Journal of Petrology

This Article Abstract FREE Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Alert me to new issues of the journal Add to My Personal Archive Download to citation manager Search for citing articles in: ISI Web of Science (48) Request Permissions Google Scholar Articles by MATVEEV, S. Articles by GREEN, D. H. Search for Related Content GeoRef GeoRef Citation Social Bookmarking          What's this?

Journal of Petrology Volume 42 Number 4 Pages 721-729 2001
© Oxford University Press 2001

Effect of Silica Activity on OH^- IR Spectra of Olivine: Implications for Low-aSiO₂ Mantle Metasomatism

S. MATVEEV^1,*,, H. ST. C. O’NEILL¹, C. BALLHAUS^2,, W. R. TAYLOR¹ and D. H. GREEN¹

¹RESEARCH SCHOOL OF EARTH SCIENCES, AUSTRALIAN NATIONAL UNIVERSITY, CANBERRA, A.C.T. 0200, AUSTRALIA
²GEOLOGY DEPARTMENT, AUSTRALIAN NATIONAL UNIVERSITY, CANBERRA, A.C.T. 0200, AUSTRALIA

Received November 2, 1999; Revised typescript accepted June 26, 2000

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES AND... DISCUSSION CONCLUSIONS REFERENCES

 
Hydrogen solubility and hydroxyl substitution mechanism in olivine at upper-mantle conditions are not only a function of pressure, temperature, water fugacity and hydrogen fugacity, but are also influenced by silica activity. Olivine synthesized in equilibrium with magnesiowüstite displays hydroxyl stretching bands in the wavenumber range from 3640 to 3430 cm^–1. In contrast, olivine in equilibrium with orthopyroxene shows absorption bands in a narrower wavenumber range from 3380 to 3285 cm^–1. The two fundamentally different spectra are assigned to hydroxyl in tetrahedral and octahedral sublattices, respectively. Olivine in equilibrium with orthopyroxene is also less capable of incorporating hydroxyl, relative to olivines in equilibrium with magnesiowüstite, by about a factor of ten. A comparison of spectra obtained as part of this study with hydroxyl spectra of natural mantle olivines shows that the latter display hydroxyl stretching patterns reminiscent of equilibrium with magnesiowüstite, although undoubtedly olivine in the Earth’s mantle coexists with orthopyroxene. This may be attributed to a metasomatic overprint by a low-silica fluid and/or melt that was in reaction relationship with orthopyroxene. A likely metasomatic agent is a carbonatitic melt. When carbonatitic melts decompose to oxides and CO₂, they may temporarily impose a low-aSiO₂ environment inherited by the olivine structure. If this suggestion proves true, Fourier transform IR spectroscopy may be used to fingerprint metasomatic episodes in the lithospheric mantle.

KEY WORDS: FTIR spectrometry; olivine; mantle; metasomatism; water

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES AND... DISCUSSION CONCLUSIONS REFERENCES

 
Since the recognition about 30 years ago that nominally anhydrous minerals (NAM) can incorporate hydrogen in lattice defects (Beran, 1970; Wilkins & Sabine, 1973; Beran & Putnis, 1983; Aines & Rossman, 1984), considerable experimental and analytical effort has been expended on estimating the OH^- storage capacity of mantle minerals and determining the key factors controlling OH^- substitution (e.g. Miller et al., 1987; Mackwell & Kohlstedt, 1990; Rossman & Smyth, 1990; Skogby et al., 1990; Bai & Kohlstedt, 1993; Kohlstedt et al., 1996). The studies revealed that structural hydroxyl significantly enhances creep rates and electrical conductivity of mantle rocks. It also influences pressure–temperature positions of phase transitions (Mackwell et al., 1985; Karato et al., 1986; Schock et al., 1989; Kawamoto et al., 1996). Emphasis was placed on olivine and its polymorphs, as (Mg,Fe)₂SiO₄ constitutes the principal phase in upper-mantle lithologies (Miller et al., 1987; Mackwell & Kohlstedt, 1990; Kohlstedt et al., 1996; Kohlstedt & Mackwell, 1998).

Hydroxyl occurrence in olivine is defined by the crystal defect structure, and therefore OH^- incorporation mechanisms are expected to be largely controlled by the diffusion rates of cations, anions, point defects, electrons and polarons. The diffusivity of a vacancy is a few orders of magnitude larger than the diffusivities of ionic species and dependent on the vacancy concentration (Jost, 1960; Ottonello, 1997):

Here, DA_A is the diffusion coefficient of ion A at a given structural site A, DV_A is the diffusion coefficient of a vacancy at site A, and [V_A] is the fraction of vacancies in the respective sublattice. The Arrhenius plot in Fig. 1 illustrates diffusion rates of the major crystal species determined for San Carlos olivine. The fastest diffusing species (excluding electrons and polarons) is hydrogen. The saturation of a 1 mm diameter crystal of San Carlos olivine with hydrogen at mantle temperatures (1000–1300°C) occurs in a matter of minutes (Kohlstedt & Mackwell, 1998). To equilibrate the same grain with respect to metal point defects will take hours to days (Kohlstedt & Mackwell, 1998, and references therein). To equilibrate the grain with respect to silicon vacancies in response to changes in silica activity (aSiO₂) may require at least a few months for the chosen temperature range. Consequently, although the mantle olivine point defect structure may survive, olivine can hardly preserve its pristine hydroxyl concentration during transport to the Earth’s surface (Miller et al., 1987).

