uncorrected proof eschweizerbart_xxx Apatites from the Kaiserstuhl Volcanic Complex, Germany: new constraints on the relationship between carbonatite and associated silicate rocks LIAN-XUN WANG 1,2, * ,MICHAEL A.W. MARKS 1 ,THOMAS WENZEL 1 ,ANETTE VON DER HANDT 3 ,JO ¨ RG KELLER 3 , HOLGER TEIBER 1 and GREGOR MARKL 1 1 Mathematisch-Naturwissenschaftliche Fakulta ¨t, FB Geowissenschaften, Universita ¨t Tu ¨bingen, 72074 Tu ¨bingen, Germany *Corresponding author, e-mail: [email protected]2 Faculty of Earth Sciences, China University of Geosciences, Wuhan 430074, China 3 Albert-Ludwigs-Universita ¨t Freiburg, Institut fu ¨r Geowissenschaften - Mineralogie-Geochemie, 79104 Freiburg, Germany Abstract: Apatites from carbonatites, related alkaline silicate rocks, a carbonate-bearing melilititic dyke rock (bergalite), and a diatreme breccia (containing both carbonate and silicate fragments) of the Miocene Kaiserstuhl Volcanic Complex, SW Germany, are used to reconstruct the petrogenetic relationship among these rocks. Apatites from carbonatites reach higher Sr and Nb contents but are generally lower in Fe, Mn, Th, U, Si, S, Cl and Br compared to apatites from associated silicate rocks, whilst Na, REE and F contents are overlapping. Apatites from bergalite show a systematic and discontinuous core-rim zonation, with the core being compositionally similar to apatites from silicate rocks and the rim corresponding to carbonatitic apatites. These observations imply that the bergalite apatites nucleated in a silicate melt and continued to crystallize from an evolving CO 2 -enriched melt probably with carbonatitic affinity. Apatites from a diatreme breccia comprise three populations: (1) similar to the apatites from silicate rocks, (2) similar to the carbonatitic apatites, and (3) resembling apatite population (1) partially replaced by apatite (2). We infer that apatite (1) was derived from silicate-rock fragments and apatite (2) crystallized from a later intruding carbonatitic melt, which metasomatized the silicate-rock fragments and caused the replacement textures as observed in apatite population (3). We conclude that apatites from the Kaiserstuhl complex preserve important information on the petrogenetic relationship between carbonatitic and silicate melts. The carbonatitic melts at the Kaiserstuhl complex are probably the products of protracted fractionation of a CO 2 -rich nephelinitic melt. Key-words: apatite, carbonatite, bergalite, Kaiserstuhl, replacement, zoning texture. 1 1. Introduction 2 From the approximately 500 carbonatite occurrences 3 worldwide, more than 75% are associated with alkaline 4 silicate rocks, ultramafic lamprophyres or kimberlites (e.g. 5 Woolley & Kjarsgaard, 2008) and, in many cases, carbo- 6 natites postdate the silicate rocks (e.g., Keller et al., 1990; 7 Bailey, 1993; Bell et al., 1999; Woolley, 2003, Woolley & 8 Kjarsgaard, 2008; Halama et al., 2005). Despite nearly a 9 century of research on carbonatites, their origin and rela- 10 tionships to associated silicate rocks is not completely 11 understood yet (Bell et al., 1999; Gittins & Harmer, 12 2003). Experimental work showed that carbonatites can 13 be generated by (1) liquid immiscibility of carbonatite and 14 silicate melts (e.g., Koster van Groos & Wyllie, 1966; 15 Freestone & Hamilton, 1980; Brooker & Kjarsgaard, 16 2011); (2) partial melting of carbonate-bearing mantle 17 peridotite (e.g., Wallace & Green, 1988; Dalton & 18 Presnall, 1998; Ghosh et al., 2009); and (3) fractional 19 crystallization of a carbonate-rich alkaline silicate 20 magma (Watkinson & Wyllie, 1971; Lee & Wyllie, 21 1994). In some localities, unusual transitional rocks (cf. 22 carbonate-rich/bearing silicate rocks), such as bergalites, 23 turjaites and okaites, are found temporally and spatially 24 close to the carbonatite and associated silicate rocks (Eby, 25 1975; Keller et al., 1990; Keller, 1991, 1997; Bell & 26 Simonetti, 1996). These rocks may provide important 27 clues for the petrogenetic link between carbonatite and 28 associated silicate rocks (Keller et al., 1990; Keller, 29 1991, 1997; Chakhmouradian & Zaitsev, 2004; Moore 30 et al., 2009). 31 Apatite is a good indicator for magma evolution (Seifert 32 et al., 2000; Piccoli & Candela, 2002; Zhang et al., 2012) 33 and occurs in a wide range of magmatic systems such as 34 mafic rocks (e.g., Brown & Peckett, 1977; Binder & Troll, 35 1989; Tribuzio et al., 1999), granitic rocks (e.g., Watson, 36 1980; Sha & Chappell, 1999; Marks et al., 2012) and 37 syenitic rocks (e.g., Liferovich & Mitchell, 2006; 0935-1221/14/0026-2377 $ 8.10 DOI: 10.1127/0935-1221/2014/0026-2377 # 2013 E. Schweizerbart’sche Verlagsbuchhandlung, D-70176 Stuttgart Eur. J. Mineral. Fast Track Article Fast Track DOI: 10.1127/0935-1221/2014/0026-2377
Transcript
unco
rrecte
d proo
f
eschweizerbart_xxx
Apatites from the Kaiserstuhl Volcanic Complex, Germany: new constraints on
the relationship between carbonatite and associated silicate rocks
LIAN-XUN WANG1,2,*, MICHAEL A.W. MARKS1, THOMAS WENZEL1, ANETTE VON DER HANDT3, JORG KELLER3,
Abstract: Apatites from carbonatites, related alkaline silicate rocks, a carbonate-bearing melilititic dyke rock (bergalite), and adiatreme breccia (containing both carbonate and silicate fragments) of the Miocene Kaiserstuhl Volcanic Complex, SW Germany, areused to reconstruct the petrogenetic relationship among these rocks. Apatites from carbonatites reach higher Sr and Nb contents butare generally lower in Fe, Mn, Th, U, Si, S, Cl and Br compared to apatites from associated silicate rocks, whilst Na, REE and Fcontents are overlapping. Apatites from bergalite show a systematic and discontinuous core-rim zonation, with the core beingcompositionally similar to apatites from silicate rocks and the rim corresponding to carbonatitic apatites. These observations implythat the bergalite apatites nucleated in a silicate melt and continued to crystallize from an evolving CO2-enriched melt probably withcarbonatitic affinity. Apatites from a diatreme breccia comprise three populations: (1) similar to the apatites from silicate rocks, (2)similar to the carbonatitic apatites, and (3) resembling apatite population (1) partially replaced by apatite (2). We infer that apatite (1)was derived from silicate-rock fragments and apatite (2) crystallized from a later intruding carbonatitic melt, which metasomatizedthe silicate-rock fragments and caused the replacement textures as observed in apatite population (3). We conclude that apatites fromthe Kaiserstuhl complex preserve important information on the petrogenetic relationship between carbonatitic and silicate melts. Thecarbonatitic melts at the Kaiserstuhl complex are probably the products of protracted fractionation of a CO2-rich nephelinitic melt.
