| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Journal of Petrology | Volume 43 | Number 3 | Pages 423-448 | 2002
© Oxford University Press 2002
Dunite Formation Processes in Highly Depleted Peridotite: Case Study of the Iwanaidake Peridotite, Hokkaido, Japan
DEPARTMENT OF EARTH AND PLANETARY SCIENCES, TOKYO INSTITUTE OF TECHNOLOGY, 2-12-1 OOKAYAMA, MEGURO-KU, TOKYO 152-8551, JAPAN
Received July 25, 2000; Revised typescript accepted September 5, 2001
 |
ABSTRACT |
|---|
Dunite formation processes in highly depleted peridotites are discussed based upon a detailed study of the Iwanaidake peridotite, Hokkaido, Japan, which consists mainly of harzburgite with a small amount of dunite. In the harzburgites, the Mg# [= 100 x Mg/(Mg + Fe2+)] of olivine ranges from 91·5 to 92·5, and the Cr# [= 100 x Cr/(Cr + Al)] of spinel from 30 to 70; in the dunites, the Mg# of olivine ranges from 92·5 to 94 and the Cr# of spinel from 60 to 85, respectively. The NiO wt % of olivine in harzburgites ranges from 0·38 to 0·44, and in dunites from 0·35 to 0·37. The Mg# and Cr# are higher and NiO wt % is lower in the dunites than in the harzburgites surrounding the dunites. The Mg# and Cr# exhibit normal depletion trends expected from simple partial melting, whereas the NiO wt % shows an abnormal trend. On the basis of mass balance calculations, dunites are considered to be derived from the harzburgites by a process involving incongruent melting of orthopyroxene (orthopyroxene olivine + Si-rich melt). Hydrous conditions were necessary to lower the solidus, and thus melting of harzburgite was probably triggered by the introduction of hydrous silicate melt. The dunite in this massif may have formed in the mantle wedge above a subduction zone.
KEY WORDS: depleted peridotite; hydrous melt; incongruent melting; residual dunite; Iwanaidake peridotite
|
INTRODUCTION |
|---|
A highly depleted peridotite, such as dunite, is one of the terminal products of partial melting in the upper mantle. Therefore, to trace the evolution of the upper mantle, it is useful to investigate this type of lithology. The Mg# [= 100 x Mg/(Mg + Fe2+)] of olivine and Cr# [= 100 x Cr/(Cr + Al)] of spinel have been used previously as measures of the degree of chemical depletion as a result of partial melting and melt extraction in the mantle (e.g. Dick & Bullen, 1984
; Arai, 1994). A high Mg# of olivine and high Cr# of spinel indicate a high degree of melting, as is suggested by peridotite melting experiments (Mysen & Kushiro, 1977; Jaques & Green, 1980). Highly depleted peridotites (with high Mg# of olivine and Cr# of spinel) are found in a number of tectonic settings, such as subduction zones (e.g. serpentinite seamounts at fore-arcs, Ishii et al., 1992; xenoliths in Japan, Kamchatka, and the Luzon–Taiwan arcs, Arai et al., 1998), cratonic regions (e.g. xenoliths in kimberlites; Boyd & Nixon, 1975; Hervig et al., 1980), or in the mantle section of ophiolites (e.g. Jaques & Chappell, 1980). In general, peridotite massifs are more suitable than peridotite xenoliths for investigation of lithological spatial relationships in the upper mantle.
The Iwanaidake peridotite, central Hokkaido, Japan, is composed of harzburgite with intercalated dunite (Niida & Kato, 1978). Peridotite massifs are exposed along the Kamuikotan belt intermittently and these massifs, including the Iwanaidake, are highly depleted in basaltic components (Kato & Nakagawa, 1986). Unlike the other peridotite massifs in the Kamuikotan belt, however, the Iwanaidake peridotite is free from extensive serpentinization. Thus, the Iwanaidake peridotite is highly suitable for a case study of the petrogenesis of highly depleted peridotite massifs. In the Iwanaidake, dunite is more depleted than harzburgite, based on Mg# of olivine and Cr# of spinel.
Recently, dunite formation processes have been extensively debated. Dunite is an important constituent of both ultramafic massifs and xenolith suites. Some dunite bodies show evidence of magma–wall rock reaction in the upper mantle and the formation of dunite would have changed the composition of both the melt and the host peridotite (e.g. Quick, 1981; Kelemen, 1990; Kelemen et al., 1995; Allan & Dick, 1996; Arai & Matsukage, 1996; Dick & Natland, 1996). For these reasons, dunite formation processes are important to understanding magma genesis and upper-mantle evolution. Three origins of dunite have been proposed: (1) residual dunite; (2) cumulative dunite; (3) replacive dunite. Residual dunite is formed after extensive partial melting of peridotite. Cumulative dunite is formed by fractionation of olivine from a mafic melt. Replacive dunite is a product of the reaction between a pyroxene-bearing host rock and an olivine-saturated magma, which dissolves orthopyroxene in the host peridotite and sometimes crystallizes olivine.
The main purpose of this paper is to propose a quantitative model of dunite formation based on a detailed petrologic study of the Iwanaidake peridotite. The relationship between dunite and harzburgite has been investigated by detailed petrological observations and chemical analysis of mineral compositions. In this massif, dunite is more depleted than harzburgite, although harzburgite itself is already depleted. Using this compositional relationship between dunite and harzburgite, it is clear that the dunite is a residue after a partial melting of the harzburgite. On the basis of this interpretation, a model is proposed that involves the injection of a hydrous melt, causing partial melting of the host harzburgite and resulting in formation of residual dunite. To verify this model, mass balance calculations using the scheme of Ozawa (1997) were carried out to estimate the change of composition and modal abundance of the main minerals. In addition, results of peridotite melting experiments under dry conditions (e.g. Mysen & Kushiro, 1977) and under hydrous conditions (Green, 1973) are used to constrain the P–T and H2O conditions. These constraints suggest that dunite formation occurred in a hydrous environment such as the mantle wedge above a subduction zone.
|
GEOLOGICAL SETTING |
|---|
The Iwanaidake peridotite is part of the Sarugawa ultramafic massif, which was emplaced in the southern part of the Kamuikotan belt in the central axial part of Hokkaido, Japan. The Kamuikotan belt extends from north to south and has a length of 320 km and a width of 20 km; it consists of high- and low-pressure metamorphic rocks with a subordinate amount of ultramafic rocks (Ishizuka et al., 1983
).
