Journal of Petrology Advance Access published online on May 28, 2007
Journal of Petrology, doi:10.1093/petrology/egm021
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
On Silica Activity and Serpentinization
1Department of Geology and Geophysics, University of Wyoming, Laramie, WY 82072, USA
2Virginia Museum of Natural History, 21 Starling Avenue, Martinsville, VA 24112, USA
Received October 3, 2006; Revised typescript accepted March 29, 2007
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONSERPENTINITES AND SILICA...OXYGEN AND SULFUR FUGACITIESSERPENTINITES AND CALCIUMTHE OCCURRENCE OF OTHER...DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Serpentinites have the lowest silica activity of common crustal rocks. At the serpentinization front, where olivine, serpentine, and brucite are present, silica activities (relative to quartz) are of the order of 10–2·5 to 10–5, depending on the temperature. Here we argue that this low silica activity is the critical property that produces the unusual geochemical environments characteristic of serpentinization. The formation of magnetite is driven by the extraction of silica from the Fe3Si2O5(OH)4 component of serpentine, producing extremely reducing conditions as evinced by the rare iron alloys that partially serpentinized peridotites contain. The incongruent dissolution of diopside to form Ca2+, serpentine, and silica becomes increasingly favored at lower T, producing the alkalic fluids characteristic of serpentinites. The interaction of these fluids with adjacent rocks produces rodingites, and we argue that desilication is also part of the rodingite-forming process. The low silica activity also explains the occurrence of low-silica minerals such as hydrogrossular, andradite, jadeite, diaspore, and corundum in serpentinites or rocks adjacent to serpentinites. The tendency for silica activity to decrease with decreasing temperature means that the presence of certain minerals in serpentinites can be used as indicators of the temperature of serpentinization. These include, with decreasing temperature, diopside, andradite and diaspore. Because the assemblage serpentine + brucite marks the lowest silica activity reached in most serpentinites, the presence and distribution of brucite, which commonly is a cryptic phase in serpentinites, is critical to interpreting the processes that lead to the hydration of any given serpentinite.
KEY WORDS: serpentinization; serpentinites; silica activity; oxygen fugacity; rodingites; magnetization of serpentinites
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONSERPENTINITES AND SILICA...OXYGEN AND SULFUR FUGACITIESSERPENTINITES AND CALCIUMTHE OCCURRENCE OF OTHER...DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Serpentinite has been long known to host unusual minerals such as awaruite (FeNi3), heazlewoodite (Ni3S2), and hydrogarnet that are rarely found in any other geological environment. In addition, the fluids evolving from serpentinites are exceptionally reduced (Abrajano et al., 1988
; Charlou et al., 1998; Kelley et al., 2001) and, with pH >10, are some of the most alkaline fluids on Earth (Barnes & O’Neil, 1969; Barnes et al., 1978). Because of the unusual geological conditions, serpentinization has long elicited interest among petrologists (Frost, 1985, and references therein). Recent discoveries of sea-floor hydrothermal vents that could have originated from serpentinization (Beard & Hopkinson, 2000; Kelley et al., 2001; Charlou et al., 2002), and the recognition that the reducing conditions of fluids emitted from serpentinization would be an ideal environment for the origin of life (Russell & Hall, 1997; Sleep et al., 2004; Russell & Arndt, 2005) have provoked an increase in studies on the importance of serpentinization to sea-floor processes (Schroeder et al., 2002; Kelley et al., 2005; Bach et al., 2006).
The authors recently had the good fortune to participate in the Integrated Ocean Drilling Program (IODP) expeditions 304 and 305, which drilled the Atlantis Massif at 20°N on the Mid-Atlantic Ridge. The >1 km of core recovered was mostly gabbro, but the core also contained significant amounts of olivine-rich troctolite. Millimeter-scale serpentine veins locally cut the troctolite and where the veins intersect plagioclase the plagioclase was altered to prehnite and grossular, assemblages common in rodingites. Because plagioclase is the only Ca-bearing phase in the troctolites, we concluded that these micro-rodingites must have formed by desilication of the plagioclase rather than the addition of CaO from outside the system (Frost et al., 2005). This insight prompted us to recognize the important role that silica activity plays in serpentinization.
|
SERPENTINITES AND SILICA ACTIVITY |
|---|
|
TOPABSTRACTINTRODUCTIONSERPENTINITES AND SILICA...OXYGEN AND SULFUR FUGACITIESSERPENTINITES AND CALCIUMTHE OCCURRENCE OF OTHER...DISCUSSIONCONCLUSIONSREFERENCES |
|---|
A note on databases and standard states
Most reactions were calculated using the THERMOCALC database of Holland & Powell (1998
). Because greenalite and saponite are not in the Holland and Powell database, reactions involving these phases were calculated at 1 bar using Geochemist's Workbench© (Bethke, 2005), which incorporates a modified version of the SUPCRT92 database (Johnson et al., 1992). In these calculations we used thermo.com.v8.6+, the updated version of the thermo database. Activities were calculated using a modified Debye–Hückel equation (Bethke, 2005). Databases were not mixed: each diagram was calculated from a single database. In any event, in the serpentinite system the huge free energy changes in the reactions swamp the small uncertainties in the thermodynamic values of the individual phases. Indeed, test calculations done using the two databases yield essentially equivalent results.
We used chrysotile as our serpentine mineral, despite the fact that Evans (2004) recently noted that it is likely to form metastably. We do this because adequate thermodynamic data are not available for lizardite, the stable form of low-T serpentine, and because antigorite, the high-T temperature form of serpentine, is only rarely seen to have formed by direct hydration of a peridotite. In actuality, the choice of which serpentine polymorph to use makes little difference; the free energy change involved in the hydration reactions studied vastly overwhelms the small energy differences between lizardite, chrysolite and antigorite.
Although mantle peridotites generally contain olivine with about 10 mol % fayalite component, we calculated most of our diagrams for the pure Mg-system. We chose to do this because our calculations showed that between the pure Mg-system and the compositions typically found in metaperidotites the location of the silica buffers is relatively insensitive to the small change in 
. For example, calculations using the QUILF program of Andersen et al. (1993) indicate that for the olivine–Opx silica buffer [equilibrium (1)], there is only a difference of 0·02 log units in silica activity between the pure system and the system where 
(the common value for olivine in the mantle).