2 From the approximately 500 carbonatite occurrences3 worldwide, more than 75% are associated with alkaline4 silicate rocks, ultramafic lamprophyres or kimberlites (e.g.5 Woolley & Kjarsgaard, 2008) and, in many cases, carbo-6 natites postdate the silicate rocks (e.g., Keller et al., 1990;7 Bailey, 1993; Bell et al., 1999; Woolley, 2003, Woolley &8 Kjarsgaard, 2008; Halama et al., 2005). Despite nearly a9 century of research on carbonatites, their origin and rela-
10 tionships to associated silicate rocks is not completely11 understood yet (Bell et al., 1999; Gittins & Harmer,12 2003). Experimental work showed that carbonatites can13 be generated by (1) liquid immiscibility of carbonatite and14 silicate melts (e.g., Koster van Groos & Wyllie, 1966;15 Freestone & Hamilton, 1980; Brooker & Kjarsgaard,16 2011); (2) partial melting of carbonate-bearing mantle17 peridotite (e.g., Wallace & Green, 1988; Dalton &18 Presnall, 1998; Ghosh et al., 2009); and (3) fractional
19crystallization of a carbonate-rich alkaline silicate20magma (Watkinson & Wyllie, 1971; Lee & Wyllie,211994). In some localities, unusual transitional rocks (cf.22carbonate-rich/bearing silicate rocks), such as bergalites,23turjaites and okaites, are found temporally and spatially24close to the carbonatite and associated silicate rocks (Eby,251975; Keller et al., 1990; Keller, 1991, 1997; Bell &26Simonetti, 1996). These rocks may provide important27clues for the petrogenetic link between carbonatite and28associated silicate rocks (Keller et al., 1990; Keller,291991, 1997; Chakhmouradian & Zaitsev, 2004; Moore30et al., 2009).31Apatite is a good indicator for magma evolution (Seifert32et al., 2000; Piccoli & Candela, 2002; Zhang et al., 2012)33and occurs in a wide range of magmatic systems such as34mafic rocks (e.g., Brown & Peckett, 1977; Binder & Troll,351989; Tribuzio et al., 1999), granitic rocks (e.g., Watson,361980; Sha & Chappell, 1999; Marks et al., 2012) and37syenitic rocks (e.g., Liferovich & Mitchell, 2006;
1 Rønsbo, 2008). In particular, carbonatite (e.g., Buhn et al.,2 2001; Brassinnes et al., 2005; Chen & Simonetti, 2013)3 and related rocks (Eby, 1975; Keller et al., 1990; Bell &4 Simonetti, 1996; Moore et al., 2009) contain appreciable5 amounts of apatite. Experimental studies imply that many6 trace elements (e.g. Sr, rare earth elements [REE], Th, U)7 exhibit distinct partitioning behavior in carbonatitic and8 silicate melt systems (Klemme & Dalpe, 2003; Prowatke &9 Klemme, 2006; Hammouda et al., 2010). Thus, chemical
10 differences between apatites from carbonatites and from11 associated silicate rocks are expected, and detailed inves-12 tigations of apatite from the transitional dyke rocks might13 help to decipher the petrogenetic relationships between the14 two rock types.15 The Kaiserstuhl Volcanic Complex (KVC) comprises a16 series of alkaline silicate rocks, associated with carbona-17 tites as well as transitional dyke rocks (bergalites) and18 polygenic diatreme breccias (mixtures of carbonatite and19 silicate-rock fragments, e.g., Keller, 1984; Schleicher20 et al., 1990; Wimmenauer, 2003). In these rocks, apatite21 is a common accessory mineral. Hence, in the present work22 we studied the apatites from various rocks of KVC with the23 aim to (1) constrain the chemical variations of these apa-24 tites; (2) shed more light on the petrogenetic relationship25 between carbonatite and associated silicate rocks; and (3)26 investigate the abundance and variability of volatile ele-27 ments (F, Cl, Br, S) in apatites from alkaline silicate rocks28 and carbonatite.
29 2. Geological setting and sample description
30 The KVC is located in the southern part of Upper Rhine31 Graben (Fig. 1a) and is part of extensive Cenozoic volcan-32 ism in Central Europe (Keller et al., 1990; Wilson &33 Downes, 1991, 2006; Riley et al., 1999; Wimmenauer,
342003). Ages of the Kaiserstuhl rocks range between 1835and 13 Ma (Lippolt et al., 1963; Baranyi et al., 1976;36Kraml et al., 1995, 2006; Keller et al., 2002). The KVC37is a sequence of early intrusion and extrusion of alkaline38silicate rocks followed by late formation of carbonatites39(Keller et al., 1990; Wimmenauer, 2003; Keller, 2008).40The silicate rocks are dominated by tephrites, essexites and41phonolites (Fig.1b). Sub-dominant rock types include oli-42vine-melilitites, olivine-nephelinites, limburgites (olivine þ43augite þ glassy groundmass), hauynophyres, syenites, shon-44kinite porphyries, and mondhaldeites (Wimmenauer, 2003;45Keller, 2008). Shonkinite porphyry and mondhaldeite are46fractionated dyke rocks from the tephritic magma.47Carbonatites consist of sovite intrusions (Badberg and48Orberg), alvikitic dykes and rare extrusive carbonatites49(Keller, 1981, 2001; Hubberten et al., 1988). These are spa-50tially and temporally associated with (1) magmatic diatreme51breccias, some of which are polygenic mixtures consisting of52mafic cumulates, carbonatite and silicate rock fragments, fill53pipe structures in the sovite bodies (Baranyi, 1977; Katz &54Keller, 1981; Hubberten et al., 1988; Keller et al., 1990;55Sigmund, 1996), and (2) bergalite dyke rocks (Hubberten56et al., 1988; Schleicher et al., 1990; Keller, 1991, 1997),57which are carbonate-bearing, silica-undersaturated rocks58with a mineral assemblage of meliliteþ sodaliteþ perovskite59þ biotite � nepheline þ apatite þ magnetite þ calcite60(Keller, 2001, 2008).61In this study we investigate apatite separates from62twelve KVC rocks, including four sovitic carbonatites63from surface outcrops and drill cores, one diatreme brec-64cia, one bergalite, one essexite, one mondhaldeite, one65shonkinite porphyry, and three phonolitic rocks66(Table 1a). Details on the petrography, mineralogy and67geochemistry of these samples are given in Hubberten68et al. (1988); Keller et al. (1990) and Schleicher et al.69(1990). Apatite separates were produced by using standard70heavy liquid methods and were subsequently mounted in
Fig. 1. Simplified geological map of the Rhine Graben district (a), and the Kaiserstuhl Volcanic Complex (b).
2 L.-X. Wang et al. Fast Track Article
unco
rrecte
d proo
f
eschweizerbart_xxx
1 epoxy, polished and carbon coated for further cathodolu-2 minescence studies (CL), electron microprobe analysis3 (EMPA), secondary ion mass spectrometry (SIMS), and4 total reflection X-ray fluorescence analysis (TXRF).
5 3. Analytical methods
6 3.1. Electron microprobe analysis
7 Major and minor element compositions of apatites were8 determined using a JEOL 8900 electron microprobe oper-9 ated in wavelength-dispersive mode at Tubingen
10 University. A beam current of 10 nA and an acceleration11 voltage of 15 kV were used in connection with a defocused12 beam diameter of 10 mm. Durango apatite was used as the13 standard for Ca, P and F. Other standards included albite14 for Na, diopside for Si, hematite for Fe, rhodonite for Mn,15 barite for S, tugtupite for Cl, La-glass (REE16G) for La,16 Ce-glass (REE16G) for Ce, and synthetic GaAs and17 SrTiO3 for As and Sr, respectively. Counting times were18 16 s for the Ca peak; 30 s for P, F, Na, Si, S and Sr peaks;19 and 60 s for Fe, Mn, As, La, Ce and Cl peaks, resulting in20 detection limits that are very similar to those reported by21 Marks et al. (2012). Data reduction was performed using22 the internal ZAF matrix correction software of JEOL23 (Armstrong, 1991).24 Cathodoluminescence imaging, a powerful tool to25 reveal internal textures of minerals (e.g., Gotze et al.,26 2013), was used prior to EMPA to document the internal
27structures in apatite crystals. According to Stormer et al.28(1993) and Goldoff et al. (2012), long-time exposure under29an electron beam might cause F and Cl diffusion depending30on the orientation of the apatite crystal. To monitor these31effects, we performed two tests. In the first test we ana-32lyzed apatite grains of sample M2 by EMPA along profiles33parallel to the c- and a-axes of a crystal. Subsequently,34these grains were exposed for around 3 minutes to the low-35intensity (10 nA) CL-imaging tool integrated into the elec-36tron microprobe. The same grains were then reanalyzed37with the analysis points set close to the previous ones. For38the second test the sample holder was re-polished and39carbon coated again. Additional apatite grains were ana-40lyzed and subsequently these grains were exposed for41around 3 minutes to the high-intensity electron gun of a42CL microscope (12–15 mA). These grains were then re-43analyzed close to the previous points with the electron44microprobe. The results from these two tests are shown in45Fig. 2. Low-intensity CL did not cause statistically signifi-46cant diffusion effects for F and Cl. In contrast, the high-47intensity CL enhanced F diffusion. Thus, we conclude that48the F and Cl diffusion effects are related to the intensity of49the electron beam, and the use of a low (,10 nA) beam50current for the CL study should not affect the quality of F51and Cl analyses.