Some of the ultramafic rocks within the Kamuikotan belt are considered to be sections of ophiolite complexes, as lithologies include pelagic sediments overlying the basal ultramafic rocks (e.g. Horokanai ophiolite, Ishizuka et al., 1983). However, most of the ultramafic rocks are not associated with an ophiolite sequence but rather occur as isolated peridotite massifs. Spatial variations in mineral assemblage and chemical composition are observed within the belt (Kato & Nakagawa, 1986). In the northern part, the Al concentrations in orthopyroxene, clinopyroxene, and spinel are low, and the Mg# of olivine is high, suggesting that the peridotites here are a more refractory residue than those in the southern part (Kato & Nakagawa, 1986). According to Niida & Kato (1978), the Sarugawa ultramafic massif is divided into western and eastern units (Fig. 1). The western unit peridotite is surrounded by the Sarugawa formation, composed mainly of mafic lavas and pyroclastic rocks with slate, limestone, chert, and sandstone. The Sarugawa formation corresponds to the Sorachi group, which is in the uppermost part of the Hidaka supergroup (Niida & Kato, 1978). The eastern unit peridotite is located within the unclassified Hidaka supergroup, which is considered to be older than the Sarugawa formation, and which is composed mainly of slate and pelitic schist with sandstone, chert, and pyroclastic rocks. Both units of the Sarugawa ultramafic massif are in fault contact with the surrounding rocks, and the two units are also separated by faults.
|
The Iwanaidake peridotite occurs within the Sarugawa ultramafic massif and crops out around the top of Mt. Iwanaidake, where the rocks are free from severe serpentinization. It comprises harzburgite with a small amount of dunite, minor websterite, orthopyroxenite and chromitite. In this study, the study area is treated as part of the Iwanaidake peridotite massif and is divided into sub-areas A, B, C, and D based on the modal abundance of clinopyroxene in harzburgite: A, 1–3 vol. %; B, <1 vol. %; C, gradually changing from 
1 vol. % in the southern part to almost 0 vol. % in the northern part; D, almost 0 vol. % (Fig. 2).
|
|
PETROGRAPHY |
|---|
In this study, the boundary between dunite and harzburgite is defined by 2 vol. % of orthopyroxene. This is different from the IUGS classification (‘dunite’ is defined as peridotite with 90–100 vol. % of olivine), but is more useful for describing detailed petrologic features in the Iwanaidake peridotite. As will be seen in the following section, dunite and harzburgite defined by this criterion can be clearly distinguished by chemical composition.
Harzburgite
Harzburgite is the most abundant rock type in the Iwanaidake peridotite and consists of olivine, orthopyroxene and a lesser amount of clinopyroxene and spinel. In most harzburgites, the modal abundance of olivine is 70–90 vol. %, orthopyroxene 10–30 vol. % and spinel 1 vol. %, and clinopyroxene is rare. The modal abundance of these minerals varies even within a single hand specimen (Fig. 3a). Serpentine, brucite, talc, opaque minerals and a small amount of tremolite are also observed.
|
Olivine is usually anhedral, and ranges in grain size from very fine (<10 µm in diameter) to very coarse (larger than a few centimeters). Orthopyroxene is usually anhedral and sometimes porphyroclastic, mostly with a grain size of 0·5–5 mm and sometimes >10 mm. Orthopyroxene crystals are usually deformed, often contain exsolution lamellae of clinopyroxene and rarely contain spinel lamellae. Although orthopyroxene is usually more resistant to serpentinization than olivine, some crystals are partly replaced by talc. Replacement of orthopyroxene by talc is preferentially observed in areas C and D, and this alteration is more intensive and selective than serpentinization of olivine. Clinopyroxene is absent or small in amount if present, usually occurring at the rim of orthopyroxene and rarely as anhedral isolated grains. Clinopyroxene exsolution lamellae in orthopyroxene are common. The grain size of clinopyroxene crystals at the rim of orthopyroxene ranges from <10 µm to 0·3 mm, and the isolated grains are 0·5 mm in size. The occurrence and abundance of clinopyroxene varies spatially. Spinel is generally anhedral with the exception of those euhedral crystals that are surrounded completely by olivine. The grain size of spinel ranges from 0·1 to 1 mm, and sometimes is larger than a few millimeters. Spinel sometimes forms in a vermicular intergrowth with orthopyroxene, and rarely occurs as lamellae in orthopyroxene. It is resistant to alteration, and usually is not altered to opaque minerals, even close to serpentine veins.
Sometimes a weak foliation and lineation, formed by orthopyroxene alignment, exists and rarely strong foliations and lineations are developed. However, the direction of these structures varies even within a single outcrop, and therefore, the regional structure of the Iwanaidake peridotite is difficult to discern. Sometimes dunite exists in parallel bands to the foliation or lineation of orthopyroxene (see Fig. 3a). Harzburgite shows deformation textures, such as porphyroclastic texture (sometimes mylonitic), kink banding in olivine, and distorted grains of orthopyroxene and clinopyroxene. Porphyroclasts are olivine and orthopyroxene, whereas neoblasts are olivine alone, indicating that olivine deforms more easily than the other minerals. Porphyroclasts are sometimes as large as a few centimeters in size, whereas mylonitic olivine grains are less than a few micrometers. In harzburgite, an equigranular texture of olivine is developed in orthopyroxene-poor domains (1 cm2 or larger), where grain size is a few tens of micrometers to 100 µm in diameter.
Dunite
The grain size of olivine is usually larger than that in harzburgite and reaches a few centimeters in diameter. The modal abundance of spinel is 1–2 vol. %. The spinel grains in dunite are usually euhedral, and are generally 0·1–1 mm in diameter although sometimes 2–3 mm. Dunite occasionally contains clinopyroxene, with or without a small amount of orthopyroxene, and the grain sizes of both pyroxenes are a few millimeters. Rare arrays of coarse-grained clinopyroxene are found in a dunite of 2 m thickness in area C. Spinel lineation exists commonly in dunite but not in harzburgite. Porphyroclastic texture in dunite consists of porphyroclasts and neoblasts of olivine.
