TY - JOUR
AU - Jackson, S. E.
AB - Abstract Silicate-bearing and silicate-free natrocarbonatites have been recently erupted at Oldoinyo Lengai. Low-pressure (20 and 100 MPa) experiments were undertaken to determine the phase relationships of a natural silicate-bearing natrocarbonatite (OL5) from Oldoinyo Lengai. Nepheline appears first on the liquidus. Over a protracted cooling interval (>200°C), only silicate and oxide phases crystallize, with nyerereite and gregoryite precipitating at temperatures <635°C. Results of this study, used in conjunction with the phase assemblage and mineral chemistry data available for natrocarbonatites from Oldoinyo Lengai, are used to test hypotheses relating silicate-bearing natrocarbonatite to the silicate-free variety. Our preferred model involves low-P (∼20 MPa) in situ fractional crystallization of silicate minerals from a silicate-bearing natrocarbonatite parent magma to produce silicate-free natrocarbonatites. Subsequent evolution of low-density, silica-poor natrocarbonatite is dominantly controlled by carbonate (nyerereite and gregoryite) crystallization and fractionation. The low density and viscosity of silicate-free natrocarbonatite allow these magmas to erupt more easily than silicate-bearing natrocarbonatite. Introduction The origin of many carbonatites, and of natrocarbonatites from Oldoinyo Lengai in particular, remains problematic. Two models are usually proposed to explain the origin of carbonatites: (1) immiscible carbonate liquids exsolved from parental CO2-bearing silicate melts (melilitite or nephelinite), that are of mantle derivation; or (2) primary carbonate melts derived directly by partial melting of the mantle. Recent experimental work (Kjarsgaard et al., 1995) supports the model that at Oldoinyo Lengai, natrocarbonatite magmas are exsolved from highly peralkaline wollastonite nephelinite magmas. It has been suggested, however, that natrocarbonatite at Oldoinyo Lengai might be derived from a primary sodic dolomitic carbonatite parent magma by fractional crystallization and wall-rock interaction (Sweeney et al., 1995). Complementing these experimental studies, Bell & Simonetti (1996) determined the Nd, Sr and Pb isotopic signature of the natrocarbonatites, but could not use these data to distinguish between a primary mantle origin for natrocarbonatites and an origin by liquid immiscibility. However, we prefer the coherence from multiple lines of evidence that support the hypothesis that natrocarbonatites are exsolved from highly peralkaline nephelinites (see Bell & Keller, 1995). There are two distinct types of natrocarbonatite at Oldoinyo Lengai: silicate-bearing and silicate-free natrocarbonatites. A representative analysis of each type is listed in Table 1. Silicate-free natrocarbonatites are defined as natrocarbonatites containing <0.5 wt % SiO2 and are dominated by phenocrysts of nyerereite and gregoryite, with extremely rare silicate phases. This type of natrocarbonatite is represented by 70% of all the available analyses of natrocarbonatite from Oldoinyo Lengai (Fig. 1). Silicate-bearing natrocarbonatites contain >0.5 wt % SiO2, although most analyses have 1.0–5.0 wt % SiO2 with an increased frequency at 3.0–4.0 wt % SiO2 (Fig. 1). These natrocarbonatites contain phenocrysts and glomeroporphyritic aggregates (silicate spheroids) of nepheline, clinopyroxene, melanite, melilite and wollastonite, plus others in addition to nyerereite and gregoryite. Table 1: Whole-rock major element analyses of representative natrocarbonatites . OL5 . OL102 . SiO2 3.07 0.16 TiO2 0.10 0.02 Al2O3 0.85 n.d. Fe2O3 n.d. 0.28 FeO 1.19 n.d. MnO 0.32 0.60 MgO 0.38 0.52 CaO 14.80 12.86 Na2O 27.90 30.42 K2O 5.81 9.14 P2O5 1.05 0.75 SrO 1.15 1.43 BaO 0.85 2.17 F 1.64 4.10 Cl 2.54 5.18 SO3 2.55 3.72 CO2 30.40 31.55 H2O n.d. 0.56 Total (–O = F,Cl) 93.32 99.12 . OL5 . OL102 . SiO2 3.07 0.16 TiO2 0.10 0.02 Al2O3 0.85 n.d. Fe2O3 n.d. 0.28 FeO 1.19 n.d. MnO 0.32 0.60 MgO 0.38 0.52 CaO 14.80 12.86 Na2O 27.90 30.42 K2O 5.81 9.14 P2O5 1.05 0.75 SrO 1.15 1.43 BaO 0.85 2.17 F 1.64 4.10 Cl 2.54 5.18 SO3 2.55 3.72 CO2 30.40 31.55 H2O n.d. 0.56 Total (–O = F,Cl) 93.32 99.12 Porphyritic silicate-bearing natrocarbonatite OL5: major element data from Simonetti et al. (1997); F, Cl, CO2, SO3 and H2O by Analytical Chemistry, GSC, Ottawa. Porphyritic sil-icate-free natrocarbonatite OL102: data from Keller & Spettel (1995). Open in new tab Table 1: Whole-rock major element analyses of representative natrocarbonatites . OL5 . OL102 . SiO2 3.07 0.16 TiO2 0.10 0.02 Al2O3 0.85 n.d. Fe2O3 n.d. 0.28 FeO 1.19 n.d. MnO 0.32 0.60 MgO 0.38 0.52 CaO 14.80 12.86 Na2O 27.90 30.42 K2O 5.81 9.14 P2O5 1.05 0.75 SrO 1.15 1.43 BaO 0.85 2.17 F 1.64 4.10 Cl 2.54 5.18 SO3 2.55 3.72 CO2 30.40 31.55 H2O n.d. 0.56 Total (–O = F,Cl) 93.32 99.12 . OL5 . OL102 . SiO2 3.07 0.16 TiO2 0.10 0.02 Al2O3 0.85 n.d. Fe2O3 n.d. 0.28 FeO 1.19 n.d. MnO 0.32 0.60 MgO 0.38 0.52 CaO 14.80 12.86 Na2O 27.90 30.42 K2O 5.81 9.14 P2O5 1.05 0.75 SrO 1.15 1.43 BaO 0.85 2.17 F 1.64 4.10 Cl 2.54 5.18 SO3 2.55 3.72 CO2 30.40 31.55 H2O n.d. 0.56 Total (–O = F,Cl) 93.32 99.12 Porphyritic silicate-bearing natrocarbonatite OL5: major element data from Simonetti et al. (1997); F, Cl, CO2, SO3 and H2O by Analytical Chemistry, GSC, Ottawa. Porphyritic sil-icate-free natrocarbonatite OL102: data from Keller & Spettel (1995). Open in new tab Immiscible natrocarbonate liquids exsolved from wollastonite nephelinite melts in the experiments of Kjarsgaard et al. (1995) contained between 1.26 and 5.15 wt % SiO2. Importantly, it should be noted that this SiO2 content is part of the ‘silicate component’, which is completely in solution in the natrocarbonate liquid. Moreover, previous experiments (at P <400 MPa and T <1000°C) on silicate–carbonate liquid immiscibility (e.g. Koster Van Groos & Wyllie, 1973; Freestone & Hamilton, 1980; Hamilton et al., 1989; Kjarsgaard & Hamilton, 1989) all demonstrated that exsolved carbonate liquids that contained >25 wt % Na2O + K2O also contained >1 wt % SiO2. These observations indicate that, for the P–T–X conditions relevant to Oldoinyo Lengai, exsolved parental natrocarbonatite melts are silicate bearing (>1 wt % SiO2). Thus a dichotomy exists between the predominance of silicate-free natrocarbonatite lavas and the compositions of parental natrocarbonate liquids produced by experiment (Fig. 1). Fig. 1. Open in new tabDownload slide Frequency–SiO2 diagram for natrocarbonatite lavas, illustrating a bimodal distribution, with a high- and a low-SiO2 population. Sources of data: Keller & Krafft (1990), Church & Jones (1995), Dawson et al. (1995, 1996), Keller & Hoefs (1995), and Simonetti et al. (1997). For comparison, the SiO2 concentration in natrocarbonatite liquids from the immiscibility experiments of C. M. Petibon (unpublished data, 1998; parameters are P ≤ 100 MPa; T ≤ 900°C) is shown for comparison. Two-liquid plus solid phase experiments (Kjarsgaard et al., 1995) demonstrated that exsolved natrocarbonate melts are in equilibrium (saturated) with (ferromagnesian) silicate phases. They also suggested that if exsolved natrocarbonatite melts separate from their silicate host, further precipitation and fractionation of these solid phases would produce a halogen-rich and SiO2-, TiO2-, Al2O3-, MgO- and FeO-depleted natrocarbonatite magma (i.e. a silicate-free natrocarbonatite). Complementary experimental studies in model silicate–carbonate systems (Lee & Wyllie, 1998) have demonstrated that an immiscible carbonate melt that separates from its conjugate silicate liquid will first precipitate silicate minerals on cooling. To further examine silicate mineral fractionation in carbonate melts, and the relationship between silicate-bearing and silicate-free natrocarbonatite, a low-pressure phase equilibrium study was undertaken on a natural silicate-bearing natrocarbonatite (OL5) from Oldoinyo Lengai. Phase relationships and mineral compositions observed in the experiments and in naturally occurring natrocarbonatites are used together to constrain a model fornatrocarbonatite magma differentiation at OldoinyoLengai. Experimental Rationale and Methods The use of liquidus studies in establishing relationships between magmas is well established (see Wyllie, 1984); none the less it is useful to reiterate some of the fundamentals. First, the experiments should reproduce or be consistent with the sequence and composition of the phases appearing in the natural rocks. This establishes that the P–T–volatile conditions used are comparable with those occurring in nature. Second, liquidus studies are by their very nature an examination of bulk, equilibrium crystallization. Subsequent comparison with natural rocks must take into account that far more complicated mechanisms may have been operative (e.g. crystal fractionation, mixing). One potential pitfall in doing liquidus studies is that the starting composition may be incorrectly chosen (e.g. does not correspond to a parent magma and/or a liquid composition). None the less, coherence in mineral composition and crystallization order between that observed in the experiments and in the natural rocks may eliminate the need for more complicated petrogenetic models. Experimental methods Approximately 0.07 g of finely ground OL5 powder was loaded into pre-cleaned gold tubes (2.5 mm o.d., 20 mm length), which were sealed by arc-welding. The experimental charges were run in Tuttle-type, externally heated pressure vessels in the experimental laboratory at the Geological Survey of Canada, Ottawa. Experiments were performed at pressures of 20 and 100 MPa, over the temperature interval 550–900°C. Argon was used as a pressure medium. Run times varied from 4.5 to 618 h. Temperature was measured with a thermocouple located in a well at the base of the vessel. A temperature correction was employed based on calibration with an internal thermocouple. Reported temperatures are believed to be accurate to ±5°C. Pressure was measured using a Bourdon tube gauge (Astrogauge), calibrated against a Heise laboratory standard gauge, and is thought to be accurate to ±2 MPa. Samples were quenched at the end of the run by a jet of compressed air. Quenching rates were on average 300°C/min for the first minute. No reversal experiments were attempted because of problems with nucleation of metastable high-Ti clinopyroxenes (e.g. Kjarsgaard et al., 1995). Analytical methods After each run, capsules were weighed, then punctured and reweighed to check for weight loss or gain. Loss of weight