Baldwin2017 Article PetrogenesisOfAlkalineBasalt-h

Contrib Mineral Petrol (2017) 172:43
DOI 10.1007/s00410-017-1362-0
ORIGINAL PAPER
Petrogenesis of alkaline basalt‑hosted sapphire megacrysts.

Petrological and geochemical investigations of in situ sapphire
occurrences from the Siebengebirge Volcanic Field, Germany
L. C. Baldwin1,2 · F. Tomaschek1 · C. Ballhaus1 · A. Gerdes3 · R. O. C. Fonseca1 ·
R. Wirth4 · T. Geisler1 · T. Nagel5
Received: 13 January 2017 / Accepted: 18 April 2017 / Published online: 23 May 2017
© The Author(s) 2017. This article is an open access publication
Abstract Megacrystic sapphires are frequently associated including carbonates, members of the Pyrochlore, Betafite
with alkaline basalts, most notably in Asia and Australia, and Columbite Groupe minerals, as well as a high abun-
although basalt is not generally normative in corundum. dance of HFSE and of gaseous low-density CO2 inclusions
Most of these sapphire occurrences are located in alluvial support a parentage of a highly evolved, MgO and FeO
or eluvial deposits, making it difficult to study the enig- deficient carbonatitic melt. We identified CO2 to be the link
matic relationship between the sapphires and their host between alkaline basaltic volcanism and the xenocrystic
rocks. Here, we present detailed petrological and geochemi- sapphires. Only alkaline volcanic suites can build up enough
cal investigations of in situ megacrystic sapphires within CO2 in this magma chamber upon fractionation so that at
alkaline basalts from the Cenozoic Siebengebirge Volcanic high degrees of fractionation a carbonatitic melt exsolves
Field (SVF) in Germany. Markedly, the sapphires show which in turn can crystallize sapphires.
several micrometer thick spinel coronas at the contact with
the host basalt, indicating chemical disequilibrium between Keywords Sapphire · Alkaline basalt · Siebengebirge
the sapphire and the basaltic melt, supporting a xenogenetic Volcanic Field · Columbite Group mineral dating ·
relationship. However, in situ U–Pb dating of a Columbite Carbonatite
Group inclusion within one Siebengebirge sapphire using
laser ablation–inductively coupled plasma–mass spectrom-
etry (LA–ICP–MS) indicates a close genetic relationship Introduction
between sapphire crystallization and alkaline mafic vol-
canism in the SVF. The syngenetic mineral inclusion suite Sapphire and ruby are gem varieties of corundum
(α-Al2O3) and belong to the most prominent and precious
gemstones, along with diamond and emerald. Deposits
Communicated by Timothy L. Grove. of gem-quality corundum are well distributed around the
world (e.g., Hughes 1997). A review of the literature on
* L. C. Baldwin gem corundum shows that corundum occurs in various
[email protected]
geological settings and seems to be generated via several
1
Steinmann Institute, University of Bonn, Meckenheimer different petrogenetic processes, such as metamorphic
Allee 169, 53111 Bonn, Germany reactions or through igneous crystallization from alumina-
2
Department of Lithospheric Research, University of Vienna, rich magmas (see references in this work and references
Althanstraße 14, 1090 Vienna, Austria therein). As to the broad variability in the depositional
3
Geozentrum, Goethe University Frankfurt, Altenhöferalle 1, type of corundum, Simonet et al. (2008) established a clas-
60438 Frankfurt, Germany sification scheme according to their geological occurrence,
4
Helmholz Zentrum Potsdam, Telegrafenberg, distinguishing between primary and secondary corundum
14473 Potsdam, Germany deposits. Primary deposits are those where corundum is
5
Department of Geoscience, University of Aarhus, entrained inside the same rock that it crystallized from. In
Høegh‑Guldbergs Gade 2, 8000 Aarhus C, Denmark secondary deposits, corundum is an inherited mineral that
13
43
Page 2 of 27 Contrib Mineral Petrol (2017) 172:43
formed in a different petrogenetic setting as to where it is but only rarely in SiO2 saturated rocks (e.g., Berger and
deposited. Berg 2006; Berg 2007, 2014; Palke et al. 2017). The petro-
Of special interest are secondary corundum deposits genesis of alkaline basalt-hosted sapphires was intensively
that are associated with rift-related alkaline basaltic vol- studied from various gem fields in the 1990s and early
canism as these are among the world’s largest gem corun- 2000s. These studies revealed that the sapphires that were
dum suppliers (e.g., Guo et al. 1996; Limtrakun et al. 2001; brought to the surface by basaltic eruptions may have vari-
Shore and Weldon 2009) and are hence economically of ous pre-depositional histories and may be of metamorphic,
great importance. Such basaltic secondary deposits are magmatic, or of metasomatic origin (e.g., Sutherland et al.
located all over the world, most notably in Asia (e.g., Guo 2009a; Giuliani et al. 2014), depending on the δ18O values
et al. 1996, Giuliani et al. 2015), and in eastern Australia, (e.g., Giuliani et al. 2005, 2007; Yui et al. 2006; Sutherland
at the western Pacific margin (e.g., Coenraads 1990; Gra- et al. 2009a, b), the mineral inclusion suite (e.g., Guo et al.
ham et al. 2008; Sutherland 1996; Sutherland et al. 1998a, 1996), their trace element geochemistry (Sutherland et al.
2002, 2009b, 2015, 2017; Saminpanya 2001; McGee 2005; 2009b, 2017), and most notably on the Ga/Mg ratio (Peucat
Abduriyim et al. 2012), but also in Europe (e.g., Malikova et al. 2007).
1999; Giuliani et al. 2009; Uher et al. 2012), Africa (Kief- Several hypotheses have been proposed to explain the
ert and Schmetzer 1987; Krzemnicki et al. 1996; Schwarz igneous crystallization of sapphires, the most prominent
et al. 2000; Pardieux et al. 2014), in South America being: (1) Crystallization from highly evolved melts such
(Sutherland et al. 2008), and in North America (e.g., Gar- as syenites (e.g., Irving 1986; Coenraads et al. 1990; Aspen
land 2002; Berg 2007, 2014; Palke et al. 2017). In most et al. 1990; Garnier et al. 2005; Giuliani et al. 2009). (2)
basalt fields, corundum is present as blue, green, or yellow Crystallization from alkaline Si- and Al-rich melts formed
sapphires, and is commonly referred to as blue–green–yel- by partial melting of amphibole-bearing lithospheric man-
low (BGY) sapphire suite (e.g., Sutherland et al. 1998a, tle (Sutherland 1996; Sutherland et al. 1998b). (3) Growth
b). These BGY sapphires may reach up to 5 cm in diam- as restitic component during partial melting of Al-rich
eter and usually occur in alluvial or eluvial placer depos- crustal protoliths such as anorthosites (Palke et al. 2016,
its. In situ occurrences where the sapphires are still inside 2017). (4) Hybrid formation where sapphire is stabilized
their host basalts are rarely found (e.g., MacNevin 1972; by reaction of a fractionated silicate with a carbonatitic
Stephenson 1976; Vichit 1987). The best known example melt at mid-crustal levels (Guo et al. 1996). This hybrid
is a basalt flow at Changle, China (Guo et al. 1992; Zelong model is based on the occurrence of a very complex min-
et al. 2007; Giuliani et al. 2015), while numerous corun- eral inclusion suite in sapphire that must have precipitated
dum, including gem crystals, were recorded in extensive from melts of two different compositions, most probably a
flows south of Jos Plateau, Nigeria (Wright 1985). carbonatitic and a highly evolved silicate melt. Guo et al.
Although this type of corundum deposit is economically (1996) proposed that evolved granite/syenitic pegmatite
the most valuable amongst corundum deposits, the origin silicate melts become Al2O3 oversaturated in response to
of these corundum is to date still poorly understood (e.g., mixing with carbonatitic melts, resulting in local precipi-
Izokh et al. 2010; Uher et al. 2012). The problem is that tation of corundum in the hybridization zone. (5) Crystal-
primitive alkali basalts are not corundum-normative when lization at shallow crustal levels from an iron-rich syen-
they leave their mantle sources, regardless of the pressure itic melt with the participation of a carbonate–H2O–CO2
of formation (Liu and Presnall 1990). Hence, unless these fluid phase (Izokh et al. 2010). This model is similar to the
sapphires precipitated via some yet unknown process out model proposed by Guo et al. (1996) but does not include
of the basaltic melts, they must instead have a xenogenetic the participation of a carbonatitic melt. Izokh et al. (2010)
relation with their host rocks. Indeed, anhedral crystal rather suggested that the syenitic melt must have had ini-
shapes, rounded edges, and macroscopically visible spi- tially high C O2 contents, as it was formed by differentia-
nel coronas at the sapphire-basalt interface (e.g., Stephen- tion of a CO2-rich basaltic melt. Differentiation of these
son 1976; Guo et al. 1996; Sutherland et al. 2002; McGee CO2 rich syenitic magmas led to the formation of alkali–
2005) indicate disequilibrium between these sapphires and carbonate complexes, so that the residual melt becomes
their host rocks, supporting a xenocrystic nature. There oversaturated in alumina and corundum may precipitate in
seems to be a consensus (see references in this work) that the most fractionated magma chambers.
the sapphires originate from a reservoir in the mantle or the In this paper, we present a detailed microchemical inves-
crust, and were later transported to the surface by ascending tigation of up to 2 cm sized sapphire xenocrysts that occur
basaltic melts. Yet, the origin of the basalt-associated sap- within primitive alkali basalts and basanites from the Sie-
phires remains a puzzle. Moreover, it is still an open ques- bengebirge Volcanic Field (SVF) in Germany. The sap-
tion as to why sapphires most often occur as xenocrystals phires from the SVF are found in situ within alkali basalt,
in alkaline basalts from intra-continental volcanic fields, a relatively rare occurrence with the potential for providing
13
Contrib Mineral Petrol (2017) 172:43 Page 3 of 27 43
important insights into how this type of sapphire formed. To the Hessian Depression, the Rhön area, Heldburg, and the
date, there are only limited investigations of sapphire that Oberpfalz (Fig. 1). Further European regions belonging to
are still attached to their host rocks using modern analytical the CEVP are the Bohemian Massif in the Czech Republic,
techniques. Hence, the Siebengebirge sapphires provide an the Eger Rift System in the Czech Republic and Poland, the
exceptional opportunity to study the genetical relationship Pannonian Basin in Hungary, and the French Massif Cen-
of the sapphires and their host basalt. We present a unique tral. Igneous activity in the CEVP lasted throughout the
petrogenetic model for the origin of the Siebengebirge sap- entire Cenozoic (Wilson and Downes 1991).
phires that may provide important implications for the gen- The SVF is located at the SE margin of the Lower
eral origin of basalt-hosted sapphires and for predicting sites Rhine Graben and is part of the Rhine rift system that
of possible further deposits. formed in Eocene to Oligocene times as a consequence
of syn- to post-orogenic extension in the Alpine fore-
land (Ziegler 1992). The NW–SE-trending Lower Rhine
Geological setting Graben cuts into the Rhenish Shield that is a Variscan
tectonic block within the Rhenohercynian Zone and has
The Siebengebirge Volcanic Field (SVF) is part of the Cen- been uplifted since Eocene times (e.g., Illies et al. 1979;
tral European Volcanic Province (CEVP). The latter forms Ziegler 1992; Oncken et al. 1999).
a 200–300 km broad Cenozoic intra-continental volcanic The total volcanic area of the SVF covers an area of
belt north of the Alpine Orogen. In Germany, volcanic about 900 km2. The volcanic activity in the Siebengebirge
fields belonging to the CEVP are (from west to east) the occurred over several eruptional phases, resulting in the
Eifel, the Siebengebirge, the Westerwald, the Vogelsberg, eruption of a very broad spectrum of both SiO2 saturated
Fig. 1 Volcanic Fields of the

Central European Volcanic
Province (modified after Wede-
pohl et al. 1994). The main
sampling localities Petersberg,
Oelberg, and Jungfernberg are
shown in the boxbox at the bot-
tom of the figure
13
43
and SiO2 undersaturated mafic and felsic melts. The most Samples
primitive mafic volcanics are basanites to alkali basalts;
more felsic volcanics are SiO2 saturated latites to trachytes Most of the sapphires studied in this work belong to the
and SiO2 undersaturated phonotephrites to tephriphonolites mineralogical collection of the Geological and Mineral-
(Vieten et al. 1988; Kolb et al. 2012; Jung et al. 2012). The ogical Department of the University of Bonn and were
period of volcanic activity in the SVF has been dated using recovered in the early twentieth century when the basalt
the K/Ar and Ar/Ar chronometers and lasted from 30 to occurrences (now protected) were still subject to quarrying.
19 Ma (Todt and Lippolt 1980; Linthout et al. 2009; Przyb- Other samples were donated by local collectors. The prin-
yla 2013). The main magmatic activity is thought to have cipal localities in the Siebengebirge from which sapphires
occurred between 26 and 24 Ma (Przybyla 2013). were recovered are the Petersberg and Oelberg alkaline
Alkaline basalts from the SVF have long been known mafic rocks. Other known localities are the Jungfernberg
to host megacrystic sapphires. They were described sci- in Heisterbacher Rott at the northern part of the Siebenge-
entifically in the late nineteenth and early twentieth cen- birge, and Unkel at the south-western extension of the
tury (Brauns 1922, and references therein). However, until Siebengebirge.
now, no effort has been made to investigate the sapphires Samples studied included 11 sapphires with an anhe-
entrained in the alkaline mafic rocks from the SVF with dral to subhedral crystal shape and sizes of 5–20 mm.
modern techniques and to identify possible genetic links to Sample descriptions, localities, and host rock lith-
their host basalts. ologies are summarized in Table 1. Colors vary from a
Table 1 Petrographic description of the Siebengebirge sapphires, their host rock locality and locations within the Siebengebirge
Sample Host rock locality Host rock lithol- Color Size (mm) Shape Reaction rim Inclusions
ogy thickness (µm)
3 Jungfernberg Basanite pale-blue, milky 3 Irregular 70–80 Calcite, Nb-rich

rutile, alkali
feldspar, spinel,
apatite, ilmenite
15 Oelberg Alkali Basalt 10 Irregular 40–100 Calcite
26 Oelberg Alkali Basalt Dark blue 10 Irregular 50–100, Calcite, plagioclase
Partly missing series, alkali
feldspar,
Columbite Group,
thorite, thorianite,
apatite
KS21 Heisterbacher Rott Alkali Basalt a) Pale blue, 7 Irregular 10–50, –
milky; Partly missing
b) Dark blue, clear
OL25 Oelberg Hawaiite Pale blue, milky 4 Irregular 100–200 Calcite, Nb-rich
rutile, ilmenite-
pyrophanite
OL60 Oelberg Hawaiite Pale blue, milky 3 Irregular 100–150, one area Plagioclase series,
with 500 spinel, calcite
OL61 Oelberg Hawaiite Dark blue, clear 3 Irregular ca. 100 Plagioclase series
UN31 Unkel Alkali Basalt Pale blue, milky 5 Irregular 40–100 –
and dark blue
areas
PE21 Petersberg Alkali Basalt Pale blue, milky 17 Hypidiomorphic 100–200 Nb-rich
with gray area in microinclusions in
the center reaction rim
PE25 Petersberg Alkali Basalt Pale blue, milky 15 Hypidiomorphic 100–200 Alkali feldspar,
with gray area in Nb-rich
one edge microinclusions in
reaction rim
S52 Unknown – Dark blue 8 Irregular 10–300 Plagioclase series,
alkali feldspar,
calcite
13
milky pale-blue to a clear dark blue, and two sapphires quite sharp, but in their contact with the basaltic matrix
have large brown to gray domains (Fig. 2), distinguish- the spinel coronas become more ragged and porous, and
ing them macroscopically from the other samples. All tend to disintegrate to discrete grains (Fig. 3a). The basalt
sapphires were still attached to their host rocks and in direct contact with the spinel coronas is enriched in
showed evidence for chemical disequilibrium with their plagioclase and depleted in modal pyroxene and titano-
host basalt, including resorbed edges, rounded shapes, magnetite (Fig. 3b, c) compared to the average basaltic
and macroscopically visible dark green spinel coronas matrix further away from the sapphire. Most sapphires
(Fig. 3a, b). The contact between sapphire and spinel is contain mineral and/or fluid inclusions. Except for two
a b plagioclase dominated basalt