The boundary between dunite and harzburgite is important in considering the formation of dunite. Both gradual boundaries (orthopyroxene fades out and the width of the transition zone ranges from a few to a few tens of centimeters) and sharp boundaries (orthopyroxene disappears immediately) exist in this massif. The shape of the dunite bodies is also important. Two types are recognized: lenticular-shaped dunites (Fig. 3b) and irregular-shaped dunites (Fig. 3c). The former is more common in this massif, corresponding to the banded texture of dunite and harzburgite described by Niida & Kato (1978). The thickness of lenticular-shaped dunites ranges from a few centimeters to a few meters. Lenticular-shaped dunite sometimes occurs in concordance with the orthopyroxene foliation or lineation of the surrounding harzburgite. The irregular-shaped dunite is uncommon in the Iwanaidake massif and occurs only in areas C and D. The structural relationship between irregular-shaped dunite bodies and the foliation or lineation of surrounding harzburgite is not clear.
|
ROCK SAMPLES AND ANALYTICAL METHODS |
|---|
Minerals in 38 samples of harzburgite and dunite were analyzed following microscopic observation. These samples were collected from sites away from lithological boundaries between dunite and harzburgite. Sampling was conducted to cover the entire massif of the Iwanaidake peridotite, and major sampling sites are shown in Fig. 2. To eliminate the effects of subsolidus re-equilibration and alteration, only the compositions of the cores of each mineral are discussed below. These samples show a compositional trend in the Iwanaidake peridotite referred to subsequently as the Iwanaidake General Trend.
To examine the relationship between dunite and harzburgite, three harzburgite–dunite sections were intensively sampled: HD2 (relatively thick lenticular-shaped dunite with broad contact zones), HD3 (thin lenticular-shaped dunite) and HD7 (irregular-shaped dunite). Further description of these three sections is given in the next section.
The modal abundance of minerals was measured by the point counting method. Mineral compositions were analyzed by electron probe micro-analysis (EPMA), using the JEOL JCMA-733Mk II and JXA-8900 systems at the Department of Earth and Planetary Science of the University of Tokyo. Olivine was analyzed at 25 kV, 5·0 x 10-8 A and for 200 s duration with the JCMA-733Mk II and corrected with the ZAF method. Compositions of orthopyroxene, clinopyroxene and spinel were analyzed at 15 kV, 1·2 x 10-8 A and 10 s duration using the JCMA-733Mk II with the correction by the Bence and Albee algorithm and the JXA-8900 with the ZAF oxide method. Average values of chemical compositions within a single thin section (about five points) and their variance shown by their standard deviation (1
) are given in Table 1. Representative compositions of olivine are given in Table 2, and those of orthopyroxene, clinopyroxene and spinel are given in Tables 3, 4 and 5, respectively. The standard deviation (1) of the measured Mg# of olivine is 0·01 and that of NiO wt % in olivine 0·001 based on repeated analysis of an internal standard, the San Carlos olivine (Arizona).
|
|
|
|
|
|
CHEMICAL COMPOSITIONAL VARIATIONS |
|---|
Iwanaidake General Trend
In the Iwanaidake peridotite, the Mg# of olivine ranges from 91·5 to 92·5 in harzburgite, and from 92·5 to 94·0 in dunite (Fig. 4a and b). Data from HD2 are plotted for comparison. The Cr# of spinel in harzburgite ranges from 28·7 to 70·0 and that in dunite ranges from 66·7 to 84·4. The compositional relationship between Mg# of olivine and Cr# of spinel is shown in Fig. 4a, in which a positive correlation is seen for harzburgite and dunite. In Fig. 4a, the olivine–spinel mantle array (OSMA) by Arai (1994)
is also shown, which represents a residual mantle peridotite trend in the spinel lherzolite field. Olivine and spinel in the Iwanaidake peridotite plot within the OSMA. The Mg# of olivine is basically constant within one thin section both in dunite and harzburgite. The NiO content of olivine in harzburgite is 0·38–0·43 wt % and that in dunite is 0·35–0·38 wt %. The Mg# and NiO wt % of olivine in dunite and harzburgite show a negative correlation (NiO decreases as Mg# increases), which is opposite to the established general compositional trends in mantle peridotite olivines [i.e. mantle olivine array (MOA) by Takahashi, 1986a; Fig. 4b]. The compositional relationships of the Iwanaidake peridotite in Fig. 4a and b are called the Iwanaidake General Trend. The data for HD2 (representing the local trend of HD2) harzburgite plot in a gap within the Iwanaidake General Trend.
|
The relationships between the modal abundance of orthopyroxene and the chemical composition of olivine and spinel are shown in Fig. 5. The Mg# of olivine and Cr# of spinel in the dunite are systematically higher, and NiO wt % of olivine in dunite is systematically lower, than in normal harzburgite (with modal abundance of orthopyroxene >10 vol. %). Also in this figure, data for HD2 fall in the gap of the Iwanaidake General Trend. Details of HD2 and the relationship between the Iwanaidake General Trend and the local trend of HD2 are discussed later.
|
Local trends (HD2, HD3 and HD7)
Three harzburgite–dunite sections that represent different dunite lithologies are examined: HD2 (Fig. 6), HD3 (Fig. 7) and HD7 (Fig. 8).
|
|
|
HD2 is a harzburgite with a relatively thick lenticular-shaped dunite and broad contact zones between, which was taken from area D (Fig. 6a). HD2 contains a dunite of 77·5 cm thickness and adjacent harzburgite. Twenty-two samples were collected from HD2 over a 150 cm distance. In the left-hand harzburgite, when observed with the naked eye, the abundance of orthopyroxene grains seems to decrease gradually toward the dunite. In thin section, however, the modal proportion of orthopyroxene is difficult to determine because of severe alteration to talc (Fig. 6f). In the right-hand harzburgite, the boundary is relatively sharp. The Mg# of olivine in the left-hand harzburgite increases gradually from the left end (92·2) to the dunite contact (92·8) and is highest in the center of the dunite (93·3, Fig. 6b). NiO wt % of olivine shows the opposite trend: it decreases from the left end of the harzburgite (0·41 wt %) to the dunite contact (0·38 wt %) and is lowest at the center of dunite (0·35 wt %, Fig. 6c). The Cr# of spinel increases gradually from the left end (59·1) to the dunite contact (72·8) and is highest in the central position of the dunite (78·0, Fig. 6d). The Cr# of spinel shows a trend similar to that of the Mg# of olivine. Fe3+/(Cr + Al + Fe3+) is very low (<0·05) in both dunite and harzburgite (Fig. 6e).