We recognize two standard states for silica. When we are comparing silica activity of serpentinites with silica activity in other rocks, we use the standard state of pure quartz at T and P of interest. This variable we call aSiO2 and it is a measure of the chemical potential change between the assemblage in question and pure quartz. When discussing the silica activity in systems that involve dissolution reactions we reference silica activity to the standard state of the dissolved form of pure quartz at the T and P of interest [i.e. the activity of H4SiO4, or SiO2(aqueous)]. This variable we call aSiO2 (aq). The difference between the two terms is shown in Fig. 1, which shows the variation of log aSiO2 vs T for pure quartz. We can convert the log aSiO2 for a given reaction to log aSiO2 (aq) by simply adding the value of log aSiO2 (aq)} for quartz at the T and P of interest. To accommodate the huge temperature dependence of the oxygen buffers, we normalize the oxygen fugacity to a common buffer. In most calculations we normalize it to the fayalite–magnetite–quartz (FMQ) buffer and call the variable 
log fO, which is defined as the deviation from the FMQ buffer. In one calculation we normalize it to the iron–magnetite (IM) buffer and in this instance we refer to the oxygen fugacity as IM, IM + 2 or IM + 4; that is, to the IM buffer or to the IM buffer + 2 or 4 log units of oxygen fugacity, respectively.
|
Silica activity in igneous rocks
Silica activity has long been used in the classification of igneous rocks (Carmichael et al., 1974
), even though it has rarely been assigned a numerical value. Thus, it is helpful to compare the silica activity in serpentinites with that of the well-known igneous buffers (Fig. 2). With decreasing silica activity the normative minerals in igneous rocks are hypersthene, olivine, nepheline and leucite, and kalsilite (see Table 1 for compositions and abbreviations of phases in this paper). The reactions that relate these phases are given in Table 2. Because peridotitic rocks do not contain feldspars or feldspathoids, the reactions involving nepheline, leucite, and kalsilite are provided in Fig. 2 merely for reference.
|
|
|
The reaction between olivine and orthopyroxene [reaction (1)], which can be applied to both basalts and peridotites, corresponds to the plane of critical silica saturation (Yoder & Tilley, 1962
). At silica activities above those of reaction (1) a basalt is a hypersthene-normative tholeiite, whereas at lower silica activities it is nepheline-normative and will be classified as an alkali basalt. Because fayalitic olivine can coexist with quartz, equilibrium (1) is dependent on 
as well as T and P. However, as noted above, because the small changes in between the pure system and the composition of minerals in the typical mantle have only a minimum effect on the location of equation (1) (and other silica buffers in the ultramafic system) we use the pure Mg system in all the calculations in this paper. Figure 2 indicates that most igneous rocks have crystallized at silica activities above 10–0·5; only in the rarely occurring kalsilite-bearing rocks are silica activities lower than 10–1. These silica buffers are moderately dependent on pressure; increasing pressure from 1 to 10 kbar decreases silica activity for the olivine–Opx assemblage by 0·2 log units and increases the silica activity for the upper stability of kalsilite by 0·5 log units.
Silica activity in serpentinites and metaperidotites
The silica activity for metaperidotites falls from that defined by the olivine–Opx assemblage to lower values with decreasing temperature and increasing hydration of the assemblage (Table 3, Fig. 3). In assemblages containing anthophyllite and talc, two relatively silica-rich minerals, this trend is not particularly steep, but it becomes dramatic in assemblages where serpentine is in equilibrium with olivine or brucite. In an equilibrium assemblage of olivine, serpentine and brucite, the silica activity is nearly 10–2·5. These values are far lower than those in igneous rocks (even when the igneous silica buffers are extended to low temperatures), making serpentinites among the most silica-deprived rocks on Earth.
|
|
The hydration of olivine is usually described by the model reaction
|
| (13) |
diagram (Fig. 4b). Serpentinization probably begins by conversion of olivine to serpentine by the reaction
|
| (8) |
|
| (15) |
|
| (9) |
|
The silica activity that accompanies low-T serpentinization can be monitored from reaction (15) (Fig. 4c). Reaction (15) is shown as a dashed line in Fig. 4c to emphasize that curve is not valid for PH O = Ptotal, as are the other curves in Fig. 4c, but that it is the trace followed by invariant point (A) with decreasing water activity. At all temperatures, however, the serpentinization reaction [reaction (13)], remains the same. Figure 4c shows that silica activity during low-T serpentinization may be more than two log units lower than it would be at temperatures where brucite + serpentine + pure water occur in equilibrium.
Of course, in most peridotites, orthopyroxene will be involved in the hydration reactions as well as olivine. In the hydration of an Opx-bearing rock the initial reaction will be to form talc by the reaction
|
| (16) |
|
| (7) |
|
| (18) |
|
To calculate the temperature dependence of the silica activity at the invariant point in Fig. 5a we can write a reaction for the assemblage Opx–Srp–Tlc using silica as a mobile component:
|
| (19) |
Figure 5a and b shows that a gradient in silica activity of a log unit or more could be present between serpentinite formed from Opx (point 1 in Fig. 5a and b) and that from surrounding olivine (point 2 in Fig. 5a and b). The extent of this gradient will depend upon the relative rates of the reactions that hydrate olivine and Opx and on the efficiency of fluid flow through the rock. The fact that talc may form rims between Opx and serpentine in bastites (Viti et al., 2005) suggests that in some serpentinites a large silica activity gradient may be present on a millimeter scale.
|
OXYGEN AND SULFUR FUGACITIES |
|---|
|
TOPABSTRACTINTRODUCTIONSERPENTINITES AND SILICA...OXYGEN AND SULFUR FUGACITIESSERPENTINITES AND CALCIUMTHE OCCURRENCE OF OTHER...DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Low oxygen and sulfur fugacities
One of the most distinctive features of serpentinite is its extremely low oxygen fugacity, as indicated by the fact that serpentinites are one of the few crustal environments where iron alloys occur (Frost, 1985
). The occurrence of iron and nickel as metallic alloys, rather than in sulfides indicates that serpentinites also have very low sulfur fugacity. As noted by Frost (1985), sulfur fugacity in metamorphic rocks is a function of oxygen fugacity because sulfur is present dominantly as H2S or SO2, rather than S2. The fugacities of H2S and SO2 are related by the reaction (Table 4)
|
| (20) |
), and, hence, far above the oxygen fugacities ambient in serpentinites. At oxygen fugacities below the sulfate–sulfide fence sulfur fugacity is controlled by the reaction
|
| (21) |
).
|
Because of reaction (21), sulfides are a sensitive monitor of oxygen fugacity in serpentinites. As oxygen fugacity falls, the sulfide minerals stable with magnetite become increasingly poorer in sulfur relative to metal. With decreasing oxygen fugacity in serpentinites, the stable sulfide ranges from millerite (NiS) to pentlandite ((FeNi)9S8) to heazlewoodite (Ni3S2) (Frost, 1985
). The sulfur-rich sulfides polydymite (Ni3S4) and vaesite (NiS2) are found only in relatively oxidizing, carbonate-bearing serpentinites (Eckstrand, 1975). Although pentlandite is common in many mafic and ultramafic rock types, heazlewoodite is found only in serpentinites (Ramdohr, 1968).
Because the low sulfur activity associated with serpentinization destabilizes sulfides, many metals are liberated from serpentinites. They are deposited in relatively more oxidized country rocks, where the increase in oxygen fugacity drives sulfur out of H2S and back into the sulfide phases (Frost, 1985). Metals that have been concentrated adjacent to serpentinites include Ni, Cu, Co and Ag (Groves & Keays, 1979; Leblanc & Lbouabi, 1988).
