JOURNAL OF PETROLOGY
VOLUME 53
NUMBER 9
PAGES 1943^1968
2012
doi:10.1093/petrology/egs039
Olivine Pseudomorphs after Serpentinized
Orthopyroxene Record Transient Oceanic
Lithospheric Mantle Dehydration
(Leka Ophiolite Complex, Norway)
OLIVER PLU«MPER1*, SANDRA PIAZOLO2y AND
H—KON AUSTRHEIM1
1
PHYSICS OF GEOLOGICAL PROCESSES (PGP), UNIVERSITY OF OSLO, PO BOX 1048, BLINDERN, N-0316, OSLO, NORWAY
2
DEPARTMENT OF GEOLOGICAL SCIENCES, STOCKHOLM UNIVERSITY, STOCKHOLM, SWEDEN
RECEIVED NOVEMBER 19, 2011; ACCEPTED MAY 18, 2012
We examine the partial survival of high-temperature mantle
microstructures throughout multi-stage hydration- and dehydrationmediated pseudomorphism at differing pressure^temperature^fluid
conditions. Throughout the harzburgitic mantle section of the Leka
Ophiolite Complex (Norway), finite domains of parallel olivine encompassed by mesh-textured olivine resembling ‘perfectly cleaved’
olivine grains were identified. Crystallographic orientation mapping,
combined with micro-computed tomography, reveals that the parallel
olivine grains are highly misoriented (up to 908) with no
crystal-preferred orientation, despite remaining parallel in three dimensions. Parallel olivine grains exhibit free dislocations with low
dislocation density, whereas within mesh-textured olivine dislocations are aligned into walls. MnO is enriched (up to 1·8 wt %)
and NiO depleted (0·21 0·24 wt %) within parallel olivine
grains compared with mesh-textured olivine (0·29 0·14 wt %
MnO; 0·38 0·19 wt % NiO). Clinopyroxene lamellae that are
crystal-plastically deformed occur sandwiched in lizardite layers between every parallel olivine grain or fully enclosed within olivine.
Al2O3 and Cr2O3 concentrations of clinopyroxene lamellae
(2·09 0·88 wt % Al2O3; 0·79 0·27 wt % Cr2O3) overlap
with those of primary clinopyroxene grains (2·43 0·69 wt %
Al2O3; 0·83 0·36 wt % Cr2O3) and are distinctly different
from those of secondary diopside found within the parallel olivine domains. Intragranular serpentine inclusions (XMg ¼ 0·95 0·01),
displaying elevated Al2O3 (3·92 4·10 wt %) and Cr2O3
(0·78 0·82 wt %) concentrations, are exclusively found within
parallel olivine grains. Lizardite (XMg ¼ 0·92 0·02) within the
*Corresponding author. E-mail: [email protected]
yPresent address: Australian Research Council Centre of Excellence for
Core to Crust Fluid Systems/GEMOC, Department of Earth and
Planetary Sciences, Macquarie University, N.S.W. 2109, Australia.
domains originates from hydration of parallel olivine and compositionally overlaps with mesh-texture lizardite. Antigorite
(XMg ¼ 0·95 0·01) replaces both types of olivine grains.
Whole-rock compositions indicate a harzburgitic composition; however, microstructural and chemical observations and the current absence of primary orthopyroxene suggest that the precursor silicate of
every parallel olivine domain was a single orthopyroxene grain that
was initially serpentinized and later dehydrated to result in the present microstructure. Although desilicification is necessary during the
transformation of orthopyroxene to olivine via a bastite stage, calculations based on whole-rock compositions imply that the released
SiO2(aq) was mobile only over micrometer to centimeter scales, reacting with the surrounding olivine directly to form serpentine.
Crosscutting relationships and serpentine compositions imply that
dehydration occurred prior to the now evident lizardite- and
antigorite-serpentinization. Comparison with the regional geological
setting indicates that dehydration may have occurred transiently
within the oceanic lithosphere prior to obduction.
microstructures; ophiolite; pseudomorphism; serpentinization; serpentine dehydration
KEY WORDS:
I N T RO D U C T I O N
Alteration and metamorphism of the Earth’s oceanic lithosphere is intimately coupled to fluid-dependent mineral
ß The Author 2012. Published by Oxford University Press. All
rights reserved. For Permissions, please e-mail: journals.permissions@
oup.com
JOURNAL OF PETROLOGY
VOLUME 53
replacement reactions. The most pronounced of these reactions that also have fundamental repercussions for the
oceanic lithosphere’s geochemical and geophysical properties are the hydration (serpentinization) of mantle peridotite and dehydration (deserpentinization) of serpentinites.
Typically, serpentinization advances along slow-spreading
ridges or bending faults during the onset of subduction,
influencing oceanic plate rheology and geochemical
inputs into subduction zones (e.g. Escartin et al., 2001;
Ru«pke et al., 2004). Upon fluid release from the subducting
slab, migration upwards induces serpentinization of the
forearc mantle wedge (e.g. Hyndman & Peacock, 2003).
Deserpentinization is thought to occur deep within the
subducting slab, expelling large amounts of fluid that act
as a flux for arc magmatism, and has been related to
intermediate-depth earthquakes (e.g. Hacker et al., 2003;
Padron-Navarta et al., 2010). Recent hydrothermal convection models, however, suggest that dehydration may also
occur in situ within the oceanic plate prior to subduction
and obduction (Iyer et al., 2010). Dehydration could also
be promoted when partly serpentinized peridotite is displaced along oceanic transform faults towards the hot adjacent ridge segment, potentially triggering seismicity (e.g.
Rutter & Brodie, 1987). No direct observations of deserpentinization at ridges or in forearc settings have been reported, possibly because meta-peridotite samples are
either dredged or are taken from shallow boreholes where
temperatures are insufficient to dehydrate serpentine minerals. This may be different at depth where upwelling of
hot asthenosphere or magma could supply the heat
required (e.g. Hyndman & Peacock, 2003; Allen &
Seyfried, 2004; Behn et al., 2007). Furthermore, breakdown
of hydrous minerals may occur during reactive fluid infiltration (e.g. blueschist eclogitization along a fluid conduit;
Beinlich et al., 2010) or if the fluid chemistry fluctuates
(e.g. changes in aH2O; e.g. Hardie, 1967).
Mantle sections of ophiolite complexes provide an extensive fossil record of the alteration processes occurring
within the oceanic lithosphere. However, deciphering
their potentially long alteration history is notoriously difficult. Unknown contributions of fluid flux from a subducting slab in supra-subduction zone ophiolites and possible
multiple overprints in response to metamorphic events
can hinder placing separate events into a geological context. The observation that serpentine is stable over a large
temperature and pressure range adds to the system’s complexity (e.g. Evans, 2004). Additional difficulties arise
from the common lack of geochronologically relevant minerals that prevents placing microstructurally distinct
events in an accurate time frame. The only exception is
the indirect dating of serpentinization via hydrothermal
zircons within rodingites (e.g. Dubinska et al., 2004).
Stable isotope and trace element geochemistry are mainly
used to constrain different fluid sources (e.g. Wenner &
NUMBER 9
SEPTEMBER 2012
Taylor, 1973; Deschamps et al., 2011), whereas the crosscutting relationships of vein networks and vein to wall-rock
patterns give insights into the sequence of different events
(e.g. Andreani et al., 2007). However, detailed microstructural analysis in terms of mineral reaction history is a promising tool to provide valuable clues about unknown
metamorphic and tectonic events occurring during the
‘life-cycle’ of the oceanic mantle.
Here we present an in-depth investigation of an extraordinary microstructure comprising olivine pseudomorphs
after serpentinized orthopyroxene grains from the mantle
section of the Leka Ophiolite Complex (Norway).The microstructure records the complete polymetamorphic evolution
of the mantle section, beginning with high-temperature, subsolidus and solid-state deformation as well as mineral
re-equilibrations during a hydration^dehydration^rehydration cycle. We particularly emphasize the survival of microstructures throughout multi-stage pseudomorphism during
succeeding metamorphic events and use the information
gained to unravel the complex mineral reaction scenarios.
GEOLOGIC A L S ET T I NG A N D
SAMPLES
The Leka Ophiolite Complex (LOC) crops out on the
island of Leka (90 km2), Nord-Trndelag, Norway, located
at 658N (Fig. 1). It is the most completely preserved
ophiolite present in the Scandinavian Caledonides and
represents part of the Upper or Uppermost Allochthon
of the Norwegian tectonostratigraphy with an age of
497 2 Ma (Dunning & Pedersen, 1988; Furnes et al.,
1988). Ophiolite formation took place in a suprasubduction setting of the North Iapetus Ocean and at
present exists in a pull-apart structure resulting from postorogenic transtension (Furnes et al., 1988, 1992; Titus et al.,
2002). Amphibolite-grade basement gneisses of the northern Vestranden region crop out south and west of Leka
and make up the dominant rock type of the adjacent mainland. There are no exposed contacts between the LOC
and the adjacent mainland rocks, and the lack of continuity between the two implies that all contacts are of tectonic
origin (Furnes et al., 1998; Titus et al., 2002). On the island
of Leka mantle meta-peridotite crops out in the northern
area of the island and is (1) in tectonic or transitional contact with ultramafic cumulates and (2) in tectonic contact
with meta-gabbro towards the east of the island (Fig. 1b).
Meta-dunite bodies of various sizes occur locally within
the meta-peridotite tectonite (Albrektsen et al., 1991). The
layered cumulates are dominated by dunite interlayered
with meta-wehrlite and locally chromite layers. The
meta-wehrlites crop out as several meter-long, lens-shaped
pods within the meta-dunites. A sharp discordant boundary separates the ultramafic cumulates from the metagabbros toward the SW part of the island. Ordovician
1944
PLU«MPER et al.
OLIVINE PSEUDOMORPHS, LEKA, NORWAY
Fig. 1. (a) Geographical location and geological overview map of the central Scandinavian Caledonides showing the tectonostratigraphic position of the Leka Ophiolite Complex (LOC) and surrounding geological units [modified from Gee et al. (1985) and Bucher-Nurminen (1991)].
BS, basement; PA, para-autochthonous; LA, lower allochthonous; MA, middle allochthonous; UA, upper allochthonous; UMA, uppermost allochthonous. (b) Detailed geological map of the LOC, showing sample localities (stars), and immediately surrounding geological units [modified from Prestvik (1972)]. True north is indicated. Sample localities cover a large part of the mantle tectonic section. In some localities more
than one sample was taken. (c) Field image of the LOC mantle meta-peridotite, displaying a local strong foliation. This foliation is not visible
throughout the entire complex. (d) Close-up of typical rough meta-peridotite weathering crosscut by millimeter- to centimeter-scale serpentine
veins.
metasedimentary rocks were deposited nonconformably on
the ophiolite complex and metamorphosed to greenschist
facies (Sturt et al., 1985). In this study we focus on samples
from 10 drill cores covering the mantle tectonite (metaperidotite) section (Fig. 1b). On the outcrop scale the
mantle meta-peridotite has a homogeneous appearance
with little to no variation in color (Fig. 1c); abundant millimeter- to centimeter-scale serpentine veins crosscut the
meta-peridotite. Meta-peridotite is easily distinguishable
from meta-dunite owing to its rougher weathering surface
(Fig. 1d).
A N A LY T I C A L T E C H N I Q U E S
Rock samples were sliced with a diamond saw and ground
in a steel mill. Major and minor element concentrations of
powdered rock samples were determined by inductively
coupled plasma atomic emission spectrometry (ICP-AES)
using a Perkin Elmer Optima 3300R system at Royal
Holloway, University of London. The analytical precision
is 1% for Si, Al, Fe, Mg and Ca, and 2% for Na, K, Ti, P
and Mn. Microstructures were examined in polished thin
sections (30 mm) using polarized light microscopy.
Back-scattered electron (BSE) images were acquired using
a JEOL 6610-LV scanning electron microscope
(SEM; Institut fu«r Mineralogie, University of Mu«nster,
Germany). Major quantitative element analyses were obtained with a Cameca SX 100 electron microprobe (EMP;
Department of Geosciences, University of Oslo, Norway).
All major element analyses were acquired by wavelengthdispersive X-ray (WDX) spectrometry using an accelerating voltage of 15 kV at a beam current of 10^20 nA. A
1945
JOURNAL OF PETROLOGY
VOLUME 53
selection of natural and synthetic minerals were used for
standardization; these were albite (Na), corundum (Al),
Cr2O3 (Cr), native Fe (Fe), NiO (Ni), MgO (Mg), orthoclase (K), pyrophanite (Mn and Ti) and wollastonite (Ca
and Si), respectively. Detection limits (3s uncertainty) in
olivine and clinopyroxene for Cr, Ni and Mn were 400^
650 ppm. Mineral analyses and subsequent calculated
values are reported with a 2s standard deviation. Matrix
corrections followed a routine implemented in the
Cameca PAP program (Pouchou & Pichoir, 1984).
Element distribution maps were obtained using operating
conditions of 15 kV and 40 nA with a step size of 1 mm.
For serpentine phase and brucite identification Raman
spectra were collected with a high-resolution Jobin Yvon
T6400 Raman spectrometer, which used the 532 nm line
of a 14 mW Nd:YAG laser as excitation source (Norwegian
Center for Raman Spectroscopy, University of Oslo,
Norway). The scattered Raman light was collected in
1808 backscattering geometry and dispersed by a 1800
grooves mm1 grating after travelling through a 100 mm
entrance slit. The spectrometer was calibrated using the
first-order Raman band of silica at 520·7 cm1 and the
epoxy resin was analyzed to check for contributions to the
mineral analyses. Raman peak positions were fitted using
the FITYK software (Wojdyr, 2010). Acquired serpentine
and brucite spectra were compared with published reference spectra (Dawson et al., 1973; Rinaudo et al., 2003;
Groppo et al., 2006). In addition, all spectra were compared with the RRUFF Project database (Downs, 2006).