spinel rim
‚normal‘ basalt
HFSE enriched
sapphire
‚normal‘ blue sapphire
1 mm
Fig. 2 Sapphire PE25 and graphical interpretation. a Photograph rounding the sapphire is spinel. The spinel rim is surrounded by a
of sapphire PE25. b Graphical interpretation. The brownish area is corona of plagioclase-rich basalt that has about the same thickness as
chemically characterized by a higher enrichment in the HFSE (Nb, the spinel rim. The plagioclase enriched basaltic matrix is surrounded
Ta, Zr, and Hf) compared to the blue area. The transition from the by ‘normal’ basaltic matrix comprising plagioclase, pyroxene, and Ti-
brown towards the blue area is extremely sharp. The black rim sur- phases
a b
‚nor
mal
‘ ba
salt
plag ic m
iocl atri
ase x
rich
spin bas
el rim alt z
dis one
inte
gra
tion
spin
el ri
m
200µm sapp
hire
Fig.  3 a Backscattered electron image of the transition from the sap- that mingle in more or less equal fractions with the basaltic matrix.
phire to the basaltic matrix. b Graphical interpretation of a, empha- The basaltic matrix in direct contact with the spinel rim comprises
sizing the zonal structure of the spinel rim. The sapphire-spinel almost pure plagioclase. Fe–Mg phases such as pyroxene are miss-
interface is well-defined whereas the transition zone from the spinel ing. The plagioclase-rich basaltic matrix is followed by the ‘normal’
towards the basaltic melt occurs via a zone of disintegrated spinels basaltic matrix with abundant plagioclase, pyroxene, and Ti-phases
13
43
samples (Fig. 3d), the spinel coronas are virtually free of normalized using 27Al as the internal standard, assuming an
inclusions. average of 98.5 wt% Al2O3 as based on the EPMA analyses.
The NIST-SRM-610 and NIST-SRM-612 standard refer-
ence materials were used for calibration as external stand-
Analytical methods ards using the preferred reference values given by Jochum
et al. (2011). Isotopes monitored were 9Be, 23Na, 24Mg,
25
EPMA Mg, 39K, 47Ti, 51V, 52Cr, 53Cr, 55Mn, 57Fe, 66Zn, 68Zn, 69Ga,
88
Sr, 89Y, 90Zr, 93Nb, 118Sn, 120Sn, 178Hf, 180Hf, 181Ta, 182W,
184
The sapphires, their reaction rims, and mineral inclusions W, 186W, 232Th, and 238U. The doubly charged cation pro-
were analyzed for major and minor elements with a JEOL duction rate monitored with 137Ba++/137Ba was kept below
JXA 8900 electron probe microanalyzer equipped with a 2%.
tungsten cathode at the Steinmann Institute, University of One Columbite Group mineral inclusion was dated on
Bonn. All measurements were performed at 15 kV accelera- the basis of the U–Pb decay system, using in situ LA–
tion voltage and a probe current of 15 nA. The major ele- ICP–MS analysis. This inclusion was analyzed for U, Th,
ments were quantified on the Kα lines. Where possible, the and Pb isotopes by LA–ICP–MS at the Institute of Geo-
trace elements Nb, Ce, Y, Ta, and Zr were quantified on the sciences, Johann Wolfgang Goethe University Frankfurt,
Lα lines, while Th, U, and Pb were quantified on their Mα using a Thermo Scientific Element II sector field ICP-
lines. For the silicates, standards used for calibration are oli- MS coupled to a New Wave Research UP-213 ultravio-
vine for Mg, Fe, and Si, jadeite for Na, microcline for K, let laser system. Data were acquired in the analog mode
anorthite for Ca, augite for Al, rutile for Ti, MnFe metal for 238U and in the counting mode for all other isotopes
for Mn, and pure metal for each Cr, and Ni. For corundum with 20 s measurement on the background followed by
analysis, except for magnetite that was used for Fe, the sample ablation of 20 s. Prior to analysis, the sample sur-
same standards were employed. Oxides of the High Field face was cleaned with 3 pulses of pre-ablation. Square
Strength Elements (HFSE) Nb, Ta, Th, and U were analyzed laser spots with an edge length of 28 µm were used. The
at the same probe running conditions. Niobium, Ce, Y, Ta, laser was operated at a repetition rate of 6 Hz and an
and Zr were analyzed on the Lα lines. The Mα lines were energy fluence of 2 J/cm2 with an energy output of 50%.
used to measure Th, U, and Pb. Standards for the HFSE and The signal was tuned to obtain maximum sensitivity for
238
REE are pure metal for Nb and Ta, synthetic phosphates U and 206Pb, and to keep the oxide production moni-
CePO4, LaPO4, and YPO4 for Ce, La, and Y, zircon for Zr, tored as 254UO/238U below 3%. Measured masses were
202
natural apatite for P and natural U
O2, ThO2, and PbO for U, Hg, 204Hg, 204Pb, 206Pb, 207Pb, 208Pb, 232Th, 235U, and
238
Th, and Pb. Matrix corrections were done online using the U. Mercury was also measured because 204Hg can be a
ZAF correction routine. minor contaminant of the Ar carrier gas, potentially caus-
ing isobaric interferences on 204Pb.
XRF The primary standard used was the GJ-1 zircon,
reported by Jackson et al. (2004) to have a TIMS age of
The sapphire crystals were still inside their host rocks, 608 ± 0.4 Ma. Isotopic ratios were corrected for ablation-
offering the opportunity to classify the basalts that car- induced elemental fractionation offset between zircon matrix
ried them to the surface. After removal of visible altera- and columbite offset between zircon matrix and columbite
tion rims, around 100 g of basaltic material was crushed matrix using a columbite–tantalite solid solution (Coltan
and ground in an agate mill. Major elements were analyzed 139) from Madagascar that is being used as an in-house
from fused lithium fluoride-based glass disks using a PAN- standard for coltan LA–ICP–MS analyses by the Federal
alytical Axios wavelength dispersive XRF. Institute for Geosciences and Natural Resources, Germany
(Gäbler et al. 2011). The usability of Coltan 139 as a matrix
LA–ICP–MS matching standard for Columbite Group Mineral LA–ICP–
MS U–Pb dating was recently demonstrated by Che et al.
Trace elements in the sapphires were measured using a (2015). Coltan 007 from Kokobin, Ghana (Melcher et al.
Resonetics Resolution M50E 193 nm laser ablation sys- 2008, 2015) was used as a secondary standard to control
tem coupled to a Thermo Scientific X Series II quadrupole the accuracy and reproducibility of the measurements. For
inductively coupled plasma mass spectrometer (Q–ICP– Coltan 139, Melcher et al. (2015) reported TIMS ages of
MS). The laser was operated at a repetition rate of 15 Hz 505.4 ± 1.0 Ma (BGR, Hannover) and 506.6 ± 2.4 Ma
and a fluence of 7 J/cm2. Spot sizes were 100 µm in diam- (University of Toronto), and an LA–ICP–MS age of
eter, and 20–30 spots were quantified depending on grain 506.2 ± 5.0 Ma (University of Frankfurt). Coltan 007 has an
size and homogeneity of the sapphire. Count rates were age of 2079.6 ± 3.1 Ma (Melcher et al. 2008, 2015).
13
A total number of 45 unknowns were analyzed, using the one from the blue area—were cut, using the focused ion
standard bracketing method. Acquired data were then pro- beam (FIB) technique (Ohnishi et al. 1990; Ishitani et al.
cessed using an in-house MS Excel© spreadsheet (Gerdes 1990) where thin membranes of an area of interest are
and Zeh 2006, 2009). A 208Pb-based common lead correc- cut from a polished sample by sputtering with G a+ ions
tion was applied, as there is very low Th in the sample. accelerated to 30 keV. The FIB sections from the brown-
ish area were cut into two different crystallographic ori-
Raman spectroscopy entations, one near parallel to [0001] and another one
inclined to [0001] of the sapphire. The section of the blue
Raman spectroscopy was used to further characterize all area was cut near parallel to [0001]. Subsequently, the FIB
types of inclusions in the sapphire, and special attention sections were thinned with a Ga ion beam to a final thick-
was paid to those that are not exposed on the polished sur- ness of 25–30 nm to guarantee electron transparency. A
face, including solid inclusions as well as fluid inclusions detailed description of the principles of the FIB technique
and their daughter minerals. As fluid inclusions were not is given by Wirth (2004, 2009). The FIB sections were sub-
recognizable in polished sapphire mounts, for the Raman sequently analyzed with an FEI Tecnai F20 X-Twin TEM
analyses, polished thick sections with a thickness of with a Schottky field emitter as electron source. The instru-
~500 µm were prepared, so that the sapphires were trans- ment is equipped with a Fishione high-angle annular dark-
lucent. Unfortunately, some sapphires were too brittle for field (HAADF) detector allowing for Z-contrast sensitive
thick sectioning. Because of this, we cannot say for cer- imaging, an EDAX energy dispersive X-ray spectroscopic
tain that all the sapphires depicted here have inclusions, or system (EDS) for the identification of nano-scale chemi-
whether the inclusions shown here are representative of all cal compositions from the excited characteristic X-rays of
SVF sapphires. the sample, as well as a Gatan Tridiem energy filter and
Raman spectroscopy was also employed to distinguish an electron energy loss spectrometer (EELS) for spectrum
between calcium carbonate polymorphs in carbonate inclu- imaging.
sions. Raman spectra were acquired with a confocal Horiba
HR800 equipped with an Olympus BX41 optical micro-
scope operated in 180° backscatter geometry. The excita- Results
tion source was an Nd-YAG Laser with a wave length of
532.18 nm. The laser beam was focused on the target with Host rock composition and corundum mineral
a ×50 objective lense. Spectra were accumulated with an chemistry
exposure time of 30 s on each point within a frequency
range of 100–4000 cm−1. The laser power was 200 mW. Host rock compositions and their CIPW norms are pre-
sented in Table 2. All rocks are nepheline normative,
TEM though to various extends, ranging from 0.3 to 10 wt%
normative nepheline. In the Total Alkali vs. Silica (TAS)
One sapphire (sample PE25) was investigated with the diagram (Le Maitre 2002), most host rocks plot in the field
transmission electron microscope (TEM) at the Geo- for alkali basalts. One sapphire was entrained in a basanite,
ForschungsZentrum (GFZ) in Potsdam, Germany. Next and three samples lie in the trachybasalt (hawaiite) field.
to the typical blue colored areas, this sapphire has a gray The Siebengebirge sapphires include up to 2 wt% FeO
to brown colored domain (Fig. 2) that is characterized by and up to 0.8 wt% TiO2. The major and trace element com-
a significant enrichment in the HFSE up to two orders positions of the sapphires, as well as the major element
of magnitude higher than the HFSE content in the blue composition of the spinel coronas, are listed in Table 3.
domains (up to 5000 ppm Nb in the brownish domain). Next to Fe and Ti as the main impurities, all sapphires
Given that some sapphire domains contain up to 5000 ppm contain detectable amounts of Mg, Mn, Ga, V, Nb, W, and
HFSE, it would be interesting to determine whether these Th. Beryllium, Na, K, Cr, Zn, Zr, Ta, and Hf were detected
elements substitute into the corundum crystal lattice, or in few analyses, whereas Sr, Y, and U are usually below
whether they are present as nano-inclusions. Therefore, a detection limit. Remarkably, the highest concentrations of
combination of the focused ion beam (FIB) and the TEM the HFSE (up to 5000 ppm Nb and up to 2900 ppm Ta), as
techniques was applied to retrieve information about the well as of Th (up to 226 ppm) were detected in the brown
crystallographic orientation relations of the sapphire and to gray domains in samples PE21 and PE25 where the con-
any potential nano-inclusions at an atomic scale, reveal- centrations are up to two orders of magnitude higher than
ing the physicochemical state of the HFSE in the sap- in the blue domains of the same sapphire. The other ele-
phires. For this purpose, three electron-transparent foils ments are not concentrated at any preferred site within
(15 × 10 × 0.15 µm)—two from the brownish areas, and the sapphire. The Ga contents range from 48 to 256 ppm.
13
43
Table 2 XRF analyses of the host rocks of the Siebengebirge sapphires

3 15 26 KS 21 OL 25 OL 60 OL 61 UN 31 PE 21 PE 25
SiO2 43.7 46.1 45.4 46.5 46.9 46.3 46.0 46.2 46.2 46.6
Al2O3 12.6 14.3 13.8 14.8 14.2 14.3 14.2 14.2 13.9 14.0
Fe2O3 12.0 11.4 11.3 11.5 11.2 11.4 11.1 11.4 12.4 12.4
MnO 0.19 0.18 0.20 0.19 0.18 0.18 0.18 0.18 0.20 0.19
MgO 12.1 9.41 9.72 8.01 8.68 9.07 9.24 9.27 9.09 8.92
CaO 10.7 9.84 10.6 9.62 10.1 9.87 9.80 9.87 9.32 9.40
Na2O 3.10 3.35 3.03 2.75 3.60 3.80 3.66 3.58 3.05 3.26
K2O 1.21 1.38 1.37 1.37 1.40 1.44 1.43 1.35 1.15 1.19
TiO2 2.28 2.23 2.21 2.35 2.25 2.24 2.23 2.26 2.19 2.20
P2O5 0.55 0.46 0.52 0.41 0.46 0.45 0.45 0.46 0.45 0.44
SO3 0.16 0.17 0.10 0.15 0.13 0.18 0.07 0.15 0.17 0.19
L.O.I. 0.45 0.73 0.62 1.34 0.16 0.53 0.47 0.63 0.77 0.90
Sum 99.0 99.6 98.9 99.0 99.3 99.8 98.8 99.6 98.9 99.7
Rock type Basanite Alkali Alkali Alkali Hawaiite Hawaiite Hawaiite Alkali Alkali Alkali Basalt
Basalt Basalt Basalt Basalt Basalt
CIPW norm
Plagioclase 25.0 37.9 34.7 48.1 37.4 34.5 34.7 36.5 35.0 42.9
Orthoclase 7.33 8.33 8.33 8.39 8.45 8.69 8.69 8.16 8.1 7.21
Nepheline 10.1 5.9 6.46 0.21 6.54 8.49 8.4 7.08 7.21 2.82
Diopside 26.7 21.0 24.1 17.4 23.4 22.9 22.8 22.2 22.8 19.2
Olivine 23.0 19.4 19.0 18.4 16.9 18.0 18.2 18.7 19.3 20.3
Ilmenite 4.44 4.33 4.31 4.62 4.35 4.33 4.35 4.39 4.35 4.27
Magnetite 1.78 1.70 1.68 1.73 1.65 1.68 1.65 1.68 1.83 1.84
Apatite 1.32 1.09 1.23 0.97 1.09 1.07 1.07 1.09 1.07 1.04
Na2SO4 0.31 0.3 0.18 0.28 0.23 0.32 0.12 0.27 0.32 0.35
Total 99.9 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0
e3+/Fetot ratio of the basalt is 0.1