During serpentinization some of the iron in the original peridotite is driven into magnetite, making serpentinites a possible source of magnetic anomalies (Coleman, 1971; Shive et al., 1988). A plot of magnetic susceptibility, which is a function of the abundance of magnetite, against density, which is a monitor of the degree of serpentinization, is a commonly used diagram from which to infer the nature of the magnetite-forming reactions in serpentinites (Toft et al., 1990; Oufi et al., 2002; Bach et al., 2006; Fig. 6). Three things are evident from this figure. First, the path followed by hydration of peridotites does not follow the path one would predict if the formation of magnetite were directly tied to a serpentinization reaction (dashed line). A reaction that directly ties magnetite formation with serpentinization would be something like:
|
| (22) |
) listed more than 20], progress along these reactions follows a path similar to that shown as the dashed line in Fig. 6. Early formation of magnetite will drive the susceptibility up by several orders of magnitude without a large change in density; only late in the hydration process does density decrease. For this reason, Toft et al. (1990) concluded that hydration of peridotite was decoupled from the formation of magnetite.
|
Another important conclusion that can be derived from Fig. 6 is that each occurrence of peridotite appears to follow its own trend. Some, such as Burro Mountains, Dun Mountain, and Josephine tend to form linear trends that originate at the density and susceptibility of fresh peridotite. Others, such as Red Mountain, Ocean Drilling Program (ODP) 1274 and ODP 895, have either curvilinear trends that originate at the conditions of fresh olivine or linear trends that reach low susceptibility at densities that are considerably lower than those of fresh peridotite. Bach et al. (2006
) postulated that these relations indicate that the formation of magnetite in peridotites commences only after a significant amount of hydration has taken place.
Another consideration in the formation of magnetite is the fact that the FeO content of serpentine from peridotites is variable. Serpentine in incipiently serpentinized peridotites tends to be more iron-rich (with XFe 
0·10, approximately that of the primary olivine), whereas in the highly serpentinized peridotites 
= 0·05–0·03 (Oufi et al., 2002). A similar relation was observed by Bach et al. (2006), who reported that mesh pseudomorphs after olivine contain serpentine and rather Fe-rich brucite without magnetite, whereas the cores of the same meshes contain serpentine, Mg-rich brucite, and magnetite. Two other key points about magnetite and serpentine composition are that magnetite is absent in bastites (serpentine pseudomorphs after Opx) and that the serpentine in the bastites generally has similar XFe to the Opx that they have replaced (Le Gleuher et al., 1990; Viti et al., 2005; Bach et al., 2006).
It is evident from the observations of Oufi et al. (2002) and Bach et al. (2006) that olivine hydrates by at least two reactions. In one it forms relatively iron-rich serpentine and brucite, whereas in the other it forms magnetite along with relatively magnesian serpentine and brucite. Figure 7a shows a possible explanation of how this might work. In the previous discussion we have treated serpentinization reactions as if they dealt solely with the magnesian end-members. This is reasonable because peridotites have XMg 0·9 and the effect of this small amount of FeO on the phase relations is minimal. Of course, we cannot continue to do this when discussing the role of magnetite, as magnetite is formed from the Fe dissolved into the silicates. In Fig. 7a we show the major silica-dependent reactions in serpentinites [reactions (7) and (9); Table 3] as reaction loops. The loops are based on the following observations. Talc is always more magnesian than serpentine; for example, when 
= 0·9, 
= 0·975 (Trommsdorff & Evans, 1972). The relatively Fe-rich serpentine formed from olivine may have XMg = 0·9 (Oufi et al., 2002). The Fe-rich brucite formed in this reaction is poorly constrained but it may be as high as XMg = 0·75 (Bach et al., 2006). The magnesian serpentine that occurs with magnetite has XMg = 0·95 and it occurs with a brucite with XMg = 0·90 (Bach et al., 2006).
|
A third reaction on this diagram, which is shown as a bold line, is the oxidation of Fe-serpentine to produce magnetite and silica:
|
| (23) |
Because we cannot write a reaction producing magnetite using only magnesian end-members, the loop for reaction (23) does not close on the right side of Fig. 7a; instead it becomes asymptotic. Constraints on the location of the magnetite limb of this loop come from Evans & Frost (1975), who showed that in serpentinites 
= 0·02.
Our explanation of magnetite-absent and magnetite-present olivine hydration reactions is as follows. In a rock where silica activity is relatively high, either through serpentinization of adjacent Opx [silica activity conditions (1) in Fig. 7a] or through external supply of silica, olivine will hydrate under conditions where magnetite is not stable [silica activity conditions (2) in Fig. 7a]. Consequently, olivine will hydrate to ‘iron-rich’ serpentine and ‘iron-rich’ brucite (open circles on Fig. 7a). It should be noted that, if silica activity is high enough and there is sufficient flux of silica in the fluid, olivine may hydrate directly to serpentine by reaction (8) (Table 3), without the formation of brucite. If Opx is consumed from the rock, or if there is limited communication between Opx-bearing and olivine-bearing portions of the rock, olivine hydration will occur at lower silica activities [silica activity condition (3) in Fig. 7a] with the production of magnetite, and relatively magnesian brucite and serpentine (filled circles on Fig. 7a). It is clear that serpentinization can take place over a range of silica activities between (2) and (3) in Fig. 7a, which would explain why serpentine and brucite from partially serpentinized peridotites show a range of XMg (Oufi et al., 2002; Bach et al., 2006).
We cannot calculate Fig. 7a with thermodynamic rigor because we lack activity–composition relations for serpentine and brucite. However, we can use existing thermodynamic data to determine if our explanation is feasible. The relations we describe in Fig. 7a can hold only if the oxidation of serpentine to magnetite + silica [reaction (23)] in a serpentine with XMg = 0·9 occurs at an oxygen fugacity and silica activity near those that attend the hydration of olivine to serpentine + brucite. Figure 7b shows the location of reaction (23) calculated for three oxygen fugacities, compared with the silica activities of the serpentine–brucite buffer when in the presence of olivine. The figure has been calculated assuming that the end-member thermodynamic properties of Fe-serpentine are similar to those of greenalite and that solution of Fe and Mg in serpentine are ideal. Figure 7b shows that the low silica activity limit for serpentine with XFe = 0·1 lies close to the silica activity of the serpentine = brucite buffer in the presence of olivine, which means that the relations we describe in Fig. 7a are plausible.