Quantitative crystallographic orientation data for olivine, clinopyroxene and magnetite were acquired via automatically indexed electron backscatter diffraction (EBSD)
patterns collected using a Philips XL-30 field-emission
(FE-)SEM equipped with an Oxford Nordlys II EBSD detector (Department of Geological Sciences, Stockholm
University, Sweden). Thin sections for EBSD analysis were
SYTONÕ-polished and not coated with carbon. Working
conditions during acquisition of EBSD patterns were
25 kV accelerating voltage, 0·8 nA beam current,
20·4 mm working distance, 708 sample tilt and low-vacuum
mode (0·7 mbar). EBSD patterns were acquired on rectangular grids by moving the electron beam at a regular
step size of 0·7^1 mm. EBSD pattern indexing was executed
using the HKL CHANNEL5 software. This software was
also used for EBSD data processing, by applying a noise reduction filter that replaces non-indexed solutions by the
most common neighbor orientation (Prior et al., 2002;
Bestmann & Prior, 2003). A modified Kuwahara filter was
applied once (3 3 filter with 58 smoothing and 18 artifact
angle).
Oxidation decoration was used to gain an insight
into the olivine dislocation density and arrangement
(Kohlstedt et al., 1976). This technique uses oxidation of Fe
present in the olivine and subsequent precipitation of iron
NUMBER 9
SEPTEMBER 2012
oxides at lattice distortions to identify the dislocation substructures on a microscopic scale. Single dislocations are
visible as points and dislocation walls (i.e. subgrain boundaries) as lines. For this technique a polished rock slab was
heated at 9008C for 45 min in air. Subsequently, a 30 mm
thin section was prepared from the polished side and
examined using an SEM in BSE mode (Karato, 1987).
For micro-computed X-ray tomography (mCT) a cylinder (1cm 0·7 cm) was cut from a rock sample that
showed the required microstructure in thin section. mCT
was executed using a XRADIA MicroXCT-400 at an acceleration voltage of 100 kV and a tube current of 100 mA
(Australian Centre for Microscopy & Microanalysis,
University of Sydney, Australia). For the measurement,
1200 projections (40 s integration time per projection)
were acquired during a 1808 rotation of the sample, resulting in a spatial resolution of 2·5 mm per voxel.
Three-dimensional (3-D) visualization was carried out
using the Visualization Sciences Group (VSG) Avizo
FireÕ program. A Hessian-based multi-scale enhancement
filter (Frangi et al., 1998; computed in MATLAB) was
used to extract the serpentine vein locations from the
mCTdata.
Mineral abbreviations follow those of Whitney & Evans
(2010).
G E O C H E M I C A L C H A R AC T E R O F
TH E LEK A MA NTLE SECTION
A N D T H E P RO T O L I T H S O F
THE LEKA SERPENTINITES
Analyses of whole-rock compositions and previously published whole-rock data (Maale, 2005; Iyer et al., 2008a)
clearly identify the meta-peridotites of the mantle tectonite
section as harzburgite with an average normative primary
mineralogy of 77 8% olivine, 22 8% orthopyroxene
and 1 1% clinopyroxene (Fig. 2a). Whole-rock compositions of dunites that are identifiable in the field (e.g.
Albrektsen et al., 1991) are also plotted in Fig. 2a for comparison. Average whole-rock compositions of harzburgite
and dunite are given in Table 1. Mg/Si and Al/Si
whole-rock element ratios, which are tracers for the protolith of the serpentinite (Hattori & Guillot, 2007), are close
to the element ratios of serpentinites from the Mariana
Trench and the Greater Antilles (Fig. 2b) consistent with
previous geochemical investigations, which indicate an
evolution within a supra-subduction zone setting (Furnes
et al., 1992). This is also supported by the composition of
primary Cr-spinel grains (Fig. 2c). Based on whole-rock
compositions and the extensive alteration we use the
name meta-peridotite when referring to rocks with a harzburgitic geochemistry.
1946
PLU«MPER et al.
OLIVINE PSEUDOMORPHS, LEKA, NORWAY
Ol
DPO-bearing
meta-peridotite
M I C RO S T RU C T U R E S A N D
M I N E R A L C H E M I S T RY O F T H E
L O C M E TA- P E R I D O T I T E
(a)
Dunite
Ol90
meta-dunite
On the basis of microstructural and microchemical analysis of the samples, three domains can be identified
throughout the LOC meta-peridotite, which are schematically illustrated in Fig. 3.
Domain I is made up of elongated, parallel olivine
grains (domains of parallel olivine: DPO) with a characteristic spacing, which are separated by thin clinopyroxene
lamellae that are generally enclosed in lizardite. Diopside
aggregates and elongated magnetite grains occur interstitially. Antigorite locally overprints the DPOs.
Domain II comprises the matrix and is composed of
mesh-textured olivine grains that are partially hydrated to
form lizardite with interstitial magnetite. Isolated antigorite replaces mesh-textured olivine relicts to varying degrees. In addition, large augitic clinopyroxene and
primary spinel grains occur.
Domain III is represented by continuous areas of massive interpenetrating antigorite blades extending for several millimeters across thin sections, obliterating all
previous microstructures.
Detailed chemical analysis of olivine, clinopyroxene, the
serpentine varieties and oxides are given in Tables 2 and 3.
Maaløe (2005)
av. peridotite
av. dunite
Ha
e
rlit
rzb
ur g
h
We
ite
Lherzolite
Ol40
b
(b)
Mariana
Forearc
Himalayas
1.4
res
MARK
idu
e
me
lt
1.2
Abyssal
peridotite
0.00
1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
1.0
Zaza, Cuba
& Puerto Rico
0.02
0.04 0.06 0.08
Al/Si (wt%)
Primitive
mantle
0.10
Domain I: detailed characterization
In total 30 DPOs in 20 thin sections were investigated in
detail. All display consistent microstructural and microchemical characteristics.
0.12
(c)
Microstructures
Within a single thin section DPOs make up 7^10% relative to mesh-textured olivine, excluding areas of massive
antigorite-serpentinization (Domain III). Isolated DPOs,
measuring 1^5 mm in diameter, are surrounded by
Domain II textures, producing an appearance similar to a
porphyroblast (Fig. 4a and b). Olivine grains in the DPOs
have complex undulose extinction patterns where single
Fo
rea
rc
1.0
Cr#
Cpx
Ab
ys
sa
l
Mg/Si (wt%)
Opx
1.6
0.8
0.6
XMg
0.4
0.2
0.0
Fig. 2. Geochemical characteristics of the LOC meta-peridotites and
meta-dunites. (a) Streckeisen triangular plot for the classification of
ultramafic rocks, showing the primary modal composition of LOC
meta-peridotites and meta-dunites calculated on a water-free basis.
DPO-bearing rocks are located within the harzburgite field, exhibiting a natural spread in composition with no major trend towards a
dunitic composition. Calculated mineral abundances from the
whole-rock geochemistry of the meta-dunites from Leka clearly show
the difference between the meta-dunites and the DPO-bearing rocks.
Whole-rock compositions originate from this study and from Iyer
et al. (2008) as well as average (av.) compositions from Maale (2005)
for comparison. (b) Weight ratios of Mg/Si vs Al/Si in peridotites of
various origins in comparison with meta-harzburgites and
meta-dunites from the Leka ophiolite [modified from Hattori &
Guillot (2007) and Saumur et al. (2010)]. Meta-harzburgites and
meta-dunites from Leka plot within the field of forearc mantle peridotites and are similar in composition to forearc serpentinites along the
Zaza fault in Cuba (Hattori & Guillot, 2007) and Puerto Rico
Trench serpentinites (Bowin et al., 1966). (c) Comparison of Cr# vs
XMg of Cr-spinel grains from the LOC upper mantle section. Data
are taken from Furnes et al. (1992) and Iyer et al. (2008a). Each point
represents the composition of the core of one grain. The field of abyssal peridotite includes abyssal spinel peridotite and plagioclase peridotite (Dick & Bullen, 1984). The forearc mantle peridotite field is
defined by Cr-spinel data from serpentinized peridotites exhumed in
the Mariana forearc (Ishii et al., 1992). All data points, except for
one, plot in the field of forearc peridotite.
1947
JOURNAL OF PETROLOGY
VOLUME 53
NUMBER 9
SEPTEMBER 2012
Table 1: Average whole-rock geochemical compositions of
investigated meta-peridotites and meta-dunites
Rock type:
Meta-peridotite
Meta-dunite
n:
16
9
wt %
SiO2
39·06 (1·09)
35·56 (1·62)
TiO2
0·00 (0·01)
0·00 (0·01)
Al2O3
0·61 (0·40)
0·20 (0·30)
Fe2O3
7·12 (1·64)
9·28 (1·35)
FeO(rec)
6·41 (1·47)
8·35 (1·21)
MgO
46·28 (1·66)
48·40 (2·44)
MnO
0·12 (0·04)
0·16 (0·03)
CaO
0·42 (0·65)
0·30 (0·82)
Na2O
0·01 (0·01)
0·01 (0·02)
K2O
0·01 (0·03)
0·01 (0·02)
P2O5
0·00 (0·01)
0·00 (0·01)
LOI
10·38 (1·33.)
9·96 (3·25)
Total
98·43 (0·61)
98·06 (1·63)
Fig. 3. Simplified thin-section sketch showing the main millimeterscale spatial mineralogical associations in the investigated samples.
Samples can be divided into three domains depending on their mineralogy and microstructures; Domain I consists solely of DPOs occurring as individual areas throughout the thin section and minor
overprinting antigorite (Atg), Domain II comprises olivine (Ol)
grains that are partially replaced by lizardite to form a mesh texture,
antigorite patches replacing mesh-textured olivine, single large clinopyroxene (Cpx) grains and primary spinel (Spl) grains. Domain III
encompasses massive antigorite occurrences that overprint all previous microstructures.
Recalculated on water-free basis
SiO2
44·37 (1·54)
40·37 (0·84)
TiO2
0·00 (0·01)
0·00 (0·01)
Al2O3
0·69 (0·46)
0·22 (0·35)
Fe2O3
8·09 (1·78)
10·53 (1·49)
FeO(rec)
7·28 (1·60)
9·48 (1·34)
MgO
46·28 (1·66)
48·40 (2·44)
MnO
0·12 (0·04)
0·16 (0·03)
CaO
0·42(0·65)
0·30 (0·82)
Na2O
0·01 (0·03)
0·01 (0·02)
K2O
0·01 (0·03)
0·01 (0·02)
P2O5
0·00 (0·01)
Total
100
0·00 (0·01)
100
Mg/Si (wt %)
1·044 (0·059)
1·199 (0·074)
Al/Si (wt %)
0·016 (0·010)
0·006 (0·009)
Values are average and 2s, including loss on ignition (LOI)
and recalculated on a water-free basis. Data from this
study and Iyer et al. (2008a). n, number of analyses; rec,
recalculated.
areas do not extinguish concurrently (Fig. 4c). Figure 4c illustrates the parallel alignment of the elongated olivine
grains, which have a characteristic periodic spacing of
16 4 mm (n ¼100). Orientation data expressed as a misorientation map of a representative central DPO area
(Fig. 5a) display no correlation between crystallographic
orientation and the parallelism of the olivine grains
(Fig. 5b). Instead, neighboring parallel olivine grains are
highly misoriented with respect to one another, with misorientation angles of up to 908 to a fixed reference point
(star in Fig. 5b). Furthermore, high misorientations are
observed either (1) in extension of an elongated olivine
grain (arrow 1 in Fig. 5b) or (2) as islands included in
larger elongated grains with the same orientation (arrow
2 in Fig. 5b). On the scale of a single olivine grain internal
distortions are minimal (0^88; Fig. 5c), and only free dislocations, without the occurrence of dislocation walls, can
be observed (Fig. 6a). However, small misorientations of
1^28 arise where fractures cut across single grains (Fig.
5c). In addition, examination of dislocation substructures
indicates lower dislocation densities of DPO olivine relative to mesh-textured olivine grains (Fig. 6). m-CT of a
centimeter-sized rock cylinder reveals that the parallelism
of the olivine grains, as well as the grains themselves, remains continuous in three dimensions over several hundred micrometers. As serpentine veins separate nearly
every elongated olivine grain, we have used these veins to
highlight the 3-D parallelism in Fig. 7. An animation of a
projection series and the 3-D visualization of Fig. 7 is available online (Movie 1, online supplementary material,
available from http://www.petrology.oxfordjournals.org).
Compositionally and spatially different lizardite and
chrysotile have also been identified in addition to the
minor antigorite rosettes and needles that overprint the
DPO microstructures (Fig. 8a and c), using Raman spectroscopy (Fig. 9). Lizardite is found homogeneously distributed between the elongated olivine grains (Fig. 8c and d)
or makes up transgranular serpentine veins crosscutting
the DPOs (Fig. 8b and c). In contrast, chrysotile occurs as
1948
PLU«MPER et al.
OLIVINE PSEUDOMORPHS, LEKA, NORWAY
Table 2: Mineral compositions mesh-textured olivine, DPO olivine (center and halo) and clinopyroxene varieties
Mineral:
Olivine
Clinopyroxene
Mesh
DPO center
DPO halo
Phenocrysts
Lamellae
Diopside aggregates
n:
109
130
58
18
24
53
52·92 (0·89)
52·32 (2·42)
wt %
SiO2
40·75 (0·73)
40·95 (0·96)
39·69 (0·96)
54·99 (1·86)
TiO2
b.d.l.
b.d.l.
b.d.l.
b.d.l.
b.d.l.
b.d.l.
Al2O3
0·01 (0·06)
0·01 (0·04)
0·02 (0·16)
2·43 (0·69)
2·09 (0·88)
0·10 (0·16)
Cr2O3
0·02 (0·05)
0·07 (0·11)
0·06 (0·08)
0·83 (0·36)
0·79 (0·27)
0·14 (0·16)
FeO(tot)
9·71 (1·18)
9·71 (1·19)
14·14 (1·32)
1·88 (0·26)
1·97 (0·46)
1·08 (0·59)
NiO
0·38 (0·19)
0·21 (0·25)
0·07 (0·20)
0·03 (0·06)
0·05 (0·06)
0·02 (0·06)
MgO
49·10 (1·14)
49·02 (1·42)
44·88 (2·43)
16·97 (0·56)
18·25 (2·86)
18·34 (2·17)
MnO
0·29 (0·14)
0·35 (0·21)
1·30 (0·57)
0·07 (0·06)
0·07 (0·05)
0·03 (0·07)
CaO
b.d.l.
0·09 (0·81)
0·03 (0·04)
24·32 (0·53)
23·95 (1·95)
25·88 (2·09)
Na2O
Total
b.d.l.
100·26 (1·46)
b.d.l.
100·41 (1·29)
b.d.l.
