Search

Scapolite is abundant in medium to high grade metamorphic rocks with evaporitic calc-silicate bulk compositions, and because it can be Cl− and SO42− rich, is able to provide a unique view into the evolution of deep brines during orogenic cycles. In the Zhongtiao Mountains, part of the Trans-North China Orogen (TNCO), scapolite occurs in two distinct mineralogical associations; scapolite-I in carbonate-bearing biotite schist and scapolite-II in amphibolites. Initially, metamorphic Cl-rich scapolite-I formed in the schist at low fluid-rock ratio and peak pressure (meionite content = 32–37%, Cl = 2.34–2.75 wt.%). This observation, together with high molar Cl/Br ratios (Cl/Br = 973–2236), implies a halite-bearing evaporitic protolith. During a second stage of scapolite formation in the calc-silicates, rims overgrew scapolite-I with distinct compositions (Me% = 31–32, Cl = 2.64–3.0 wt.%, Cl/Br = 408–892). The lower molar Cl/Br ratios in scapolite-I rims imply that the later fluid was distinctly different to the earlier metamorphic fluid, and may have evolved through a stage of halite saturation induced by phase separation. Halite saturation would drive Br enrichment in the remaining fluid because Cl is preferentially sequestered in halite. Also at this time, amphibolites were subjected to intense open system fluid-rock interactions resulting in formation of scapolite-II (Me% = 30–32, Cl = 2.74–2.97 wt.%, Cl/Br = 291–1426), similar to that of the scapolite-I. S6+/∑S ratios in scapolite obtained via synchrotron XANES indicate that equilibration between fluid and scapolite would buffer the fluid to oxidized conditions, with oxygen fugacity (log fO2) beyond −15. The scapolitic rocks in this region were thus able to play a key role in controlling the salinity and oxidation state of the deep metamorphic fluid, thereby creating enhanced potential for scavenging of a range of metals within the metamorphic terrane and promoting syn-orogenic hydrothermal ore formation.

This paper provides a report of a study of a series of boudined bodies of mineralogically zoned calc-silicate bodies, some over 1 m thick, enclosed in granulite facies charnockitic gneisses on West Ongul Island, Lützow-Holm Complex, East Antarctica. Zircons extracted from the centre of one of the calc-silicate bodies indicate that they were metamorphosed around 532.3 ± 0.8 Ma and that the cores of these zircons were inherited from igneous rocks intruded 738.5 ± 1.8 Ma and possibly 605.0 ± 1.2 Ma. The meionite content of scapolite indicates that the peak metamorphic temperature reached a minimum of 790 °C. In the outermost rim of the calc-silicate scapolite is partially replaced by clinopyroxene, probably as a result of the influx of Mg bearing fluids from the surrounding charnockite gneiss during uplift and decompression. An influx of hydrous fluid resulted in the almost complete retrogressive alteration of plagioclase to sericite throughout the calc-silicates, with minor growth of amphibole after clinopyroxene and calcite after scapolite only in the innermost zone.

We report on the petrology and metamorphic history of boudined bodies of zoned calc-silicate rock, up to 1 m in width, which are enclosed in gneisses, transitional between upper amphibolite to granulite facies, at Akarui Point, Lützow-Holm Complex, East Antarctica. Zircons extracted from a closely related body of Mn-exhalative derived calc-silicates indicate that the metamorphism occurred in two separate events, at 546.6 ± 0.6 and 524.3 ± 0.6 Ma, during the Pan African Orogeny. Based on the meionite content of scapolite from different mineral zones in the calc-silicates, we derive a metamorphic history that involved a peak temperature of a minimum 810 – 820 °C, followed by a fall to 760-800 °C and then to 690-765 °C during uplift and decompression to 5.5 – 7 kbar. GAP geothermo-barometry indicates that the enclosing gneiss partially re-equilibrated during retrogression at 500-580 °C.

