Petrogenesis of the Ultrapotassic Fanshan Intrusion in the North China Craton: Implications for Lithospheric Mantle Metasomatism and the Origin of Apatite Ores Tong Hou 1,2,3 , Zhaochong Zhang 1 *, Jakob K. Keiding 4 and Ilya V. Veksler 4,5 1 State Key Laboratory of Geological Process and Mineral Resources, China University of Geosciences, Beijing 100083, China, 2 Institut fu ¨ r Mineralogie, Leibniz Universita ¨t Hannover, Callinstrasse 3, Hannover, D-30167, Germany, 3 Department of Petrology and Economic Geology, Geological Survey of Denmark and Greenland (GEUS), Øster Voldgade 10, DK-1350 Copenhagen K, Denmark, 4 Helmholtz Centre Potsdam GFZ German Research Centre for Geosciences, Section 4.2, Telegrafenberg, Potsdam, D-14473, Germany and 5 Department of Mineralogy, Technical University Berlin, Ackerstrasse 71–76, Berlin 13555, Germany *Corresponding author. Telephone: þ86 10 82322195. Fax: þ86 10 82323419. E-mail: [email protected]. Received February 23, 2014; Accepted April 8, 2015 ABSTRACT The Fanshan intrusion in the North China Craton (NCC) is concentrically zoned with syenite in the core (Unit 1), surrounded by ultramafic rocks (clinopyroxenite and biotite clinopyroxenite; Unit 2), and an outer rim of garnet-rich clinopyroxenite and orthoclase clinopyroxenite and syenite (Unit 3). The intrusive rocks are composed of variable amounts of Ca-rich augite, biotite, orthoclase, melanite, garnet, magnetite and apatite, with minor primary calcite. Monomineralic apatite rocks, nelsonite and glimmerite exclusively occur in Unit 2. Geochemically, the Fanshan rocks are highly enriched in light rare earth elements (LREE) and large ion lithophile elements (LILE), moderately depleted in high field strength elements (HFSE), and have a limited range of Sr–Nd–O isotopic compositions. The similar mineralogy, mineral compositions, and trace element characteristics of the three units sug- gest that all the rocks are co-magmatic. The parental magma is ultrapotassic and is akin to kamafu- gite. Very low-degree partial melting of metasomatized lithospheric mantle best explains the geo- chemistry and petrogenesis of the parental magmas of the Fanshan intrusion. We propose that the mantle source may have been metasomatized by a hydrous carbonate-bearing melt, which has im- printed the enriched Sr–Nd isotopic signature and incompatible element enrichment with conspicu- ous negative Nb–Ta–Zr–Hf–Ti anomalies and LREE enrichments. The mantle source enrichment may be correlated with oceanic sediment recycling during southward subduction of the Paleo-Asian oce- anic plate during the Carboniferous and Permian. We propose that crystal settling and mechanical sorting combined with repeated primitive magma replenishment and mixing with previously fractio- nated magma is the predominant process responsible for the formation of the apatite ores. Key words: apatite; ultrapotassic; Fanshan Intrusion; North China Craton; mantle metasomatism INTRODUCTION The study of ultrapotassic rocks, which are rare and volumetrically minor, has been justified by their genetic link with terrestrial mantle evolution and specific tectonic settings (e.g. Foley et al., 1987; Miller et al., 1999; Conticelli et al., 2007, 2009, 2013), as well as their role in the generation of economic mineral deposits (e.g. Dill, 2010). Although they can provide valuable V C The Author 2015. Published by Oxford University Press. All rights reserved. For Permissions, please e-mail: [email protected]1 J OURNAL OF P ETROLOGY Journal of Petrology, 2015, 1–26 doi: 10.1093/petrology/egv021 Original Article Journal of Petrology Advance Access published May 26, 2015 at Georgetown University on May 26, 2015 http://petrology.oxfordjournals.org/ Downloaded from
26
Embed
Petrogenesis of the Ultrapotassic Fanshan Intrusion in the ...
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Petrogenesis of the Ultrapotassic Fanshan
Intrusion in the North China Craton:
Implications for Lithospheric Mantle
Metasomatism and the Origin of Apatite Ores
Tong Hou1,2,3, Zhaochong Zhang1*, Jakob K. Keiding4 and
Ilya V. Veksler4,5
1State Key Laboratory of Geological Process and Mineral Resources, China University of Geosciences, Beijing
100083, China, 2Institut fur Mineralogie, Leibniz Universitat Hannover, Callinstrasse 3, Hannover, D-30167, Germany,3Department of Petrology and Economic Geology, Geological Survey of Denmark and Greenland (GEUS), Øster
Voldgade 10, DK-1350 Copenhagen K, Denmark, 4Helmholtz Centre Potsdam GFZ German Research Centre for
Geosciences, Section 4.2, Telegrafenberg, Potsdam, D-14473, Germany and 5Department of Mineralogy, Technical
University Berlin, Ackerstrasse 71–76, Berlin 13555, Germany
Received February 23, 2014; Accepted April 8, 2015
ABSTRACT
The Fanshan intrusion in the North China Craton (NCC) is concentrically zoned with syenite in the
core (Unit 1), surrounded by ultramafic rocks (clinopyroxenite and biotite clinopyroxenite; Unit 2),
and an outer rim of garnet-rich clinopyroxenite and orthoclase clinopyroxenite and syenite (Unit 3).
The intrusive rocks are composed of variable amounts of Ca-rich augite, biotite, orthoclase, melanite,
garnet, magnetite and apatite, with minor primary calcite. Monomineralic apatite rocks, nelsonite
and glimmerite exclusively occur in Unit 2. Geochemically, the Fanshan rocks are highly enriched in
light rare earth elements (LREE) and large ion lithophile elements (LILE), moderately depleted in highfield strength elements (HFSE), and have a limited range of Sr–Nd–O isotopic compositions. The
similar mineralogy, mineral compositions, and trace element characteristics of the three units sug-
gest that all the rocks are co-magmatic. The parental magma is ultrapotassic and is akin to kamafu-
gite. Very low-degree partial melting of metasomatized lithospheric mantle best explains the geo-
chemistry and petrogenesis of the parental magmas of the Fanshan intrusion. We propose that the
mantle source may have been metasomatized by a hydrous carbonate-bearing melt, which has im-printed the enriched Sr–Nd isotopic signature and incompatible element enrichment with conspicu-
ous negative Nb–Ta–Zr–Hf–Ti anomalies and LREE enrichments. The mantle source enrichment may
be correlated with oceanic sediment recycling during southward subduction of the Paleo-Asian oce-
anic plate during the Carboniferous and Permian. We propose that crystal settling and mechanical
sorting combined with repeated primitive magma replenishment and mixing with previously fractio-
nated magma is the predominant process responsible for the formation of the apatite ores.