The CIPW norm was calculated on the assumption that the F
In terms of their Ga/Mg ratio and the total Fe concentra- of secondary origin if it is xenomorphic and restricted to
tion, the Siebengebirge sapphires can be divided into two cracks in the sapphire.
groups: (1) Sapphires with low Ga/Mg (<10) and high Fetot The Siebengebirge sapphires contain a broad spectrum
and (2) sapphires with high Ga/Mg (>10) ratios and lower of mineral inclusions, including silicates, carbonates, and
Fetot. a wide variety of Nb, Ta and Th oxides. While some sap-
phires are virtually inclusion free, others have a high abun-
Inclusions in sapphire dance and compositional variability among the mineral
inclusions. Notably, there is no correlation between the
Inclusions have high potential for providing important mineral inclusion suite and the minor and trace element
information on the origin of sapphires. If the inclusions compositions of the host sapphires. Interestingly, many
are primary, i.e., crystallized during growth of the sap- inclusions in the sapphires are carbonates, which have only
phires, then they may represent melts, fluids, or accessory been described previously in sapphires associated with
minerals that coexisted with the sapphires during crystal alkaline basaltic volcanism from the Bo Phloi gemfield in
growth. Secondary inclusions may provide information Kanchanaburi, western Thailand (Khamloet et al. 2014).
about a potential post-magmatic alteration of the sapphires. Furthermore, the U-bearing inclusions thorite, thorianite
Inclusions are interpreted to be primary, if they are com- and Pyrochlore and Columbite Group minerals are of great
pletely enclosed within the sapphire and are not located at interest, as they potentially allow to date the age of the
any visible cracks. An inclusion has to be regarded to be sapphire crystallization using the U–Pb chronometer. With
13
Table 3 Major element (EPMA) and trace element (LA–ICP–MS) composition of the Siebengebirge sapphires and spinel coronas
3 15 26 KS21 OL25 OL60
Crn Spl Crn Spl Crn Spl Crn Spl Crn Spl Crn Spl
wt%
SiO2 0.02 0.03 0.02 0.1 0.02 0.02 0.01 0.02 n.a. bdl 0.02 0.02
TiO2 0.10 0.11 0.13 0.09 0.12 0.07 0.18 0.23 0.11 0.12 0.06 0.08
Al2O3 97.6 65.4 98.1 65.6 99.5 65.6 98.6 64.8 99.4 65.5 99.7 65.4
FeO 1.17 16.8 2.02 18.5 0.87 18.9 1.52 21.1 1.57 20.6 0.43 17.8
MnO 0.01 0.27 0.01 0.28 0.02 0.3 0.01 0.27 0.01 0.33 bdl 0.24
MgO bdl 16.4 0.01 16.1 bdl 15.8 bdl 14.2 bdl 15.2 bdl 16.7
CaO bdl 0.02 bdl 0.02 0.01 0.01 0.02 0.04 bdl bdl 0.01 0.01
Na2O bdl bdl 0.01 0.02 0.01 0.01 0.02 bdl bdl bdl 0.03 0.01
K2O bdl bdl bdl 0.01 0.01 bdl bdl bdl bdl n.a. 0.01 bdl
Cr2O3 bdl 0.03 bdl 0.02 bdl 0.01 0.01 0.01 0.01 0.01 0.02 0.01
Total 98.9 99.0 100.3 100.6 100.6 100.7 100.4 100.6 101.1 101.7 100.2 100.3
n 3 4 3 5 3 7 2 3 20 10 3 5
Spinel 63 60 60 55 57 64
Hercynite 37 40 40 45 43 36
ppm (n = 32) (n = 30) (n = 31) (n = 32) (n = 30) (n = 16)
Na 15–198 4–144 8–373 bdl–144 bdl–359 8–1007
K 6–94 10–1152 12–283 bdl bdl–222 14–39
Be 0.06–2.3 0.06–7 0.1–2 bdl–3 bdl–2 0.3–26
Mg 22–10,803 39–150 5–253 4–39 7–90 2–860
Sr bdl bdl bdl bdl bdl bdl
Ti 158–2880 171–1182 78–1292 158–1316 76–5154 60–659
V 8–15 7–17 2–12 3–15 5–19 2–12
Cr bdl 1–15 2–8 bdl–2 bdl–2 0.8–2
Mn 1.2–212 2–29 4–92 1–13 bdl–49 3–18
Fe 5559–18295 8565–14,958 2967–16,746 4599–15,967 3177–12,122 1875–8549
Zn 1.9–212 2–10 2–9 bdl–714 bdl–10 2–11
Ga 127–238 91–138 149–256 135–252 87–147 135–210
Sn 0.13–9 0.1–1.8 0.1–2 0.2–20 0.1–2 bdl–10
Y bdl bdl bdl bdl bdl–0.2 bdl
Zr 0.1–1.94 bdl bdl bdl–4 bdl–0.6 bdl
Nb 0.75–706 0.2–57 0.5–710 bdl–806 0.6–96 0.3–814
Ta 0.43–362 bdl–26.4 0.1–25 0.03–342 bdl–46 0.2–370
ppb
Hf 10–600 10–70 10–130 bdl–600 bdl–0.1 bdl–200
W 10–290 10–130 10–560 bdl–1735 bdl–750 bdl–600
Th 10–7030 10–850 bdl–890 bdl–14 (ppm) 10–5310 10–9960
U bdl–10 bdl–200 bdl–470 bdl–20 bdl–120 bdl–20
Ratios 2σ Ratios 2σ Ratios 2σ Ratios 2σ Ratios 2σ Ratios 2σ
Ga/Mg 3.6 4.1 1.5 0.8 17.1 16.2 16.0 31.4 6.3 7.9 23.4 57.8
OL61 UN31 PE21 PE25 S52
Crn Spl Crn Spl Crn Spl Crn Spl Crn Spl
wt%
SiO2 0.02 0.01 0.01 0.01 0.02 0.01 0.02 0.05 0.02 0.02
TiO2 0.03 0.1 0.2 0.14 0.75 0.27 0.25 0.15 0.1 0.08
13
43
Table 3 continued
OL61 UN31 PE21 PE25 S52
Crn Spl Crn Spl Crn Spl Crn Spl Crn Spl
Al2O3 97.8 64.7 98.2 65.0 98.1 63.7 99.4 65.1 97.8 65.1
FeO 1.31 17.0 1.46 18.8 0.86 20.7 0.92 21.3 1.67 18.2
MnO 0.02 0.21 bdl 0.26 0.01 0.26 0.02 0.3 bdl 0.28
MgO 0.01 16.1 0.01 15.6 bdl 14.2 0.01 14.1 0.01 15.9
CaO 0.01 0.01 0.01 0.01 0.01 0.03 0.01 0.02 0.01 0.02
Na2O 0.01 0.01 bdl 0.01 0.01 bdl 0.01 0.02 bdl 0.02
K2O 0.01 bdl bdl bdl 0.01 0.01 0.01 0.01 0.01 bdl
Cr2O3 0.01 0.01 bdl bdl n.d. 0.02 0.02 0.02 0.01 0.01
Total 99.2 98.1 99.9 99.8 99.8 99.2 100.7 101.0 99.7 99.6
n 2 3 3 4 3 5 3 2 3 5
Spinel 63 60 56 54 61
Hercynite 37 40 44 46 39
ppm (n = 30) (n = 30) (n = 30) (n = 36) (n = 31)
Na bdl–469 bdl–60 bdl–79 bdl–149 bdl–1644
K bdl–732 bdl bdl–52 bdl–109 bdl–427
Be bdl–0.7 bdl–1 bdl–10 0.7–11 bdl–0.8
Mg 10–45 61–174 10–34 13–243 42–152
Sr bdl–8 bdl bdl–0.2 bdl–0.2 bdl–11
Ti 75–1205 641–2149 1045–6231 876–6226 150–490
V 6–10 28–66 13–37 12–29 6–12
Cr bdl–4 bdl–22 bdl–20 bdl–21 bdl–5
Mn bdl–10 bdl–3.7 bdl–196 bdl–5 bdl–33
Fe 4598–10,728 6545–12,701 3301–6693 3853–6168 6519–10,434
Zn 2–14 2 -11 bdl–5 bdl–7 bdl–5
Ga 107–167 133–205 113–193 116–188 48–90
Sn bdl–0.6 0.3–9 2–21 0.4–14 bdl–0.1
Y bdl bdl bdl–0.5 bdl–0.5 bdl–4
Zr bdl–0.6 bdl–3 0.4–22 0.3–22 bdl–1
Nb bdl–9 bdl–748 3–3998 12–2771 bdl–18
Ta 0.1–4 0.1–169 4–1699 8–1011 bdl–1992
ppb
Hf bdl–35 bdl–552 113–4825 33–4085 bdl–14
W bdl–376 bdl–1770 995–41 (ppm) 117–23 (ppm) bdl–388
Th bdl–74 bdl–21.3 (ppm) 1–166 (ppm) 0.6–226 (ppm) bdl–3264
U bdl–24 bdl–40 bdl–419 bdl–0.6 bdl–741
Ratios 2σ Ratios 2σ Ratios 2σ Ratios 2σ Ratios 2σ
Ga/Mg 1.3 2.0 1.6 0.7 5.8 3.3 6.6 4.1 1.0 0.3
such age information, the sapphires may be temporally found associated with calcite, either as discrete mineral
related with to their host basalts, as the age of the basal- grains attached to calcite or as a thin rim around the carbon-
tic volcanism in the SVF is well known (Todt and Lippolt ate phase. These associations are interpreted to be primary.
1980; Linthout et al. 2009; Przybyla 2013). In sample 26, the feldspars are located in vesicles along
cracks within sapphire and associated with clay minerals
Silicates and zeolites; hence, these are considered to be secondary in
origin. This population of silicate mineral inclusions may
Silicate mineral inclusions are potassic feldspar and lab- not be representative for all investigated sapphires, as only
radoritic plagioclase. In one sample (OL25), feldspar was few polished surfaces were studied. Yet, it is striking that
13
Fig. 4 Backscattered electron a
images of selected carbonate b Crn
inclusions in some Siebenge-
birge sapphires. a Calcite inclu- Crn
sion in sample 3. b Dolomite
inclusion in sample 26. c
Vesicular secondary aragonite
in alkali feldspar located at Cal Dol
cracks in sample 26. d Spherical
calcite inclusions in phonolitic
silicate glass in sample 26.
Mineral abbreviations are after
Whitney and Evans (2010) 100 µm 20 µm
c d Crn
Arg
Cal
Afs
100 µm Crn 20 µm
apparently only feldspar has been incorporated as a crystal- of pyrochlore Ca2Nb2O7, thorite ThO2, and thorianite
line silicate mineral, but not pyroxene or olivine, which are ThSiO4 (Fig. 5b). Abundant base metal oxides are spi-
ubiquitous liquidus phases in the host basalt samples. nel—hercynite solid solutions (Mg,Fe)Al2O4, ilmenite—
pyrophanite solid solutions (Fe,Mn)TiO3 with exsolved
Carbonates hercynite and Nb-rutile (Nb,Ti,Fe)O2, and ilmenite
FeTiO3. Figure 5c shows a complex oxide inclusion
Carbonate inclusions are observed in samples 3, 15, 26, where ilmenite-pyrophanite ss showing with Nb-rich
OL25, and S52. Compositions range from calcium carbon- rutile and hercynite exsolutions is rimmed by spinel at the
ate to magnesian carbonate. Vibrational spectroscopic stud- contact to corundum, indicating a reaction of the ilmen-
ies of the carbonate inclusions in sample 26 using Raman ite-pyrophanite ss with the host corundum at magmatic
spectroscopy revealed that these carbonates are either temperatures accounting for a primary origin. In addition,
dolomite, calcite, or aragonite. Calcite and dolomite inclu- one Nb-rutile–Betafite Group (Ca,U)2(Ti,Nb,Ta)2O6(OH)
sions can be regarded as primary as these are completely composite aggregate was found in sample OL25. The
included in the host sapphire. By contrast, aragonite tends Columbite Group inclusion from sample 26 (Fig. 5a) was
to form vesicular textures within secondary potassic feld- used for in situ U–Pb dating using LA–ICP–MS.
spar in the sapphire. Consequently, aragonite is considered
a secondary low-temperature alteration phase. Figure 4 Melt inclusions
shows selected carbonate inclusions inside the Siebenge-
birge sapphires. Sample 26 contains composite silicate glasses of two
different compositions (Fig. 6), interpreted to represent
Oxides silicate melts that were trapped by the growing sap-
phires. One composition (melt 1) forms vesicular struc-
Samples 3, 15, 26, and OL25 have various oxide inclu- tures within a second glass referred to as melt 2. Melt 1
sions. All oxide inclusions are completely isolated inside is quite complex and heterogeneous in composition, and
the sapphire, and are not located at any visible cracks appears to represent a carbonated silicate melt, while
(Fig. 5). Hence, they may be regarded as primary inclu- melt 2 is close to an aluminous phonolite. The aver-
sions that formed cogenetically with the host sapphires. age compositions of the silicate glasses are compiled in
Figure 5a shows a Columbite Group (Mn)(Fe,Nb)2O6 Table 4. Most studies on basalt-hosted sapphires (e.g.,
inclusion that in turn includes a complex aggregate McGee 2005; Pakhomova et al. 2006; Izokh et al. 2010;
13
43
Fig. 5 Complex composite
oxide inclusions in the Sie- a b
bengebirge sapphires. a and b
Pcl
are inclusions in sample 26, c Col
and d are inclusions in sample
OL25. Bet Betafite Group, Crn
Crn corundum, Col Columbite Thr
Group, Herc hercynite, Ilm
ilmenite, Ilme Nb-rich rutile
(ilmenorutile), Pcl Pyrochlore Col
Group, Pph Pyrophanite,
Spl spinel, Thr thorite, Thrn Thrn
thorianite
100 µm 10 µm
c Ilme Crn d
Crn
Ilme
Ilm-Pph Bet
Herc
Spl
20 µm 50 µm
Fig. 6 Inclusions of glasses that

