Click
Here
GEOPHYSICAL RESEARCH LETTERS, VOL. 33, L12S13, doi:10.1029/2006GL025858, 2006
for
Full
Article
Ferric iron in Al-bearing post-perovskite
Ryosuke Sinmyo,1 Kei Hirose,1 Hugh St. C. O’Neill,2 and Eiji Okunishi3
Received 27 January 2006; revised 3 May 2006; accepted 9 May 2006; published 10 June 2006.
[1] The Fe3+/SFe ratios in both (Al, Fe)-bearing MgSiO3
post-perovskite phase and Ca-ferrite-type Al-phase,
synthesized in a natural mid-oceanic ridge basalt (MORB)
composition at 113 GPa and 2240 K, were determined by
electron energy-loss near-edge structure (ELNES)
spectroscopy. The results demonstrate that post-perovskite
has high proportions of Fe3+ with Fe3+/SFe ratio of
0.65 ± 0.04. The high Fe3+ concentration in post-perovskite
may have significant effects on its physical properties, phase
stability, and iron partitioning. In contrast, the Ca-ferritetype Al-phase, which is the second Fe-bearing phase in a
subducting MORB crust, is enriched in Fe2+ rather than Fe3+
with Fe3+/SFe ratio of 0.15 and 0.29. Citation: Sinmyo, R.,
K. Hirose, H. St. C. O’Neill, and E. Okunishi (2006), Ferric
iron in Al-bearing post-perovskite, Geophys. Res. Lett., 33,
L12S13, doi:10.1029/2006GL025858.
1. Introduction
[2] A number of experimental studies demonstrated that
(Al, Fe)-bearing MgSiO3 perovskite, the dominant mineral in
the lower mantle, contains substantial amounts of Fe3+ [e.g.,
McCammon et al., 1992; McCammon, 1997; Lauterbach et
al., 2000; Gloter et al., 2000]. The high Fe3+ content in
perovskite has large effects on electrical conductivity [Xu and
McCammon, 2002], radiative thermal conductivity [Keppler
et al., 1994], elasticity [Andrault et al., 2001], and Mg-Fe
partitioning [e.g., Frost and Langenhorst, 2002; McCammon
et al., 2004]. Recently Murakami et al. [2004] discovered a
novel phase transition in MgSiO3 from perovskite to postperovskite under the lowermost mantle conditions. An important issue is whether the post-perovskite phase can also
accommodate significant amounts of Fe3+. The major element composition of post-perovskite formed in natural peridotitic mantle and MORB compositions were reported by
Murakami et al. [2005] and Hirose et al. [2005], but the
valence state of iron has not been reported yet.
[3] By analogy with perovskite, Fe3+ in post-perovskite
has possibly significant effects on its physical and chemical
properties. In addition, it may have different effect from that
of Fe2+ on the stability of post-perovskite [Mao et al.,
2004]. Moreover, iron metal is likely present in the lower
mantle together with Fe3+-rich perovskite [e.g., Miyajima et
al., 1999; Lauterbach et al., 2000; Frost et al., 2004]. The
Fe3+/SFe ratio in post-perovskite is also important for the
fate of such metallic iron in the D00 region.
1
Department of Earth and Planetary Sciences, Tokyo Institute of
Technology, Tokyo, Japan.
2
Research School of Earth Sciences, Australian National University,
Canberra, Australia.
3
JEOL Ltd., Tokyo, Japan.
Copyright 2006 by the American Geophysical Union.
0094-8276/06/2006GL025858$05.00
[4] Recently, the Fe3+/SFe ratio has been quantitatively
determined with a nm-scale spatial resolution by ELNES
spectroscopy [van Aken et al., 1998; Lauterbach et al.,
2000; van Aken and Liebscher, 2002; Garvie et al., 2004].
Here we determined the Fe3+/SFe ratios in both (Al, Fe)bearing MgSiO3 post-perovskite and Ca-ferrite-type Alphase formed in a MORB bulk composition, based on
similar analytical techniques. The results show that postperovskite contains large amounts of Fe3+ comparable to
that in Al-bearing perovskite.
2. Analytical Procedures
[5] The sample was synthesized from a gel starting
material with a chemical composition of normal MORB
[Hirose et al., 2005]. In order to reduce all Fe3+ to Fe2+,
small amounts of gel powder was heated at 1273 K for 1 hr
in a H2-CO2 gas-mixing furnace in which oxygen fugacity
was controlled to be slightly above the iron-wüstite buffer
(3 log units below the QFM buffer). The high-pressure
phases were formed at 113 GPa and 2240 K in a diamondanvil cell (DAC) by heating with a Nd:YAG laser for
120 mins. Other experimental details were described by
Hirose et al. [2005].