100·19 (1·72)
0·11 (0·09)
0·09 (0·07)
0·05 (0·07)
99·56 (1·19)
99·58 (2·83)
100·63 (1·96)
Structural formula based on 4 oxygen for olivine and 6 oxygen for clinopyroxene
Si
0·996 (0·011)
1·000 (0·021)
0·995 (0·012)
1·930 (0·025)
1·899 (0·055)
1·976 (0·055)
Ti
b.d.l.
b.d.l.
b.d.l.
b.d.l.
b.d.l.
b.d.l.
Al
0·000 (0·001)
0·000 (0·001)
0·001 (0·004)
0·105 (0·030)
0·090 (0·039)
0·004 (0·007)
Cr
0·000 (0·000)
0·001 (0·002)
0·001 (0·002)
0·024 (0·010)
0·023 (0·008)
0·004 (0·005)
Fe
0·199 (0·025)
0·198 (0·025)
0·296 (0·032)
0·057 (0·008)
0·060 (0·014)
0·033 (0·018)
Ni
0·008 (0·002)
0·004 (0·005)
0·001 (0·004)
0·001 (0·002)
0·002 (0·002)
0·001 (0·002)
Mg
1·790 (0·023)
1·785 (0·040)
1·677 (0·065)
0·923 (0·030)
0·987 (0·146)
0·982 (0·115)
Mn
0·006 (0·001)
0·007 (0·004)
0·028 (0·012)
0·002 (0·002)
0·002 (0·002)
0·001 (0·002)
Ca
b.d.l.
0·002 (0·022)
0·001 (0·002)
0·950 (0·017)
0·932 (0·077)
0·996 (0·078)
Na
b.d.l.
b.d.l.
b.d.l.
0·008 (0·006)
0·006 (0·005)
0·003 (0·005)
XMg
0·90 (0·01)
0·90 (0·01)
0·85 (0·01)
0·94 (0·01)
0·94 (0·01)
0·97 (0·01)
Values are average and 2s. n, number of analyses; b.d.l., below detection limit; tot, total; XMg ¼ Mg/(Mg þ Fe).
micrometer-sized patches within the aforementioned lizardite or as inclusions, fully enclosed within the elongated
olivine grains (Fig. 10).
Thin clinopyroxene lamellae, measuring1^3 mm in thickness and sandwiched between lizardite, were observed in
all investigated DPOs (Fig. 8d and e). However, clinopyroxene lamellae are also commonly found completely
enclosed within olivine grains without microscopically
visible lizardite at the clinopyroxene^olivine contacts
(Fig. 8f). All the lamellae have a discontinuous lens-shape
and vary in length from 10 to 500 mm. The crystallographic
plane of the clinopyroxene lamellae that is parallel to
the elongated olivine^serpentine interphase boundary
was found to be consistently (100) using EBSD analysis
(Fig. 8e). Crystallographic analysis of a representative
clinopyroxene lamella reveals a significant orientation
spread in both [100] and [001] axes following a circle
around the [010] axis, whereas the latter displays only
minor dispersion (Fig. 11). Some DPOs exhibit an apparent
curvature where single, continuous clinopyroxene lamellae
are bent (Fig. 8a). Associated olivine grains follow this
bending but in a discontinuous manner and are separated
by fractures. In contrast to the clinopyroxene lamellae,
mosaic-like, randomly oriented aggregates of diopside
grains (Fig. 8c and d) were observed intergrown with elongated olivine grains (Fig. 8e). The occurrence of these diopside grains varies in quantity from a sporadic
appearance (Fig. 8c and d) to massive replacement of the
DPOs (Fig. 12). Concomitantly, the abundance of magnetite increases with increasing diopside occurrence (compare
Figs 8b and 12).
Magnetite grains display an elongated habit following
the parallelism within the DPOs, both in two (Figs 4d
and 8) and three dimensions (Fig. 7, Supplementary Data
1949
JOURNAL OF PETROLOGY
VOLUME 53
NUMBER 9
SEPTEMBER 2012
Table 3: Mineral compositions of serpentine varieties and Fe-oxides
Mineral:
n:
Serpentine
Fe-oxides
Lizardite (mesh)
Lizardite (DPO)
Chrysotile inclusions
Antigorite
Cr-magnetite
Magnetite
85
54
71
87
30
21
36·45 (3·89)
36·26 (4·18)
37·85 (2·09)
43·65 (2·77)
wt %
SiO2
TiO2
b.d.l.
b.d.l.
b.d.l.
b.d.l.
0·11 (0·20)
0·31 (0·39)
b.d.l.
b.d.l.
0·03 (0·04)
Al2O3
0·01 (0·08)
0·04 (0·28)
3·92 (4·10)
0·74 (0·65)
0·10 (0·09)
Cr2O3
0·03 (0·07)
0·05 (0·08)
0·78 (0·82)
0·16 (0·33)
21·36 (2·31)
0·65 (0·20)
FeO(tot)
6·12 (2·40)
6·35 (2·27)
3·40 (1·34)
2·78 (1·26)
45·66 (2·81)
89·17 (1·20)
Fe2O3(rec)
n.c.
n.c.
n.c.
n.c.
68·72 (3·61)
67·08 (0·67)
FeO (rec)
n.c.
n.c.
n.c.
n.c.
27·58 (1·32)
28·74 (0·84)
NiO
0·17 (0·31)
0·26 (0·26)
0·78 (0·82)
0·14 (0·28)
0·18 (0·12)
0·58 (0·15)
MgO
41·75 (3·69)
41·53 (2·86)
38·92 (2·80)
39·43 (1·22)
1·11 (0·43)
0·66 (0·38)
MnO
0·20 (0·14)
0·22 (0·14)
0·08 (0·10)
0·93 (0·54)
0·07 (0·04)
b.d.l.
CaO
0·13 (0·17)
0·43 (1·62)
b.d.l.
b.d.l.
b.d.l.
b.d.l.
Na2O
b.d.l.
b.d.l.
b.d.l.
b.d.l.
b.d.l.
b.d.l.
Total
84·84 (7·68)
85·14 (2·61)
85·73 (3·93)
86·90 (2·03)
97·03 (1·47)
98·12 (0·87)
Structural formula based on 7 oxygen for lizardite, chrysotile and antigorite and 4 oxygen for magnetite
Si
1·747 (0·139)
1·737 (0·199)
1·817 (0·1209)
2·064 (0·135)
0·004 (0·008)
0·012 (0·015)
Ti
b.d.l.
b.d.l.
b.d.l.
b.d.l.
b.d.l.
b.d.l.
Al
0·001 (0·005)
0·002 (0·016)
0·221 (0·229)
0·041 (0·036)
0·005 (0·004)
0·002 (0·002)
Cr
0·001 (0·003)
0·002 (0·003)
0·029 (0·031)
0·006 (0·012)
0·655 (0·072)
0·020 (0·006)
Fe3þ
n.c.
n.c.
n.c.
n.c.
1·333 (0·076)
1·964 (0·014)
Fe2þ
0·245 (0·081)
0·254 (0·091)
0·136 (0·049)
0·110 (0·053)
0·894 (0·037)
0·934 (0·032)
Ni
0·007 (0·012)
0·006 (0·013)
0·004 (0·008)
0·004 (0·011)
0·006 (0·004)
0·018 (0·004)
Mg
2·983 (0·095)
2·966 (0·170)
2·784 (0·165)
2·780 (0·133)
0·064 (0·025)
0·038 (0·022)
Mn
0·008 (0·005)
0·009 (0·006)
0·003 (0·004)
b.d.l.
0·030 (0·018)
0·002 (0·001)
Ca
0·007 (0·009)
0·022 (0·083)
b.d.l.
b.d.l.
b.d.l.
b.d.l.
Na
b.d.l.
b.d.l.
b.d.l.
b.d.l.
b.d.l.
b.d.l.
XMg
0·92 (0·02)
0·92 (0·02)
0·954 (0·014)
0·038 (0·016)
n.c.
n.c.
Values are average and 2s. n, number of analyses; b.d.l., below detection limit; tot, total; rec, recalculated; n.c.,
not calculated; XMg ¼ Mg/(Mg þ Fe).
Movie 1), and are closely associated with the clinopyroxene
lamellae (Fig. 8). Rarely, spherical magnetite is observed.
Comparison of DPO and magnetite grain orientations in
Fig. 13 for the area shown in Fig. 5b reveals that, although
olivine orientations vary, magnetite grains consistently exhibit two specific orientations across the entire area.
Mineral chemistry
Olivine grains within DPO centers have an average XMg of
0·900 0·012 and show a strong correlation with MnO content, with decreasing XMg as MnO concentrations increase
(Fig. 14a). In contrast, NiO concentration decreases sharply
from 0·5 to 0 wt % (average NiO ¼ 0·21 0·24 wt %)
within a relatively small XMg range (Fig. 14b).
Surrounding these centers Mn þ Fe-rich, Ni-poor haloes of
olivine grains (XMg ¼ 0·850 0·026) that have increased
electron backscattering intensity develop (Fig. 8a and b).
Within these olivine grain haloes MnO remains negatively
correlated with XMg, whereas NiO is independent of XMg
(Fig. 14a and b).
The serpentine varieties show a distinct difference in Al
and Cr concentrations (Fig. 14c and d). Lizardite grains
(XMg ¼ 0·92 0·02) and antigorite (XMg ¼ 0·96 0·02)
are low in Al2O3 as well as Cr2O3. However, chrysotile
grains enclosed in the DPO olivine (XMg ¼ 0·95 0·02)
display elevated Al2O3 and Cr2O3 concentrations.
1950
PLU«MPER et al.
OLIVINE PSEUDOMORPHS, LEKA, NORWAY
Fig. 4. (a, b) Optical photomicrographs under crossed polars (variable polarizer and analyser positions) giving an overview of two DPOs
(within dashed borderline). (c) Close-up of the DPO in (b) highlighting the parallelism of the olivine (Ol) grains with varying second-order
interference colors and undulose extinction. (d) Photomicrograph in plane-polarized light showing magnetite (Mag) grains following the parallel north^south elongation trend of the olivine grains. (e) Photomicrograph under crossed polars of olivine grains within the vicinity of a
DPO (upper left side within dashed borderline) that are partially replaced by lizardite (Lz) to form a mesh texture. Because of the retention
of crystallographic continuity during serpentinization the approximate grain size of primary mantle olivine grains, measuring 6^7 mm, is
still visible. (f) Magnified photomicrograph under crossed polars of an area in (e), highlighting the relationships between lizardite and antigorite
(Atg), where the latter overprints the former.
1951
JOURNAL OF PETROLOGY
VOLUME 53
NUMBER 9
SEPTEMBER 2012
Fig. 5. (a) BSE image displaying a representative central area of a DPO and (b) corresponding orientation map, acquired using the EBSD
technique. Angular misorientation threshold is 08 to 908 with respect to the reference point (star). White lines display theoretical grain boundaries. Even though olivine (Ol) grains are parallel, misorientations between areas are high and no specific crystallographic plane can be associated with the ‘pseudo-cleavage’ appearance. High misorientations are also observed either in extension of an elongated olivine grain
(arrow 1) or as islands included into larger elongated olivine grains with the same orientation (arrow 2). In contrast, no internal distortions
within grains are evident, as illustrated by the orientation map in (c) with an angular misorientation threshold of 08 to 88. The map in (c) is
taken of the large blue area within the DPO shown in (a). (d) Orientation map [same angular threshold as in (c)] of partly serpentinized
mesh-textured olivine grains displaying internal distortions that are attributed to plastic deformation-induced dislocation walls. (e) Extracted
pole figures from (d) across dislocation walls (equal area projection, upper hemisphere), highlighting the crystal-plastic deformation-induced
origin of the dislocation walls with dispersion or misorientation angles of 48. (Note the rotation axis where data points do not show significant
dispersions, whereas data points of the other corresponding axes describe a progressive dispersion following a circle around the rotation axis.)
Blue phases in (b), (c) and (d) are magnetite (Mag) grains. The reference frame of the pole figure is chosen to coincide with the x (horizontal)
and y (vertical) axes of the image in (d). Brc, brucite; Cpx, clinopyroxene; Lz, lizardite; lam., lamella.
Comparison of SiO2 wt % with XMg of lizardite grains
also implies the cryptic presence of brucite.
All clinopyroxene lamellae display elevated Al2O3
(2·09 0·90 wt %) and Cr2O3 (0·79 0·30 wt %) contents, which is highlighted in the element distribution
maps in Fig. 10. Mosaic-like diopside aggregates are nearly
pure diopside, with only minor concentrations of Al2O3
(0·10 0·16 wt %) and Cr2O3 (0·14 0·16 wt %). A comparison between the clinopyroxene varieties can be found
in Fig. 14e and f. Associated magnetite grains have Cr2O3
concentrations of up to 25 wt %; however, small magnetite
grains (45 mm) in this area are commonly low in Cr.
Domain II: detailed characterization
Microstructures
Lizardite brucite (verified using Raman spectroscopy)
replaces olivine grains, measuring 5^7 mm diameter,
forming a typical mesh texture in two (Fig. 4e and f) and
three dimensions (Fig. 7). Lizardite-serpentinization varies
locally on the centimeter scale from410 to 50%. In addition, olivine grains are penetrated by crosscutting antigorite blades, further separating the olivine grains into
smaller grain domains. In both textures, currently separated grains can be assigned to an originally single grain
owing to their retention of optical continuity (Fig. 4e).
Single olivine grain relicts commonly display intragranular kink bands and undulose optical extinction.
Throughout Domain II several inclusion trails of serpentine and locally chlorite, with a trail width of 5^20 mm,
are found to transect olivine domains (Fig. 15).
Dislocation microstructure analysis of mesh-textured
olivine grains reveals organized dislocation walls, with
areas of high free dislocation densities between them
(Fig. 6b). The dislocation wall spacing is 5 mm on average,
1952
PLU«MPER et al.
OLIVINE PSEUDOMORPHS, LEKA, NORWAY
Fig. 6. BSE images showing the dislocation microstructures of
(a) olivine within a DPO and (b) mesh-textured olivine after oxygen
decoration. The dislocations are indicated by dots and lines with elevated backscattering intensity compared with the surrounding olivine.
Only free dislocations, such as those marked with arrows, with a low
density are visible in (a). The parallel lines in (a) that are aligned
bottom left to top right are part of the pseudo-cleavage commonly exhibiting clinopyroxene lamellae (Cpx lam.; compare with Fig. 5) and
are not subgrain boundaries formed by crystal-plastic deformation.