represent quenched products of
a b
two immiscible melts. Melt 1
which has the composition of melt 1 melt 1
an inhomogeneous carbonated
silicate melt, forms spherical
droplets in melt 2 which has a
melt 2
high-alumina phonolitic compo-
melt 2
sition. Halos around melt 2 are
fluorescence effects
50 µm 20 µm
c d
melt 1
melt 2
melt 2
melt 1
20 µm 20 µm
Palke et al. 2016, 2017) report melt inclusions (MI) that those from the Siebengebirge sapphires. Similar compo-
are SiO2 saturated silicates ranging from trachyandesitic sitions of MI were reported by Song and Hu (2009) from
to rhyolitic in composition and, are, hence different than basalt-hosted sapphires from Changle, China, which are
13
Table 4 Results of EPMA analyses of the immiscible melt inclusions in sample 26. Values for CO2 are calculated from the difference to the
totals of the EPMA analyses
Melt 1 Melt 2
1a 1b 1c 1d 2a 2b 3a 3b 3c 4a 4b 4c 5a 5b 5c 1 2 3 4 5
n 2 2 3 3 5
SiO2 56.6 17.0 2.45 48.8 47.3 4.9 58.4 43.4 32.2 32.5 38.5 61.5 38.6 39.5 43.8 58.8 59.6 60.6 60.3 59.6
Al2O3 11.9 28.1 31.9 15.0 11.1 31.7 25.9 26.4 29.6 32.0 27.2 13.9 17.4 16.7 16.5 25.9 25.5 26.0 25.3 25.2
FeO 0.63 0.53 0.63 0.56 0.65 0.44 0.67 0.72 0.49 0.64 0.66 0.56 0.35 0.32 0.30 0.61 0.42 0.56 0.55 0.47
MnO 0.17 0.08 0.10 0.17 0.08 0.07 0.12 0.19 0.08 0.08 0.11 0.12 0.12 0.12 0.24 0.10 0.07 0.15 0.11 0.12
MgO 0.05 0.01 0.01 0.06 0.18 0.02 0.15 0.16 0.19 0.11 0.27 0.17 9.19 8.63 9.93 0.06 0.06 0.07 0.06 0.06
CaO 0.64 0.65 0.66 0.64 0.57 0.60 0.34 0.61 0.61 0.55 0.66 0.54 1.40 1.40 1.40 0.67 0.45 0.50 0.49 0.48
Na2O 0.17 5.82 5.89 0.70 1.57 9.32 6.17 7.13 9.00 9.52 7.77 2.41 0.94 1.50 1.66 7.15 8.40 8.25 8.04 6.78
K2O 5.11 6.89 6.73 6.47 6.57 6.66 3.05 6.90 6.63 6.81 6.70 6.31 0.92 1.16 1.24 6.58 7.18 6.95 7.12 7.02
Total 75.2 59.0 48.3 72.4 68.0 53.7 94.7 85.5 78.8 82.2 81.8 85.6 68.9 69.3 75.1 99.9 101.6 103.0 102.0 99.7
CO2 24.8 41.0 51.7 27.6 32.0 46.3 5.3 14.5 21.2 17.8 18.2 14.4 31.1 30.7 24.9
rich in carbonates and sulfates, and by Vysotskij and fluid trapped in the inclusions was a complex sulfate–
Barkar (2009) who found primary MI of phonolitic com- phosphate–H2O bearing C O2 fluid with some dissolved
position in alluvial corundum from the Primorje Region, cations, notably alumina and alkalis. Similar melt inclu-
Russia. sions daughter phases including sulfates and carbonates
were reported by Song and Hu (2009) from the Changle
Fluid inclusions sapphires.
Fluid inclusions were found in samples 15, 26, S52, and Nano‑inclusions
KS21. They occur either randomly distributed within the
whole sapphire without any preferred crystallographic Selected TEM images of the brown (Nb–Ta-enriched) and
orientation, or arranged along trails (Fig. 7a). Sizes of the blue sapphire domains of sample PE25 are shown in Fig. 9.
fluid inclusions are highly variable and range from <1 µm The brown sapphire domains exhibit a very high population
to several tens of micrometers. Many fluid inclusions density of nano-inclusions that are probably responsible for
show negative crystal shapes (Fig. 7b) following the crys- the brown coloration. The inclusions form ~10 to 20 nm long
tal symmetry of the host sapphire, indicating them to be rods with a shape-preferred orientation having the long axes
of primary origin. of the needles near parallel to corundum [0001], though the
Many fluid inclusions seem to be pure CO2 with scatter of the orientation of the long axes (Fig. 9 a, b; TEM
two strong vibrational bands at 1285 and at 1388 cm−1 bright-field and dark-field images) demonstrates that the
(Rosso and Bodnar 1995). Some fluid inclusions con- nano-inclusions do not show a strict orientation relation rela-
tain daughter phases (Fig. 7c, d) whose compositions tive to the crystal lattice of the host sapphire and thus must
are hard to characterize with Raman spectroscopy due be discrete phases. The crystallinity of the nano-inclusions
to the overlap between the vibrational spectra of multi- is verified by electron diffraction and dark-field imaging
ple phases at low wavenumbers as well as with the spec- using a reflection from the inclusions (Fig. 9b; TEM dark-
trum of the host sapphire. One strong band at 992.4 cm−1 field image), lattice fringes from the inclusions and by the
repeatedly occurs in the vibrational spectrum of many Moirée interference patterns visible in the high-resolution
fluid inclusions. This band is characteristic for miner- image (HRTEM) in Fig. 9c. The latter are produced by the
als such as natroalunite NaAl3(SO4)2(OH6), thenar- superposition of the crystal lattices of the nano-inclusions
dite Na2(SO4), aphthitalite K 3Na(SO4)2, and blödite and that of the host sapphire. The observed diffraction con-
Na2Mg(SO4)2 * 4H2O (Fig. 8). Further identified daugh- trast around the inclusions is characteristic of the so-called
ter minerals are burkeite N a4(SO4)(CO3), paravauxite Guinier–Preston Zones which are nano-crystalline platelets
Fe2+Al2(PO4)2(OH)2*8H2O, and dawsonite NaAl(CO3) in a host material. The compositions of the inclusions were
(OH)2. Although an unambiguous identification of the qualitatively estimated by energy dispersive methods. All
daughter minerals is not possible from the vibrational analyzed inclusions are high in Nb, Ta, Ti, and Fe. Given
spectrum, it can be stated with some confidence that the that Nb-rich rutile (Nb, Ti, Fe)O2 is a common macroscopic
13
43
Fig. 7 Fluid inclusions in Sie-

bengebirge sapphires. a Crystal- a b
lographically oriented elongated
fluid inclusions. b Trails of fluid
inclusions. c, d Multiple phase
fluid inclusions
20 µm 20 µm
c d
20 µm 20 µm
mineral inclusion in the Siebengebirge sapphires, we con- distribution of larger rutile inclusions that have a needle-
sider it most likely that Ta-bearing Nb-rich rutile is the phase like shape and show a strong crystallographic orientation
forming the majority of the nano-inclusions. relationship with the host sapphire (Fig. 9e, f).
The nano-inclusions may be either solid-state exsolu-
tion products or they may be primary crystals that grew
epitaxially along with the sapphire. The lack of a preferred Geochronology
crystallographic orientation relation between the nano-
inclusions and the host sapphire lattice supports the latter The Columbite Group mineral dated in this study
possibility, as it has also been argued for platinum group is a 200–500 µm sized crystal with the formula
mineral nano-inclusions (nPGM) in pentlandite from the Fe0.2Mn0.5Nb2.3O6, suggesting columbite-Mn and was
Merensky Reef platinum group element ore deposit of the found as an inclusion in sapphire 26 from the Oelberg.
Bushveld Complex in South Africa by Wirth et al. (2013) As it contained a complex oxide aggregate (Fig. 5a, b),
and Junge et al. (2014). An additional argument opposing it was repolished until it was free of inclusions which
the nano-inclusions to be solid-state exsolutions is provided allowed for in situ LA–ICP–MS dating. The reliabil-
by their large-scale distribution within the sapphire. Fig- ity of in situ U–Pb dating with LA–ICP–MS has previ-
ure 9d (TEM bright-field image) shows that the population ously been demonstrated by Smith et al. (2004), Dill
density of inclusions varies periodically with inclusion-rich et al. (2007), Melcher et al. (2008, 2015), and Che et al.
zones intermittently alternating with inclusion-poor zones. (2015) who dated coltan (columbite–tantalite solid solu-
Strikingly, the boundaries between the zones with differ- tion series) from different pegmatites in China using
ent population densities are very sharp and almost perfectly LA–ICP–MS.
parallel to each other. The orientation of the lamellar zone/ The returned weighted mean ages of the colum-
boundaries is near parallel to [0001] of the corundum lat- bite-Mn inclusion (Table 5) are 24.74 ± 0.24 Ma
tice and hence may be parallel to the (11–20) hexagonal (MSWD = 3.4) (Fig. 10). The accuracy and reproduc-
prism of corundum. The distribution of the nano-inclu- ibility were checked by repeated analyses of Coltan 007
sions within the sapphire differs significantly from the (n = 8) which yields an intercept age of 2074.1 ± 7.5 Ma
13
Questions addressed in the discussion of this paper are

the nature of the parental melt that crystallized the sap-
cm -1
phires, as well as the relationship between the sapphires
3.3
cm -1
99
and the host primitive alkaline basalts, as sapphires only
7.5
occur in such primitive basalts from intra-continental set-
41
cm -1
tings, but never in basalts from other geotectonic settings.
m -1
2.7
1c
Blödite 45
62 Magmatic vs. metamorphic origin

Thenardite
cm -1
m -1
cm -1
cm -1
5c
Intensity (cps, arbitrary units)
Accepting a xenogenetic relation between the sapphires
.7
cm -1 642.
87
1.5
8.7
Aphthitalite
10
75
37
1c
m -1
and the host basalt, the first question to be answered is
the geological nature of the Siebengebirge sapphires, i.e.,
5.5
Corundum
81
57
if they formed via igneous or metamorphic processes.

The distinction between sapphire crystals of magmatic
and of metamorphic origin is mainly based on their Ga/
KS21-1
Mg ratio (Peucat et al. 2007), but also on the mineral
0
inclusion suite (e.g., Sutherland and Schwarz 2001; Gra-
200 400 600 800 1000 1200 1400
wavenumber (cm-1) ham et al. 2004). According to Peucat et al. (2007), mag-
matic blue sapphires have rather high Ga/Mg ratios (>10)
Fig. 8 Raman spectrum of fluid inclusion KS21-1 with the reference
and metamorphic or metasomatic blue sapphires have
spectra of possible daughter phases. The raman spectrum of KS21-1 rather low Ga/Mg ratios (<10). According to that dis-
is a mixed spectrum of several minerals. Many bands represent the crimination, the Siebengebirge sapphires would mostly
corundum matrix. Other bands cannot be unambiguously assigned to fall into the “magmatic” field, whereas three samples (26,
a certain mineral. But certainly, the daughter phase must be a sodium
sulfate. The reference spectra are from the RRUFF Database. The fol-
KS21, OL60) would clearly plot in the “metamorphic”
lowing reference spectra were used: blödite: R050341; thenardite: field. This discrimination between magmatic and meta-
R040178; aphthitalite: R050651; corundum: R060020 morphic sapphires by Peucat et al. (2007) is based on
the fractionation of Ga/Al and of Mg in sapphires during
crystallization. In the corundum crystal lattice, Ga substi-
(MSWD = 2.9) that is in excellent agreement with the tutes for Al, whereas Mg is located at interstitial lattice
published value (Melcher et al. 2008, 2015). sites. Wahlen et al. (1987) found that magmatic corun-
dum generally has higher Ga/Al ratios than metamorphic
corundum, explaining this by the extraction of Ga from
Discussion the metamorphic rocks by circulating F-rich fluids dur-
ing partial melting of granulites. During fluid circula-
All sapphires investigated in this study are rimmed by tion, F− would complex with G a3+ to form GaF3−6 ions
tens to several hundred micrometer sized spinel rims, that are extracted, leaving the restitic granulite depleted
indicating chemical disequilibrium with the host basalt in Ga over Al. If this process implicated by Whalen et al.
and supporting a xenocrystic nature of the sapphires. This (1987) controlled the Ga/Mg ratio of a sapphire and
is supported by the phase relations in the CMAS system hence indicated a metamorphic or magmatic origin, the
(Sen and Presnall 1984; Milholland and Presnall 1998; Ga/Mg ratio should be a function of the Ga content of the
Liu and Presnall 1990, 2000), showing that a hypotheti- sapphire. Such a Ga control of the Ga/Mg ratio cannot be
cal cooling path of a basaltic model melt will never inter- observed for the Siebengebirge sapphires. In contrast, the
sect the corundum stability field (Fig. 11). In the pres- Ga/Mg ratio is correlated with the bulk Mg content of the
ence of MgO (in CMAS) or MgO+FeO (in basalt), stable sapphire (Fig. 12). Therefore, we consider it unlikely that
aluminous phases co-existing with near-liquidus melts the Siebengebirge sapphires represent a polymodal suite,
are spinel at 1–2 GPa, garnet at ~3 GPa, or anorthite at i.e., that some samples were formed via metamorphic
pressures <1 GPa. Hence, the stability of corundum is not processes and others are igneous precipitates.
only a function of the alumina content of a melt, but is Support for a monogenetic, magmatic, origin for all
also sensitive to the presence of MgO and FeO. In a prim- Siebengebirge sapphires is provided by several other min-
itive basaltic melt with high MgO and FeO contents, the eralogical characteristics which are (1) syngenetic mineral
corundum stability field in the CMAS system will never inclusions of Pyrochlore Group, Columbite Group, cal-
be intersected upon cooling from magmatic temperatures, cite and feldspar series members that are more typical for
so that corundum will never be a stable alumina phase. a magmatic origin, (2) the presence of glassy inclusions
13
43
Fig. 9 TEM images of the brownish sapphire domains (a–d) and patterns. d TEM bright-field image showing a low-population density
of the blue sapphire domain (e, f). a TEM bright-field image, and zone between two zones of high nano-inclusion population density.
b TEM dark-field image of a zone with a high population density The zone-boundary is sharp and oriented parallel to the [0001] zone
in nano-inclusions, showing that the inclusions have roughly the axis of the host sapphire. e TEM bright-field image of a blue sapphire
same shape orientation, but are not in a strict shape-preferred orienta- domain showing strictly crystallographically oriented needle-like
tion relationship with the host sapphire. Halos around the inclusions ilmenite inclusions. f HRTEM image of a blue domain showing the
are nano-crystalline platelets in the host rock. c high-resolution (HR) lattice fringe of an ilmenite inclusion within corundum
TEM lattice fringe image of a nano-inclusion showing typical Moirée
which were likely melts that were trapped as the corundum of their inclusions. If the sapphires and hence their min-
crystals grew, and (3) the presence of numerous Nb-rich eral inclusions had been of metamorphic origin, then they
nano-inclusions in the brownish domain of sapphire PE25 should yield Devonian ages (i.e., the age of the metamor-
which are most likely magmatic precipitates that co-pre- phic basement below the SVF). However, our own in situ
cipitated with the sapphire host during crystal growth. This dating resulted in Cenozoic ages which, when taking into
lamellar distribution of intermittently alternating inclusion- account the uncertainty, are the same age as the host basalt
poor and inclusion-rich zones of the nano-inclusions par- (24.74 ± 0.24 Ma), indicating a clear magmatic origin for
allel to the prism faces of the host sapphire supports an these inclusions and hence for their host sapphires.
epitaxial growth of the inclusions on the sapphire crystal Based on this discussion, we state that the discrimination
faces. Such oscillatory patterns are ubiquitous in many geo- between magmatic and metamorphic sapphires based on the
chemical systems (L’Heureux 2013) and are either inter- Ga/Mg ratio as proposed by Peucat et al. (2007) has to be
preted to be externally controlled and to reflect periodic regarded with caution and may not be applied to any sap-
chemical changes in the crystallizing environment, or to be phire without additional information provided by the mineral
internally controlled by intrinsic self-organization in a non- inclusion suite. The Ga/Mg ratio may be used as a discrimi-
equilibrium system with positive feedback (e.g., Ortoleva nation feature for the sapphire samples investigated in the
1994; Merino and Wang 2001). study from Peucat et al. (2007), but at least for the sapphires
Perhaps the most compelling evidence for a magmatic presented in this study this may not be applicable. Similar
origin for the Siebengebirge sapphires comes from the age discrepancies for the discrimination between “magmatic”
13
Table 5 U-Th–Pb isotopic data for the columbite-Mn inclusion

