Journal of Petrology Advance Access originally published online on February 9, 2005
Journal of Petrology 2005 46(5):1013-1044; doi:10.1093/petrology/egi009
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Petrographic, Chemical and B-Isotopic Insights into the Origin of Tourmaline-Rich Rocks and Boron Recycling in the Martinamor Antiform (Central Iberian Zone, Salamanca, Spain)
1 DEPARTAMENTO DE MINERALOGÍA Y PETROLOGÍA, UNIVERSIDAD DEL PAIS VASCO, 48080 BILBAO, SPAIN
2 DEPARTAMENTO DE MINERALOGÍA Y PETROLOGÍA, UNIVERSIDAD DE GRANADA, 18002 GRANADA, SPAIN
3 STATE KEY LABORATORY FOR MINERAL DEPOSITS RESEARCH, DEPARTMENT OF EARTH SCIENCES, NANJING UNIVERSITY, NANJING 210093, CHINA
RECEIVED JANUARY 16, 2004; ACCEPTED DECEMBER 8, 2004
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONGEOLOGICAL SETTINGANALYTICAL TECHNIQUESTOURMALINE OCCURRENCES AND...WHOLE-ROCK CHEMISTRYMINERAL CHEMISTRYBORON ISOTOPESDISCUSSIONSUPPLEMENTARY DATAREFERENCES |
|---|
Tourmaline in the Martinamor antiform occurs in tourmalinites (rocks with >15–20% tourmaline by volume), clastic metasedimentary rocks of the Upper Proterozoic Monterrubio formation, quartz veins, pre-Variscan orthogneisses and Variscan granitic rocks. Petrographic observations, back-scattered electron (BSE) images, and microprobe data document a multistaged development of tourmaline. Overall, variations in the Mg/(Mg + Fe) ratios decrease from tourmalinites (0·36–0·75), through veins (0·38–0·66) to granitic rocks (0·23–0·46), whereas Al increases in the same order from 5·84–6·65 to 6·22–6·88 apfu. The incorporation of Al into tourmaline is consistent with combinations of x
Al(NaR)–1 and AlO(R(OH))–1 exchange vectors, where x represents X-site vacancy and R is (Mg + Fe2+ + Mn). Variations in x/(x + Na) ratios are similar in all the types of tourmaline occurrences, from 0·10 to 0·53, with low Ca-contents (mostly <0·10 apfu). Based on field and textural criteria, two groups of tourmaline-rich rocks are distinguished: (1) pre-Variscan tourmalinites (probably Cadomian), affected by both deformation and regional metamorphism during the Variscan orogeny; (2) tourmalinites related to the synkinematic granitic complex of Martinamor. Textural and geochemical data are consistent with a psammopelitic parentage for the protolith of the tourmalinites. Boron isotope analyses of tourmaline have a total range of 
11B values from –15·6 to 6·8
; the lowest corresponding to granitic tourmalines (–15·6 to –11·7) and the highest to veins (1·9 to 6·8). Tourmalines from tourmalinites have intermediate 11B values of –8·0 to +2·0. The observed variations in 11B support an important crustal recycling of boron in the Martinamor area, in which pre-Variscan tourmalinites were remobilized by a combination of mechanical and chemical processes during Variscan deformation, metamorphism and anatexis, leading to the formation of multiple tourmaline-bearing veins and a new stage of boron metasomatism. KEY WORDS: tourmalinites; metamorphic and granitic rocks; mineral chemistry; whole-rock chemistry; boron isotopes
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTINGANALYTICAL TECHNIQUESTOURMALINE OCCURRENCES AND...WHOLE-ROCK CHEMISTRYMINERAL CHEMISTRYBORON ISOTOPESDISCUSSIONSUPPLEMENTARY DATAREFERENCES |
|---|
Minerals that combine an extensive compositional range with chemical zoning can provide valuable information for understanding the degree of solid solution and possible substitution mechanisms, the nature of reactions involving these minerals, the physical–chemical conditions of formation and the genetic environment. This is the case for minerals of the tourmaline group. Because of its refractory nature and occurrence in a wide variety of geological environments, tourmaline has received great interest in recent years and is believed to be a useful petrogenetic indicator and a potential exploration tool for ore genesis (Henry & Guidotti, 1985
; Henry & Dutrow, 1996; Slack, 1996; Jiang, 2001). Experimental studies document a large P–T–X stability range for tourmaline, from surface conditions to granulite facies and anatexis (e.g. Bénard et al., 1985; Vorbach, 1989; Henry & Dutrow, 1996; Werding & Schreyer, 1996; von Goerne et al., 1999). However, the stability of tourmaline is critically dependent upon the composition of the coexisting fluid phase (Frondel & Collette, 1957; Weisbrod et al., 1986; Morgan & London, 1989).