523.2. Secondary-ion mass spectrometry
53Trace elements were analyzed using an upgraded54CAMECA ims-3f ion microprobe at the Max-Planck-55Institute for Chemistry in Mainz. The primary beam con-56sisted of negative oxygen ions at a nominal accelerating57potential of 12.5 kV and a beam current of 20 nA, resulting58in a sputtering surface of 15–20 mm. Positive secondary59ions of 16O, 23Na, 30Si, 44Ca, 88Sr, 89Y, 90Zr, 93Nb, 139La,60
174Yb, 232Th, 238U were extracted in that order, using an62acceleration potential of 4.5 kV with a 25 eV energy63window and fully open entrance and exit slits. Ions were64counted in a peak jumping mode and ratioed to 44Ca to65quantify element abundances. Each measurement con-66sisted of a six-cycle routine. At the beginning of each67measurement, the energy distribution of 16O and subse-68quent peak centers for 44Ca, 88Sr, 14�Ce and 232Th were69determined by scanning the peak in 20 steps across a 1.570per mille wide B-field and the neighboring masses adjusted71to these new peak centers. Measurement times were 30 s72for Nb, Er, Yb; 20 s for Sm, Eu, Gd, Dy, U; 8 s for Y and73Nd, 5 s for Ba, La, Ce, Pr, Th and 1 s for all other elements.74Ion yields in phosphates are comparable to those from75silicate glasses (Sano et al., 2002) and sensitivity factors76were determined with MPI-DING reference glasses KL2-77G, ML3B (Jochum et al., 2006) and NIST 610 (Jochum &78Stoll, 2008). Energy filtering of molecular ion species was79achieved with an offset of �80 eV. In apatite, the only80molecular ions in the mass range of the REE that are not81suppressed by energy filtering are monoxides and fluor-82ides. However, the only substantial fluoride interference in
Table 1. (a) Investigated apatite samples of this study and (b)abbreviations for the various apatite types. All samples wereanalyzed by EPMA, some samples were additionally analyzedby SIMS (in italics) and by TXRF (in bold).
(b) Illustration of the abbreviationsC.Ap apatite from sovitic carbonatitesS.Ap apatite from alkaline silicate rocksD.Ap lower CL brightness apatites from diatreme
brecciaB.Ap higher CL brightness apatites from diatreme
brecciaR.Ap replaced apatites from diatreme brecciaR.Ap.rel relict part of the R.ApR.Ap.sec secondary replaced areas of the R.Ap
Fast Track Article Signatures of apatites from Kaiserstuhl 3
unco
rrecte
d proo
f
eschweizerbart_xxx
1 the apatites was found at mass 159 where 140Ce19F inter-
2 feres with Tb whereas hydride, hydroxide, and chloride3 peaks were found to be insignificant (Crozaz & Zinner,4 1985; Zinner & Crozaz, 1986). Accordingly, Tb was not5 analyzed and oxide interferences on Eu, Gd, Dy, Er and Yb6 were corrected offline. Here, the general light-REE-7 enriched apatite pattern produces significant interferences8 where BaOþ interferes with 153Euþ, 141PrOþ with 157Gdþ;9
147SmOþ with 163Dyþ; 151EuOþ with 167Erþ; and both10
158GdOþ and 158DyOþ with 174Ybþ. The MOþ to Mþ
11 ratios used for the correction are 0.057, 0.165, 0.127,12 0.058, 0.049, 0.145 and 0.127, respectively. The largest13 uncertainties are introduced by the correction of GdO and14 DyO interferences on Yb.
15 3.3. Total reflection X-ray fluorescence analysis
16 These analyses were performed using a S2 PICOFOX17 TXRF (Bruker AXS) at Universitat Tubingen. Hand-18 picked apatite separates were milled to a fine-grained19 powder (, 15 mm). Around 1 mg apatite powder was20 dispensed in 1 ml Triton (1 vol.%) as solvent in an21 Eppendorf tube; 10 ml of a 10 mg/l Se solution was22 added as an internal standard. These suspensions were23 then homogenized for about 2 minutes, and aliquots of24 these suspensions were pipetted onto quartz glass sam-25 ple discs and dried on a hot plate at 70�C. The dried26 samples where then analyzed for 1000 s (live time)27 using a Mo X-ray tube with an operating voltage of28 50 kV and a beam current of 600 mA. The spectra were29 fitted with the Spectra 6.2 software (Bruker Nano30 GmbH). Further technical details can be found in31 Marks et al. (2012). TXRF is barely used to analyze32 solid materials (minerals and rocks) and available ana-33 lytical approaches are limited (e.g., Garcıa-Heras et al.,
341997; Misra et al., 2002; Marks et al., 2012). Here, we35compare the results and uncertainties of apatite compo-36sitions analyzed by EPMA, SIMS and TXRF. Our data37show good consistency between the three methods (see38supplementary material, linked to this article and freely39available online on the GSW website of the journal:40http://eurjmin.41geoscienceworld.org/), demonstrating that TXRF is reli-42able in quantifying REE.
434. Results
444.1. CL imaging
45The observed variations in CL intensity and color46(Fig. 3) are generally related to the contents of lumi-47nescence activator elements, e.g. REE, Mn and Eu48(Mariano, 1988; Koberski & Keller, 1995; Mitchell49et al., 1997; Waychunas, 2002; Dempster et al.,502003). To better illustrate the different apatite types51from various KVC rocks, we use the abbreviations52listed in Table 1b.
534.1.1. Apatite from silicate rocks and carbonatites
54Apatites from silicate rocks (S.Ap) generally show com-55plex zonations, either oscillatory or patchy (Fig. 3a-c),56which, however, do not reflect any detectable systematic57compositional variation. Apatites from carbonatites58(C.Ap) appear brighter than those from silicate rocks and59occasionally contain micro-holes and calcite inclusions60(Fig. 3d-e). Sample KB3, which has an unusual orbicular61texture in hand-specimen, shows varying CL intensities in62apatite separates, and at least two types of apatite (brightly63luminescent and dull) can be distinguished (Fig. 3f).
1.8
2.0
2.2
2.4
2.6
2.8
(a)
(001)
(100)
before LE CL
after LE CL
F (
wt%
)
1.8
2.0
2.2
2.4
2.6
2.8before HE CL
after HE CL
(001)
(100)
F (
wt%
)
(c)
0.0
0.2
0.4
0.6
(001) (100)
before LE CL
after LE CL
Cl (
wt%
)
(b)
0.0
0.2
0.4
0.6before HE CL
after HE CL
Cl (
wt%
)
(001)(100)
(d)
Fig. 2. Comparison of F and Cl contents in apatites from sample M2 before and after EMPA integrated Low intensity CL (LE CL, 10 nA; a &b) and external High intensity CL (HE CL, 12–15 mA; c & d).