HD3 is a harzburgite with very thin lenticular-shaped dunite of 3·2 cm thickness, which was taken from area D (Fig. 7a). In a small drill core sample, 2·4 cm in diameter, the modal abundance of orthopyroxene changes abruptly at the boundaries between dunite and harzburgite (Fig. 7a and f). The grain size of minerals in HD3 dunite ranges from a few micrometers to a few millimeters, which is not very different from that of the surrounding harzburgite. Two core samples have both dunite and harzburgite parts and they were examined separately. Therefore, two samples of dunite and four samples of harzburgite were taken from HD3. The modal abundance of orthopyroxene in the harzburgite ranges from 11·0 to 19·7 (Fig. 7f). Mineral compositions are homogeneous throughout the dunite and harzburgite. In the harzburgite, the Mg# of olivine ranges from 91·8 to 92·0, NiO wt % of olivine ranges from 0·39 to 0·40, and Cr# of spinel ranges from 60·1 to 61·2, whereas these values range from 92·0 to 92·1, from 0·38 to 0·39 and from 61·6 to 63·6, respectively, in the dunite (Fig. 7b, c and d, respectively). The Mg# of olivine and Cr# of spinel are higher and NiO wt % of olivine is lower in the dunite than in the harzburgite, although the differences are very small. Fe3+/(Cr + Al + Fe3+) is very low (<0·05) in both dunite and harzburgite (Fig. 7e).
HD7 is a harzburgite with an irregular-shaped dunite, which was taken from area D (Fig. 8a). HD7 is divided into a dunite–harzburgite alternation zone and a host harzburgite zone. Twenty-seven samples were taken from a section of 130 cm length and three samples (one from the left-most side and two from the right-most side) are classified as host harzburgite. In the dunite–harzburgite alternation zone, dunite and harzburgite alternate three times in a narrow band (80 cm in thickness). Harzburgite in the alternation zone is orthopyroxene poor, the modal abundance being 5 vol. % and at most 13·5 vol. % (Fig. 8f). In the harzburgite of the alternation zone, the Mg# of olivine is nearly constant (92·4–92·5), whereas the NiO wt % of olivine and Cr# of spinel show small variations (Fig. 8b, c and d). The NiO wt % of olivine and Cr# of spinel range from 0·35 to 0·38 and from 68·7 to 78·2, respectively. In the alternation zone, the Mg# of olivine in the dunite does not differ from that in the harzburgite. The thickest central dunite in HD7 has a lower NiO wt % of olivine and Cr# of spinel than the adjacent harzburgite, and in particular, the center of this dunite has the lowest values (NiO wt % of olivine: 0·35; Cr# of spinel: 65·1). Spinel in the other dunites does not have low Cr# of spinel, but shows slightly lower NiO wt %. In the host harzburgite, the Mg# of olivine ranges from 92·2 to 92·5, which is not very different from that in the alternation zone. The Cr# of spinel ranges from 69·7 to 70·3, which is slightly lower than that in the harzburgite in the alternation zone. Likewise, the NiO wt % of olivine ranges from 0·37 to 0·39, which is slightly higher than in the harzburgite in the alternation zone. Fe3+/(Cr + Al + Fe3+) is very low (<0·05) in both dunite and harzburgite; however, the ratio seems relatively higher in dunite than harzburgite (Fig. 8e).
|
SIZE EFFECTS ON THE CHEMICAL COMPOSITION OF DUNITE BODIES |
|---|
To discuss primary magmatic stages recorded in peridotites, it is necessary to eliminate the effect of subsequent subsolidus processes. In this section, deformation and diffusion in the subsolidus stage is discussed.
Formation of lenticular-shaped dunite bodies
As discussed above, a variety of sizes and lithologies of dunite bodies are present in the Iwanaidake peridotite. Lenticular-shaped dunite bodies may have formed from dunite bodies with more complex original shapes. This interpretation is supported by the concordant structure of the lenticular-shaped dunite bodies with the foliation or lineation of the ambient harzburgite. In a peridotite undergoing intense deformation, dunite bodies of any shape could be changed into a lenticular shape by flattening and stretching (Nicolas, 1984). Therefore, the thin lenticular-shaped dunite (e.g. HD3, Fig. 7) may have been generated by deformation of a thicker body (e.g. HD2, Fig. 6), and the irregular-shaped dunite (e.g. HD7, Fig. 8) may preserve the original undeformed structure of the dunite.
Diffusion effects
To evaluate the effects of diffusion at the dunite–harzburgite contact, interdiffusion of Mg–Fe in olivine in the vicinity of a planar contact between dunite and harzburgite is modeled. A one-dimensional diffusion model is considered with the concentration profile being initially step-like: Mg# of olivine is 91 in the harzburgite and 93 in the dunite plate (sandwiched between harzburgite). According to Crank (1956), the Mg# of olivine (CMg2SiO4) can be expressed as a function of place (x) and time (t):
 |
The change in the Mg# of olivine by Mg–Fe interdiffusion is examined for two dunite bodies of different thicknesses: 2 cm and 200 cm, respectively. A comparison between the two shows that the time taken for the dunite of 200 cm thickness to homogenize with the surrounding harzburgite is 10 000 times longer than for the dunite of 2 cm thickness. As an example, a temperature of 1100°C is considered, which corresponds to the lowest temperature at which dunite can be formed at 1 GPa (estimated from the experimental melting of pyrolite under water-saturated conditions; Green, 1973). Ranges of Mg–Fe interdiffusion coefficients have been reported for olivine (Buening & Buseck, 1973; Misener, 1974; Chakraborty, 1997). According to Chakraborty (1997), the diffusion coefficient at 1100°C is 10-17 m2/s. In this case the Mg# of olivine in the center of the 2 cm dunite decreases and homogenizes with the surrounding harzburgite after 10 My. In contrast, that in the 200 cm dunite remains unchanged even after 1 Gy. The diffusion coefficients of Buening & Buseck (1973) and Misener (1974) at 1100°C are both 10-15 m2/s. In this case, the Mg# of olivine in the center of the 2 cm dunite is homogenized after 0·1 My and that in the 200 cm dunite after 1 Gy. It is difficult to estimate the duration of the annealing time because of very large uncertainty of diffusion coefficients and temperatures involved in the Iwanaidake. It is suggested that it takes a geologically significant time for a thick dunite body to be homogenized with the surrounding harzburgite. A thick dunite body may have retained its original Mg# in the center whereas the Mg# in the center of a thinner dunite can easily change, even by solid-state diffusion, during the geological history of the Iwanaidake peridotite.