Tourmaline is the main host for boron in crustal rocks and can play an important role in the behavior of boron during metamorphism. In sedimentary and metamorphic rocks, boron abundances reflect interactions between the coexisting fluid phase, sheet silicates and tourmaline (Shaw, 1996). Clay minerals are the main carrier of boron in low-grade metamorphic conditions (Henry & Dutrow, 1996), but dehydration reactions may lead to boron depletion during prograde metamorphism (Moran et al., 1992; Bebout et al., 1993). Tourmaline, however, tends to remain unreactive up to the highest grades of metamorphism and provides an important control on boron concentration in anatectic melts (London et al., 1996). The genesis of B-rich anatectic melts is expected to depend on the thermal history and metamorphic grade of the source rocks with respect to the stabilities of B-bearing host minerals (Leeman & Sisson, 1996). However, B-bearing magmas may not ultimately yield tourmaline granites because the boron may be lost from the granite in late-stage fluids to the proximal country rocks where tourmaline-rich rocks can be developed. Tourmaline is typical of granite-related hydrothermal systems in which boron and other volatile components are involved in a number of processes, including differentiation, fluid phase evolution, wall-rock alteration, and metal transport and deposition (e.g. Pollard et al., 1987). However, whether the association of boron with metals such as Sn and W results directly from the influence of boron or is simply a consequence of geochemical processes is a matter of debate (London et al., 1996). The presence of B and other volatiles such as F and P in granitic melts causes a depression of both the solidus and liquidus temperatures, and an increase of the solubility of H2O in the melt (Manning, 1981; London et al., 1993; Dingwell et al., 1996). Consequently, such fluxing components enhance the solubility of Sn, W, Nb, Ta and related incompatible elements, thus promoting the concentration of metal species in residual melts and the formation of magmatic-hydrothermal ore deposits (Pollard et al., 1987).
Tourmaline is abundant in the Martinamor antiform of western Spain, where tourmaline-rich rocks related to clastic metasedimentary rocks, veins of different styles, tourmaline-bearing granitic rocks, pegmatites and W (± Sn) mineral deposits occur. Tourmaline in these occurrences has been described in relation to mica schists (Martínez-García & Nicolau, 1973), granites (Gonzalo et al., 1975), scheelite mineralization (Duong et al., 1980) and tourmaline-rich rocks adjacent to pegmatites (García de Figuerola et al., 1983). However, there are no detailed studies on the origin and significance of tourmaline-rich rocks in this area. The geological setting and the relationship between the tourmaline-rich rocks and tectonic events support the formation of tourmaline both before and during Variscan deformation and metamorphism. Hence, the presence of tourmaline in a variety of rock types suggests an important spatial and temporal redistribution of boron. In this paper, the authors report petrographic, mineralogical and geochemical data on different occurrences of tourmaline in the Martinamor antiform (Central Iberian Zone, Salamanca). Interpretation of the major and trace-element compositions of the tourmalines and their host rocks, combined with field and petrographic observations and the boron isotopic compositions of the tourmaline, suggest that boron was recycled from pre-Variscan tourmalinites by metamorphic processes and anatexis during the Variscan orogeny. The authors believe that these pre-Variscan tourmalinites formed during the Cadomian period by the interaction of psammopelitic rocks with granite-derived B-rich magmatic fluids.
|
GEOLOGICAL SETTING |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTINGANALYTICAL TECHNIQUESTOURMALINE OCCURRENCES AND...WHOLE-ROCK CHEMISTRYMINERAL CHEMISTRYBORON ISOTOPESDISCUSSIONSUPPLEMENTARY DATAREFERENCES |
|---|
The Martinamor antiform is a small area within the larger Central Iberian Zone, which is located on the northern margin of the Domain of Vertical Folds (Díez Balda et al., 1990
) about 15 km south of Salamanca (Fig. 1). Three main lithological assemblages crop out in this area: (1) a sequence of pre-Ordovician stratified rocks belonging to the Schist–Greywacke Complex (SGC) of Upper Precambrian to Lower Cambrian age (Díez Balda, 1986); (2) granitic rocks; (3) Tertiary sedimentary rocks (Fig. 1). Strata of the SGC were grouped into the Monterrubio and Aldeatejada formations by Díez Balda (1986). The former unit includes psammopelites, quartzites and conglomerates of volcanic affinity (‘porphyroids’), with subordinate calc-silicate rocks, whereas the latter consists chiefly of pelitic rocks. The tourmaline-rich rocks occur in the Monterrubio Formation.
|
The tectonic evolution of the Martinamor antiform is complex. According to Díez Balda (1986)
and Díez Balda et al. (1995), three main phases of Variscan deformation (D1, D2 and D3) are superimposed on the rocks, accompanied by regional metamorphism. Although D1 gave rise to the most important folds in the Domain of Vertical Folds, in the Martinamor antiform, the effects of D1 are significantly masked by D2, whose structural features and distribution are closely related to the metamorphic zonation. D2 is characterized by the development of isoclinal to sheath folds with a subhorizontal penetrative foliation (S2). It has been related to ductile-shearing processes of extensional origin that account for the mylonitic foliation and C–S fabric in the granitic rocks (Díez Balda et al., 1995). D3 produced open, upright folds and is believed to be responsible for the main antiformal structure of the Martinamor area. During D1 and D2, the rocks followed a clockwise P–T path that reached pressures of >4 kbar at temperatures of 600–700°C, followed by a nearly isothermal decompression during D2, leading to low-pressure conditions (
2 kbar) recorded by the stability of andalusite. Later fluid infiltration produced retrograde effects responsible for the transformation of biotite to white mica and chlorite. A more detailed description of the structural and metamorphic evolution of the region has been given by Díez Balda (1986), Díez Balda et al. (1990, 1995).