View larger version (18K): [in this window] [in a new window]   Fig. 1. Diffusivities of the major crystal species determined for San Carlos olivine. H(a), hydrogen diffusion //[100] (Kohlstedt & Mackwell, 1998); VMe'', Fe and Mg, diffusivities of metal vacancies, Fe2+ and Mg2+ //[001], respectively (Hermeling & Schmalzried; 1984; Chakraborty et al., 1994; Kohlstedt & Mackwell, 1998); VSi'''', diffusivity of silicon vacancies assuming DVSi'''' Me2+, where Me2+ denotes coefficient of chemical diffusion (i.e. Hermeling & Schmalzried; 1984); O and Si, diffusivities of O2- and Si4+ //[001] (Gerard & Jaoul, 1989; Houlier et al., 1990).

 

Numerous experimental studies have been performed to estimate the water storage capacity in olivine at mantle conditions (Bai & Kohlstedt, 1992, 1993; Kohlstedt et al., 1996; Kohlstedt & Mackwell, 1998). In these studies grains of San Carlos olivine with surfaces polished parallel to the principal crystallographic directions were placed in a fine-grained matrix of talc plus brucite, anticipating that the matrix, upon thermal decomposition, would supply water and produce orthopyroxene, thus defining water fugacity (fH₂O) and aSiO₂ at fixed pressure, temperature and oxygen fugacity (fO₂). To test the effect of fO₂ on the hydroxyl solubility in olivine, Bai & Kohlstedt (1993) also pre-annealed their starting olivines at 1300°C and 1 atm at various fO₂ values from 10^-11 to 10^-4 MPa. Subsequent high-pressure ‘hydration’ experiments showed that olivines, pre-annealed at high fO₂, incorporate more hydrogen than olivine pre-annealed at low fO₂ (Bai & Kohlstedt, 1993), suggesting that ferric iron plays a key role in producing defects in the ‘dry’ olivine structure. Hydroxyl solubility in olivine was found to rise gradually with increasing fH₂O, reaching 1200 ppm H₂O at log fH₂O 6·5 and 13 GPa, 1100°C (Kohlstedt et al., 1996). Most of the hydroxyl was assigned to vacant metal sites, and the amount of hydroxyl on tetrahedral sites was considered to be negligible (Kohlstedt & Mackwell, 1998).

The problem is that experimental OH^- absorption spectra synthesized up to now do not fully reproduce the spectra observed in natural olivine. Experimental olivines tend to show strong IR absorption around 3600 cm^-1, whereas some natural mantle olivines show almost exclusively IR absorption around 3340 cm^-1. Miller et al. (1987), who noted this peculiarity, concluded that the mechanisms of hydrogen incorporation in experiment might differ from those observed in nature.

In this experimental study, we succeeded in producing two fundamentally different OH^- IR spectra characterized by peaks located around either 3600 cm^-1 or 3340 cm^-1, thus reproducing the whole range of IR spectra observed in natural olivines. In contrast to previous experiments, we took a more straightforward experimental approach by adding the silica buffer endmembers (either orthopyroxene or magnesiowüstite) and water directly to the charge. We also performed our experiments at higher temperatures close to the ‘wet’ solidus (1300°C) allowing solution–precipitation processes to facilitate equilibration of the olivine structure at the imposed aSiO₂. It is shown that the key factor in controlling OH^- substitution in olivine is silica activity (aSiO₂). An OH^- IR absorption spectrum of olivine may serve as an aSiO₂ sensor for mantle processes, and may prove useful to identify metasomatic episodes in the Earth’s mantle.

	EXPERIMENTAL PROCEDURES AND RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES AND... DISCUSSION CONCLUSIONS REFERENCES

 
The starting minerals were single-crystal olivines (Fo_90–91) and equilibrium orthopyroxenes (En_90–91) separated from xenoliths collected from Mt Porndon, Victoria (Australia). Average grain diameters were chosen to be 700 µm, to ensure that after an experiment, grain sizes were still large enough for quantitative Fourier transform IR (FTIR) analysis. To vary aSiO₂, two sets of experiments were performed:

1. high-aSiO₂ experiments in which olivine was embedded in a finely powdered matrix of olivine and orthopyroxene (1:1) from crystals of the same occurrence; aSiO₂ is controlled by the equilibrium Mg₂SiO₄ + SiO₂ = Mg₂Si₂O₆;
   
2. low-aSiO₂ experiments in which olivine was embedded in a finely powdered matrix of synthetic magnesiowüstite (M₈₀W₂₀); aSiO₂ is controlled by the equilibrium Mg₂SiO₄ = 2MgO + SiO₂.
   

All experiments were performed in platinum capsules. In both sets of experiments fO₂ was set to the Re–ReO₂ buffer equilibrium (1·2–1·4 log-bars above the FMQ equilibrium depending on total pressure), by adding Re and ReO₂ directly to the charge. Water was added by microsyringe. fH₂O was varied by adding to some experiments known quantities of silver oxalate (Ag₂C₂O₄) in addition to water, which decomposes to CO₂. To prevent the capsules from rupturing during decompression–quenching, total fluid was kept below 4–5 mg.

All experiments were performed in a 0·5 inch (1·25 cm) piston-cylinder press at ANU at 1573 K and 2–4 GPa using talc outer sleeves and fired pyrophyllite crucibles as pressure media (Table 1). Temperature was monitored with a Pt₇₀Rh₃₀–Pt₉₄Rh₆ thermocouple and kept within ±5 K. Pressure was adjusted manually to within ±500 MPa. To correct for talc friction, a -10% pressure correction was applied (Green et al., 1966). Run times of 24 h were considered long enough to approach point defect equilibrium (see Beran & Putnis, 1983; Miller et al., 1987; Young et al., 1993; Kohlstedt et al., 1996). Experiments were quenched by turning off power supply.