after puncturing indicated the presence of a coexisting fluid phase. After examination with a binocular microscope, the run products were mounted in epoxy and polished. Samples were polished in liquid paraffin and cleaned in acetone to avoid any contact with water or alcohol, which dissolve alkali-bearing carbonates and salts. Samples were first studied by scanning electron microscopy (SEM) at the Geological Survey of Canada (Cambridge S200 and S360) to identify the different phases and to obtain back-scattered images used to locate electron microprobe (EMP) spots. Major elements for different phases were determined using a Cameca SX50 electron microprobe. Samples were analysed at the Geological Survey of Canada, using WDS technique, and at Memorial University of Newfoundland, using combined WDS–EDS techniques. Analyses were made using an accelerating potential of 15 kV and a beam current of 10 nA. Crystals and silicate glass were analysed with a slightly defocused beam (10 μm), and quenched carbonate liquids with a 40 μm × 50 μm rastered beam. Detection limits are ∼0.10 oxide wt % for elements analysed by EDS technique and ∼500 ppm for elements analysed by WDS technique. Relative standard deviations are ∼0.2–1% for crystals and silicate glass. Volatilization problems arose when analysing carbonate solid phases. To quantify this error, analyses of nyerereite and gregoryite in natrocarbonatite CML5 (kindly donated by Tony Peterson) analysed for this study were compared with the high-quality stoichiometric analyses of Peterson (1990). A correction factor was calculated based on the results of the two studies, and subsequently applied to the raw data for Na2O, CaO and K2O for the nyerereite and gregoryite analyses presented in this paper. Results Petrography of silicate-bearing natrocarbonatite OL5 A sample of silicate-bearing natrocarbonatite OL5 was kindly provided by Keith Bell. It was sampled from the extremely blocky Chaos Crags flow (Simonetti et al., 1997), which erupted in 1993. Major element concentrations are reported in Table 1. OL5 is a porphyritic silicate-bearing natrocarbonatite. Phenocryst minerals and silicate spheroids are set in a groundmass (Fig. 2a) that consists of fine-grained intergrowths of sylvite and Na-rich gregoryite. Phenocryst minerals present in OL5 include nyerereite, gregoryite, clinopyroxene, wollastonite, nepheline, melanite garnet, apatite, and pyrrhotite. In addition, one melilite phenocryst was observed. No obvious zoning was observed in silicate phenocryst minerals during examination by polarized light microscopy and by back-scattered electron image (BSEI) on the scanning electron microscope, i.e. the crystals are very homogeneous in composition. Silicate spheroids represent ∼8–10 vol. % of the lava (Simonetti et al., 1997). The spheroids have cores composed of a single phenocryst, or glomeroporphyritic phenocryst clusters. Mineral phases observed in the spheroids include clinopyroxene, melanite garnet, wollastonite and nepheline; mutual inclusion relationships amongst these phases are common, suggesting co-precipitation. The spheroid cores are rimmed by a siliceous mesostasis that contains microphenocrysts of nepheline, clinopyroxene, melanite garnet and a sodium-rich gregoryite. Fig. 2. Open in new tabDownload slide (a) Back-scattered electron image (BSEI) of OL5, showing a silicate spheroid in porphyritic natrocarbonatite. The silicate spheroid is cored by a phenocryst of clinopyroxene (Cpx), which contains inclusions of melanite garnet (Gt) and nepheline (Ne). Microphenocrysts in the groundmass of the spheroid include nepheline, clinopyroxene, melanite garnet, wollastonite (Wo) and Na-rich gregoryite. The groundmass of the spheroids is of wollastonite nephelinite composition. The surrounding porphyritic natrocarbonatite consists of phenocrysts of nyerereite (NY) and gregoryite (GRE) in a groundmass composed of small intergrowths of sylvite and Na-rich carbonate. An isolated phenocryst of wollastonite is also present in the natrocarbonatite. (b) BSEI of experiment CP45 (20 MPa, 850°C) showing crystals of melilite, nepheline and perovskite (Pvk) in equilibrium with quenched carbonate liquid. (c) BSEI of experiment CP107 (100 MPa, 600°C) showing crystals of nepheline, clinopyroxene and nyerereite in quenched carbonate liquid. It should be noted that the size and shape of the nyerereite phenocrysts are easily distinguished from the blade- and needle-shaped nyerereite quench crystals. (d) BSEI of experiment CP108 (20 MPa, 600°C) showing crystals of clinopyroxene, melanite garnet, nyerereite and gregoryite in quenched carbonate liquid. Similar porphyritic silicate-bearing natrocarbonatite samples have been previously described by Dawson et al. (1994, 1996) and Church & Jones (1995). Dawson et al. (1996) also reported rare melilite and titanite phenocrysts in the spheroids, and melilite, titanite, pyrrhotite, perovskite and Ti-magnetite microphenocrysts in the matrix. Experiments Run products Details of the experimental conditions (P, T, t) and the phase assemblage present in each run are listed in Table 2. Run products consisted of quench carbonate liquid ± crystals of nepheline, melanite garnet, clinopyroxene, apatite, wollastonite, melilite, nyerereite, gregoryite and a coexisting vapour (fluid) phase (Fig. 2b, c and d). Minor crystal phases observed include perovskite, sodalite, K-feldspar and Fe-rich spinel. A P–T diagram constructed from the experimental data is presented in Fig. 3. Experiments indicate that pressure has a significant effect on mineral stability. In the temperature range studied, nepheline, clinopyroxene, melanite garnet and nyerereite are present in both at 20 and 100 MPa experiments, whereas melilite, wollastonite and gregoryite are present only in 20 MPa experiments (Table 2 and Fig. 3). At 100 MPa, nyerereite first precipitates between 600 and 625°C. In contrast, at 20 MPa nyerereite precipitates at the slightly higher temperatures (625–650°C; Fig. 3). Gregoryite first crystallizes between 600 and 625°C at 20 MPa, but is not observed in 100 MPa experiments (Fig. 3). Decreasing pressure (20 vs 100 MPa) raises the temperature at which silicate phases first precipitate, with the exception of clinopyroxene, which first crystallizes at a higher temperature in the 100 MPa experiments (Fig. 3). Table 2: Run data Sample . P (MPa) . T (°C) . Time (h) . Phase assemblage . CP88 100 900 4.5 LC + Ne + F CP106 100 850 20 LC + Ne + F CP79 100 800 116 LC + Ne + F CP96 100 750 382 LC + Ne + F CP98 100 700 161 LC + Ne + K-Fd + F CP94 100 645 618 LC + Ne + Cpx + Mela + F CP118 100 625 87 LC + Ne + Cpx + F CP107 100 600 114 LC + Ne + Cpx + NY + Ap + F CP128 100 575 255 LC + Ne + Cpx + NY + Ap + F CP127 100 550 114 LC + Ne + Cpx + NY + Ap + F CP90 20 900 4 LC + Ne + Fe-spinel + F CP45 20 850 46 LC + Ne + Melilite + Perovskite + Fe-spinel + F CP61 20 800 109 LC + Ne + Melilite + Fe–Mn-spinel + F CP80 20 775 65 LC + Ne + Mela + Melilite + Fe-spinel + F CP51 20 750 116 LC + Ne + Mela + Wo + Melilite + F CP57 20 700 164 LC + Ne + Mela + Wo + F CP71 20 650 255 LC + Ne + Mela + Wo + F CP129 20 625 236 LC + Ne + Cpx + Mela + Wo + NY + Ap + F CP108 20 600 114 LC + Ne + Cpx + Mela + NY + GRE + Ap + F CP126 20 575 114 LC + Ne + Cpx + Mela + NY + GRE + Ap + F CP112 20 550 88 LC + Cpx + Mela + NY + GRE + sodalite + Ap + F Sample . P (MPa) . T (°C) . Time (h) . Phase assemblage . CP88 100 900 4.5 LC + Ne + F CP106 100 850 20 LC + Ne + F CP79 100 800 116 LC + Ne + F CP96 100 750 382 LC + Ne + F CP98 100 700 161 LC + Ne + K-Fd + F CP94 100 645 618 LC + Ne + Cpx + Mela + F CP118 100 625 87 LC + Ne + Cpx + F CP107 100 600 114 LC + Ne + Cpx + NY + Ap + F CP128 100 575 255 LC + Ne + Cpx + NY + Ap + F CP127 100 550 114 LC + Ne + Cpx + NY + Ap + F CP90 20 900 4 LC + Ne + Fe-spinel + F CP45 20 850 46 LC + Ne + Melilite + Perovskite + Fe-spinel + F CP61 20 800 109 LC + Ne + Melilite + Fe–Mn-spinel + F CP80 20 775 65 LC + Ne + Mela + Melilite + Fe-spinel + F CP51 20 750 116 LC + Ne + Mela + Wo + Melilite + F CP57 20 700 164 LC + Ne + Mela + Wo + F CP71 20 650 255 LC + Ne + Mela + Wo + F CP129 20 625 236 LC + Ne + Cpx + Mela + Wo + NY + Ap + F CP108 20 600 114 LC + Ne + Cpx + Mela + NY + GRE + Ap + F CP126 20 575 114 LC + Ne + Cpx + Mela + NY + GRE + Ap + F CP112 20 550 88 LC + Cpx + Mela + NY + GRE + sodalite + Ap + F LC, carbonate liquid; F, fluid (CO2–H2O mixture, CO2 rich); Ne, nepheline; K-Fd, potassic feldspar; Cpx, clinopyroxene; Mela, melanite garnet; NY, nyerereite; Ap, apatite; Wo, wollastonite; GRE, gregoryite. Open in new tab Table 2: Run data Sample . P (MPa) . T (°C) . Time (h) . Phase assemblage . CP88 100 900 4.5 LC + Ne + F CP106 100 850 20 LC + Ne + F CP79 100 800 116 LC + Ne + F CP96 100 750 382 LC + Ne + F CP98 100 700 161 LC + Ne + K-Fd + F CP94 100 645 618 LC + Ne + Cpx + Mela + F CP118 100 625 87 LC + Ne + Cpx + F CP107 100 600 114 LC + Ne + Cpx + NY + Ap + F CP128 100 575 255 LC + Ne + Cpx + NY + Ap + F CP127 100 550 114 LC + Ne + Cpx + NY + Ap + F CP90 20 900 4 LC + Ne + Fe-spinel + F CP45 20 850 46 LC + Ne + Melilite + Perovskite + Fe-spinel + F CP61 20 800 109 LC + Ne + Melilite + Fe–Mn-spinel + F CP80 20 775 65 LC + Ne + Mela + Melilite + Fe-spinel + F CP51 20 750 116 LC + Ne + Mela + Wo + Melilite + F CP57 20 700 164 LC + Ne + Mela + Wo + F CP71 20 650 255 LC + Ne + Mela + Wo + F CP129 20 625 236 LC + Ne + Cpx + Mela + Wo + NY + Ap + F CP108 20 600 114 LC + Ne + Cpx + Mela + NY + GRE + Ap + F CP126 20 575 114 LC + Ne + Cpx + Mela + NY + GRE + Ap + F CP112 20 550 88 LC + Cpx + Mela + NY + GRE + sodalite + Ap + F Sample . P (MPa) . T (°C) . Time (h) . Phase assemblage . CP88 100 900 4.5 LC + Ne + F CP106 100 850 20 LC + Ne + F CP79 100 800 116 LC + Ne + F CP96 100 750 382 LC + Ne + F CP98 100 700 161 LC + Ne + K-Fd + F CP94 100 645 618 LC + Ne + Cpx + Mela + F CP118 100 625 87 LC + Ne + Cpx + F CP107 100 600 114 LC + Ne + Cpx + NY + Ap + F CP128 100 575 255 LC + Ne + Cpx + NY + Ap + F CP127 100 550 114 LC + Ne + Cpx + NY + Ap + F CP90 20 900 4 LC + Ne + Fe-spinel + F CP45 20 850 46 LC + Ne + Melilite + Perovskite + Fe-spinel + F CP61 20 800 109 LC + Ne + Melilite + Fe–Mn-spinel + F CP80 20 775 65 LC + Ne + Mela + Melilite + Fe-spinel + F CP51 20 750 116 LC + Ne + Mela + Wo + Melilite + F CP57 20 700 164 LC + Ne + Mela + Wo + F CP71 20 650 255 LC + Ne + Mela + Wo + F CP129 20 625 236 LC + Ne + Cpx + Mela + Wo + NY + Ap + F CP108 20 600 114 LC + Ne + Cpx + Mela + NY + GRE + Ap + F CP126 20 575 114 LC + Ne + Cpx + Mela + NY + GRE + Ap + F CP112 20 550 88 LC + Cpx + Mela + NY + GRE + sodalite + Ap + F LC, carbonate liquid; F, fluid (CO2–H2O mixture, CO2 rich); Ne, nepheline; K-Fd, potassic feldspar; Cpx, clinopyroxene; Mela, melanite garnet; NY, nyerereite; Ap, apatite; Wo, wollastonite; GRE, gregoryite. Open in new tab The liquidus temperature of OL5 was not determined, because of temperature limitations of the externally heated pressure vessel. The very low modal proportion of nepheline in the experimental charge at 900°C and 100 MPa suggests the liquidus temperature is slightly higher (∼925°C). At 20 MPa and 900°C, however, there is a increased modal proportion of nepheline in the experimental charge, suggesting a higher liquidus temperature at 20 MPa than at 100 MPa. Some minor phases are present in the experiments that were not observed in OL5. These include perovskite, Fe-rich spinel, K-feldspar and sodalite. The absence of perovskite, Fe-spinel, K-feldspar and sodalite in lava OL5 is suggested to be due to their P–T–dependent stability. Typical silicate-free natrocarbonatite eruption temperatures of 590°C have been recorded by Norton & Pinkerton (1997). Perovskite and Fe-rich spinels precipitate at low pressure and high temperature and are not observed in runs at T <750°C. K-feldspar is observed in only one experiment, at 100 MPa and 700°C. Similarly, sodalite was observed in only one run at 20 MPa and 550°C. The absence of these minor phases in lava OL5 is consistent with recorded eruption temperatures and the phase assemblages produced in the experiments of this study. It should be noted that pyrrhotite exists in lava OL5, but was not observed in the experiments. We suggest this is a consequence of poor control over sulphur and oxygen fugacity in the experiments. Comparison of mineral compositions in experiments and OL5 Nyerereite (Na0.8K0.2Ca0.5CO3) compositions are reported in Table 3. The proportions of CaCO3 (Cc), Na2CO3 (Nc) and K2CO3 (Kc) were used to plot the compositions of nyerereite from experiments and lavas on a Cc–Nc–Kc ternary plot (Fig. 4). Nyerereite crystals from the 20 MPa experiments have compositions similar to that of nyerereite phenocrysts from silicate-bearing natrocarbonatite OL5 and those studied by Church & Jones (1995) and Keller & Krafft (1990), although they contain slightly less K2O (Table 3). In contrast, nyerereite from 100 MPa experiments contains ≤4.6 wt % K2O, which is significantly lower than the reported concentration of K2O in the 20 MPa experiments and the lavas (Table 3 and Fig. 4). Little variation in nyerereite composition is shown with decreasing T at fixed P. Fig. 3. Open in new tabDownload slide Pressure–temperature diagram for silicate-bearing natrocarbonatite OL5 at 100 and 20 MPa over the temperature range 900–550°C. Gregoryite (Na1.6K0.1Ca0.15CO3) crystals from the 600 and 575°C experiments at 20 MPa (Table 3, Fig. 4) are extremely similar in composition to gregoryite phenocrysts from silicate-bearing natrocarbonatite OL5 and the gregoryite phenocrysts in lavas studied by Keller & Krafft (1990) and Church & Jones (1995). Gregoryite from the silicate spheroids in OL5 contains more sodium than the gregoryite phenocrysts in the matrix, but it is not as sodium rich as the gregoryite observed by Dawson et al. (1996) in groundmass intergrowths with fluorite (Fig. 4). Gregoryite compositions show increasing K2CO3 and CaCO3 with decreasing Na2CO3 over the temperature interval 600–550°C at 20 MPa (Fig. 3), consistent with previous studies (Cooper et al., 1975). Nepheline major element compositions and proportions of the different end-members [calculated utilizing the method of Peterson, (1989)] are reported in Table 4, and their compositions are plotted on an Ne–Qz–Ks ternary diagram (Fig. 5). Nepheline composition in 100 MPa experiments is temperature dependent. Over a 300°C cooling interval, there is continuous increase in K2O content. In contrast, Na2O in nepheline initially decreases from 17.5 to 16.1 wt % over the temperature interval 850–645°C and then subsequently increases to 16.7 wt % at 600°C and remains at near-constant concentration to 550°C. In contrast, at 20 MPa there is a more limited compositional change in nepheline with decreasing temperature (Table 4). Nepheline crystals from the experiments have a range of major element compositions similar to those exhibited by nepheline phenocrysts in silicate-bearing natrocarbonatites OL5 (this study) and those examined by Dawson et al. (1996). The isolated nepheline phenocrysts and the nepheline phenocrysts within the silicate spheroids in OL5 are similar in composition. In the spheroid groundmass, nepheline microphenocrysts are more potassic and slightly richer in iron as compared with the nepheline phenocrysts in the groundmass, and best resemble nepheline analyses from the 20 MPa experiments. Clinopyroxene major element compositions are reported in Table 5. They are mainly low-Al, low-Ti diopside–hedenbergite–aegirine solid solutions. Clinopyroxenes from the experiments have fairly variable compositions and there is some scatter in the data (e.g. CP118, CP126), such that mineral chemistry variation trends are difficult to quantify. However, for the 100 MPa runs, with cooling from 645 to 550°C clinopyroxenes exhibit enrichment in Na2O and MnO, with depletion in Al2O3 and TiO2. The behaviour of MgO and FeO, although exhibiting some scatter, shows decreasing mg-number (to 575°C) reversing to higher mg-number at 550°C. The 20 MPa data are somewhat more problematic to interpret, again because of scatter in the data. There appears to be a weak enrichment in Al2O3 and TiO2 with decreasing temperature; the opposite effects are observed in clinopyroxene from the 100 MPa experiments. At 20 MPa, MgO and FeO concentration variation in clinopyroxene produces decreasing mg-number (625–575°C) reversing to higher mg-number at 550°C. Compositions of clinopyroxene from OL5 are slightly richer in MgO and poorer in Na2O as compared with those from the experiments. Isolated clinopyroxene phenocrysts and the clinopyroxene phenocrysts in the silicate spheroids from OL5 have similar compositions (Table 5). Clinopyroxene microphenocrysts in OL5 contain higher Ti and Al abundances compared with the phenocrysts, an observation also noted by Dawson et al. (1996). Based on comparisons with the experimental data of this study, the higher concentrations of Ti and Al would be consistent with clinopyroxene microphenocrysts precipitating at lower pressure (e.g. 20 MPa) than the phenocrysts. Melanite garnet has a relatively restricted composition variation, with TiO2 abundances ranging from 8.7 to 13.5 wt % (Table 6, Fig. 6). For plotting purposes, all Fe was arbitrarily recast as Fe3+, because recalculation and assignment of cations by charge balance is problematic (Kjarsgaard et al., 1995). From Fig. 6 it can be seen that the composition of melanite from the experiments overlaps those from OL5, and also the melanite analyses reported by Dawson et al. (1996). The isolated melanite garnet phenocrysts and the melanite garnet phenocrysts of the silicate spheroids in OL5 have similar compositions. Compared with the phenocrysts, the melanite garnet microphenocrysts in the silicate spheroid groundmass contain higher Na2O, and lower TiO2, consistent with the observations of Dawson et al. (1996). Table 3: Representative microprobe analyses of nyerereite and gregoryite phenocrysts of experimental changes and OL5 Sample: . CP107 . CP128 . CP127 . CP129 . CP108 . CP126 . CP112 . OL5 . CP108 . CP126 . CP112 . OL5 . OL5 . Phase: . NY . NY . NY . NY . NY . NY . NY . NY . GRE . GRE . GRE . GRE . GRE . Type: . ec . ec . ec . ec . ec . ec . ec . ph-c . ec . ec . ec . ph-c . microph-c . SiO2 b.d. 0.2 0.3 b.d. b.d. b.d. b.d. 0.3 b.d. 0.3 b.d. 0.2 0.5 CaO 26.5 26.4 25.9 25.9 25.7 25.5 24.8 25.1 10.8 9.9 10.8 10.1 6.6 Na2O 24.3 24.3 24.6 26.2 25.7 22.8 24.9 22.5 46.9 46.5 43.2 45.4 51.7 K2O 4.6 4.5 4.6 5.9 6.0 5.9 6.5 7.7 3.6 3.8 3.9 3.8 2.7 P2O5 1.1 0.7 0.6 1.0 0.7 0.8 0.5 0.6 3.5 3.1 2.7 4.8 3.2 F b.d. b.d. b.d. b.d. b.d. b.d. b.d. b.d. 0.2 0.3 0.8 0.1 b.d. Cl 0.1 b.d. b.d. 0.1 0.1 0.1 b.d. 0.2 0.2 0.3 0.3 0.6 0.7 SO3 1.2 1.2 1.2 1.1 1.0 1.2 1.2 1.0 4.2 4.7 5.6 3.3 4.1 BaO b.d. 0.5 0.6 b.d. 0.5 b.d. 0.5 0.8 b.d. b.d. 0.4 0.7 0.7 SrO 2.3 2.3 2.4 2.4 2.5 2.7 2.6 3.2 1.0 0.7 1.0 1.0 1.0 Total 60.0 60.1 60.2 62.6 62.1 59.0 61.0 61.3 70.4 69.6 68.7 70.0 71.2 CaF2 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.01 0.01 0.02 0.00 0.00 Na2Cl2 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.01 0.01 CaSO4 0.02 0.02 0.02 0.01 0.01 0.02 0.02 0.01 0.05 0.06 0.07 0.04 0.05 Na2P2O6 0.01 0.01 0.00 0.01 0.01 0.01 0.00 0.00 0.02 0.02 0.02 0.03 0.02 FeCO3 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 SrCO3 0.02 0.02 0.02 0.02 0.03 0.03 0.03 0.03 0.01 0.01 0.01 0.01 0.01 BaCO3 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.01 0.00 0.00 0.00 0.00 0.00 CaCO3 0.49 0.49 0.48 0.46 0.46 0.48 0.45 0.47 0.14 0.11 0.11 0.14 0.07 Na2CO3 0.41 0.41 0.42 0.43 0.42 0.40 0.42 0.38 0.73 0.74 0.71 0.71 0.80 K2CO3 0.05 0.05 0.05 0.06 0.07 0.07 0.07 0.09 0.04 0.04 0.04 0.04 0.03 Sample: . CP107 . CP128 . CP127 . CP129 . CP108 . CP126 . CP112 . OL5 . CP108 . CP126 . CP112 . OL5 . OL5 . Phase: . NY . NY . NY . NY . NY . NY . NY . NY . GRE . GRE . GRE . GRE . GRE . Type: . ec . ec . ec . ec . ec . ec . ec . ph-c . ec . ec . ec . ph-c . microph-c . SiO2 b.d. 0.2 0.3 b.d. b.d. b.d. b.d. 0.3 b.d. 0.3 b.d. 0.2 0.5 CaO 26.5 26.4 25.9 25.9 25.7 25.5 24.8 25.1 10.8 9.9 10.8 10.1 6.6 Na2O 24.3 24.3 24.6 26.2 25.7 22.8 24.9 22.5 46.9 46.5 43.2 45.4 51.7 K2O 4.6 4.5 4.6 5.9 6.0 5.9 6.5 7.7 3.6 3.8 3.9 3.8 2.7 P2O5 1.1 0.7 0.6 1.0 0.7 0.8 0.5 0.6 3.5 3.1 2.7 4.8 3.2 F b.d. b.d. b.d. b.d. b.d. b.d. b.d. b.d. 0.2 0.3 0.8 0.1 b.d. Cl 0.1 b.d. b.d. 0.1 0.1 0.1 b.d. 0.2 0.2 0.3 0.3 0.6 0.7 SO3 1.2 1.2 1.2 1.1 1.0 1.2 1.2 1.0 4.2 4.7 5.6 3.3 4.1 BaO b.d. 0.5 0.6 b.d. 0.5 b.d. 0.5 0.8 b.d. b.d. 0.4 0.7 0.7 SrO 2.3 2.3 2.4 2.4 2.5 2.7 2.6 3.2 1.0 0.7 1.0 1.0 1.0 Total 60.0 60.1 60.2 62.6 62.1 59.0 61.0 61.3 70.4 69.6 68.7 70.0 71.2 CaF2 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.01 0.01 0.02 0.00 0.00 Na2Cl2 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.01 0.01 CaSO4 0.02 0.02 0.02 0.01 0.01 0.02 0.02 0.01 0.05 0.06 0.07 0.04 0.05 Na2P2O6 0.01 0.01 0.00 0.01 0.01 0.01 0.00 0.00 0.02 0.02 0.02 0.03 0.02 FeCO3 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 SrCO3 0.02 0.02 0.02 0.02 0.03 0.03 0.03 0.03 0.01 0.01 0.01 0.01 0.01 BaCO3 