The granitic rocks, which mostly crop out in the eastern part of the antiform within the Monterrubio formation, have been described in several studies (e.g. Gonzalo et al., 1975; Saavedra et al., 1984). Two groups are recognized: (1) the biotitic orthogneiss of San Pelayo; (2) the Martinamor complex (Fig. 1). The orthogneiss of San Pelayo has a strong foliation parallel to S2. Based on geological and available geochronological data [Rb–Sr whole-rock ages of 430 ± 30 Ma (Linares et al., 1987)], Díez Balda et al. (1995) argued in support of a Cadomian age for their intrusion. The Martinamor complex [Rb–Sr and K–Ar whole-rock ages of 360 ± 25 Ma and 334 ± 10 Ma, respectively (Linares et al., 1987)] comprises a swarm of lenticular tourmaline-bearing leucogranitic and pegmatite bodies, synkinematic to D2, which alternate with metamorphic rocks parallel to S2. Within the same area, a small granitic stock (1 km2), the two-mica granite of Sta. Genoveva intruded during D3. Galibert (1984) reported a zircon U–Pb age of 298 ± 28 Ma for this granite. Abundant quartz veins occur throughout the antiform.
|
ANALYTICAL TECHNIQUES |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTINGANALYTICAL TECHNIQUESTOURMALINE OCCURRENCES AND...WHOLE-ROCK CHEMISTRYMINERAL CHEMISTRYBORON ISOTOPESDISCUSSIONSUPPLEMENTARY DATAREFERENCES |
|---|
Chemical analyses of 30 representative bulk-rock samples (metasediments, tourmalinites, orthogneisses and granites) from the Martinamor antiform and 14 tourmaline mineral separates from tourmalinites, granites and quartz veins were performed at Granada University. Rock samples were crushed in a steel jaw crusher and pulverized in a tungsten carbide shatterbox. Major elements were determined by X-ray fluorescence (XRF) using a Philips PW1404/10 X-ray spectrometer by fusing with lithium tetraborate and casting into glass discs. The trace element Zr was also measured by X-ray fluorescence using pellets of pressed rock powder. X-ray counts were converted into concentrations by a computer program based on the fundamental parameters method of de Jongh (1973)
. Precision was of 2–5% for major elements, except Mn and P (5–10%), and 2–5% for Zr. Trace elements, except Zr, were determined by inductively coupled plasma–mass spectrometry (ICP–MS), after HNO3 + HF digestion of 100 mg of sample powder in a Teflon-lined vessel at 180°C and 200 p.s.i. for 30 min, evaporation to dryness and subsequent dissolution in 100 ml of 4 vol. % HNO3. Precision is better than ±5% for analyte concentrations of 10 ppm. Calibration was done with international standards PM–S and WS–E (Govindaraju et al., 1994), dissolved in the same way (see Torres-Ruiz et al., 2003). Tourmaline separates were prepared using the procedure of Pesquera & Velasco (1997). The purity of these separates is estimated to be >99%; two separates (samples 3 and 4, Table 1) have accessory minerals such as zircon, xenotime and/or monazite. Selected tourmalines were analyzed for their boron isotopic composition following the analytical procedure described by Jiang et al. (2003). During the analytical period, about 20 analyses of the NIST SRM 951 standard yielded an average 11B/10B ratio of 4·0089 ± 0·7 (2
).
|
Mineral compositions were determined at the University of Granada with a Cameca SX50 electron microprobe, equipped with four wavelength-dispersive spectrometers, using both natural and synthetic standards: natural fluorite (F), natural sanidine (K), synthetic MnTiO3 (Ti, Mn), natural diopside (Ca), synthetic Fe2O3 (Fe), natural albite (Na), natural periclase (Mg), synthetic SiO2 (Si), synthetic Cr2O3 (Cr) and synthetic Al2O3 (Al). An accelerating voltage of 20 kV and beam current of 30 nA, with a 1–2 µm focused electron beam, were used to analyze tourmaline and associated minerals. Counting times on peaks were twice those of backgrounds, with 15 s for Na and K; 20 s for Ti and Ca; 25 s for Fe, Si and Al; and 30 s for Mg. Data were reduced using the procedure of Pouchou & Pichoir (1985)
. Analytical errors are estimated to be on the order of ±1–2% for major elements and ±10% for minor elements. The mineral chemistry database (available as Electronic Appendix, which may be downloaded from the Journal of Petrology website at http://www.petrology.oupjournals.org) includes over 700 analyses of all tourmaline types, and over 300 analyses of associated minerals (micas, garnet, apatite, rutile and ilmenite). Five tourmaline samples were also analyzed by Mössbauer spectroscopy to assess the iron oxidation state. The Mössbauer analyses of powdered tourmaline samples were carried out in the Electricity and Electronic Department of the University of the Basque Country. Mössbauer spectra were performed at room temperature using a standard spectrometer with Co–Rh source. The isomer shift is reported relative to metallic iron. All of the spectra were collected in a multi-channel analyzer with 512 channels, and the experimental data were evaluated by means of a least-squares fitting.
|
TOURMALINE OCCURRENCES AND PETROGRAPHY |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTINGANALYTICAL TECHNIQUESTOURMALINE OCCURRENCES AND...WHOLE-ROCK CHEMISTRYMINERAL CHEMISTRYBORON ISOTOPESDISCUSSIONSUPPLEMENTARY DATAREFERENCES |
|---|
Tourmaline is widespread throughout the Martinamor antiform. It has been found in three main geological occurrences: (1) tourmalinites; (2) quartz veins; (3) granitic rocks.
Tourmalinites
The tourmalinites represent a conspicuous lithology within the Martinamor antiform. They occur in the Upper Proterozoic Monterrubio Formation and form discontinuous, ESE–WNW-trending stratiform bodies within psammopelitic rocks and quartzites in the staurolite zone (Fig. 1). Scheelite-bearing calc-silicate rocks are associated with tourmalinites in some areas. Unfortunately, the poor outcrop conditions, together with the intense deformation, make it difficult to estimate the strike length and thickness of the tourmalinites. Tourmaline-rich rocks also occur in contact with lenticular granitic and pegmatite bodies of the Martinamor complex.