View this table: [in this window] [in a new window]   Table 1: Experimental details and water hydroxyl expressed as H/106 Si and H2O in ppm

 

After quenching, the capsules were weighed and pierced to release the fluid. Orthopyroxene-buffered runs emanated a rather thick yellowish gel presumably caused by elevated dissolved SiO₂ and MgO contents. In contrast, the fluid in magnesiowüstite-buffered runs was optically clear, apparently because the solubility of oxide species at low aSiO₂ is lower (Manning, 1994). CO₂, the more volatile endmember, was determined by weighing the capsule before and immediately after piercing at 298 K. Water content was estimated by reweighing the CO₂-free capsule after drying for 5 h at 400 K. Agreement between added and weighed relative and absolute H₂O and CO₂ amounts was found to be within ±0·2 mg.

After drying, olivine single crystals were separated from the matrix and the buffer assemblage. The buffer was checked with X-ray diffraction (XRD) and was found to be intact after experiment. The largest olivine grains were doubly polished to platelets of 80–150 µm thickness. As a result of the anhedral character of the recovered olivine grains, all samples had to be polished crystallographically unoriented.

Attainment of equilibrium
A good test that point defect equilibrium was reached is to check that the olivine run products are homogeneous and in FeMg_-1 exchange equilibrium with the orthopyroxene or magnesiowüstite matrix, given that vacancies are anticipated to diffuse faster than Fe²⁺ and Mg²⁺ cations. If FeMg_-1 exchange equilibrium was reached one may in turn assume vacancy equilibrium. Selected images of starting olivine grains and run products are shown in Fig. 2. In orthopyroxene-buffered runs, a large proportion of the original starting olivine grains completely recrystallized, judging from the poikilitic textures and the ubiquitous presence of euhedral fluid inclusions. The mechanism proposed is solution–precipitation, consistent with the fact that the fluid in orthopyroxene-buffered runs was oxide enriched. In contrast, olivine grains equilibrated with magnesiowüstite largely remained intact. They do trap fluid inclusions, but these are randomly shaped and generally aligned along healed cracks. In addition, FeMg_-1 exchange with magnesiowüstite is more sluggish than with orthopyroxene. Larger grains often preserve Fe/Mg ratios (and OH^- absorption spectra) inherited from starting grains, notably in cores (Fig. 2c). Obviously, in the absence of solution–precipitation, which is only evident in the high-aSiO₂ runs, the rate-limiting step for FeMg_-1 exchange equilibrium is solid-state diffusion.

View larger version (92K): [in this window] [in a new window]   Fig. 2. Characteristic images of starting and annealed olivines. (a) Transmitted light image of starting olivine; (b) transmitted light image of olivine equilibrated with orthopyroxene (note poikilitic textures and primary fluid inclusions); (c) back-scattered electron image of olivine partially equilibrated with magnesiowüstite; light core is in MgFe-1 exchange disequilibrium with magnesiowüstite; EPMA profile [AB] is shown in Fig. 6.

 

A summary of electron probe microanalysis (EPMA) results is shown in Fig. 3. As a rule, all olivines turned out to be more magnesian than in the starting material, notably those equilibrated with magnesiowüstite. In the high-aSiO₂ runs, the degree of Mg enrichment may be attributed to preferential fractionation of iron into the fluid or minor Fe loss to the Pt capsule via the fluid phase. As orthopyroxene-buffered runs with mixed H₂O–CO₂ fluid (reduced fH₂O) did not experience Fe loss to the same extent, although run conditions were similar, we consider the first proposition the more likely. K_DFe–Mg^ol–opx was found to be close to unity. In the low-aSiO₂ runs, it is readily apparent that magnesiowüstite (M₈₀W₂₀) of the starting mix was not in exchange equilibrium to start with (Fig. 2c). None the less, equilibrated grains, usually those with diameters <200 µm, were found to be homogeneous within analytical error.

View larger version (18K): [in this window] [in a new window]   Fig. 3. Typical olivine compositions before and after experiments with orthopyroxene and magnesiowüstite.

 

FTIR spectrometry
Single-crystal FTIR measurements were carried out with a BOMEM spectrometer and a Spectratech IR-plan microscope. Care was taken that crystal areas exposed to the beam were free of cracks, free of visible fluid inclusions, and in FeMg_-1 exchange equilibrium with the silica buffer (see Fig. 3). Beam diameters ranged from 50 to 100 µm. To compensate for the lack of crystallographic orientation, we analysed at least five grains of every run to obtain consistent absorption patterns. All single-crystal IR spectra were collected at room temperature in the wavenumber range from 600 to 4000 cm^-1 at a resolution of 4 cm^-1. The acquired spectra were normalized to 1 cm sample thickness and processed with the OPUS 2.0 Brucker software.

Experimental results are included in Table 1, and typical IR spectra for the two starting setups are shown in Fig. 4. Olivine grains in equilibrium with magnesiowüstite displayed OH^- stretching bands in the wavenumber range from 3640 to 3430 cm^-1, along with a single peak located at 3295 cm^-1 (Fig. 4a). In contrast, olivine in equilibrium with orthopyroxene showed much weaker IR absorption bands in the wavenumber range from 3380 to 3285 cm^-1. The underlying ‘plateau’ of elevated background in Fig. 4a was also noted by Miller et al. (1987), who attributed it to molecular water located in submicroscopic defects and/or fluid micro-inclusions. It is noteworthy that there is little overlap in terms of absorption bands in olivines equilibrated at different aSiO₂; none of the peaks found in olivine equilibrated with orthopyroxene were also observed in olivine equilibrated with magnesiowüstite, or vice versa.