[6] Angle-dispersive X-ray diffraction measurements in
situ at high pressure showed that the sample was composed
of post-perovskite, a-PbO2-type SiO2 phase, CaSiO3 perovskite, and Ca-ferrite-type Al-phase. Both post-perovskite
and CaSiO3 perovskite converted to amorphous phase
during decompression to atmospheric pressure. On the other
hand, crystal structures of the Ca-ferrite-type Al-phase and
SiO2 phase were preserved at ambient condition. The
sample recovered from DAC was then Ar ion-thinned under
liquid nitrogen cooling for transmission electron microscope
(TEM) analysis. A typical TEM image is shown in Figure 1.
The chemical composition of each constituent mineral was
Figure 1. TEM micrograph of the sample synthesized at
113 GPa and 2240 K in a MORB bulk composition. MP,
MgSiO3-rich post-perovskite; CF, Ca-ferrite-type Al-phase;
CP, CaSiO3 perovskite; SiO2, a-PbO2-type silica phase.
L12S13
1 of 3
L12S13
SINMYO ET AL.: FERRIC IRON IN AL-BEARING POST-PEROVSKITE
Figure 2. Typical Fe L2,3-edge ELNES spectra of silicate
glass samples. Data reduction included the correction for
dark current and channel-to-channel gain variation of the
detector, and the subtraction of an inverse power law
background.
determined by energy-dispersive (EDS) analytical system
and was reported by Hirose et al. [2005].
[7] We obtained Fe L2,3-edge ELNES spectra from
(Al, Fe)-bearing post-perovskite and Ca-ferrite-type
Al-phase by using JEM-2100F TEM with GIF-Tridiem
energy filters. A couple of silicate glass samples were also
analyzed in order to check consistency with the previous
Mössbauer spectroscopy measurements [Jayasuriya et al.,
2004]. We measured thin part but avoided analysis of very
edge of the sample. The ELNES spectra were collected
with energy dispersion of 0.2 eV per channel, energy
resolution of 0.8 – 0.9 eV (measured as width of the
zero-loss peak at half maximum), and integration time of
0.2– 1.0 sec. The incident beam current was about 0.5 nA,
and the fluence rate was 3.0*104 e/Å2/sec. The timeseries measurements were made for silicate glass samples.
Spectra were corrected for dark current and channel-tochannel gain variation. The quantitative determination of
Table 1. Results of ELNES Measurements on Silicate Glassesa
Sample: D31/7/01
Sample: C14/12/01
a
Integration
Time
Fe3+/SFe
by ELNES
0.2 s
0.2 s
0.2 s
Average
0.5 s
0.5 s
0.5 s
Average
1.0 s
1.0 s
1.0 s
1.0 s
Average
All average
0.5 s
1.0 s
1.0 s
Average
0.80
0.72
0.84
0.79±0.06
0.81
0.85
0.74
0.80±0.06
0.83
0.75
0.81
0.81
0.80±0.04
0.80±0.04
0.47
0.44
0.39
0.43 ± 0.04
Error value indicates one standard deviation.
b
Data from Jayasuriya et al. [2004].
Fe3+/SFe
by Mössbauerb 15
L12S13
Figure 3. Representative Fe L2,3-edge ELNES spectra of
(Al,Fe)-bearing MgSiO3 post-perovskite.
the Fe3+/SFe ratio is based on the white line intensities at
the Fe L2,3-edge, following the method described by van
Aken et al. [1998].
3. Results
[8] The ELNES spectra were obtained from two silicate
glasses; sample C14/12/01 with 16.36 wt% FeO and sample
D31/7/01 with 27.06 wt% FeO (Figure 2), showing the
Fe3+/SFe ratios of 0.43 ± 0.04 and 0.80 ± 0.04, respectively
(Table 1). These results are consistent with the previous
Mössbauer analyses, in which the Fe3+/SFe ratios were
determined to be 0.38 and 0.83, respectively [Jayasuriya et
al., 2004]. Results of time-series measurements on these
silicate glasses are also shown in Table 1. No remarkable
differences were observed in Fe3+/SFe ratio with increasing
collection time from 0.2 to 1.0 sec. This suggests that the
effect of beam-induced oxidation was negligible during
integration times of 1.0 sec.
