, and REE (Volborth 1962, Kwok 1987, Anderson 1989). However, the anorogenic granites are Al-rich (some are peraluminous), only marginally alkaline, and have REE distributions with large negative Eu anomalies (Anderson 1983), suggesting a genesis very different from that of the ultrapotassic rocks. Most of the data indicate that the anorogenic granites are of crustal origin (Anderson & Bender 1989), although they are associated with anorthosite, likely a mantle-derived product (Emslie 1978). Both the anorogenic granite and ultrapotassic series may have been produced during a single extended Middle Proterozoic thermal episode originating in the mantle. Mantle upwelling under a thick supercontinent has been postulated to account for the Middle Proterozoic anorogenic magmatism (Anderson & Bender 1989, Hoffman 1989), and mantle metasomatism is also thought to occur under thick continental crust (Foley et al. 1987).
The narrow zone of Mountain Pass ultrapotassic rocks may have been focused by local deep-seated crustal rupturing within the wide northeast-trending anorogenic magmatic belt. In this regard, the time of emplacement of the ultrapotassic rocks may be significant; it is approximately equivalent to the age postulated for the opening of Middle Proterozoic basins (e.g., the Belt basin in the northwestern United States: Obradovich et al. 1984). The Mountain Pass belt of ultra potassic rocks is inboard, but near and roughly parallel to, the western boundary of the North American Precambrian continent (Kistler 1990), which is considered to have separated from the Siberian platform (Piper 1983) during a much later event (0.6 Ga). Middle Proterozoic ultrapotassic magmatism in southeastern California may have been associated with intracontinental rifting that was not consummated in continental breakup.
I thank Chevron Mining Inc. for permission to utilize data generated during the time that I was an employee of the Mountain Pass operation (1984–1988). Chevron Mining Inc. underwrote the cost of color pages in this paper. I also acknowledge Molycorp Inc. geologists Richard L. Sherer, Stephen Levine, Geoffrey Nason, and William Buckovic for their important geological contributions, and Anthony N. Mariano for his work on Mountain Pass mineralogy. In addition, I thank W. Brown (Mountain Pass geological assistant, 1986–1987) for help with petrography and pit mapping, Kris Pizarro (Nevada Bureau of Mines and Geology) for assistance with some illustrations, and Gordon Haxel and Ed DeWitt (U.S. Geological Survey) for discussions and information. I am indebted to Anthony N. Mariano and Michael J. Le Bas for helpful reviews of the manuscript, and to David R. Lentz for editorial comments. Finally, I dedicate this paper to John Gittins in recognition of his contributions to the understanding of carbonatites.
- Received February 26, 2007.
- Revised manuscript accepted November 10, 2007.
- © 2008 Mineralogical Association of Canada