The formation and stability of iron alloys
The oxygen necessary for the formation of magnetite is extracted from the decomposition of water [reaction (24)] and the hydrogen produced by this reaction makes serpentinites among the most reduced rocks on Earth. Partially serpentinized peridotites are so reducing that they commonly contain iron alloys. Awaruite (FeNi3) is the most commonly reported alloy, but the more iron-rich, variable-composition Fe–Ni alloy, taenite (Rosetti & Zuchetti, 1998) and even pure native iron (Chamberlain et al., 1965) are known. Although iron alloys are reported from serpentinites, something other than the displaced magnetite–iron buffer [reaction (25)] must form a ‘floor’ for oxygen fugacity in serpentinites. Native iron alloys are scarce in serpentinites; they usually occur as minute grains and make up an insignificant volume of the rock. Not only that, if the presence of native alloys did form the floor, then the oxygen-conserved reaction
|
| (26) |
Some other equilibrium must be operating in serpentinites that keeps the low silica activity from depleting Fe from the silicates while producing abundant magnetite and native iron. Two reasonable candidates for this are the following reactions, which cause a stability field for Fe-bearing brucite to lie between the stability field for magnetite and that for native iron:
|
| (27) |
|
| (28) |
). In the presence of olivine, reduced water activity reduces the stability fields of serpentine and Fe-brucite such that awaruite is stable. When olivine is depleted from the rock, water activity increases, and the fields for both serpentine and brucite expand, and, in doing so, they consume any alloys that may have formed early in the serpentinization process.
|
diagram showing the stabilities of magnetite, Fe(OH)2 in brucite, Fe3Si2O5(OH)4 (in serpentine). Shaded fields refer to conditions where water activity is fixed by the presence of olivine (plus serpentine + brucite, magnetite, or iron). Stippled fields refer to conditions where olivine is not present and water activity is unity (or nearly so). It should be noted how the increase in water activity after the depletion of olivine from the rock will cause the stability fields of brucite and serpentine to expand, consuming the field for native metals. |
SERPENTINITES AND CALCIUM |
|---|
|
TOPABSTRACTINTRODUCTIONSERPENTINITES AND SILICA...OXYGEN AND SULFUR FUGACITIESSERPENTINITES AND CALCIUMTHE OCCURRENCE OF OTHER...DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Diopside stability and fluid chemistry
Low-T fluids issuing from serpentinite-hosted vents are characteristically alkaline (commonly pH >10) and Ca-rich (Barnes & O’Neil, 1969
; Barnes et al., 1978; Palandri & Reed, 2004). Both of these features are attributable to the breakdown of clinopyroxene during serpentinization. Specifically, the absence of a stable Ca mineral requires that unbalanced dissolution of Ca2+ in the serpentinizing fluid is accommodated by OH– production (Li et al., 2004; Palandri & Reed, 2004). The loss of Ca2+ to a fluid phase is reflected by the Ca depletion observed in many serpentinized peridotites (Coleman, 1963; Puga et al., 1999; Shervais et al., 2005).
We argue here that the low silica activity of serpentinites plays a critical role in destabilizing clinopyroxene in serpentinizing fluids (Fig. 9; Table 5). At aH2 O buffered by the serpentine–brucite equilibrium [reaction (9)] diopside is unstable in fluids with the nominal composition of seawater at temperatures below about 200°C. The temperature of this reaction point decreases as pH and 
increase. The tremolite-forming reaction
|
| (30) |
). Other Ca-silicates either are unstable in the high-Mg environments of serpentinites (e.g. larnite, wollastonite) or require substantial amounts of Al (e.g. zoisite, grossular) or ferric iron (andradite). Thus the reaction
|
| (31) |
). These fluids are forming in the presence of tremolite (Allen & Seyfried, 2003) and their acidic character is consistent with the equilibria shown in Fig. 9a and b. It should also be noted from reaction (31) that diopside–serpentine equilibria are not affected by Mg concentration in the fluid as long as serpentine is stable.
|
). (b) pH vs T diagram showing how the solution of diopside on the serpentine–brucite buffer produces progressively more alkalic fluids with decreasing temperature.
|
Rodingites
Rodingites are rocks (most often basalts and gabbros, but rodingitized metasedimentary rocks and granites are also reported) that have undergone intensive metasomatism as a consequence of the serpentinization of surrounding peridotite (Coleman, 1963
, 1977; Schandl et al., 1989; O’Hanley, 1996). The metasomatism is usually described in terms of Ca enrichment, and, indeed, this is observed in many or most rodingites. However, an equally common and important component of the metasomatism is Si depletion (e.g. Coleman, 1963). Many rodingites are polymetamorphic rocks that have undergone several stages of both metamorphism and metasomatism (Frost, 1975; Evans et al., 1979; Schandl et al., 1989; O’Hanley et al., 1992; Puga et al., 1999; El-Shazly & Al-Belushi, 2004; Li et al., 2004). We concern ourselves here with metasomatic and metamorphic processes that are directly related to serpentinization sensu stricto. In a typical gabbroic rodingite, plagioclase is replaced by various Si-poor, usually hydrated Ca–Al silicates including grossular (and/or hydrogrossular), (clino)zoisite, prehnite, vesuvianite, wollastonite, and xonotlite [Ca6Si6O17(OH)2]. The commonest Fe–Mg silicates are usually chlorite and amphibole (especially tremolite) although serpentine minerals may also be present, especially if the protolith contained olivine. If one examines rodingite mineralogy with the assumption that Si is not fixed, it is apparent that the reaction of clinopyroxene and plagioclase in the presence of water will yield a variety of the most common rodingite assemblages (e.g. grossular–chlorite; Fig. 10a). These assemblages are the hydrated chemical equivalents of Cpx–plagioclase except that they are silica deficient. This becomes clear when silica is plotted explicitly (Fig. 10b). It is also clear from the stoichiometry of reactions (37)–(39) (Table 5). These reactions may be isochemical with respect to other components, but they are not isochemical with respect to silica.
|
The silica activity imposed by the brucite–serpentine buffer oversteps the rodingite reaction:
|
| (37) |
Hydrogrossular, a mineral that is virtually characteristic of rodingites and serpentinites and found in few other geological environments, is related to grossular by the reaction
|
| (40) |
Although silica activity may be the driving force for many rodingite-forming reactions, it is clear from bulk chemistry that elements besides Si, especially Ca, are also mobile (e.g. O’Hanley, 1996). The addition of Ca to rodingites is certainly a consequence of the high Ca content of serpentinizing fluids as discussed above. For rodingites formed from gabbroic rocks, much of the Ca addition probably results from the replacement of Na–Ca plagioclase with Ca-silicates such as grossular or prehnite. In these circumstances, Na is lost to the fluid and/or reprecipitated as Na minerals such as albite, jadeite or analcime, either locally or elsewhere in the hydrothermal system (Whitmarsh et al., 1998; Beard et al., 2002; Li et al., 2004).
|
THE OCCURRENCE OF OTHER LOW-SILICA MINERALS IN SERPENTINITES |
|---|
|
TOPABSTRACTINTRODUCTIONSERPENTINITES AND SILICA...OXYGEN AND SULFUR FUGACITIESSERPENTINITES AND CALCIUMTHE OCCURRENCE OF OTHER...DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Andradite and hydroandradite
Andraditic garnet containing varying amounts of Cr2O3, TiO2, and H2O has been found in a number of serpentinite parageneses including serpentine–awaruite–andradite ± magnetite ± brucite (Botto & Morrison, 1976
; Onuki et al., 1981; Beard & Hopkinson, 2000), serpentine + diopside + magnetite (Amthauer et al., 1974; Muntener & Hermann, 1994) and serpentine + carbonate + magnetite (Peters, 1965). Evidently it is stable in serpentinites over a range of oxygen fugacities and Ca activities.