The crystal structures of marialite (Me6) from Badakhshan, Afghanistan and meionite (Me93) from Mt. Vesuvius, Italy were obtained using synchrotron high-resolution powder X-ray diffraction (HRPXRD) data and Rietveld structure refinements. Their structures were refined in space groupsI4/mandP42/n, and similar results were obtained. TheMe6sample has a formula Ca0.24Na3.37K0.24[Al3.16Si8.84O24]Cl0.84(CO3)0.15, and its unit-cell parameters area=12.047555(7),c=7.563210(6) Å, andV=1097.751(1) Å3. The average ⟨T1-O⟩ distances are 1.599(1) Å inI4/mand 1.600(2) Å inP42/n, indicating that the T1 site contains only Si atoms. InP42/n, the average distances of ⟨T2-O⟩=1.655(2) and ⟨T3-O⟩=1.664(2) Å are distinct and are not equal to each other. However, the mean ⟨T2,3-O⟩=1.659(2) Å inP42/nand is identical to the ⟨T2′-O⟩=1.659(1) Å inI4/m. The ⟨M-O⟩ [7]=2.754(1) Å (M site is coordinated to seven framework O atoms) and M-A=2.914(1) Å; these distances are identical in both space groups. TheMe93sample has a formula of Na0.29Ca3.76[Al5.54Si6.46O24]Cl0.05(SO4)0.02(CO3)0.93, and its unit-cell parameters area=12.19882(1),c=7.576954(8) Å, andV=1127.535(2) Å3. A similar examination of theMe93sample also shows that both space groups give similar results; however, the C–O distance is more reasonable inP42/nthan inI4/m. Refining the scapolite structure nearMe0orMe100inI4/mforces the T2 and T3 sites (both with multiplicity 8 inP42/n) to be equivalent and form the T2′ site (with multiplicity 16 inI4/m), but ⟨T2-O⟩ is not equal to ⟨T3-O⟩ inP42/n. Using different space groups for different regions across the series implies phase transitions, which do not occur in the scapolite series.

Scapolite oikocrysts, up to 3 cm in diameter, occur in cognate, cumulate nodules from the 2 Ma rift-related, alkali-basaltic Kula Volcanic Province in Western Turkey. Three scapolite-bearing nodules were found within deposits originating from two adjacent cones. They contain primocrysts of clinopyroxene, biotite, apatite and titanite, together with interstitial kaersutite, biotite, plagioclase and scapolite. Kula scapolites fall mid-way between the meionite (Ca4[Al6Si6O24]CO3) and marialite (Na4[Al3Si9O24]Cl) end-members (∼Me45—71). We believe this to be the first report of interstitial magmatic scapolite.

The scapolite structure consists of two interpenetrating frameworks: [(Si,Al)12O24] and {Na,Ca}{Cl,(CO3)}. The unstrained dimensions of each framework are significantly different (as determined by the valence-matching principle and distance-least-squares refinement); there is an intrinsic steric stress between the two frameworks, the magnitude of which is a function of chemical composition. Evidence of the resulting structural strain is the large U eq values for Cl at the A site across the marialite–meionite series and the large U eq values for (Na,Ca) at the M site in marialite. Local bond-valence considerations indicate extensive SRO (Short-Range Order) about both Cl and (CO3). The short-range bond-valence requirements of the (CO3) group are satisfied by coordination with Ca4 at ~ Me 100, NaCa3 at ~ Me 80, Na2Ca2 at ~ Me 55, Na3Ca at ~ Me 40, and Na4 at ~ Me 20. The driving mechanism for the I 4/ m → P 42/ n phase-transitions is the coupling of SRO between the [(Si,Al)12O24] and {Na,Ca}{Cl,(CO3)} frameworks. At a composition of ~ Me 50, local bond-valence requirements force the local clusters Na2Ca2 to occur in a trans configuration {NaCaNaCa} around the A site. Anion bond-valence requirements also force these clusters to link to each other through Na–O–Na or Ca–O–Ca linkages, and hence the combination of SRO and bond-valence requirements gives rise to LRO (long-range order). In turn, Al and Si are constrained to order at different tetrahedra, an ordering that breaks the topological I 4/ m symmetry and gives rise to two distinct tetrahedra, T (2) and T (3), dominated by Al and Si, respectively. The result is the P 42/ n scapolite structure. As Na or Ca become dominant, the number of Na2Ca2 clusters decreases, the effectiveness of the symmetry-lowering mechanism lessens, and the degree of order of Al and Si over the T (2) and T (3) sites decreases toward both the marialite and meionite ends of the series. At the compositions ~ Me 22 and Me 78, the clusters NaCa3 and Na3Ca dominate, the driving mechanism for lower symmetry vanishes, and the structures revert to maximal I 4/ m symmetry. The nonlinear variations in bulk composition across the marialite–meionite series are also the result of SRO of M -site Na and Ca around the A site [which is occupied by Cl in end-member marialite and (CO3) in end-member meionite]. Local bond-valence requirements indicate extensive SRO about both Cl and (CO3) in scapolite. In particular, Cl cannot occur in I 4/ m meionite except where K is present; in this case, chemical variations indicate that Cl enters the I 4/ m meionite structure as {ClK2Ca2} clusters.