Key words: apatite; ultrapotassic; Fanshan Intrusion; North China Craton; mantle metasomatism
INTRODUCTION
The study of ultrapotassic rocks, which are rare and
volumetrically minor, has been justified by their geneticlink with terrestrial mantle evolution and specific
tectonic settings (e.g. Foley et al., 1987; Miller et al.,
1999; Conticelli et al., 2007, 2009, 2013), as well as their
role in the generation of economic mineral deposits
(e.g. Dill, 2010). Although they can provide valuable
VC The Author 2015. Published by Oxford University Press. All rights reserved. For Permissions, please e-mail: [email protected] 1
J O U R N A L O F
P E T R O L O G Y
Journal of Petrology, 2015, 1–26
doi: 10.1093/petrology/egv021
Original Article
Journal of Petrology Advance Access published May 26, 2015
conformably by Mesoproterozoic to Cenozoic sediment-
ary strata (Zhao et al., 2001).The northern margin of the NCC was strongly influ-
enced by the southward subduction of the Paleo-Asian
oceanic plate during Carboniferous to Permian times
(Xiao et al., 2003) with development of an Andean-style
continental margin during the Late Carboniferous–Early
Permian (Zhang et al., 2009). The final closure of the
North China Craton
Qilianshan Orogen
Qinling-Dabie OrogenYangtze Craton
Su-Lu Orogen240-210Ma
LA-ICP-MS
224±4MaSHRIMP
234±2MaLA-ICP-MS
Fanshan218±2MaSHRIMP
220±2MaSHRIMP
225-209MaLA-ICP-MS
220±5MaSHRIMP
Beijing
223±2Ma;222±4MaLA-ICP-MS300km
N40°
N30°
E100° E110° E120° E130°
N40°
N30°
Late Triassicalkaline intrusion
231±1MaSHRIMP
221±5MaSHRIMP
224±2MaSHRIMP
Solonker sutureCentral Asian Orogenic Belt
EAST CHINA SEA
Fault
Datong
Yaojiazhuang
Fig. 1. Simplified tectonic map of North China showing the locations of Late Triassic alkaline intrusions [modified from Ren et al.(2009)]. Age data and analytical methods are compiled from Zhang et al. (2012) and references therein.
Paleo-Asian Ocean and amalgamation of Mongolian arc
terranes with the NCC along the Solonker suture
occurred between the Late Permian and earliest Triassic
(Zhang et al., 2012). Following this closure, post-colli-
sional Triassic (250–200 Ma) alkaline intrusions wereemplaced along an east–west-trending belt parallel to
the northern margin of the NCC (Fig. 1). The Early
Triassic alkaline rocks consist mainly of monzogranite,
K-feldspar granite and minor monzonite, whereas the
Middle–Late Triassic alkaline rocks include syenite and
peralkaline granite, as well as associated mafic–ultra-
mafic intrusions such as the Fanshan and Yaojiazhuangintrusions, which host magmatic apatite ores (Fig. 1).
Coeval lamprophyre and carbonatite dyke swarms have
been recognized in the Datong region of Shanxi prov-
ince, close to the northern margin of the NCC (Shao
et al., 2003; Fig. 1). In contrast to the Fanshan intrusion,
which contains nearly monomineralic apatite layers, apa-
tite is present throughout the Yaojiazhuang intrusion, lo-
cally in modal proportions up to 15% (Chen et al., 2013).
FANSHAN INTRUSION
GeologyThe Fanshan intrusion was emplaced at c. 218 Ma (Ren
et al., 2009; Niu et al., 2012), at the intersection between
pre-existing NNW–SSE- and east–west-trending faults,
into Meso- to Neoproterozoic limestones and clastic
rocks of the Wumishan Formation. It is an �6� 5 km
oval-shaped body in plan (Fig. 2). However, much of the
pluton is covered by over 100 m of Quaternary sedi-ments, and thus the present geometry of the intrusion
Syenite dykeClinopyroxenesyenite
Fe-P orebodies
Clinopyroxenitedominated rocks
Orthoclaseclinopyroxenite
Garnet-richclinopyroxenite
Late Cretaceous granodiorite
Z2w
Z2w
N
Z2wMesoproterozoic limestones
Z2w
BA
-200
200
600(m)
Height
(a)
Quaternary sediments
Drill hole
of samples from units 2-3
Samples from unit 1(b)
LENDEND
B
A
0 1km
Unit 1
Unit 2
Unit 3
Unit 1
Unit 2
Unit 3
Fig. 2. Geological map (a) and cross-section (b) of the Fanshan intrusion. Modified from Jiang et al. (2004). Most of the intrusion iscovered by Quaternary sediments, and the geological map is largely inferred from borehole and geophysical data. The dashed linein (a) indicates the approximate near-surface locations of samples collected from Units 2 and 3. A–B indicates the line of section in(b). The position of the borehole from which the clinopyroxene syenite samples of Unit 1 were obtained is indicated.
Fig. 3. Schematic stratigraphy of the Fanshan intrusion showing the stratigraphic position of the samples of Unit 2 and garnet clino-pyroxenite in Unit 3 [modified from Cheng & Sun (2003)]. The orthoclase clinopyroxenite samples of Unit 3 were collected from atunnel in which a complete lithological succession is exposed through mining, whereas for the syenite in Unit 1, four core samples15–25 cm long were collected randomly from a recently drilled borehole. The 0 m reference level corresponds to the contact zonebetween garnet clinopyroxenite and Mesoproterozoic limestones of the Wumishan Formation.
preferred orientation, but are usually bent and show
undulose extinction. Magnetite is rare.
Monomineralic apatite rocks and nelsonite dominate
the ore horizons (Fig. 6) and are composed mainly ofapatite and magnetite, with minor Ca-rich augite and
biotite. The monomineralic apatite rocks are light yel-
low to green in colour and very friable. They are me-
dium grained, granular, with a subhedral texture;
apatite contents can be up to 95 modal % (Fig. 6b). In
some of these rocks apatite displays 120� triple junc-tions (Fig. 5f). In those rocks containing biotite, apatite
typically occurs between euhedral or subhedral biotite
grains, and both show a preferred orientation. Most bio-
tite grains are bent and have undulose extinction. Ca-
rich augite is a minor interstitial phase and occurs as
subhedral or anhedral grains that crystallized later than
the biotite and apatite. Some inclusions composed ofbiotite and Ca-rich augite (Fig. 5g) with minor carbonate
(dolomite and calcite) have been observed within
apatite.
Nelsonite is also an important component of the ore
horizons; it is gray–black in colour (Fig. 4a), and con-
tains less apatite and biotite but more magnetite (up to40 modal %; Fig. 6d) and Ca-rich augite (up to 10 modal
%) than the monomineralic apatite rocks. Apatite is eu-
hedral and ranges from 0�5 to 5 mm in cross-section,
and up to 10 mm in length. Ca-rich augite occurs as eu-
hedral to subhedral aggregates and locally contains eu-
hedral apatite grains. In some samples biotite forms
coarse poikilitic plates enclosing small Ca-rich augitegrains. K-feldspar is absent in the nelsonite.
In most samples, Ca-rich augite occurs as a primoc-
ryst phase suggesting that it crystallized first, followed
by apatite, biotite and K-feldspar. Magnetite and calcite
are interstitial to the main rock-forming minerals in theultramafic rocks, and crystallized at a late stage. Apatite
accumulates together with magnetite in Units 2 and 3
(Fig. 5h), forming the layered Fe–P-rich lithologies
(Figs 3, 4a and 6).
Unit 3Unit 3 consists predominantly of orthoclase- and gar-
net-clinopyroxenite and syenite. Compared with the cli-
nopyroxenites in Unit 2, the orthoclase-clinopyroxenite
in this unit contains more interstitial K-feldspar, al-
though the contents of K-feldspar and Ca-rich augite
vary considerably. Locally, K-feldspar is the most abun-
dant mineral in lithologies such as Ca-rich augite syen-ite. The garnet-rich clinopyroxenites typically contain
�5 modal % melanite garnet, which locally reaches as
much as 35 modal %. They are composed of subhedral
garnet, Ca-rich augite, biotite, orthoclase and magnet-
ite. Euhedral apatite is common and titanite is also
present.