Tourmalinites are fine-grained (usually <1 mm), dark brown to black rocks, with variable amounts of tourmaline (up to >90% by volume). They are characterized by a prominent layering consisting of a finely laminated sequence in which tourmaline-rich laminae alternate with quartz-rich laminae (Fig. 2a). The layering is coplanar to the bedding and the main foliation (S2). Quartz and variable amounts of plagioclase, muscovite and biotite are the most abundant minerals associated with tourmaline in the tourmalinites. Some layers contain appreciable amounts of apatite (up to 5% by volume) and rutile (up to 2% by volume). Fine-grained, subhedral garnet was found exclusively in a tourmalinite adjacent to a thin albitite layer. In contrast, subhedral to anhedral garnet is relatively common in the surrounding psammopelitic rocks. Small grains of zircon and monazite are accessory minerals.
|
Most of the tourmalinites have been subjected to the regional deformation and metamorphism that affected the Martinamor area. Evidence includes features such as: (1) sheath and isoclinal folds (Fig. 2b and c); (2) mylonites showing tourmaline porphyroclasts (‘augen’ tourmalinite) (Fig. 2d and e); (3) overprinting relations between foliations (Figs 2f and 3a); (4) stretching and boudinage of the tourmaline layers and crystals; (5) fracturing and vein-filling processes (Fig. 2b, d and f); (6) extensive recrystallization. Augen tourmalinite contains tourmaline porphyroclasts, separated by alternating laminae of recrystallized quartz ± plagioclase and tourmaline, the laminae defining the foliation (S2) that wraps the tourmaline porphyroclasts and intrafolial folds (Figs 2d and 3b). It is difficult to distinguish the axial-planar foliation associated with the isoclinal folding and the mylonitic foliation, suggesting that both structures record the same deformational event. A preferred orientation is present in the tourmaline layers, as a result of the elongated shape of the grains. Tourmaline c-axis fabric patterns for the tourmaline laminae indicate an orientation of [001] parallel to the S2 foliation in a fan-like arrangement, and with a gentle inclination relative to the stretching lineation (Fig. 4a).
|
|
In detail, the porphyroclasts are aggregates of numerous (typically >300 grains), very small (8–200 µm) tourmaline crystals that may be associated with quartz ± plagioclase and rutile ± pyrite along grain boundaries. They define a straight to curved internal fabric surrounded by the mylonitic foliation (S2) or in continuity with it. Tourmaline c-axis fabric patterns for the tourmaline porphyroclasts display a preferred orientation of c-axes at a high angle to foliation (Fig. 4b). Some porphyroclasts exhibit deflection-fold and millipede microstructures (Figs 2d and 3c) like those of Bell & Rubenach (1980)
and Passchier & Speck (1994). In sections cut parallel to the mineral lineation and perpendicular to the foliation, some porphyroclasts appear as tourmaline ‘pull-aparts’ wherein quartz ± plagioclase have infilled between segments of tourmaline (Fig. 2d). In this case, the porphyroclasts are up to 6 mm thick and 15 mm long. Tourmaline c-axes are mostly oriented perpendicular to the long dimension of the ‘pull-aparts’. Porphyroclast-forming tourmaline may have formed by mechanical breakdown of single porphyroblasts or by coalescence processes. Cataclasis of porphyroblasts is unlikely to lead to a systematic preferred orientation of tourmaline grains. A model for complex garnet porphyroblast development by multiple nucleation and coalescence has been postulated by Spiess et al. (2001). In this case, although the growth and coalescence of tourmaline grains may be favoured along certain orientations, it is difficult to imagine similar fabric patterns resulting from porphyroclast-forming coalescence processes. The authors believe, alternatively, that tourmaline porphyroclasts evolved from old cleavage domains of tourmaline into domains with a lesser degree of alignment or microlithons (Barker, 1990) surrounded by zones of progressive shearing. Tails of tourmaline adjacent to the porphyroclasts that extend along the mylonitic foliation may evolve into lens- or rod-shaped domains of fine-grained tourmaline. Locally, a discrete crenulation cleavage defined by tourmaline-bearing dark seams of stylolitic appearance overprints the main foliation (S2). The dark seams are thought to result from the concentration of insoluble material along dissolution surfaces by solution transfer processes. Recrystallized quartz may occur in the necks of the tourmaline boudins and characteristic reddish-brown tourmaline overgrowths tend to mend the separated boudins. The microstructures developed in the necks are similar to stretching veins in places. Evidence for intracrystalline deformation and recovery in tourmaline comes from the presence of crystalline-bending, deformation bands, undulose extinction and subgrains (Fig. 3d). Aggregates of subgrains in association with new grains of tourmaline may account for dynamic recrystallization processes that, in part, were controlled by subgrain rotation. Likewise, irregular lobes at grain boundaries suggest a bulge-assisted recrystallization mechanism (Fig. 3d). Granoblastic–polygonal textures in which undulose extinction is very rare indicate pervasive static recrystallization (Fig. 3e). Interstitial, atoll to sponge-like poikiloblastic tourmaline in quartzites seems to represent a sequence of consecutive steps during growth along the margins of quartz grains.
Tourmaline-rich rocks in which tourmaline is syn- to post-tectonic with respect to D2, and therefore formed during the Variscan orogeny, are believed to be the result of a new stage of tourmalinization related to the emplacement of granites and pegmatites of the Martinamor complex. Tourmaline crystals are oriented in the foliation plane defined by the preferred orientation of micas, mainly aligned along two perpendicular directions (Fig. 4c).