Synchrotron high-resolution powder X-ray diffraction (HRPXRD) data and Rietveld structure refinements were used to examine the crystal structures of quartz, sodalite, tremolite, and meionite Me79.6, and compare them with those obtained by single-crystal diffraction (SXTL). The purpose is to illustrate that crystal structures obtained by HRPXRD are comparable to those obtained by SXTL. The comparisons indicate important and significant differences between the structures obtained by the two methods. The cell parameters obtained by Rietveld refinements using HRPXRD data appear of superior quality to those obtained by SXTL. The distances in pure SiO4 tetrahedron in quartz, sodalite, and tremolite are 1.6081(3), 1.6100(2), and 1.620(1) A, respectively. These values are affected by interstitial cations. In meionite Me79.6, the average and distances are 1.647(1) and 1.670(1) A, respectively, and they indicate that the occupancies are (Al0.28Si0.72) for T 1 where the atoms are partially ordered and (Al0.45Si0.55) for T 2 site where the atoms are nearly disordered, based on sodalite Si–O and Al–O distances of 1.6100(2) and 1.7435(2) A, respectively.

Scapolite-group minerals are a group of rock-forming framework aluminosilicates common in a wide variety of igneous and metamorphic rocks. They have the general formula M 4 [ T 12 O 24 ] A , where M represents Na, K, Ca, Sr, Ba and Fe 2+ , T stands for Si, Al, (Fe 3+ ), and A represents Cl, (CO 3 ), (SO 4 ). They constitute a non-binary solid-solution series involving the end members marialite (Na 4 [Al 3 Si 9 O 24 ] Cl, Ma ) and meionite (Ca 4 [Al 6 Si 6 O 24 ] CO 3 , Me ). The crystal structures of eighteen scapolite-group minerals from Me 4.9 to Me 92.8 [ Me = 100 ∑ M 2+ /∑( M 2+ + M + )] have been refined with Mo K α X-radiation. The crystals, a 12.045–12.208, c 7.564–7.586 A, V 1099.7–1130.1 A 3 , I 4/ m or P 4 2 / n , Z = 2, were analyzed with an electron microprobe subsequently to collection of the X-ray intensity data. There are subtle differences between the behavior of the distinct T tetrahedra in the I 4/ m and P 4 2 / n structures. For Si and Al, determinative curves for the assignment of site populations from T –O> distances were developed for each T site in each structure type, and T -site populations were assigned from these curves. In I 4/ m marialite, there is no Al at the T (1) site. In P 4 2 / n scapolite, Al is strongly ordered at T (2), and Si is strongly ordered at T (3), but this order is never complete. The A -site cation (Na, Ca, K) is [8]-coordinated in end-member marialite and is (on average) [8.5]-coordinated in end-member meionite. Each ( M ϕ n ) polyhedron shares two edges with adjacent ( M ϕ n ) polyhedra, forming a three-dimensional framework that interpenetrates the [ T 12 O 24 ] silicate framework. In (CO 3 )-bearing structures, the model proposed by previous workers, four (CO 3 ) groups arranged about the 4 axis ( I 4/ m ) or 4 axis ( P 4 2 / m ), is consistent across the series from marialite to meionite. Any (SO 4 ) present is disordered by rotation of 90° about the central 4-fold axis; in I 4/ m structures, each arrangement is 50% occupied, whereas in P 4 2 / n structures, the occupancies of the two arrangements are not required to be equal.

Spectacular reaction textures in poikiloblastic scapolitite boudins, within marbles in the continental crust exposed in the Lutzow–Holm Complex, East Antarctica, provide insights into the changing fluid composition and movement of fluid along grain boundaries and fractures. Petrographic and geochemical features indicate scapolite formation under contrasting fluid compositions. Core composition of scapolite poikiloblasts (ScpI) are marialitic (Cl = 0.7 apfu) whereas rims in contact with biotite or clinopyroxene are meionite rich. Fine-grained recrystallized equigranular scapolite (ScpII) shows prominent chemical zoning, with a marialitic core and a meionitic rim (Cl = 0.36 apfu). Scapolite poikiloblasts are traversed by ScpIII reaction zones along fractures with compositional gradients. Pure CO2 fluid inclusions are observed in healed fractures in scapolite poikiloblasts. These negative crystal-shaped fluid inclusions are moderately dense, and are believed to be coeval with ScpIII formation at temperatures >600 °C and a minimum pressure of c. 3.8 kbar. Grain-scale LA-ICPMS studies on trace and rare earth elements on different textural types of scaplolites and a traverse through scapolite reaction zone with compositional gradient suggest a multistage fluid evolution history. ScpI developed in the presence of an internally buffered, brine-rich fluid derived probably from an evaporite source during prograde to peak metamorphism. Recrystallization and grain size reduction occurred in the presence of an externally sourced carbonate (CaCO3)-bearing fluid, resulting in the leaching of Cl, K, Rb and Ba from ScpI along fractures and grain boundaries. Movement of fluids was enhanced by micro-fracturing during the transformation of ScpI to ScpIII. Fractures in fluorapatite are altered to chlorapatite proving evidence for the pathways of escaping Cl-bearing fluids released from ScpI. The present study thus provides evidence for the usefulness of scapolite in fingerprinting changing volatile composition and trace element contents of fluids that percolate within the continental crust.