4 L.-X. Wang et al. Fast Track Article
unco
rrecte
d proo
f
eschweizerbart_xxx
1 4.1.2. Apatite from bergalite
2 Most apatite grains from bergalite show regular zoning with3 a gray inner core, darker outer core, and a bright rim4 (Fig. 3g-h). The boundaries between these zones are gener-5 ally abrupt and often display euhedral outlines. In rare cases,6 the core appears to be partially resorbed (rounded shape;7 Fig. 3h). Apparently unzoned grains show a similar CL8 brightness as the cores (Fig. 3g) and have similar chemical9 compositions (see below). We assume that this is an effect of
10 different orientations of the crystals in the sample mount.
11 4.1.3. Apatite from diatreme breccia
12 Apatites from a diatreme breccia comprise three different13 populations (Fig. 3i): (1) grains with lower CL brightness14 (D.Ap); (2) higher-brightness grains with occasional micro-15 holes and calcite inclusions (B.Ap, Fig. 3j); and (3) grains16 with replacement textures (R.Ap), resembling apatite popula-17 tion (1) partially replaced by apatite (2) preferentially around18 the margins of crystals and along cracks (Fig. 3k-l).
19 4.2. Compositional variation
20 4.2.1. Comparison of C.Ap and S.Ap
21 Strontium is generally higher in C.Ap than in S.Ap (Fig.22 4a; Table 2). Iron and Mn are low in C.Ap (mostly , 0.005
23apfu) with more than 50 % of values below the respective24detection limits (360 ppm for Fe, 220 ppm for Mn),25whereas they are higher in S.Ap (mostly . 0.005 apfu,26Fig. 4a). The C.Ap and S.Ap datasets generally overlap in27their REE and Na contents (Fig. 4b and 4e), with apatites28from sovite KB3 showing a REE variation (from 0.06 to290.15 apfu) that is distinct from other C.Ap (Table 2).30Thorium in C.Ap is commonly lower than 10 ppm with31some cases below the detection limit of 0.4 ppm (Table 3,32Fig. 4f). Uranium in C.Ap is always below the detection33limit (0.04 ppm). In contrast, the S.Ap has much higher Th34(40 to 200 ppm) and U (6 to 72 ppm) contents (Fig. 4f).35Silicon and S contents are generally lower in C.Ap than in36S.Ap (Fig. 4c), reaching 0.17 apfu Si and 0.09 apfu S in37S.Ap. An exception, again, is sovite KB3, with Si contents38as high as 0.27 apfu. Arsenic is higher in C.Ap (23–4739ppm) than in S.Ap (,10 ppm; Table 3). Niobium in C.Ap40is always higher than 2 ppm, compared to , 1ppm in S.Ap41(Fig. 4f).42Variation in F is very similar in C.Ap and S.Ap (Fig. 4d),43whereas Cl is significantly lower in C.Ap (,0.01 apfu)44than in S.Ap (0.02-0.2 apfu). In both groups, F and Cl are45negatively correlated to each other (Fig. 4d). Bromine is46always below the detection limit of TXRF in C.Ap (around470.5 ppm), but generally between 1 and 3 ppm in S.Ap with48an exceptionally high Br content of 12 ppm in apatites from49a phonolitic dyke (Table 4).
e
b c
k l
g
Zonation
j
Calcite
dCalcite
a
h
Rim
Outer-coreInner-core
iB. Ap
D. Ap
R.Ap
f
Fig. 3. Cathodoluminescence and topographic images of apatites from the Kaiserstuhl Volcanic Complex. (a) Patchy zoning in apatite fromessexite (Pulverbuck). (b) Core-rim zoning texture in apatite from phonolite dome M15. (c) Complex zoned apatite crystals from phonolitetuff M2. (d) Calcite inclusions and holes in apatite from sovite KB3. (e) Patchy zoned apatite from sovite at Badloch. (f) Overview of apatitesfrom sovite KB3, showing populations with high and low CL brightness, respectively. (g) Apatites from bergalite, mostly consisting of core-mantle-rim zoned crystals. A few unzoned apatites are similar in CL brightness to the core of the zoned crystals. (h) Core-mantle-rim zonedapatite crystal from bergalite with relatively broad rim. (i) Three groups of breccia apatites: B.Ap¼ bright apatites; D.Ap¼ dark apatites; anda third group, partly replaced crystals (R.Ap). (j) Bright crystals from breccias contain micro-holes and calcite inclusions as found in sovitesamples (Figs. 3d&f). (k) & (l) Partially replaced dark grains. The replacement occurs preferably around the rim and along cracks in thecrystals. In several images (a, b, c, e, g, h, i and l) of this figure, analysis spots are visible.
Fast Track Article Signatures of apatites from Kaiserstuhl 5
unco
rrecte
d proo
f
eschweizerbart_xxx
1 4.2.2. Apatite from bergalite
2 The CL zonation in bergalite apatites (Fig. 3i-k) corre-3 lates with compositional differences (Fig. 5; Table 2).4 The inner cores are characterized by low Sr and REE and5 intermediate Na contents (Fig. 5; Table 2). The rims are6 highest in Sr and lowest in Na. The outer cores generally7 overlap with the inner cores in terms of Sr, REE and Na.8 Silicon contents increase from inner to outer core and9 decrease towards the rims (Fig. 5). Sulfur concentrations
10 in the inner cores show a large variation and overlap11 considerably with those in the outer cores. The rims,12 however, are notably poorer in S (Table 2). The average13 F content in the inner cores (0.35 apfu) is slightly lower14 than in the outer cores (0.47 apfu; Fig. 5). The highest15 values were found in the rims (0.75 apfu). The contents16 of Cl and calculated OH are lower in the rims than in the17 cores (Table 2).
184.2.3. Apatite from diatreme breccia
19The B.Ap grains show large variation in many elements20and commonly overlap with the data for D.Ap (Fig. 6).21Fluorine and Cl contents in both groups exhibit significant22differences. R.Ap.rel always shows the same composi-23tional range as defined by D.Ap, whereas R.Ap.sec is24compositionally similar to B.Ap.
255. Discussion
265.1. Interpretation of the compositional differences27between C.Ap and S.Ap
28Chemical variation of magmatic apatites depends on apa-29tite–melt partitioning coefficients, melt composition,
0.01
0.02
0.02 0.04 0.06Sr (apfu)
Fe+
Mn
(ap
fu)
0.02
0.04
0.06
0.08
0.02 0.04 0.06La + Ce (apfu)
Na
(ap
fu)
0.02
0.04
0.06
0.08
0.04 0.08 0.12 0.16Si (apfu)
S (
apfu
)
0.3
0.4
0.5
0.6
0.7
0.01 0.1Cl (apfu)
F (
apfu
)
C.Ap
S.Ap
(a) (b)
(c)(d)
1E1
1E2
1E3
1E4
1E5
La Ce Pr Nd Sm Eu Gd Dy Er Yb
Co
nce
ntr
atio
n/ C
ho
nd
rite
C.ApS.Ap
(e)
C.ApS.Ap
(f)
0.01
0.10
1
10
100
0 1 10 100 1000
N b
(p
pm
)
Th (ppm)
1
d.l.
0.000.00
0.03 0.10
0.08
0.00d.l.
0.18 0.0010.2
0.10
0.00
Fig. 4. Comparison of the chemical composition of carbonatitic apatites (C.Ap) and apatites from associated silicate rocks (S.Ap) based onEMPA (a - d) and SIMS (e & f) data, respectively. Sovite KB3 has a special orbicular texture in hand specimen (Keller et al. 1990). It is asomewhat exotic sample of the C.Ap group regarding its Si and REE contents (Figs. 6 & 8; Table 2) and is not shown here. d.l. ¼ detectionlimit. Data below the detection limit are not shown. Chondrite-normalization after Sun and McDonough (1989).