[9] We obtained the ELNES spectra from four different
grains of post-perovskite with 23.2 wt% total FeO content
(Figure 3) [Hirose et al., 2005, Table 2]. The Fe3+/SFe
ratios range from 0.59 to 0.69 (ave. 0.65 ± 0.04) (Table 2),
indicating that the post-perovskite phase contains more
Fe3+ than Fe2+. The high Fe3+ concentration in postperovskite is also inferred from the EDS analyses of major
element composition; the sum of cations considerably
exceeds 2.0 when normalized to 3 oxygen if all Fe is
assumed to be Fe2+ [Hirose et al., 2005].
[10] The Fe L2,3-edge ELNES spectra of the Ca-ferritetype Al-phase are relatively weak (Figure 4), because the
total iron content (9.6 wt%) [see Hirose et al., 2005, Table 2]
Table 2. Results of ELNES Measurements on Post-Perovskite and
Ca-Ferrite-Type Al-Phase
0.83
Post-perovskite
0.38
Ca-ferrite-type Al-phase
2 of 3
Integration Time
Fe3+/SFe by ELNES
1.0 s
1.0 s
1.0 s
1.0 s
Average
1.0 s
1.0 s
0.67
0.63
0.69
0.59
0.65 ± 0.04
0.29
0.15
L12S13
SINMYO ET AL.: FERRIC IRON IN AL-BEARING POST-PEROVSKITE
L12S13
[14] Acknowledgments. We thank N. Takafuji for preparation of
TEM samples. The constructive reviews by R. Tronnes and anonymous
referee were helpful to improve the manuscript.
References
Figure 4. Fe L2,3-edge ELNES spectra of Ca-ferrite-type
Al-phase.
is much less than that of post-perovskite. Analyses of two
different grains revealed the Fe3+/SFe ratios of 0.15 and 0.29
(Table 2). Fe2+ is therefore dominant in the Ca-ferrite-type
Al-phase, contrary to post-perovskite.
4. Discussions
[11] The high Fe3+ concentration in post-perovskite might
result from disproportionation of Fe2+ in the starting material to produce Fe3+ and metallic Fe [e.g., Lauterbach et al.,
2000; Frost et al., 2004]. Metallic iron was, however, not
observed under the TEM in this study. Oxidation during
heating, which easily occurs in a LHDAC [e.g., Dubrovinsky
et al., 1998], may be a likely scenario in our experiments.
Such iron metal formed by disproportionation of Fe2+ was not
commonly observed in the previous multi-anvil experiments
either [Frost and Langenhorst, 2002].
[12] The Fe3+/SFe ratio in Mg-perovskite has been
extensively studied, showing that the concentration of
Fe3+ strongly correlates with the Al content and is independent from the oxygen fugacity [e.g., McCammon, 1997;
Lauterbach et al., 2000; Frost and Langenhorst, 2002; Frost
et al., 2004]. Our analyses demonstrated that post-perovskite
of the present study contains 0.254 Al3+ and 0.223 Fe3+ in a
two-cation formula. These Al and Fe3+ concentrations are
quite consistent with the Al-Fe3+ relation in perovskite
[Frost and Langenhorst, 2002]. The Al and Fe3+ in postperovskite may have positive correlation similarly to those in
perovskite, although the effect of oxygen fugacity on Fe3+
content is not known for post-perovskite.
[13] The Fe3+ content in perovskite formed in a MORB
composition was not analyzed in this study, but it could be
about 0.25 Fe3+ in a two-cation formula estimated from the
Al content [Hirose et al., 2005] and Al-Fe3+ relationship
[Frost and Langenhorst, 2002]. This amount is comparable
to that included in the post-perovskite phase. The analyses
of Ca-ferrite-type Al-phase, which is another Fe-bearing
phase in MORB, demonstrate that Fe2+ is dominant in this
phase with much less Fe3+ concentrations than that in postperovskite. These suggest that the total Fe3+ content in a
subducting MORB crust does not change remarkably at the
post-perovskite phase transition in the lowermost mantle.
Andrault, D., N. Bolfan-Casanova, and N. Guignot (2001), Equation of
state of lower mantle (Al, Fe)-MgSiO3 perovskite, Earth Planet. Sci.
Lett., 193, 501 – 508.
Dubrovinsky, L., et al. (1998), Structure of b-iron at high temperature and
pressure, Science, 281, 11.
Frost, D. J., and F. Langenhorst (2002), The effect of Al2O3 on Fe-Mg
partitioning between magnesiowüstite and magnesium silicate perovskite,
Earth Planet. Sci. Lett., 199, 227 – 241.
Frost, D. J., C. Liebske, F. Langenhorst, C. A. McCammon, R. G.
Trønnes, and D. C. Rubie (2004), Experimental evidence for the
existence of iron-rich metal in the Earth’s lower mantle, Nature,
428, 409 – 412.