SAMPLE PREPARATION AND ANALYTICALMETHODS
Owing to the different conditions in the mine workings,we sampled the Fanshan intrusion using different meth-
ods. A total of 41 samples of Unit 2, 24 samples of Unit
Apatite rock
Nelsonite
(a) (b)
Glimmerite
Apatite rock
Syenite
Clinopyroxenite
(c) (d)
Garnet-richclinopyroxenite
Syenite
Fig. 4. Field photographs of Fanshan rocks from underground excavations. (a) Rhythmic layered rocks comprising alternatinglayers of monomineralic apatite rock and nelsonite, Unit 2. (b) Monomineralic apatite rocks occurring as enclaves in glimmerite,Unit 2. (c) Clinopyroxenite intruded by a syenite vein, Unit 2. (d) Garnet-rich clinopyroxenite intruded by a syenite vein, Unit 3.
Fig. 6. Mineral modes (see Supplementary Data) of (a) Ca-rich augite, (b) apatite, (c) biotite, (d) magnetite and (e) K-feldspar in Unit2 of the Fanshan intrusion as a function of stratigraphic height and stratigraphy of cumulus (grey) and intercumulus (white) phasesin Unit 2. The grey bands in (a)–(e) indicate the ore horizons.
the ore horizons (Fig. 7c). On an Fe3þ–Fe2þ–Mg diagram
most of the Fanshan biotites suggest crystallization
under relatively normal redox conditions for plutonic
rocks close to the fayalite-magnetite–quartz–magnetite
oxygen buffer (FMQ; Fig. 8c).
Melanite, magnetite and apatiteAll the analysed garnets are Ti-rich (1�8–12�3 wt %) and
classified as melanite (i.e. Fe3þ>Ti in the octahedral
position; Deer et al., 1992) with Ti¼ 0�142–0�778, Ca¼2�898–3�018 and Fe3þ¼ 1�193–1�620 (a.p.f.u.). Magnetite
shows a wide compositional range with TiO2 contents
ranging from 0�1 to 10�9 wt %, 0�0–3�6 wt % MgO and0�2–1�6 wt % MnO. However, no systematic trend was
found between the units and within the layered rocks
(Fig. 7f). Apatites in Units 2 and 3 are predominantly flu-
orapatite characterized by high F (1�8–2�1 wt %) and low
Cl contents (0�04–0�06 wt %); the F/Cl ratios are constant
(35–53) throughout Unit 2 (Fig. 7g).
Bulk-rock major and trace element compositionsAll analysed samples have low or negligible LOI values
except those containing considerable amounts of apatite
and biotite (Table 4). As one would expect from stronglymodally layered cumulates, the bulk-rocks exhibit large
compositional variations (Table 4). Specifically, the
(biotite-)clinopyroxenites in Unit 2 have low SiO2 and
high CaO, consistent with the high content of apatite in
these rocks. They also have variable Mg-numbers
(Table 4). The Al2O3, K2O and Na2O contents are lowerthan those of other rock types in the Fanshan intrusion.
The two glimmerite samples (FS-07 and FS-1-10) are
characterized by high MgO contents and relatively low
contents of total Fe2O3 compared with the clinopyroxen-
ites. Generally, the compositional variations can be
attributed to the varying proportions of Ca-rich augite
and biotite and to the variable amounts of intercumulus
phases (mainly orthoclase, apatite, and magnetite).Compared with the clinopyroxenite and glimmerite in
Unit 2, the syenites in Unit 1 have much higher SiO2,
Al2O3, K2O and Na2O contents, but lower CaO, MgO and
total Fe2O3, consistent with the dominance of orthoclase
in these rocks. The garnet-rich clinopyroxenites and
orthoclase-clinopyroxenites in Unit 3 show transitional
chemical compositions between the ultramafic rocks [i.e.(biotite-)clinopyroxenite and glimmerite] and the syenites
(Table 4), but the former exhibit dispersion of the data for
Na2O and TiO2, possibly owing to the presence of melan-
ite. In particular, the garnet-rich clinopyroxenites have
low Na2O contents and are characterized by higher TiO2
contents, with limited variation in SiO2, Al2O3, K2O andMgO, but variable CaO and total Fe2O3. The monominer-
alic apatite rocks have the highest P2O5 contents, corres-
ponding to almost pure apatite; the nelsonites also have
high Fe2O3 and P2O5 contents (Table 4).
Representative trace element compositions of the
Fanshan intrusive rocks are given in Table 4 and illus-
trated in chondrite-normalized and mantle-normalizeddiagrams in Figs 10 and 11. All the rock types are char-
acterized by significant enrichment in large ion litho-
phile elements (LILE), such as Sr, Ba and Rb, and LREE,
and display prominent troughs in Nb, Ta, Zr, Hf and Ti.
The garnet-clinopyroxenite and syenite samples
(Fig. 11e and f) show convex-upward REE patterns with
1500m
1400m
1300m
1200m
1100m
1000m
0.6 0.7 0.8 0.9 1.0Mg# in Cpx
0 0.5 1.0 1.5 2.0TiO2(wt.%) in Cpx
10 20 300MgO(wt.%)in Bt
K2O(wt.%)in Bt
9 9.5 10 10.5TiO2(wt.%)in Bt
0 2 4 6 0 5 10 15TiO2(wt.%)in Mt
0 20 40 60F/Cl in Ap
(a) (b) (c) (d) (e) (f) (g)
DIST
ANCE
Fig. 7. Major element compositional variations of (a) Ca-rich augite (Mg#), (b) Ca-rich augite (TiO2), (c) biotite (MgO), (d) biotite(K2O), (e) biotite (TiO2), (f) magnetite (TiO2), and (g) apatite (F/Cl ratio) with stratigraphic position in Unit 2 of the Fanshan intrusion.The grey bands indicate the ore horizons.
n, number of analysed points that are adjacent to the core of the crystals. Partition coefficient values are fromBedard (2014); the composition of a kamafugite from West Qinling, North China Craton is from Guo et al. (2014).
similar to that of Group II kamafugites. However, the
notable differences in Nb–Ta–Hf concentration between
kamafugites and our calculated parent magma may in-
dicate a different magma source or degree of crustalcontamination as discussed below.
Crustal contaminationThe Fanshan rocks are characterized by crustal-like
trace element and isotopic compositions such as the
relative depletion in high field strength elements (HFSE)(Figs 10 and 11) and enriched Sr–Nd isotopic compos-
itions [87Sr/86Srt¼ 0�70513–0�70601, eNd(t)¼ –6�8 to
–5�5]. Niu et al. (2012) proposed that the Fanshan parent
magma experienced contamination by mafic lower
crust, on the basis of Os isotope data. The relatively
high d18O values (þ7 to þ9%; Table 4) also indicate thepossibility of crustal contamination. Because the intru-
sion is emplaced into limestones of the Wumishan
Formation, contamination by limestone seems inev-
itable. Assimilation of limestone will drive the
crystallization of Ca-rich clinopyroxene, resulting in
desilication of the melt and an increase of Si-under-
saturation (Gaeta et al., 2006; Mollo & Vona, 2014), as
supported by several experimental studies (e.g. Iacono-Marziano et al., 2007, 2008, 2009; Freda et al., 2008;
Conte et al., 2009). However, the assimilation of carbon-
ate will have no significant effect on the 87Sr/86Sr and
LREE/HREE ratios (Conticelli et al., 2002; Perini et al.,
2004), and may be visible only in terms of oxygen
isotope ratios (Gaeta et al., 2006). Assimilation andfractional crystallization (AFC) modelling of the
Wo
En Fs
Wo(a)
(c)
Or60
80
30Ab
An(b)
Syenite, unit 1Clinopyroxenite, unit 2(Garnet-)clinopyroxenite, unit 3
Alm+Sp
from potassic and ultrapotassicrocks in central Italy
MH
NNOFMQ
Fe2+
Mg
Fe3+ Fe3+
Fig. 8. Compositional variations of minerals from the Fanshanintrusion. (a) Clinopyroxene (Ca-rich augite); field of clinopyr-oxene from potassic and ultrapotassic rocks in central Italy isfrom Cellai et al. (1994), Gaeta et al. (2006), Melluso et al.(2008) and Mollo & Vona (2014). (b) K-feldspar. (c) Biotite(Wones & Eugster 1965). Data for rocks from Unit 3 from Jianget al. (2004) and Niu et al. (2012) are also plotted.