In contrast, dislocations are aligned into walls (subgrain boundaries,
arrows) in (b) with a high density of free dislocations between the
walls. EBSD analysis across these subgrain boundaries is shown in
Fig. 5e. The brightest phase in both images is magnetite and the darkest phase is serpentine.
but can vary between 2 and 10 mm. Straight dislocations
are also observed at the rims of the mesh-textured olivine
grains that are nearly perpendicular to the grain margins.
Crystallographic orientation mapping shows that areas between dislocation walls have intragranular misorientations
of several degrees (Fig. 5d). Pole figure representation of
crystal orientations across dislocation walls shows systematic dispersions of the two main olivine crystallographic
axes following a half-circle around the third main axis,
which itself does not display any significant dispersion
(Fig. 5e).
Magnetite grains in Domain II are dispersed and follow the mesh texture in two (Fig. 8a, lower left) and
Fig. 7. Three-dimensional visualization of an interface between a
single DPO and the surrounding mesh-textured olivine (Ol). Only
serpentine veins (Srp; translucent brown) and magnetite (Mag;
blue) have been depicted in the 3D image to aid visualization of the
parallel structures; however, in the structure the gaps between these
minerals are filled with olivine. Magnetite grains within the DPO
are aligned into planes following the parallel serpentine veins within
the DPO, whereas magnetite grains within the mesh-textured olivine
follow the crosscutting serpentine veins. (Note the abrupt microstructural change across the DPO^mesh interface. Serpentine veins
within the DPO are planar across several hundred micrometers and
are almost never crosscut by other serpentine veins. The cuboid
shown has a volume of 1580 mm 1190 mm 265 mm. An animated
version for better DPO parallelism visualization is available online at
http://www.petrology.oxfordjournals.org.
three dimensions (Fig. 7; Supplementary Data Movie 1).
Comparison of crystallographic orientations between magnetite grains and associated mesh-textured olivine shows
that single magnetite grain orientations display a topotactic relationship in two different orientations, which is dependent on the host olivine orientation (Fig. 13b). Single
clinopyroxene grains (Fig. 8a and b), measuring 0·6^
1mm diameter, occur within Domain II and are generally
located in the vicinity of DPOs (Domain I). All large
clinopyroxene grains are partially pseudomorphically serpentinized along fractures and cleavage planes and are occasionally replaced by pure diopside.
Mineral chemistry
Olivine grain relicts within Domain II are compositionally
homogeneous with average XMg, NiO and MnO concentration of 0·900 0·012, 0·38 0·20 wt% and 0·29 0·14 wt%, respectively. The large clinopyroxene grains in
this area are Al- and Cr-bearing with a similar composition to the DPO clinopyroxene lamellae (Fig. 14e and f).
Dispersed magnetite grains are typically low in Cr, although grains with elevated Cr content can also be found.
Cores of hypidiomorphic Cr-spinel grains display
1953
JOURNAL OF PETROLOGY
VOLUME 53
NUMBER 9
SEPTEMBER 2012
Fig. 8. (a, b) BSE images of DPOs encompassed by mesh-textured olivine (Ol) grains that are partly serpentinized and large, partly altered,
primary clinopyroxene (Cpx) grains. A universal feature of the DPOs is the occurrence of an Mn-rich olivine halo encompassing the central
parallel area (within dashed borderlines). The apparent bending of the entire DPO in (a) should be noted (bending is traced with dashed
lines). (c^e) BSE images of the central DPO areas displaying the occurrence of diopside aggregates (Di agg.), clinopyroxene lamellae (lam.) between every single parallel olivine ‘pseudo-cleavage’ [arrows without an adjoining label in (d]), and lizardite brucite (Lz Brc) either
within the ‘pseudo-cleavage’ or as crosscutting fractures. Clinopyroxene lamellae consistently have their (100) plane parallel to one another as
shown in (e). Magnetite (Mag) grains frequently follow the olivine grain parallelism and are closely associated with the clinopyroxene lamellae.
Diopside aggregates occur primarily within the olivine or at olivine^lizardite grain boundaries. (f) High-resolution BSE image of the central
area in (d), illustrating the frequent appearance of fully enclosed clinopyroxene lamellae within olivine. Atg, antigorite.
1954
PLU«MPER et al.
230
376
683
measured
reference
Intensity (arbitary unit)
462 527
OLIVINE PSEUDOMORPHS, LEKA, NORWAY
1045
(a) Atg
383
Cpx lam.
687
228
527
350
(b) Lz
692
388
232
Ol (DPO)
347
527
1102
622
(c) Ctl
200
400
600
800
1000
1200
Raman shift (cm-1)
Fig. 9. Representative Raman spectra of the serpentine varieties (a)
antigorite (Atg), (b) lizardite (Lz) and (c) chrysotile (Ctl) occurring
in association with the DPOs. Antigorite occurs as massive overgrowth. Lizardite is found either as the filling product of meshtextured olivine grains or between every parallel olivine grain within
the DPO. Chrysotile is present as inclusions within the parallel olivine
grains. The mode at 1011cm1 (Si^O stretching vibrations) in the
lizardite spectrum (b) originates from the closely associated clinopyroxene lamellae. The phases were identified by comparison with reference spectra (gray) from the RRUFF Project (Downs, 2006) and
literature spectra (e.g. Rinaudo et al., 2003; Groppo et al., 2006). The
serpentine varieties are distinguishable by the difference in Si^Ob^Si
vibrations at about 683 cm1 for antigorite and 690 2 cm1 for lizardite and chrysotile as well as the distinct antisymmetric Si^Ob^Si
stretching modes at 1045 cm1 for antigorite. Lizardite and chrysotile
are distinguishable by the SiO4 tetrahedra bending, where the characteristic Raman shift for lizardite (here 383 cm1) is at lower wavenumbers than for chrysotile (here 388 cm1). In addition, chrysotile
exhibits an antisymmetric Si^Onb stretching mode at 1102 cm1.
negatively correlated Cr2O3 (40^55 wt %) and Al2O3
(8^30 wt %) concentrations (Cr# ¼ 0·5^0·8). MnO contents of Cr-spinel grains range from 0·2 to 1·0 wt %.
However, most Cr-spinel grains are altered to ferritchromite and display chlorite alteration corona.
DISCUSSION
Microstructure origin
Perfectly cleaved olivine, magmatic replacement or
fluid-mediated, multi-stage pseudomorphism?
At first sight the DPO microstructures resemble ‘perfectly
cleaved’ olivine grains that are thought to be the result of
a deformation-induced, crystallographically controlled
parting, which typically occurs along the (100) and (010)
planes with no significant misorientation between olivine
segments (Hawkes, 1946; Kuroda & Shimoda, 1967; Kutty
et al., 1983; Aikawa & Tokonami, 1987; Ohara & Ishii,
1998; Murata et al., 2009; Nozaka & Ito, 2011). However,
there are several microstructural characteristics of the
DPOs that cannot be explained by this simple
crystallographically controlled parting model. First, the
DPOs occur in isolation, encompassed by mesh-textured
olivine grains (Fig. 4). Second, although some elongated
olivine grains form single domains with minimal misorientation within the DPOs (blue area in Fig. 5b), others lack
any crystal-preferred orientation associated with the parting and are highly misoriented with respect to one another.
Third, the mCT visualization of magnetite and serpentine
reveals that the DPO olivine grains are parallel in three dimensions and are completely surrounded by the
mesh-textured olivine (Fig. 7).
In contrast, olivine grains in a shear zone that display
‘perfect cleavage’ are not found in isolation, but as many
grains that have been aligned during deformation.
Nozaka & Ito (2011) interpreted the ‘perfect cleavage’ of
these olivine grains to be the consequence of brittle failure
along dislocation walls with contemporaneous antigorite
penetration. This scenario would produce a more systematic rock fabric where multiple neighboring grains are
rotated towards the stable shear plane and develop a parting (e.g. Lister & Snoke, 1984), rather than the observed
isolated DPOs. Moreover, the parted segments of these
grains would have a low degree of misorientation between
each other or systematic crystallographic deviations.
Similarly, parting as a result of shock-induced deformation
would also lead to planar fracturing along specific planes
without the induction of single domain rotational deformation, and thus cannot explain the observed DPO structures either (e.g. Mu«ller & Horneman, 1969). In fact, none
of the deformation processes that produce parting in olivine can account for the progressive rotation of single elongated olivine grains in the DPOs with retention of a
parallel habit in the third dimension (Fig. 7), because discontinuous olivine rods within a heterogeneous stress field
would be necessary to produce this structure. Furthermore,
the development of a hierarchical mesh-texture network
surrounding the DPOs indicates that serpentinization was
independent of far-field (tectonic) stresses on a grain-scale
(e.g. Iyer et al., 2008b). During syn-metamorphic deformation, a preferentially oriented fracture network would
probably develop, affecting both the DPOs and the encompassing mesh-textured olivine grains. Furthermore, dislocation density and low-angle subgrain boundaries
indicative of high-temperature crystal-plastic deformation
are found only within olivine of Domain II (compare
with Arndt et al., 2010; Soustelle et al., 2010) and do not
occur in DPO olivine grains (Figs 5c and 6a). A similar
absence of microstructural markers for dislocation creep
has recently been observed in metamorphic olivine after
antigorite (Padron-Navarta et al., 2010) and has been interpreted to be indicative of secondary olivine formation. In
addition, crystallographic relationships of magnetite and
olivine throughout the samples indicate that the DPOs
are significantly different from the surrounding
1955
JOURNAL OF PETROLOGY
VOLUME 53
NUMBER 9
SEPTEMBER 2012
Fig. 10. (a) BSE image of a representative internal DPO area and corresponding WDX element distribution maps of Al (b) and Cr (c).
Clinopyroxene lamellae (Cpx lam.) are characterized by elevated Al and Cr contents compared with the low Al content of the diopside aggregates (Di agg.). (d) BSE image and corresponding Al (e) and Cr (f) element maps highlighting the high-Al and -Cr serpentine inclusions that
are commonly fully enclosed within the DPO olivine. Ellipses in both BSE images and the element maps highlight their positions. Phases
with high backscattering contrast in (a) and (d) are Cr-rich magnetite (Mag) grains. Brc, brucite; Lz, lizardite; Ol, olivine.
mesh-textured olivine grain matrix (Fig. 13). Consequently,
we propose that the DPOs are of metamorphic and
secondary origin in comparison with surrounding relicts
of predominantly primary mantle olivine. Comparisons
of MnO and NiO concentrations within the meshtextured olivine and DPO olivine grains (Fig. 14a and b)
with those from previous investigations of welldocumented secondary olivine after serpentine support
the secondary nature of the DPO olivine grains (e.g.
Pinsent & Hirst, 1977; Vance & Dungan, 1977; O’Hanley,
1996).
Al- and Cr-rich chrysotile found exclusively as heterogeneously distributed inclusions within the DPOs (Fig. 10)
can help to constrain the protolith mineral. If these serpentine inclusions formed after DPO olivine via the infiltration of Al- and Cr-bearing fluids they would be
widespread in occurrence. However, we did not observe
any serpentine with similar Al and Cr concentrations, or
microstructural associations, outside the DPOs; thus they
are directly related to the protolith mineral. As Al is
found neither as a major nor trace component within
olivine, whereas serpentine after orthopyroxene is known
to incorporate considerable amounts of Al and Cr (e.g.
Dungan, 1979; Wicks & Plant, 1979; Viti & Mellini, 1998),
we propose that the protolith mineral was an orthopyroxene grain.
Mantle orthopyroxene grains commonly exhibit microscopic and submicroscopic solid-state diffusion-driven
exsolution features, originating from cooling or decompression of the originally high-temperature and -pressure
orthopyroxene (e.g. Buseck et al., 1980). Of these features,
exsolution of (100) clinopyroxene lamellae is a common
subsolidus feature in orthopyroxene grains, which often
remain nearly undisturbed during serpentinization (Viti
et al., 2005). In contrast, if the DPO clinopyroxene lamellae
were exsolved directly from the DPO olivine, as observed
in some natural occurrences (e.g. Markl et al., 2001), the intensive brittle deformation described earlier that is needed
to explain the observed DPO misorientations should
affect both the exsolution lamellae and the host mineral.
However, within the DPOs, clinopyroxene lamellae
display systematic lattice distortions indicative of
1956
PLU«MPER et al.
OLIVINE PSEUDOMORPHS, LEKA, NORWAY
Fig. 11. Representative crystallographic orientation analysis of crystal-plastically deformed clinopyroxene lamellae (Cpx lam.) within a
DPO using the EBSD technique. (a) Combined textural component
map of the analyzed area highlighting the clinopyroxene lamellae between parallel olivine (Ol) grains. (b) Pole figure (stereographic,
upper hemisphere) displaying the lack of significant dispersion of
points around the rotation axis [010], unlike data points of the [100]
and [001] axes, which describe a progressive dispersion following a
circle around the rotation axis, confirming that the clinopyroxene
lamellae were crystal-plastically deformed. The reference frame of
the pole figure is chosen to coincide with the x (horizontal axis) and
y (vertical axis) of the image in (a).
high-temperature crystal-plastic deformation (Fig. 11),
which are absent within the surrounding DPO olivine
(Figs 5c and 6a). Furthermore, the DPO clinopyroxene
lamellae are consistently aligned in the (100) orientation
(Fig. 8e), without any relationship to the varying orientations of the DPO olivine grains. If the lamellae were
hydrothermal in origin their Al2O3 and Cr2O3 concentrations should be comparable with those of the pure diopside
aggregates that also occur within the DPOs, which have
Fig. 12. (a, b) BSE images of pervasive replacement of DPOs by
diopside aggregates (Di agg.) retaining the underlying DPO microstructure including the clinopyroxene lamellae (Cpx lam.), which
remain visible even when diopside almost completely replaces the
DPO as displayed in (b). Magnetite (Mag) abundance increases
with increasing degree of diopside replacement. Magnetite/diopside
ratios are calculated from equal-sized areas. Atg, antigorite; Brc, brucite; Lz, lizardite; Ol, olivine.
the characteristic clinopyroxene composition of hydrothermal secondary origin within meta-peridotites (e.g. Frost,
1975; Evans, 1977; Peretti et al., 1992; Li et al., 2004;
McCollom & Bach, 2009). However, the lamellae have
Al2O3 and Cr2O3 concentrations comparable with those
of published clinopyroxene exsolution lamellae within
orthopyroxene grains (Al2O3 2·5^7·5 wt %, Cr2O3 1^
2 wt %; Cannat et al., 1992; Viti et al., 2005; Dijkstra et al.,
2010) and those of primary single clinopyroxene grains
from the investigated samples, which are higher than the
hydrothermal diopside, supporting the origin of the lamellae as a ‘primary’ solid-state exsolution from a now extinct
orthopyroxene grain (Fig. 14e and f).