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.01 0.00 0.00 0.00 0.00 0.00 CaCO3 0.49 0.49 0.48 0.46 0.46 0.48 0.45 0.47 0.14 0.11 0.11 0.14 0.07 Na2CO3 0.41 0.41 0.42 0.43 0.42 0.40 0.42 0.38 0.73 0.74 0.71 0.71 0.80 K2CO3 0.05 0.05 0.05 0.06 0.07 0.07 0.07 0.09 0.04 0.04 0.04 0.04 0.03 End-members determined by the method of Peterson (1990). ec, experimental charge; ph-c, phenocryst; microph-c, microphenocryst; NY, nyerereite; GRE, gregoryite. Open in new tab Table 3: Representative microprobe analyses of nyerereite and gregoryite phenocrysts of experimental changes and OL5 Sample: . CP107 . CP128 . CP127 . CP129 . CP108 . CP126 . CP112 . OL5 . CP108 . CP126 . CP112 . OL5 . OL5 . Phase: . NY . NY . NY . NY . NY . NY . NY . NY . GRE . GRE . GRE . GRE . GRE . Type: . ec . ec . ec . ec . ec . ec . ec . ph-c . ec . ec . ec . ph-c . microph-c . SiO2 b.d. 0.2 0.3 b.d. b.d. b.d. b.d. 0.3 b.d. 0.3 b.d. 0.2 0.5 CaO 26.5 26.4 25.9 25.9 25.7 25.5 24.8 25.1 10.8 9.9 10.8 10.1 6.6 Na2O 24.3 24.3 24.6 26.2 25.7 22.8 24.9 22.5 46.9 46.5 43.2 45.4 51.7 K2O 4.6 4.5 4.6 5.9 6.0 5.9 6.5 7.7 3.6 3.8 3.9 3.8 2.7 P2O5 1.1 0.7 0.6 1.0 0.7 0.8 0.5 0.6 3.5 3.1 2.7 4.8 3.2 F b.d. b.d. b.d. b.d. b.d. b.d. b.d. b.d. 0.2 0.3 0.8 0.1 b.d. Cl 0.1 b.d. b.d. 0.1 0.1 0.1 b.d. 0.2 0.2 0.3 0.3 0.6 0.7 SO3 1.2 1.2 1.2 1.1 1.0 1.2 1.2 1.0 4.2 4.7 5.6 3.3 4.1 BaO b.d. 0.5 0.6 b.d. 0.5 b.d. 0.5 0.8 b.d. b.d. 0.4 0.7 0.7 SrO 2.3 2.3 2.4 2.4 2.5 2.7 2.6 3.2 1.0 0.7 1.0 1.0 1.0 Total 60.0 60.1 60.2 62.6 62.1 59.0 61.0 61.3 70.4 69.6 68.7 70.0 71.2 CaF2 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.01 0.01 0.02 0.00 0.00 Na2Cl2 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.01 0.01 CaSO4 0.02 0.02 0.02 0.01 0.01 0.02 0.02 0.01 0.05 0.06 0.07 0.04 0.05 Na2P2O6 0.01 0.01 0.00 0.01 0.01 0.01 0.00 0.00 0.02 0.02 0.02 0.03 0.02 FeCO3 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 SrCO3 0.02 0.02 0.02 0.02 0.03 0.03 0.03 0.03 0.01 0.01 0.01 0.01 0.01 BaCO3 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.01 0.00 0.00 0.00 0.00 0.00 CaCO3 0.49 0.49 0.48 0.46 0.46 0.48 0.45 0.47 0.14 0.11 0.11 0.14 0.07 Na2CO3 0.41 0.41 0.42 0.43 0.42 0.40 0.42 0.38 0.73 0.74 0.71 0.71 0.80 K2CO3 0.05 0.05 0.05 0.06 0.07 0.07 0.07 0.09 0.04 0.04 0.04 0.04 0.03 Sample: . CP107 . CP128 . CP127 . CP129 . CP108 . CP126 . CP112 . OL5 . CP108 . CP126 . CP112 . OL5 . OL5 . Phase: . NY . NY . NY . NY . NY . NY . NY . NY . GRE . GRE . GRE . GRE . GRE . Type: . ec . ec . ec . ec . ec . ec . ec . ph-c . ec . ec . ec . ph-c . microph-c . SiO2 b.d. 0.2 0.3 b.d. b.d. b.d. b.d. 0.3 b.d. 0.3 b.d. 0.2 0.5 CaO 26.5 26.4 25.9 25.9 25.7 25.5 24.8 25.1 10.8 9.9 10.8 10.1 6.6 Na2O 24.3 24.3 24.6 26.2 25.7 22.8 24.9 22.5 46.9 46.5 43.2 45.4 51.7 K2O 4.6 4.5 4.6 5.9 6.0 5.9 6.5 7.7 3.6 3.8 3.9 3.8 2.7 P2O5 1.1 0.7 0.6 1.0 0.7 0.8 0.5 0.6 3.5 3.1 2.7 4.8 3.2 F b.d. b.d. b.d. b.d. b.d. b.d. b.d. b.d. 0.2 0.3 0.8 0.1 b.d. Cl 0.1 b.d. b.d. 0.1 0.1 0.1 b.d. 0.2 0.2 0.3 0.3 0.6 0.7 SO3 1.2 1.2 1.2 1.1 1.0 1.2 1.2 1.0 4.2 4.7 5.6 3.3 4.1 BaO b.d. 0.5 0.6 b.d. 0.5 b.d. 0.5 0.8 b.d. b.d. 0.4 0.7 0.7 SrO 2.3 2.3 2.4 2.4 2.5 2.7 2.6 3.2 1.0 0.7 1.0 1.0 1.0 Total 60.0 60.1 60.2 62.6 62.1 59.0 61.0 61.3 70.4 69.6 68.7 70.0 71.2 CaF2 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.01 0.01 0.02 0.00 0.00 Na2Cl2 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.01 0.01 CaSO4 0.02 0.02 0.02 0.01 0.01 0.02 0.02 0.01 0.05 0.06 0.07 0.04 0.05 Na2P2O6 0.01 0.01 0.00 0.01 0.01 0.01 0.00 0.00 0.02 0.02 0.02 0.03 0.02 FeCO3 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 SrCO3 0.02 0.02 0.02 0.02 0.03 0.03 0.03 0.03 0.01 0.01 0.01 0.01 0.01 BaCO3 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.01 0.00 0.00 0.00 0.00 0.00 CaCO3 0.49 0.49 0.48 0.46 0.46 0.48 0.45 0.47 0.14 0.11 0.11 0.14 0.07 Na2CO3 0.41 0.41 0.42 0.43 0.42 0.40 0.42 0.38 0.73 0.74 0.71 0.71 0.80 K2CO3 0.05 0.05 0.05 0.06 0.07 0.07 0.07 0.09 0.04 0.04 0.04 0.04 0.03 End-members determined by the method of Peterson (1990). ec, experimental charge; ph-c, phenocryst; microph-c, microphenocryst; NY, nyerereite; GRE, gregoryite. Open in new tab Fig. 4. Open in new tabDownload slide Projection of the compositions of nyerereite and gregoryite in the ternary system Na2CO3–K2CO3–CaCO3. Compositional trend of gregoryite from 20 MPa experiments at 600–550°C as indicated in figure. Source of data from other studies: Keller & Krafft (1990), Peterson (1990), Church & Jones (1995) and Dawson et al. (1996). Apatite was observed in the experiments at temperatures ≤625°C. The major element concentration is broadly similar to those in OL5 (see Table 8, below); both contain significant fluorine (2.6 wt % F in apatite from OL5; >3 wt % F in apatite from an experiment). There is also minor substitution of Si and Na noted in apatite from both OL5 and experiment (britholite end member). Wollastonite crystals from the experiments contain neither FeO nor MgO, but significant MnO (see Table 8, below). In contrast, two analysed wollastonite crystals in OL5 contained FeO (0.93 and 0.95 wt %), MgO (both 0.2 wt %) and MnO (0.44 and 0.38 wt %), similar to levels reported by Dawson et al. (1996). Wollastonite phenocrysts in the groundmass and in the silicate spheroids in OL5 have similar compositions. Table 4: Representative microprobe analyses of nepheline from experimental charges and silicate-bearing natrocarbonatite OL5 Sample: . CP106 . CP79 . CP96 . CP98 . CP94 . CP118 . CP107 . CP128 . CP127 . CP90 . Type: . ec . ec . ec . ec . ec . ec . ec . ec . ec . ec . SiO2 43.5 43.5 42.9 43.0 42.5 43.1 43.1 42.5 43.0 43.4 Al2O3 33.3 33.0 32.9 32.7 31.9 32.5 32.9 32.3 32.9 30.7 FeOt 1.5 1.5 1.7 1.6 1.3 2.1 1.3 1.6 1.4 2.4 Na2O 17.5 17.2 17.1 16.6 16.1 16.7 16.6 16.7 16.7 17.7 K2O 6.2 6.3 6.7 6.6 6.9 6.9 6.9 6.8 7.1 6.1 Total 102.1 101.5 101.3 100.7 99.3 101.6 100.9 100.5 101.0 100.7 Ks 18.7 19.1 20.5 20.4 21.6 21.1 21.3 21.2 21.7 18.9 Nf 2.7 2.6 3.1 2.8 2.5 3.8 2.4 2.9 2.5 4.3 Ne 77.5 76.9 76.5 74.5 73.9 73.8 75.0 75.6 75.2 78.6 An 2.5 1.0 0.0 1.0 1.3 1.0 0.0 1.8 0.0 2.1 Cn 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Qz 0.8 1.9 1.4 2.4 2.5 1.8 2.5 0.8 0.2 0.6 Sample: . CP106 . CP79 . CP96 . CP98 . CP94 . CP118 . CP107 . CP128 . CP127 . CP90 . Type: . ec . ec . ec . ec . ec . ec . ec . ec . ec . ec . SiO2 43.5 43.5 42.9 43.0 42.5 43.1 43.1 42.5 43.0 43.4 Al2O3 33.3 33.0 32.9 32.7 31.9 32.5 32.9 32.3 32.9 30.7 FeOt 1.5 1.5 1.7 1.6 1.3 2.1 1.3 1.6 1.4 2.4 Na2O 17.5 17.2 17.1 16.6 16.1 16.7 16.6 16.7 16.7 17.7 K2O 6.2 6.3 6.7 6.6 6.9 6.9 6.9 6.8 7.1 6.1 Total 102.1 101.5 101.3 100.7 99.3 101.6 100.9 100.5 101.0 100.7 Ks 18.7 19.1 20.5 20.4 21.6 21.1 21.3 21.2 21.7 18.9 Nf 2.7 2.6 3.1 2.8 2.5 3.8 2.4 2.9 2.5 4.3 Ne 77.5 76.9 76.5 74.5 73.9 73.8 75.0 75.6 75.2 78.6 An 2.5 1.0 0.0 1.0 1.3 1.0 0.0 1.8 0.0 2.1 Cn 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Qz 0.8 1.9 1.4 2.4 2.5 1.8 2.5 0.8 0.2 0.6 Sample: . CP45 . CP80 . CP51 . CP57 . CP71 . CP129 . CP126 . OL5 . OL5 . OL5 . Type: . ec . ec . ec . ec . ec . ec . ec . ph-c . ph-c sph. . microph-c . SiO2 43.6 42.8 44.0 41.7 41.3 43.6 43.5 43.5 43.3 43.1 Al2O3 33.0 33.0 32.0 32.2 30.7 31.7 33.1 32.9 33.0 31.2 FeOt 1.3 1.2 1.7 2.0 1.7 2.1 1.7 1.7 1.5 2.0 Na2O 16.9 16.7 16.0 16.1 16.9 16.8 16.6 16.8 17.0 15.6 K2O 6.3 7.0 6.4 7.2 6.8 7.0 7.4 6.8 6.5 7.2 Total 101.3 101.0 100.3 99.7 97.4 101.4 102.3 101.9 101.8 99.3 Ks 19.2 21.5 19.5 22.5 21.8 21.3 22.5 20.6 19.9 22.5 Nf 2.4 2.2 30.8 3.6 3.1 3.9 3.0 3.0 2.7 3.7 Ne 75.6 76.1 71.3 73.0 78.8 74.6 73.7 74.4 75.9 70.2 An 1.0 1.3 1.7 1.6 1.1 1.3 0.0 2.0 2.2 1.6 Cn 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Qz 3.0 0.9 5.1 0.8 0.0 1.8 2.1 1.7 1.3 3.7 Sample: . CP45 . CP80 . CP51 . CP57 . CP71 . CP129 . CP126 . OL5 . OL5 . OL5 . Type: . ec . ec . ec . ec . ec . ec . ec . ph-c . ph-c sph. . microph-c . SiO2 43.6 42.8 44.0 41.7 41.3 43.6 43.5 43.5 43.3 43.1 Al2O3 33.0 33.0 32.0 32.2 30.7 31.7 33.1 32.9 33.0 31.2 FeOt 1.3 1.2 1.7 2.0 1.7 2.1 1.7 1.7 1.5 2.0 Na2O 16.9 16.7 16.0 16.1 16.9 16.8 16.6 16.8 17.0 15.6 K2O 6.3 7.0 6.4 7.2 6.8 7.0 7.4 6.8 6.5 7.2 Total 101.3 101.0 100.3 99.7 97.4 101.4 102.3 101.9 101.8 99.3 Ks 19.2 21.5 19.5 22.5 21.8 21.3 22.5 20.6 19.9 22.5 Nf 2.4 2.2 30.8 3.6 3.1 3.9 3.0 3.0 2.7 3.7 Ne 75.6 76.1 71.3 73.0 78.8 74.6 73.7 74.4 75.9 70.2 An 1.0 1.3 1.7 1.6 1.1 1.3 0.0 2.0 2.2 1.6 Cn 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Qz 3.0 0.9 5.1 0.8 0.0 1.8 2.1 1.7 1.3 3.7 End-members determined by the method of Peterson (1989). ec, experimental charge; ph-c, phenocryst; ph-c sph., phenocryst in spheroid; microph-c, microphenocryst. Open in new tab Table 4: Representative microprobe analyses of nepheline from experimental charges and silicate-bearing natrocarbonatite OL5 Sample: . CP106 . CP79 . CP96 . CP98 . CP94 . CP118 . CP107 . CP128 . CP127 . CP90 . Type: . ec . ec . ec . ec . ec . ec . ec . ec . ec . ec . SiO2 43.5 43.5 42.9 43.0 42.5 43.1 43.1 42.5 43.0 43.4 Al2O3 33.3 33.0 32.9 32.7 31.9 32.5 32.9 32.3 32.9 30.7 FeOt 1.5 1.5 1.7 1.6 1.3 2.1 1.3 1.6 1.4 2.4 Na2O 17.5 17.2 17.1 16.6 16.1 16.7 16.6 16.7 16.7 17.7 K2O 6.2 6.3 6.7 6.6 6.9 6.9 6.9 6.8 7.1 6.1 Total 102.1 101.5 101.3 100.7 99.3 101.6 100.9 100.5 101.0 100.7 Ks 18.7 19.1 20.5 20.4 21.6 21.1 21.3 21.2 21.7 18.9 Nf 2.7 2.6 3.1 2.8 2.5 3.8 2.4 2.9 2.5 4.3 Ne 77.5 76.9 76.5 74.5 73.9 73.8 75.0 75.6 75.2 78.6 An 2.5 1.0 0.0 1.0 1.3 1.0 0.0 1.8 0.0 2.1 Cn 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Qz 0.8 1.9 1.4 2.4 2.5 1.8 2.5 0.8 0.2 0.6 Sample: . CP106 . CP79 . CP96 . CP98 . CP94 . CP118 . CP107 . CP128 . CP127 . CP90 . Type: . ec . ec . ec . ec . ec . ec . ec . ec . ec . ec . SiO2 43.5 43.5 42.9 43.0 42.5 43.1 43.1 42.5 43.0 43.4 Al2O3 33.3 33.0 32.9 32.7 31.9 32.5 32.9 32.3 32.9 30.7 FeOt 1.5 1.5 1.7 1.6 1.3 2.1 1.3 1.6 1.4 2.4 Na2O 17.5 17.2 17.1 16.6 16.1 16.7 16.6 16.7 16.7 17.7 K2O 6.2 6.3 6.7 6.6 6.9 6.9 6.9 6.8 7.1 6.1 Total 102.1 101.5 101.3 100.7 99.3 101.6 100.9 100.5 101.0 100.7 Ks 18.7 19.1 20.5 20.4 21.6 21.1 21.3 21.2 21.7 18.9 Nf 2.7 2.6 3.1 2.8 2.5 3.8 2.4 2.9 2.5 4.3 Ne 77.5 76.9 76.5 74.5 73.9 73.8 75.0 75.6 75.2 78.6 An 2.5 1.0 0.0 1.0 1.3 1.0 0.0 1.8 0.0 2.1 Cn 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Qz 0.8 1.9 1.4 2.4 2.5 1.8 2.5 0.8 0.2 0.6 Sample: . CP45 . CP80 . CP51 . CP57 . CP71 . CP129 . CP126 . OL5 . OL5 . OL5 . Type: . ec . ec . ec . ec . ec . ec . ec . ph-c . ph-c sph. . microph-c . SiO2 43.6 42.8 44.0 41.7 41.3 43.6 43.5 43.5 43.3 43.1 Al2O3 33.0 33.0 32.0 32.2 30.7 31.7 33.1 32.9 33.0 31.2 FeOt 1.3 1.2 1.7 2.0 1.7 2.1 1.7 1.7 1.5 2.0 Na2O 16.9 