Several workers have reported Ni diffusion coefficients in olivine (Morioka, 1981; Nagasawa & Morioka, 1996; Ito et al., 1999). These values are slightly smaller than the Mg–Fe interdiffusion coefficients in olivine. Therefore, NiO wt % of olivine in dunite would change simultaneously with Mg# of olivine, or more slowly. It may be difficult to change the Cr# of spinel in dunite by Al–Cr interdiffusion because the spinel fraction is very small and spinel grains are isolated from each other.
|
PROCESS OF DUNITE FORMATION |
|---|
Similarity of HD2 with the Iwanaidake General Trend
The local trend of HD2 (with a relatively thick dunite) shows a wide range of variation in mineral compositions, suggesting that the core of this dunite may preserve the original mineral compositions. Within the massif as a whole, a correlation is evident between the thickness of the dunite layers and chemical composition [Figs 6 and 7, and Tamura et al. (1999)
]. Figure 4 shows the relationship between the Mg# of olivine, Cr# of spinel and NiO wt % of olivine in the harzburgite and dunite, compared with HD2. Figure 5 shows the relationship between the modal abundance of orthopyroxene and mineral compositions for the same samples. The HD2 harzburgite appears to bridge the gap between dunite and harzburgite in the Iwanaidake General Trend. No apparent relationship is evident between the modal abundance of orthopyroxene and chemical compositions in the Iwanaidake General Trend; however, in the local trend of HD2, a negative correlation is seen with Mg# of olivine and Cr# of spinel, and positive correlation is seen with NiO wt % of olivine. However, considering that the modal abundance of orthopyroxene in HD2 covers the gap between dunite and harzburgite in the General Trend, HD2 seems to represent the intermediate way to make the General Trend. Therefore, it can be considered that dunite in the General Trend was produced by the same formation process as HD2-type dunite.
Incongruent melting of orthopyroxene
First, Mg# of olivine and Cr# of spinel are examined using experimental data on peridotite melting (e.g. Mysen & Kushiro, 1977; Jaques & Green, 1980). According to the experiments, as the degree of melting is increased, the rock type changes from lherzolite to harzburgite and finally to dunite, and the Mg# of olivine and Cr# of spinel in the residue increase gradually (Mysen & Kushiro, 1977; Jaques & Green, 1980). Because dunite has a higher Mg# of olivine and Cr# of spinel than harzburgite in the local trend of HD2 (and also in the Iwanaidake General Trend), it is suggested that dunite may have been derived from harzburgite by melting. However, the lower NiO wt % of olivine in dunite than in harzburgite conflicts with this simple melting model. Melting of harzburgite should increase the NiO wt % of olivine in residues, because the partition coefficient between olivine and melt is usually larger than unity (Kinzler et al., 1990; Beattie et al., 1991). Although the replacive dunite model (e.g. Kelemen, 1990), in which reaction between pyroxene-bearing host rock and olivine-saturated magma produces dunite, can explain the Mg#–NiO wt % negative correlation, this model seems not to explain adequately the Mg#–Cr# positive correlation simultaneously. To solve this problem, an incongruent melting model of orthopyroxene is proposed here; that is, the reaction orthopyroxene 
olivine + SiO2-rich melt. The phase diagram in the simple system MgSiO3–H2O (Kushiro et al., 1968) shows that orthopyroxene melts incongruently to liquid and olivine at pressures less than 0·5 GPa under dry conditions and that incongruent melting occurs at 3 GPa under water-saturated conditions. If the mass of olivine increases as a result of the above reaction, the concentration of NiO in olivine should decrease because of the dilution effect.
Calculation of melting mode
To verify this hypothesis of orthopyroxene incongruent melting, model calculations of mineral compositional change using a variety of melting modes were carried out using the calculation scheme developed by Ozawa (1997). To consider the melting processes of HD2, the most fertile harzburgite (the left-most side with the lowest Cr# of spinel) in HD2 was used as the starting composition and cation ratio for this calculation (Table 6). Because clinopyroxene is rare in HD2, only olivine, orthopyroxene and spinel are used. The relationships between minerals and melt are described by one net-transfer reaction. Six elements (Mg, Fe, Ca, Al, Cr and Ni) are used for this calculation and the cation ratio of Si is given by subtracting the others from unity. The exchange distribution coefficients for Mg–Fe (MF), Al–Cr (AC) and Mg–Ca (MC) between melt and mineral (
) are expressed as K(MF,AC,MC) and given as
 |
 |
 |
|
The values of exchange partition coefficients are modified to satisfy the chemical compositional relationships in the starting sample of HD2. For Ni, a simple Nernst-type partition coefficient was used:
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
Here, three melting models have been computed:
- 0·497 orthopyroxene + 0·5 olivine + 0·003 spinel 1 melt;
- 0·997 orthopyroxene + 0·003 spinel 1 melt;
- 1·497 orthopyroxene + 0·003 spinel 1 melt + 0·5 olivine.
Case (1) corresponds to eutectic melting of orthopyroxene and olivine in nearly equal amount. Case (3) corresponds to the case of extreme incongruent melting; one-third of orthopyroxene reacts to form olivine, yielding a very silica-rich melt. Case (2) corresponds to the intermediate case where orthopyroxene melts almost congruently. By using these three chemical reactions, both batch melting and fractional melting (in this calculation, 1% incremental melting is used as a proxy for fractional melting) were examined. The calculations were continued until orthopyroxene disappeared (i.e. until dunite was formed).
The results of the calculations are shown in Fig. 9. Case (3) with 1·497 orthopyroxene + 0·003 spinel 1 melt + 0·5 olivine by batch melting (Fig. 9a and b) reproduces the trend where the Cr# of spinel increases and the NiO wt % of olivine decreases, as the Mg# of olivine increases. This trend is similar to the HD2 local trend (and hence, the Iwanaidake General Trend). In this chemical reaction, orthopyroxene melts incongruently to olivine and melt, and the Mg# of olivine and Cr# of spinel increase and NiO wt % of olivine decreases, as the degree of melting increases. Although the parameters used in the calculation have some uncertainty and we have to consider the possibilities of re-equilibrium between the new melt and the surrounding harzburgite, the agreement of the calculated trends with those observed in HD2 (and the Iwanaidake peridotite) supports the hypothesis that dunite formed by batch partial melting of harzburgite with the incongruent melting of orthopyroxene.
|
Effect of water in dunite formation
The supply of hydrous melt into the host harzburgite is the most likely triggering mechanism for formation of the dunite bodies in the Iwanaidake peridotite. This idea is based on the following reasons. Because harzburgite in the Iwanaidake has been itself depleted, judging from its high Mg# of olivine and Cr# of spinel, it would be difficult to form even more depleted dunite by simple partial melting of this refractory harzburgite under anhydrous conditions. However, it has been demonstrated that the solidus of peridotite is decreased, the temperature of the orthopyroxene-out reaction is decreased, and the degree of melting is increased by the addition of the water. Green (1973) showed that in a pyrolite bulk composition at 1 GPa, orthopyroxene melts out at 1400°C under dry conditions and 1100°C under water-saturated conditions. In addition, the field of orthopyroxene incongruent melting expands up to 3 GPa (Kushiro et al., 1968) under hydrous conditions, whereas it occurs only at lower pressures <0·5 GPa under dry conditions. Therefore, orthopyroxene-incongruent melting and dunite formation in the Iwanaidake peridotite may have been stimulated by the presence of water. Dunite bodies in the Iwanaidake peridotite are scattered in the host harzburgite, suggesting that the host harzburgite melted only locally. This pattern of occurrence can be explained by injection of hydrous melt into the host harzburgite.