Nominally the stability of andradite relative to clinopyroxene is governed by the reaction
|
| (41) |
|
|
).
In serpentinites where native iron is dissolved into Fe–Ni alloys (generally awaruite), andradite may form from diopside by the following oxygen-independent reactions (Fig. 11a):
|
| (42) |
|
| (43) |
|
| (44) |
|
–T diagram comparing the stability of andradite in serpentinite with that of diopside. Inset shows how the assemblage serpentine + andradite + brucite + iron could form by the reaction between fluids generated from low-T serpentinization of Cpx-bearing peridotite with overlying serpentinite.If reaction (44) controls the occurrence of andradite in a rock, the resulting assemblages would be Srp–Brc–Adr–Di–I or Srp–Brc–Adr–Mag–I. The serpentinites from ODP hole 1068A described by Beard & Hopkinson (2000
) have the assemblage Srp–Brc–Adr–Aw and lack both magnetite and diopside. Although reaction (44) limits the stability of andradite, the assemblage itself must have formed by another process. Beard & Hopkinson (2000) interpreted the serpentinites from hole 1068A as having formed in contact with fluids being expelled from an active serpentinization front at depth. The unusual magnetite-free assemblage Srp–Brc–Adr–Aw appears to be related to this fluid flow. Specifically, breakdown of clinopyroxene at depth via reaction (31) yields a high-pH, Ca-rich fluid, which allows the reaction
|
| (45) |
). It should be noted that, as for hydrogrossular, hydroandradite should be stable at silica activities lower than those at which andradite proper is stable.
Diaspore and corundum
Aluminum hydroxide (diaspore and/or gibbsite) occurs in hydrothermally altered submarine serpentinites from the Iberian margin and can occur in other serpentinized peridotites (Beard & Hopkinson, 2000; Kovacs et al., 2002). Diaspore plus brucite can form under extremely low Si activities by desilication of chlorite (Fig. 12). These silica activities, however, also require the desilication of serpentine and are not likely to be regularly attained in serpentinizing peridotite. The surprising presence of diaspore in the Iberian rocks, therefore, is probably a consequence of a second, low-T diaspore-forming reaction (see Fig. 12):
|
| (46) |
|
). Shaded area shows a small field for diaspore stability at low T and relatively high silica activity.Although corundum never occurs in serpentinites—chlorite is always more stable than serpentine + corundum—it is commonly found in hydrothermally altered rocks associated with metaperidotites (Schreyer et al., 1981
; Kerrich et al., 1987; Grapes & Palmer, 1996). Although these rocks have clearly been altered metasomatically sometime in their history, the timing of this metasomatism and, hence the origin of these rocks, is controversial [compare Schreyer et al. (1981) and Kerrich et al. (1987)]. Recently, Bucher et al. (2005) demonstrated that the corundum-bearing rocks probably formed by metasomatic alteration of aluminous metasedimentary rocks adjacent to serpentinite (or metaperidotite). The low silica activity necessary for the formation of corundum is caused by the extraction of silica from these rocks into the adjacent metaperidotite.
Jadeite
Although jadeite is a characteristic mineral of high-P/T metamorphic terranes, it is well known that serpentinization plays a key role in the formation in massive jadeite, including all known occurrences of precious jade. Specifically, jadeitite forms by crystallization from a Na-rich fluid with a low aSiO2 mediated by the presence of actively serpentinizing peridotite (Harlow & Sorensen, 2005). As noted by Harlow & Sorensen (2005), the relationship between jadeite formation and low aSiO2 related to serpentinization was anticipated by Coleman (1961). Harlow & Sorensen (2005) noted that quartz is almost unknown in precious jade deposits and that late-stage silication [analogous to that which occurs under the fluid-dominated conditions in fully serpentinized peridotite (Frost, 1985)] commonly produces rims of albitite or other relatively siliceous assemblages around jadeitite bodies.
Under even moderately low silica activities, the stability of jadeite is governed by its hydration to analcime [reaction (58)], rather than its reaction to albite + quartz [reaction (57)] (Fig. 13). Figure 13 shows that in these low-silica environments jadeite is stable to lower pressure than it would be in quartz-bearing rocks, but that it still requires moderate pressures (4–6 kbar) to stabilize jadeite relative to analcime. This explains why most jadeite deposits are formed in blueschist terranes and why analcime is present instead in low-P environments (e.g. Whitmarsh et al., 1998).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONSERPENTINITES AND SILICA...OXYGEN AND SULFUR FUGACITIESSERPENTINITES AND CALCIUMTHE OCCURRENCE OF OTHER...DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Recognizing the important role that silica activity plays in serpentinization not only provides petrologists with a conceptual framework to understand the complex mineralogical changes that accompany the hydration of peridotite rocks, it also presents us with important tools to unlock the processes that are involved. It provides us with a way to constrain the temperatures at which serpentinization occurs and with a means to characterize the pathways that fluid followed during serpentinization.
We recognize three minerals that may indicate the temperature at which serpentinization occurred (Table 6). The first of these is diopside. The assemblage serpentine (antigorite or lizardite)–brucite–diopside is considered to be indicative of the lowest temperatures of metamorphism (Evans, 1977). Unfortunately, most serpentinites lack diopside because, as temperatures fall, diopside becomes increasingly soluble in serpentinites. Between 300°C and 100°C the Ca2+ activity in the assemblage Di–Srp–Brc increases by more than three orders of magnitude (Fig. 9). The occurrence of diopside in a serpentinite, therefore, is a function of both the temperature of serpentinization and the fluid/rock ratio, with diopside being more stable at high T and low fluid/rock ratios.
|
As indicated in Fig. 11, andradite in brucite-bearing serpentinites is stable only around 225°C and thus the assemblage andradite–brucite–serpentine may indicate low-T serpentinization. However, the presence of andradite alone in serpentinites is not a temperature indicator, because andradite stability is strongly dependent on silica activity; it may be stable in brucite-free serpentinite at higher T. Finally, diaspore rarely occurs in serpentinites and its occurrence should indicate very low-T serpentinization. It is also stable only at relatively high silica activity and cannot occur with brucite.
Understanding that silica activity plays an important role in serpentinization provides petrologists with a key to interpreting the textures of serpentinites and, hence, to understanding the process by which a given rock body was hydrated. The assemblage brucite–serpentine controls the low silica activity in serpentinites, but reactions such as the hydration of Opx, the formation of magnetite and the dissolution of diopside contribute silica to the rock that will consume brucite. Therefore, the distribution of Opx and Cpx (or relict Opx and Cpx), magnetite, and brucite within a serpentinite may provide key information about the scale over which silica activity varied within a serpentinite, and this will provide important clues to the process by which serpentinization occurred.
Although it is simple to document the occurrences of Opx, magnetite, and diopside in serpentinites, it is much more difficult to constrain the spatial distribution of brucite. In some serpentinites brucite occurs in relatively large grains (albeit still microscopic) (Beard & Hopkinson, 2000). In others it is either absent or cryptically present as interlayers within serpentine. Petrologists, therefore, cannot assume brucite is absent from a serpentinite even if it is not evident petrographically. Future studies should either use microprobe analyses to identify serpentine–brucite mixtures [as was done by Bach et al. (2006)], or use X-ray mapping to find the distribution of brucite in a rock.