Spot Age Age
207
Pba Ub Pbb Thb 206
Pbcc 206
Pbd ±2σ 207
Pbd ±2σ 207
Pbd ±2σ rhoe 206
Pb ±2σ 207
Pb ±2σ
238 235 206 238 235
(cps) (ppm) (ppm) U (%) U (%) U (%) Pb (%) U (Ma) U (Ma)
A06 569 685 2.9 0.66 1.4 0.00368 3.3 0.0229 9.5 0.0451 8.9 0.35 23.7 0.8 23.0 2.2
A07 367 363 1.5 0.47 1.8 0.00383 3.8 0.0248 8.0 0.0470 7.1 0.47 24.7 0.9 24.9 2.0
A08 513 687 3.0 0.61 0.9 0.00397 3.5 0.0253 9.1 0.0462 8.3 0.39 25.5 0.9 25.3 2.3
A09 402 419 1.8 0.44 2.1 0.00397 3.7 0.0250 9.2 0.0457 8.5 0.40 25.6 0.9 25.1 2.3
A10 26,135 384 20.7 0.58 25.9 0.00385 11 0.0270 17.3 0.0509 13 0.63 24.8 2.7 27.1 4.6
A11 375 249 1.1 0.34 4.0 0.00394 4.3 0.0248 12.0 0.0456 11 0.35 25.4 1.1 24.9 3.0
A12 1230 616 3.5 0.85 10.1 0.00393 3.9 0.0255 12.1 0.0471 11 0.32 25.3 1.0 25.5 3.1
A13 813 570 2.7 0.50 7.0 0.00383 3.8 0.0232 13.8 0.0440 13 0.27 24.6 0.9 23.3 3.2
A14 727 623 3.0 0.69 3.9 0.00387 3.3 0.0240 8.6 0.0450 8 0.38 24.9 0.8 24.1 2.0
A15 849 946 4.8 1.02 0.3 0.00401 3.2 0.0266 5.8 0.0481 4.8 0.56 25.8 0.8 26.7 1.5
A16 7954 544 7.9 0.68 25.0 0.00378 5.4 0.0243 18.6 0.0467 18 0.29 24.3 1.3 24.4 4.5
A17 282 269 1.1 0.28 5.1 0.00383 4.3 0.0221 16.6 0.0419 16 0.26 24.7 1.1 22.2 3.6
A18 373 365 1.5 0.34 1.4 0.00375 3.9 0.0245 8.9 0.0473 8.1 0.43 24.2 0.9 24.5 2.2
A19 507 230 0.9 0.25 3.0 0.00388 3.3 0.0238 11.7 0.0445 11 0.28 25.0 0.8 23.9 2.8
A20 865 402 1.6 0.49 3.3 0.00368 3.2 0.0256 9.8 0.0504 9.3 0.33 23.7 0.8 25.6 2.5
A21 1198 492 2.0 0.29 4.4 0.00376 3.2 0.0251 7.3 0.0484 6.5 0.44 24.2 0.8 25.2 1.8
A22 1251 422 1.9 0.34 6.2 0.00400 3.2 0.0264 7.4 0.0480 6.6 0.44 25.7 0.8 26.5 1.9
A23 1242 371 1.7 0.56 7.2 0.00370 3.7 0.0234 11.2 0.0459 11 0.33 23.8 0.9 23.5 2.6
A24 694 325 1.4 0.51 2.4 0.00377 3.2 0.0241 8.9 0.0463 8.3 0.35 24.3 0.8 24.2 2.1
A25 771 387 1.6 0.43 1.7 0.00372 3.6 0.0247 8.3 0.0482 7.4 0.43 23.9 0.9 24.8 2.0
A26 719 406 1.6 0.44 1.7 0.00365 3.4 0.0226 8.6 0.0450 7.9 0.40 23.5 0.8 22.7 1.9
A27 444 457 2.1 0.57 1.1 0.00406 3.4 0.0263 9.5 0.0471 8.9 0.36 26.1 0.9 26.4 2.5
A28 422 451 1.9 0.44 2.4 0.00390 3.6 0.0238 9.7 0.0442 9 0.37 25.1 0.9 23.9 2.3
A29 417 428 1.8 0.57 1.2 0.00375 3.7 0.0242 9.4 0.0469 8.6 0.39 24.1 0.9 24.3 2.3
A30 448 505 2.2 0.52 0.9 0.00383 3.7 0.0248 9.2 0.0470 8.5 0.40 24.6 0.9 24.9 2.3
A31 591 566 2.4 0.52 1.7 0.00375 3.4 0.0236 7.3 0.0456 6.4 0.47 24.1 0.8 23.6 1.7
A32 567 491 2.2 0.47 4.3 0.00395 3.2 0.0257 12.8 0.0471 12 0.25 25.4 0.8 25.7 3.3
A38 483 441 2.0 0.62 2.1 0.00378 3.7 0.0236 12.3 0.0453 12 0.30 24.3 0.9 23.7 2.9
A39 995 666 2.8 0.74 3.0 0.00374 3.4 0.0233 9.9 0.0451 9.3 0.34 24.1 0.8 23.4 2.3
A40 441 208 0.9 0.48 1.9 0.00402 3.9 0.0257 10.0 0.0464 9.2 0.39 25.9 1.0 25.8 2.5
A41 473 233 0.9 0.28 1.0 0.00381 3.5 0.0264 7.8 0.0502 7 0.44 24.5 0.8 26.4 2.0
A42 725 177 0.9 0.46 14.4 0.00389 3.5 0.0258 15.4 0.0480 15 0.23 25.0 0.9 25.8 3.9
A43 556 273 1.1 0.42 1.9 0.00364 3.2 0.0236 10.8 0.0470 10 0.30 23.5 0.8 23.7 2.5
A44 769 368 1.6 0.51 1.8 0.00399 3.0 0.0259 7.9 0.0471 7.3 0.39 25.6 0.8 26.0 2.0
A45 1158 338 1.5 0.45 10.2 0.00364 3.7 0.0226 9.9 0.0451 9.2 0.38 23.4 0.9 22.7 2.2
A46 866 379 1.7 0.47 4.4 0.00392 3.4 0.0258 9.1 0.0478 8.4 0.38 25.2 0.9 25.9 2.3
A47 1007 400 1.8 0.44 5.4 0.00393 3.2 0.0241 8.8 0.0445 8.2 0.36 25.2 0.8 24.2 2.1
A48 2132 478 2.5 0.45 12.6 0.00399 3.2 0.0252 9.7 0.0458 9.1 0.34 25.7 0.8 25.3 2.4
A49 625 343 1.4 0.37 1.1 0.00382 3.2 0.0259 8.7 0.0493 8.1 0.36 24.6 0.8 26.0 2.2
A50 942 432 1.8 0.45 3.1 0.00384 3.1 0.0252 7.7 0.0476 7.1 0.41 24.7 0.8 25.3 1.9
A51 614 276 1.1 0.32 3.3 0.00388 3.5 0.0238 10.7 0.0445 10 0.32 25.0 0.9 23.9 2.5
13
43
Table 5 continued
Spot Age Age
207
Pba Ub Pbb Thb 206
Pbcc 206
Pbd ±2σ 207
Pbd ±2σ 207
Pbd ±2σ rhoe 206
Pb ±2σ 207
Pb ±2σ
238 235 206 238 235
(cps) (ppm) (ppm) U (%) U (%) U (%) Pb (%) U (Ma) U (Ma)
A52 693 313 1.2 0.29 3.0 0.00373 3.5 0.0255 7.6 0.0496 6.8 0.46 24.0 0.8 25.5 1.9
A53 1023 643 2.8 0.70 1.3 0.00388 3.1 0.0266 7.3 0.0497 6.5 0.43 25.0 0.8 26.7 1.9
A54 865 345 1.5 0.33 4.0 0.00409 3.1 0.0263 10.5 0.0467 10 0.30 26.3 0.8 26.4 2.7
A55 1920 385 2.1 0.61 11.2 0.00405 3.3 0.0292 12.4 0.0524 12 0.27 26.0 0.9 29.3 3.6
A56 41,353 1981 145.2 0.02 0.0 0.07940 2.6 0.6281 2.8 0.0574 0.99 0.93 492.6 12.3 494.9 10.9
Spot size = 19 and 28 µm, respectively; depth of crater ~20 µm. 206Pb/238U error is the quadratic addition of the within run precision (2 SE) and
an estimated repeatability of 2% (2 SD). 207Pb/206Pb error propagation (207Pb signal dependent) following Gerdes and Zeh (2009). 207Pb/235U
error is the quadratic addition of the 207Pb/206Pb and 206Pb/238U uncertainty
a
Within run background-corrected mean 207Pb signal in cps (counts per second)
b
U and Pb content and Th/U ratio were calculated relative to GJ-1 reference zircon
c
Percentage of the common Pb on the 206Pb. b.d.l. below detection limit
d
Corrected for background, within-run Pb/U fractionation (in case of 206Pb/238U) and common Pb using Stacey and Kramers (1975)
model Pb composition and subsequently normalised to GJ-1 (ID-TIMS value/measured value); 207Pb/235U calculated using 207Pb/206Pb/
(238U/206Pb*1/137.88)
e
Rho is the 206Pb/238U/207Pb/235U error correlation coefficient
f
Degree of concordance = 206Pb/238U age/207Pb/206Pb age × 100
g
Accuracy and reproducibility were checked by repeated analyses (n = 8 and 9) of reference columbite 139 and 007; data given as mean with 2
standard deviation uncertainties
data-pint error ellipses are 2 σ

28
0.0044
27
0.0042
26
Age (Ma)
0.0040
25
Pb/238U
0.0038
24
206
0.0036 23
0.0034 22 Mean = 24.74±0.24 [0.96%] 95% conf.

Wtd by data-pt errs only, 0 of 45 rej.
MSWD = 3.4
0.0032 21
0.018 0.022 0.026 0.030 0.034
Pb/ U
207* 235
Fig. 10 Age calculations for the columbite-Mn inclusion in the Sie- ratio. The gray bars are the errors of the calculated age; the red line
bengebirge sapphire sample 26. a Data plotted in a Wetherill concor- represents the mean age of all analyses
dia diagram. b Weighted mean age based on the calculated 206Pb/238U
and “metamorphic” sapphires based on their Ga content, Montana, USA, lead Palke et al. (2017) to suggest the trace
and the Ga/Mg ratio as proposed by Peucat et al. (2007) element classification scheme for sapphires to be revisited
were pointed out by Izokh et al. (2010), Uher et al. (2012), and revised. Based on our trace element data on the Sie-
Sutherland et al. (2015), and Palke et al. (2015, 2017). The bengebirge sapphires, we support the necessity of the revi-
discrepancies in trace element data and the resulting ambi- sion of the Peucat et al. (2007) trace element-based discrimi-
guity about the origin of sapphires from alluvial deposits in nation for the identification of the origin of sapphires.
13
CaAl2Si2O8 the HFSE are well known from highly evolved silicate
melts and from carbonatitic melts. Most authors (e.g.,
2 GPa Sutherland et al. 1998b) relate the HFSE enrichment of
Crn
the sapphires and the presence of Columbite Group, Pyro-
chlore Group and of Nb-rich rutile mineral inclusions to
Spr An
highly evolved SiO2 undersaturated silicate melts such as
t
ten
Spl syenites, or nepheline syenites. A further observation lead-

on
ing these authors to favor syenites as parental melts is that

Al-c
repeatedly melt inclusions of syenitic composition were

ng
asi
reported from magmatic sapphires (e.g., Pakhomova et al.

re
inc
2006; Zaw et al. 2006; Izokh et al. 2010). Up until now,

En
carbonatites have not been considered as potential parental
melts, although many authors (e.g., Coenraads et al. 1990;
Fo Qtz
Van Long et al. 2004; McGee 2005; Nechaev et al. 2009;
Izokh et al. 2010; Palke et al. 2017) reported that CO2 must
Mg2SiO4 MgSiO3 wt. % SiO2 have played a significant role in the formation of these sap-
phires, as CO2-rich fluid inclusions are commonly observed
in many magmatic sapphires. Palke et al. (2017) interpreted
Fig. 11 Phase relations in the CMAS system at 2 GPa (Liu und
Presnall 1990) pressures. Abbreviations after Whitney and Evans the presence of analcime–calcite inclusions in alluvial
(2010). The star in the forsterite field represents a typical composi- magmatic sapphires from Montana, USA as an indication
tion of an alkali basalt. Upon cooling, the melt will follow the red for that the sapphires crystallized from a melt that simul-
path. The final eutectic composition of the melt is given by the small
taneously exsolved a carbonatite melt, and fractions of that
stars at the anorthite–enstatite–quartz triple point
carbonatite melt were trapped by the growing sapphire
and later recrystallized to the analcime-calcite paragen-
a 100 b 100 esis. However, in their genetic model, the carbonatite melt
was only a “by-product” that resulted from the carbonated
10 10 nature of the source rock, and they did not state a compel-
ling necessity of the presence of C O2 or a carbonatite melt
Ga/Mg
Ga/Mg
1 1 for the sapphire formation. Only Guo et al. (1996) and

Limatrakun et al. (2001) assign carbonatitic melts a role in
0.1 0.1
sapphire petrogenesis. Yet, the only role that was assigned
0.01 0.01
to carbonatites in these studies was their function as a CO2
100 1000 1 10 100 1000 10 000 source, but no direct parentage was proposed, mainly due
Ga (ppm) Mg (ppm) to the lack of primary evidence for the presence of a car-
bonatitic melt during sapphire crystallization.
Fig.  12 a Ga/Mg vs. Ga (ppm) and b Ga/Mg vs. Mg (ppm) plots. The Siebengebirge sapphires by contrast do show direct
There is no visible effect of Ga on the Ga/Mg ratio, whereas the total evidence for the participation of a carbonatitic melt in the
Mg content in the sapphires is highly correlated with the Ga/Mg ratio,
indicating that the strong variability of the Ga/Mg ratio in the Sie- sapphire petrogenesis which is given by the several pri-
bengebirge sapphires is controlled by changes in the Mg concentra- mary carbonate inclusions and by the glasses in sample
tion, and largely unaffected by changes in the Ga content 26. These glasses have two different compositions and are
either a carbonated silicate or a phonolitic glass, and prob-
ably represent quenched liquids that formed via liquid
Constraints on the parental melt immiscibility from a carbonated silicate melt, as indicated
by the textures shown in Fig. 6. We, therefore, suggest that
Important indicators for the composition of the parental liquid immiscibility in the carbonatite–silicate melt system
melt of the Siebengebirge sapphires are the trace element played a major role in the genesis of magmatic sapphires,
composition and the composition of syngenetic mineral, in a similar fashion as stated before by Palke et al. (2017)
fluid, and glass inclusions, as these directly sample the for magmatic sapphires from Montana, USA.
parental melt. One key characteristic of the Siebengebirge The liquid miscibility relations in the carbonatite–
sapphires is the abundance of HFSE-rich mineral inclu- silicate system have been studied intensively in the past
sions and of HFSE-rich nano-inclusions in the brownish decades (Koster VanGroos and Wyllie 1963, 1966; Ham-
domains, indicating that the sapphires’ parental melt must ilton and Freestone 1979; Freestone and Hamilton 1980;
have been rich in the HFSE. Such high concentrations in Kjarsgaard and Hamilton 1988; Brooker and Hamilton
13
43
Na2O + K2O would be a carbonatite containing ≤5 wt% SiO2+Al2O3.