The existence of water in the Iwanaidake peridotite has been demonstrated by studies of fluid inclusions in olivine (Arai & Hirai, 1985; Hirai & Arai, 1987). Crystal phases in the fluid inclusions are brucite and serpentine, a combination implying that the fluid comprised mainly H2O. Of the range of possible environments in the upper mantle in which water might be present, the mantle wedge above a subduction zone seems the most likely tectonic setting. The existence of water in the mantle wedge above subduction zones is substantiated by studies of mantle peridotite xenoliths entrained in arc magmas. Hydrous minerals, including amphibole, were reported in peridotite xenoliths from the Ichinomegata crater, NE Japan (Takahashi, 1986b). They were derived from the mantle wedge beneath the NE Japan arc.
Both relatively high temperatures (1100°C) and water-excess conditions are necessary to melt harzburgite to produce dunite as described above. Many models for the temperature distribution and mantle lithologies in subduction zones have been suggested (Tatsumi, 1995; Green & Falloon, 1998; Iwamori, 1998), and it is estimated that the main dehydration of the slab finishes within the fore-arc region and the released water is trapped by the mantle peridotite above the slab. This suggests that the supply of large quantities of water into mantle peridotite occurs mainly at relatively shallow depth near the trench. These models also suggest that the mantle wedge is warmed by upwelling of asthenospheric mantle material and that the high-temperature field >1100°C extends towards the trench side of the volcanic front. Therefore, a part of the fore-arc may satisfy both the high-temperature and water-excess conditions. The Iwanaidake peridotite may have undergone injection of hydrous melts in the fore-arc, as a result of which small patches of dunite were formed. Tamura et al. (1999) also proposed a similar hydrous melting process for the Takadomari, Nukabira and Iwanaidake massifs in the Kamuikotan belt.
The model of dunite formation in the Iwanaidake peridotite
Figure 10a illustrates the proposed model of dunite formation in the Iwanaidake peridotite. Let us suppose that the hydrous melt was in equilibrium with olivine of Mg# 91, NiO wt % 0·40 wt % and spinel of Cr# 60 and was undersaturated with respect to SiO2. This composition corresponds to the melt in equilibrium with the surrounding harzburgite (with olivine of Mg# 91, NiO wt % 0·40 wt % and spinel of Cr# 60) at greater depth. When the hydrous melt upwells into the shallower harzburgite, the melt would no longer be in equilibrium with the surrounding harzburgite, and the solidus of the harzburgite would be lowered by the addition of the H2O-rich melt. Both the effects of the silica-poor melt and the presence of H2O would cause the orthopyroxene in the harzburgite to melt incongruently and to produce olivine. The incongruent melting causes the melt to become more silica rich. In this case, the NiO wt % of olivine would decrease as the mass of olivine increased, and as the melt fraction increased by incongruent melting, Mg# of olivine and Cr# of spinel would increase.
|
The relationships between the shapes of the dunite bodies and the mineral compositional trends may represent the history of melt supply and deformation in the Iwanaidake peridotite. Figure 10b illustrates the deformation history. First, HD2-type dunites were formed (as illustrated in Fig. 10a), which corresponds to the formation of the Iwanaidake General Trend. Second, deformation occurred in the Iwanaidake peridotite and the observed dunites were stretched or flattened, forming the lenticular shapes. Thinner dunites (e.g. HD3) may have been formed from thicker dunites by this deformation process. Finally, after deformation, irregular-shaped dunites such as HD7 were formed. Because HD7 has a different trend of mineral compositions from the other lenticular-shaped dunites, as shown in Fig. 8, another dunite formation process, which was different from that of the Iwanaidake General Trend (e.g. precipitation of olivine from Si-poor melt), may have occurred during the final stage of the dunite formation processes in the Iwanaidake.
|
COMPARISON WITH PERIDOTITES IN KNOWN TECTONIC SETTINGS |
|---|
 TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTINGPETROGRAPHYROCK SAMPLES AND ANALYTICAL...CHEMICAL COMPOSITIONAL...SIZE EFFECTS ON THE...PROCESS OF DUNITE FORMATION COMPARISON WITH PERIDOTITES IN... CONCLUSIONREFERENCES |
|---|
Data compilations for peridotites from known tectonic settings have been reported previously (Dick & Bullen, 1984
; Arai, 1994). Arai (1994) demonstrated the relationship between Mg# of olivine and Cr# of spinel for peridotite xenoliths from known tectonic settings (ocean floor, fore-arc, oceanic hotspot, Japan arcs, continent or African craton). For abyssal peridotites, the Mg# of olivine ranges from 89·5 to 91·5 and Cr# of spinel ranges from 8 to 60 (Dick & Bullen 1984; Arai, 1994). These values are lower than most of the Iwanaidake peridotites. The most fertile harzburgites in Iwanaidake correspond to the most depleted abyssal peridotite. Therefore, the host peridotite (= fertile harzburgite in this massif) may have been an abyssal peridotite, but the more depleted harzburgite and dunite may have been formed in another tectonic setting. In view of the upper limit of Cr# of spinel (up to 80; Arai, 1994), the depleted harzburgites are rather similar to fore-arc peridotites. Recently, new data for the peridotites from seamounts in the Izu–Bonin arc have been reported (Ishii et al., 1992; Parkinson & Pearce, 1998). The Cr# of spinel ranges from 38 to 83, the Mg# of olivine ranges from 91 to 94 (Ishii et al., 1992) and the Mg# of olivine and Cr# of spinel are positively correlated (Parkinson & Pearce, 1998), resembling those in the Iwanaidake peridotite. Therefore, at the stage of dunite formation, the Iwanaidake peridotite may have been located in the mantle wedge in the fore-arc region. It has been proposed recently that the entire Kamuikotan belt originated in a subduction zone tectonic setting (Tamura et al., 1999). |
CONCLUSION |
|---|
The Iwanaidake peridotite consists mainly of harzburgite with a small amount of dunite. Although the harzburgite is itself depleted in basaltic components, the dunite is even more depleted than harzburgite. When compared with the harzburgites, the dunites usually have higher Mg# of olivine, higher Cr# of spinel and lower NiO wt % of olivine. This highly depleted dunite may have been formed by intensive melting of harzburgite with incongruent melting of orthopyroxene being triggered by injection of hydrous melts into the host harzburgite. The melting mode is estimated to be approximately 1·5 orthopyroxene
0·5 olivine + 1 melt. A supply of water would have been necessary to lower the solidus of the harzburgite and the incongruent melting temperature of orthopyroxene. The mantle wedge above a subduction zone is the most likely tectonic setting where hydrous melt can be formed and upwell. The Iwanaidake peridotite may have been emplaced in the mantle wedge at a subduction zone when the dunites were formed.