The distribution of brucite in a rock is important because it may provide evidence of fluid pathways that operated during serpentinization and may indicate the direction of fluid flow. Fluid that is moving downward, either seawater or fluid that is in equilibrium with gabbro, will have higher silica activity than the assemblage brucite + serpentine. Thus, prolonged downward fluid flow should consume brucite. Fluids in equilibrium with metagabbro may even have high enough silica activity to alter serpentine to talc. Downward-moving fluids, therefore, should move through channels that are brucite-free and, especially near contacts with gabbros, may be talc-bearing. In contrast, fluids that are moving upward from a serpentinization front should have low silica activity and their pathways should be marked by the presence of abundant brucite. If the temperature were low enough and if the fluids were rich enough in Ca2+, their presence may also be marked by the occurrence of andradite (Beard & Hopkinson, 2000).
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONSERPENTINITES AND SILICA...OXYGEN AND SULFUR FUGACITIESSERPENTINITES AND CALCIUMTHE OCCURRENCE OF OTHER...DISCUSSIONCONCLUSIONSREFERENCES |
|---|
We have shown that the distinctive petrological and geochemical properties of serpentinites—their highly magnetic nature, the reducing conditions, the calcic, high-pH fluids issuing from them—are all tied to the low silica activity of these rocks. The hydration of forsterite to form serpentine and brucite and this reaction's attendant effects on silica activity are ultimately responsible for all these distinctive mineralogical and hydrochemical features. The very low silica activities of the assemblage serpentine–brucite provide the chemical potential needed for virtually all of the geological and mineralogical oddities associated with serpentinization. First, low silica activity lowers the stability of the Fe3Si2O5(OH)4 component in serpentine, such that some of the ferrous iron in the primary olivine must go into magnetite (in most cases), metallic iron phases, or andradite. The formation of oxidized iron phases, especially magnetite, from the ferrous iron in silicates is the root cause of the characteristically reduced conditions found in serpentinites. A secondary effect of low oxygen fugacity is the reduction of sulfur, leading the stabilization of low-sulfur sulfides (e.g. heazlewoodite) and sulfur-free metal alloys (e.g. awaruite). Second, the instability of clinopyroxene at low silica activity results in its incongruent dissolution, producing the Ca-rich fluids associated with serpentinization. The pH of these fluids is a function of T, being acidic at high T but alkaline at the lower temperatures of most serpentinizing systems. Third, many of the unusual minerals that characterize serpentinites (e.g. hydrogarnet and jadeite) are low-Si phases stabilized by low silica activity. Finally, metasomatic processes that accompany serpentinization, including the formation of rodingite and blackwall alteration, are, first and foremost, desilication processes. This observation demonstrates that important geochemical variables, including pH and oxygen fugacity, may be derivative features that are controlled by chemical potentials to which they are only indirectly related.
|
ACKNOWLEDGEMENTS |
|---|
The authors would like to thank the Joint Oceanographic Institutions for allowing us to sail on IODP Expeditions 304 and 305 and providing us access to the samples. Without the chance to interact and puzzle over the formation of serpentinite in Hole U1309D we would never have come up with the insights necessary to write this paper. This work was supported by the grant JOI 48299 to B.R.F. and by the Virginia Museum of Natural History. The authors are particularly thankful for perceptive reviews by Wolfgang Bach, Bernard Evans, and Richard Laurent, which greatly improved the quality of this paper. Funding to pay the Open Access publication charges for this artical was provided by NSF grant EAR 06398980.
*Corresponding author. 307-766-4290. 307-766-6679. rfrost{at}uwyo.edu
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONSERPENTINITES AND SILICA...OXYGEN AND SULFUR FUGACITIESSERPENTINITES AND CALCIUMTHE OCCURRENCE OF OTHER...DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Abrajano TA, Sturchio NC, Bohlke JK, Lyon GL, Poreda RJ, Stevens CM. Methane–hydrogen seeps, Zambales Ophiolite, Philippines: Deep or shallow origin? Chemical Geology (1988) 71:211–222.[CrossRef][Web of Science]
Allen DE, Seyfried WE. Compositional controls on vent fluids from ultramafic-hosted hydrothermal systems at midocean ridges: An experimental study at 400°C, 500 bars. Geochimica et Cosmochimica Acta (2003) 67:1531–1542.[CrossRef][Web of Science]
Amthauer G, Kurtz W, Rost F, Schloemer H. Chemismus und Genese des Andradits aus dem Serpentinit des Val Malenco Schweiz. Mineralogische und Petrographische Mitteilungen (1974) 54:691–706.
Andersen DJ, Lindsley DH, Davidson PM. QUILF: A PASCAL program to assess equilibria among Fe, Mg, Mn, Ti oxides, pyroxenes, olivine, and quartz. Computers and Geosciences (1993) 19:1333–1350.[CrossRef]
Bach W, Paulick H, Garrido CJ, Ildefonse B, Meurer WP, Humphris SE. Unraveling the sequence of serpentinization reactions: petrography, mineral chemistry and petrophysics of serpentinites from MAR 15°N (ODP Leg 209, Site 1274). Geophysical Research Letters (2006) 33. L13306, doi:10.1029/2006GL025681.
Barnes I, O'Neil JR. The relationship between fluids in some fresh alpine-type ultramafics and possible modern serpentinization. Geological Society of America Bulletin (1969) 80:1947–1960.[Web of Science]
Barnes I, O'Neil JR, Trescases JJ. Present day serpentinization in New Caledonia, Oman, and Yugoslavia. Geochimica et Cosmochimica Acta (1978) 42:144–145.[CrossRef][Web of Science]
Beard JS, Hopkinson L. A fossil, serpentinization-related hydrothermal vent, Ocean Drilling Program Leg 173, Site 1068 (Iberia Abyssal Plain): Some aspects of mineral and fluid chemistry. Journal of Geophysical Research (2000) 105:16527–16540.[CrossRef]
Beard JS, Fullagar PD, Sinha AK. Gabbroic pegmatite intrusions, Iberia Abyssal Plain, ODP leg 173, Site 1070: Magmatism during a transition from non-volcanic rifting to seafloor spreading. Journal of Petrology (2002) 43:885–905.
Bethke CM. The Geochemists Workbench®, Version 6.0, GWB Essentials Guide (2005) Urbana, IL: Hydrogeology Program, University of Illinois. 76.
Botto RI, Morrison GH. Josephinite: a unique nickel–iron. American Journal of Science (1976) 276:241–274.
Bucher K, De Capitani C, Grapes R. The development of a margarite–corundum blackwall by metasomatic alteration of a slice of mica schist in ultramafic rocks, Kvesölen, Norwegian Caledonides. Canadian Mineralogist (2005) 43:129–156.[CrossRef][Web of Science]
Carmichael IS, Turner FJ, Verhoogen J. Igneous Petrology (1974) New York: McGraw–Hill.