c
1.0 GPa a The prerequisite for a liquid that lies along the a–b two
liquid field boundary to cross the silicate liquidus surface
CO2-saturated
and reach the silicate–carbonate field boundary to precip-
itate carbonate minerals is the physical separation of the
Silicate carbonatitic melt from the co-existing silicate melt (e.g.,
liquidus
surface Kjarsgaard and Peterson 1991; Lee and Wyllie 1998).
When the necessity for the presence of carbonatitic
%
Carbonate
liquidus melt is combined with the abundance of carbonate min-
wt.
Two-liquid field surface

eral inclusions and C O2-rich fluid inclusions in the sap-
b phires, it seems most probable that the sapphires crystal-
lized from the carbonatite that separated from the melts
now present as glass inclusions within the sapphires.
After physical separation of the putative carbonatitic
Al2O3 + SiO2 CaO + MgO
+ FeO melt from the co-existing silicate melt, the carbonatitic
melt will evolve separately. It is likely that the sapphires
from the SVF crystallized from these carbonatitic melts
Fig. 13 Location of the glass compositions (melt 1: green stars;
melt 2: red star) in the schematic pseudoternary phase diagram of the that exsolved from a highly evolved melt of phonolitic
Hamilton projection at 1 GPa (Lee and Wyllie 1998). The line ‘a–b’ composition.
represents the two liquid field boundary at the N
a2O, K2O rich side, Additional support to the idea that sapphires crys-
and line ‘b–c’ represents the silicate–carbonate field boundary. The
tallized from a carbonatitic parental melt is given by
composition of both glasses plot in the same field (silicate liquidus
surface at the S iO2+Al2O3 rich apex), ruling out that they formed the extremely low viscosity of carbonatitic melts of
solely via liquid immiscibility ~1.5 × 10−2 – 5 × 10−3 Pa s (Dobson et al. 1996) which
would promote rapid ion transport through the melt that
is needed to form megacrystic corundum at low degrees
1990; Lee and Wyllie 1997, 1998; Brooker 1998; Brooker of alumina oversaturation.
and Kjarsgaard 2010). From the phase relations shown in A possible origin of the sapphires from carbonatitic
Fig. 13, it becomes obvious that the two carbonated sili- melts could be identified employing oxygen isotope
cate and the phonolitic melts do not represent two melts measurements on the sapphires, as carbonate-associated
that exsolved from each other, as two liquids that lie in corundum has generally higher δ18O values than oth-
the field of liquid immiscibility would exsolve along the ers (Giuliani et al. 2005). Yet, interpreting oxygen iso-
cotectic lines, so that one composition would lie at the tope data from sapphires that prospectively precipitated
lowermost (SiO2 and Al2O3 rich) two field boundary, and from a carbonatitic melt would be quite difficult as there
the co-existing composition would lie at the opposite are no data available about oxygen isotope fractionation
(Na2O and K2O/CaO rich) two field boundary. Yet, the between corundum and carbonatitic melt.
glasses show striking evidence for liquid immiscibility,
such as the vesicular shape of the innermost glass (melt Genetic relation between the sapphires and their host
1). Consequently, if these glasses do represent quenched rocks
products of two melts that formed via liquid immiscibil-
ity, as indicated by the vesicular shape, then there must In situ LA–ICP–MS U–Pb dating of a syngenetic colum-
have been an additional, N a2O- and K 2O-rich liquid that bite-Mn inclusion within a xenocrystic sapphire hosted
simultaneously separated from the co-existing S iO2- and by alkaline basalt from the Oelberg in the SVF yielded
Al2O3-rich liquids, but is now not present in the sap- an age of 24.73 ± 0.35 Ma. This age closely matches the
phire. Evidence for the existence of such an N a2O- and age of alkaline mafic volcanism in the area. The alkali
K2O-rich liquid is given by the alkali-rich daughter basalts from the Oelberg have an 40Ar–39Ar isochron
phases in the fluid inclusions that were observed in some age of 24.91 ± 0.45 Ma (Przybyla 2013) and can thus be
of the sapphires. This Na2O- and K2O-rich liquid would regarded as cogenetic with the Siebengebirge sapphires.
have a composition that lies somewhere on the ‘a–b’ The Oligocene age of the sapphires confirms a magmatic
two liquid field boundary in Fig. 13. Upon cooling, the origin, as a sapphire that originated from the basement
melt will evolve from a to b via crystallization of silicate below the SVF would be expected to yield a Variscan age.
phases, and will then evolve along the silicate–carbonate The agreement between the age of the Siebengebirge sap-
field boundary ‘b–c’ with the co-precipitation of silicate phires and the alkaline basaltic volcanism provides cru-
and carbonate phases (Lee and Wyllie 1998). Such a melt cial evidence for a genetic link between the processes
13
triggering alkaline basaltic magmatism and the crystalli- 1.5

zation of sapphire megacrysts. The genetic link between
the sapphire crystallization and host alkaline basalt must
Pressure (GPa)
be via C O2. As mentioned in the introduction, the BGY 1
sapphires are restricted in occurrence to a very specific Siebengebirge
geotectonic setting, i.e., to intra-continental rift-related Nezametnoye
alkaline basalts. It has been inferred by many previous 0.5
studies (e.g., Hirose 1997; Green and Falloon 1998; Das- Weldborough
gupta et al. 2007; Zeng et al. 2010) that the mantle source Kings Plains
generating continental intra-plate alkaline volcanics must
be either carbonate-bearing, or metasomatically enriched
500 1000
by CO2 bearing fluids. Also, the alkaline mafic melts from
Temperature (°C)
the SVF are suggested to be generated from refertilized
carbonated spinel peridotite (Jung et al. 2012). This infers
that BGY sapphires are only generated in intra-continen- Fig. 14 P–T stability field estimated for the Siebengebirge sapphires.
The field for the Siebengebirge sapphires is drawn based on assumed
tal settings where the mantle source is enriched in CO2, average temperatures of 700–1000 °C for the carbonatites that crys-
so that primary melts can evolve to phonolites and later tallized the sapphires (Dobson et al. 1996; Genge et al. 1995; Wolff
expell carbonatites which in turn may crystallize sap- 1994). For comparison, the P–T fields for the Weldborough (McGee
phires. The necessity of a C O2 enriched mantle source to 2005) and Kings Plains (Abduriyim et al. 2014) sapphires in Aus-
tralia, and sapphires from the Nezametnoye deposit in the Primorsky
produce melts that are capable of precipitating BGY sap- Region, Far East Russia (Pakhomova et al. 2006) are shown
phires explains the lack of sapphires in S iO2 saturated
basalts such as tholeiites.
of 700–1000 °C (Dobson et al. 1996; Genge et al. 1995;
Wolff 1994), the pressure of sapphire crystallization
Barometry must have been between 0.1 and 0.4 GPa. As we cannot
exclude the possibility that during ascent-related decom-
When a magmatic mineral contains CO2-bearing fluid pression of the sapphires some fluid inclusions decrepi-
inclusions, the density of the inclusions may be used tated to form inclusions of lower density than the original
to quantify the pressure during crystallization (Bertrán one, we therefore assume that the highest estimated pres-
1983). The density of C O2-bearing fluid inclusions can sure of 0.4 GPa to be the minimum pressure of fluid trap-
be calculated from the distance between the C O2 main ping and hence of sapphire crystallization. This pressure
peaks at ~1388.0 cm (upper band) and ~1285.5 cm−1
−1
corresponds to a depth of ~12 km.
(lower band) of its vibrational spectrum obtained using Based on fluid inclusions, the lithostatic pressure dur-
Raman spectroscopy (e.g., Garrabos et al. 1980; Rosso ing crystallization has been previously obtained for sap-
and Bodnar 1995; Yamamoto and Kagi 2006; Wang et al. phires from Weldborough and Kings Plains, Australia
2011). For the Siebengebirge sapphires, we used the (McGee 2005; Abduriyim et al. 2014), and from the
equation of Wang et al. (2011) that relates the distance Nezametnoye deposit in the Primorsky Region, Far East
between the C O2 upper and lower band to the density Russia (Pakhomova et al., 2006), and are estimated to be
of a CO2-bearing fluid inclusion, and obtained densi- 0.45 GPa (Weldborough), 0.07–0.54 GPa (Kings Plains),
ties in the range of 0.3–0.9 g/cm3. A systematic relation and 0.17–0.3 GPa (Nezametnoye), corresponding to mid-
between the estimated density of a fluid inclusion and its to lower-crustal levels.
location within the host sapphire was not observed. The The pressures estimated for the Siebengebirge sapphires
pressure during entrapment of the fluid by the host crys- agree well with pressures estimated for sapphires from
tal can be calculated from the volumetric properties of Weldborough and Kings Plains, Australia (McGee 2005;
the fluid inclusion, given that composition and trapping Abduriyim et al. 2014) and the Nezametnoye deposit in the
temperature are known and given that the system was Primorsky Region (Pakhomova et al. 2006) (Fig. 14). The
closed. Pressure and temperature vary along isochores. temperature estimates for the Nezametnoye (780–820 °C)
The P–V–T relations of pure C O2 were calculated by sapphires are also well in accord with a putative crystalliza-
Yamamoto et al. (2002) from the equation of state for tion from carbonatites at low magmatic temperatures. All
CO2 by Pitzer and Sterner (1994). Assuming that the Sie- P–T estimates from the literature, and from this study are
bengebirge sapphires crystallized from carbonatitic melts well in accord with a crystallization of the magmatic sap-
and taking average liquidus temperatures for carbonatites phires from carbonatitic melts at mid-crustal levels.
13
43
Petrogenetic model for the Siebengebirge sapphires. Still, although these studies comprise detailed
sapphires investigations on the sapphires including trace element
analyses, analysis of mineral inclusions as well as melt
The formation of the alkaline basalt-hosted sapphires and/or fluid inclusion and oxygen isotope measurements,
from the SVF is a result of a complex line of magmatic no consensus was reached about the origin of these sap-
processes that elapsed during a very short time scale. phires. Moreover, these studies cannot explain the inti-
During Miocene times, extensive active volcanism started mate association of blue–green–yellow (BGY) sapphires
in the SVF, producing a broad spectrum of SiO2 saturated with alkaline basaltic volcanism. This study on the BGY
and undersaturated primitive and evolved volcanic rocks. sapphires hosted by alkaline basalts from the Siebenge-
The alkaline mafic melts probably formed due to melt- birge Volcanic Field (SVF) elucidates many yet unknown
ing of amphibole and phlogopite-bearing peridotite in the details about the formation of this special type of gem-
spinel peridotite stability zone (Jung et al. 2012; Schubert stone. The main conclusions from this study can be sum-
et al. 2015), at pressures between 1 and 2.5 GPa and tem- marized as follows:
peratures between 1150 and 1350 °C (Fig. 15a). These
P–T conditions correspond to the lithospheric mantle. 1) Sapphire megacrysts are xenogenetic to the host
It has been repeatedly suggested that the alkaline mafic basalts, as indicated by the spinel corona surrounding
rocks were formed from a carbonatite-metasomatized each sapphire at the contact towards the basalt.
mantle source (Jung et al. 2012; Pfänder et al. 2012). As 2) The parental melt was probably a carbonatitic melt
a result, the melts must have been enriched in volatiles that exsolved from a highly evolved phonolitic melt,
including CO2. With ensuing fractionation of the primi- as evidenced by carbonate inclusions, strong enrich-
tive melts towards more evolved phonolitic compositions, ment in the HFSE, by the glass inclusions of car-
the partial pressure of C O2 (pCO2) in the melt increases bonated silicate in contact with phonolite, and by
(Fig. 15b) until at high degrees of differentiation and CO2-bearing fluid inclusions.
high pCO2, a carbonatitic melt exsolves from the pho- 3) There is a close genetic relation between the sap-
nolite (Kogarko 1997). The carbonatitic melt exsolving phires and their host alkaline mafic rocks, as evi-
from a highly evolved phonolite will be CaO- and vola- denced by the overlapping age of the Siebengebirge
tile-rich but is expected to be depleted in S
iO2, MgO, and sapphires dated in this work and the age of the alka-
FeO (Kogarko 1997). The carbonatite will cool and start line mafic volcanism in the area (Przybyla 2013). We
to crystallize carbonate phases when it is physically sepa- identified the genetic link between magmatic sap-
rated from the co-existing silicate melt (Lee and Wyllie phires and the host basalt that has been a puzzle to
1998). The physical separation of the carbonatitic melt date, to bear upon CO2. Only alkaline volcanic suites
from the silicate melt will probably occur rapidly after can build up enough CO2 in the magma chamber
exsolution due to the significant viscosity and density dif- upon fractionation so that at high degrees of fraction-
ference of both melts. The crystallization of the sapphires ation a carbonatitic melt exsolves which in turn can
from highly fractionated carbonatite occurred at mid- crystallize sapphires at a late stage of evolution. Dur-
dle- to upper crustal levels (Fig. 15c) after the physical ing ascent, fresh alkaline basaltic melts that encoun-
separation of the carbonatite and the phonolite melt, as ter carbonatite melt pockets will decompose the car-
indicated by the presence of carbonate inclusions in the bonatite and incorporate the sapphires to carry them
sapphires. A fresh pulse of hot ascending alkali basalts to the surface.
later encounters the carbonatite melt pocket in the crust
on their way upwards, decomposes the carbonatitic melt Our findings from the Siebengebirge sapphires raise the
and incorporates the sapphires to transport them to the question about an overall applicability of the genetic model
surface (Fig. 15d). to worldwide occurrences of magmatic sapphires. It is hard
to tell if all magmatic sapphires from the world must have
precipitated from a carbonatitic melt that exsolved from
Summary and conclusions highly fractionated SiO2 undersaturated silicate melts. If so,
one would expect more sapphires from other locations to
Magmatic sapphires are well known to occur associated show similar features evidencing for a carbonatitic parental
with intra-continental alkaline mafic volcanism. Around melt. Yet, it is important to notice that even those sapphires
the world, predominantly in Asia and Australia, many from this study do not all show the whole range of features
alkaline basalt fields are reported to carry megacrysts of pointing towards a carbonatitic parentage, i.e., carbonate
magmatic sapphires. To date, many hypotheses have been and HFSE-rich mineral inclusions + CO2 fluid inclusions
proposed to investigate the nature of these so-called BGY with carbonate daughter phases + silicate/carbonate melt
13
Fig. 15 Petrogenetic model for a