Although the whole-rock geochemistry implies the
former presence of orthopyroxene (Fig. 2a) we did not observe any primary orthopyroxene grains or bastite after
orthopyroxene within the LOC meta-peridotite. The only
1957
JOURNAL OF PETROLOGY
VOLUME 53
NUMBER 9
SEPTEMBER 2012
Fig. 13. Pole figures illustrating the crystallographic relationship between (a) magnetite (Mag) and olivine (Ol) grains within a DPO as well
as (b) magnetite and partially serpentinized mesh-textured olivine grains. For simplicity only the [001] projection of magnetite and olivine
are shown. Clustering magnetite orientations are depicted as squares and hexagons. In (a) it should be noted that all magnetite grains
(nMag ¼102) consistently show the same two orientations even though olivine orientations within the DPO vary. (b) Magnetite grains
(nMag1 ¼67; nMag2 ¼ 35) within different olivine areas exhibit two orientations depending on the original olivine domain orientation (circle).
Comparison of (a) and (b) indicates that magnetite grains in DPOs are independent of the DPO olivine orientation, but show a topotactic
relationship with mesh-textured olivine grains.
bastite identified in the LOC is restricted to metaorthopyroxenite dykes within the layered cumulate section
of the ophiolite (Iyer et al., 2008a). Orthopyroxene can be
directly replaced by olivine during ‘dunitization’, either
through partial melting and basaltic melt extraction
(e.g. Kelemen, 1990), or the infiltration of a silicaundersaturated, carbonatite melt (e.g. Yaxley et al., 1991).
The former is unlikely, as partial melting would shift the
whole-rock composition to dunite (Fig. 16) and the latter
should leave substantial traces of accessory minerals such
as apatite and titanite, which are absent in the LOC
meta-peridotite. Reactions indicative of these processes
would also affect the clinopyroxene lamellae and the
primary clinopyroxene grains; however, they remain predominantly unaltered. Moreover, both aforementioned
scenarios would lead to a systematic ‘replacement’ without
the observed misorientations (Fig. 5b) or Al- and Cr-rich
serpentine inclusions that are exclusively found within the
DPO olivine grains (Fig. 10).
Therefore, we suggest that the precursor silicate of
each DPO was a single orthopyroxene grain with clinopyroxene exsolution lamellae that was initially pseudomorphically serpentinized to form a so-called bastite
(Dungan, 1979) and later completely dehydrated to result
in the secondary olivine that dominates the present
microstructure.
Pseudomorphism during orthopyroxene hydration and
bastite dehydration
Preservation of protolithic orthopyroxene features in the
DPOs, such as the bent clinopyroxene lamellae visible
over several hundred micrometers, indicates that both
hydration and dehydration proceeded to a great extent
pseudomorphically (Fig. 8). This is also indicated by the
retention of morphological outlines. Indeed, average DPO
sizes, measuring 1^5 mm diameter, are in agreement
with typical mantle orthopyroxene grain sizes (e.g.
Mercier & Nicolas, 1975; Ceuleneer et al., 1988; Soustelle
et al., 2010). Mesh-textured olivine grains (6 mm diameter; Fig. 5e) from less intensively antigorite-serpentinized
areas, are also in agreement with primary mantle grain
sizes. Moreover, Mn-rich olivine haloes observed at the
DPO margins probably represent an area of disequilibrium
at former olivine^orthopyroxene interphase boundaries
(Fig. 5a and b); however, their timing within the alteration
sequence is not resolvable.
The presence of mineral pseudomorphs in retrograde
metamorphic rocks but their absence in prograde rocks
led Carmichael (1969) to argue that prograde transformation mechanisms, generally involving dehydration, do
not involve simple replacement. From this observation
Ferry (2000) concluded that the force of crystallization
plays a crucial role during pseudomorphic mineral replacement. Force of crystallization is the mechanical work
exerted from a growing crystal onto its surroundings if the
surrounding has a finite yield strength. Ferry (2000)
inferred that a negative total reaction molar volume
change, taking the large molar volume of free H2O into
account (e.g. Vm,calc H2O ¼ 20 cm3 mol1 at 3508C and 2
kbar; Duan et al., 2008), will result in a negative strain
free energy promoting pseudomorphism. In contrast,
during dehydration reactions the total molar volume
change is positive and force of crystallization prohibits
pseudomorphic replacement. This approach, however, considers only a closed system, neglecting the preservation of
the solid component’s volume. Particularly in the case of
mineral dehydration (assuming solely thermal dehydration), water needs to be removed to drive the reaction
away from equilibrium (e.g. Ague et al., 1998). Thus, our
observations
of
pseudomorph
retention
during
1958
PLU«MPER et al.
OLIVINE PSEUDOMORPHS, LEKA, NORWAY
(a)
1.80
1.60
Olivine
1.20
1.00
0.30
0.80
0.20
NiO (wt%)
0.40
Mesh
DPO center
DPO halo
1.40
MnO (wt%)
(b) 0.50
0.60
0.10
0.40
0.20
0.84
0.86
0.88
0.90
0.92
0.82
0.84
0.86
0.90
0.00
0.92
XMg
XMg
(c)
7.00
(d) 1.50
Serpentine
6.00
Al2O3 (wt%)
0.88
1.25
Lizardite (Mesh)
Lizardite (DPO)
Chrysotile inclusions
Antigorite
5.00
4.00
3.00
1.00
0.75
0.50
Cr2O3 (wt%)
0.82
2.00
0.25
1.00
0.00
0.00
0.90
0.91
0.92
0.93
0.94
0.95
0.96
0.97
0.89
0.90
0.91
0.92
XMg
0.94
0.95
0.96
0.97
XMg
3.50
Al2O3 (wt%)
0.93
(e)
(f)
3.50
3.00
3.00
2.50
2.50
2.00
2.00
1.50
1.50
Clinopyroxene
Large grain
Lamella
Aggregate
1.00
0.50
1.00
0.50
0.00
0.00
Al2O3 (wt%)
0.89
0.00
0.20
0.40
0.60
0.80
1.00
0.93
Cr2O3 (wt%)
0.94
0.95
0.96
0.97
0.98
XMg
Fig. 14. Compositional comparison of mesh-textured olivine vs DPO olivine in (a) and (b), the three identified serpentine varieties in (c)
and (d), and large clinopyroxene grains, clinopyroxene lamellae as well as diopside aggregates in (e) and (f). With respect to (a) and
(b), DPOs are subdivided into central areas (DPO center) and the Mn þ Fe-rich, Ni-poor olivine halo (DPO halo) at the margin of the domains (see Fig. 8a and b). In (c) and (d) analyses of lizardite from both the mesh-textured olivine grains and between the parallel olivine
grains within the DPOs have the same compositional characteristics. In (e) and (f) large clinopyroxene grains outside the DPOs and clinopyroxene lamellae within the DPOs overlap in composition (gray area), whereas diopside aggregate compositions do not. Dashed lines indicate
0 wt %.
1959
JOURNAL OF PETROLOGY
VOLUME 53
NUMBER 9
SEPTEMBER 2012
orthopyroxene^bastite dehydration indicate that the molar
volume of water would have only a minor influence on
the force of crystallization and pseudomorphs may well develop during prograde metamorphism.
Although orthopyroxene has been observed to hydrate
directly to form talc and secondary olivine at temperatures
well within the stability field of olivine (46008C), via a
reaction such as (Kimball et al., 1985)
5MgSiO3 þ H2 O ! Mg3 Si4 O10 ðOHÞ2 þ Mg2 SiO4
ðEnÞ
ðTlcÞ
ðOlÞ
ð1Þ
Fig. 15. BSE images of inclusion trails composed of serpentine within
mesh-textured olivine (Ol) grains of Domain II. In the vicinity of
Cr-spinel grains chlorite is also found as inclusions. (a) Lowmagnification images of an interface (dashed line) between DPO
(right) and mesh-textured olivine grains (left), exhibiting several
crosscutting inclusion trails within the mesh-textured olivine grain
assemblages (white oval). (b) Close-up of an olivine domain
within a mesh showing several crosscutting inclusion trails, denoted
by dashed lines, that is comparable with the area shown in (a).
Trails are typically 10^30 mm wide and have the appearance of a
healed mesh texture. (c) BSE image highlighting the common
complex internal structure of the inclusion trails that also frequently
contain magnetite. The post-dating serpentine veins that crosscut the
inclusion trails should be noted. Brc, brucite; Lz, lizardite; Mag,
magnetite.
it is unlikely that this occurred in our samples, as we do not
find evidence for concomitant talc co-evolution with the
secondary olivine. Even if talc was present and subsequently overprinted, reaction (1) would produce 40%
olivine, whereas in our samples the observable area per
cent of olivine in the DPOs is 65^70%. Fully enclosed
clinopyroxene lamellae within olivine, in conjunction with
the compositional characteristics of the DPO lizardite
(Fig. 14c and d), imply that prior to lizardite formation
the amount of olivine was 10^20% more than what is observable today.
A variety of studies have shown that lizardite forms as
the major bastite component with minor chrysotile talc,
although the MgO^SiO2^H2O (MSH) phase diagram
predicts that orthopyroxene should form antigorite þ talc
at high temperatures if water is present. However, water
infiltration into unserpentinized peridotite will be at low
flow rates owing to their low permeability and porosity,
hindering the hydration of orthopyroxene to antigorite þ talc. Nevertheless, even if water is able to penetrate
to the reaction site, lizardite may form owing to the stabilizing effect of trivalent cations (Al3þ, Cr3þ, Fe3þ) present
in lizardite after orthopyroxene (e.g. Wicks & Whittaker,
1977; Dungan, 1979; Viti & Mellini, 1998). Although the
only bastite^serpentine relicts identified in our samples
are Al- and Cr-rich chrysotile inclusions (Figs 9 and 10),
we cannot rule out the presence of lizardite with similar
Al and Cr concentrations within the DPOs, which may
simply be a result of the amount of Raman spectroscopic
analyses executed. Nevertheless, the absence of antigorite
and talc, as well as the survival of some Al- and Cr-rich
chrysotile (and lizardite?) as inclusions within the DPO
olivine, suggests that in the investigated samples orthopyroxene serpentinization proceeded as described in previous
studies (e.g. Dungan, 1979). Balancing the reaction on
volume results in the pseudomorphic replacement (approximate Vm of enstatite and lizardite are 25 and
100 mol cm1, respectively)
2þ
4MgSiO3 þ 2Hþ
ðaqÞ þ H2 O ! Mg3 Si2 O5 ðOHÞ4 þMgðaqÞ þ 2SiO2ðaqÞ
1960
ðEnÞ
ðLz=CtlÞ
ð2Þ
PLU«MPER et al.
54
-16%
52
-12%
pure SiO2-loss
MgO (wt%)
therefore, it is likely that the released SiO2(aq) reacted
with surrounding olivine to form serpentine without brucite via a reaction such as
SiO2-loss
50
-6%
-10%
48
46
OLIVINE PSEUDOMORPHS, LEKA, NORWAY
MgO-loss
3Mg2 SiO4 þ SiO2ðaqÞ þ 4H2 O ! 2Mg3 Si2 O5ðOHÞ4 :
ðOlÞ
Srp
-2%
ð4Þ
model harzburgite
(80 Ol, 15 Opx, 5 Cpx)
44
model dunite
42
LOC meta-peridotite
LOC meta-dunite
(95 Ol, 5 Cpx)
36
38
40
42
44
SiO2 (wt%)
46
48
Fig. 16. SiO2 vs MgO diagram showing how the whole-rock compositions of LOC meta-peridotite and meta-dunite would change during
large-scale desilicification and extraction of silica from the system.
A model harzburgite (forsterite, enstatite and diopside end-member
compositions) was used as a reference point to calculate the SiO2and MgO-loss isopleths. Whole-rock compositions of LOC metaperidotites exhibit a natural spread around the model harzburgite
composition and thus do not show any trend towards a more dunitic
composition expected during large-scale extraction of silica from the
system. The figure was constructed on a water-free basis (Table 1).
Cpx, clinopyroxene; Ol, olivine; Opx, orthopyroxene.
indicating that Mg and Si are lost to the fluid during
orthopyroxene serpentinization. In turn, on the basis of
the observations above, orthopyroxene^bastite dehydration also proceeds primarily via an isovolumetric reaction
such as (approximate Vm of lizardite and olivine are
100 cm3 mol1 and 50 cm3 mol1, respectively)
þ
Mg3 Si2 O5 ðOHÞ4 þMg2þ
ðaqÞ ! 2Mg2 SiO4 þ 2HðaqÞ þ H2 O:
ðLzÞ
ðOlÞ
ð3Þ
Reactions (2) and (3) are a rough estimate, as we do not
know the composition of the fluid, the amount of transiently generated porosity during replacement and whether
talc formed. However, comparison of both reactions highlights that overall only SiO2(aq) would need to be removed
to maintain constant volume during a sequential alteration
scenario of orthopyroxene hydration and bastite dehydration. Comparison of calculated model harzburgite
with a whole-rock composition of DPO-containing metaperidotite indicates that the released SiO2(aq) was mobile
on only a micrometer to centimeter scale, because
large-scale desilicification with migration of SiO2(aq) out
of the system would shift the whole-rock composition considerably towards a dunitic composition (Fig. 16). Strong
gradients in silica activity (aSiO2(aq)) are known to develop between orthopyroxene and olivine during serpentinization (Frost & Beard, 2007; Klein et al., 2009);
This reaction is possible above 4008C; thus if water is
able to penetrate to the orthopyroxene grains the released
SiO2(aq) would aid olivine serpentinization, even within
the olivine stability field (e.g. Frost & Beard, 2007; Evans,
2010). A‘back-of-the-envelope’ calculation, assuming a peridotite composition of 80% olivine and 20% orthopyroxene, where all orthopyroxene has been serpentinized
following reaction (2), indicates that the released SiO2(aq)
could aid in the hydration of 40% olivine directly to serpentine without the production of brucite.