16.7 16.0 16.1 16.9 16.8 16.6 16.8 17.0 15.6 K2O 6.3 7.0 6.4 7.2 6.8 7.0 7.4 6.8 6.5 7.2 Total 101.3 101.0 100.3 99.7 97.4 101.4 102.3 101.9 101.8 99.3 Ks 19.2 21.5 19.5 22.5 21.8 21.3 22.5 20.6 19.9 22.5 Nf 2.4 2.2 30.8 3.6 3.1 3.9 3.0 3.0 2.7 3.7 Ne 75.6 76.1 71.3 73.0 78.8 74.6 73.7 74.4 75.9 70.2 An 1.0 1.3 1.7 1.6 1.1 1.3 0.0 2.0 2.2 1.6 Cn 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Qz 3.0 0.9 5.1 0.8 0.0 1.8 2.1 1.7 1.3 3.7 Sample: . CP45 . CP80 . CP51 . CP57 . CP71 . CP129 . CP126 . OL5 . OL5 . OL5 . Type: . ec . ec . ec . ec . ec . ec . ec . ph-c . ph-c sph. . microph-c . SiO2 43.6 42.8 44.0 41.7 41.3 43.6 43.5 43.5 43.3 43.1 Al2O3 33.0 33.0 32.0 32.2 30.7 31.7 33.1 32.9 33.0 31.2 FeOt 1.3 1.2 1.7 2.0 1.7 2.1 1.7 1.7 1.5 2.0 Na2O 16.9 16.7 16.0 16.1 16.9 16.8 16.6 16.8 17.0 15.6 K2O 6.3 7.0 6.4 7.2 6.8 7.0 7.4 6.8 6.5 7.2 Total 101.3 101.0 100.3 99.7 97.4 101.4 102.3 101.9 101.8 99.3 Ks 19.2 21.5 19.5 22.5 21.8 21.3 22.5 20.6 19.9 22.5 Nf 2.4 2.2 30.8 3.6 3.1 3.9 3.0 3.0 2.7 3.7 Ne 75.6 76.1 71.3 73.0 78.8 74.6 73.7 74.4 75.9 70.2 An 1.0 1.3 1.7 1.6 1.1 1.3 0.0 2.0 2.2 1.6 Cn 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Qz 3.0 0.9 5.1 0.8 0.0 1.8 2.1 1.7 1.3 3.7 End-members determined by the method of Peterson (1989). ec, experimental charge; ph-c, phenocryst; ph-c sph., phenocryst in spheroid; microph-c, microphenocryst. Open in new tab Melilite crystals from the experiments have a range of major element compositions, which overlap the composition of the single melilite phenocryst observed (with nepheline and melanite in the core of a spheroid) and analysed in OL5 (Table 7). The composition of this one grain is also very similar to the melilite phenocryst analyses presented by Dawson et al. (1996). Melilite phenocrysts from the lavas most closely resemble melilite from the 775 and 750°C experiments at 20 MPa (Fig. 7). Perovskite was encountered as a stable phase in only one experiment (100 MPa, 850°C). Nb, Zr, and rare earth elements were not analysed, hence the low total (Table 8). Discussion The phenocrysts present in silicate-bearing natrocarbonatite OL5 include nyerereite, gregoryite, clinopyroxene, wollastonite, nepheline, melanite garnet, melilite, apatite, and pyrrhotite. In the 100 MPa experiments, wollastonite, melilite and gregoryite did not precipitate and melanite garnet crystallized only over a 50°C degree interval. The phenocryst minerals noted in OL5, however, do occur as solid phases in the 20 MPa experiments. Furthermore, the best congruence of mineral compositions from lava OL5 is with solid phases from the 20 MPa experiments, not the 100 MPa experiments. Fig. 5. Open in new tabDownload slide Plot of nepheline analyses from experiments and silicate-bearing natrocarbonatite lavas onto part of the nepheline–kalsilite–quartz triangular plot (mol %). Source of data additional to this study: Dawson et al. (1996). The mineral assemblage of OL5 cannot be exactly duplicated at 20 MPa because of the absence of pyrrhotite in the experiments (suggested to be related to control of sulphur fugacity). However, we suggest that OL5 in fact records a silicate-bearing natrocarbonatite magma undergoing equilibrium cooling from 775 to 615°C at 20 MPa. The solitary (relict?) silicate spheroid that contains melilite, nepheline and melanite phenocrysts represents a temperature of 775°C at 20 MPa (Fig. 3). Phenocrysts of melanite, wollastonite and nepheline in OL5 are equivalent to the co-precipitating phase assemblage in 20 MPa experiments over the temperatures interval from 750 to 635°C. However, if the lone melilite-bearing spheroid is ignored (as a high-T relict), then the phase assemblage of OL5, i.e. nepheline, wollastonite, melanite, clinopyroxene, nyerereite, apatite and gregoryite, is exactly replicated at 615°C and 20 MPa (Fig. 3). Thus we believe that OL5 preserves a complete record of its equilibrium cooling history from 775 to 615°C at 20 MPa. Does OL5 represent a silicate-bearing natrocarbonatite parent magma? OL5 was obtained from the Chaos Crags flow erupted in June 1993. The lavas from this flow are very thick and viscous, and contain ∼3 wt % SiO2. This silica is mainly represented by silicate phenocrysts (both isolated and in the spheroids), and also by the siliceous mesostasis of the spheroids. Nepheline, melanite, clinopyroxene, wollastonite and melilite crystals in OL5 are euhedral and not corroded, suggesting they are equilibrium with the surrounding natrocarbonatite melt. Melilite from OL5 (and other silicate-bearing natrocarbonatites; e.g. Dawson et al., 1996) differ in composition from those from the 1966 eruption (Dawson et al., 1992; see Fig. 8, below). The very explosive 1966 eruption has been shown to be the product of physical mixing between natrocarbonatite and silicate magmas (Dawson et al., 1992) and pyroclastic deposits contain up to 25 wt % SiO2. Silicate minerals in the pyroclastic rocks from the 1966 eruption show disequilibrium textures with the hostnatrocarbonatite, e.g. corroded crystals of wollastonite and clinopyroxene are surrounded by coronas containing combeite, melilite and Ca-silicates (Dawson et al., 1992). Table 5: Representative microprobe analyses of clinopyroxene from experimental charges and silicate-bearing natrocarbonatite OL5 Sample: . CP94 . CP118 . CP107 . CP128 . CP127 . CP129 . CP108 . CP126 . CP112 . OL5 . OL5 . OL5 . Type: . ec . ec . ec . ec . ec . ec . ec . ec . ec . ph-c . ph-c sph. . microph-c . SiO2 51.4 51.0 52.4 50.9 53.5 52.6 52.0 49.3 51.3 50.4 52.1 46.7 TiO2 0.7 0.6 0.4 0.5 0.1 0.2 0.5 1.0 0.6 1.2 0.6 3.1 Al2O3 1.3 1.4 0.5 0.5 b.d. 0.6 0.8 0.5 1.1 2.2 1.1 5.0 FeO 7.6 10.4 8.8 9.9 3.1 8.3 8.4 6.1 9.4 8.3 7.5 5.5 Fe2O3 6.1 7.8 6.6 8.7 9.5 7.8 7.5 16.3 6.6 4.1 6.6 8.1 MnO 0.4 0.5 0.5 0.5 1.8 0.4 0.5 3.0 0.6 0.4 0.4 0.4 MgO 9.7 7.7 9.1 7.1 10.3 8.9 8.7 3.2 8.7 10.4 9.8 9.5 CaO 21.8 20.6 20.2 19.5 20.1 19.6 20.0 15.2 21.3 22.5 21.9 22.2 Na2O 1.9 2.3 2.6 3.0 3.3 3.0 2.8 6.0 2.0 1.2 2.0 1.7 Total 100.9 102.4 101.1 100.6 101.7 101.3 101.1 100.6 101.6 100.7 102.0 102.2 Mg 58.2 45.6 51.2 42.2 59.3 50.3 50.4 19.7 51.5 61.3 57.8 63.6 Fe2+ + Mn 26.8 36.5 29.4 34.9 16.1 27.8 28.8 31.6 33.0 29.3 26.4 21.7 Na 15.0 17.9 19.4 22.9 24.6 21.9 20.9 48.7 15.5 9.4 15.8 14.7 Sample: . CP94 . CP118 . CP107 . CP128 . CP127 . CP129 . CP108 . CP126 . CP112 . OL5 . OL5 . OL5 . Type: . ec . ec . ec . ec . ec . ec . ec . ec . ec . ph-c . ph-c sph. . microph-c . SiO2 51.4 51.0 52.4 50.9 53.5 52.6 52.0 49.3 51.3 50.4 52.1 46.7 TiO2 0.7 0.6 0.4 0.5 0.1 0.2 0.5 1.0 0.6 1.2 0.6 3.1 Al2O3 1.3 1.4 0.5 0.5 b.d. 0.6 0.8 0.5 1.1 2.2 1.1 5.0 FeO 7.6 10.4 8.8 9.9 3.1 8.3 8.4 6.1 9.4 8.3 7.5 5.5 Fe2O3 6.1 7.8 6.6 8.7 9.5 7.8 7.5 16.3 6.6 4.1 6.6 8.1 MnO 0.4 0.5 0.5 0.5 1.8 0.4 0.5 3.0 0.6 0.4 0.4 0.4 MgO 9.7 7.7 9.1 7.1 10.3 8.9 8.7 3.2 8.7 10.4 9.8 9.5 CaO 21.8 20.6 20.2 19.5 20.1 19.6 20.0 15.2 21.3 22.5 21.9 22.2 Na2O 1.9 2.3 2.6 3.0 3.3 3.0 2.8 6.0 2.0 1.2 2.0 1.7 Total 100.9 102.4 101.1 100.6 101.7 101.3 101.1 100.6 101.6 100.7 102.0 102.2 Mg 58.2 45.6 51.2 42.2 59.3 50.3 50.4 19.7 51.5 61.3 57.8 63.6 Fe2+ + Mn 26.8 36.5 29.4 34.9 16.1 27.8 28.8 31.6 33.0 29.3 26.4 21.7 Na 15.0 17.9 19.4 22.9 24.6 21.9 20.9 48.7 15.5 9.4 15.8 14.7 FeO and Fe2O3 determined by stoichiometry utilizing the technique of Droop (1987). ec, experimental charge; ph-c, phenocryst; ph-c sph., phenocryst in spheroid; microph-c, microphenocryst. Open in new tab Table 5: Representative microprobe analyses of clinopyroxene from experimental charges and silicate-bearing natrocarbonatite OL5 Sample: . CP94 . CP118 . CP107 . CP128 . CP127 . CP129 . CP108 . CP126 . CP112 . OL5 . OL5 . OL5 . Type: . ec . ec . ec . ec . ec . ec . ec . ec . ec . ph-c . ph-c sph. . microph-c . SiO2 51.4 51.0 52.4 50.9 53.5 52.6 52.0 49.3 51.3 50.4 52.1 46.7 TiO2 0.7 0.6 0.4 0.5 0.1 0.2 0.5 1.0 0.6 1.2 0.6 3.1 Al2O3 1.3 1.4 0.5 0.5 b.d. 0.6 0.8 0.5 1.1 2.2 1.1 5.0 FeO 7.6 10.4 8.8 9.9 3.1 8.3 8.4 6.1 9.4 8.3 7.5 5.5 Fe2O3 6.1 7.8 6.6 8.7 9.5 7.8 7.5 16.3 6.6 4.1 6.6 8.1 MnO 0.4 0.5 0.5 0.5 1.8 0.4 0.5 3.0 0.6 0.4 0.4 0.4 MgO 9.7 7.7 9.1 7.1 10.3 8.9 8.7 3.2 8.7 10.4 9.8 9.5 CaO 21.8 20.6 20.2 19.5 20.1 19.6 20.0 15.2 21.3 22.5 21.9 22.2 Na2O 1.9 2.3 2.6 3.0 3.3 3.0 2.8 6.0 2.0 1.2 2.0 1.7 Total 100.9 102.4 101.1 100.6 101.7 101.3 101.1 100.6 101.6 100.7 102.0 102.2 Mg 58.2 45.6 51.2 42.2 59.3 50.3 50.4 19.7 51.5 61.3 57.8 63.6 Fe2+ + Mn 26.8 36.5 29.4 34.9 16.1 27.8 28.8 31.6 33.0 29.3 26.4 21.7 Na 15.0 17.9 19.4 22.9 24.6 21.9 20.9 48.7 15.5 9.4 15.8 14.7 Sample: . CP94 . CP118 . CP107 . CP128 . CP127 . CP129 . CP108 . CP126 . CP112 . OL5 . OL5 . OL5 . Type: . ec . ec . ec . ec . ec . ec . ec . ec . ec . ph-c . ph-c sph. . microph-c . SiO2 51.4 51.0 52.4 50.9 53.5 52.6 52.0 49.3 51.3 50.4 52.1 46.7 TiO2 0.7 0.6 0.4 0.5 0.1 0.2 0.5 1.0 0.6 1.2 0.6 3.1 Al2O3 1.3 1.4 0.5 0.5 b.d. 0.6 0.8 0.5 1.1 2.2 1.1 5.0 FeO 7.6 10.4 8.8 9.9 3.1 8.3 8.4 6.1 9.4 8.3 7.5 5.5 Fe2O3 6.1 7.8 6.6 8.7 9.5 7.8 7.5 16.3 6.6 4.1 6.6 8.1 MnO 0.4 0.5 0.5 0.5 1.8 0.4 0.5 3.0 0.6 0.4 0.4 0.4 MgO 9.7 7.7 9.1 7.1 10.3 8.9 8.7 3.2 8.7 10.4 9.8 9.5 CaO 21.8 20.6 20.2 19.5 20.1 19.6 20.0 15.2 21.3 22.5 21.9 22.2 Na2O 1.9 2.3 2.6 3.0 3.3 3.0 2.8 6.0 2.0 1.2 2.0 1.7 Total 100.9 102.4 101.1 100.6 101.7 101.3 101.1 100.6 101.6 100.7 102.0 102.2 Mg 58.2 45.6 51.2 42.2 59.3 50.3 50.4 19.7 51.5 61.3 57.8 63.6 Fe2+ + Mn 26.8 36.5 29.4 34.9 16.1 27.8 28.8 31.6 33.0 29.3 26.4 21.7 Na 15.0 17.9 19.4 22.9 24.6 21.9 20.9 48.7 15.5 9.4 15.8 14.7 FeO and Fe2O3 determined by stoichiometry utilizing the technique of Droop (1987). ec, experimental charge; ph-c, phenocryst; ph-c sph., phenocryst in spheroid; microph-c, microphenocryst. Open in new tab Table 6: Representative microprobe analyses of melanite garnet from experimental charges and silicate-bearing natrocarbonatite OL5 Sample: . CP94 . CP80 . CP51 . CP57 . CP71 . CP129 . CP108 . CP126 . CP112 . OL5 . OL5 . OL5 . Type: . ec . ec . ec . ec . ec . ec . ec . ec . ec . ph . ph-c sph. . microph-c . SiO2 30.0 29.1 29.7 29.1 29.4 32.0 29.7 30.7 31.1 29.2 30.2 31.2 TiO2 13.3 10.1 13.5 12.8 13.6 8.7 13.3 11.0 11.7 13.1 12.8 10.5 Al2O3 1.1 0.7 0.4 0.8 0.5 0.6 1.2 0.7 0.6 0.7 0.9 0.7 Fe2O3 23.9 26.8 23.7 24.1 24.1 26.4 23.5 24.9 25.6 23.7 24.2 24.8 MnO 0.4 0.3 0.3 0.3 b.d. b.d. 0.4 0.4 b.d. 0.4 0.4 0.5 MgO 0.7 0.5 0.5 0.6 0.7 0.6 0.6 0.5 0.4 0.6 0.6 0.3 CaO 31.8 31.3 30.8 32.3 31.7 32.0 31.6 31.6 31.8 31.0 31.7 31.6 Na2O b.d. b.d. 0.7 0.4 0.5 0.5 0.7 0.3 0.7 0.7 0.8 0.7 Total 101.2 98.8 99.5 100.2 100.5 100.8 101.0 100.1 101.8 99.4 101.6 100.3 Fe3+ 61.2 70.4 62.7 63.1 62.6 