The crystal structure of scapolite from the Liitzow-Holm Bay region, East Antarctica with cell formular (Na_Ca_K_Mg_) _ (Al_Si_Fe_) _O_ (Cl_(CO_3) _) _, Me_, was determined in tetragonal, P4_2/n, a=12.122(2); c=7.585(2) A by the single crystal X-ray diffraction method. The structure is the same as that of meionite but with some difference in atomic coordinates and Al percentage in the tetrahedon. The CO_3 groups were mainly disordered on the xy plane and off-centred from the Anion site. The cation site was 8-coordinate rather than 6-coordinate.

Rietveld structure refinement, 29 Si magic angle spinning (MAS) and 27 Al satellite transition (SATRAS) NMR spectroscopic data are given for three natural samples of meionite and for synthetic end-member meionite. Trends in structural parameters and NMR spectra are discussed for the solid-solution series from marialite to meionite. The a cell dimension and 27 Al SATRAS NMR spectra show a continuous change through the series, whereas the space group, c cell dimension and 29 Si MAS NMR spectra show different trends for the three subseries, Si 9.0–8.3, Si8.3–7.4 and Si7.4–6.0. Over the interval Si 9.0–8.3, the space group is I4/m, the c cell dimension increases with Al content, and 29 Si spectra show multiple well-resolved peaks; the anion site is occupied by Cl – , and the T(1) site, by Si. Si–Al order between the T(2) and T(3) sites causes the space group for Si 8.3–7.4 to be P42/n and the 29 Si MAS NMR spectra to be dominated by Si(1) (3Si1Al) and Si(3) (1Si3Al) peaks. Here, there is a negative correlation between c cell dimension and Al content. Over the interval Si 7.4–6.0, the c cell parameter increases with Al content, the space group is I4/m, and multiple 29 Si MAS NMR peaks become broad and poorly resolved. Ca 2+ continues to replace monovalent cations in the M site

Two types of zoned scapolite in calc-silicate gneisses from Skallen, East Antarctica, were found. One type varies from Me_ in the core to Me_ at the rim. Associated minerals are mainly diopside, pargasite and spinel. The other varies from Me_ in the core to Me_ at the rim. Associated minerals are mainly diopside, phlogopite and K-feldspar (Or_). Scapolite in this specimen contains characteristically K, and the substitution of K in the M site is 8-12%. The unit cell dimensions of scapolite (Me_) from Skallen are a=12.121 (1), c=7.585 (2) A, and those (Me_) from Sri Lanka are a=12.164 (1), c=7.568 (1) A. These scapolites belong to different isomorphous series. It is likely that the zonation and K-substitution of scapolites from Skallen appear to be formed by a reaction with minerals in direct contact and/or fluids in later stage alteration.

Silvialite, ideally Ca4Al6Si6O24SO4, is tetragonal,I4/m, Z = 2, witha= 12.160(3),c= 7.560(1) Å,V= 1117.9(8) Å3,c:a= 0.6217:1, ω = 1.583, ε = 1.558 (uniaxial negative), Dm= 2.75 g/cm3, Dcalc= 2.769 g/cm3and H (Mohs) = 5.5. It is transparent and slightly yellow, has a good {100} cleavage, chonchoidal fracture, white streak and a vitreous lustre. It occurs in upper-mantle garnet-granulite xenoliths hosted by olivine nephelinite, from McBride Province, North Queensland, Australia. The empirical formula, derived from electron-microprobe analysis, is (Na1.06Ca2.86)(Al4.87Si7.13)O24[(SO4)0.57(CO3)0.41]. Crystal-structure refinement shows disordered carbonate and sulfate groups along the fourfold axis. Silvialite is a primary cumulate phase precipitated from alkali basalt at 900–1000°C and 8–12 kbar under highfSO2andfO2. The name silvialite, currently used in literature to describe the sulfate analogue of meionite, was suggested by Brauns (1914).