Although it is possible to define a narrow temperature
field (350^4008C) in which lizardite/chrysotile (Mg-endmember) dehydration occurs contemporaneously with
olivine hydration to antigorite, the presence of Al in lizardite/chrysotile after orthopyroxene, which is inferred to be
several weight per cent (Fig. 10), stabilizes lizardite/chrysotile to temperatures above 5008C (Caruso & Chernosky,
1979). Both the Al and Cr must have been mobile during
orthopyroxene^bastite dehydration as neither is incorporated into the resulting olivine. Whereas Cr was probably
incorporated into the abundant Cr-magnetite grains, Al
shows no evidence of being assimilated into new mineral
phases, neither in interfacial areas in the DPOs nor in
their surroundings. Thus, we have to assume that Al was
mobile on at least the thin section-scale and was either
transported away or diluted in subsequent hydration
events to be incorporated into, for example, newly forming
serpentine minerals. Although Al has long been regarded
as an inert element in geological processes (e.g.
Carmichael, 1969; Ragnarsdottir & Walther, 1985;
Walther, 1997), more recent natural and experimental investigations have indicated that it can be mobile during
fluid^rock interaction in the deep crust and mantle (e.g.
Tagirov & Schott, 2001; Newton & Manning, 2008).
Placing DPO formation within the
alteration sequence
Within the investigated samples there are three distinguishable sets of microstructures that can be assigned to
different stages in the metamorphic evolution of the investigated meta-peridotite. Although the characteristics of
the first, high-temperature stage do not have to be simultaneous, it led to the exsolution of clinopyroxene lamellae,
the formation of crystal-plastic deformation structures
that remain visible owing to the bending of some DPOs
(Fig. 8a; Domain I), and the formation of subgrain boundaries within some of the now mesh-textured olivine
1961
JOURNAL OF PETROLOGY
VOLUME 53
(Domain II). The second and third stages can be deduced
from examination of reaction microstructures and crosscutting relationships of serpentine varieties within
Domain II, which imply that two extensive serpentinization stages occurred that can be traced throughout the
entire mantle section: (1) partial lizardite-serpentinization
of olivine resulting in mesh-texture formation followed by
(2) a later antigorite-serpentinization of olivine relicts
and massive antigorite-overprinting of all previous microstructures (Figs 4 and 8). At first glance it appears that
the orthopyroxene hydration could be assigned to the
lizardite-serpentinization stage and the dehydration to the
antigorite-serpentinization stage. However, there are several observations that indicate that DPO formation
occurred prior to the aforementioned serpentinization
stages.
In the most recent stage, antigorite randomly overprints
the secondary olivine within the DPOs (Fig. 8), therefore
postdating the DPO formation. Furthermore, the temperature range of the antigorite-serpentinization stage (300^
4508C) is estimated to be below the expected stability
range of Al-rich serpentine (45008C; Caruso &
Chernosky, 1979) observed as inclusions within the DPO
olivine (Fig. 10). We do not observe any talc þ olivine in
the investigated samples, which can be formed through
lizardite/chrysotile or antigorite breakdown. According to
Gibbs’ free energy considerations, the lack of such a transformation suggests that the antigorite-serpentinization
stage probably did not exceed 450^5008C (Evans, 2004,
Fig. 6). This is supported by the absence of tremolite formation after primary or secondary clinopyroxene (Fig. 17b).
There is also compelling evidence that antigorite brucite
was formed directly from primary mantle olivine grains
during fluid infiltration (Plu«mper et al., 2012), rather than
solely as a result of thermal, H2O-conserving lizardite/
chrysotile recrystallization. This suggests that the antigorite-serpentinization temperature may not have exceeded
even 4008C because olivine (pure forsterite) is stable
above this temperature when excess H2O is present
(Evans, 2004). Therefore, the stability of Al-rich lizardite/
chrysotile, originating from orthopyroxene serpentinization, exceeds the upper temperature limit for the
antigorite-serpentinization stage and thus would not be expected to dehydrate. This is in agreement with observed
lizardite (1^5 wt % Al2O3) survival well beyond the
antigorite-out reaction (Frost, 1975).
Lizardite brucite veins crosscut the DPO microstructures, indicating that DPO olivine formation also pre-dates
the lizardite-serpentinization event. Lizardite replacement
of the DPO olivine is also evident along the DPO ‘pseudocleavage’, because the Al-free lizardite found between the
parallel olivine domains (Domain I) shows a compositional overlap with lizardite that forms the mesh texture
in olivine grains in Domain II (Fig. 14c and d). Moreover,
NUMBER 9
SEPTEMBER 2012
the cryptic presence of brucite in the Al-free lizardite in
Domain I implies that it formed after the DPO olivine
and not during primary orthopyroxene hydration, as brucite formation during orthopyroxene serpentinization is
unfavorable owing to an increased aSiO2(aq) hydration environment (e.g. Frost & Beard, 2007). In this case the interphase boundary between the metamorphic olivine and
clinopyroxene exsolution lamellae was a preferred site for
fluid infiltration and serpentinization initiation, creating
the observed ‘pseudo-cleavage’ appearance with the sequence olivine^lizardite^clinopyroxene^lizardite^olivine.
As lizardite-serpentinization temperatures (50^3008C, Fig.
14) can be expected to have been lower than those of the
antigorite-serpentinization event, orthopyroxene^bastite
dehydration could have not occurred during this stage
either.
On the basis of the microstructural and microchemical
investigations, as well as phase stability considerations, we
suggest that two metamorphic events occurred in addition
to the observed lizardite- and antigorite-serpentinization
episodes: (1) a hydration stage that led to the orthopyroxene serpentinization and serpentinization of the surrounding olivine grains aided by the released SiO2(aq) upon
orthopyroxene serpentinization and (2) a transient dehydration stage that dehydrated orthopyroxene^bastites and
‘healed’ any mesh textures after olivine (Fig. 15). Although
direct indications for the latter are scare, re-hydration and
subsequent serpentinization probably reused previous,
pre-existing fluid pathways to obliterate the microstructural evidence. None the less, both metamorphic stages
that pre-date the evident lizardite- and antigoriteserpentinization stages are recorded by the DPO microstructure (Fig. 17).
Searching for the dehydration event
Deserpentinization has been suggested to be either
the result of contact metamorphic events, such as magma
infiltration into meta-peridotites or -serpentinites
(Trommsdorff & Evans, 1972; Vance & Dungan, 1977), as a
consequence of hydrated oceanic lithosphere subduction
(e.g. Hacker et al., 2003; Padron-Navarta et al., 2011), or
due to Barrovian-type metamorphism during mountain
building (e.g. Evans, 1977). However, in the LOC
large-scale magmatic intrusions that could induce dehydration are absent. Similarly, an entire ophiolite sequence
that has undergone subduction and later exhumation
would be expected to be at least partially dismembered
and should show blueschist- and/or eclogite-facies associations within the crustal section (e.g. Angiboust et al.,
2011), but neither are observed on Leka (Fig. 1). It has
been suggested that the antigorite-serpentinization overprint is a consequence of Caledonian regional metamorphism that produced a regular Barrovian-type
metamorphic pattern throughout the mantle fragments in
the Scandinavian Caledonides, with an increase of
1962
PLU«MPER et al.
OLIVINE PSEUDOMORPHS, LEKA, NORWAY
MS1
Decompression
Rela
tive
chro
nolo
MS2
Hydration
gy o
(100)
MS3
Dehydration
f mi
cros
truc
ture
evo
MS4
Re-hydration
Orthopyroxene
serpentinization
(bastite formation) &
primary olivine
serpentinization
Mineralogy &
Structure
~200 < T < ~400 °C
Primary (P)
Secondary (S)
OlS
OlP
CpxP
Atg
(b)
MS2
MS4
Bastite dehydration
to secondary olivine
& dehydration of
serpentine after
primary olivine
T > ~500 °C
Lz/
Lz/
Ctl(Opx) Ctl(Ol,Di)
Di
Grain/
Interphase
boundary
MS3
200
300
~300 < T < ~400 °C
Atg
Fo+Tlc+
H2 O
?
Phase
Pha diagram
(C)M(A)SH-system
(C)M(
O
2
t
(C
Fo+ tl)/Lz
Tlc+
Chl
+H XAl =0.2
2O
(Ctl
)/Lz
Fo+
Tlc+
Chl
XA =0.5
+H
l
2O
Ctl
Fo+T
lc+H
-ou
Atg
Ol-in
Trin
1
Atg+D
i
Tr+Fo
+H O
2
Atg+Brc
Fo+H O
2
2
Ctl/Lz+Brc
Fo+H O
Ctl/Lz
Fo+Atg+H O
2
Lz
Atg+Brc
Ctl
suboceanic
MOHO
Atg+Brc
Pressure (kbar)
3
2
Antigorite
serpentinization of
primary &
secondary olivine
?
MS5
6
4
~200 < T < ~300 °C
MS = Microstructure
Mag
7
5
Lizardite
serpentinization of
primary &
secondary olivine
Tlc+Fo
En+H2O
T > ~1000 °C
(a)
n
MS5
Hydration
Solid-state
re-equilibration:
clinopyroxene
exsolution
from orthopyroxene
Opx w/
Cpx ex.
lamella
lutio
400
500
Temperature (°C)
U
Univariant
reactions
r
stable
metastable
600
Fig. 17. (a) Relative chronology of the microstructure evolution summarizing the observed transient dehydration within the mantle
meta-peridotite of the Leka Ophiolite Complex.The evolution is illustrated using the three main rock-forming minerals orthopyroxene (Opx), olivine (Ol) and clinopyroxene (Cpx); spinel is excluded because of a negligible contribution to the DPO evolution. (b) Phase diagram in the system
MgO^SiO2^H2O (MSH) displaying the stability of antigorite (Atg), lizardite (Lz) and chrysotile (Ctl) with varying temperature and pressure,
using a compilation of available data from experimental, theoretical and naturally constrained data sources (Andreani et al., 2007, and references
therein).White curve above 4008C is added from the system CaO^MgO^SiO2^H2O (CMSH) and dashed curves from MgO^Al2O3^SiO2^H2O
(MASH) to illustrate the breakdown of diopside þ antigorite (Di þ Atg) to tremolite (Tr-in) andthe enlarged stability field of lizardite (and possibly
chrysotile) as a function of its Al content (XAl ¼ 0·2 is 3·7 wt % Al2O3 and XAl ¼ 0·5 is 9·2 wt % Al2O3), respectively.Temperature ranges for the different evolutionary stages of the DPO microstructure are indicated above the phase diagram. MS1 is not shown in (b) as this microstructure is the
result of a decompression event occurring at higher temperatures. Brc, brucite; Chl, chlorite; En, enstatite; Fo, forsterite; Tlc, talc.
1963
JOURNAL OF PETROLOGY
VOLUME 53
metamorphic grade from SE to NW (Bucher-Nurminen,
1991). Thus, if the observed dehydration was a consequence
of Barrovian-type metamorphism, its transient nature and
the subsequent lower temperature rehydration to lizardite
prior to antigorite-serpentinization (Fig. 17a) need to be
accounted for. Alternatively, both the lizardite- and
antigorite-serpentinization could be synchronous or
post-date Caledonian metamorphism, although this is unlikely as the LOC is situated in the Upper or Uppermost
Allochthon and hence, on the basis of the Norwegian tectonostratigraphy, the antigorite-serpentinization should represent the lowest grade (54008C) of Caledonian
metamorphism (Bucher-Nurminen, 1991).
Deserpentinization can also occur in lower pressure settings; for example, where fault movements displace serpentinized mantle to come into contact with the hotter end of
an adjacent ridge segment (Rutter & Brodie, 1987). In situ,
low-pressure transient dehydration has been proposed to
occur owing to the feedback between mantle hydration
and hydrothermal convection at oceanic spreading centers.
Simulations of such feedback, which take into account
evolving permeability, indicate that in situ serpentine hydration^dehydration^rehydration can be active in transient
phases as convective cells are formed and merge together
(Iyer et al., 2010). Subsequently the deserpentinized mantle
can once again be re-hydrated when a quasi-steady-state
regime is established and temperatures favor serpentinization, overprinting initial deserpentinization episodes.
Alternatively, influx of new hot asthenosphere or fluid
flux contributions from a subducting slab may also be
common in the supra-subduction zone settings of evolving
ophiolites (Shervais, 2001; Hyndman & Peacock, 2003),
contributing to the thermal fluctuations required.
At present, we are unable to unequivocally answer what
initiated the dehydration event. Further isotope and trace
element data analyses are required to elucidate the origin
of the fluid(s) that induced serpentinization and thus identify the metamorphic event that caused dehydration of the
meta-peridotite. During ‘conventional’ serpentinization,
fully serpentinized orthopyroxene grains are typically surrounded by considerably serpentinized olivine grains;
hence, detailed analysis of the olivine serpentinization
prior to dehydration is also required to determine the
extent of the first hydration stage. With respect to the
latter, it is interesting to note that examinations of serpentinized meta-dunite within the ultramafic cumulate section
(Fig. 1) do not exhibit microstructures that indicate extensive serpentine dehydration.