Under the microscope, some tourmalines show a distinctive optical zoning with subhedral to euhedral, reddish-orange cores and green rims. Most of the tourmaline grains, however, display a weak optical zonation with gradational changes in color from core to rim: changing from greenish or reddish-brown to yellowish-brown. Irregular patches of green to blue color are also observed. Fine-growth lamellae, like those decribed by Taylor & Slack (1984), are observed in some grains and probably represent a relict primary microstructure. Zonation in tourmaline is more clearly evident in SEM back-scattered electron images where more or fewer complex zoning patterns are observed (Fig. 5). Oscillatory and sector zoning, cellular textures, interior zones embayed by outer zones, microfracturing and overgrowths are typical features of tourmaline in the tourmalinites (Fig. 5a–e). Fine-scale zonation is commonly truncated at the corroded edges and perturbed by cellular zones. Similar replacive features provoked by the activity of reactive fluids during the deformation and metamorphism have been described in a tourmalinite vein from the Tauern Window, Eastern Alps (Henry et al., 2002) and in tourmalinites from the Sierra Nevada, Southeastern Spain (Torres-Ruiz et al., 2003). Tourmaline crystals that have undergone stretching and boudinage may show multiple narrow, bluish-green to reddish-brown bands (‘fuzzy bands’), perpendicular to the crystalline elongation (Fig. 5f). In some cases, the back-scattered electron images reveal possible coalescence phenomena in which tourmaline grains include two or more subhedral to subrounded cores, mantled by a subhedral outer zone.
|
Tourmaline ± quartz veins
An additional component of the tourmalinites involves a variety of tourmaline ± quartz veins with variable shape, orientation and dimension that range from submillimetric to metric scale. With existing data, it is very difficult to reconstruct the sequence of veining, but different generations and styles related to the regional deformation can be recognized. For example, tourmaline occurs in: (1) syn- to late-tectonic veins exhibiting a blocky texture, which truncate fold structures and are themselves unfolded (Fig. 2b); (2) brecciated veins; (3) necks of boudinaged crystals; (4) tensional stepover sites associated with mylonites; (5) slickenfibres in which tourmaline fibres are slightly inclined to the vein margins; (6) fibrous, elongate, blocky and stretched veins (Figs 2f, 3b and f) showing similar characteristics to those described by Bons (2000)
. Elongate blocky veins of tourmaline and quartz show inclusion bands that suggest a crack-seal mechanism for their formation. Irregularly spaced surfaces (40–200 µm) rich in small tourmaline grains occur parallel to the contacts. Needle-shaped crystals of tourmaline are arranged sub-perpendicular to the vein walls and to the successive inclusion bands (Fig. 3f). Accordingly, if each band represents one cycle of cracking and sealing, following the mechanism of Ramsay (1980), this means that the veins record several tourmaline-forming episodes, i.e. a multistage development of tourmaline at vein scale. This is consistent with formation via fluids in successive vein-forming cycles as proposed Yardley & Bottrell (1992) for quartz veins in Connemara (Ireland). Stretched tourmaline veins appear to have formed via multiple crack-seal events, each crack being of the order of a few microns wide. Evidence in support of this comes from the occurrence of stretched crystals displaying a fine-scale optical banding oriented normal to the elongation and parallel to the vein walls, which accounts for the compositional variations. In thin section, orange–brown bands alternate with bluish to pale green bands, with the boundaries between them commonly marked by trails of solid inclusions. Tourmalines from late-tectonic veins commonly show an oscillatory zoning with numerous bands (Fig. 5g).
Gneisses, granites and pegmatites
Tourmaline is relatively abundant in the Martinamor granitic complex. It occurs as black, fine-to-medium-grained crystals (<2 mm) individually disseminated through the granite, but also concentrated in ovoid or lenticular clots (normally <1 cm) with long axes aligned according to the granitic foliation. In thin section, tourmaline crystals are unzoned or weakly zoned with euhedral to anhedral shapes, and are typically embayed by quartz ± feldspar. In back-scattered electron images (Fig. 5h and i), many tourmalines display a zonation characterized by euhedral to subhedral cores and subhedral to anhedral rims. Fine-scale zonation, as thin as a few microns, may be also superimposed on the main zones, which suggests a formation mechanism related to the kinetics of crystal growth more than to sudden changes in P, T or melt composition. Pleochroism varies from pale or bluish-green to green. Small euhedral grains may be enclosed by muscovite and garnet. In the San Pelayo orthogneiss, pale green to green tourmaline is an accessory mineral and the tourmaline clots are absent. It occurs as very fine-grained, subhedral crystals and as irregularly shaped clusters associated with quartz and feldspar. Tourmaline megacrysts surrounded by biotite ± muscovite and feldspars are also present. Small biotite crystals appear partially or completely enclosed by tourmaline.
In tourmaline-bearing pegmatites, tourmaline is an accessory or minor constituent and forms black, fine- to coarse-grained crystals, in a matrix of quartz, feldspar and muscovite. In thin section, the tourmaline crystals exhibit a subhedral shape with green to bluish-green colors. Tourmaline-bearing quartz veins are also commonly associated with the granitic rocks.