100
1000
Th Nb Ta La Ce Pr Nd Hf Sm Eu Gd Tb Dy Y Ho Er Tm Yb Lu
10
Calculated parent
Kamafugitefrom Western Qinling
1
Th Nb Ta La Ce Pr Nd Hf Sm Eu Gd Tb Dy Y Ho Er Tm Yb Lu
(a)
(c)
Ca-rich augite from unit 2
La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu
Sam
ple/
Chro
ndrit
e
Ca-rich augite from unit 2
(b)1000
100
10
1
0.1
100
10
1
Fig. 9. Chondrite-normalized REE patterns (a) and primitivemantle-normalized trace element patterns (b) for the Ca-richaugite in Unit 2. (c) Calculated parent magma compositionscompared with that of kamafugite from Western Qinling in theNCC (Guo et al., 2014).
LOI, weight loss on ignition at 1000�C. Total iron oxide expressed as Fe2O3 (Fe2O3-t); Mg#¼ [molar Mg/(MgþFe2þ)]�100, assum-ing 15% of total iron is ferric. Chondritic uniform reservoir (CHUR) values [(143Sm/144Nd)CHUR
0¼0�512638, (143Nd/144Nd)CHUR0¼0
�1967] are used for the calculation. kRb¼1�42�10�11 a–1 (Steiger & Jager, 1977), kSm¼6�5�10�12 a–1 (Lugmair & Harti, 1978).(87Sr/86Sr)t and eNd(t) were calculated at 218 Ma. Rock type abbreviations as in Table 1.
final temperature of 760�C, H2O¼ 4 wt % and a pressure
of 10 kbar. The MELTS modelling shows that the
sequence of mineral crystallization is augite–biotite–
apatite–magnetite–garnet. This is consistent with petro-
graphic observations from Unit 2 and some of the rocks
in Unit 3. However, K-feldspar, which is an interstitialphases in these two units, is not present in the model
assemblage. It is probable that owing to the high water
content in the model parent magma most of the potas-
sium entered biotite. To test this, we conducted another
model run using the same starting composition with an
elevated K2O content (3 wt %) but low water content
(0�5 wt %) under lower pressure (2 kbar). The resultsshow that sanidine (K-feldspar) crystallizes after augite
and biotite, consistent with the mineral assemblage in
Unit 1. Therefore, our modelling suggests that the par-
ental magma generating the Fanshan intrusion may be
composionally similar to the kamafugite from Western
Qinling but with higher K2O and lower H2O contents.
Magma generation and nature of the mantlesourceLithospheric mantlePartial melting of metasomatized subcontinental litho-
spheric mantle is widely regarded as the most likelyprocess to explain the origin of Group II potassic and
ultrapotassic rocks (Peccerillo, 2005). Indeed, the Sr and
Nd isotope compositions of the investigated samples
(Fig. 12) fall far outside the ranges for oceanic basalts
[mid-ocean ridge basalt (MORB) and ocean-island bas-
alt (OIB)]. This argues against exclusively astheno-
spheric or mantle plume sources. The enrichment of
LREE and LILE and depletion of HFSE (Figs 10 and 11)support an origin from the lithospheric mantle. The
Fanshan rocks have La/Yb and Nb/La ratios consistent
with an origin from the lithospheric mantle (Fig. 13b;
Condie, 1997).
Subduction-related metasomatismThe enrichment of LILE (Rb, K, Th, U, Sr, and Pb) and
depletion of the high-field strength elements (HFSE; Nb
and Ti) and the HREE (Yb) are characteristic features of
magmas generated in suprasubduction-zone settings
(e.g. Wilson, 1989; Castillo & Newhall, 2004). The high
Th/Yb ratios above the MORB–OIB array (Fig. 15a) are
presumed to reflect the influence of subduction-zonefluids or melts enriched in Th in their petrogenesis. The
Fanshan intrusive rocks plot in the fields of arc volcanic
rocks in Fig. 15a and b. Thus, a likely scenario for the
petrogenesis of the ultrapotassic magmas is that a fluid
or melt derived from subducted pelagic or terrigeneous
sediments was channelled in the overlying lithosphericmantle, forming a zone of hybrid veined mantle
(Conticelli et al., 2013). As stated above, the northern
Fig. 10. Primitive mantle-normalized trace element and chondrite-normalized rare earth element (REE) patterns for the clinopyrox-ene syenite of Unit 1, and the monomineralic apatite rocks and nelsonites of Unit 2 of the Fanshan intrusion. The primitive mantleand chondrite normalizing values are from Sun & McDonough (1989).
Fig. 11. Primitive mantle-normalized trace element and chondrite-normalized rare earth element (REE) patterns of the (biotite-)clino-pyroxenite and glimmerite of Unit 2, the orthoclase clinopyroxenite, garnet-rich clinopyroxenite and syenite of Unit 3 and syeniticdykes from the Fanshan intrusion. The primitive mantle and chondrite normalizing values are from Sun & McDonough (1989).
suggesting a genetic link to the Palaeozoic subduction
of the Paleo-Asian oceanic slab. Furthermore, an exten-
sional tectonic regime probably developed in the north-
ern margin of the NCC during the late Triassic following
the final collision of the Mongolian oceanic arc terraneswith the NCC. We thus propose that the parental mag-
mas of the Fanshan intrusion were generated by
decompression melting of enriched mantle peridotite in
a post-collision, extensional tectonic setting. The en-
riched lithospheric mantle source was probably meta-
somatized by infiltration of subduction zone fluids
Fig. 12. Variation of eNd(t) vs (87Sr/86Sr)t for the Fanshan intrusion. Plotted for comparison are Sr–Nd isotopic compositions ofPrecambrian basement rocks from the northern North China Craton (calculated at 218 Ma) and those of Late Paleozoic–earlyMesozoic intrusions (calculated at 218 Ma) in the northern margin of the NCC from Jiang (2005) and Zhang et al. (2009, 2012). Thecomposition of the lower continental crust is after Jahn et al. (1999). The composition of Paleozoic lithospheric mantle in the NCC(represented by kimberlite-hosted xenoliths; Zheng & Lu, 1997; Xu et al., 2004), the Roman Magmatic Province (Prelevic et al., 2008,2013; Boari et al., 2009) and Yaojiazhuang intrusion (Chen et al., 2013) are shown for comparison. The compositions of igneousrocks of the Devonian Kola Alkaline Carbonatite Province in NW Russia and eastern Finland are compiled from Downes et al. (2005)and Lee et al. (2006).
Fig. 14. Result of MELTS modelling (Ghiorso & Sack, 1995)using a starting magma composition based on a kamafugitefrom Western Qinling, NCC (Guo et al., 2014) assumingfO2¼FMQ, a starting temperature of 1250�C and final tempera-ture of 730�C: (a) H2O¼3 wt %, pressure of 10 kbar (�30 km);(b) H2O¼0�5 wt %, pressure of 2 kbar. The sequence of appear-ance of mineral phases in each of the MELTS runs is displayedby the grey bars. In (b) K2O content of the starting magma com-position has been increased to 3 wt %.
