The hydroxyl orientations in the 2H2 polytype of amesite, Mg2Al(SiAl)O5(OH)4, have been determined by minimizing the electrostatic potential energy as a function of OH orientation. The angles, ρ, between the hydroxyls and (001) vary between 81.5 and 88.8°. All surface hydroxyls form hydrogen bonds with oxygens of the adjacent layer. The OHs tend to tilt away from the higher charged A1VI ions, and the spread in ρ values is due to variations in the positions of the receptor oxygens. The inner hydroxyls noticeably weaken the interlayer bonding. The substitution of Al in both T and M sites creates a dipolar layer charge and the resulting attraction between layers forms an important part of the interlayer bonding. The 2H2 polytype of amesite has substantial interlayer bonding even with all surface hydroxyls replaced by fluorine. This is not true for a composition of Mg3Si2O5(OH)4 where a similar F for OH substitution destroys all interlayer bonding as in kaolinite.

Sepiolite from Vallecas-Vicálvaro, Spain, contains 1.3% fluorine. Laser microprobe mass spectrometry of this sepiolite suggests the presence of fragments of (SiO2)nOMgF and (SiO2)nOMgOH, which are typical of the sepiolite structure. During thermal dehydroxylation, the fluorine in this sepiolite is removed simultaneously with OH groups at about 750°C. Nuclear magnetic resonance spectroscopy (NMR) of 19F indicates that the fluorine is located in the interior of the sepiolite structure, probably substituting for OH groups, and is homogeneously distributed. In the Vallecas-Vicálvaro sepiolite, about one of every four OH groups bound to Mg2+ is substituted by fluorine. The kinetics of extraction of Mg2+ and F- ions by acid treatment (1 N HCl) shows a more rapid extraction of Mg2+, with a monotonous decrease of the Mg/F ratio as the extent of extraction increases. These results support the internal location of the fluorine, as suggested by the NMR data.

A series of natural clays, including 1:1 layer silicates (serpentines, kaolin minerals), smectites, vermiculite, micas, talc, pyrophyllite, sepiolite, and palygorskite, were studied by 19F magic-angle spinning (MAS) nuclear magnetic resonance (NMR) spectroscopy. The 19F chemical shift in these layer silicates is characteristic of the structure, in particular, to the local octahedral cation occupancy. Fluoride ions bonded to three Mg octahedral cations have a chemical shift of about −177 ppm and those bonded to two Al cations and a vacancy have a chemical shift of about −134 parts per million (ppm). The shift at −182.8 ppm in hectorite is apparently associated with fluoride bonded to two Al cations and a Li cation. Surprisingly, the difference in chemical shift of the interlayer and inner fluoride in 1:1 layer silicates is insufficient to distinguish these sites. Based on trends in chemical shift, it appears that fluoride substitution for inner hydroxyls in clays with octahedral substitution is not random. Fluoride is apparently preferentially associated with Mg rather than Al in the octahedral sheet as no resonance due to a fluoride bonded to two Al cations and a vacancy is observed in clays such as SAz-1.

Expandable fluorine micas were synthesized using talc and Na2SiF6 at 800°C for 2 hours in air, nitrogen, argon, and under vacuum. Gaseous SiF4, generated from Na2SiF6, and the resultant amorphous sodium silicofluoride formed during the reaction between talc and Na2SiF6 below 900°C are taking active part in the formation of expandable micas because the intensity of the 12.5 Å reflection of expandable micas decreases as the gas flow increases in the furnace. Expandable micas seem to be formed by the transformation from talc taking place without the entire disruption of the original atomic arrangement. This takes place with the loss of one Mg2+ from an octahedral site and by the intercalation of every two Na+ into the interlayer site of talc. Infrared absorption and thermal analyses show that expandable micas include a small amount of OH− in their structures.

High-resolution solid-state, fluorine-19, magic-angle spinning-nuclear magnetic resonance spectroscopy (MAS-NMR) was used to study natural and synthetic fluorinated 2:1 layer silicates of known composition. This technique enabled us to determine directly the coordination of structural fluorine and it was found to be sensitive to both the chemical nature of the octahedral elements (Al, Mg, Li) and the type of octahedral sheet (di- or trioctahedral). The observed chemical shifts at −132, −152, −176 and −182 ppm (relative to CFC13) were assigned to different environments of fluorine. The results were then used to characterize synthetic 2:1 layer silicates with unknown octahedral composition.