73.0 60.5 67.1 66.8 62.4 62.9 68.0 Al 4.8 3.1 1.8 3.3 2.2 3.0 5.3 3.2 2.7 3.1 4.0 3.3 Ti 34.1 26.5 35.6 33.6 35.3 24.0 34.3 29.7 30.5 34.5 33.2 28.8 Sample: . CP94 . CP80 . CP51 . CP57 . CP71 . CP129 . CP108 . CP126 . CP112 . OL5 . OL5 . OL5 . Type: . ec . ec . ec . ec . ec . ec . ec . ec . ec . ph . ph-c sph. . microph-c . SiO2 30.0 29.1 29.7 29.1 29.4 32.0 29.7 30.7 31.1 29.2 30.2 31.2 TiO2 13.3 10.1 13.5 12.8 13.6 8.7 13.3 11.0 11.7 13.1 12.8 10.5 Al2O3 1.1 0.7 0.4 0.8 0.5 0.6 1.2 0.7 0.6 0.7 0.9 0.7 Fe2O3 23.9 26.8 23.7 24.1 24.1 26.4 23.5 24.9 25.6 23.7 24.2 24.8 MnO 0.4 0.3 0.3 0.3 b.d. b.d. 0.4 0.4 b.d. 0.4 0.4 0.5 MgO 0.7 0.5 0.5 0.6 0.7 0.6 0.6 0.5 0.4 0.6 0.6 0.3 CaO 31.8 31.3 30.8 32.3 31.7 32.0 31.6 31.6 31.8 31.0 31.7 31.6 Na2O b.d. b.d. 0.7 0.4 0.5 0.5 0.7 0.3 0.7 0.7 0.8 0.7 Total 101.2 98.8 99.5 100.2 100.5 100.8 101.0 100.1 101.8 99.4 101.6 100.3 Fe3+ 61.2 70.4 62.7 63.1 62.6 73.0 60.5 67.1 66.8 62.4 62.9 68.0 Al 4.8 3.1 1.8 3.3 2.2 3.0 5.3 3.2 2.7 3.1 4.0 3.3 Ti 34.1 26.5 35.6 33.6 35.3 24.0 34.3 29.7 30.5 34.5 33.2 28.8 ec, experimental charge; ph-c, phenocryst; ph-c sph., phenocryst in spheroid; microph-c, microphenocryst. Open in new tab Table 6: Representative microprobe analyses of melanite garnet from experimental charges and silicate-bearing natrocarbonatite OL5 Sample: . CP94 . CP80 . CP51 . CP57 . CP71 . CP129 . CP108 . CP126 . CP112 . OL5 . OL5 . OL5 . Type: . ec . ec . ec . ec . ec . ec . ec . ec . ec . ph . ph-c sph. . microph-c . SiO2 30.0 29.1 29.7 29.1 29.4 32.0 29.7 30.7 31.1 29.2 30.2 31.2 TiO2 13.3 10.1 13.5 12.8 13.6 8.7 13.3 11.0 11.7 13.1 12.8 10.5 Al2O3 1.1 0.7 0.4 0.8 0.5 0.6 1.2 0.7 0.6 0.7 0.9 0.7 Fe2O3 23.9 26.8 23.7 24.1 24.1 26.4 23.5 24.9 25.6 23.7 24.2 24.8 MnO 0.4 0.3 0.3 0.3 b.d. b.d. 0.4 0.4 b.d. 0.4 0.4 0.5 MgO 0.7 0.5 0.5 0.6 0.7 0.6 0.6 0.5 0.4 0.6 0.6 0.3 CaO 31.8 31.3 30.8 32.3 31.7 32.0 31.6 31.6 31.8 31.0 31.7 31.6 Na2O b.d. b.d. 0.7 0.4 0.5 0.5 0.7 0.3 0.7 0.7 0.8 0.7 Total 101.2 98.8 99.5 100.2 100.5 100.8 101.0 100.1 101.8 99.4 101.6 100.3 Fe3+ 61.2 70.4 62.7 63.1 62.6 73.0 60.5 67.1 66.8 62.4 62.9 68.0 Al 4.8 3.1 1.8 3.3 2.2 3.0 5.3 3.2 2.7 3.1 4.0 3.3 Ti 34.1 26.5 35.6 33.6 35.3 24.0 34.3 29.7 30.5 34.5 33.2 28.8 Sample: . CP94 . CP80 . CP51 . CP57 . CP71 . CP129 . CP108 . CP126 . CP112 . OL5 . OL5 . OL5 . Type: . ec . ec . ec . ec . ec . ec . ec . ec . ec . ph . ph-c sph. . microph-c . SiO2 30.0 29.1 29.7 29.1 29.4 32.0 29.7 30.7 31.1 29.2 30.2 31.2 TiO2 13.3 10.1 13.5 12.8 13.6 8.7 13.3 11.0 11.7 13.1 12.8 10.5 Al2O3 1.1 0.7 0.4 0.8 0.5 0.6 1.2 0.7 0.6 0.7 0.9 0.7 Fe2O3 23.9 26.8 23.7 24.1 24.1 26.4 23.5 24.9 25.6 23.7 24.2 24.8 MnO 0.4 0.3 0.3 0.3 b.d. b.d. 0.4 0.4 b.d. 0.4 0.4 0.5 MgO 0.7 0.5 0.5 0.6 0.7 0.6 0.6 0.5 0.4 0.6 0.6 0.3 CaO 31.8 31.3 30.8 32.3 31.7 32.0 31.6 31.6 31.8 31.0 31.7 31.6 Na2O b.d. b.d. 0.7 0.4 0.5 0.5 0.7 0.3 0.7 0.7 0.8 0.7 Total 101.2 98.8 99.5 100.2 100.5 100.8 101.0 100.1 101.8 99.4 101.6 100.3 Fe3+ 61.2 70.4 62.7 63.1 62.6 73.0 60.5 67.1 66.8 62.4 62.9 68.0 Al 4.8 3.1 1.8 3.3 2.2 3.0 5.3 3.2 2.7 3.1 4.0 3.3 Ti 34.1 26.5 35.6 33.6 35.3 24.0 34.3 29.7 30.5 34.5 33.2 28.8 ec, experimental charge; ph-c, phenocryst; ph-c sph., phenocryst in spheroid; microph-c, microphenocryst. Open in new tab Fig. 6. Open in new tabDownload slide Plot of melanite garnet analyses from experiments and natural lavas from Oldoinyo Lengai onto an Al–Fe3+–Ti triangular plot (atomic %). Source of data additional to this study: Dawson et al. (1996). Table 7: Representative microprobe analyses of melilite from experimental charges and silicate-bearing natrocarbonatite OL5 Sample: . CP45 . CP61 . CP80 . OL5 . Type: . ec . ec . ec . ph-c . SiO2 43.7 42.5 43.9 43.4 TiO2 b.d. 0.0 0.0 b.d. Al2O3 7.5 6.6 7.3 7.9 FeO 4.0 1.9 2.9 4.3 Fe2O3 2.8 5.6 4.0 3.3 MnO 0.7 0.9 0.9 b.d. MgO 4.7 4.3 4.4 4.3 CaO 30.7 30.2 29.8 30.7 Na2O 5.1 5.5 5.8 5.7 K2O 0.2 0.3 0.1 b.d. SrO 2.6 2.4 2.3 n.d. Total 102.0 100.1 101.6 99.6 Na 36.1 46.6 48.5 47.3 Fe 35.3 25.5 23.3 25.4 Mg 28.7 27.9 28.2 27.8 Sample: . CP45 . CP61 . CP80 . OL5 . Type: . ec . ec . ec . ph-c . SiO2 43.7 42.5 43.9 43.4 TiO2 b.d. 0.0 0.0 b.d. Al2O3 7.5 6.6 7.3 7.9 FeO 4.0 1.9 2.9 4.3 Fe2O3 2.8 5.6 4.0 3.3 MnO 0.7 0.9 0.9 b.d. MgO 4.7 4.3 4.4 4.3 CaO 30.7 30.2 29.8 30.7 Na2O 5.1 5.5 5.8 5.7 K2O 0.2 0.3 0.1 b.d. SrO 2.6 2.4 2.3 n.d. Total 102.0 100.1 101.6 99.6 Na 36.1 46.6 48.5 47.3 Fe 35.3 25.5 23.3 25.4 Mg 28.7 27.9 28.2 27.8 ec, experimental charge; ph-c, phenocryst. Open in new tab Table 7: Representative microprobe analyses of melilite from experimental charges and silicate-bearing natrocarbonatite OL5 Sample: . CP45 . CP61 . CP80 . OL5 . Type: . ec . ec . ec . ph-c . SiO2 43.7 42.5 43.9 43.4 TiO2 b.d. 0.0 0.0 b.d. Al2O3 7.5 6.6 7.3 7.9 FeO 4.0 1.9 2.9 4.3 Fe2O3 2.8 5.6 4.0 3.3 MnO 0.7 0.9 0.9 b.d. MgO 4.7 4.3 4.4 4.3 CaO 30.7 30.2 29.8 30.7 Na2O 5.1 5.5 5.8 5.7 K2O 0.2 0.3 0.1 b.d. SrO 2.6 2.4 2.3 n.d. Total 102.0 100.1 101.6 99.6 Na 36.1 46.6 48.5 47.3 Fe 35.3 25.5 23.3 25.4 Mg 28.7 27.9 28.2 27.8 Sample: . CP45 . CP61 . CP80 . OL5 . Type: . ec . ec . ec . ph-c . SiO2 43.7 42.5 43.9 43.4 TiO2 b.d. 0.0 0.0 b.d. Al2O3 7.5 6.6 7.3 7.9 FeO 4.0 1.9 2.9 4.3 Fe2O3 2.8 5.6 4.0 3.3 MnO 0.7 0.9 0.9 b.d. MgO 4.7 4.3 4.4 4.3 CaO 30.7 30.2 29.8 30.7 Na2O 5.1 5.5 5.8 5.7 K2O 0.2 0.3 0.1 b.d. SrO 2.6 2.4 2.3 n.d. Total 102.0 100.1 101.6 99.6 Na 36.1 46.6 48.5 47.3 Fe 35.3 25.5 23.3 25.4 Mg 28.7 27.9 28.2 27.8 ec, experimental charge; ph-c, phenocryst. Open in new tab Table 8: Representative microprobe analyses of miscellaneous minerals from experimental charges and silicate-bearing natrocarbonatite OL5 Phase: . Wo . Pvk . Fe-sp . Fe-sp . K-Fd . Ap . Ap . Sod . Wo . Wo . Ap . Py . Sample: . CP129 . CP45 . CP61 . CP80 . CP98 . CP107 . CP112 . CP112 . OL5 . OL5 . OL5 . OL5 . Type: . ec . ec . ec . ec . ec . ec . ec . ec . ph-c . ph-c sph. . ph-c . ph-c . SiO2 50.10 2.1 0.2 0.4 64.7 0.9 0.7 37.8 51.9 52.2 1.7 0.7 TiO2 b.d. 39.8 2.5 4.0 b.d. b.d. b.d. 0.5 0.1 0.1 b.d. b.d. Al2O3 b.d. 0.6 0.3 b.d. 18.1 b.d. b.d. 28.0 b.d. b.d. 0.2 b.d. FeOt b.d. 5.3 80.1 86.2 0.7 b.d. b.d. 2.9 0.9 1.0 b.d. 44.9* MnO 0.80 b.d. 3.6 3.0 b.d. b.d. b.d. b.d. 0.4 0.4 b.d. 2.9 MgO b.d. 0.3 0.7 0.8 b.d. b.d. b.d. b.d. 0.2 0.2 b.d. b.d. CaO 46.10 28.0 0.4 0.2 b.d. 54.0 53.3 5.7 46.9 47.1 51.7 0.2 Na2O 0.10 1.6 b.d. 0.5 1.6 0.3 b.d. 21.6 0.3 b.d. 0.4 3.8 K2O 0.70 0.2 b.d. b.d. 15.0 0.1 b.d. 2.4 0.1 b.d. 0.1 11.3 P2O5 b.d. b.d. 0.3 b.d. b.d. 42.1 42.4 b.d. b.d. b.d. 42.3 b.d. F b.d. 0.2 b.d. b.d. b.d. 3.0 4.0 b.d. b.d. b.d. 2.6 b.d. Cl 0.40 0.2 b.d. b.d. b.d. b.d. b.d. 5.8 b.d. b.d. b.d. b.d. SO3 b.d. b.d. b.d. b.d. b.d. b.d. b.d. 0.7 b.d. b.d. b.d. 34.3* BaO b.d. b.d. b.d. b.d. 0.9 b.d. b.d. b.d. b.d. b.d. b.d. b.d. SrO b.d. 2.0 0.3 b.d. b.d. b.d. 0.7 b.d. b.d. b.d. b.d. b.d. Total 99.1 80.2 88.5 95.0 101.0 100.6 101.1 101.4 100.5 100.8 99.0 98.1 Phase: . Wo . Pvk . Fe-sp . Fe-sp . K-Fd . Ap . Ap . Sod . Wo . Wo . Ap . Py . Sample: . CP129 . CP45 . CP61 . CP80 . CP98 . CP107 . CP112 . CP112 . OL5 . OL5 . OL5 . OL5 . Type: . ec . ec . ec . ec . ec . ec . ec . ec . ph-c . ph-c sph. . ph-c . ph-c . SiO2 50.10 2.1 0.2 0.4 64.7 0.9 0.7 37.8 51.9 52.2 1.7 0.7 TiO2 b.d. 39.8 2.5 4.0 b.d. b.d. b.d. 0.5 0.1 0.1 b.d. b.d. Al2O3 b.d. 0.6 0.3 b.d. 18.1 b.d. b.d. 28.0 b.d. b.d. 0.2 b.d. FeOt b.d. 5.3 80.1 86.2 0.7 b.d. b.d. 2.9 0.9 1.0 b.d. 44.9* MnO 0.80 b.d. 3.6 3.0 b.d. b.d. b.d. b.d. 0.4 0.4 b.d. 2.9 MgO b.d. 0.3 0.7 0.8 b.d. b.d. b.d. b.d. 0.2 0.2 b.d. b.d. CaO 46.10 28.0 0.4 0.2 b.d. 54.0 53.3 5.7 46.9 47.1 51.7 0.2 Na2O 0.10 1.6 b.d. 0.5 1.6 0.3 b.d. 21.6 0.3 b.d. 0.4 3.8 K2O 0.70 0.2 b.d. b.d. 15.0 0.1 b.d. 2.4 0.1 b.d. 0.1 11.3 P2O5 b.d. b.d. 0.3 b.d. b.d. 42.1 42.4 b.d. b.d. b.d. 42.3 b.d. F b.d. 0.2 b.d. b.d. b.d. 3.0 4.0 b.d. b.d. b.d. 2.6 b.d. Cl 0.40 0.2 b.d. b.d. b.d. b.d. b.d. 5.8 b.d. b.d. b.d. b.d. SO3 b.d. b.d. b.d. b.d. b.d. b.d. b.d. 0.7 b.d. b.d. b.d. 34.3* BaO b.d. b.d. b.d. b.d. 0.9 b.d. b.d. b.d. b.d. b.d. b.d. b.d. SrO b.d. 2.0 0.3 b.d. b.d. b.d. 0.7 b.d. b.d. b.d. b.d. b.d. Total 99.1 80.2 88.5 95.0 101.0 100.6 101.1 101.4 100.5 100.8 99.0 98.1 ec, experimental charge; ph-c, phenocryst; ph-c sph., phenocryst in spheroid; microph-c, microphenocryst. Wo, wollastonite; Pvk, perovskite; Fe-sp, Fe-spinel; K-Fd, potassium feldspar; Ap, apatite; Sod, sodalite; Py, pyrrhotite. * SO3 recalculated to S and FeO recalculated to Fe for pyrrhotite. Open in new tab Table 8: Representative microprobe analyses of miscellaneous minerals from experimental charges and silicate-bearing natrocarbonatite OL5 Phase: . Wo . Pvk . Fe-sp . Fe-sp . K-Fd . Ap . Ap . Sod . Wo . Wo . Ap . Py . Sample: . CP129 . CP45 . CP61 . CP80 . CP98 . CP107 . CP112 . CP112 . OL5 . OL5 . OL5 . OL5 . Type: . ec . ec . ec . ec . ec . ec . ec . ec . ph-c . ph-c sph. . ph-c . ph-c . SiO2 50.10 2.1 0.2 0.4 64.7 0.9 0.7 37.8 51.9 52.2 1.7 0.7 TiO2 b.d. 39.8 2.5 4.0 b.d. b.d. b.d. 0.5 0.1 0.1 b.d. b.d. Al2O3 b.d. 0.6 0.3 b.d. 18.1 b.d. b.d. 28.0 b.d. b.d. 0.2 b.d. FeOt b.d. 5.3 80.1 86.2 0.7 b.d. b.d. 2.9 0.9 1.0 b.d. 44.9* MnO 0.80 b.d. 3.6 3.0 b.d. b.d. b.d. b.d. 0.4 0.4 b.d. 2.9 MgO b.d. 0.3 0.7 0.8 b.d. b.d. b.d. b.d. 0.2 0.2 b.d. b.d. CaO 46.10 28.0 0.4 0.2 b.d. 54.0 53.3 5.7 46.9 47.1 51.7 0.2 Na2O 0.10 1.6 b.d. 0.5 1.6 0.3 b.d. 21.6 0.3 b.d. 0.4 3.8 K2O 0.70 0.2 b.d. b.d. 15.0 0.1 b.d. 2.4 0.1 b.d. 0.1 11.3 P2O5 b.d. b.d. 0.3 b.d. b.d. 42.1 42.4 b.d. b.d. b.d. 42.3 b.d. F b.d. 0.2 b.d. b.d. b.d. 3.0 4.0 b.d. b.d. b.d. 2.6 b.d. Cl 0.40 0.2 b.d. b.d. b.d. b.d. b.d. 5.8 b.d. b.d. b.d. b.d. SO3 b.d. b.d. b.d. b.d. b.d. b.d. b.d. 0.7 b.d. b.d. b.d. 34.3* BaO b.d. b.d. b.d. b.d. 0.9 b.d. b.d. b.d. b.d. b.d. b.d. b.d. SrO b.d. 2.0 0.3 b.d. b.d. b.d. 0.7 b.d. b.d. b.d. b.d. b.d. Total 99.1 80.2 88.5 95.0 101.0 100.6 101.1 101.4 100.5 100.8 99.0 98.1 Phase: . Wo . Pvk . Fe-sp . Fe-sp . K-Fd . Ap . Ap . Sod . Wo . Wo . Ap . Py . Sample: . CP129 . CP45 . CP61 . CP80 . CP98 . CP107 . CP112 . CP112 . OL5 . OL5 . OL5 . OL5 . Type: . ec . ec . ec . ec . ec . ec . ec . ec . ph-c . ph-c sph. . ph-c . ph-c . SiO2 50.10 2.1 0.2 0.4 64.7 0.9 0.7 37.8 51.9 52.2 1.7 0.7 TiO2 b.d. 39.8 2.5 4.0 b.d. b.d. b.d. 0.5 0.1 0.1 b.d. b.d. Al2O3 b.d. 0.6 0.3 b.d. 18.1 b.d. b.d. 28.0 b.d. b.d. 0.2 b.d. FeOt b.d. 5.3 80.1 86.2 0.7 b.d. b.d. 2.9 0.9 1.0 b.d. 44.9* MnO 0.80 b.d. 3.6 3.0 b.d. b.d. b.d. b.d. 0.4 0.4 b.d. 2.9 MgO b.d. 0.3 0.7 0.8 b.d. b.d. b.d. b.d. 0.2 0.2 b.d. b.d. CaO 46.10 28.0 0.4 0.2 b.d. 54.0 53.3 5.7 46.9 47.1 51.7 0.2 Na2O 0.10 1.6 b.d. 0.5 1.6 0.3 b.d. 21.6 0.3 b.d. 0.4 3.8 K2O 0.70 0.2 b.d. b.d. 15.0 0.1 b.d. 2.4 0.1 b.d. 0.1 11.3 P2O5 b.d. b.d. 