Fig. 13. (a) Theoretical two-component mixing curves for d18Ovs initial 87Sr/86Sr. Ratios shown on each curve denote the pro-portion of Sr in the mantle or mantle-derived end-member tothe proportion of Sr in the crustal contaminant or slab-derivedfluid (after James, 1981). (b) Nb/La vs La/Yb variation inthe Fanshan samples. The black lines separating fields of as-thenospheric, lithospheric and mixed lithospheric–astheno-spheric mantle are from Abdel-Rahman (2002). Symbols are asin Fig. 12.
lithospheric mantle source that was metasomatized by
subduction-related melts or fluids prior to magmageneration.
Experimental studies suggest that small degrees of
melting (<10%) of mantle peridotite can yield alkali-rich
primary magmas (Hirschmann et al., 1998) with LREE-
enriched REE patterns (Henderson, 1984). During partial
melting, apatite and hydrous phases are rapidly con-sumed over a small temperature interval close to the
solidus (Wilson, 1989; Hammouda et al., 2010).
Experimental melting studies of phlogopite-bearing
harzburgite or lherzolite indicate that under F-rich con-
ditions and elevated pressure (>12 kbar for harzburgite
or >18 kbar for lherzolite) melt compositions change
from silica-saturated to silica-undersaturated (Melzer &
Foley, 2000, and references therein). A decrease in the
degree of silica-saturation of potassic melts has alsobeen observed experimentally under F-poor, H2O-rich
conditions at elevated pressure (Foley, 1992, 1993). The
degree of silica-saturation of primary potassic melts,
however, is also controlled by the fluid composition
during partial melting. A predominance of CO2 over
H2O during magma generation will suppress the stabil-ity field of olivine, favouring the formation of silica-
undersaturated melts (Wendlandt & Eggler, 1980).
Therefore, low-degree melting of such a mantle source
under high-pressure conditions could well explain the
geochemical characteristics of the Fanshan parental
magmas, analagous to models proposed for the origin
of Group II ultrapotassic rocks in Central Italy and
glimmerite,
Fig. 15. Variation of (a) Th/Yb vs Nb/Yb, (b) Ba/Nb vs La/Nb, and (c) [Ta/La]N vs [Hf/Sm]N for the Fanshan intrusion. The trends ofsubduction- and carbonatite-related metasomatism are from LaFleche et al. (1998) and references therein. (d) Variation of Tb/Yband La/Sm normalized to primitive mantle values (Sun & McDonough, 1989). The boundary between products of spinel- and gar-net-dominated melting is from Wang et al. (2002) and references therein; OIB from Sun & McDonough (1989).
Western Qinling in the NCC (e.g. Conticelli et al., 2013;
Guo et al., 2014).
Origin of the apatite oresThe Fanshan intrusion includes a distinctive type of
magmatic apatite-rich rocks comprising layers of mono-
mineralic apatite. Cumulates with very high apatitemodes approaching monomineralic facies have been
reported from other alkaline intrusions such as the
Kihibina Complex in the Kola Peninsula (e.g. Notholt,
1979; Veksler et al., 1998; Zaitsev et al., 2014) but the
Fanshan apatite-rich rocks are to our knowledge unpar-
alleled in terms of size and the unique cumulus assem-blage characterized by the coexistence of glimmerite
with apatite 6 magnetite rocks. Like silica-poor phoscor-
ites (e.g. Zaitsev et al., 2014), the almost silica-free char-
acteristics of the Fanshan ores (monomineralic apatite
rocks and nelsonite) probably preclude the possibility
that these rocks crystallized from immiscible P-rich con-jugates (e.g. Veksler et al., 1998, 2006, 2008).
Fractional crystallization and magmareplenishment?It is notable that the monomineralic apatite rocks andnelsonites contain biotite and are often associated with
glimmerite. According to our studies, the Fanshan rocks
exhibit a large range of major element compositional
variation (Table 4), yet all have similar Sr–Nd isotopic
and trace element characteristics (Figs 10–12), suggest-
ing that the parental magmas have experienced varyingdegrees of fractional crystallization and crystal accumu-
lation after emplacement. The order of crystallization
can be deduced from the field relations, petrographic
observations and MELTS modelling. Except for melan-
ite, the following crystallization sequence is proposed:
Ca-rich augite is the earliest phase on the liquidus,
followed by biotite, apatite, magnetite and finallyK-feldspar. The absence of apatite in the early formed
clinopyroxenite suggests undersaturation of apatite
during the early stage. Further fractionation of silicate
minerals after magma emplacement may have driven
the elevated concentrations of phosphorus until the
monomineralic apatite rocks and nelsonites started toform (Tollari et al., 2006). However, MELTS modelling
suggests that apatite crystallized simultaneously with
several other silicate phases including Ca-rich augite
and biotite. Therefore, simple fractional crystallization is
incapable of explaining the formation of monomineralic
rocks.
The combination of large variations in modal min-eralogy and evidence of fractionation suggest that even
Unit 2 does not exhibit closed-system behaviour. The
first and most important line of evidence is the presence
of several Mg-number reversals in Ca-rich augite com-
position from the base of Unit 2 upwards. At these
stratigraphic levels, the concentration of Ti in Ca-richaugite shifts to higher values (Fig. 7b), consistent with
abrupt increases of these elements in the crystallizing
magma. Thus, these intervals probably record the re-
charge and mixing of more primitive magmas with the
fractionating magma in the chamber. Such open-
system behaviour has previously been proposed for
other phoscorite intrusions (e.g. alkaline plutons of theKola Peninsula, Russia; Verhulst et al., 2000).
In the high-level Fanshan magma chamber the dens-
ity of the evolved magma crystallizing Ca-rich augite,
biotite, apatite and magnetite is likely to be lower than
that of the primitive magma replenished from below.
Thus, when a new pulse of primitive magma arrived it
would have formed a layer at the base of the magmachamber (Campbell & Turner, 1989; Snyder & Tait,
1995). The widespread planar foliation and lineation
shown by the main minerals in the rocks of Unit 2 may
indicate that magmatic currents and a laminar flow re-
gime may have resulted from the replenishment of
magma (e.g. Wager & Brown, 1967; Irvine, 1987;Conrad & Naslund, 1989). Thus the thick monomineralic
apatite rocks could be produced by apatite crystalliza-
tion from fractionated P-rich magmas frequently
recharged with more primitive magma from a deeper
crustal magma chamber. If this is the case, the Fanshan
magma is required to be low in viscosity to facilitatemagma flow. Addition of H2O is known to reduce melt
We are grateful to B. Ronald Frost, Dejan Prelevic, and
Editor Marjorie Wilson for their thoughtful and con-structive comments. Tonny Bernt Thomsen at the
Geological Survey of Denmark and Greenland, Paul
Carpenter of Washington University in St. Louis and
Oona Appelt at the Helmholtz Centre GFZ Potsdam are
thanked for their assistance with laser ablation and elec-
tron microprobe analysis; Ziliang Jin and Liu Han arethanked for their help in the field. Zhenhui Bian is
acknowledged for provision of logistical support in the
Fanshan Phosphorus Mine.
FUNDING
Parts of this work were supported by 973 Program(2012CB416806), the National Natural Science Founda-
tion of China (Nos 40925006 and 40821061), the ‘Funda-
mental Research Funds for the Central Universities’, the
111 Project (B07011), and PCSIRT, and DFG grant VE 619/
2-1. I.V.V. also acknowledges support by the Russian Sci-
ence Foundation grant No. 14-17-00200.
SUPPLEMENTARY DATA
Supplementary data for this paper are available at
Journal of Petrology online.