Compositional trends of natural samples of scapolite extend toward the ideal (anhydrous) end-members marialite (Ma) Na4Al3Si9O24Cl and meionite (Me) Ca4Al6Si6O24CO3. Intermediate members do not follow the plagioclase substitution, although Si contents closely correlate with Me% [calculated as 100{Σ(divalent cations)}/4]. The solid solution may be explained by substitutional interactions among the M, T and A sites, as constrained by net charge-balance and maximum site-occupancy requirements of the structural formula M4T12O24A. Two changes in compositional trends located at [(Na3.4Ca0.6)(Al3.6Si8.4)O24]+1 and [(Na1.4Ca2.6)(Al4.7Si7.3)O24]+1.9 divide the series into three portions. Nomenclature consistent with these divisions is marialite (Me 0–15), calcian marialite (15 < Me ≤ 50), sodian meionite (50 < Me < 65) and meionite (Me 65–100). A large number of scapolite occurrences have Si values near 7.3 apfu, corresponding to the 14/m-P42/n symmetry transition and to the stabilization of scapolite in a Cl-poor environment. Trends of Cl with Si or (Na + K) apfu indicate complex substitutional mechanisms, and should not be used to suggest that “mizzonite” or “dipyre” are end-members. Calculation of excess positive charge from measurement of M and T cations indicates that scapolite typically has the maximum content of Cl expected by charge-balance. Anion substitution is complicated by substition of H as hydroxyl or bicarbonate-bisulfate anions. The calculation CO2 = 1−ClFS may apply only to anhydrous scapolites which are rare, otherwise it over-estimates CO2 because it ignores the role of H. CO2 content calculated by charge-balance equations gives better agreement with constraints of the known substitutional mechanism. More H is present than is required by stoichiometry, and molecular H2O may be present in channels along the c-axis.

Mesozoic shale, siltstone, sandstone and carbonate rocks and Eocene volcanic, volcaniclastic and pyroclastic rocks are intruded by late Eocene granitic batholith and related acidic dikes. As a result of hydrothermal activity associated with the granitoid, two distinct skarn zones of calcic skarn zone and magnesian skarn zone were formed in the carbonate rocks. Cl-bearing amphiboles of ferroactinolite to potassium rich ferrohastingsite composition are restricted to the calcic skarn zone and associating with salite, grandite, magnetite, calcite and sometimes with scapolite. The Cl content of amphiboles shows positive relationships with Fe2+/(Fe2++Mg) and K/(K+Na) and a negative relation with Si/(Si+AlIV). In contrast, F bearing phiogopite is a major mineral in the magnesian skarn. F content exhibits positive relationship with wt% of SiO2 and MgO. The estimated log(fH2O/fHF) equilibrating with the phlogopite ranges from 4.6 to 5.5 at 400°C. Cl rich scapolite is widespread in both calcic and magnesian skarn zones. The scapolite has a compositional range of 23 to 40 mole% meionite in marialite-meionite solid solution series. XC1 of the scapolite ranges from 0.44 to 0.98 (mostly more than 0.75). Scapolite of XC1=0.98 from the hornfels is the most Cl-rich scapolite ever reported. Cl content of the scapolite positively correlate with Na/(Na+Ca+K). Calculated XNaCl values for fluid in equilibrium with scapolite from the Sangan deposits reach up to more than 55 wt% equivalent NaCl. The Cl-rich hastingsite, Cl-rich scapolite, F-rich phiogopite, and fluid inclusions with high salinity (up to 70 wt% eq. NaCl) in granitic rocks and in skarn zones suggest that hydrothermal fluids were derived from the granitoid magma. The ore elements such as Fe and Cu could be transported by solutions as metal-chloride complexes. Large amounts of cations such as Al, Si, and Ca incorporated in the fluids should be derived from the country rocks and from the breakdown products of early skarn minerals.