View larger version (26K): [in this window] [in a new window]   Fig. 4. Typical OH- IR spectra measured on (a) olivine in equilibrium with orthopyroxene (2 GPa); (b) olivines in equilibrium with magnesiowüstite at 4 GPa (the upper spectrum) and 2 GPa.

 

Water contents in olivine
Absolute hydroxyl concentrations in olivine were estimated following the procedure of Paterson (1982), applying an orientation factor of 1/3 as suggested for unoriented grains (Paterson, 1982). The IR spectra were integrated over the critical wavenumber range from 3640 to 3100 cm^-1. For olivine coexisting with orthopyroxene (Fig. 4a) we integrated the sharp absorption peaks in the wavenumber range from 3380 to 3285 cm^-1, including the elevated background under the peaks but excluding the broad stretching area attributed to molecular water (see Miller et al., 1987). For olivine coexisting with magnesiowüstite we integrated the spectra over the whole wavenumber range with hydroxyl stretching bands, as the ratio of oriented to molecular water is much larger. Calculated hydroxyl concentrations are included in Table 1 and illustrated in Fig. 5. Also shown in Fig. 5 is the semi-empirical equation of Kohlstedt et al. (1996). It is apparent that the low-aSiO₂ olivines are by a factor of 2·5 richer in hydroxyl than predicted by that equation whereas olivines in equilibrium with orthopyroxene are by about the same factor lower in water. At given temperature and fH₂O, a low-aSiO₂ olivine is around 10 times more capable of incorporating hydrogen than a high-aSiO₂ olivine. It should be noted, however, that our absolute water contents as quoted in Table 1 and Fig. 5 should be considered as semiquantitative only, because (1) the spectra were acquired on randomly oriented grains, and (2) there were uncertainties as to how to incorporate potential molecular water in orthopyroxene-buffered runs. None the less, the differences in spectra as well as in absolute water contents as a function of aSiO₂ remain significant.

View larger version (12K): [in this window] [in a new window]   Fig. 5. Calculated hydroxyl concentrations in olivines from Table 1. , olivine equilibrated with orthopyroxene; , olivine in equilibrium with magnesiowüstite. The dashed line is the semi-empirical fit by Kohlstedt et al. (1996) relating OH- in olivine to fH2O. For reasons of clarity, runs with identical conditions and results are represented by only an average symbol. Horizontal error bars reflect uncertainties in adding of silver oxalate to the charge.

 

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES AND... DISCUSSION CONCLUSIONS REFERENCES

 
This study has demonstrated that OH^- substitution and OH^- concentration in olivine is, amongst other factors, a function of aSiO₂. For the first time, it is possible to reproduce the range of OH^- IR spectra observed in natural mantle olivines. In this section, we attempt to interpret the spectra in terms of the olivine defect structure. We discuss reasons why previous experimental work may have failed to reproduce the OH^- IR spectra of natural olivine. We then address the petrological implications of our results for estimating maximum water storage capacity in the Earth’s mantle and fingerprinting episodes of metasomatic overprint in the lithospheric upper mantle.

Hydroxyl substitution in olivine
Hydroxyl-defect substitution mechanisms in olivine have been discussed by Beran & Putnis (1983), Bai & Kohlstedt (1993), Sykes et al. (1994), Libowitzky & Beran (1995), Kohlstedt et al. (1996) and Kohlstedt & Mackwell (1998). Beran & Putnis (1983), and later Libowitzky & Beran (1995), suggested that high-frequency stretching OH^- bands (3500–3700 cm^-1) may be due to hydroxyl dipoles balancing silicon vacancies, aligned along former Si–O bonds, but did not rule out that bands from 3500 to 3600 cm^-1 could also reflect hydroxyl in octahedral vacancies. Bai & Kohlstedt (1993) discriminated two groups of peaks; group 1 from 3640 to 3445 cm^-1 and group 2 from 3445 to 3170 cm^-1. These correspond to our low-aSiO₂ and high-aSiO₂ absorption bands, respectively, shown in Fig. 4. Interestingly, Bai & Kohlstedt encountered the two groups of peaks together, i.e. in single olivine grains supposedly in aSiO₂ equilibrium with orthopyroxene. The group 1 peaks were assigned to hydroxyl balancing metal vacancies in octahedral positions and/or hydrogen bonded with doubly charged oxygens in interstitial positions. The group 2 peaks were thought to reflect hydroxyls balancing silicon tetrahedral vacancies and/or hydrogen to singly charged oxygen interstitials. Later, however, Kohlstedt et al. (1996) and Kohlstedt & Mackwell (1998) attributed all absorption bands indiscriminately to hydroxyls balancing metal vacancies.

Assuming aSiO₂ and fO₂ as independent thermodynamic characteristics, the nature of major olivine point defects can be described by two reactions (Nakamura & Schmalzried, 1983):

Following the Kröger & Vink (1956) notation, Fe_Me^x and Fe_Me^• denote ferrous and ferric iron on octahedral M-sites, respectively, V_Me'' denotes vacancies in an M-site, and Si_Si^x and Fe_Si' stand for silicon and ferric iron in tetrahedral position. Because site fractions of cations in their normal positions are close to unity, the following mass balance equations result:

Equations (3) and (4) illustrate that increasing aSiO₂ enhances defects in octahedral metal positions while reducing defect concentration in tetrahedral sites. Hydrogen incorporation in the olivine structure involves reduction of Fe_Me^• and compensates negatively charged point defects (Wright & Catlow, 1994; Kohlstedt et al., 1996):

Equation (5) implies that OH^- will substitute in olivine only if there is ferric iron. However, Libowitzky & Beran (1995) and Kohn (1996) have shown that even pure forsterite can incorporate appreciable OH^- in its structure if it crystallized at high fH₂O. Clearly, this implies that hydrogen can also substitute directly for cations in the M1 and M2 sites, questioning whether hydroxyl incorporation necessarily requires the presence of ferric iron.