6 L.-X. Wang et al. Fast Track Article
unco
rrecte
d proo
f
eschweizerbart_xxx
Tab
le2
.A
ver
age
EM
PA
dat
a(w
t%)
for
apat
ites
fro
md
iffe
ren
tro
cks
of
the
Kai
sers
tuh
lV
olc
anic
Co
mp
lex
.
C.A
pA
pat
ites
fro
mB
erg
alit
e
Bad
loch
KK
68
0S
chel
ing
enK
B3
M1
2
L.B
.CL
H.B
.CL
I.C
ore
O.C
ore
Rim
Un
zon
en¼
39
n¼
20
n¼
50
n¼
12
n¼
16
n¼
30
n¼
8n¼
11
n¼
3
SiO
20
.43
(16
)0
.49
(16
)0
.43
(16
)2
.66
(28
)1
.07
(40
)1
.29
(23
)2
.22
(29
)0
.95
(23
)1
.42
(36
)L
a 2O
30
.28
(8)
0.2
9(8
)0
.24
(5)
1.6
5(1
8)
0.9
0(2
0)
0.2
5(1
0)
0.3
8(9
)0
.36
(7)
0.2
6(1
5)
Ce 2
O3
0.4
8(1
2)
0.5
3(1
4)
0.4
7(8
)2
.75
(29
)1
.62
(36
)0
.47
(12
)0
.69
(12
)0
.55
(8)
0.5
1(1
8)
FeO
0.0
5(2
)0
.04
(1)
0.0
6(2
)0
.06
(3)
0.0
6(3
)0
.07
(2)
0.0
9(4
)0
.04
(1)
0.0
5(2
)M
nO
0.0
3(1
)0
.03
(1)
0.0
4(1
)b
.d.l
.b
.d.l
.0
.03
(1)
b.d
.l.
0.0
2(1
)0
.02
(1)
CaO
54
.05
(44
)5
5.1
7(6
4)
54
.12
(38
)5
1.7
7(3
3)
53
.04
(76
)5
4.2
3(4
9)
53
.8(3
5)
53
.53
(46
)5
4.1
2(5
3)
Na 2
O0
.09
(2)
0.1
7(5
)0
.27
(9)
0.1
0(2
)0
.16
(6)
0.0
8(4
)0
.12
(3)
0.0
5(1
)0
.09
(2)
SrO
0.8
8(1
0)
0.6
1(6
)0
.64
(5)
0.8
0(5
)0
.94
(7)
0.5
5(1
4)
0.6
8(5
)1
.2(1
9)
0.6
8(1
1)
P2O
54
0.5
9(4
8)
40
.25
(34
)4
0.3
1(6
2)
36
.26
(40
)3
9.4
7(6
8)
39
.18
(53
)3
7.0
2(4
9)
39
.79
(59
)3
9.0
4(6
2)
As 2
O5
b.d
.l.
b.d
.l.
b.d
.l.
b.d
.l.
b.d
.l.
b.d
.l.
b.d
.l.
b.d
.l.
b.d
.l.
SO
30
.08
(3)
0.2
1(6
)0
.26
(7)
0.2
5(6
)0
.07
(2)
0.6
1(2
0)
0.5
9(6
)0
.24
(4)
0.5
6(1
6)
Cl
0.0
2(1
)0
.03
(1)
0.0
4(1
)0
.04
(1)
0.0
2(1
)0
.15
(3)
0.1
7(2
)0
.05
(2)
0.1
4(3
)F
2.2
4(1
3)
1.5
4(8
)1
.47
(17
)1
.65
(13
)2
.49
(15
)1
.31
(21
)1
.71
(20
)2
.75
(14
)1
.33
(13
)T
ota
l9
9.2
29
9.3
69
8.3
59
7.9
99
9.8
49
8.2
29
7.4
69
9.5
29
8.2
2
Fo
rmu
lab
ase
do
n8
oxy
gen
ato
ms
Ca
4.9
44
.98
4.9
44
.86
4.8
84
.95
4.9
74
.92
4.9
5S
r0
.04
0.0
30
.03
0.0
40
.05
0.0
30
.03
0.0
60
.03
Fe
0.0
00
.00
0.0
00
.00
0.0
00
.00
0.0
10
.00
0.0
0M
n0
.00
0.0
00
.00
0.0
00
.00
0.0
00
.00
0.0
00
.00
La
0.0
10
.01
0.0
10
.05
0.0
30
.01
0.0
10
.01
0.0
1C
e0
.02
0.0
20
.01
0.0
90
.05
0.0
10
.02
0.0
20
.02
Na
0.0
20
.03
0.0
40
.02
0.0
30
.01
0.0
20
.01
0.0
2S
um
A5
.03
5.0
75
.04
5.0
65
.04
5.0
25
.07
5.0
25
.02
P2
.93
2.8
72
.90
2.6
92
.87
2.8
32
.70
2.8
92
.82
Si
0.0
40
.04
0.0
40
.23
0.0
90
.11
0.1
90
.08
0.1
2A
s0
.00
0.0
00
.00
0.0
00
.00
0.0
00
.00
0.0
00
.00
S0
.00
0.0
10
.02
0.0
20
.00
0.0
40
.04
0.0
20
.04
Su
mB
2.9
72
.93
2.9
62
.94
2.9
62
.98
2.9
32
.98
2.9
8C
l0
.00
0.0
00
.01
0.0
10
.00
0.0
20
.03
0.0
10
.02
F0
.60
0.4
10
.40
0.4
60
.68
0.3
50
.47
0.7
50
.36
OH
(cal
.)0
.39
0.5
90
.60
0.5
40
.32
0.6
20
.51
0.2
50
.62
Fast Track Article Signatures of apatites from Kaiserstuhl 7
unco
rrecte
d proo
f
eschweizerbart_xxx
Tab
le2
.C
on
tin
ued
S.A
pA
pat
ites
fro
mB
recc
ia
M1
3M
14
Pu
lver
bu
ckM
15
SM
12
M2
23
48
1-3
a/3
c
B.A
pD
.Ap
R.A
p.s
ecR
.Ap
.rep
n¼
44
n¼
43
n¼
75
n¼
26
n¼
21
n¼
33
n¼
16
n¼
21
n¼
3n¼
3
SiO
20
.52
(6)
0.9
9(3
5)
0.7
7(2
6)
1.0
1(4
1)
1.1
3(3
0)
1.1
2(4
1)
0.5
1(3
4)
0.9
2(1
2)
0.1
6(5
)1
.19
(17
)L
a 2O
30
.13
(2)
0.3
0(2
0)
0.2
4(9
)0
.28
(18
)0
.24
(14
)0
.31
(19
)0
.29
(12
)0
.18
(4)
0.5
1(8
)0
.19
(1)
Ce 2
O3
0.1
9(4
)0
.57
(34
)0
.46
(13
)0
.46
(29
)0
.46
(23
)0
.59
(32
)0
.51
(23
)0
.41
(9)
0.8
2(4
)0
.45
(3)
FeO
0.2
3(3
)0
.10
(4)
0.0
8(3
)0
.08
(4)
0.0
7(3
)0
.07
(4)
0.0
7(5
)0
.08
(3)
0.0
6(3
)0
.07
(1)
Mn
O0
.06
(2)
0.0
4(1
)0
.04
(1)
0.0
3(2
)0
.04
(1)
0.0
4(2
)0
.04
(1)
0.0
4(2
)0
.03
(2)
0.0
4(2
)C
aO5
4.5
7(3
8)
54
.28
(64
)5
4.4
9(4
3)
54
.68
(65
)5
4.9
3(4
3)
54
.33
(64
)5
4.5
5(7
9)
54
.25
(41
)5
3.2
7(6
6)
54
.58
(32
)N
a 2O
0.2
4(2
)0
.17
(8)
0.1
4(5
)0
.16
(4)
0.1
5(5
)0
.12
(3)
0.1
5(1
3)
0.1
6(2
)0
.43
(1)
0.1
6(1
)S
rO0
.26
(3)
0.5
5(1
5)
0.4
5(1
2)
0.5
3(2
1)
0.5
4(1
7)
0.7
6(2
9)
0.6
1(3
4)
0.3
5(6
)0
.62
(4)
0.3
4(2
)P
2O
54
0.1
2(3
2)
39
.23
(85
)3
9.8
5(8
0)
39
.07
(99
)3
8.9
2(6
5)
39
.25
(76
)4
0.7
6(8
8)
39
.57
(46
)4
0.6
2(3
0)
39
.33
(47
)A
s 2O
5b
.d.l
.b
.d.l
.b
.d.l
.b
.d.l
.b
.d.l
.b
.d.l
.b
.d.l
.b
.d.l
.b
.d.l
.b
.d.l
.S
O3
1.0
3(9
)0
.74
(29
)0
.86
(26
)1
.06
(37
)1
.11
(32
)0
.91
(25
)0
.28
(16
)0
.29
(7)
0.1
2(2
)0
.36
(13
)C
l1
.27
(3)
0.5
6(2
6)
0.4
3(1
0)
0.2
4(1
2)
0.2
9(1
0)
0.3
2(1
3)
0.0
4(2
)0
.20
(4)
0.0
3(1
)0
.19
(1)
F1
.31
(6)
1.4
8(2
5)
2.0
8(2
1)
2.2
9(3
8)
2.1
5(3
2)
2.0
8(1
0)
2.3
0(4
8)
0.8
7(2
5)
1.9
5(2
0)
1.1
2(6
)T
ota
l9
9.9
39
9.0
19
9.8
99
9.8
91
00
.03
99
.90
10
0.1
19
7.3
29
8.6
29
8.0
2
Fo
rmu
lab
ase
do
n8
oxy
gen
ato
ms
Ca
4.9
44
.95
4.9
54
.96
4.9
74
.94
4.9
44
.97
4.8
84
.97
Sr
0.0
10