the Siebengebirge sapphires.
a Alkaline basaltic volcanism
is initiated in the SVF. b Dif-
ferentiation of alkaline mafic
magmas leads to the formation
of highly evolved phonolites,
which is accompanied by a MOHO
build-up of p CO2 in the melt. c
In the highly evolved phonolite
that is now saturated in C O2, a
carbonatitic melt exsolves and
separates from the phonolite. Lithosphere
Sapphires precipitate from the Asthenosphere
carbonatite after physical sepa-
ration. d Fresh pulses of alka- b
line mafic melts ascend. When
encountering the carbonatite
lenses, the alkaline mafic melts
decompose the carbonatite,
incorporate the sapphires and
carry them to the surface
MOHO
Lithosphere
Asthenosphere
MOHO
Lithosphere
Asthenosphere
primitive alkali
MOHO basaltic melt
fractionated
phonolitic melt
CO2
Lithosphere sapphire
Asthenosphere
spinel-rimmed
sapphire
13
43
inclusions. It is rather the sum of single pieces of mineral- Brooker RA, Hamilton DL (1990) Three-liquid immiscibility and the
ogical evidence from the whole sample set that inferred a origin of carbonatites. Nature 346:459–462
Brooker RA, Kjarsgaard BA (2010) Silicate-carbonate liq-
carbonatite source for the Siebengebirge sapphires. There- uid immiscibility and phase relations in the system
fore, it cannot be excluded that other magmatic sapphires SiO2-Na2O-Al2O3-CaO-CO2 at 0.1–2.5 GPa with applications
that do not bear mineralogical evidence for a carbonatite to carbonatite genesis. J Petrol 52:1281–1305
origin do not have the same petrogenetic history. Hence, an Che X-D, Wu F-Y, Wang R-Ch, Gerdes A, Ji W-Q, Zhao Z-H, Yang
J-H, Zhu Z-Y (2015) In situ U-Pb isotopic dating of columbite-
overall applicability of our petrogenetic model for the origin tantalite by LA–ICP–MS. Ore Geol Rev 65:979–989
of magmatic sapphires should be considered, but cannot be Coenraads RR (1990) Key areas for alluvial diamond and sapphire
stated with certainty. However, from the findings of not only exploration in the New England gem fields, New South Wales,
our study, but also from referenced herein, we can state with Australia. Econ Geol 85:1186–1207
Coenraads RR, Sutherland FL, Kinny PD (1990) The origin of sap-
some confidence that in general, sapphire crystallizing melts phires: U-Pb dating of zircon inclusions shed new light. Miner
were produced via melting of C O2-rich, probably carbon- Mag 54:113–122
atite metasomatized, source rocks. Dasgupta R, Hirschmann MM, Smith ND (2007) Partial melting
experiments of peridotite+CO2 at 3 GPa and genesis of alkalic
Acknowledgements Open access funding provided by University ocean island basalts. J Petrol 48:2093–2124
of Vienna. We would like to thank Renate Schumacher for providing Dill HG, Gerdes A, Weber B (2007) Cu-Fe-U phosphate mineraliza-
the sapphire samples from the mineralogical collection of the Univer- tion of the Hagendorf-Pleystein pegmatite province, Germany:
sity of Bonn, and Nils Jung for sample preparation. Furthermore, we with special reference to laser-ablation-inductive-coupled-
would like to thank two anonymous reviewers for their constructive plasma mass spectrometry (LA–ICP–MS) of limonite-cored
comments, helping to improve the manuscript. This is contribution torberite. Miner Mag 71:423–439
no 44 of the DFG funded LA–ICP–MS laboratory of the Steinmann Dobson DP, Jones AP, Rabe R, Toshimori S, Kurita K, Taniguchi
Institute, University of Bonn. T, Kondo T, Kato T, Shimomura O, Urakawa S (1996) In-situ
measurement of viscosity and density of carbonate melts at high
pressure. Earth Planet Sci Lett 143:207–215
Open Access This article is distributed under the terms of the Crea- Freestone IC, Hamilton DL (1980) The role of liquid immiscibility
tive Commons Attribution 4.0 International License (http://crea- in the genesis of carbonatites—an experimental study. Contrib
tivecommons.org/licenses/by/4.0/), which permits unrestricted use, Miner Petr 73:105–117
distribution, and reproduction in any medium, provided you give Gäbler H-E, Melcher F, Graupner T, Bahr A, Sitnikova M, Henjes-
appropriate credit to the original author(s) and the source, provide a Kunst F, Oberthür T, Brätz H, Gerdes A (2011) Speeding up the
link to the Creative Commons license, and indicate if changes were analytical workflow for coltan finger-printing by an integrated
made. mineral liberation analysis/LA–ICP–MS approach. Geostand
Geoanal Res 35:431–448
Garland MI (2002) The alluvial Sapphire Deposits of Western Mon-
References tana. (Ph.D. Thesis). University of Toronto
Garnier V, Ohnenstetter D, Giuliani G, Fallick AE, Phan Trong T,
Hoang Quang V, Van Pham L, Schwarz D (2005) Basalt petrol-
Abduriyim A, Sutherland FL, Belousova EA (2012) U-Pb age and ori- ogy, zircon ages and sapphire genesis from Dak Nong, southern
gin of gem zircon from the New England sapphire fields, New Vietnam. Miner Mag 69:21–38
South Wales, Australia. Aust J Earth Sci 59:1067–1081 Garrabos Y, Tufeu R, Le Neindre B, Zalczer G, Beysens D (1980)
Abduriyim A, Kamegata N, Nonguchi N, Kagi H, Sutherland FL, Rayleigh and raman scattering near the critical point of carbon
Coldham T (2014) Residual pressure distribution and visualiza- dioxide. J Chem Phys 72:4637–4651
tion of mineral inclusions in corundum: application of photolu- Genge MJ, Price GD, Jones AP (1995) Molecular dynamics simula-
minescence spectroscopy in relation to sapphires from New Eng- tions of C aCO3 melts to mantle pressures and temperatures:
land, New South Wales, Australia. Aust Gemmol 25:245–254 implications for carbonatite magmas. Earth Planet Sci Lett
Aspen P, Upton BGJ, Dicken AP (1990) Anorthoclase, sanidine and 131:225–238
associated megacrysts in Scottish alkali basalts: high pres- Gerdes A, Zeh A (2006) Combined U-Pb and Hf isotope LA-
sure syenitic debris from upper mantle sources? Eur J Mineral (MC- )-ICP-MS analyses of detrital zircons: comparison with
2:503–517 SHRIMP and new constraints for the provenance and age of
Berg RB (2007) Sapphires in the Butte-Deer Lodge Area, Montana. an Armorican metasediment in Central Germany. Earth Planet
Montana Bur Mines Geol Bull 134, p 59 Sci Lett 249:47–61
Berg RB (2014) Sapphires in the southwestern part of the Rock Creek Gerdes A, Zeh A (2009) Zircon formation versus zircon altera-
sapphire district. Montana, Bur Mines Geol Bull, p 135 tion—new insights from combined U-Pb and Lu-Hf in situ
Berger AL, Berg RB (2006) The Silver Bow sapphire occurrence, LA–ICP–MS analyses, and consequences for the interpreta-
Montana: evidence for a volcanic bedrock source for Montana’s tion of Archean zircon from the Central Zone of the Limpopo
alluvial sapphire deposits. Econ Geol 101:679–684 Belt. Chem Geol 261:230–243
Bertrán JF (1983) Study of the Fermi doublet v1-2v2 in the Raman spec- Giuliani G, Fallick AE, Garnier V, France-Lanord Ch, Ohnestetter
tra of C
O2 in different phases. Spectrochim Acta 39A:119–121 D, Schwarz D (2005) Oxygen isotope composition as a tracer
Brauns R (1922) Die Mineralien der Niederrheinischen Vulkangebi- for the origin of ruby and sapphires. Geology 33:249–252
ete mit besonderer Berücksichtigung ihrer Bildung und Umbil- Giuliani G, Fallick AE, Rakotondrazafy M, Ohnestetter D,
dung. Schweizerbart, Stuttgart Andriamamonjy A, Rakotosamizanay S, Ralantoarison
Brooker RA (1998) The effect of CO2 saturation on immiscibility Th, Razanatseheno M, Dungaigre Ch, Schwarz D (2007)
between silicate and carbonate liquids: an experimental study. J Oxygen isotope systematic of gem corundum deposits in
Petrol 39:1905–1915
13
Madagascar: relevance for their geological origin. Miner Junge M, Oberthür T, Melcher F (2014) Cryptic variation of chromite
Depos 42:251–270 chemistry, platinum group element and platinum group mineral
Giuliani G, Fallick A, Ohnestetter D, Pegere G (2009) Oxygen iso- distribution in the UG-2 chromitite: an example from the Karee
topes composition of sapphires from the French Massif Central: Mine, western Bushveld Complex, South Africa. Econ Geol
implications for the origin of gem corundum in basaltic fields. 109:795–810
Miner Depos 44:221–231 Khamloet P, Pisutha-Arnond V, Sutthirat C (2014) Mineral inclusions
Giuliani G, Ohnestetter D, Fallick AE, Groat LA, Fagan J (2014) in sapphire from the basalt-related deposit in Bo Phloi, Kan-
The geology and genesis of gem corundum. In: Groat LA (ed) chanaburi, western Thailand: indication of their genesis. Russ
Geology of gem deposits, 2nd edn. Mineralogica Association of Geol Geophys 55:1087–1102
Canada, Short Course Series 44, pp 29–112 Kiefert L, Schmetzer K (1987) Blue and yellow sapphire from
Giuliani G, Pivin M, Fallick AE, Ohnestetter D, Song Y, Demaiffe D Kaduna Province, Nigeria. J Gemmol 22:484–496
(2015) Geochemical and oxygen isotope signatures of mantle Kjarsgaard B, Hamilton DL (1988) Liquid immiscibility and the ori-
corundum megacrysts from the Mbuji-Mayi kimberlite, Demo- gin of alkali-poor carbonatites. Miner Mag 52:43–55
cratic Republic of Congo, and the Changle alkali basalt, China. Kjarsgaard B, Peterson T (1991) Nephelinite-carbonatite liquid
CR Geosci 347:24–34 immiscibility at Shombole volcanoe, East Africa: petrographic
Graham IT, Sutherland FL, Web GB, Fanning CM (2004) Polygenetic and experimental evidence. Miner Petrol 43:293–314
corundums from New South Wales gemfields. In: Khanchuk AI, Kogarko LN (1997) Role of CO2 on differentiation of ultramafic alka-
Gonevchuk GA, Mitrokhin AN, Simaneko IF, Cook NJ, Selt- line series: liquid immiscibility in carbonate-bearing phonolitic
mann R (eds) Metallogeny of the Pacific Northwest: Tectonics, dykes (Polar Siberia). Miner Mag 61:549–556
magmatism and metallogeny of active continental margins. Dal- Kolb M, Paulik H, Kirchenbaur M, Münker C (2012) Petrogenesis of
nauka, Vladivostok, pp 336–339 mafic to felsic lavas from the Oligocene Siebengebirge Volcanic
Graham IT, Sutherland L, Zaw K, Nechav V, Khanchuk A (2008) Field (Germany): implications for the origin of intracontinental
Advances in our understanding of the gem corundum deposits volcanism in Central Europe. J Petrol 53:2349–2379
of the West Pacific continental margins intraplate basaltic fields. Koster van Groos AK, Wyllie PJ (1963) Experimental data bearing on
Ore Geol Rev 34:200–215 the role of liquid immiscibility in the genesis of carbonatites.
Green DH, Falloon TJ (1998) Pyrolite: a Ringwood concept and its Nature 199:801–802
current expression. In: Jackson I (ed) The Earth’s mantle: com- Koster van Groos AK, Wyllie PJ (1966) Liquid immiscibility in the
position, structure, and evolution. Cambridge University Press, system Na2O–Al2O3–SiO2–CO2 at pressures to 1 kilobar. Am J
Cambridge, pp 311–378 Sci 264:234–255
Guo J, Wang F, Yakoumelos G (1992) Sapphires from Changle in Krzemnicki M, Hänni HA, Guggenheim R, Mathys D (1996) Investi-
Shandong Province, China. Gems Gemol 28:255–260 gations on sapphires from an alkali basalt, South West Rwanda.
Guo J, O’Reilly SY, Griffin WL (1996) Corundum from basaltic ter- J Gemmol 25:90–106
rains: a mineral inclusion approach to the enigma. Contrib Le Maitre RW (2002) A classification and glossary of terms; recom-
Miner Petr 122:368–386 mendations of the International Union of Geological Sciences:
Hamilton DL, Freestone IC (1979) Origin of carbonatites by liquid Subcommision on the systematics of igneous rocks. Cambridge
immiscibility. Nature 279:52–54 University Press, Cambridge
Hirose K (1997) Partial melt compositions of carbonated peridotite at Lee W-J, Wyllie PJ (1997) Liquid immiscibility between nephelinite
3 GPa and role of CO2 in alkali-basalt magma generation. Geo- and carbonatite from 1.0 to 2.5 GPa compared with mantle melt
phys Res Lett 24:2837–2840 compositions. Contrib Miner Petr 127:1–16
Hughes RW (1997) Ruby and sapphire. RWH Publishing, Boulder Lee W-J, Wyllie PJ (1998) Process of crustal carbonatite formation
Co, USA by liquid immiscibility and differentiation, elucidated by model
Illies JH, Prodehl C, Schmincke H-U, Semmel A (1979) The quar- systems. J Petrol 39:2005–2013
ternary uplift of the rhenish shield in Germany. Tectonophysics L’Heureux I (2013) Self-organized rhythmic patterns in geochemi-
61:197–225 cal systems. Philos Trans R Soc Lon Math Phys Eng Sci
Irving AJ (1986) Polybaric magma mixing in alkali basalts and kim- 371(2004):20120356
berlites: evidence from corundum, zircon and ilmenite meg- Limtrakun P, Zaw K, Ryan CG, Mernagh TP (2001) Formation of the
acrysts. In: 4th International Kimberlite Conference, Perth, Denchai gem sapphires, northern Thailand: evidence from min-
Geological Society of Australia Abstr. Ser., vol. 16. Blackwell eral chemistry and fluid/melt inclusion characteristics. Miner
Scientific, Oxford, pp 262–264 Mag 65:725–735
Ishitani T, Ohnishi T, Kawanami Y (1990) Micromachining and Linthout K, Paulick H, Wijbrans JR (2009) Provenance of basalt
device transplantation using focused ion beam. Jpn J Appl Phys blocks from Roman sites in Vleuten-De Meern (The Nether-
29(10R):2283 lands) traced to the Tertiary Siebengebirge (Germany): a geoar-
Izokh AE, Smirnov SZ, Egorova VV, Anh TT, Kovyazin SV, Phuong cheological quest using petrological and geochemical methods.
NT, Kalinina VV (2010) The conditions of formation of sap- Geol Mijnbouw 88:78–99
phire and zircon in the areas of alkali-basaltoid volcanism in Liu T-C, Presnall DC (1990) Liquidus phase relationships on the join
Central Vietnam. Russ Geol Geophys 51:719–733 anorthite-forsterite-quartz at 20 kbar with applications to basalt
Jackson SE, Pearson NJ, Griffin WL, Belousova EA (2004) The applica- petrogenesis and igneous sapphirine. Contrib Miner Petrol
tion of laser ablation-inductively coupled plasma-mass spectrom- 104:735–742
etry to in situ U-Pb zircon geochronology. Chem Geol 211:47–69 Liu T-C, Presnall DC (2000) Liquidus phase relations in the system
Jochum KP, Weis U, Stoll B et al (2011) Determination of reference CaO–MgO–Al2O3–SiO2 at 2.0 GPa: applications to basalt frac-
values for NIST SRM 610–617 glasses following ISO guide- tionation, eclogites, and igneous sapphirine. J Petrol 41:3–20
lines. Geostan Geoanal Res 35:397–429 MacNevin AA (1972) Sapphires in the New England district, New
Jung S, Vieten K, Romer RL, Mezger K, Hoernes S, Satir M (2012) South Wales. Records of the geological survey of New South
Petrogenesis of tertiary alkaline magmas in the Siebengebirge, Wales 14:19–35
Germany. J Petrol 53:2381–2409
13
43
Malikova P (1999) Origin of the alluvial sapphires from the Jizerska Rosso KM, Bodnar RJ (1995) Microthermometric and Raman spec-
Louka alluvial deposits in North Bohemia, Czech Republic, troscopic detection limits of C O2 in fluid inclusions and the
Europe. Aust Gemol 20:202–206 Raman spectroscopic characterization of C O2. Geochim Cos-
McGee BM (2005) Characteristics and origin of the Weldborough mochim Ac 59:3961–3975
sapphire, NE Tasmania. (Unpublished BSc thesis): University Saminpanya S (2001) Ti-Fe mineral inclusions in star sapphires from
of Tasmania, School of Earth Science, Hobart, Australia Thailand. Aust Gemmol 21:125–128
Melcher F, Sitnikova MA, Graupner T, Martin N, Oberthür T (2008) Schubert S, Jung S, Pfänder JA, Hauff F, Garbe-Schönberg D (2015)
Analytical fingerprint of columbite-tantalite (coltan) minerali- Petrogenesis of Tertiary continental intra-plate lavas between
zation in pegmatites: focus on Africa. In: Proceedings, Ninth Siebengebirge and Westerwald, Germany: constraints from
International Congress for Applied Mineralogy (ICAM), pp trace element systematics and Nd, Sr and Pb isotopes. J Vol-
615–624 canol Geoth Res 305:84–99
Melcher F, Graupner T, Gäbler H-E, Sitnikova M, Henjes-Kunst F, Schwarz D, Kanis J, Schmetzer K (2000) Sapphires from Ant-
Oberthür T, Gerdes A, Dewaele S (2015) Tantalum-(niobium- siranana Province, Northern Madagascar. Gems Gemol
tin) mineralisation in African pegmatites and rare metal gran- 36:216–233
ites: constraints from Ta-Nb oxide mineralogy, geochemistry Sen G, Presnall DC (1984) Liquidus phase relations on the join
and U-Pb geochronology. Ore Geol Rev 64:667–719 anorthite-forsterite-quartz at 10 kbar with applications to basalt
Merino E, Wang Y (2001) Geochemical self-organization in rocks: petrogenesis. Contrib Mineral Petrol 85:404–408
occurrences, observations, modeling, testing—with emphasis Shore R, Weldon JL (2009) Ruby and sapphire production and distri-
on agate genesis. Yearb Self-Org 11:13–45 bution: a quarter century of change. Gems Gemol 45:236–259
Milholland CS, Presnall DC (1998) Liquidus phase relations in the Simonet C, Fritsch E, Lasnier B (2008) A classification of gem corun-
CaO-MgO-Al2O3-SiO2 system at 3.0 GPa: the aluminous dum deposits aimed towards gem exploration. Ore Geol Rev
pyroxene thermal divide and high-pressure fractionation of pic- 34:127–133
ritic and komatiitic magmas. J Petrol 39:3–27 Smith SR, Foster GL, Romer RL, Tindle AG, Kelley SP, Noble SR,
Nechaev VP, Nechaeva EV, Chashchin AA, Vysotskiy SV, Graham IT, Horstwood M, Breaks FW (2004) U-Pb columbite-tantalite
Sutherland FL (2009) New isotopic data on late Cenozoic age chronology of rare-element pegmatites using TIMS and Laser
and mantle origin of gem corundum from placers of Primorye, Ablation-Multi-Collector-ICP-MS. Contrib Miner Petrol
Russia. Geology 429A:1426–1429 147:549–564
Ohnishi T, Kawanami Y, Ishitani T (1990) Proposal for device Song Y, Hu W (2009) Carbonates and sulfates-bearing melt inclusions
transplantation using a focused ion beam. Japan J Appl Phys in corundum megacrysts from Changle basalts of Shandong
29(1A):L188 province and their implications. Acta Petrol Miner 28:349–363
Oncken O, Winterfeld C, Dittmar U (1999) Accretion of a rifted pas- Stacey JT, Kramers JD (1975) Approximation of terrestrial lead iso-
sive margin: the Late Paleozoic Rhenohercynian fold and thrust tope evolution by a two-stage model. Earth Planet Sci Lett
belt (Middle European Variscides). Tectonics 18:75–91 26(2):207–221
Ortoleva PJ (1994) Geochemical self-organization (p 411). Oxford Stephenson PJ (1976) Sapphire and zircon in some basaltic rocks
University Press, Clarendon Press, Oxford from Queensland, Australia. In: Abstracts 25th International
Pakhomova VA, Zalishchak BL, Odarichenko EG, Lapina MI, Kar- Geological Congress, vol. 2
manov NS (2006) Study of melt inclusions in the Nezametnoye Sutherland FL (1996) Alkaline rocks and gemstones, Australia: a
corundum deposit, Primorsky region of the Russian Far East: review and synthesis. Aust J Earth Sci 43:323–343
petrogenetic consequences. J Geochem Explor 89:302–305 Sutherland FL, Schwarz D (2001) Origin of gem corundums from
Palke AC, Renfro NB, Berg RB (2015) Glassy melt inclusions in basaltic fields. Aust Gemmol 21:30–33
sapphires from the Rock Creek and Missouri River deposits in Sutherland FL, Schwarz D, Jobbins EA, Coenraads RR, Webb G
Montana, USA. Geol Soc Am Abstr Progr 47:200 (1998a) Distinctive gem corundum suites from discrete basalt
Palke AC, Renfro ND, Berg RB (2016) Origin of sapphires from a fields: a comparative study of Barrington, Australia, and West
lamprophyre dike at Yogo Gulch, Montana, USA: clues from Pailin, Cambodia, gemfields. J Gemmol 26:65–85
their melt inclusions. Lithos 260:339–344 Sutherland FL, Hoskin POW, Fanning CM, Coenraads RR (1998b)
Palke AC, Renfro ND, Berg RB (2017) Melt inclusions in alluvial sap- Models of corundum origin from alkali basaltic terrains: a reap-
phires from Montana, USA: formation of sapphires as a restitic praisal. Contrib Miner Petrol 133:356–372
component of lower crustal melting? Lithos 278–281:43–53 Sutherland FL, Graham IT, Pogson RE, Schwarz D, Webb GB,
Pardieux V, Sansawong S, Mayal J, Sturman N (2014) Blue sapphires Coenraads RR, Fanning CM, Hollis JD, Allen TC (2002) The
from the Mambilla Plateau, Taraba State, Nigeria. A prelimi- Tumbarumba basaltic gem field, New South Wales. In relation
nary exploration. GIA News from Research, GIA Laboratory to sapphire-ruby deposits of eastern Australia. Rec Aust Mus
Bangkok 54:215–248
Peucat JJ, Ruffault P, Fritsch E, Bouhnik-Le Coz M, Simonet C, Las- Sutherland FL, Duroc-Danner JM, Meffre S (2008) Age and origin of
nier B (2007) Ga/Mg ratio as a new geochemical tool to dif- gem corundum and zircon megacrysts from the Mercades-Rio
ferentiate magmatic from metamorphic blue sapphires. Lithos Mayo area, South-west Colombia, South America. Ore Geol
98:261–274 Rev 34:155–168
Pfänder JA, Jung S, Münker C, Stracke A, Mezger K (2012) A pos- Sutherland FL, Giuliani G, Fallick AE, Garland M, Webb G (2009a)
sible high Nb/Ta reservoir in the continental lithospheric man- Sapphire and ruby characteristics, West Palin, Cambodia: clues
tle and consequences on the global Nb budget—Evidence from to their origin based on trace element and O isotope analysis.
continental basalts from Central Germany. Geochim Cosmo- Aust Gemmol 23:373–432
chim Ac 77:232–251 Sutherland FL, Zaw K, Meffre S, Giuliani G, Fallick AE, Graham IT,
Pitzer KS, Sterner SM (1994) Equations of state valid continuously Webb GB (2009b) Gem-corundum megacrysts from eastern
from zero to extreme pressures for H 2O and C
O2. J Chem Phys Australian basalt fields: trace elements, oxygen isotopes and
101:3111–3116 origins. Aust J Earth Sci 6:1003–1022
Przybyla M (2013) 40Ar/39Ar-Datierungen an Vulkaniten des Sie- Sutherland FL, Coenraads RR, Abduriym A, Meffre S, Hoskin
bengebirges. Msc thesis, University of Cologne, Germany PWO, Giuliani G, Beattie R, Wuhrer R, Sutherland GB (2015)
13
Corundum (sapphire) and zircon relationships, Lava Plains gem Wirth R (2004) Focused Ion Beam (FIB) A novel technology for
fields, NE Australia: integrated mineralogy, geochemistry, age advanced application of micro- and nanoanalysis in geosciences
determination, genesis and geographical typing. Miner Mag and applied mineralogy. Eur J Miner 16:863–876
79:545–581 Wirth R (2009) Focused Ion Beam (FIB) combined with SEM and
Sutherland FL, Graham IT, Harris SJ, Coldham T, Powell W, Belou- TEM: advanced analytical tools for studies of chemical compo-
sova EA, Martin L (2017) Unusual ruby-sapphire transition in sition, microstructure and crystal structure in geomaterials on a
alluvial megacrysts, Cenozoic basaltic gem field, New England, nanometre scale. Chem Geol 261:217–229
New South Wales, Australia. Lithos 278:347–360 Wirth R, Reid D, Schreiber A (2013) Nanometer-sized platinum-
Todt W, Lippolt HJ (1980) K-Ar age determination on Tertiary vol- group minerals (PGM) in base metal sulfides: new evidence
canic rocks: V Siebengebirge, Siebengebirge-Graben. J Geo- for an orthomagmatic origin of the Merensky Reef PGE
phys-Z Geophys 48:18–27 ore deposit, Bushveld Complex, South Africa. Can Miner
Uher P, Giuliani G, Szakall S, Fallick A, Strunga V, Vaculovic T, 51:143–155
Ozdin D, Greganova M (2012) Sapphires related to alkali Wolff JA (1994) Physical properties of carbonatite magmas inferred
basalts from the Cerová Highlands, Wesatern Carpathians from molten salt data, and application to extraction pat-
(southern Slovakia): composition and origin. Geol Carpath terns from carbonatite-silicate magma chambers. Geol Mag
63:71–82 131:145–153
Van Long P, Quang Vinh H, Garnier V, Giuliani G, Ohnestetter D, Wright JB (1985) Geology and mineral resources of West Africa.
Lhomme T, Schwarz D, Fallick AE, Dubessy J, Trong Trinh Allen and Unwin, London
P (2004) Gem corundum deposits in Vietnam. J Gemmol Yamamoto J, Kagi H (2006) Extended micro-Raman densimeter for
29:129–147 CO2 applicable to mantle-originated fluid inclusions. Chem Lett
Vichit P (1987) Gemstones in Thailand. J Geol Soc Thailand 35:610–611
9:108–133 Yamamoto J, Kagi H, Kaneoka I, Lai Y, Prikhod’ko VS, Arai S (2002)
Vieten K, Hamm HM, Grimmeisen W (1988) Tertiärer Vulkanismus Fossil pressures of fluid inclusions in mantle xenoliths exhib-
des Siebengebirges. Fortschr Der Mineral Beih 66:1–42 iting rheology of mantle minerals: implications for the geo-
Vysotskij SV, Barkar AV (2009) Sapphire of the Primorje Region. barometry of mantle minerals using micro-Raman spectroscopy.
Dal’nauka, Vladivostok (in Russian) Earth Planet Sci Lett 198:511–519
Wang X, Chou I-M, Hu W, Burruss RC, Sun Q, Song Y (2011) Raman Yui TF, Wu CM, Limtrakum P, Sricharn W, Boonsoog A (2006) Oxy-
spectroscopic measurements of C O2 density: experimental cali- gen isotope studies on placer sapphire and ruby in the Chant-
bration with high-pressure optical cell (HPOC) and fused silica aburi-Trat alkali basaltic gemfield, Thailand. Lithos 86:197–211
capillary capsule (FSCC) with application to fluid inclusion Zaw K, Sutherland FL, Dellapasqua F, Ryan CG, Yui T-F, Mernagh
observations. Geochim Cosmochim Ac 75:4080–4093 TP, Duncan D (2006) Contrasts in gem corundum characteris-
Wedepohl KH, Gohn E, Hartmann G (1994) Cenozoic alkali basaltic tics, eastern Australian basaltic fields: trace elements, fluid/melt
magmas of western Germany and their products of differentia- inclusions and oxygen isotopes. Miner Mag 70:669–687
tion. Contrib Miner Petrol 115:253–278 Zelong D, Xiaoming C, Wenxuan H, RuCheng W, Ming Z (2007)
Whalen JB, Curie KL, Chappell BW (1987) A-type granites: geo- Coronas of corundum megacrysts in the Neogene Changle
chemical characteristics, discrimination and petrogenesis. Con- basalt and its forming model. Acta Petrol Sinca 23:805–816
trib Miner Petrol 95:407–419 Zeng G, Chen L-H, Xu X-S, Jiang S-Y, Hofmann AW (2010) Carbon-
Whitney D, Evans B (2010) Abbreviations for names of rock-forming ated mantle sources for Cenozoic intra-plate alkaline basalts in
minerals. Am Miner 95:185–187 Shandong, North China. Chem Geol 273:35–45
Wilson M, Downes H (1991) Tertiary-Quarternary extension related Ziegler PA (1992) European Cenozoic rift system. Tectonophysics
alkaline magmatism in western and central Europe. J Petrol 208:91–111
32:811–849
13