Recognition and survival of microstructures during multi-stage, fluid-mediated
pseudomorphism
Metamorphic rocks can undergo a prolonged history,
which can be partly or completely deciphered from the
NUMBER 9
SEPTEMBER 2012
present-day microstructures, either as an evolution along
a polyphase P^T path or simply because of multiple fluid
infiltration events. For example, fluid-mediated mineral replacement reactions often retain the external form (habit)
and crystallographic details of the parent. Although complete replacement reactions make it difficult to identify
the metamorphic nature of the system, requirements for
pseudomorphism can be used to help understand the reactions because of their ability to retain crucial information
about the replaced mineral (e.g. the retention of
micron-sized twinning; Putnis, 2009). As seen in Fig. 8,
this information can also be the survival of a specific mineral phase, which may regulate the microstructure from
an early stage in the rock’s history. In the DPOs, although
the clinopyroxene lamellae originate from the subsolidus
re-equilibration of the former orthopyroxene grain, they
controlled the evolution of the microstructure by providing
fluid pathways and defining the single elongated secondary
olivine grains (Fig. 8). Misinterpretation of fully enclosed
clinopyroxene lamellae within olivine (Fig. 8f) could lead
to erroneous conclusions; for example, that the clinopyroxene lamellae were the result of solid-state exsolution from
olivine, implying incorrect formation conditions and thus
metamorphic history (see Ren et al., 2008). Even when
these mineral relicts are obscured by the hydrothermal
equivalent of the relict phase (diopside aggregates), certain
characteristics such as the chemistry and crystallographic
orientation can still be identified (Fig. 12). Detailed investigations, coupling crystallographic and chemical data, may
lead to more widespread recognition of such ‘ghost’ mineral
microstructures. For example, the occurrence of oriented
lamellae in minerals, especially those that cannot be explained by classical closed-system solid-state diffusiondriven unmixing, may represent such ‘ghost’microstructure
and not, as suggested, solid-state exsolution of either an
exotic precursor phase, occurring at ultrahigh-pressure
conditions (Dobrzhinetskaya et al., 2009), or a precipitate
caused by diffusional exchange with the matrix through
an intact host crystal lattice (open-system precipitation;
Proyer et al., 2009). In a fluid-driven reaction environment
the mechanism of textural re-equilibration will also be
through dissolution and reprecipitation, thus ‘closure temperatures’ at which microstructures cease to evolve will be
much lower than during solid-state diffusion (Putnis,
2009; Putnis & Austrheim, 2010). Hence, evidence for
microstructures as a result of an interface-coupled dissolution^reprecipitation mechanism may be only cryptically
preserved or not at all. However, careful investigations of
mineral replacement reactions by applying a multi-method
approach that encompasses a combination of microstructural, crystallographic and microchemical analytical
methods may enable us to see through multiple sequential
events.
1964
PLU«MPER et al.
OLIVINE PSEUDOMORPHS, LEKA, NORWAY
S U M M A RY A N D C O N C L U S I O N S
We have examined the pseudomorphism of single orthopyroxene grains via serpentine^bastites to secondary olivine, recording a transient dehydration stage within the
oceanic lithospheric mantle section of the Leka Ophiolite
Complex (Norway). On the basis of observed 2-D and
3-D mineral replacement microstructures, their consecutive relationships, crystallography, mineral chemistry and
phase stabilities, we can summarize the observed hydration^dehydration^rehydration sequence in terms of the following proposed evolutionary sequence (Fig. 17).
(1) During subsolidus solid-state re-equilibration at high
temperature orthopyroxene grains exsolve clinopyroxene lamellae (MS1). The clinopyroxene lamellae exhibit crystal-plastic deformation-induced bending,
indicating that the rock was subject to asthenospheric
mantle flow during or after clinopyroxene exsolution.
(2) Orthopyroxene hydrates to form serpentine (bastite;
MS2) with partial serpentinization of surrounding
olivine grains [200^4008C(?)], probably aided by the
SiO2(aq) released during pseudomorphic orthopyroxene serpentinization. Clinopyroxene exsolution lamellae survive.
(3) Dehydration of orthopyroxene^bastites (MS3) causes
serpentine breakdown to secondary olivine enclosing
Al- and Cr-rich bastite^serpentine relicts (45008C).
Clinopyroxene exsolution lamellae prevail and control the parallelism of the secondary olivine grains to
result in the DPO microstructure. Dehydration
‘heals’ any mesh textures formed in the primary olivine grains.
(4) Re-hydration at low temperatures (c. 53008C) initiates lizardite-serpentinization of the DPOs as well as
the primary mantle olivine and clinopyroxene grains
(MS4). Lizardite-serpentinization probably re-uses
previous, pre-existing fluid pathways to obliterate the
microstructural evidence of ‘healed’ mesh textures.
(5) Later higher temperature hydration (300^4008C) initiates antigorite-serpentinization replacing both relict
mantle olivine and secondary DPO olivine, destroying all former underlying microstructures (MS5).
We have shown that microstructures that have undergone multiple overprinting processes can still retain evidence of the earlier metamorphic stages during the
history of a rock. Although some events are easily distinguishable, others are evident only because of their crystallographic and petrological relationships with respect to
later events. Thus, a detailed examination of the mineral
chemistry, reaction microstructures and crystallographic
associations was required, utilizing a wide range of analytical techniques. The restriction of intensive deformation to
shear zones during fluid^rock interaction preserves
markers for pseudomorphism and the original spatial relationships between phases, aiding the determination of the
different processes that took place. Mineral tracers, such
as magnetite and serpentine in the DPOs, are critical
when deciphering the 3-D parallelism of the DPO olivine
structure and their isolation within the surrounding mesh
texture. The determination of the dehydration event demonstrates that although the microstructures can provide
clues to the rock history, further research with respect to
isotope and trace element analysis is required. It is crucial
to identify the origin of fluid(s) that induced serpentinization, to relate the different serpentinization episodes to
possible infiltration of circulating seawater, released subduction zone fluids or fluids infiltrating during
Caledonian regional metamorphism and thus further unravel the origin of the dehydration event.
Although our understanding of the mineralogical
changes observed in this study is restricted to the investigated area, it is a logical progression that transient dehydration could be a much more widespread phenomenon
within the oceanic lithospheric mantle than is currently
acknowledged. Furthermore, the survival of microstructures and minerals during succeeding metamorphic
events results in complex structures that are prone to misinterpretation. In particular, fluid-mediated (pseudomorphic) replacement reactions that have gone to
completion will probably leave only cryptic traces of their
possibly long and complex history.
AC K N O W L E D G E M E N T S
We thank B. W. Evans, K. A. Evans and T. Nozaka for their
constructive reviews, and B. R. Frost for editorial management. The paper greatly benefited from discussions with
H. E. King, J. Krause, A. Putnis, T. B. Andersen, B.
Jamtveit and the Delta-Min team, who we thank for stimulating thoughts about mineral replacement reactions. We
thank M. Erambert for technical assistance, A. Rath for
visualization assistance, and K. Iyer for providing his
sample collection. We thank A. Jones of the Australian
Centre for Microscopy and Microanalysis (ACMM,
University of Sydney, Australia) for help during the X-ray
tomography data acquisition. O. Plu«mper acknowledges
his scientific mobility opportunities. This is Contribution
45 from the ARC Centre of Excellence for Core to Crust
Fluid Systems (www.ccfs.mq.edu.au) and Contribution
810 in the GEMOC Key Centre (www.gemoc.mq.edu.au).
FU N DI NG
This research was funded by the European Commission
through the EU Early Stage Training Network Delta-Min
(Mechanisms of Mineral Replacement Reactions; www
.delta-min.com) contract no. PITN-GA-2008-215360. S.
Piazolo acknowledges the Macquarie University New
1965
JOURNAL OF PETROLOGY
VOLUME 53
Staff Grant. The Knut och Alice Wallenberg stiftelse is
thanked for funding for the used SEM set-up.
S U P P L E M E N TA RY DATA
Supplementary data for this paper are available at Journal
of Petrology online.
R EF ER ENC ES
Ague, J. J., Park, J. & Rye, D. M. (1998). Regional metamorphic dehydration and seismic hazard. Geophysical Research Letters 25,
4221^4224.
Aikawa, N. & Tokonami, M. (1987). Crystal structure and cation distribution of ‘cleavable’ olivines. Mineralogical Journal 13, 271^279.
Albrektsen, B. A., Furnes, H. & Pedersen, R. B. (1991). Formation of
dunites in mantle tectonites, Leka Ophiolite Complex, Norway.
Journal of Geodynamics 13, 205^220.
Allen, D. E. & Seyfried, W. E. (2004). Serpentinization and heat generation: Constraints from Lost City and Rainbow hydrothermal
systems. Geochimica et Cosmochimica Acta 68, 1347^1354.
Andreani, M., Mevel, C., Boullier, A. M. & Escartin, J. (2007).
Dynamic control on serpentine crystallization in veins:
Constraints on hydration processes in oceanic peridotites.
Geochemistry, Geophysics, Geosystems 8, doi:10.1029/2006GC001373.
Angiboust, S., Agard, P., Raimbourg, H., Yamato, P. & Huet, B. (2011).
Subduction interface processes recorded by eclogite-facies shear
zones (Monviso, W. Alps). Lithos 127, 222^238.
Arndt, N. T., Guitreau, M., Boullier, A. M., Le Roex, A.,
Tommasi, A., Cordier, P. & Sobolev, A. (2010). Olivine, and the
origin of kimberlite. Journal of Petrology 51, 573^602.
Behn, M. D., Boettcher, M. S. & Hirth, G. (2007). Thermal structure
of oceanic transform faults. Geology 35, 307^310.
Beinlich, A., Klemd, R., John, T. & Gao, J. (2010). Trace-element mobilization during Ca-metasomatism along a major fluid conduit:
Eclogitization of blueschist as a consequence of fluid^rock interaction. Geochimica et Cosmochimica Acta 74, 1892^1922.
Bestmann, M. & Prior, D. J. (2003). Intragranular dynamic recrystallization in naturally deformed calcite marble: diffusion accommodated grain boundary sliding as a result of subgrain rotation
recrystallization. Journal of Structural Geology 25, 1597^1613.
Bowin, C. O., Nalwalk, A. J. & Hersey, J. B. (1966). Serpentinized
peridotite from the north wall of the Puerto Rico Trench. Geological
Society of America Bulletin 77, 257^269.
Bucher-Nurminen, K. (1991). Mantle fragments in the Scandinavian
Caledonides. Tectonophysics 190, 173^192.
Buseck, P. R., Nord, G. L. & Veblen, D. R. (1980). Subsolidus phenomena in pyroxenes. In: Prewitt, C. T. (ed.) Pyroxenes. Reviews in
Mineralogy. Mineralogical Society of America 7, 177^211.
Cannat, M., Bideau, D. & Bougault, H. (1992). Serpentinized peridotites and gabbros in the Mid-Atlantic Ridge Axial Valley at
15837’N and 16852’N. Earth and Planetary Science Letters 109, 87^106.
Carmichael, D. M. (1969). On the mechanism of prograde metamorphic reactions in quartz-bearing pelitic rocks. Contributions to
Mineralogy and Petrology 20, 244^267.
Caruso, L. & Chernosky, J. V. (1979). The stability of lizardite.
Canadian Mineralogist 17, 757^769.
Ceuleneer, G., Nicolas, A. & Boudier, F. (1988). Mantle flow patterns
at an oceanic spreading center: The Oman peridotites record.
Tectonophysics 151, 1^26.
NUMBER 9
SEPTEMBER 2012
Dawson, P., Hadfield, C. D. & Wilkinson, G. R. (1973). Polarized infrared and Raman spectra of Mg(OH)2 and Ca(OH)2. Journal of
Physics and Chemistry of Solids 34, 1217^1225.
Deschamps, F., Guillot, S., Godard, M., Andreani, M. & Hattori, K.
(2011). Serpentinites act as sponges for fluid-mobile elements in
abyssal and subduction zone environments. Terra Nova 23, 171^178.
Dick, H. J. B. & Bullen, T. (1984). Chromian spinel as a petrogenetic
indicator in abyssal and Alpine-type peridotites and spatially associated lavas. Contributions to Mineralogy and Petrology 86, 54^76.
Dijkstra, A. H., Sergeev, D. S., Spandler, C., Pettke, T., Meisel, T. &
Cawood, P. A. (2010). Highly refractory peridotites on Macquarie
Island and the case for anciently depleted domains in the Earth’s
mantle. Journal of Petrology 51, 469^493.
Dobrzhinetskaya, L. F., Wirth, R., Rhede, D., Liu, Z. & Green, H. W.
(2009). Phlogopite and quartz lamellae in diamond-bearing
diopside from marbles of the Kokchetav massif, Kazakhstan:
exsolution or replacement reaction? Journal of Metamorphic Geology
27, 607^620.
Downs, R. T. (2006). The RRUFF Project: an integrated study of the
chemistry, crystallography, Raman and infrared spectroscopy of
minerals. Program and Abstracts of the 19th General Meeting of the
International Mineralogical Association in Kobe, Japan, O03^13.
Duan, Z. H., Hu, J. W., Li, D. D. & Mao, S. D. (2008). Densities of
the CO2^H2O and CO2^H2O^NaCl systems up to 647 K and 100
MPa. Energy and Fuels 22, 1666^1674.
Dubinska, E., Bylina, P., Kozlowski, A., Dorr, W., Nejbert, K.,
Schastok, J. & Kulicki, C. (2004). U^Pb dating of serpentinization:
hydrothermal zircon from a metasomatic rodingite shell (Sudetic
ophiolite, SW Poland). Chemical Geology 203, 183^203.
Dungan, M. A. (1979). Bastite pseudomorphs after orthopyroxene,
clinopyroxene and tremolite. Canadian Mineralogist 17, 729^740.
Dunning, G. R. & Pedersen, R. B. (1988). U/Pb ages of ophiolites and
arc-related plutons of the Norwegian Caledonidesçimplications
for the development of Iapetus. Contributions to Mineralogy and
Petrology 98, 13^23.
Escartin, J., Hirth, G. & Evans, B. (2001). Strength of slightly serpentinized peridotites: Implications for the tectonics of oceanic lithosphere. Geology 29, 1023^1026.
Evans, B. W. (1977). Metamorphism of alpine peridotite and serpentinite. Annual Review of Earth and Planetary Sciences 5, 397^447.
Evans, B. W. (2004). The serpentinite multisystem revisited: Chrysotile
is metastable. International Geology Review 46, 479^506.
Evans, B. W. (2010). Lizardite versus antigorite serpentinite: magnetite,
hydrogen, and life(?). Geology 38, 879^882.
Ferry, J. M. (2000). Patterns of mineral occurrence in metamorphic
rocks. American Mineralogist 85, 1573^1588.
Frangi, A. F., Niessen, W. J., Vincken, K. L. & Viergever, M. A. (1998).
Multiscale vessel enhancement filtering. Medical Image Computing
and Computer-Assisted InterventionçMiccai’98 1496, 130^137.
Frost, B. R. (1975). Contact metamorphism of serpentinite, chloritic blackwall and rodingite at Paddy-Go-Easy Pass, Central
Cascades, Washington. Journal of Petrology 16, 272^313.
Frost, B. R. & Beard, J. S. (2007). On silica activity and serpentinization. Journal of Petrology 48, 1351^1368.
Furnes, H., Pedersen, R. B. & Stillman, C. J. (1988). The Leka
Ophiolite Complex, central Norwegian Caledonidesçfield characteristics and geotectonic significance. Journal of the Geological Society,
London 145, 401^412.