As aSiO₂ was the only variable in our experiments that could have influenced the defect structure, we conclude that the two different types of OH^- IR spectra result from hydroxyl substituting in different structural sites. In keeping with equations (1) and (2), the low-frequency absorption bands in the 3380–3285 cm^-1 range are assigned to OH^- substitution in the metal sublattice (V_Me''), whereas the bands in the high-frequency 3640–3430 cm^-1 range, as well as the single peak at 3295 cm^-1, are attributed to OH^- substitution in the tetrahedral sublattice (Fe_Si' and V_Si''''). This assignment is basically consistent with the works of Sykes et al. (1994) and Libowitzky & Beran (1995).

Comparison with previous experimental olivine IR spectra
Previous researchers (e.g. Bai & Kohlstedt, 1992; Kohlstedt et al., 1996) have reported OH^- IR absorption in the 3640–3430 cm^-1 wavenumber range, in olivine supposedly in equilibrium with orthopyroxene. They have failed to encounter the sharp discrimination that we relate to variations in aSiO₂. The reason may lie in different experimental strategies.

Kohlstedt and coworkers used San Carlos olivine as starting material. San Carlos is known to absorb in the 3640–3630 cm^-1 wavenumber range (Miller et al., 1987), suggesting a defect structure in apparent equilibrium with magnesiowüstite. The size of starting olivine single crystals used by Kohlstedt et al. (1996) was close to 1 mm in diameter. Orthopyroxene and water, necessary to buffer aSiO₂ and incorporate OH^- in the structure, had to be produced during experiment by dehydration and reaction of talc and brucite. Most of the experiments of Kohlstedt et al. were performed at 1100°C with run times ranging from 2 to 24 h. Electron and transmitted light images showed that, except for near-pure forsterite overgrowth rims, the starting olivine single crystals remained essentially intact during experiment. They preserved both their original shape and their composition (Fo₉₀). In light of the data and experience presented here, this is an important observation. It implies that (1) the cores failed to reach FeMg_-1 exchange equilibrium with the matrix, (2) solution–precipitation was limited to grain boundaries (pure forsterite), and (3) the cores, later analysed with FTIR spectrometry, may have kept their starting defect structure.

To underline this point, Fig. 6 shows an Fe–Mg profile across the partially equilibrated olivine grain displayed in Fig. 2c, along with two FTIR spectra. The forsteritic rim (Fo₉₇), in exchange equilibrium with magnesiowüstite, returned a low-aSiO₂ absorption spectrum. The unequilibrated core (Fo₉₁) gave a spectrum in apparent equilibrium with orthopyroxene, presumably inherited from the starting crystal. We therefore conclude that silicon vacancies may diffuse about as fast as the chemical Fe–Mg exchange. By implication, one cannot expect point defect equilibrium to be reached if temperature and run time were insufficient to reach FeMg_-1 exchange equilibrium.

View larger version (37K): [in this window] [in a new window]   Fig. 6. (a) EPMA profile through partially equilibrated olivine (see Fig. 2c); (b) OH IR spectrum measured on olivine unequilibrated core; (c) OH- IR spectrum measured on olivine in FeMg-1 exchange equilibrium with magnesiowüstite.

 

Fingerprinting mantle metasomatism?
Miller et al. (1987) presented IR absorption spectra from natural olivines. The majority of their spectra showed absorption bands in the 3430–3640 cm^-1 range, in apparent equilibrium with magnesiowüstite. Out of six mantle olivines they analysed, only two samples showed absorption in the 3285–3380 cm^-1 range, although from the petrographic information provided olivine must have coexisted with orthopyroxene. Absorption bands around 3340 cm^-1 are plausible, given that aSiO₂ in the upper mantle is normally buffered by the forsterite–enstatite equilibrium (Green & Falloon, 1998). Absorption around 3600 cm^-1, however, is enigmatic and requires explanation.

We suggest that the low-aSiO₂ absorption bands may be indicative of infiltration of a metasomatic melt intrinsically low in aSiO₂ and not in equilibrium with orthopyroxene. One possibility is a carbonatitic melt. Wallace & Green (1988) and Green & Wallace (1988) have demonstrated that at high CO₂ partial pressure, the first melt increment on the solidus of orthopyroxene-saturated fertile peridotite at pressures >2·1 GPa is carbonatitic in composition. The stability of this melt is very sensitive to pressure and bulk mantle composition. When it leaves its stability field, e.g. by decompression or by infiltrating mantle domains that are too depleted to sustain a carbonatitic melt, it will decompose according to the reaction

liberating oxides and giving off CO₂ as a vapour phase. The oxides will temporarily impose a low-aSiO₂ environment in the reaction front, i.e. near the olivine–magnesiowüstite equilibrium, to be later consumed by reaction with orthopyroxene to form clinopyroxene and/or olivine (Green & Wallace, 1988; Dalton & Presnall, 1998).