Baldwin2017 Article PetrogenesisOfAlkalineBasalt-h

Uploaded by

Copyright:

Available Formats

Baldwin2017 Article PetrogenesisOfAlkalineBasalt-h

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Baldwin2017 Article PetrogenesisOfAlkalineBasalt-h

Uploaded by

Copyright:

Available Formats

Contrib Mineral Petrol (2017) 172:43

Petrogenesis of alkaline basalt‑hosted sapphire megacrysts.

Fig. 1 Volcanic Fields of the

3 Jungfernberg Basanite pale-blue, milky 3 Irregular 70–80 Calcite, Nb-rich

a b plagioclase dominated basalt

‚normal‘ blue sapphire

Table 2 XRF analyses of the host rocks of the Siebengebirge sapphires

­ e3+/Fetot ratio of the basalt is 0.1

Fig. 6 Inclusions of glasses that

Fig. 7 Fluid inclusions in Sie-

Questions addressed in the discussion of this paper are

62 Magmatic vs. metamorphic origin

Accepting a xenogenetic relation between the sapphires

if they formed via igneous or metamorphic processes.

Table 5 U-Th–Pb isotopic data for the columbite-Mn inclusion

data-pint error ellipses are 2 σ

0.0034 22 Mean = 24.74±0.24 [0.96%] 95% conf.

Spl syenites, or nepheline syenites. A further observation lead-

ing these authors to favor syenites as parental melts is that

repeatedly melt inclusions of syenitic composition were

reported from magmatic sapphires (e.g., Pakhomova et al.

2006; Zaw et al. 2006; Izokh et al. 2010). Up until now,

1 1 for the sapphire formation. Only Guo et al. (1996) and

Na2O + K2O would be a carbonatite containing ≤5 wt% ­SiO2+Al2O3.

Two-liquid field surface

triggering alkaline basaltic magmatism and the crystalli- 1.5

Fig. 15 Petrogenetic model for a

You might also like

Fig. 1 Volcanic Fields of the

Table 2 XRF analyses of the host rocks of the Siebengebirge sapphires

e3+/Fetot ratio of the basalt is 0.1

Fig. 6 Inclusions of glasses that

Fig. 7 Fluid inclusions in Sie-

Table 5 U-Th–Pb isotopic data for the columbite-Mn inclusion

Na2O + K2O would be a carbonatite containing ≤5 wt% SiO2+Al2O3.

Fig. 15 Petrogenetic model for a