|
WHOLE-ROCK CHEMISTRY |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTINGANALYTICAL TECHNIQUESTOURMALINE OCCURRENCES AND...WHOLE-ROCK CHEMISTRYMINERAL CHEMISTRYBORON ISOTOPESDISCUSSIONSUPPLEMENTARY DATAREFERENCES |
|---|
Tourmaline-bearing rocks in the Martinamor antiform include granites, psammopelites, quartzites and tourmalinites. Major- and trace-element data for representative samples of tourmaline-rich rocks, metasediments and granites are given in Tables 1 and 2. According to the classification scheme of Frost et al. (2001)
, the granitic rocks are ferroan, alkali-calcic and peraluminous (ASI = 1·20–1·40), with Fe* (FeOtot/FeOtot + MgO) values of 0·76–0·86 and MALI (Na2O + K2O – CaO) values of 6·88–7·94. The orthogneiss samples also are peraluminous (ASI = 1·37–1·40) but with a magnesian (Fe* = 0·71) and calc-alkalic (MALI = 5·78–6·00) character. Tourmaline-rich rocks show distinctive chemical compositions, with B2O3-contents of 0·82–9·02 wt %. Overall, these rocks have large variations in Al2O3 (16·71–30·55 wt %) and Fe2O3 total (4·5–12·78 wt %). Smaller variations are observed for MgO (2·63–5·98 wt %), CaO (0·20–1·26 wt %), Na2O (0·74–3·16 wt %), K2O (0·04–1·84 wt %) and P2O5 (<1 wt %). Tourmalinites ‘sensu stricto’ are considered to contain in excess of 1·5–2 wt % B2O3 (Slack, 1982). The metasediments have higher SiO2 and K2O, and lower Al2O3, TiO2, Fe2O3 total, MgO, P2O5 and F than the tourmaline-rich rocks (Table 2). Some tourmalinite samples have high CaO (3·34 wt %), P2O5 (2·51 wt %), Na2O (up to 4·86 wt %) or K2O (up to 4·82 wt %) as a result of abundances of apatite, plagioclase or muscovite, respectively.
|
Variation diagrams of major elements plotted against Al2O3 for the metasediments (Fig. 6) show linear trends with high degrees of correlation for Al2O3 vs TiO2 (correlation coefficient, r = 0·98), Fe2O3total (0·94) and MgO (0·97). The linear trends of the data for the metasediments are believed to reflect mixing lines inherited from clay-rich and quartz-rich sedimentary precursors that have been preserved through metamorphism (Argast & Donnelly, 1987
). Positive correlations between TiO2 and Al2O3, and TiO2 versus TiO2/Zr (not shown) suggest that Ti-bearing minerals were bound to the clay fraction of the primary sediments. A negative correlation of TiO2 versus TiO2/Zr would indicate that Ti was preferably transported via heavy minerals during sedimentation (Häussinger et al., 1993). Chemical variations in the tourmalinites show only linear trends, with positive correlations for Al2O3 vs Fe2O3 total (r = 0·77) and MgO (r = 0·88). TiO2 vs Al2O3 exhibits considerable scatter (Fig. 6) taking into account that these elements are traditionally considered immobile during geological processes (Ague, 1991; Slack et al., 1993; Walter, 1997). This may be related to: (1) differences in processes of detrital sedimentation between the protoliths of the tourmalinites and metasediments; (2) the abundance of Ti in tourmaline-rich rocks being strongly affected by the distribution of rutile, whereas Ti in the metasediments is mostly harboured in the micaceous fraction; (3) mobility of Al to some degree in metamorphic fluids, which can contribute to the tourmaline present in the metamorphic veins. This could explain the lack of correlation between Al2O3 and Yb (r = 0·05), and the weak positive correlation between Al2O3 and Th (r = 0·40) (Fig. 6). Instead, better correlations for TiO2 vs Yb (r = 0·43) and Th (r = 0·63) (not shown) denote less mobility of Ti during hydrothermal and metamorphic processes. Evidence for Al or Ti mobility in such processes has been documented by some workers (e.g. Tracy et al., 1983; Gieré, 1990; Kerrick, 1990; Moine et al., 1998). Likewise, the solubility of Ti-bearing minerals appears to be considerable in some halogen-rich fluids (Ayers & Watson, 1993). CaO, Na2O and K2O also lack correlations (not shown) with Al2O3. Broad positive correlations (Fig. 6) between B2O3 and Al2O3 (r = 0·70), Fe2O3total (0·72) and MgO (0·83) are largely controlled by the modal-proportion tourmaline in the analyzed rocks.
|
The trace-element compositions of the tourmaline-rich rocks are characterized by relatively low contents of Rb, Cs and Ba, enrichment of Be, Sr, Sc, V, Cr, Zn, Y, Zr and Ni, and strong enrichment of Sn and W compared with the psammopelites (Table 1). Although fluorapatite is present in the metasediments and tourmaline-rich rocks, the covariation of Al2O3 with B2O3 and F (r = 0·70 and 0·83, respectively) (Fig. 6) indicates that F is also contained in tourmaline and micas, as is evidenced from the microprobe data (Tables 3 and 4). The covariation of F with Sn (r = 0·86) (Fig. 6), W (0·66), Li (0·61) and Be (0·64) suggests a link to granite-derived magmatic-hydrothermal fluids. Whereas Sn and W are mainly accomodated by rutile, Li mostly partitions into micas as indicated by the low Li-contents of the mica-free tourmalinites and the positive correlations of Li with K2O and Ba (r = 0·78 and 0·74, respectively). Partitioning of Li into micas is favored by relatively high mica–fluid distribution coefficient for Li (Volfinger & Robert, 1980
).