The fluoride ion is crystal chemically very similar to the hydroxyl ion, substituting for hydroxyl in many minerals in which hydrogen bonding is not important. Fluoride substitutions are particularly common in 2:1 layer silicates, such as micas, illites and smectites. The brick and tile industries, which use naturally occurring clays as their primary raw materials, have devoted considerable effort to understanding fluorine evolution during firing of the raw materials due to increasingly stringent fluorine emission regulations. In order to understand fluorine evolution from ceramic raw materials, we have studied a number of phyllosilicate materials used in making bricks. X-ray powder diffraction and fluorine analyses were combined with heating experiments and thermogravimetric analysis to evaluate the chemical and structural changes taking place on heating. Fluorine remained in 2:1 layer silicates to higher temperatures than did hydroxyl, but it behaved identically to hydroxyl in the kaolinite studied. In all cases, fluorine evolution coincided with structural breakdown of the clay host. These results show that fluorine evolution will consistently occur during firing of clay raw materials, and the problems of fluorine emission cannot be readily solved by simple variations of firing temperatures or times.

The cosmic origin of fluorine is still under debate. Asymptotic giant branch (AGB) stars are among the few suggested candidates to efficiently synthesis F in our Galaxy, however their relative contribution is not clear. In this paper, we briefly review the theoretical studies from stellar yield models of the F synthesis and chemical equilibrium models of the F-containing molecules in the outflow around AGB stars. Previous detections of the F-bearing species towards AGB and post-AGB stars are also highlighted. We suggest that high-resolution ALMA observations of the AlF, one of the two main carriers of F in the outflow of AGB stars, can provide a reliable tracer of the F-budget in AGB stars. This will be helpful to quantify the role of AGB stars in the Galactic F budget.

We present a new analytical method, which allows the simultaneous analysis of fluorine (F), chlorine (Cl), bromine (Br), and iodine (I) in geological samples. To account for interferences of Fe on the spectral lines of F, of Al on Br-lines, and of Ca on I-lines, we prepared four new halogen-free calibration glasses. The new method is used to analyze various glass reference materials and crystal-hosted melt inclusions from the Azores. Our results show that our new method allows reliable and reproducible analyses of all four halogens in silicate glasses.

Khademite, ideally Al(SO4)F(H2O)5, from the Monte Arsiccio mine, Apuan Alps, Tuscany, Italy, has been characterised through quantitative electron microprobe analysis, micro-Raman spectroscopy and single-crystal X-ray diffraction. Khademite occurs as colourless to whitish tabular crystals, up to 5 mm. Electron microprobe analysis (in wt.%, average of 20 spot analyses) gave: SO3 35.43, Al2O3 21.27, F 6.92, H2Ocalc 39.73, sum 103.35, –O = F 2.92, total 100.43. On the basis of 10 anions per formula unit, assuming the occurrence of 5 H2O groups and 1 (F+OH) atom per formula unit, its chemical formula can be written as Al0.96S1.02O4[F0.84(OH)0.16]Σ1.00⋅5H2O. The Raman spectrum of khademite is characterised by the occurrence of vibrational modes of SO4 groups and by broad and strong bands due to the O–H stretching modes. Khademite is orthorhombic, space group Pcab, with unit-cell parameters a = 11.1713(2), b = 13.0432(3), c = 10.8815(2) Å, V = 1585.54(5) Å3 and Z = 8. The crystal structure refinement converged to R1 = 0.0293 on the basis of 2359 unique reflections with Fo > 4σ(Fo) and 152 refined parameters. The crystal structure of khademite is characterised by the alternation, along b, of two distinct kinds of {010} layers, one formed by [001] rows of isolated Al-centred octahedra, connected to each other through H bonds, and the other showing isolated SO4 groups. Along b, oxygen atoms belonging to SO4 groups act as acceptor of H bonds from H2O groups coordinating Al atoms. The new data improved the description of the H bonds in khademite and led us to discuss about the possible existence of its (OH)-analogue, rostite. In addition, Raman spectroscopic data were collected on the same crystal used for the crystal-chemical characterisation, allowing a comparison with previous results.

The fluorine route is thoroughly investigated for the hydrothermal synthesis of montmorillonite in the Na2O-MgO-Al2O3-SiO2-H2O system. Using the optimal conditions suggested by Reinholdt et al. (2001) for the crystallization of pure montmorillonites with the formula Na2x(Al2(1-x)Mg2x☐)Si4O10(OH)2, several parameters (x, Mg content, duration of crystallization, F/Si atomic ratio, pH, nature of counterbalance cation) are varied independently from their ideal values. The products are analysed by various techniques (X-ray diffraction, thermogravimetric analysis-differential thermal analysis, 29Si, 27Al and 19F magic angle spinning-nuclear magnetic resonance). It appears that a pure montmorillonite can only be obtained within a narrow x range (0.10 ≤ x ≤ 0.20). The presence of F in the starting hydrogel and the crystallization time also have significant effects on the purity of the final products. It is shown that a small amount of fluorine is needed for the crystallization of pure montmorillonite phyllosilicates.