0.3 b.d. b.d. 42.1 42.4 b.d. b.d. b.d. 42.3 b.d. F b.d. 0.2 b.d. b.d. b.d. 3.0 4.0 b.d. b.d. b.d. 2.6 b.d. Cl 0.40 0.2 b.d. b.d. b.d. b.d. b.d. 5.8 b.d. b.d. b.d. b.d. SO3 b.d. b.d. b.d. b.d. b.d. b.d. b.d. 0.7 b.d. b.d. b.d. 34.3* BaO b.d. b.d. b.d. b.d. 0.9 b.d. b.d. b.d. b.d. b.d. b.d. b.d. SrO b.d. 2.0 0.3 b.d. b.d. b.d. 0.7 b.d. b.d. b.d. b.d. b.d. Total 99.1 80.2 88.5 95.0 101.0 100.6 101.1 101.4 100.5 100.8 99.0 98.1 ec, experimental charge; ph-c, phenocryst; ph-c sph., phenocryst in spheroid; microph-c, microphenocryst. Wo, wollastonite; Pvk, perovskite; Fe-sp, Fe-spinel; K-Fd, potassium feldspar; Ap, apatite; Sod, sodalite; Py, pyrrhotite. * SO3 recalculated to S and FeO recalculated to Fe for pyrrhotite. Open in new tab Fig. 7. Open in new tabDownload slide Plot of melilite analyses from experiments and natural lavas in terms of Fe (Fe-åkermanite), Mg (åkermanite) and Na (soda-melilite) end members. ◇, 20 MPa experiments, with 850–750°C down-temperature trend as indicated in figure. +, melilite phenocryst in OL5. Sources of additional melilite mineral data from Donaldson & Dawson (1978), Dawson et al. (1989, 1996) and Keller & Krafft (1990). Textural observations by Dawson et al. (1996) on silicate-bearing natrocarbonatites suggested that the matrix of the silicate spheroids was liquid or plastic when they were in contact with the carbonatitic melt, i.e. the spheroids are not accidental clasts (‘xenoliths’) eroded from the vent walls. In OL5, the major element composition of the phenocrysts in the spheroids is not significantly different from that of isolated phenocrysts in the natrocarbonatite. These observations suggest that both types of phenocrysts are in equilibrium with natrocarbonatite melt, and are not xenoliths. This is consistent with the observation by Bell & Simonetti (1996) that the spheroids have similar Sr and Nd isotopic signatures to their natrocarbonatite host. As the isolated phenocrysts and the phenocryst assemblages of the silicate spheroids are here considered to be stable crystallizing phases in a silicate-bearing natrocarbonatite magma, this in turn indicates that the SiO2 content of OL5 (3.07 wt %) is probably the original, or close to the original level of SiO2 in the parent magma. Hence we suggest that SiO2 contents of 2–5 wt % in silicate-bearing natrocarbonatite are best interpreted as a primary feature of parental natrocarbonatite magma. Origin of the silicate spheroids The silicate spheroids in silicate-bearing natrocarbonatite have been previously interpreted as immiscible silicate liquids (Church & Jones, 1995), and as the result of mixing between natrocarbonatite and wollastonite nephelinite (Dawson et al., 1996). Phase relationships presented in this paper indicate that the silicate-bearing natrocarbonatite OL5 starting composition does not intersect the two-liquid field. Absence of an immiscibility relationship is consistent with previous observations that the silicate–carbonate two-liquid field decreases in width with decreasing pressure (Koster van Groos & Wyllie, 1966; C. M. Petibon, unpublished data, 1998). We therefore concur with Dawson et al. (1996) that it is difficult to reconcile an origin for the spheroids by immiscibility (e.g. Church & Jones, 1995). The experimental observation that silicate-bearingnatrocarbonatite parental magma has nepheline appear first on the liquidus, with an extensive interval of silicate mineral co-precipitation, removes the requirement of adding silicate phases from a wollastonite nephelinite to a silicate-free natrocarbonatite (Dawson et al., 1996). The silicate spheroids are suggested to form from the agglomeration of silicate crystals that precipitated in equilibrium with natrocarbonatite magma. The extremely homogeneous compositions of the silicate phenocysts in OL5 suggest this occurred over a restricted temperature interval. We agree with the suggestion of Dawson et al. (1996) that mixing can be important in the formation of some spheroid-bearing natrocarbonatites, but suggest that the mixing occurs between batches of silicate-bearing natrocarbonatite magmas and their silicate-rich cumulates and/or residual liquids. In some silicate spheroids, there is evidence of disequilibrium (corroded crystals) between silicate minerals and host natrocarbonate magma (Dawson et al., 1992), as well as zoned silicate phenocrysts (Dawson et al., 1994), which could be attributed to mixing processes. Formation of natrocarbonatites—magma chamber processes Experimental studies at appropriate P–T–X (Kjarsgaard et al., 1995) illustrated that natrocarbonatite melts generated by liquid immiscibility from nephelinitic hosts contained significant SiO2 (1.26–5.15 wt %; see also Fig. 1) in solution. This suggests that silicate-free natrocarbonatites (<0.5 wt % SiO2) cannot be directly generated by immiscibility at relevant P–T–X, but must be differentiates from silicate-bearing natrocarbonatite parent magmas. Thus natrocarbonatite magma chamber processes require examination if crystal fractionation-differentiation processes are of petrologic importance at Oldoinyo Lengai. Carbonatitic magma has low density and viscosity. The density of (silicate-free) natrocarbonatite magma has been estimated to be (2.0–2.1) × 103 kg/m3 by Wolff (1994). Norton & Pinkerton (1997) measured the viscosity of natrocarbonatites from the 1988 eruption. These measurements showed that degassed, phenocryst-poor lavas have viscosities ranging from 1 to 5 Pa, i.e. considerably lower than in basaltic lavas; whereas a highly vesicular lava with a high phenocryst content had a viscosity of 120 Pa. However, Dawson et al. (1994) calculated that the poorly vesicular and crystal-rich lavas from the Chaos Crags flow (June 1993) had an apparent viscosity of 3 × 107 to 7 × 108 Pa, i.e. the effect of high crystallinity in the flows is to increase the viscosity to values similar to those observed for rhyolites. The physical properties of natrocarbonatites are relevant to their potential behaviour in magma chamber(s). Sparks et al. (1984) suggested that in situ nucleation on the wall of the magma chamber is the most probable site for crystallization, and that when crystallizationoccurs, the melt immediately adjacent to the growing crystals can convect away from its point of origin. Sparks et al. (1984) termed this process convective fractionation, and indicated that during crystallization along the margins of a chamber, highly fractionated magmas can be generated without requiring large amounts of crystallization, because the removal and concentration of chemical components affects only a small fraction of the total magma. Sparks et al. (1984) suggested that, if the density of the differentiated liquid is low, these liquids can efficiently segregate to the top of the magma chamber. Convective fractionation processes are likely to be very efficient on low-density and -viscosity natrocarbonatite melts. Petrogenetic model A petrogenetic model for the formation of silicate-bearing natrocarbonatite and silicate-free natrocarbonatite differentiates, as constrained by experimental and field studies, is illustrated in Fig. 8 and discussed in greater detail below. Previous experimental studies (Kjarsgaard et al., 1995) constrained exsolution of silicate-bearing natrocarbonatite to temperatures of 700–750°C at 100 MPa. Any exsolved silicate-bearing natrocarbonatite melt may carry nepheline, wollastonite, clinopyroxene, melilite and melanite garnet, which precipitated in equilibrium with two liquids (Kjarsgaard et al., 1995; C. M. Petibon, unpublished data, 1998); hence, a variety of silicate crystals may be available early on in the cooling history. Any silicate-bearing natrocarbonatite parent magma that separates from its nephelinite parent will evolve independently. At 100 MPa, the natrocarbonate melt would immediately precipitate nepheline on cooling (Fig. 3). However, based on the new experimental data of this paper, we propose that once the silicate-bearing natrocarbonatite has exsolved from its parent and separated, it evolves at 20 MPa pressure. During the temperature interval between exsolution (750°C) and eruption (615°C) of silicate-bearing natrocarbonatites, silicate minerals co-precipitate for the first 125°C of cooling, being joined by carbonate crystals at temperatures ≤635°C. Fig. 8. Open in new tabDownload slide Petrogenetic model. Silicate-bearing natrocarbonatite magma is exsolved from a wollastonite nephelinite parent magma at 100 MPa(∼3.3 km depth). Because of its low viscosity and density, the silicate-bearing natrocarbonatite efficiently segregates from its host, and then separates and collects in high-level 20 MPa (∼0.6 km depth) magma chambers. Convection in this chamber keeps precipitating silicate phases and aggregates (‘silicate spheroids’) in suspension. In situ crystallization of silicate crystals occurs because of nucleation on the chamber walls.Low-density, silica-poor natrocarbonatite differentiates migrate to the roof of the magma chamber, and carbonate phases precipitate. Thisdifferentiated magma subsequently moves to the small near-surface chambers, continuously precipitating nyerereite, gregoryite and apatite before eruption. It should be noted that eruptions of silicate-bearing natrocarbonatite originate directly from the high-level magma chamber. In high-level (∼20 MPa) magma chambers, in situ crystallization of silicate minerals on the walls of the magma chamber produces residual silicate-free natrocarbonatite magma. Because of its low density this natrocarbonatite melt segregates to the top of the chamber, where nyerereite, gregoryite and apatite precipitate. Because of the highly effective manner in which convective fractionation works (Sparks et al., 1984), the silicate-bearing natrocarbonatite parent magma is dominant in the magma chamber, where mainly silicate mineral phases precipitate. If convective velocities in the 20 MPa magma chamber are sufficiently high, silicate crystals may be retained in the magma. However, these crystals may also be fractionated out, as the residual silicate-bearing natrocarbonatite moves up into the volcanic edifice. Pyle et al. (1995) have suggested that at Oldoinyo Lengai there is a very shallow plumbing system, composed of small (10 m radius) chambers containing transient foam layers, which in turn are connected at depth to a larger chamber. We interpret that silica-poor natrocarbonatite magma segregated at the top of the 20 MPa magma chamber subsequently moves into these 10 