REFERENCES
Abdel-Rahman, A. M. (2002). Mesozoic volcanism in the MiddleEast: geochemical, isotopic and petrogenetic evolution ofextension-related alkali basalts from central Lebanon.Geological Magazine 139, 621–640.
Baker, D. R. & Vaillancourt, J. (1995). The low viscosities ofFþH2O-bearing granitic melts and implications for melt ex-traction and transport. Earth and Planetary Science Letters132, 199–211.
Baker, M. B. & Wyllie, P. J. (1992). High-pressure apatite solubil-ity in carbonate-rich liquids: implications for mantlemetasomatism. Geochimica et Cosmochimica Acta 56,3409–3422.
Bedard, J. H. (2014). Parameterizations of calcic clinopyroxene—Melt trace element partition coefficients. Geochemistry, Geo-physics, Geosystems 15, 303–336, doi:10.1002/2013GC005112.
Boari, E., Tommasini, S., Laurenzi, M. A. & Conticelli, S. (2009).Transition from ultrapotassic kamafugitic to sub-alkalinemagmas: Sr, Nd, and Pb isotope, trace element and40Ar/39Ar age data from the Middle Latin Valley volcanicfield, Roman Magmatic Province, Central Italy. Journal ofPetrology 50, 1327–1357.
Campbell, I. H. & Turner, J. S. (1989). Fountains in magmachambers. Journal of Petrology 30, 885–923.
Castillo, P. R. & Newhall, C. G. (2004). Geochemical constraintson possible subduction components in lavas of Mayon andTaal volcanoes, southern Luzon, Philippines. Journal ofPetrology 45, 1089–1108.
Cellai, D., Conticelli, S. & Menchetti, S. (1994). Crystal-chemistryof clinopyroxene from potassic and ultrapotassic rocks incentral Italy: Implications on their genesis. Contributions toMineralogy and Petrology 116, 301–315.
Chen, B., Niu, X., Wang, Z., Gao, L. & Wang, C. (2013).Geochronology, petrology, and geochemistry ofthe Yaojiazhuang ultramafic–syenitic complex from theNorth China Craton. Science China: Earth Sciences 56,1294–1307.
Cheng, C. & Sun, S. H. (2003). The Fanshan apatite–magnetitedeposit in the potassic ultramafic layered intrusions, NorthChina. Resource Geology 53, 163–174.
Condie, K. (1997). Sources of Proterozoic mafic dyke swarms:constraints from Th/Ta and La/Yb ratios. PrecambrianResearch 81, 3–14.
Conrad, M. E. & Naslund, H. R. (1989). Modally-graded rhythmiclayering in the Skaergaard intrusion. Journal of Petrology30, 251–269.
Conte, A. M., Dolfi, D., Gaeta, M., Misiti, V., Mollo, S. & Perinelli,C. (2009). Experimental constraints on evolution of leucite-basanite magma at 1 and 10–4 GPa: implications for parentalcompositions of Roman high-potassium magmas.European Journal of Mineralogy 21, 763–782.
Conticelli, S., D’Antonio, M., Pinarelli, L. & Civetta, L. (2002).Source contamination and mantle heterogeneity in thegenesis of Italian potassic and ultrapotassic volcanicrocks: Sr–Nd–Pb isotope data from the Roman provinceand southern Tuscany. Mineralogy and Petrology 74,189–222.
Conticelli, S., Carlson, R. W., Widom, E. & Serri, G. (2007).Chemical and isotopic composition (Os, Pb, Nd, and Sr) ofNeogene to Quaternary calc-alkalic, shoshonitic, and ultra-potassic mafic rocks from the Italian peninsula: inferenceson the nature of their mantle sources. In: Beccaluva, L.,Bianchini, G. & Wilson, M. (eds) Cenozoic Volcanism in theMediterranean Area. Geological Society of America, SpecialPapers 418, 171–202.
Conticelli, S., Guarnieri, L., Farinelli, A., Mattei, M., Avanzinelli,R., Bianchini, G., Boari, E., Tommasini, S., Tiepolo, M.,Prelevic, D. & Venturelli, G. (2009). Trace elements andSr–Nd–Pb isotopes of K-rich, shoshonitic, and calc-alkalinemagmatism of the Western Mediterranean Region:genesis of ultrapotassic to calc-alkaline magmatic associ-ations in a post-collisional geodynamic setting. Lithos 107,68–92.
Conticelli, S., Avanzinelli, R., Poli, G., Braschi, E. & Giordano, G.(2013). Shift from lamproite-like to leucititic rocks: Sr–Nd–Pbisotope data from the Monte Cimino volcanic complex vs.the Vico stratovolcano, Central Italy. Chemical Geology 353,246–266.
Deer, W. A., Howie, R. A. & Zussman, J. (1992). Introduction tothe Rock-forming Minerals. Prentice Hall, 712 pp.
Dill, H. G. (2010). The ‘chessboard’ classification scheme of min-eral deposits: Mineralogy and geology from aluminum tozirconium. Earth-Science Reviews 100, 1–420.
Dingwell, D. B. & Hess, K. U. (1998). Melt viscosities in thesystem Na–Fe–Si–O–F–Cl: Contrasting effects of F and Cl inalkaline melts. American Mineralogist 83, 1016–1021.
Downes, H., Balaganskaya, E., Beard, A., Liferovich, R. &Demaiffe, D. (2005). Petrogenetic processes in the ultra-mafic, alkaline and carbonatitic magmatism in the KolaAlkaline Province: A review. Lithos 85, 48–75.
Foley, S. F. (1992). Vein-plus-wall-rock melting mechanism inthe lithosphere and the origin of potassic alkaline magmas.Lithos 28, 435–453.
Foley, S. F. (1993). An experimental study of olivine lamproite:first results from the diamond stability field. Geochimica etCosmochimica Acta 57, 483–489.
Foley, S. F., Venturelli, G., Green, D. H. & Toscani, L. (1987). Theultrapotassic rocks: characteristics, classification and con-straints for petrogenetic models. Earth-Science Reviews 24,81–134.
Freda, C., Gaeta, M., Misiti, V., Mollo, S., Dolfi, D. & Scarlato, P.(2008). Magma–carbonate interaction: an experimentalstudy on ultrapotassic rocks from Alban Hills (Central Italy).Lithos 101, 397–415.
Gaeta, M., Freda, C., Christensen, J. N., Dallai, L., Marra, F.,Karner, D. B. & Scarlato, P. (2006). Time-dependent geo-chemistry of clinopyroxene from the Alban Hills (CentralItaly): clues to the source and evolution of ultrapotassicmagmas. Lithos 86, 330–346.
Ghiorso, M. S. & Sack, R. O. (1995). Chemical mass transfer inmagmatic processes IV. A revised and internally consistentthermodynamic model for the interpolation and extrapola-tion of liquid–solid equilibria in magmatic systems atelevated temperatures and pressures. Contributions toMineralogy and Petrology 119, 197–212.
Giordano, D., Russell, J. K. & Dingwell, D. B. (2008). Viscosity ofmagmatic liquids: A model. Earth and Planetary ScienceLetters 271, 123–134.
Guo, P., Niu, Y. & Yu, X. (2014). A synthesis and new perspec-tive on the petrogenesis of kamafugites from West Qinling,China, in a global context. Journal of Asian Earth Sciences79, 86–96.
Hammouda, T., Chantel, J. & Devidal, J. L. (2010). Apatite solu-bility in carbonatitic liquids and trace element partitioningbetween apatite and carbonatite at high pressure.Geochimica et Cosmochimica Acta 74, 7220–7235.