In reviewing the distribution of fluorine in igneous rocks it is clear that F abundance is related to alkalinity and to some extent to volatile contents. Two important F-bearing series are recognized: (1) the alkali basalt—ultrapotassic rocks in which F increases with increasing K2O and decreasing SiO2 contents; and (2) the alkali basalt—phonolite—rhyolite series with F showing positive correlation with both total alkalis and SiO2. Detailed studies of series (1) show that F abundance in ultrapotassic magmas (lamproite, kamafugite, lamprophyre) occurs in descending order in the sequence phlogopite>apatite>amphibole>glass. Fluorine contents in the same minerals from fresh and altered mantle xenoliths may be several orders of magnitude less than those in the host kamafugite. For many lamproites, F contents correlate with higher mg# suggesting that F is highest in the more primitive magmas.

Experiments at mantle conditions (20 kbar, 900–1400°C) on simplified F-bearing mineral systems containing phlogopite, apatite, K-richterite, and melt show that F is generally a compatible element. Additionally, low F abundance in minerals from mantle xenoliths suggests that F may not be available in mantle source regions and hence is unlikely to partition into the melt phase on partial melting. Melting experiments on the compositions of F-free and F-bearing model phlogopite harzburgite indicate that even small variations in F content produce melts similar in composition to those of lamproite.

A magmatic calcic amphibole (M4Ca≥1.50) characterized by large Sr and F contents (SrO>1 wt.%, F>2.5 wt.%) is described. According to the crystal-chemical formula, the amphibole can be classified as a strontian fluoro-magnesiohastingsite and the presence of Sr cations in the A sites is suggested. The amphibole occurs in the groundmass of ultrapotassic lavas from the Alban Hills Volcanic District (Central Italy). This peculiar chemical composition is due to the lavas' groundmass compositional trend occurring under low silica-activity conditions. We suggest that the occurrence of F amphiboles allows us to consider the Alban Hills Volcanic District magmas to be as rich in F as other ultrapotassic magmas.

Melt generation during granulite-grade metamorphism is believed to be controlled by the stability temperatures of biotite, whose breakdown provides H2O and controls fluid-absent melting in the lower crust. In a simple KMASH system, the restite minerals crystallising due to incongruent melting of phlogopite depend upon the bulk composition. In an alumina-poor and silica-rich portion of the system (Phl + Qtz), enstatite appears with the melt, while in an alumina-rich system (Phl + Sil + Qtz) cordierite appears first instead of enstatite. Since the temperature of biotite stability is believed to be strongly controlled by its F and Ti content, it will have significant effect on the fluid-absent melting reactions during granulite-grade metamorphism of mica-containing granites as well as pelitic rocks in the deeper crust.

To understand such effects in an aluminous portion of the KMASH system, experiments were performed (between 850 and 1100°C and at 7, 10 and 12 kbar) with bulk composition containing 2Phl-6Sil- 9Qtz, where natural phlogopite with F/(F+OH) = 0.39 and Mg/(Mg+Fe) = 0.96 was used. In runs with this charge and containing 5 wt.% of excess water, cordierite appeared around 920°C at 7 kbar and 990°C at 12 kbar, and it disappeared at about 1080°C with the appearance of 221 sapphirine. In fluid-absent runs, these boundaries marginally shift to higher temperatures (30-50°C). The enstatite which was distinctly absent in H2O-saturated runs, crystallises in the high-temperature sapphirine field with up to 12 wt.% Al2O3 in H2O-undersaturated runs. The enstatite formation with cordierite is perhaps inhibited due to the Al consumption by cordierite and instability of Al-free enstatite at temperatures of cordierite stability. Re-equilibrated phlogopite persists in both the cordierite and sapphirine fields. The temperatures of the beginning of phlogopite breakdown are about 100-140°C above those reported for reaction Phl + Qtz → En + Sa + L (Vielzeuf and Clemens, 1992) with F and Ti-free phlogopite, but are ≈50–100°C lower than the temperatures reported (Tareen et al., 1995; Dooley and Patino Douce, 1996) for the same reaction containing F- and Ti- bearing phlogopite. The combined effect of the F and Ti content in phlogopite on its stability temperatures in the KMASH system has been found to be additive in relation to those containing only F or Ti. H2O-saturated runs produced per-aluminous melts with ≈27 wt.% Al2O3 in the cordierite field and ≈23% Al2O3 in the sapphirine field. The H2O-undersaturated runs produced melts rich in K2O (≈10 wt.%), SiO2 (72.5 wt.%) and relatively poor Al2O3 (12 wt.%).