The heat capacities of five mineral samples from the scapolite solid-solution series, Na 1 Al 3 Si 9 O 24 Cl (marialite) to Ca 4 Al 6 Si 6 O 24 CO 3 (meionite), were measured by the adiabatic method from T = 8 K to T = 350 K and by the differential scanning calorimetry (d.s.c.) method from T = 300 K to T = 1000 K. The meionite (Me) content in per cent {Me = 100 Ca*/(Ca* + Na*)} (where the asterisk indicates that possible substituents are included) and molar heat capacity ( C p.m / R ) at T = 298.15 K for each sample is: Me 28 , 82.07; Me 44 , 82.09; Me 55 , 83.95; Me 69 , 85.80; Me 88 , 84.54. The standard molar entropies, { S m o (298.15 K) - S m o (0K)} R ?1 ( R = 8.31451 J·K -1 ·mol -1 ), at T = 298.15 K for the respective compositions are : 85.05 ± 0.26, 83.78 ± 0.50, 85.22 ± 0.24, 85.76 ± 0.21, and 84.17 ± 0.59. The calculated standard molar entropies (as above) at T = 298.15 K for the end-members marialite and meionite, and for an intermediate composition (mizzonite=Me 75 ) are 84.85, 83.94 and 86.15, respectively. Values of the coefficients in the equation C p.m / R = a + bT + cT 2 + dT ?1/2 + eT -2 (valid from T = 300 K to T = 1000 K) are: (Me x , a , b /K, c /K 2 , d /K ?1/2 , e /K -2 ) Me 88 , 315.580, −0.0795676, 1.52825·10 -5 , −3954·83, 1808460; Me 69 , 261.285, −0.0415017, 8.73053·10 -7 , −3028.28, 1083666; Me 55 , 232.236, −0.0352222, 6.49875·10 -6 , 2505.99, 601750; Me 44 , 276.696, −0.0756614, 2.39722·10 -5 , -3210.40, 1044363; Me 28 , 149.917, 0.0229399, -1.23180·10 -5 , 1208.87; -318470. Smoothed thermodynamic functions for the five samples are also presented. The enthalpies of solution for five natural scapolites were measured in 2PbO·B 2 O 3 melts at T = 973 K by Calvet-type calorimetry.

In Madagascar, hibonite occurs as a rather frequent mineral within thorianite-bearing skarns which are widespread in the Pan African granulitic formations constituting the S-E part of the Island (Tranomaro area). In these skarns, leucocratic segregations made up of CO3-scapolite to meionite (Anequivalent=89–95% which implies T≥850° C), spinel and corundum were formed at stage 1 of metasomatism in a titanite-bearing matrix consisting of scapolite (Aneq=77–88) and aluminous diopside. During stage 2 of metasomatism, scapolite from the lenses were altered to anorthite+calcite while the less calcic scapolite remained stable which indicates T≈800° C. Hibonite crystallized at the expense of corundum and spinel. Expressed as mol% of the CaAl12O19/Ca(Al10TiR2+)O19/REE(Al11R2+)O19 [+Th (Al10R2+ 2)O19] end-members (R 2+=Mg, Fe2+, Zn2+; Al=Al, Fe3+; Ti=Ti, Si), its composition varies from 26/72/2 to 50/23/27. The ideal activity of the CaAl12O19 component is about 0.25. Fluid inclusions in corundum, hibonite and anorthite are composed of nearly pure CO2. In corundum, the isochores for primary inclusions are in agreement with the P-T estimates for regional metamorphism and stage 1 metasomatism (T≈850° C, P≈5 kbar). Inclusions with the highest density in hibonite and anorthite constrain P to about 3–3.5 kbar for T=800° C. Thermodynamic calculations indicate that, in addition to a low activity of CaAl12O19, stability of hibonite in equilibrium with anorthite and calcite implies an extremely low activity of silica (below the zircon-baddeleyite buffer). By contrast the activity of CO2 may be high, in agreement with the observed fluid compositions. These results are corroborated by a short comparison with the other granulite occurrences of hibonite in Tanzania and South India.

Scapolites are tetragonal framework aluminosilicates with the general formula M4T 12024A, the major constituents being M = Na + Ca, T = Si + AI, and A = Cl + C03 + S-bearing anions, along with minor Fe, Sr, Ba, K, and H. Substitutions among the M, T, and A sites are independent and constrained only by net charge balance and maximum site occupancy. The cell parameters vary primarily with framework chemistry and Al-Si order and are not a strong function of the M- or A-site substitutions. End-member marialite has a = 12.06(1) and c = 7.551(5) A, and meionite has a = 12.20(1) and c = 7.556(5) A. Changes in cellparameter trends in the series occur at AI3.6Sis.4024 and AI4.7Si7.3024 and are correlated with the locations of P421n-141m phase transitions. The scapolite group can be divided into three isomorphous series: 9.0 > Si > 8.4,8.4 > Si > 7.3, and 7.3 > Si > 6.0, each with its own patterns of Al-Si order. Because scapolites with the same Si content have variable cell parameters, there might be variable degrees of Al-Si order or a minor, secondary influence of the M- or A-site substituents.