Furnes, H., Pedersen, R. B., Hertogen, J. & Albrektsen, B. A. (1992).
Magma development of the Leka Ophiolite Complex, Central
Norwegian Caledonides. Lithos 27, 259^277.
Gee, D. G., Kumpulainen, R., Roberts, D., Stephens, M. B.,
Thon, A. & Zachrisson, E. (1985). Scandinavian Caledonides;
1966
PLU«MPER et al.
OLIVINE PSEUDOMORPHS, LEKA, NORWAY
tectonostratigraphic map. In: Gee, D. G. & Sturt, B. A. (eds)
The Caledonide OrogençScandinavia and Related Areas. Chichester:
Wiley.
Groppo, C., Rinaudo, C., Cairo, S. & Gastaldi, D. (2006). MicroRaman spectroscopy for a quick and reliable identification of serpentine minerals from ultramafics. European Journal of Mineralogy
18, 319^329.
Hacker, B. R., Peacock, S. M., Abers, G. A. & Holloway, S. D. (2003).
Subduction factoryç2. Are intermediate-depth earthquakes in
subducting slabs linked to metamorphic dehydration reactions?
Journal of Geophysical ResearchçSolid Earth 108, doi:10.1029/
2001JB001129.
Hardie, L. A. (1967). Gypsum^anhydrite equilibrium at one atmosphere pressure. American Mineralogist 52, 171^200.
Hattori, K. H. & Guillot, S. (2007). Geochemical character of serpentinites associated with high- to ultrahigh-pressure metamorphic
rocks in the Alps, Cuba, and the Himalayas: Recycling of elements
in subduction zones. Geochemistry, Geophysics, Geosystems 8,
doi:10.1029/2007GC001594.
Hawkes, H. E. (1946). Olivine from northern California showing
perfect cleavage. American Mineralogist 31, 276^283.
Hyndman, R. D. & Peacock, S. M. (2003). Serpentinization of the
forearc mantle. Earth and Planetary Science Letters 212, 417^432.
Ishii, T., Robinson, P. T., Maekawa, H. & Fiske, R. (1992). Petrological
studies of peridotites from diapiric serpentinite seamounts in the
Izu^Ogasawara^Mariana forearc, Leg 125. In: Fryer, P., Pearce, J.
A. & Stokking, L. B. et al. (eds) Proceedings of the Ocean Drilling
Program, Scientific Results, 125. College Station, TX: Ocean Drilling
Program, pp. 445^486.
Iyer, K., Austrheim, H., John, T. & Jamtveit, B. (2008a).
Serpentinization of the oceanic lithosphere and some geochemical
consequences: Constraints from the Leka Ophiolite Complex,
Norway. Chemical Geology 249, 66^90.
Iyer, K., Jamtveit, B., Mathiesen, J., Malthe-Sorenssen, A. & Feder, J.
(2008b). Reaction-assisted hierarchical fracturing during serpentinization. Earth and Planetary Science Letters 267, 503^516.
Iyer, K., Rupke, L. H. & Morgan, J. P. (2010). Feedbacks between
mantle hydration and hydrothermal convection at ocean spreading
centers. Earth and Planetary Science Letters 296, 34^44.
Karato, S. (1987). Scanning electron-microscope observation of
dislocations in olivine. Physics and Chemistry of Minerals 14,
245^248.
Kelemen, P. B. (1990). Reaction between ultramafic rock and fractionating basaltic magma. 1. Phase relations, the origin of calcalkaline magma series, and the formation of discordant dunite.
Journal of Petrology 31, 51^98.
Kimball, K. L., Spear, F. S. & Dick, H. J. B. (1985). High-temperature
alteration of abyssal ultramafics from the Islas Orcadas Fracture
Zone, South Atlantic. Contributions to Mineralogy and Petrology 91,
307^320.
Klein, F., Bach, W., Jons, N., McCollom, T., Moskowitz, B. &
Berquo, T. (2009). Iron partitioning and hydrogen generation
during serpentinization of abyssal peridotites from 158N on the
Mid-Atlantic Ridge. Geochimica et Cosmochimica Acta 73, 6868^6893.
Kohlstedt, D. L., Goetze, C., Durham, W. B. & Vander Sande, J.
(1976). New technique for decorating dislocations in olivine. Science
191(4231), 1045^1046.
Kuroda, Y. & Shimoda, S. (1967). Olivine with well-developed cleavagesçits geological and mineralogical meanings. Journal of the
Geological Society of Japan 73, 377^388.
Kutty, T. R. N., Subbanna, G. N. & Iyer, G. V. A. (1983). Electronmicroscopy of olivine with perfect cleavage. Proceedings of the Indian
Academy of SciencesçEarth and Planetary Sciences 92, 283^295.
Li, X. P., Rahn, M. & Bucher, K. (2004). Serpentinites of the
Zermatt^Saas ophiolite complex and their texture evolution.
Journal of Metamorphic Geology 22, 159^177.
Lister, G. S. & Snoke, A. W. (1984). S^C mylonites. Journal of Structural
Geology 6, 617^638.
Maale, S. (2005). The dunite bodies, websterite and orthopyroxenite
dikes of the Leka Ophiolite Complex, Norway. Mineralogy and
Petrology 85, 163^204.
Markl, G., Marks, M. & Wirth, R. (2001). The influence of T, a(SiO2),
and f(O2) on exsolution textures in Fe^Mg olivine: An example
from augite syenites of the Ilimaussaq Intrusion, South
Greenland. American Mineralogist 86, 36^46.
McCollom, T. M. & Bach, W. G. (2009). Thermodynamic constraints
on hydrogen generation during serpentinization of ultramafic
rocks. Geochimica et Cosmochimica Acta 73, 856^875.
Mercier, J. C. C. & Nicolas, A. (1975). Textures and fabrics of
upper-mantle peridotites as illustrated by xenoliths from basalts.
Journal of Petrology 16, 454^487.
Mu«ller, W. F. & Horneman, U. (1969). Shock-induced planar deformation structures in experimentally shock-loaded olivines and in
olivines from chondritic meteorites. Earth and Planetary Science
Letters 7, 251^264.
Murata, K., Maekawa, H., Yokose, H., Yamamoto, K., Fujioka, K.,
Ishii, T., Chiba, H. & Wada, Y. (2009). Significance of serpentinization of wedge mantle peridotites beneath Mariana forearc,
Western Pacific. Geosphere 5, 90^104.
Newton, R. C. & Manning, C. E. (2008). Solubility of corundum in
the system Al2O3^SiO2^H2O^NaCl at 8008C and 10 kbar.
Chemical Geology 249, 250^261.
Nozaka, T. & Ito, Y. (2011). Cleavable olivine in serpentinite mylonites
from the Oeyama ophiolite. Journal of Mineralogical and Petrological
Sciences 106, 36^50.
O’Hanley, D. S. (1996). Serpentinites: Records of Tectonic and Petrological
History. Oxford: Oxford University Press.
Ohara, Y. & Ishii, T. (1998). Peridotites from the southern Mariana
forearc: Heterogeneous fluid supply in mantle wedge. Island Arc 7,
541^558.
Padron-Navarta, J. A., Tommasi, A., Garrido, C. J., SanchezVizcaino, V. L., Gomez-Pugnaire, M. T., Jabaloy, A. &
Vauchez, A. (2010). Fluid transfer into the wedge controlled by
high-pressure hydrofracturing in the cold top-slab mantle. Earth
and Planetary Science Letters 297, 271^286.
Padron-Navarta, J. A., Sanchez-Vizcaino, V. L., Garrido, C. J. &
Gomez-Pugnaire, M. T. (2011). Metamorphic record of highpressure dehydration of antigorite serpentinite to chlorite harzburgite in a subduction setting (Cerro del Almirez, Nevado^
Fila¤bride Complex, Southern Spain). Journal of Petrology 52,
2047^2078.
Peretti, A., Dubessy, J., Mullis, J., Frost, B. R. & Trommsdorff, V.
(1992). 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 112, 329^340.
Pinsent, R. H. & Hirst, D. M. (1977). The metamorphism of Blue
River Ultramafic Body, Cassiar, British Columbia, Canada.
Journal of Petrology 18, 567^594.
Plu«mper, O., King, H. E., Vollmer, C., Ramasse, Q., Jung, H. &
Austrheim, H. (2012). The legacy of crystal-plastic deformation in
olivine: high-diffusivity pathways during serpentinization.
Contributions to Mineralogy and Petrology 163(4), 701^724.
Pouchou, J. L. & Pichoir, F. (1984). A new model for quantitative
X-ray microanalysis. 1. application to the analysis of homogeneous
samples. Recherche Aerospatiale 3, 167^192.
1967
JOURNAL OF PETROLOGY
VOLUME 53
Prestvik, T. (1972). Alpine-type mafic and ultramafic rocks of Leka,
Nord-Trndelag. Norges Geologiske Underskelse 273, 23^34.
Prior, D. J., Wheeler, J., Peruzzo, L., Spiess, R. & Storey, C. (2002).
Some garnet microstructures: an illustration of the potential of
orientation maps and misorientation analysis in microstructural
studies. Journal of Structural Geology 24, 999^1011.
Proyer, A., Krenn, K. & Hoinkes, G. (2009). Oriented precipitates of
quartz and amphibole in clinopyroxene of metabasites from the
Greek Rhodope: a product of open system precipitation during
eclogite^granulite^amphibolite transition. Journal of Metamorphic
Geology 27, 639^654.
Putnis, A. (2009). Mineral replacement reactions. In: Oelkers, E. H. &
Schott, J. (eds) Thermodynamics and Kinetics of Water^Rock Interaction.
Mineralogical Society of America and Geochemical Society, Reviews in
Mineralogy and Geochemistry 70, 87^124.
Putnis, A. & Austrheim, H. (2010). Fluid-induced processes: metasomatism and metamorphism. Geofluids 10, 254^269.
Ragnarsdottir, K. V. & Walther, J. V. (1985). Experimental determination of corundum solubilities in pure water between 400^7008C
and 1^3 kbar. Geochimica et Cosmochimica Acta 49, 2109^2115.
Ren, Y. F., Chen, F. Y., Yang, J. S. & Gao, Y. H. (2008). Exsolutions of
diopside and magnetite in olivine from mantle dunite, Luobusa
ophiolite, Tibet, China. Acta Geologica Sinica, English Edition 82,
377^384.
Rinaudo, C., Gastaldi, D. & Belluso, E. (2003). Characterization of
chrysotile, antigorite and lizardite by FT-Raman spectroscopy.
Canadian Mineralogist 41, 883^890.
Ru«pke, L. H., Morgan, J. P., Hort, M. & Connolly, J. A. D. (2004).
Serpentine and the subduction zone water cycle. Earth and
Planetary Science Letters 223, 17^34.
Rutter, E. H. & Brodie, K. H. (1987). On the mechanisms of oceanic
transform faults. AnnalesTectonicae 1, 87^96.
Saumur, B. M., Hattori, K. H. & Guillot, S. (2010). Contrasting origins
of serpentinites in a subduction complex, northern Dominican
Republic. Geological Society of America Bulletin 122, 292^304.
Shervais, J. W. (2001). Birth, death, and resurrection: The life cycle of
suprasubduction zone ophiolites. Geochemistry, Geophysics, Geosystems
2, doi:10.1029/2000GC000080.
Soustelle, V., Tommasi, A., Demouchy, S. & Ionov, D. A. (2010).
Deformation and fluid^rock interaction in the supra-subduction
mantle: Microstructures and water contents in peridotite xenoliths
from the Avacha Volcano, Kamchatka. Journal of Petrology 51,
363^394.
Sturt, B. A., Andersen, T. B. & Furnes, H. (1985). The Skei Group,
Leka: an unconformable clastic sequence overlying the Leka
NUMBER 9
SEPTEMBER 2012
Ophiolite. In: Gee, D. G. & Sturt, B. A. (eds) The Caledonide
OrogençScandinavia and Related Areas. Chichester: John Wiley,
pp. 395^405.
Tagirov, B. & Schott, J. (2001). Aluminum speciation in crustal fluids
revisited. Geochimica et Cosmochimica Acta 65, 3965^3992.
Titus, S. J., Fossen, H., Pedersen, R. B., Vigneresse, J. L. & Tikoff, B.
(2002). Pull-apart formation and strike-slip partitioning in an obliquely divergent setting, Leka Ophiolite, Norway. Tectonophysics
354, 101^119.
Trommsdorff, V. & Evans, B. W. (1972). Progressive metamorphism
of antigorite schist in Bergell Tonalite Aureole (Italy). American
Journal of Science 272, 423^437.
Vance, J. A. & Dungan, M. A. (1977). Formation of peridotites by
deserpentinization in Darrington and Sultan Areas, Cascade
Mountains, Washington. Geological Society of America Bulletin 88,
1497^1508.
Viti, C. & Mellini, M. (1998). Mesh textures and bastites in the
Elba retrograde serpentinites. European Journal of Mineralogy 10,
1341^1359.
Viti, C., Mellini, M. & Rumori, C. (2005). Exsolution and hydration of
pyroxenes from partially serpentinized harzburgites. Mineralogical
Magazine 69, 491^507.
Walther, J. V. (1997). Experimental determination and interpretation of the solubility of corundum in H2O between 350 and
6008C from 0·5 to 2·2 kbar. Geochimica et Cosmochimica Acta 61,
4955^4964.
Wenner, D. B. & Taylor, H. P. (1973). Oxygen and hydrogen isotope
studies of serpentinization of ultramafic rocks in oceanic environments and continental ophiolite complexes. American Journal of
Science 273, 207^239.
Whitney, D. L. & Evans, B. W. (2010). Abbreviations for names of
rock-forming minerals. American Mineralogist 95, 185^187.
Wicks, F. J. & Plant, A. G. (1979). Electron-microprobe and X-raymicrobeam studies of serpentine textures. Canadian Mineralogist 17,
785^830.
Wicks, F. J. & Whittaker, E. J. W. (1977). Serpentine textures and serpentinization. Canadian Mineralogist 15, 459^488.
Wojdyr, M. (2010). Fityk: a general-purpose peak fitting program.
Journal of Applied Crystallography 43, 1126^1128.
Yaxley, G. M., Crawford, A. J. & Green, D. H. (1991). Evidence for
carbonatite metasomatism in spinel peridotite xenoliths from
Western Victoria, Australia. Earth and Planetary Science Letters 107,
305^317.
1968