.03
0.0
20
.03
0.0
30
.04
0.0
30
.02
0.0
30
.02
Fe
0.0
20
.01
0.0
10
.01
0.0
10
.01
0.0
10
.01
0.0
00
.00
Mn
0.0
00
.00
0.0
00
.00
0.0
00
.00
0.0
00
.00
0.0
00
.00
La
0.0
00
.01
0.0
10
.01
0.0
10
.01
0.0
10
.01
0.0
20
.01
Ce
0.0
10
.02
0.0
10
.01
0.0
10
.02
0.0
20
.01
0.0
30
.01
Na
0.0
40
.03
0.0
20
.03
0.0
20
.02
0.0
20
.03
0.0
70
.03
Su
mA
5.0
25
.04
5.0
25
.05
5.0
55
.03
5.0
25
.04
5.0
45
.04
P2
.87
2.8
32
.86
2.8
02
.78
2.8
22
.92
2.8
62
.94
2.8
3S
i0
.04
0.0
80
.06
0.0
90
.10
0.1
00
.04
0.0
80
.01
0.1
0A
s0
.00
0.0
00
.00
0.0
00
.00
0.0
00
.00
0.0
00
.00
0.0
0S
0.0
70
.05
0.0
50
.07
0.0
70
.06
0.0
20
.02
0.0
10
.02
Su
mB
2.9
82
.96
2.9
82
.95
2.9
52
.97
2.9
82
.96
2.9
62
.96
Cl
0.1
80
.08
0.0
60
.03
0.0
40
.05
0.0
10
.03
0.0
00
.03
F0
.35
0.4
00
.56
0.6
10
.57
0.5
60
.61
0.2
40
.53
0.3
0O
H(c
al.)
0.4
70
.52
0.3
80
.35
0.3
80
.40
0.3
80
.74
0.4
70
.67
b.d
.l.¼
bel
ow
det
ecti
on
lim
it.F
orm
ula
calc
ula
tio
ns
are
bas
edo
na
tota
lof
eig
htc
atio
ns
inth
eA
and
Xsi
tes.
OH
was
calc
ula
ted
assu
min
gst
oic
hio
met
ry,i
.e.Fþ
Clþ
OH¼
1in
the
Zsi
te.
L.B
.CL¼
Lo
wb
rig
htn
ess
inC
Lim
age;
H.B
.CL¼
Hig
hb
rig
htn
ess
inC
Lim
age.
8 L.-X. Wang et al. Fast Track Article
unco
rrecte
d proo
f
eschweizerbart_xxx
1 charge-balance constraints for multi-element substitutions,2 and possibly other factors (e.g., Peng et al., 1997; Sha &3 Chappell, 1999; Harlov et al., 2002). In the following, we4 evaluate the influences of these factors on the chemical5 differences between C.Ap and S.Ap.
6 5.1.1. Apatite–melt partition coefficients and host melt7 compositions
8 According to experimental work (Klemme & Dalpe, 2003;9 Prowatke & Klemme, 2006; Hammouda et al., 2010),
10 apatite in the carbonatitic system has lower Dapatite-melt
11 values for Sr, La and Ce and higher values for Nb in12 comparison with those from the silicate-melt system13 (Supplementary Table; Klemme & Dalpe, 2003;14 Prowatke & Klemme, 2006; Hammouda et al., 2010).15 Indeed, C.Ap from the Kaiserstuhl has higher Nb concen-16 trations than S.Ap (Table 3), implying that partition coeffi-17 cients play an important role. However, the Sr, La and Ce18 concentrations in C.Ap are not lower than those in S.Ap,19 although they have lower Dapatite-melt values in C.Ap20 (Table 2). We attribute this to the effect of melt composi-21 tion, since carbonatitic melts are usually enriched in Sr and22 REE compared to silicate melts (Hubberten et al., 1988;23 Keller et al., 1990; Martin et al., 2013). For instance,24 according to Keller et al., 1990 and unpublished), carbo-25 natites from the KVC contain up to 16 000 ppm Sr, whereas26 the silicate rocks contain , 1000 ppm.
27 5.1.2. Substitution mechanisms
28 A positive correlation between S and Si for C.Ap and S.Ap29 (Fig. 7a) indicates that sulfur is mainly incorporated30 according to the substitution
Si4þþS6þ ¼ 2P5þ
3232(Pan & Fleet, 2002). However, the high Si contents in33apatite from sovite KB3 cannot be explained this way.34Apatites from this sample also show the highest La þ Ce35contents, which clearly correlate with Si (Fig. 7b), pointing36to the importance of the substitution
REE3þþSi4þ ¼ Ca2þþP5þ;
3838as proposed for apatites from other localities (e.g., Comodi39et al., 1999; Harlov et al., 2002, 2005). The combination of40these two substitutions (Fig. 7c) demonstrates that both41mechanisms can explain the variation observed in C.Ap42and S.Ap. However, for sample KB3, a significant devia-43tion from the 1:1 line exists. This is probably because only44La þ Ce were analyzed by EPMA, thereby underestimat-45ing the total REE. Alternatively, other substitution46mechanisms may play a role for apatites from this sample.47The substitution mechanism REE3þþNaþ ¼ 2 Ca2þ
48(Harlov et al., 2002, 2005), however, does not play a49major role since Na does not correlate with La þ Ce50(Fig. 4b) and apatites from KB3 are not exceptionally51Na-rich (Table 2).
525.2. Interpretation of the zoned apatites from bergalite
53Zoning patterns in apatite may be related to a variety of54processes such as fractional crystallization, magma mix-55ing, magmatic degassing, diffusion, disequilibrium crys-56tallization, partial dissolution, and recrystallisation (e.g.,57Jolliff et al., 1989; Brenan, 1994; Sha & Chappell, 1999;58Tepper & Kuehner, 1999; Boyce & Hervig, 2008, 2009;59Chakhmouradian et al. 2008; Rønsbo, 2008). The zoning
Table 3. Average SIMS data (ppm) for apatites from different rocks of the Kaiserstuhl Volcanic Complex.