If this proves true, hydrogen absorption bands in mantle olivine may be a more sensitive indicator for mantle metasomatism than petrographically ambiguous alteration rims around orthopyroxene (i.e. Zinngrebe & Foley, 1994; Ionov, 1998); the latter are easily overlooked or misinterpreted as deuteric in origin. Thus, not only is natural olivine unable to serve as indicator for primary water contents in the mantle (Miller et al., 1987), it also does not even seem to record the primary silica activity, which was initially in equilibrium with orthopyroxene. It remains to be seen how widespread carbonatitic metasomatism in the Earth’s mantle really is, especially in samples with no petrographically visible evidence for alteration.

	CONCLUSIONS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES AND... DISCUSSION CONCLUSIONS REFERENCES

 
The incorporation of hydroxyl groups in the olivine structure is a strong function of silica activity, both in terms of defect substitution mechanism and absolute hydroxyl concentration. Olivine in equilibrium with magnesiowüstite incorporates about a factor of 10 more hydroxyl than olivine coexisting with orthopyroxene. Previous experimental studies commonly reported major absorption bands around 3600 cm^-1 for single grain olivines coexisting with orthopyroxene, leading us to suspect that previous experimentalists failed to equilibrate the olivine point defect structure with respect to aSiO₂. Our interpretation of single olivine grain OH^- IR spectra suggests that estimates regarding the water storage capacity of the upper mantle as well as hydrogen diffusion coefficients, based on previously reported experimental spectra, must be considered with caution.

A comparison of our OH^- IR spectra with spectra measured on olivines from natural mantle samples shows that many natural olivines yield OH^- absorption spectra more akin to the presence of magnesiowüstite, although they demonstrably coexist with orthopyroxene. It is speculated that such samples have experienced a metasomatic overprint by a carbonatitic melt. When a carbonatitic melt decomposes upon decompression it reacts with orthopyroxene to produce clinopyroxene and/or olivine, temporarily imposing in the reaction front a silica activity near the forsterite–magnesiowüstite equilibrium. As such, it may preserve a record of a metasomatic episode in the mantle, provided that the episode comes with fluctuation in aSiO₂.

	ACKNOWLEDGEMENTS

 
This work would have been impossible without the invaluable technical help by B. Hibberson and N. Ware. We are grateful to T. Mernagh (AGSO) and D. Royd (Forensic Division of the Australian Federal Police in Canberra) for letting us use their FTIR facilities. H. Keppler kindly reanalysed selected experimental olivines with FTIR spectrometry at Bayerisches Geoinstitut to cross-check and verify the results presented here. Discussions with H. Palme, S. Chakraborty, A. Putnis and A. Borisov, and reviews by D. Kohlstedt, G. Rossman, E. Johnson and A. Sobolev substantially improved the manuscript. S. Matveev and C. Ballhaus acknowledge generous financial support through DFG Grants Ba 964/4-1, 7-1 and 7-2, as well as through ARC Grant A39602510 to R. Arculus.

	FOOTNOTES

 
^*Corresponding author. Telephone: +49-(251)-833-3048. Fax: +49-(251)-833-8397. E-mail: matveev{at}nwz.uni-muenster.de

Present address: Institut für Mineralogie, Westfälische Wilhelms-Universität Münster, Correnstr. 24, D-48149 Münster, Germany.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES AND... DISCUSSION CONCLUSIONS REFERENCES

 
Aines, R. D. & Rossman, G. R. (1984). Water in minerals? A peak in the infrared. Journal of Geophysical Research 89, 4059–4071.

Bai, Q. & Kohlstedt, D. L. (1992). Substantial hydrogen solubility in olivine and implications for water storage in the mantle. Nature 357, 672–674.

Bai, Q. & Kohlstedt, D. L. (1993). Effects of chemical environment on the solubility and incorporation mechanism for hydrogen in olivine. Physics and Chemistry of Minerals 19, 460–471.

Beran, A. (1970). Über (OH)-Gruppen in Olivin. (The OH^- groups in olivine.) Anzeiger der Österreichischen Akademie der Wissenschaften, Mathematisch Naturwissenschaftliche Klasse 106, 73–74.

Beran, A. & Putnis, A. (1983). A model of the OH^- positions in olivine, derived from infrared-spectroscopic investigations. Physics and Chemistry of Minerals 9, 57–60.

Chakraborty, S., Farver, J. R., Yund, R. A. & Rubie, D. C. (1994). Mg tracer diffusion in synthetic forsterite and San Carlos Olivine as a function of P, T and fO₂. Physics and Chemistry of Minerals 21, 489–500.

Dalton, J. A. & Presnall, D. C. (1998). Carbonatitic melts along the solidus of model lherzolite in the system CaO–MgO–Al₂O₃–SiO₂–CO₂ from 3 to 7 GPa. Contributions to Mineralogy and Petrology 131, 123–135.

Gerard, O. & Jaoul, O. (1989). Oxygen diffusion in San Carlos Olivine. Journal of Geophysical Research 94, 4119–4128.

Green, D. H. & Falloon, T. J. (1998). Pyrolite; a Ringwood concept and its current expression. In: Jackson, I. (ed.) The Earth’s Mantle; Composition, Structure, and Evolution. Cambridge: Cambridge University Press, pp. 311–378.

Green, D. H. & Wallace, M. E. (1988). Mantle metasomatism by ephemeral carbonatite melts. Nature 336, 459–462.

Green, T. H., Ringwood, A. E. & Major, A. (1966). Friction effects and pressure calibration in a piston-cylinder apparatus at high pressure and temperature. Journal of Geophysical Research 71, 3589–3594.