|
ACKNOWLEDGEMENTS |
|---|
Thanks are due to Professor Hiroko Nagahara and Assistant Professor Hikaru Iwamori for their guidance and encouragement, and to Professors Richard J. Arculus and David H. Green for critical readings of the manuscript. I am also grateful to Professors Eiichi Takahashi and Kazuhito Ozawa for valuable suggestions. I thank Dr Kyoko Matsukage and Dr Natsue Abe for discussion. Thanks are also due to Mr Hideto Yoshida for his assistance with EPMA analysis.
|
FOOTNOTES |
|---|
*Corresponding author. Telephone: +81-3-5734-2338. Fax: +81-3-5734-3538. E-mail: kkubo{at}geo.titech.ac.jp
|
REFERENCES |
|---|
Allan, J. F. & Dick, H. J. B. (1996). Cr-rich spinel as a tracer for melt migration and melt-wall interaction in the mantle: Hess Deep, Leg 147. In: Mével, C., Gillis, K. M. et al.(eds), Proceedings of the Ocean Drilling Program, Scientific Results 147. College Station, TX: Ocean Drilling Program, pp. 157–172.
Arai, S. (1994). Characterization of spinel peridotites by olivine–spinel compositional relationships: review and interpretation. Chemical Geology 113, 191–204.[Web of Science]
Arai, S. & Hirai, H. (1985). Relics of H2O fluid inclusions in mantle-derived olivine. Nature 318, 276–277.[Web of Science]
Arai, S. & Matsukage, K. (1996). Petrology of the gabbro–troctolite–peridotite complex from Hess Deep, Equatorial Pacific: implications for mantle–melt interaction within the oceanic lithosphere. In: Mével, C., Gillis, K. M. et al.(eds), Proceedings of the Ocean Drilling Program, Scientific Results 147. College Station, TX: Ocean Drilling Program, pp. 135–155.
Arai, S., Abe, N. & Hirai, H. (1998). Petrological characteristics of the sub-arc mantle: an overview on petrology of peridotite xenoliths from the Japan arcs. Trends in Mineralogy 2, 39–55.
Beattie, P., Ford, C. & Russell, D. (1991). Partition coefficients for olivine–melt and orthopyroxene–melt systems. Contributions to Mineralogy and Petrology 109, 212–224.
Boyd, F. R. & Nixon, P. H. (1975). Origins of the ultramafic nodules from some kimberlites of northern Lesotho and the Monastery Mine, South Africa. Physics and Chemistry of the Earth 9, 431–454.
Buening, D. K. & Buseck, P. R. (1973). Fe–Mg lattice diffusion in olivine. Journal of Geophysical Research 78, 6852–6862.
Chakraborty, S. (1997). Rates and mechanisms of Fe–Mg interdiffusion in olivine at 980–1300°C. Journal of Geophysical Research 102, 12317–12331.
Crank, J. (1956). The Mathematics of Diffusion. Oxford: Oxford University Press, 414 pp.
Dick, H. J. B. & Bullen, T. (1984). Chromian spinel as petrogenetic indicator in abyssal and alpine-type peridotites and spatially associated lavas. Contributions to Mineralogy and Petrology 86, 54–76.[Web of Science]
Dick, H. J. B. & Natland, J. H. (1996). Late-stage melt evolution and transport in the shallow mantle beneath the East Pacific Rise. In: Mével, C., Gillis, K. M. et al.(eds), Proceedings of the Ocean Drilling Program, Scientific Results 147. College Station, TX: Ocean Drilling Program, pp. 103–134.
Green, D. H. (1973). Experimental melting studies on a model upper mantle composition at high pressure under water-saturated and under water-undersaturated conditions. Earth and Planetary Science Letters 19, 37–53.[Web of Science]
Green, D. H. & Falloon, T. J. (1998). Pyrolite: a Ringwood concept and its current expression. In: Jackson, J. (ed.) The Earth’s Mantle. Cambridge: Cambridge University Press, pp. 311–380.
Hervig, R. L., Smith, J. V. Steele, I. M. & Dawson, J. B. (1980). Fertile and barren Al–Cr-spinel harzburgites from the upper mantle: ion and electron probe analyses of trace elements in olivine and orthopyroxene: relation to lherzolites. Earth and Planetary Science Letters 50, 41–58.
Hirai, H. & Arai, S. (1987) H2O–CO2 fluids supplied in alpine-type mantle peridotites: electron petrology of relic fluid inclusions in olivines. Earth and Planetary Science Letters 85, 311–318.[Web of Science]
Ishii, T., Robinson, P. T., Maekawa, H. & Fiske, R. (1992). Petrological studies of peridotites from diapiric serpentinite seamounts in the Izu–Ogasawara–Mariana forearc, Leg 125. In: Fryer, P., Pearce, J. A., Stokking, L. B. et al.(eds), Proceedings of the Ocean Drilling Program, Scientific Results 125. College Station, TX: Ocean Drilling Program, pp. 445–485.
Ishizuka, H., Imaizumi, M., Gouchi, N. & Banno, S. (1983). The Kamuikotan zone in Hokkaido, Japan: tectonic mixing of high-pressure and low-pressure metamorphic rocks. Journal of Metamorphic Geology 1, 263–275.
Ito, M., Yurimoto, H., Morioka, M. & Nagasawa, H. (1999). Co2+ and Ni2+ diffusion in olivine determined by secondary ion mass spectrometry. Physics and Chemistry of Minerals 26, 425–431.[Web of Science]
Iwamori, H. (1998). Transportation of H2O and melting in subduction zones. Earth and Planetary Science Letters 160, 65–80.
Jaques, A. L. & Chappell, B. W. (1980). Petrology and trace element geochemistry of the Papuan ultramafic belt. Contributions to Mineralogy and Petrology 75, 55–70.