m radius near-surface magma chambers, where nyerereite, gregoryite and apatite continue to crystallize, before eruption. Conclusion Comparison of the results obtained by petrological and experimental studies on silicate-bearing natrocarbonatite lava OL5 indicates that this sample is a suitable candidate for a natrocarbonatite parent magma. A critical feature of the experimental phase relationships for OL5 is that nepheline crystallizes first, and a variety of silicate minerals co-precipitate over a protracted cooling interval. Crystallization at 20 MPa is dominated by ferromagnesian silicates and nepheline down to 635°C. Silicate-free natrocarbonatite lavas are the typical eruption product at Oldoinyo Lengai. Their predominance is thought to be due to their low density and viscosity. Our preferred model for the origin of these lavas is that they are differentiates that form in a high-level (20 MPa) magma chamber by the in situ crystallization of silicate-bearing natrocarbonatite. The removal of the silicate component of the parent magma by sidewall crystallization gives rise to a low-density magma, which rises along the wall to the top of the chamber. Between the high-level chamber (20 MPa) and the surface these magmas may undergo further differentiation in very small diameter, near-surface magma chambers. Acknowledgements Funding for this research was provided by an NSERC Operating Grant to G.A.J., an NSERC–IOR grant providing travel and stipend funds to C.M.P., a Memorial University scholarship to C.M.P., and by the Geological Survey of Canada (GSC) for the experimental studies. We thank Tony Peterson for discussion and samples, and Keith Bell, who provided our piece of OL5. Technical assistance from a number of people in Ottawa at the GSC and in St John's at Memorial University was highly appreciated. Helpful comments by reviewers D. H. Green, R. Sweeney, K. Bell and K. Moore were greatly appreciated. This is GSC Contribution 1998101. References Bell K. , Keller J. . , Carbonatite Volcanism: Oldoinyo Lengai and the Petrogenesis of Natrocarbonatites. IAVCEI Proceedings in Volcanology 4 , 1995 Berlin Springer-Verlag Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Bell K. , Simonetti A. . Carbonatite magmatism and plume activity: implications for the Nd, Pb and Sr isotope systematics of Oldoinyo Lengai , Journal of Petrology , 1996 , vol. 37 (pg. 1321 - 1339 ) Google Scholar Crossref Search ADS WorldCat Church A. A. , Jones A. P. . silicate–carbonate immiscibility at Oldoinyo Lengai , Journal of Petrology , 1995 , vol. 36 (pg. 869 - 889 ) Google Scholar Crossref Search ADS WorldCat Cooper A. F. , Gittins G. , Tuttle O. F. . The system Na2CO3–K2CO3-CaCO3 at 1 kilobar and its significance in carbonatite petrogenesis , American Journal of Science , 1975 , vol. 275 (pg. 534 - 560 ) Google Scholar Crossref Search ADS WorldCat Dawson J. B. , Smith J. V. , Steele I. M. . Combeite (Na2.3Ca1.74 others0.12)Si3O9 from Oldoinyo Lengai, Tanzania , Journal of Geology , 1989 , vol. 97 (pg. 365 - 372 ) Google Scholar Crossref Search ADS WorldCat Dawson J. B. , Smith J. V. , Steele I. M. . 1966 ash eruption of the carbonatite volcano Oldoinyo Lengai: mineralogy of lapilli and mixing of silicate and carbonate magmas , Mineralogical Magazine , 1992 , vol. 56 (pg. 1 - 16 ) Google Scholar Crossref Search ADS WorldCat Dawson J. B. , Pinkerton H. , Pyle D. M. , Nyamweru C. . June 1993 eruption of Oldoinyo Lengai, Tanzania: exceptionally viscous and large carbonatite lava flows and evidence for coexisting silicate and carbonate magmas , Geology , 1994 , vol. 22 (pg. 799 - 802 ) Google Scholar Crossref Search ADS WorldCat Dawson J. B. , Pinkerton H. , Norton G. E. , Pyle D. M. , Browning P. , Jackson D. , Fallick A. E. . Bell K. , Keller J. . Petrology and geochemistry of Oldoinyo Lengai lavas extruded in November 1988: magma source, ascent and crystallization , Carbonatite Volcanism: Oldoinyo Lengai and the Petrogenesis of Natrocarbonatites. IAVCEI Proceedings in Volcanology 4 , 1995 Berlin Springer-Verlag (pg. 47 - 69 ) Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Dawson J. B. , Pyle D. M. , Pinkerton H. . Evolution of natrocarbonatite from a wollastonite nephelinite parent: evidence from the June 1993 eruption of Oldoinyo Lengai, Tanzania , Journal of Geology , 1996 , vol. 104 (pg. 41 - 54 ) Google Scholar Crossref Search ADS WorldCat Donaldson C. H. , Dawson B. . Skeletal crystallization and residual glass compositions in a cellular alkalic pyroxenite nodule from Oldoinyo Lengai , Contributions to Mineralogy and Petrology , 1978 , vol. 67 (pg. 139 - 149 ) Google Scholar Crossref Search ADS WorldCat Droop G. T. R. . A general equation for estimating Fe3+concentrations in ferromagnesian silicates and oxides from microprobe analyses, using stoichiometric criteria , Mineralogical Magazine , 1987 , vol. 51 (pg. 431 - 435 ) Google Scholar Crossref Search ADS WorldCat Freestone I. C. , Hamilton D. H. . The role of liquid immiscibility in the genesis of carbonatites: an experimental study , Contributions to Mineralogy and Petrology , 1980 , vol. 73 (pg. 105 - 117 ) Google Scholar Crossref Search ADS WorldCat Hamilton D. L. , Bedson P. , Esson J. . Bell K. . The behaviour of trace elements in the evolution of carbonatites , Carbonatites: Genesis and Evolution , 1989 London Unwin Hyman (pg. 405 - 427 ) Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Keller J. , Hoefs J. . Bell K. , Keller J. . Stable isotope characteristics of recent natrocarbonatites from Oldoinyo Lengai June 1988 , Carbonatite Volcanism: Oldoinyo Lengai and the Petrogenesis of Natrocarbonatites. IAVCEI Proceedings in Volcanology 4 , 1995 Berlin Springer-Verlag (pg. 113 - 123 ) Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Keller J. , Krafft M. . Effusive natrocarbonatite activity of Oldoinyo Lengai, June 1988 , Bulletin Volcanologique , 1990 , vol. 52 (pg. 629 - 645 ) Google Scholar Crossref Search ADS WorldCat Keller J. , Spettel B. . Bell K. , Keller J. . The trace element composition and petrogenesis of natrocarbonatites , Carbonatite Volcanism: Oldoinyo Lengai and the Petrogenesis of Natrocarbonatites. IAVCEI Proceedings in Volcanology 4 , 1995 Berlin Springer-Verlag (pg. 70 - 86 ) Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Kjarsgaard B. A. , Hamilton D. L. . Bell K. . The genesis of carbonatites by immiscibility , Carbonatites: Genesis and Evolution , 1989 London Unwin Hyman (pg. 388 - 404 ) Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Kjarsgaard B. A. , Hamilton D. L. , Peterson T. D. . Bell K. , Keller J. . Peralkaline nephelinite/carbonatite liquid immiscibility: comparison of phase compositions in experiments and natural lavas from Oldoinyo Lengai , Carbonatite Volcanism: Oldoinyo Lengai and the Petrogenesis of Natrocarbonatites. IAVCEI Proceedings in Volcanology 4 , 1995 Berlin Springer-Verlag (pg. 163 - 190 ) Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC van Groos , Koster A. F. , Wyllie P. J. . Liquid immiscibility in the system Na2O–Al2O3–SiO2–CO2 at pressures to 1 kilobar , American Journal of Science , 1966 , vol. 264 (pg. 234 - 255 ) Google Scholar Crossref Search ADS WorldCat Koster van , Groos A. F. , Wyllie P. J. . Liquid immiscibility in the join NaAlSiO8–CaAl2Si2O8–Na2CO3–H2O , American Journal of Science , 1973 , vol. 273 (pg. 465 - 487 ) Google Scholar Crossref Search ADS WorldCat Lee W. , Wyllie P. J. . Petrogenesis of carbonatite magmas from mantle to crust, constrained by the system CaO–(MgO + FeO*)–(Na2O + K2O)–(SiO2+ Al2O3+ TiO2)–CO2 , Journal of Petrology , 1998 , vol. 39 (pg. 495 - 517 ) Google Scholar Crossref Search ADS WorldCat Norton G. , Pinkerton H. . Rheological properties of natrocarbonatite lavas from Oldoinyo Lengai, Tanzania , European Journal of Mineralogy , 1997 , vol. 9 (pg. 351 - 364 ) Google Scholar Crossref Search ADS WorldCat Peterson T. D. . Peralkaline nephelinites. I. Comparative petrology of Shombole and Oldoinyo Líengai, East Africa , Contributions to Mineralogy and Petrology , 1989 , vol. 101 (pg. 458 - 478 ) Google Scholar Crossref Search ADS WorldCat Peterson T. D. . Petrology and genesis of natrocarbonatite , Contributions to Mineralogy and Petrology , 1990 , vol. 105 (pg. 143 - 155 ) Google Scholar Crossref Search ADS WorldCat Pyle D. M. , Pinkerton H. , Norton G. E. , Dawson J. B. . Bell K. , Keller J. . The dynamics of degassing at Oldoinyo Lengai , Carbonatite Volcanism: Oldoinyo Lengai and the Petrogenesis of Natrocarbonatites. IAVCEI Proceedings in Volcanology 4 , 1995 Berlin Springer-Verlag (pg. 37 - 46 ) Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Simonetti A. , Bell K. , Shrady C. . Trace and rare-earth-element geochemistry of the June 1993 natrocarbonatite lavas, Oldoinyo Lengai (Tanzania): implications for the origin of carbonatite magmas , Journal of Volcanology and Geothermal Research , 1997 , vol. 75 (pg. 89 - 106 ) Google Scholar Crossref Search ADS WorldCat Sparks R. S. J. , Huppert H. E. , Turner F. R. S. . The fluid dynamics of evolving magma chambers , Philosophical Transactions of the Royal Society of London, Series A , 1984 , vol. 310 (pg. 511 - 534 ) Google Scholar Crossref Search ADS WorldCat Sweeney R. J. , Falloon T. J. , Green D. H. . Bell K. , Keller J. . Experimental constraints on the possible mantle origin of natrocarbonatites , Carbonatite Volcanism: Oldoinyo Lengai and the Petrogenesis of Natrocarbonatites. IAVCEI Proceedings in Volcanology 4 , 1995 Berlin Springer-Verlag (pg. 191 - 207 ) Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Wolff J. A. . Physical properties of carbonatite magmas inferred from molten salt data, and application to extraction patterns from carbonatite-silicate magma chambers , Geological Magazine , 1994 , vol. 131 (pg. 145 - 153 ) Google Scholar Crossref Search ADS WorldCat Wyllie P. J. . Constraints imposed by experimental petrology on possible and impossible magma sources and products , Philosophical Transactions of the Royal Society of London, Series A , 1984 , vol. 310 (pg. 439 - 456 ) Google Scholar Crossref Search ADS WorldCat © Oxford University Press 1998 © Oxford University Press 1998
TI - Phase Relationships of a Silicate-bearing Natrocarbonatite from Oldoinyo Lengai at 20 and 100 MPa
JO - Journal of Petrology
DO - 10.1093/petroj/39.11-12.2137
DA - 1998-11-01
UR - https://www.deepdyve.com/lp/oxford-university-press/phase-relationships-of-a-silicate-bearing-natrocarbonatite-from-6b3C2FywCR
SP - 2137
EP - 2151
VL - 39
IS - 11-12
DP - DeepDyve
ER -