Hermeling, J. & Schmalzried, H. (1984). Tracer diffusion of the Fe-cations in olivine (Fe_XMg_1–X)₂SiO₄ (III). Physics and Chemistry of Minerals 11, 161–166.

Houlier, B., Cheraghmakani, M. & Jaoul, O. (1990). Silicon diffusion in San Carlos Olivine. Physics of the Earth and Planetary Interiors 62, 329–340.

Ionov, D. (1998). Trace element composition of mantle-derived carbonates and coexisting phases in peridotite xenoliths from alkali basalts. Journal of Petrology 39, 1931–1941.

Jost, W. (1960). Diffusion in Solid State Physics. New York: Academic Press.

Karato, S.-I., Paterson, M. S. & FitzGerald, J. D. (1986). Rheology of synthetic olivine aggregates: influence of grain size and water. Journal of Geophysical Research 91, 8151–8176.

Kawamoto, T., Hervig, R. L. & Holloway, J. R. (1996). Experimental evidence for a hydrous transition zone in the early Earth’s mantle. Earth and Planetary Science Letters 142, 587–592.

Kohlstedt, D. L. & Mackwell, S. J. (1998). Diffusion of hydrogen and intrinsic point defects in olivine. Zeitschrift für Physikalische Chemie (International Journal of Research in Physical Chemistry and Chemical Physics) 2, 147–162.

Kohlstedt, D. L., Keppler, H. & Rubie, D. C. (1996). Solubility of water in the alpha, beta and gamma phases of (Mg,Fe)₂SiO₄ Contributions to Mineralogy and Petrology 123, 345–357.[Web of Science]

Kohn, S. C. (1996). Solubility of H₂O in nominally anhydrous mantle minerals using H-1 MAS NMR. American Mineralogist 81, 1523–1526.[Abstract]

Kröger, F. A. & Vink, H. J. (1956). Relation between the concentration of imperfections in crystalline solids. In: Seitz, F. & Turnhall, D. (eds) Solid State Physics 3. New York: Academic Press, pp. 367–435.

Libowitzky, E. & Beran, A. (1995). OH^- defects in forsterite. Physics and Chemistry of Minerals 22, 387–392.

Mackwell, S. J. & Kohlstedt, D. L. (1990). Diffusion of hydrogen in olivine: implications for water in the mantle. Journal of Geophysical Research 95, 5079–5088.

Mackwell, S. J., Kohlstedt, D. L. & Paterson, M. S. (1985). The role of water in the deformation of olivine single crystals. Journal of Geophysical Research 90, 11319–11333.

Manning, C. E. (1994). The solubility of quartz in H₂O in the lower crust and upper mantle. Geochimica et Cosmochimica Acta 58, 4831–4839.[Web of Science]

Miller, G. H., Rossman, G. R. & Harlow, G. E. (1987). The natural occurrence of hydroxide in olivine. Physics and Chemistry of Minerals 14, 461–472.

Nakamura, A. & Schmalzried, H. (1983). On the nonstoichiometry and point defects of olivine. Physics and Chemistry of Minerals 10, 27–37.

Ottonello, G. (1997). Principles of Geochemistry. New York: Columbia University Press.

Paterson, M. (1982). The determination of hydroxyl by infrared absorption in quartz, silicate glasses and similar materials. Bulletin de Minéralogie 105, 20–29.[Web of Science]

Rossman, G. R. & Smyth, J. R. (1990). Hydroxyl contents of accessory minerals in mantle eclogites and related rocks. American Mineralogist 75, 775–780.[Abstract]

Saxena, S. K. & Fei, Y. (1987). High pressure and high temperature fluid fugacities. Geochimica et Cosmochimica Acta 51, 783–791.

Schock, R. N., Duba, A. G. & Schankland, T. J. (1989). Electrical conduction in olivine. Journal of Geophysical Research 94, 5829–5839.

Skogby, H., Bell, D. R. & Rossman, G. R. (1990). Hydroxide in pyroxene: variations in the natural environment. American Mineralogist 75, 764–774.[Abstract]

Sykes, D., Rossman, G. R., Veblen, D. R. & Grew, E. S. (1994). Enhanced H and F incorporation in borian olivine. American Mineralogist 79, 904–908.[Abstract]

Wallace, M. E. & Green, D. H. (1988). An experimental determination of primary carbonatite magma composition. Nature 335, 343–346.

Wilkins, R. W. T. & Sabine, W. (1973). Water content of some nominally anhydrous silicates. American Mineralogist 58, 508–516.

Wright, K. & Catlow, C. R. A. (1994). A computer simulation study of (OH) defects in olivine. Physics and Chemistry of Minerals 20, 515–518.

Young, T. E., Green, H. W., Hofmeister, A. M. & Walker, D. (1993). Infrared spectroscopic investigation of hydroxyl in ß-(Mg,Fe)₂SiO₄ and coexisting olivine: implications for mantle evolution and dynamics. Physics and Chemistry of Minerals 19, 409–422.

Zinngrebe, E. & Foley, S. (1994). Metasomatism in two natural peridotites; effects of low-aSiO₂ melts. Mineralogical Magazine 58A, 1006–1007.

CiteULike    Connotea    Del.icio.us    What's this?