|
|
Contents of rare earth elements (REE) are low in the granites compared with the orthogneisses, metasediments and tourmaline-rich rocks (Table 1). The Zr-contents in metasediments and tourmalinites are within a range of values (200 ± 100 ppm) that suggests a minor or insignificant influence of zircon on REE distribution (see McLennan, 1989
). This feature, together with the poor correlation between Zr and HREE (r < 0·40) (not shown) and the covariation of Ca (r = 0·73), Y (r = 0·74) and HREE (r = 0·50–0·60) with P (not shown), could indicate that the REE distribution is dependent on other accessory minerals (e.g. apatite, xenotime and monazite). Based on the significant relationships existing between Ti–Nb (r = 0·72) and Zr–Nb (0·75) (not shown) and the positive correlations among Th, Ce, Y and Yb (Fig. 6), it is probable that the accessory minerals containing these elements originally occurred in the clay-sized fraction rather than in the sand-sized fraction of the sediments (see Caggianelli et al., 1992). Both the metasediments and tourmaline-rich rocks show a good correlation between Y and Yb (r = 0·81) (Fig. 6), which is consistent with a coherent geochemical behavior during hydrothermal and metamorphic processes. On a plot of Ce versus Th (Fig. 6), however, the greater scatter in the data for the tourmaline-rich rocks may be due to variable mobility of Ce, relative to Th, during such processes. Chondrite-normalized plots of REE concentrations are shown in Fig. 7. In general, the granites display small positive Eu anomalies (Eu/Eu* = 1·06–1·16) and low La/LuCN ratios (1·80–6·45), compared with the orthogneisses that have uniformly negative Eu anomalies (Eu/Eu* = 0·41–0·53) and high La/LuCN ratios (13·63–19·38). Most of the tourmalinites display high La/LuCN ratios (5·3–16·2), moderate negative Eu anomalies (Eu/Eu* = 0·54–0·83) and total REE abundances ranging from 79·20 to 286·8. Two tourmalinites have the lowest REE abundances (30·31–42·31) with positive Eu anomalies and REE patterns similar to those of tourmaline separates from veins, suggesting high fluid/rock ratios during tourmalinization (Fig. 7). The positive Eu anomalies may record a hydrothermal fluid component from a reduced, Eu2+-bearing solution at >200–250°C (Bau, 1991). Relative to tourmalinites, the metasediments show a narrow range of REE contents (196·0–220·4), lower Eu/Eu* values (0·44–0·74) and higher La/LuCN ratios (7·3 to 24·6). REE patterns and Y/Ho ratios for the majority of the tourmalinites (av. Y/Ho = 27 ± 1·79) are similar to those of the psammopelites (av. Y/Ho = 26 ± 1·91), which is consistent with a psammopelitic parentage for the protolith of these rocks.
|
|
MINERAL CHEMISTRY |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTINGANALYTICAL TECHNIQUESTOURMALINE OCCURRENCES AND...WHOLE-ROCK CHEMISTRYMINERAL CHEMISTRYBORON ISOTOPESDISCUSSIONSUPPLEMENTARY DATAREFERENCES |
|---|
Tourmaline composition
Average electron microprobe analyses of tourmaline from the Martinamor antiform are grouped in Table 3 according to the host rock, i.e. tourmaline-rich rocks, granitic rocks and veins. Overall, significant variations are observed for FeO (4·66–12·87 wt %), MgO (2·12–8·29 wt %) and TiO2 (0·11–2·66 wt %). Smaller variations occur for SiO2 (34·57–38·15 wt %), Al2O3 (29·52–36·07 wt %), Na2O (1·46–2·41 wt %) and CaO (0·01–1·29 wt %). Amounts of MnO (<0·30 wt % MnO) and K2O (<0·20 wt %) are very low. Fluorine ranges to over 1·23 wt %; contents of Cr and Cl are below detection limit. Some significant components such as Fe3+, Li and H2O cannot be directly determined by electron microprobe analysis.
Tourmaline is a complex borosilicate with a general formula of XY3Z6(T6O18)(BO3)3V3W (Hawthorne & Henry, 1999), where X = Na, Ca, K, vacancy; Y = Mg, Fe2+, Mn, Al, Li, Fe3+, Ti4+, Cr3+, V3+; Z = Al, Mg, Fe3+, Cr3+, V3+; T = Si, Al, (B); B = B, vacancy; V = OH, O; W = OH, F, O. Several normalization procedures for tourmaline are possible (Henry & Dutrow, 1996). In this paper, structural formulae were calculated on the basis of 15 (T + Z + Y) cations, assuming that boron has a stoichiometric value of 3 apfu. This normalization gives a cation charge in excess of 49 that to a first approximation suggests a negligible amount of Fe3+. In general, the microprobe data indicate that most of the tourmalines are Al-saturated and belong to the alkali group (Hawthorne & Henry, 1999), with small amounts of Ca and significant amounts of vacancies in the X site; compositional variations are summarized in terms of an Al–Fe–Mg diagram (Fig. 8). As the concentrations of Li are <190 ppm (Table 5), the tourmalines can be described within Li-poor compositional space in which seven components [schorl (S), dravite (D), uvite (U), feruvite (FV), foitite (F), magnesiofoitite (MF), olenite (L)] are required to specify the chemical variability. End-member molecules were calculated by fitting mineral compositions based on the ideal formulae of IMA-approved natural tourmaline (Table 3) to the tourmaline analyses, assuming
 | (1) |
 | (2) |
 | (3) |
 | (4) |
 | (5) |
 | (6) |
 | (7) |
 | (8) |
|
|
Tourmaline chemistry in tourmaline-rich rocks