Garvie, L. A. J., T. J. Zega, P. Rez, and R. Buseck (2004), Nanometer-scale
measurements of Fe3+/SFe by electron energy-loss spectroscopy: A cautionary note, Am. Mineral., 89, 1610 – 1616.
Gloter, A., F. Guyot, I. Martinez, and C. Colliex (2000), Electron energyloss spectroscopy of silicate perovskite-magnesiowüstite high-pressure
assemblages, Am. Mineral., 85, 1452 – 1458.
Hirose, K., N. Takafuji, N. Sata, and Y. Ohishi (2005), Phase transition and
density of subducted MORB crust in the lower mantle, Earth Planet. Sci.
Lett., 237, 239 – 251.
Jayasuriya, K. D., H. St. C. O’Neill, A. J. Berry, and S. J. Campbell (2004),
A Mössbauer study of the oxidation state of Fe in silicate melts, Am.
Mineral., 89, 1597 – 1609.
Keppler, H., C. A. McCammon, and D. C. Rubie (1994), Crystal-field and
charge-transfer spectra of (Mg, Fe)SiO3 perovskite, Am. Mineral., 79,
1215 – 1218.
Lauterbach, S., C. A. McCammon, P. van Aken, F. Langenhorst, and
F. Seifert (2000), Mössbauer and ELNES spectroscopy of (Mg,Fe)
(Si,Al)O3 perovskite: A highly oxidised component of the lower
mantle, Contrib. Miner. Petrol., 138, 17 – 26.
Mao, W. L., G. Shen, V. B. Prakapenka, Y. Meng, A. J. Cambell, D. Heinz,
J. Shu, R. J. Hemley, and H. K. Mao (2004), Ferromagnesian postperovskite silicates in the D00 layer of the earth, Proc. Natl. Acad. Sci. U. S.
A., 101, 15,867 – 15,869.
McCammon, C. (1997), Perovskite as a possible sink for ferric iron in the
lower mantle, Nature, 387, 694 – 696.
McCammon, C. A., D. C. Rubie, C. R. Ross II, F. Seifert, and H. St. C.
O’Neill (1992), Mössbauer spectra of 57Fe0. 05Mg0. 95SiO3 perovskite at
80 and 298 K, Am. Mineral., 77, 894 – 897.
McCammon, C. A., S. Lauterbach, F. Seifert, F. Langenhorst, and P. A. van
Aken (2004), Iron oxidation state in lower mantle mineral assemblages I.
Empirical relations derived from high-pressure experiments, Earth Planet. Sci. Lett., 222, 435 – 449.
Miyajima, N., K. Fujino, N. Funamori, T. Kondo, and T. Yagi (1999),
Garnet-perovskite transformation under conditions of the Earth’s lower
mantle: An analytical transmission electron microscopy study, Phys.
Earth Planet. Inter., 116, 117 – 131.
Murakami, M., K. Hirose, K. Kawamura, N. Sata, and Y. Ohishi (2004),
Post-perovskite phase transition in MgSiO3, Science, 304, 855 – 858.
Murakami, M., K. Hirose, N. Sata, and Y. Ohishi (2005), Post-perovskite
phase transition and mineral chemistry in the pyrolitic lowermost mantle,
Geophys. Res. Lett., 32, L03304, doi:10.1029/2004GL021956.
van Aken, P. A., and B. Liebscher (2002), Quantification of ferrous/ferric
ratios in minerals: New evaluation schemes of Fe L23 electron energy-loss
near-edge spectra, Phys. Chem. Miner., 29, 188 – 200.
van Aken, P. A., B. Liebscher, and V. J. Styrsa (1998), Quantitative determination of iron oxidation states in minerals using Fe L2,3 – edge electron
energy-loss near-edge structure spectroscopy, Phys. Chem. Miner., 25,
323 – 327.
Xu, Y., and C. McCammon (2002), Evidence for ionic conductivity in
lower mantle (Mg, Fe) (Si,Al)O 3 perovskite, J. Geophys. Res.,
107(B10), 2251, doi:10.1029/2001JB000677.
K. Hirose and R. Sinmyo, Department of Earth and Planetary Sciences,
Tokyo Institute of Technology, 2-12-1 Ookayama, Tokyo 152-8551, Japan.
([email protected])
E. Okunishi, JEOL Ltd., 3-1-2 Musashino, Tokyo 196-8558, Japan.
H. St. C. O’Neill, Research School of Earth Sciences, Australian National
University, Building 61 Mills Road, Canberra ACT 0200, Australia.
3 of 3