This article has been cited by other articles:

				A. Hushur, M. H. Manghnani, J. R. Smyth, F. Nestola, and D. J. Frost Crystal chemistry of hydrous forsterite and its vibrational properties up to 41 GPa American Mineralogist, May 1, 2009; 94(5-6): 751 - 760. [Abstract] [Full Text] [PDF]

				R. Sundvall, H. Skogby, and R. Stalder Dehydration-hydration mechanisms in synthetic Fe-poor diopside European Journal of Mineralogy, January 1, 2009; 21(1): 17 - 26. [Abstract] [Full Text] [PDF]

				J. Gose, P. Reichart, G. Dollinger, and E. Schmadicke Water in natural olivine--determined by proton-proton scattering analysis American Mineralogist, October 1, 2008; 93(10): 1613 - 1619. [Abstract] [Full Text] [PDF]

				I. Kovacs, J. Hermann, H. St. C. O'Neill, J. F. Gerald, M. Sambridge, and G. Horvath Quantitative absorbance spectroscopy with unpolarized light: Part II. Experimental evaluation and development of a protocol for quantitative analysis of mineral IR spectra American Mineralogist, May 1, 2008; 93(5-6): 765 - 778. [Abstract] [Full Text] [PDF]

				C. Aubaud, A. C. Withers, M. M. Hirschmann, Y. Guan, L. A. Leshin, S. J. Mackwell, and D. R. Bell Intercalibration of FTIR and SIMS for hydrogen measurements in glasses and nominally anhydrous minerals American Mineralogist, May 1, 2007; 92(5-6): 811 - 828. [Abstract] [Full Text] [PDF]

				R. Stalder and T. Ludwig OH incorporation in synthetic diopside European Journal of Mineralogy, May 1, 2007; 19(3): 373 - 380. [Abstract] [Full Text] [PDF]

				K. D. Litasov and E. Ohtani Effect of water on the phase relations in Earth's mantle and deep water cycle Geological Society of America Special Papers, January 1, 2007; 421(0): 115 - 156. [Abstract] [Full Text] [PDF]

				A. M. WALKER, S. DEMOUCHY, and K. WRIGHT Computer modelling of the energies and vibrational properties of hydroxyl groups in {alpha}- and {beta}-Mg2SiO4 European Journal of Mineralogy, October 1, 2006; 18(5): 529 - 543. [Abstract] [Full Text] [PDF]

				J. L. Mosenfelder, N. I. Deligne, P. D. Asimow, and G. R. Rossman Hydrogen incorporation in olivine from 2-12 GPa American Mineralogist, February 1, 2006; 91(2-3): 285 - 294. [Abstract] [Full Text] [PDF]

				K. Wright Atomistic Models of OH Defects in Nominally Anhydrous Minerals Reviews in Mineralogy and Geochemistry, January 1, 2006; 62(1): 67 - 83. [Full Text] [PDF]

				A. Beran and E. Libowitzky Water in Natural Mantle Minerals II: Olivine, Garnet and Accessory Minerals Reviews in Mineralogy and Geochemistry, January 1, 2006; 62(1): 169 - 191. [Full Text] [PDF]

				H. Keppler and N. Bolfan-Casanova Thermodynamics of Water Solubility and Partitioning Reviews in Mineralogy and Geochemistry, January 1, 2006; 62(1): 193 - 230. [Full Text] [PDF]

				A. J. Berry, J. Hermann, H. S.C. O'Neill, and G. J. Foran Fingerprinting the water site in mantle olivine Geology, November 1, 2005; 33(11): 869 - 872. [Abstract] [Full Text] [PDF]

				N. Bolfan-Casanova Water in the Earth's mantle Mineralogical Magazine, June 1, 2005; 69(3): 229 - 257. [Abstract] [Full Text] [PDF]

				A. Kurka, M. Blanchard, and J. Ingrin Kinetics of hydrogen extraction and deuteration in grossular Mineralogical Magazine, June 1, 2005; 69(3): 359 - 371. [Abstract] [Full Text] [PDF]

				S. MATVEEV, M. PORTNYAGIN, C. BALLHAUS, R. BROOKER, and C. A. GEIGER FTIR Spectrum of Phenocryst Olivine as an Indicator of Silica Saturation in Magmas J. Petrology, March 1, 2005; 46(3): 603 - 614. [Abstract] [Full Text] [PDF]

				D. R. BELL, G. R. ROSSMAN, and R. O. MOORE Abundance and Partitioning of OH in a High-pressure Magmatic System: Megacrysts from the Monastery Kimberlite, South Africa J. Petrology, August 1, 2004; 45(8): 1539 - 1564. [Abstract] [Full Text] [PDF]

				K. Regenauer-Lieb, K. Regenauer-Lieb, and T. Kohl Water solubility and diffusivity in olivine: its role in planetary tectonics Mineralogical Magazine, August 1, 2003; 67(4): 697 - 715. [Abstract] [Full Text] [PDF]

This Article Abstract FREE Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Alert me to new issues of the journal Add to My Personal Archive Download to citation manager Search for citing articles in: ISI Web of Science (48) Request Permissions Google Scholar Articles by MATVEEV, S. Articles by GREEN, D. H. Search for Related Content GeoRef GeoRef Citation Social Bookmarking          What's this?

Online ISSN 1460-2415 - Print ISSN 0022-3530

Copyright © 2009 Oxford University Press

Oxford Journals Oxford University Press

• Site Map
• Privacy Policy
• Frequently Asked Questions

Other Oxford University Press sites: Oxford University Press Oxford Journals China Oxford Journals Japan American National Biography Booksellers' Information Service Children's Fiction and Poetry Children's Reference Corporate & Special Sales Dictionaries Dictionary of National Biography Digital Reference English Language Teaching Higher Education Textbooks Humanities International Education Unit Law Medicine Music Online Products Oxford English Dictionary Reference Rights and Permissions Science School Books Social Sciences Very Short Introductions World's Classics