Thermodynamic and phase equilibrium data for scapolite have been used to calculate CO2 activities (aCO2) and to evaluate the presence or absence of a fluid phase in high-grade scapolite bearing meta-anorthosite, granulites, calc-silicates, and mafix xenoliths. The assemblage scapolite-plagioclase-garnet±quartz may be used to calculate or limit aCO2 by the reaction Meionite+Quartz = Grossular+Anorthite+CO2. Granulites from four high-grade terranes (Grenville Province, Canada; Sargut Belt, India; Furua Complex, Tanzania; Bergen Arcs, Norway) yield aCO2=0.4-1, with most >0.7. For scapolite-bearing granulites from the Furua Complex, in which aCO2≥0.9, calculated H2O activities (aH2O) based on phlogopite dehydration equilibria are uniformly low (0.1–0.2). The aCO2 calculated for meta-anorthosite from the Grenville Province, Ontario, ranges from 0.2 to 0.8. For Grenville meta-anorthosite also containing epidote, the aH2O calculated from clinozoisite dehydration ranges from 0.2 to 0.6. Calc-silicates from the Grenville, Sargur, and Furua terranes mostly yield aCO2 1). The calculated fluid activities are consistent with metamorphism (1) in the presence of a mixed CO2−H2O fluid phase in which CO2 is the dominant fluid species but other C−O−H−S species are minor, (2) in the absence of a bulk fluid phase (“fluid-absent metamorphism”), or (3) in the presence of a fluid-bearing melt phase. The results for many granulites and Grenville meta-anorthosite are consistent with the presence of a CO2-rich, mixed CO2−H2O fluid phase. In contrast the relatively restricted and low values of aCO2 for calc-silicates require an H2O-rich fluid or absence of a fluid phase during metamorphism. The range of values for xenoliths are most consistent with absence of a fluid phase. The primary implication of these results is that a CO2-rich fluid accounts for the reduced aH2O in scapolite-bearing granulites. However, scapolite may be stable with a wide range of fluid compositions or in the absence of a fluid phase, and the presence of scapolite is not a priori evidence of a CO2-rich fluid phase. In addition, close association of scapolite-free mafic granulites with scapolite-bearing granulites having identical mineral compositions in the Furua Complex, and the absence of scapolite from most granulite terranes implies that a CO2-rich fluid phase is not pervasive on an outcrop scale or common to all granulite terranes.

Scapolite solid solution has been synthesized at 750°C and 4 kbar and is stable relative to plagioclase + calcite + halite over the range of plagioclase compositions from Ab 85 An 15 to Ab 70 An 30 , although albite + halite is stable relative to marialite, Na 4 Al 3 Si 9 O 24 Cl, and anorthite + calcite is stable relative to meionite, Ca 4 Al 6 Si 6 O 24 CO 3 . A chloride-free scapolite, mizzonite, has been synthesized at the approximate composition NaCa 3 Al 5 Si 7 O 24 CO 3 (Ab. 2An. CaCO 3 ). In the absence of chloride, a three-phase invariant assemblage, sodic plagioclase (~Ab 60 An 40 ) + scapolite + calcite is stable relative to plagioclase + calcite over the approximate range of plagioclase composition Ab 60 An 40 -Ab 35 An 65 and another three-phase invariant assemblage, calcic plagioclase (~Ab 15 An 85 ) + scapolite + calcite is stable over the approximate range Ab 30 An 70 -An 15 An 85 . Unit-cell dimensions and refractive indices have been determined for the scapotite synthesized in these experiments and are compared with values for chemically analyzed natural scapolites. Scapolite must be regarded as a ternary solid solution in which, at a given equivalent An-content, the Cl/CO 3 ratio in the large anion site can vary as a function of NaCl and CaCO 3 activities.

The stability field of the end-member scapolite meionite was determined in piston-cylinder apparatus. Meionite has very high thermal stability at high pressures, exceeding 1500° C at 20 kbar. Below 6 kbar and 1270 ° C scapolite breakdown is subsolidus, to an-orthite + gehlenite + wollastonite + CO2, with a slope of 20 bars/degree. An extrapolation of existing thermodynamic data for CO2 permits calculation of ΔGFo=-2384.5 kcal/mol for meionite at 1270 ° C, very close to the value for 3 anorthite + calcite. Above 1270 ° C, scapolite begins to melt to An+Geh+Liq+CO2, and as pressure increases the melting curve steepens, the Geh and An being progressively replaced by Liq+corundum with Al in 6-coordination. At pressures >25kbar dp/dt becomes negative, corundum is the only crystalline product, and CO2 bubbles disappear from the quenched glass, indicating a solubility of CO2 under these conditions of about 5 wt. percent in the liquid.