Chamberlain JA, McLeod CR, Traill RJ, Lachance GR. Native metals in the Muskox intrusiton. Canadian Journal of Earth Sciences (1965) 2:188–215.
Charlou JL, Fouquet Y, Bougault H, Donval JP, Etoubleau J, Jean-Baptiste P, Dapoigny A, Appriou P, Rona PA. Intense CH4 plumes generated by serpentinization of ultramafic rocks at the intersection of the 15°20'N fracture zone and the Mid-Atlantic Ridge. Geochimica et Cosmochimica Acta (1998) 62:2323–2333.[CrossRef][Web of Science]
Charlou JL, Donval JP, Fouquet Y, Jean-Baptiste P, Holm N. Geochemistry of high H2 and CH4 vent fluids issuing from ultramafic rocks at the Rainbow hydrothermal field, (36°14'N, MAR). Chemical Geology (2002) 191:345–359.[CrossRef][Web of Science]
Coleman RG. Jadeite deposits of the Clear Creek area, New Idria district, San Benito County, California. Journal of Petrology (1961) 2:209–247.
Coleman RG. Serpentinites, Rodingites, and Tectonic Inclusions in Alpine-type Mountain Chains. Geological Society of America, Special Papers (1963) 73.
Coleman RG. Petrologic and geophysical nature of serpentinites. Geological Society of America Bulletin (1971) 82:897–918.
Coleman RG. Ophiolites, Ancient Oceanic Lithosphere? (1977) Berlin: Springer.
Eckstrand OR. The Dumont serpentinite: a model for control of nickeliferous opaque mineral assemblages by alteration reactions in ultramafic rocks. Economic Geology (1975) 70:183–201.
El-Shazly AK, Al-Belushi M. Petrology and chemistry of metasomatic blocks from Bawshir, northeastern Oman. International Geology Review (2004) 46:904–938.[Web of Science]
Evans BW. Metamorphism of alpine peridotite and serpentinite. Annual Review of Earth and Planetary Sciences (1977) 5:397–447.[CrossRef][Web of Science]
Evans BW. The serpentinite multisystem revisited; chrysotile is metastable. International Geology Review (2004) 46:479–506.[Web of Science]
Evans BW, Frost BR. Chrome-spinel in progressive metamorphism—a preliminary analysis. Geochimica et Cosmochimica Acta (1975) 39:959–972.[CrossRef][Web of Science]
Evans BW, Trommsdorff V, Richter W. Petrology of an eclogite–metarodingite suite at Cima di Gagnone, Ticino, Switzerland. American Mineralogist (1979) 64:15–31.[Abstract]
Frost BR. Contact metamorphism of serpentinite, chloritic blackwall, and rodingite at Paddy-Go-Easy-Pass, Central Cascades, Washington. Journal of Petrology (1975) 16:272–313.
Frost BR. On the stability of sulfides, oxides, and native metals in serpentinite. Journal of Petrology (1985) 26:31–63.
Frost BR, Beard JS, Abratis MW, Andreani M, Delacour A, Drouin M, Fryer P, McCaig A, Nozaka T, Ohara Y. the shipboard scientific parties of IODP expeditions 304 and 305. In: Importance of silica activity to the serpentinization processes: Insights from microrodingites in IODP hole U1309D (2005) EOS Transactions of the American Geophysical Union, Abstract V51B-1483.
Grapes R, Palmer K. (Ruby–sapphire)–chromian mica–tourmaline rocks from Westland, New Zealand. Journal of Petrology (1996) 37:293–315.
Groves DJ, Keays RR. Mobilization of ore-forming elements during alteration of dunites, Mt. Keith–Betheno, Western Australia. Canadian Mineralogist (1979) 17:373–389.
Harlow GE, Sorensen SS. Jade (nephrite and jadeite) and serpentinite: Metasomatic connections. International Geology Review (2005) 47:116–146.
Holland TJB, Powell R. An internally consistent thermodynamic dataset for phases of petrologic interest. Journal of Metamorphic Geology (1998) 16:309–343.[CrossRef][Web of Science]
Johnson JW, Oelkers EH, Helgeson HC. SUPCRT92: A software package for calculating the standard molal thermodynamic properties of minerals, gases, aqueous species, and reactions from 1 to 5000 bar and 0 to 1000°C. Computers and Geosciences (1992) 18:899–947.[CrossRef]
Kelley DS, Karson JA, Blackman DK, Früh-Green GL, Butterfield DA, Lilley MD, Olson EJ, Schrenk MO, Roe KK, Lebon GT, Rivizzigno P. An off-axis hydrothermal vent field near the Mid-Atlantic Ridge at 30. Nature (2001) 412:145–149.[CrossRef]
Kelley DS, Karson JA, Früh-Green GL, Yoerger DR, Shank TM, Butterfield DA, Hayes JM, Schrenk MO, Olson EJ, Proskurowski G, Jakuba M, Bradley A, Larson B, Ludwig K, Glickson D, Buckman K, Bradley AS, Brazelton WJ, Roe K, Elend MJ, Delacour A, Bernasconi SM, Lilley MD, Baross JA, Summons RE, Sylva SP. A serpentinite-hosted ecosystem; the Lost City hydrothermal field. Science (2005) 307:1428–1434.
Kerrich R, Fyfe WS, Barrett RL, Blair BB, Willmore LM. Corundum, Cr-muscovite rocks at O’Briens, Zimbabwe: the conjunction of hydrothermal desilification and LIL-element enrichment—geochemical and isotopic evidence. Contributions to Mineralogy and Petrology (1987) 95:481–498.[CrossRef][Web of Science]
Kovacs G, Raucsik B, Horváth P. Mineral composition of the Gyód serpentinite body, southern Transdanubia, Hungary. Acta Mineralogica–Petrographica (2002) 43:71–77.
Leblanc M, Lbouabi M. Native silver mineralization along a rodingite tectonic contact between serpentine and quartz diorite (Bou Azzer, Morocco). Economic Geology (1988) 83:1379–1391.
Le Gleuher M, Livi KJT, Veblen DR, Noack Y, Amouric M. Serpentinization of enstatite from Pernes, France: Reaction microstructures and the role of system openness. American Mineralogist (1990) 75:813–824.[Abstract]
Li X-P, Rahn M, Bucher K. Metamorphic processes in rodingites of the Zermatt–Saas ophiolites. International Geology Review (2004) 46:28–51.[Web of Science]
Muntener O, Hermann J. Titanian andradite in a metapyroxenite layer from the Malenco ultramafics (Italy): implications for Ti-mobility and low oxygen fugacity. Contributions to Mineralogy and Petrology (1994) 116:156–168.[CrossRef][Web of Science]
O'Hanley DS. Serpentinites: Records of Tectonic and Petrological History. Oxford Monographs on Geology and Geophysics (1996) 34:277.