Hellstrom, J., Paton, C., Woodhead, J. & Hergt, J. (2008). Iolite:Software for spatially resolved LA- (quad and MC) ICPMSanalysis. In: Sylvester, P. (ed.) Laser Ablation ICP-MS in theEarth Sciences: Current Practices and Outstanding Issues.Mineralogical Association of Canada, Short Course Series40, 343–348.
Henderson, P. (ed.) (1984). Rare Earth Element Geochemistry.Elsevier, 510 pp.
Hirschmann, M. M., Ghiorso, M. S., Asimow, P. D., Wasylenki,L. E. & Stolper, E. M. (1998). Calculation of peridotite partialmelting from thermodynamic models of minerals and melts.I. Review of methods and comparison with experiments.Journal of Petrology 39, 1091–1115.
Iacono Marziano, G., Gaillard, F. & Pichavant, M. (2007).Limestone assimilation and the origin of CO2 emissions atthe Alban Hills (Central Italy): Constraints from experimentalpetrology. Journal of Volcanology and GeothermalResearch 166, 91–105.
Iacono-Marziano, G., Gaillard, F. & Pichavant, M. (2008).Limestone assimilation by basaltic magmas: an experimen-tal re-assessment and application to Italian volcanoes.Contributions to Mineralogy and Petrology 155, 719–738.
Iacono-Marziano, G., Gaillard, F., Scaillet, B., Pichavant, M. &Chiodini, G. (2009). Role of non-mantle CO2 in the dynamics
of volcano degassing: The Mount Vesuvius example.Geology 37, 319–322.
Irvine, T. N. (1987). Layering and related structures in the DukeIsland and Skaergaard intrusions: Similarities, differences,origins. In: Parsons, I. (ed.) Origins of Igneous Layering.D. Reidel, pp. 185–245.
Jacobsen, S. B. & Wasserburg, G. J. (1980). Sm–Nd isotopicevolution of chondrites. Earth and Planetary Science Letters50, 139–155.
Jahn, B. M., Wu, F. Y., Lo, C. H. & Tsai, C. H. (1999). Crust–mantle interaction induced by deep subduction of the con-tinental crust: Geochemical and Sr–Nd isotopic evidencefrom post-collisional mafic–ultramafic intrusions of thenorthern Dabie complex, central China. Chemical Geology157, 119–146.
James, D. E. (1981). The combined use of oxygen andradiogenic isotopes as indicators of crustalcontamination. Annual Review of Earth and PlanetarySciences 9, 311–344.
Jiang, N. (2005). Petrology and geochemistry of theShuiquangou syenitic complex, northern margin of theNorth China craton. Journal of the Geological Society,London 162, 203–215.
Jiang, N., Chu, X. L., Mizuta, T., Ishiyama, D. & Mi, J. G. (2004).A magnetite–apatite deposit in the Fanshan alkalineultramafic complex, Northern China. Economic Geology 99,397–408.
LaFleche, M. R., Camire, G. & Jenner, G. A. (1998).Geochemistry of post-Acadian, Carboniferous continentalintraplate basalts from the Maritimes Basin, MagdalenIslands, Quebec, Canada. Chemical Geology 148, 115–136.
Lee, M. J., Lee, J. I., Hur, S. D., Kim, Y., Moutte, J. &Balaganskaya, E. (2006). Sr–Nd–Pb isotopic compositions ofthe Kovdor phoscorite–carbonatite complex, KolaPeninsula, NW Russia. Lithos 91, 250–261.
Le Roex, A. P. & Lanyon, R. (1998). Isotope and trace elementgeochemistry of Cretaceous Damaraland lamprophyres andcarbonatites, Northwestern Namibia: evidence forplume–lithosphere interactions. Journal of Petrology 39,1117–1146.
Liu, D. Y., Nutman, A. P., Compston, W., Wu, J. S. & Shen, Q. H.(1992). Remnants of �3800 Ma crust in the Chinese part ofthe Sino-Korean Craton. Geology 20, 339–342.
Lugmair, G. W. & Harti, K. (1978). Lunar initial 143Nd/144Nd: dif-ferential evolution of the lunar crust and mantle. Earth andPlanetary Science Letters 39, 349–357.
McBirney, A. R. & Hunter, R. H. (1995). The cumulate paradigmreconsidered. Journal of Geology 103, 114–122.
Melluso, L., Lustrino, M., Ruberti, E., Brotzu, P., Gomes, C. B.,Morbidelli, L., Morra, V., Svisero, D. P. & d’Amelio, F. (2008).Major- and trace-element composition of olivine, perovskite,clinopyroxene, Cr–Fe–Ti oxides, phlogopites and host kama-fugites and kimberlites, Alto Paranaıba, Brazil. CanadianMineralogist 46, 19–40.
Melzer, S. & Foley, S. F. (2000). Phase relations and fraction-ation sequences in potassic magma series modelled in thesystem CaMgSi2O6–KAlSiO4–Mg2SiO4–SiO2–F2O–1 at 1 barto 18 kbar. Contributions to Mineralogy and Petrology 138,186–197.
Miller, C., Schuster, R., Klotzli, U., Frank, W. & Purtscheller, F.(1999). Post-collisional potassic and ultrapotassic magma-tism in SW Tibet: geochemical and Sr–Nd–Pb–O isotopicconstraints for mantle source characteristics and petrogen-esis. Journal of Petrology 40, 1399–1424.
Mollo, S. & Vona, A. (2014). The geochemical evolution of clino-pyroxene in the Roman Province: A window on decarbona-tion from wall-rocks to magma. Lithos 192, 1–7.
Mu, B., Jiang, P., Zeng Y. & Yan, G. (1988). The Fanshan igneouscomplex and apatite–magnetite deposit in Hebei Province,China. Peking University Press (in Chinese).
Niu, X., Chen, B., Liu, A., Suzuki, K. & Ma, X. (2012). Petrologicaland Sr–Nd–Os isotopic constraints on the origin of theFanshan ultrapotassic complex from the North ChinaCraton. Lithos 149, 146–158.
Norman, M. D., Pearson, N. J., Sharma, A. & Griffin, W. L.(1996). Quantitative analysis of trace elements in geologicalmaterials by laser ablation ICPMS: instrumental operatingconditions and calibration values of NIST glasses.Geostandards Newsletter 20, 247–261.
Notholt, A. J. G. (1979). The economic geology and develop-ment of igneous phosphate deposits in Europe and theUSSR. Economic Geology 74, 339–350.
Paton, C., Hellstrom, J. C., Paul, P., Woodhead, J. D. & Hergt, J.M. (2011). Iolite: Freeware for the visualisation and process-ing of mass spectrometric data. Journal of AnalyticalAtomic Spectrometry 26, 2508–2518.
Peccerillo, A. (2005). Plio-Quaternary Volcanism in Italy:Petrology, Geochemistry, Geodynamics. Springer, 365 pp.
Perini, G., Ponticelli, S., Francalanci, L. & Davidson, J. P. (2004).Evolution and genesis of magmas from Vico volcano,Central Italy: multiple differentiation pathways and variableparental magmas. Journal of Petrology 45, 139–182.
Prelevic, D., Foley, S. F., Romer, R. & Conticelli, S. (2008).Mediterranean Tertiary lamproites derived from multiplesource components in postcollisional geodynamics.Geochimica et Cosmochimica Acta 72, 2125–2156.
Prelevic, D., Jacob, D. E. & Foley, S. F. (2013). Recycling plus: anew recipe for the formation of Alpine–Himalayan orogenicmantle lithosphere. Earth and Planetary Science Letters 362,187–197.
Prelevic, D., Brugmann, G., Barth, M., Bozovic, M., Cvetkovic,V., Foley, S. F. & Maksimovic, Z. (2014). Os-isotope con-straints on the dynamics of orogenic mantle: The case of theCentral Balkans. Gondwana Research http://dx.doi.org/10.1016/j.gr.2014.02.001.