Humite group minerals occur in the marbles of granulite grade at Ambasamudram, southern India. Detailed mineralogical and mineral chemical characterization indicate the mineral is a flourine-rich titanian-poor variety of clinohumite. Average MTi/Si values of 2.22 with typical XRD pattern indicate the mineral to be clinohumite. Petrological constraints on the clinohumite formation shows a high temperature (>700°C) and low aco2 during its formation. The fluorine content of these clinohumites is the highest reported in any environment, with F/(F+OH) ratio reaching a value of 0.70. The high fluorine content reflect the high P-T condition of formation. The OH content of the clinohumites is around 0.59 mole fraction. Preliminary FTIR spectra of the clinohumites show eight sharp absorption peaks between wave numbers 3700 and 3400 cm−1 and a broad absorption band with a peak at 3840 cm−1. The sharp peaks are due to OH in the clinohumite. The high fluorine content of the Ambasamudram clinohumites possibly resulted from the isochemical reactions involving (OH-F) silicates such as amphiboles or phlogopites. The internal fluid buffering is also supported by the stable isotope as well as the petrological studies of the marble assemblages.

Several K-Fe sulphide minerals have been described from natrocarbonatite lavas erupted from the Tanzanian volcano Oldoinyo Lengai but uncertainty remains about their exact identity. They do not appear to be any of the established K-Fe sulphides according to Dawson et al. (1995) and Mitchell (1997). Here, we describe yet another variant which is a Mn- and F-bearing variety of rasvumite. It appears to have formed by reaction of Fe alabandite grains with natrocarbonatite magma and occurs as minute rims on the alabandite. Its formula is: (K0.94Na0.06)(Fe1.91Mn0.13Ca0.01Sr0.01)(S2.06F0.24Cl0.04).

Three Angolan carbonatites were selected to evaluate the change in composition of pyrochlores during magmatic evolution: the Tchivira carbonatites occur in a plutonic complex, the Bonga carbonatites represent hypabyssal carbonatites and the Catanda carbonatites are volcanic in origin. In Tchivira pyrochlore, zoning is poorly developed; fluorine is dominant at the Y site; chemical zoning may arise as a result of substitutions for Nb in the B site; and the rare earth element (REE), U, Th and large-ion lithophile element (LILE) contents are very low. Pyrochlores from Bonga show oscillatory zonation; the F and Na contents are lower than those in the pyrochlores from Tchivira; and as substitution of Na at the A site increases, the Th, U, REE contents and inferred vacancies also increase. Pyrochlores from Catanda display complex textures. They generally have a rounded corroded core, which is mantled by two or three later generations. The core composition is similar to the Bonga pyrochlores. The rims are enriched in Zr, Ta, Th, Ce and U, but depleted in F and Na. In pyrochlores from the Angolan carbonatites, the F and Na contents decrease from plutonic to volcanic settings and there is enrichment of Th, U and REE in the A site and Ta and Zr in the B site. Zoning may be explained by changes in the activity of F, due to the crystallization of fluorite or apatite in the plutonic and hypabyssal carbonatites, or to volatile exsolution in the volcanic carbonatites.

Euhedral crystals of complexly zoned niobian titanite (up to 0.3 mm) are enclosed in hedenbergite (Hd53—81Di15—43Jh3-5) and quartz from a hedenbergite vein skarn at Kamenne doly near Pisek, Czech Republic. They are associated with minor clinozoisite-epidote (Ps3—22), calcite, plagioclase (An95). scapolite (Me80—82), scheelite, pyrrhotite, fluorapatite, arsenopyrite, native bismuth and Bi,Te-minerals. The following textural and compositional subtypes were recognized: (I) Nb-rich titanite, (II) Nb-moderate titanite in the central zone, (III) Nb-poor, Sn-enriched titanite and (IV) Nb-poor, Al,F-rich titanite in the outer zone. The substitution Al(Nb,Ta)Ti—2 is dominant in subtypes I and II, the titanite subtype I being characterized by elevated contents of Al ≤ 0.257 atoms per formula unit (a.p.f.u.), Nb (≤ 0.161 a.p.f.u.) and Ta (≤ 0.037 a.p.f.u.). Amounts of Al, Nb and Ta in subtype II are smaller and more variable. The minor substitution SnTi—1 occurs chiefly in titanite subtype III with a content of Sn ≤ 0.039 a.p.f.u.. The substitution Al(F,OH)(TiO)_i is typical for titanite subtype IV exhibiting elevated contents of Al (s£ 0.221 a.p.f.u.), F (≤ 0.196 a.p.f.u.) and Fe (≤ 0.039 a.p.f.u.).

The negative relationship of substitutions Al(F,OH)(TiO)_i vs. SnTi—1 and Al(Nb,Ta)Ti—1 is constrained chiefly by crystal structure rather than by the composition of parent medium alone. Textural relations suggest that the Nb-moderate titanite in the core zone and entire outer zone are products of fluids-induced dissolution-reprecipitation processes. The studied niobian titanite represents a new F-enriched type from a medium-grade, calc-silicate rock.