O'Hanley DS, Schandl ES, Wicks FJ. The origin of rodingites from Cassiar, British Columbia and their use to estimate T and P(H2O) during serpentinization. Geochimica et Cosmochimica Acta (1992) 56:97–108.[CrossRef][Web of Science]
Onuki H, Yoshida T, Nodachi M. Notes on petrography and rock-forming mineralogy: X. Awaruite and other accessory minerals coexisting with Ti-rich hydroandradite in metamorphosed ultramafics of the Sanbagawa belt. Journal of the Japanese Association of Mineralogists, Petrologists, and Economic Geologists (1981) 76:372–375.
Oufi O, Cannat M, Horen H. Magnetic properties of variably serpentinized abyssal peridotites. Journal of Geophysical Research (2002) 107. doi:10.1029/2001JB000549.
Palandri JL, Reed MH. Geochemical models of metasomatism in ultramafic systems: Serpentinization, rodingitization, and sea floor carbonate chimney precipitation. Geochimica et Cosmochimica Acta (2004) 68:1115–1133.[CrossRef][Web of Science]
Peretti A, Dubessy J, Mullis J, Frost BR, Trommsdorff V. Highly reducing conditions during Alpine metamorphism of the Malenco peridotite (Sondrio, northern Italy) indicated by mineral paragenesis and H2 in fluid inclusions. Contributions to Mineralogy and Petrology (1992) 112:329–340.[CrossRef][Web of Science]
Peters T. A water-bearing andradite from the Totalp serpentine (Davos, Switzerland). American Mineralogist (1965) 50:1482–1486.[Web of Science]
Puga E, Nieto JM, Diaz de Federico A, Bodinier JL, Morten L. Petrology and metamorphic evolution of ultramafic rocks and dolerite dykes of the Betic ophiolitic association (Mulhacen Complex, SE Spain): evidence of eo-Alpine subduction following an ocean-floor metasomatic process. Lithos (1999) 49:23–56.[CrossRef][Web of Science]
Ramdohr P. Ore Minerals and their Intergrowths (1968) Oxford: Pergamon.
Rosetti P, Zuchetti S. Occurrence of native iron, Fe–Co and Ni–Fe alloys in the serpentinite from the Balangero asbestos mine (Western Italian Alps). Ofioliti (1998) 13:43–56.
Russell MJ, Arndt NT. Geodynamic and metabolic cycles in the Hadean. Biogeosciences (2005) 2:97–111.[Medline]
Russell MJ, Hall AJ. The emergence of life from iron monosulphide bubbles at a submarine hydrothermal redox and pH front. Journal of the Geological Society, London (1997) 154:377–402.
Schandl ES, O'Hanley DS, Wicks FJ. Rodingites in serpentinized ultramafic rocks of the Abitibi greenstone belt. Ontario. Canadian Mineralogist (1989) 27:579–591.[Web of Science]
Schreyer W, Werding C, Abraham K. Corundum–fuchsite rocks in greenstone belts of Southern Africa: petrology, geochemistry and possible origin. Journal of Petrology (1981) 22:191–231.
Schroeder T, John B, Frost BR. Geologic implications of seawater circulation through peridotite exposed at slow-spreading mid-ocean ridges. Geology (2002) 30:367–370.
Shervais JW, Kolesar P, Andreasen K. A field and chemical study of serpentinization—Stonyford, California: chemical flux and mass balance. International Geology Review (2005) 47:1–28.[Web of Science]
Shive PN, Frost BR, Peretti A. The magnetic properties of metaperidotitic rocks as a function of metamorphic grade: implications for crustal magnetic anomalies. Journal of Geophysical Research (1988) 93:12187–12195.[CrossRef]
Sleep NH, Meibom A, Fridriksson Th., Coleman RG, Bird DK. H2-rich fluids from serpentinization: Geochemical and biotic implications. Proceedings of the National Academy of Sciences of the USA (2004) 101:12818–12823.
Toft PB, Arkanai-Hamed J, Haggerty SE. The effects of serpentinization on density and magnetic susceptibility: a petrophysical model. Physics of the Earth and Planetary Interiors (1990) 65:137–157.[CrossRef][Web of Science]
Trommsdorff V, Evans BW. Progressive metamorphism of antigorite schist in the Bergell tonalite aureole (Italy). American Journal of Science (1972) 272:423–437.[Abstract]
Viti C, Mellini M, Rumori C. Exsolution and hydration of pyroxenes from partially serpentinized harzburgites. Mineralogical Magazine (2005) 69:491–507.
Whitmarsh RB, Beslier M-O, Wallace P, the Leg 173 Shipboard Scientific Party. Sites 1065–1070, Iberia Abyssal Plain. In: Proceedings of the Ocean Drilling Program, Initial Reports, 173—Whitmarsh RB, Beslier M.-O, Wallace PJ, et al, eds. (1998) College Station, TX: Ocean Drilling Program.
Yoder HS Jr, Tilley CE. Origin of basalt magmas: An experimental study of natural and synthetic rock systems. Journal of Petrology (1962) 3:342–532.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  B. W. Evans, S. M. Kuehner, and A. Chopelas Magnetite-free, yellow lizardite serpentinization of olivine websterite, Canyon Mountain complex, N.E. Oregon American Mineralogist, November 1, 2009; 94(11-12): 1731 - 1734. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Morishita, K. Hara, K. Nakamura, T. Sawaguchi, A. Tamura, S. Arai, K. Okino, K. Takai, and H. Kumagai Igneous, Alteration and Exhumation Processes Recorded in Abyssal Peridotites and Related Fault Rocks from an Oceanic Core Complex along the Central Indian Ridge J. Petrology, July 1, 2009; 50(7): 1299 - 1325. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. H. Ahmed, S. Arai, and M. Ikenne Mineralogy and Paragenesis of the Co-Ni Arsenide Ores of Bou Azzer, Anti-Atlas, Morocco Economic Geology, March 1, 2009; 104(2): 249 - 266. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. S. Beard, B. R. Frost, P. Fryer, A. McCaig, R. Searle, B. Ildefonse, P. Zinin, and S. K. Sharma Onset and Progression of Serpentinization and Magnetite Formation in Olivine-rich Troctolite from IODP Hole U1309D J. Petrology, March 1, 2009; 50(3): 387 - 403. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Klein and W. Bach Fe-Ni-Co-O-S Phase Relations in Peridotite-Seawater Interactions J. Petrology, January 1, 2009; 50(1): 37 - 59. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Pelletier, F. Vils, A. KALT, and K. Gmeling Li, B and Be Contents of Harzburgites from the Dramala Complex (Pindos Ophiolite, Greece): Evidence for a MOR-type Mantle in a Supra-subduction Zone Environment J. Petrology, December 4, 2008; (2008) egn057v1. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. W. Evans Control of the Products of Serpentinization by the Fe2+Mg-1 Exchange Potential of Olivine and Orthopyroxene J. Petrology, October 16, 2008; (2008) egn050v1. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. R. Frost, J. S. Beard, A. McCaig, and E. Condliffe The Formation of Micro-Rodingites from IODP Hole U1309D: Key To Understanding the Process of Serpentinization J. Petrology, September 1, 2008; 49(9): 1579 - 1588. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||