Putirka, K. D. (2008). Thermometers and barometers for vol-canic systems. In: Putirka, K. D. & Tepley, F. J., III (eds)Minerals, Inclusions and Volcanic Processes. MineralogicalSociety of America and Geochemical Society, Reviews inMineralogy and Geochemistry 69, 61–120.
Ren, R., Mu, B. L., Han, B. F., Zhang, L., Chen, J. F., Xu, Z. &Song, B. (2009). Zircon SHRIMP U–Pb dating of the Fanshanpotassic alkaline ultramafite–syenite complex in Hebei prov-ince, China. Acta Petrologica Sinica 25, 588–594 (in Chinesewith English abstract).
Rudnick, R. L. & Gao, S. (2003). Composition of the continentalcrust. In: Rudnick, R. L. (ed.) Treatise on Geochemistry 3.Elsevier–Pergamon, pp. 1–64.
Rudnick, R. L., McDonough, W. F. & Chappell, B. W. (1993).Carbonatite metasomatism in the northern Tanzanian man-tle: petrographic and geochemical characteristics. Earth andPlanetary Science Letters 114, 463–475.
Shao, J., Zhang, Y., Zhang, L., Mu, B., Wang, P. & Guo, F. (2003).Early Mesozoic dike swarms of carbonatites and lampro-phyres in Datong area. Acta Petrologica Sinica 19, 93–104(in Chinese with English abstract).
Snyder, D. & Tait, S. (1995). Replenishment of magma cham-bers: comparison of fluid-mechanic experiments with fieldrelations. Contributions to Mineralogy and Petrology 122,230–240.
Steiger, R. H. & Jager, E. (1977). Subcommission on geochron-ology: convention on the use of decay constants in geo-chronology and cosmochronology. Earth and PlanetaryScience Letters 36, 359–362.
Sun, S. S. & McDonough, W. F. (1989). Chemical and isotopicsystematics of oceanic basalts; implications for mantle com-position and processes. In: Saunders, A. D. & Norry, M. J.(eds) Magmatism in the Ocean Basins. Geological Society,London, Special Publications 42, 313–345.
Tollari, N., Toplis, M. J. & Barnes, S. J. (2006). Predicting phos-phate saturation in silicate magmas: an experimental studyof the effects of melt composition and temperature.Geochimica et Cosmochimica Acta 70, 1518–1536.
Veksler, I. V., Nielsen, T. F. D. & Sokolov, S. V. (1998).Mineralogy of crystallized melt inclusions fromGardiner and Kovdor ultramafic alkaline complexes: impli-cations for carbonatite genesis. Journal of Petrology 39,2015–2031.
Veksler, I. V., Dorfman, A. M., Danyushevsky, L. V., Jakobsen, J.K. & Dingwell, D. B. (2006). Immiscible silicate liquidpartition coefficients: implications for crystal–melt elementpartitioning and basalt petrogenesis. Contributions toMineralogy and Petrology 152, 685–702.
Veksler, I. V., Dorfman, A. M., Borisov, A., Wirth, R. & Dingwell,D. B. (2008). Liquid unmixing kinetics and the extent of im-miscibility in the system K2O–CaO–FeO–Al2O3–SiO2.Chemical Geology 256, 119–130.
Vennemann, T. W. & Smith, H. S. (1990). The rate and tempera-ture of reaction of ClF3 with silicate minerals, and theirrelevance to oxygen isotope analysis. Chemical Geology 86,83–88.
Verhulst, A., Balaganskaya, E., Kirnarsky, Y. & Demaiffe, D.(2000). Petrological and geochemical (trace elements andSr–Nd isotopes) characteristics of the Paleozoic Kovdorultramafic, alkaline and carbonatite intrusion (KolaPeninsula, NW Russia). Lithos 51, 1–25.
Wager, L. R. & Brown, G. M. (1967). Layered Igneous Rocks.Freeman.
Wang, K., Plank, T., Walker, J. D. & Smith, E. I. (2002). A mantlemelting profile across the basin and range, SW USA.Journal of Geophysical Research—Solid Earth 107,doi:10.1029/2001JB0002092.
Wendlandt, R. F. & Eggler, D. (1980). The origins of potassicmagmas: 1. Melting relations in the system KAlSiO4–Mg2SiO4–SiO2 and KAlSiO4–MgO–SiO2–CO2 to 30 kilobars.American Journal of Science 280, 385–420.
Wilson, M. (1989). Igneous Petrogenesis. Unwin Hyman.Wones, D. R. & Eugster, H. P. (1965). Stability of biotite: experi-
ment, theory and application. American Mineralogist 50,1228–1272.
Xiao, W., Windley, B. F., Hao, J. & Zhai, M. G. (2003). Accretionleading to collision and the Permian Solonker suture, InnerMongolia, China: termination of the central Asian orogenicbelt. Tectonics 22(6), 1069, http://dx.doi.org/10.1029/2002TC001484.
Xu, Y. G. (2001). Thermo-tectonic destruction of the Archeanlithospheric keel beneath the Sino-Korean Craton in China:evidence, timing and mechanism. Physics and Chemistry ofthe Earth (A) 26, 747–757.
Xu, Y. G., Ma, J. L., Huang, X. L., Iizuka, Y., Chung, S. L., Wang,Y. B. & Wu, X. Y. (2004). Early Cretaceous gabbroic complexfrom Yinan, Shandong Province: petrogenesis and mantledomains beneath the North China Craton. InternationalJournal of Earth Sciences 93, 1025–1041.
Zaitsev, A. N., Williams, C. T., Jeffries, T. E., Strekopytov, S.,Moutte, J., Ivashchenkova, O. V., Spratt, J., Petrov, S. V.,Wall, F., Seltmann, R. & Borozdin A. P. (2014). Rare earthelements in phoscorites and carbonatites of the DevonianKola Alkaline Province, Russia: Examples from Kovdor,Khibina, Vuoriyarvi and Turiy Mys complexes. Ore GeologyReviews 61, 204–225.
Zhang, S. H., Zhao, Y., Song, B. & Liu, D. Y. (2007).Petrogenesis of the middle Devonian Gushan dioritepluton on the northern margin of the North China block andits tectonic implications. Geological Magazine 144, 553–568.
Zhang, S. H., Zhao, Y., Song, B., Hu, J. M., Liu, S. W., Yang, Y.H., Chen, F. K., Liu, X. M. & Liu, J. (2009). ContrastingLate Carboniferous and Late Permian–Middle Triassicintrusive suites from the northern margin of the NorthChina Craton: geochronology, petrogenesis and tectonicimplications. Geological Society of America Bulletin 121,181–200.
Zhang, S. H., Zhao, Y., Ye, H., Hou, K. J. & Li, C. F. (2012). EarlyMesozoic alkaline complexes in the northern North China
Craton: Implications for cratonic lithospheric destruction.Lithos 155, 1–15.
Zhao, G. C., Cawood, P. A. & Wilde, S. A. (2001). High-pressuregranulites (retrograded eclogites) from the Hengshan com-plex, North China Craton: petrology and tectonic implica-tions. Journal of Petrology 42, 1141–1170.
Zheng, J. P. & Lu, X. F. (1997). A study on the Sr–Nd isotopiccompositions of mantle xenoliths bearing in kimberlites ineastern China. Earth Science 18(Supplement), 15–17 (inChinese with English abstract).
Zimova, M. & Webb, S. L. (2006). The combined effects of chlor-ine and fluorine on the viscosity of aluminosilicate melts.Geochimica et Cosmochimica Acta 71, 1553–1562.