Near the village of Engeln, Laacher See area, garnet-bearing pyriclasite and pyribolite ejecta were recognized as constituents of alkaline basaltic tuffs; they are interpreted as fragments of the lowermost crust. During the first main stage of granulite facies metamorphism, assemblages with garnet (Alm 47 Pyr 34 Spess 2 Gross + Andr 17 ), clinopyroxene (Wo 37 En 35 Fs 15 Ts 8.5 Jd 4.5 ), orthopyroxene I (En 34 Fs 38 Ts 4 Jd 2 ), and plagioclase I (An 40 -An 60 ) were formed in a temperature range of 730–850°C and rock pressures somewhere between 6.5 and 12 kb, P tot > P H 2 O > 0 . The rare sulfate-rich meionite, and at least a part of the ubiquitous brown hornblende were presumably also formed during this stage. A retrograde metamorphic event under slightly lower pressures and temperatures led to the breakdown of the assemblage garnet + clinopyroxene thereby forming coronas of plagioclase II ( An 75 ) + orthopyroxene II + Ti-magnetite ± brown hornblende .

X-ray diffraction studies reveal that most scapolites belong to space groupP4 2 /n. The mean intensity of weak reflections violating body-centered symmetry, reaches a maximum around 37 per cent meionite and extrapolates to zero at the pure end-members. All intermediate scapolites of the marialitemeionite series are pseudobody-centered, and as the composition approaches that of the end-members the structure becomes truly body-centered (I4/m). The weakh+k+l odd reflections are directly related to (1) orderdisorder of Al and Si in the tetrahedral framework and (2) structural distortion from body-centered symmetry. Scapolites may be regarded as a continuous, homogeneous solid solution. The cell dimensions are only approximately linearly related to composition and there is a maximum deviation from linearity, for those scapolites (25% Me ≈ 45% Me) whose symmetry departs the most from body-centered symmetry. These phenomena are interpreted in terms of the structural characteristic of the scapolites.

This paper reports several new localities of wollastonite- and scapolite-bearing calc-silicate assemblages from the granulite-facies supracrustal Kerala Khondalite Belt (KKB), southern India. Based on mineralogy, these calc-silicate rocks are classified into four types: Type I, lacking wollastonite and grossular; Type II, wollastonite-bearing but grossular-absent; Type III, wollastonite- and grossular-bearing; and Type IV, dolomitic marbles. Detailed petrographic studies reveal a variety of reaction textures overprinting the polygonal granoblastic peak metamorphic assemblages in these rocks. The Type II calc-silicate rocks preserve reaction textures, including meionite breaking down to anorthite-calcite-quartz, wollastonite breaking down to calcite-quartz and meionite-quartz symplectites after K-feldspar and wollastonite. Type III calc-silicate rocks have porphyroblastic and coronal grossular. Grossular-quartz coronas separating wollastonite and anorthite and the development of grossular within the anorthite-calcite-quartz pseudomorphs of meionite form important retrograde reaction textures in this type. In Type IV dolomitic marble assemblages, meionite forming in grain boundaries of calcite and feldspars, forsterite rimmed by diopside-dolomite and the formation of grossular in feldspar-rich zones are the important textures. Calculated partial petrogenetic grids in the CaOAl 2 O 3 SiO 2 CO 2 system are used to deduce the pressure-temperature-fluid evolution of the calc-silicate rocks. The Type II assemblages provide CO 2 activity estimates of > 0.5, with a peak metamorphic temperature of about 790°C. Initial cooling followed by later CO 2 influx can be deduced from reaction modelling in these calc-silicate rocks. Type III assemblages are characterized by internal fluid buffering throughout their tectonic history. The formation of coronal grossular indicates an initial cooling from peak metamorphic temperatures of about 830°C deduced from vapour-absent meionite and grossular equilibria. Type IV marble assemblages also indicate internal fluid buffering followed by localized CO 2 influx. Overall, the calc-silicate rocks of the KKB define peak metamorphic temperatures in the range of 790–850°C, with an internally buffered fluid composition during the peak conditions. Initial cooling was followed by localized carbonic fluid influx that also post-dated decompression deduced from other rock types in the KKB.