Fast Track Article Signatures of apatites from Kaiserstuhl 9
unco
rrecte
d proo
f
eschweizerbart_xxx
1 textures found in the bergalite apatites (Fig. 5) display2 discontinuous concentration changes in both volatile (F,3 Cl and S) and non-volatile elements (e.g., Sr and Si).4 Therefore, it seems unlikely that magma degassing can5 fully explain these zonations.6 Clear compositional differences between inner and outer7 cores are only observed for Si (Fig. 5). The difference may8 be caused by fractionation of the parental melt combined9 with changing trace element substitution mechanisms in
10 apatite. However, the observed CL textures and chemical11 profiles show abrupt changes (Fig. 3g-h) and rounded12 shapes and embayments in some cores (Fig. 3g-h). This13 can be explained by the stirring of magmas during ascent in14 high-level magma bodies. Such stirring could result in
15partial thermal and chemical changes of the magma and16lead to disequilibrium crystallization of apatite.17The chemical differences between core and rim probably18rather reflect the evolving magma composition during berga-19lite crystallization. The rims of the bergalite apatites are very20similar to those of C.Ap (e.g., relatively high CL brightness,21low Si, Cl and S, but high Sr contents; Fig. 8), whereas the22cores are chemically similar to S.Ap. To further evaluate the23compositional similarity, we calculated Sr concentrations of24the related parental melts from the apatite compositions. We25assume that the cores crystallized from a melilite nephelinite26melt, whereas the rims crystallized from a late-stage melt27with carbonatitic affinity. The DSr partition coefficients28used here are 5.1 for apatite/phonolite, 1.56 for apatite/
Table 4. TXRF data (ppm) for apatites from different rocks of the Kaiserstuhl Volcanic Complex.
1 tephrite and 0.53 for apatite/carbonatite (Prowatke &2 Klemme, 2006; Hammouda et al., 2010). A comparison of3 the modeling result with the whole-rock Sr contents (Keller,4 unpublished data) shows that the calculated Sr concentrations5 are in agreement with, or close to, those from whole-rock data6 (Fig. 9). Moreover, it also supports the hypothesis that the
7cores of the bergalite apatites correspond to S.Ap., whereas8the rims correspond to C.Ap.9The overgrowth of C.Ap-like rims on S.Ap-like
10cores in the apatite crystals from bergalite can be11explained by the following processes: (1) fractional12crystallization of a melilite nephelinite magma; and
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08
Sr
0 0.005 0.01 0.015 0.02
La
0.01 0.015 0.02 0.025 0.03
Ce
0 0.005 0.01 0.015 0.02 0.025 0.03
Na
apfu
0.25 0.35 0.45 0.55 0.65 0.75 0.85 0.95
F
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07
S
0 0.05 0.1 0.15 0.2 0.25
Si
0 0.005 0.01 0.015 0.02 0.025 0.03 0.035
Cl
apfu
Rim
Inner-core
Outer-core
Rim
Inner-core
Outer-core
Fig. 5. Chemical variations (EMPA data) between different zones in apatites from bergalite.
0 0.02 0.04 0.06
Sr
0 0.005 0.01 0.015
La
0.01 0.02 0.03
Ce
0 0.02 0.04 0.06 0.08
Na
0 0.04 0.08 0.12
Si0 0.01 0.02 0.03 0.04
S0.2 0.4 0.6
F0 0.01 0.02 0.03 0.04
Cl
B.Ap
D.Ap
Rp.A
p.rel
Rp.
Ap.
sec
apfu apfu apfu apfu
0
0
B.Ap
D.Ap
Rp.Ap.rel
Rp.Ap.sec
Fig. 6. Chemical variations (EMPA data) of different apatite populations from a diatreme breccia. B.Ap. apatites showing bright CL intensity;D.Ap, apatites showing dark CL intensity; R.Ap.sec, secondary bright replacement of the apatites; R.Ap.rel, relics of the replaced apatites.
Fast Track Article Signatures of apatites from Kaiserstuhl 11
unco
rrecte
d proo
f
eschweizerbart_xxx
1 (2) mixing of two independent magmas, a silicate and2 a carbonatitic one. These two magmas will either mix3 to form a homogeneous intermediate product, or4 remain immiscible. Watkinson & Wyllie (1971) pro-5 duced calcite þ cancrinite þ melilite at 1 kbar using a
6carbonated nepheline-rich liquid at ca. 600 � C, indi-7cating that carbonatite can be generated by fractional8crystallization of a carbonated nephelinite magma.9Accordingly, it was assumed that the bergalite melt
10fractionated from a parental CO2-rich olivine melilite11nephelinite mantle-derived magma (Keller et al., 1990;12Keller, 2008). This fractionation caused the enrichment13of CO2, Sr, and REE in the evolved melt (Keller, 1991,141997). The residual melt from which the C.Ap-like15rims crystallized would, in this case, have carbonatitic16affinity. Dempster et al. (2003) demonstrated that dis-17continuously zoned apatite can monitor changing per-18meability in granites, in which apatite cores grew early19in the magma chamber, whereas rims record late-stage20crystallization within more isolated interstitial melt21pockets in a highly solidified and compacted crystal22mush. Similarly, C.Ap-like rims may represent the late-23stage crystallization from carbonate-rich melt pockets.24Mixing of carbonatitic and melilititic magmas could25theoretically yield an intermediate product such as26bergalite. However, the Sr contents in the rims do27not support this assumption, because they generally28plot outside the interval defined by the two assumed29end-members, C.Ap and S.Ap (Fig. 8). Moore et al.30(2009) interpreted the genesis of silicocarbonatite31dykes by mingling between carbonatitic and silicate32magmas, involving liquid immiscibility. If we assume33a carbonatitic melt was injected into a melilititic melt34with immiscibility during the late stage, it is still not35clear whether core-rim zoned apatite or replaced apa-36tite (as observed in diatreme breccias; see below) will37form during this mingling process.38In summary, the zoning textures in bergalitic apatites39imply that they initially nucleated in a silicate melt and40later developed a rim while in equilibrium with a carbona-41titic melt. We favour hypothesis (1) to explain the observed42apatite textures.
435.3. Interpretation of the replacement textures in44apatite from diatreme breccias
45Mineral replacement reactions are generally related to dis-46solution-reprecipitation processes resulting from chemical47weathering, leaching, alteration, metamorphism or meta-48somatism (Putnis, 2002; Engvik et al., 2009). Replacement49of chlorapatite by hydroxy-fluorapatite during metasoma-50tism in metagabbro has been described by Harlov et al.51(2002) and Engvik et al. (2009) and has been experimen-52tally reproduced under alkaline hydrothermal conditions53(Yanagisawa et al., 1999; Harlov et al., 2002). However,54the replacement of hydroxyapatite by fluorapatite, as docu-55mented experimentally by Rendon-Angeles et al. (2000),56has so far not been reported for natural apatites. In the57present study, the replacement textures found in apatites58from the diatreme breccia reveal hydroxy-fluorapatite59replaced by fluorine-hydroxyapatite (Figs. 3i-l and 7).60The R.Ap.sec represents a late-stage fluid/melt re-61equilibration product. Compositional and CL similarities
0.10
S (
apfu
)
C.ApS.ApKB3
0.04
0.08
0.12
0.16
0.10 0.20 0.300.00
0.00
La +
Ce
(apf
u)
Si (apfu)
1:1
0.04
0.08
0.12
0.16
0.10 0.200.00
0.20
0.00 0.30
La +
Ce
+ S
(ap
fu)
Si (apfu)
1:1
0.02
0.04
0.06
0.08
0.10 0.200.000.00
0.30
Si (apfu)
(a)
(b)
(c)
1:1
Fig. 7. Correlations between REE (LaþCe) and S with Si contents ofapatite samples from the Kaiserstuhl Volcanic Complex.