Jaques, A. L. & Green, D. H. (1980). Anhydrous melting of peridotite at 0–15 kb pressure and the genesis of tholeiitic basalts. Contributions to Mineralogy and Petrology 73, 287–310.[Web of Science]
Kato, T. & Nakagawa, M. (1986). Tectogenesis of ultramafic rocks in the Kamuikotan tectonic belt, Hokkaido, Japan. Geology and tectonics of Hokkaido. Association for the Geological Collaboration in Japan, Monograph 31, 119–135.
Kelemen, P. B. (1990). Reaction between ultramafic rock and fractionating basaltic magma I. Phase relations, the origin of calc-alkaline magma series, and the formation of discordant dunite. Journal of Petrology 31, 55–98.
Kelemen, P. B., Shimizu, N. & Salters, V. J. M. (1995). Extraction of mid-ocean-ridge basalt from the upwelling mantle by focused flow of melt in dunite channels. Nature 375, 747–753.
Kinzler, R. J., Grove, T. L. & Recca, S. I. (1990). An experimental study on the effect of temperature and melt composition on the partitioning of nickel between olivine and silicate melt. Geochimica et Cosmochimica Acta 54, 1255–1265.
Kushiro, I., Yoder, H. S. & Nishikawa, M. (1968). Effect of water on the melting of enstatite. Geological Society of America Bulletin 79, 1685–1692.
Misener, D. J. (1974). Cationic diffusion in olivine to 1400°C and 35 kb. In: Hofmann, A. W. et al.(eds), Geochemical Transport and Kinetics: Papers. Carnegie Institution of Washington, Publication 634, pp. 117–129.
Morioka, M. (1981). Cation diffusion in olivine–II. Ni–Mg, Mn–Mg, Mg and Ca. Geochimica et Cosmochimica Acta 45, 1573–1580.
Mysen, B. O. & Kushiro, I. (1977). Compositional variations of coexisting phases with degree of melting of peridotite in the upper mantle. American Mineralogist 62, 843–865.[Abstract]
Nagasawa, H. & Morioka, M. (1996). Synthesis and diffusion measurement of double-layered single crystals of olivine. Chikyukagaku (Geochemistry) 30, 17–25.
Nicolas, A. (1984). Principles of Rock Deformation. Dordrecht: D. Reidel, 208 pp.
Niida, K. & Kato, T. (1978). Ultramafic rocks in Hokkaido. Report of Research Group of Geology 21, 61–81.
Ozawa, K. (1997). Mechanism of magma generation constrained by mantle peridotites: solid-dominant open magma system. Bulletin of the Volcanological Society of Japan 42, special number ‘Magmalogy’, 61–85.
Parkinson, I. J. & Pearce, J. A. (1998). Peridotites from the Izu–Bonin–Marina Forearc (ODP Leg 125): evidence for mantle melting and melt–mantle interaction in a supra-subduction zone setting. Journal of Petrology 39, 1557–1618.
Quick, J. E. (1981). The origin and significance of large, tabular dunite bodies in the Trinity peridotite, northern California. Contributions to Mineralogy and Petrology 78, 413–422.
Takahashi, E. (1986a). Origin of basaltic magmas—implications from peridotite melting experiments and an olivine fractionation model. 30th anniversary issue. Bulletin of the Volcanological Society of Japan 2(30), 17–40.
Takahashi, E. (1986b). Genesis of calc-alkali andesite magma in a hydrous mantle–crust boundary: petrology of lherzolite xenoliths from the Ichinomegata crater, Oga peninsula, northeast Japan, part II. Journal of Volcanology and Geothermal Research 29, 355–395.
Tamura, A., Makita, M. & Arai, S. (1999). Petrogenesis of ultramafic rocks in the Kamuikotan belt, Hokkaido, northern Japan. Memoirs of Journal of Geological Society of Japan 52, 53–68.
Tatsumi, Y. (1995). Subduction Zone Magmatism. Tokyo: University of Tokyo Press, 186 pp.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  D.G. PEARSON and N. WITTIG Formation of Archaean continental lithosphere and its diamonds: the root of the problem Journal of the Geological Society, September 1, 2008; 165(5): 895 - 914. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. Uysal, M. Kaliwoda, O. Karsli, M. Tarkian, M. B. Sadiklar, and C. J. Ottley COMPOSITIONAL VARIATIONS AS A RESULT OF PARTIAL MELTING AND MELT-PERIDOTITE INTERACTION IN AN UPPER MANTLE SECTION FROM THE ORTACA AREA, SOUTHWESTERN TURKEY Can Mineral, December 1, 2007; 45(6): 1471 - 1493. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Rehfeldt, D. E. Jacob, R. W. Carlson, and S. F. Foley Fe-rich Dunite Xenoliths from South African Kimberlites: Cumulates from Karoo Flood Basalts J. Petrology, July 1, 2007; 48(7): 1387 - 1409. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Bernstein, P. B. Kelemen, and K. Hanghoj Consistent olivine Mg# in cratonic mantle reflects Archean mantle melting to the exhaustion of orthopyroxene Geology, May 1, 2007; 35(5): 459 - 462. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Arai, Y. Shimizu, S. A. Ismail, and A. H. Ahmed Low-T formation of high-Cr spinel with apparently primary chemical characteristics within podiform chromitite from Rayat, northeastern Iraq Mineralogical Magazine, October 1, 2006; 70(5): 499 - 508. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Arai, K. Kadoshima, and T. Morishita Widespread arc-related melting in the mantle section of the northern Oman ophiolite as inferred from detrital chromian spinels Journal of the Geological Society, September 1, 2006; 163(5): 869 - 879. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M.-F. ZHOU, P. T. ROBINSON, J. MALPAS, S. J. EDWARDS, and L. QI REE and PGE Geochemical Constraints on the Formation of Dunites in the Luobusa Ophiolite, Southern Tibet J. Petrology, March 1, 2005; 46(3): 615 - 639. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Tamura and S. Arai Unmixed spinel in chromitite from the Iwanai-dake peridotite complex, Hokkaido, Japan: A reaction between peridotite and highly oxidized magma in the mantle wedge American Mineralogist, February 1, 2005; 90(2-3): 473 - 480. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. A. Coish and P. Gardner Suprasubduction-zone peridotite in the northern USA Appalachians: evidence from mineral composition Mineralogical Magazine, August 1, 2004; 68(4): 699 - 708. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. N. Matsukage, K. N. Matsukage, and K. Kubo Chromian spinel during melting experiments of dry peridotite KLB-1) at 1.0-2.5 GPa American Mineralogist, August 1, 2003; 88(8-9): 1271 - 1278. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||