12 L.-X. Wang et al. Fast Track Article
unco
rrecte
d proo
f
eschweizerbart_xxx
100
1000
10000
100000
100 1000 10000 100000
Sr calculated (ppm)
Outer-core (bergalite)
Rim (bergalite)
S.Ap
C.Ap
Sr
wh
ole
ro
ck (
pp
m)
1:1line
Inner-core (bergalite)
Fig. 9. Calculated Sr contents of carbonatite and silicate melts based on their apatite compositions (EMPA results). Partition coefficients(DSr) between apatite and silicate melt are from Prowatke & Klemme (2006), those between apatite and carbonatite melt are from Hammoudaet al. (2010). Whole rock data for silicate rocks are from Keller (unpublished), those for sovites from Keller et al. (1990). The parental melt inwhich the apatite core from bergalite formed is assumed to be of silicate composition, whereas the rim is thought to have crystallized from acarbonatitic melt.
Fig. 8. CL characteristics and compositional variations of apatites from the KVC. The rims of the bergalite apatites and B.Ap and R.Ap.sec frombreccias, show compositional similarities with C. Ap. In contrast, the cores of bergalite apatites and D.Ap and R.Ap.rel are similar to S.Ap.
Fast Track Article Signatures of apatites from Kaiserstuhl 13
unco
rrecte
d proo
f
eschweizerbart_xxx
1 between R.Ap.rel and S.Ap indicate that the former repre-2 sent relict grains of early-formed apatites crystallized from3 a silicate melt. In contrast, R.Ap.sec is similar to C.Ap,4 implying formation from a carbonatitic melt. These obser-5 vations suggest that a carbonatitic magma captured sili-6 cate-rock fragments and rapidly rose to the surface to form7 the diatreme breccias. During this process, metasomatism8 of the silicate xenoliths by the carbonatitic melt caused the9 replacement textures observed in some of the apatites.
10 5.4. Abundances and variability of volatile elements11 (F, Cl, Br, S) in apatite
12 Apatites from the KVC are mostly fluorine-hydroxya-13 patite or hydroxy-fluorapatite (Fig. 10). Their Cl con-14 tents are quite low, especially in the C.Ap (Fig. 11a).15 This agrees well with previous studies where apatites16 from carbonatite complexes were reported as Cl-poor17 with mostly less than 0.1 wt.% Cl (e.g., Eby, 1975;18 Hogarth, 1989; Seifert et al., 2000; Patino Douce et al.,19 2011). The low Cl content was explained by Cl parti-20 tioning into an aqueous fluid phase, which generally21 coexisted with the carbonatitic melt (Gittins, 1989;22 Seifert et al., 2000). Bromine contents generally corre-23 late with Cl: in Cl-poor C.Ap Br is below the detection24 limit but reaches around 2.5 ppm in relatively Cl-rich25 S.Ap. This is consistent with previously found positive
26correlations between Cl and Br (O’Reilly & Griffin,272000; Marks et al., 2012), implying Cl and Br behave28similarly during incorporation by apatite.29Sommerauer & Katz-Lehnert (1985) determined carbon30concentrations in some apatites from Kaiserstuhl using cou-31lometric titration and Fourier-transformed infrared (FTIR)32spectroscopy, revealing up to 3.9 wt.% CO2 in carbonatitic33apatites and up to 0.9 wt.% in apatites from the silicate rock.34However, the calcite inclusions in C.Ap observed in the35present work and the strong zoning in apatite from the36bergalite samples make the interpretation of Sommerauer37& Katz-Lehnert (1985) problematic, because their work was38performed on bulk powder samples.39Sulfur is incorporated in apatite as sulfate (Pan & Fleet,402002), and most natural apatites contain ,0.5 wt.% SO3
41(Broderick et al., 2007), although concentrations of up to 242wt.% have been reported (Imai et al., 1993; Streck &43Dilles, 1998; Broderick et al., 2007; Parat et al., 2011).44The S.Ap from KVC are rich in SO3 (0.74–1.11 wt.%),45indicating that the KVC silicate magma was enriched in S46and crystallized under relatively oxidizing conditions47(Imai et al., 1993; Peng et al., 1997; Parat et al., 2002).48In contrast, C.Ap is relatively poor in SO3 (0.02–0.5 wt.%).49These low contents of both SO3 and Cl (Fig. 11b) probably50reflect the relative depletion of carbonatitic magma in both51elements. It is unlikely to represent a Cl and S loss during52degassing, because when nephelinitic magma fractionated53to carbonatitic magma, non-volatile elements (e.g., Si)
F
Cl OH
F
OH
C.Ap
S.Ap Bergalite:Core
Rim
Breccia:
B.Ap
D.Ap
(a) (b) (c)
(c)
(b)(a)
Fig. 10. Triangular plot of the halogen and hydroxyl contents in apatite from the KVC. (a) C.Ap and S.Ap, (b) bergalite, (c) diatreme breccia.
14 L.-X. Wang et al. Fast Track Article
unco
rrecte
d proo
f
eschweizerbart_xxx
1 vary with Cl and S as well (see the compositional variation2 in core-rim zoned apatite from bergalite in Fig. 5).
3 6. Conclusions
4 Textural and chemical variations of apatites from the KVC5 provide important information for the genesis of their host6 rocks: Apatites from bergalite allow for the reconstruction7 of the genetic relations between carbonatites and asso-8 ciated alkaline silicate rocks. Their textural and composi-9 tional core-rim zonation implies that these apatites initially
10 crystallized from a silicate melt and further developed in a11 melt with carbonatitic affinity. This carbonatitic melt is12 likely a product of prolonged fractional crystallization of13 an initial CO2-rich melilite nephelinite melt. On the other14 hand, apatites from silicate rock fragments found in a15 diatreme breccia were subsequently metasomatized by16 late-injected carbonatitic melt, as shown by their replace-17 ment textures. Overall, our study shows that apatite is an18 equally sensitive monitor for primary magmatic processes19 (e.g., fractional crystallization) and secondary processes,20 such as metasomatic overprint.
21 Acknowledgements: Dr. M. Rahn and Prof. H.22 Schleicher are gratefully thanked for kindly providing23 some of the investigated apatite separates and for24 insightful discussions. Prof. Ch. Ma is thanked for25 the helpful suggestions and comments. We also26 thank B. Walter for assistance with apatite separation27 and M. Mangler for discussions about the TXRF28 method. The China Scholarship Council (CSC) is29 thanked for granting scholarship (2010641006) to the30 first author. H. Teiber is supported by the Deutsche31 Forschungsgemeinschaft (grant MA 2563/3-1), which32 is gratefully acknowledged. We appreciate the33 detailed reviews of A. Chakhmouradian and T.34 Hammouda as well as the editorial handling of R.35 Giere.
36References
37Armstrong, J.T. (1991): Quantitative elemental analysis of indivi-
38dual microparticles with electron beam instruments. in
39‘‘Electron Probe Quantitation’’, K.F.J. Heinrich & D.E.
40Newbury ed. Plenum, New York, 261–315.
41Bailey, D.K. (1993): Carbonate magmas. J. Geol. Soc., 50, 637–651.
42Baranyi, I. (1977): Petrography and geochemistry of the subvolcanic
43breccias of the Kaiserstuhl volcano (SW-Germany). N. Jb.
44Mineral. Abh., 128, 254–284.
45Baranyi, I., Lippolt, H.J., Todt, W. (1976): K-Ar
46Altersbestimmungen an tertiaren Vulkaniten des
47Oberrheingraben-Gebietes: II Die Alterstraverse vom Hegau