Tourmaline compositions from tourmalinites are subdivided depending on the dominant mineral assemblage, i.e. tourmaline + quartz (TQ), tourmaline + plagioclase + quartz (TPQ) and tourmaline + mica (TM) (Table 3). In general, the tourmalines show variable X/(X + Na) ratios of 0·10–0·51 and Mg/(Mg + Fe) ratios of 0·36–0·75, both ratios increasing and decreasing, respectively, from TM to TQ assemblages (Table 3, Fig. 9). Tourmaline from TQ and TPQ assemblages contain enough Al to fill the Z site. By contrast, many tourmalines from TM assemblages have an Al-deficit and relatively high Ti-contents, which reflect the influence of the uvite component and the substitution TiRAl–2 where R is (Mg + Fe2+ + Mn) (Table 3, Fig. 9). The lower Al-contents in the TM tourmalines could also be controlled by some substitution of Fe3+ in accordance with the FeAl–1 exchange vector. However, the fact that the average charge excess of the TM tourmalines is similar to the TQ and TPQ tourmalines, together with the absence of compositions with charge deficiency on the xs charge vs (Fe + Mg + X) diagram (Medaris et al., 2003
) (Fig. 9), are not consistent with the presence of significant amounts of Fe3+. Mössbauer spectra support this contention. The spectra show three Fe2+ doublets that are considered by Dyar et al. (1998) to represent Fe2+ in three distinct types of Y sites (Y1, Y2 and Y3) (Fig. 10). The proportion of alkali-free tourmaline (foitite + magnesiofoitite) increases from TM (av. 15·18 mol %) to TQ (av. 34·68 mol %), whereas the average olenite proportion varies from 2·24 mol % in TM to 6·23 mol % in TPQ. In the TQ and TPQ tourmalines, the xAl(NaR)–1 and AlO(R(OH))–1 exchange vectors, where x represents X-site vacancy, appear to be the dominant mechanisms of Al-incorporation, judging from the Altot vs X-site vacancy diagrams (Fig. 9). By contrast, the very shallow slope of the TM data in the Altot vs X-site vacancy diagram (Fig. 9) denotes a minor importance of the Al(NaR)–1 exchange vector. Tourmalines from TM show the highest Mg/(Mg + Fe) ratios and Ca-contents. Accordingly, they have the largest dravite (av. 41·88 mol %), uvite (12·14 mol %) and feruvite (av. 5·84 mol %) components, each decreasing from TM to TQ assemblages (Table 3). Prevailing mechanisms of Ca-incorporation in TM tourmalines are consistent with one or more of the substitutions CaMg(NaAl)–1, CaO(Na(OH))–1 and CaMg2OHNa–1Al–2O–1 (Fig. 9). In TQ and TPQ tourmalines, the incorporation of Ca could also be controlled by other mechanisms, such as the substitutions CaMg–1Al–2 and CaMgO[(Al(OH)]–1 (Fig. 9).
|
|
values are believed to represent Fe2+ in Y sites (Y1 and Y2), and the Fe2+ doublet with the lower Tourmaline records some compositional change as a result of regional deformation and metamorphism (Table 6). Porphyroclasts have more Al (av. 6·46 apfu) and less Ti (av. 0·07 apfu), Fe (av. 0·92 apfu), Mg (av. 1·52 apfu) and F (av. 0·16 apfu) than tourmalines defining the S1 foliation (av. 6·39, 0·09, 0·95, 1·54 and 0·18 apfu, respectively). Tourmalines that define the S2 foliation contain lower Mg (av. 1·28 apfu) but higher Ti (av. 0·11 apfu), Fe (av. 1·17 apfu) and F (av. 0·20 apfu) than the tourmaline porphyroclasts and S1-tourmalines. Poikiloblastic, post-S2-tourmalines have lower Al (av. 6·36 apfu) and Mg (av. 1·16 apfu) but higher Ti (av. 0·15 apfu), Fe (av. 1·29 apfu) and F (av. 0·31 apfu). Like poikiloblastic- and S2-tourmalines, the overgrowths that tend to mend the separated boudins of tourmaline have lower Mg (av. 1·25 apfu) but higher Fe (av. 1·30 apfu), Ti (av. 0·16 apfu) and F (av. 0·21 apfu) (Table 6).
|
The tourmalines in the tourmalinites are characterized by three main types of chemical zonation: (1) a discontinuous core to rim zoning where Na, F and Ti increase as Al decreases (common in TQ and TPQ tourmalines) (Figs 5c and 11a); (2) a discontinuous oscillatory zoning defined by steep compositional gradients in Al, Ti, Mg, Fe, Na, Ca and F, however, with non-systematic chemical variations from core to rims of the crystals (Figs 5d and 11b); (3) compositional fluctuations, particularly in Al, Na, Mg, Ti and F, associated with overgrowths and ‘fuzzy bands’ in elongated and boudinaged tourmaline crystals (Figs 5e and f, and 11c and d). These compositional trends reflect the influence of various substitutions, in particular
AlNa–1R–1, AlOR–1OH–1, TiRAl–2, CaMg(NaAl)–1 and CaMgO(Al(OH))–1.
|
Tourmaline chemistry in veins
Tourmalines from veins are divided into syn-tectonic (STV) and late-tectonic (LTV) groups, based on vein morphology and relationship to regional deformation. Overall, these vein tourmalines display variable Mg/(Mg + Fe) and x
/(x + Na) ratios of 0·38–0·66 and 0·16–0·53, respectively (Fig. 9). STV tourmalines have higher schorl (av. 30·24 mol %) than those from LTV (av. 25·04 mol %). Compared with the TM and TPQ tourmalines, the content of alkali-free tourmaline in both STV and LTV is relatively high (av. 31·06 and 35 mol %, respectively) (Table 3). The olenite component is lower in STV (av. 2·89 mol %) than in LTV (av. 7·31 mol %). Variations in Al are controlled by the Al(NaR)–1 and AlO(R(OH))–1 exchange vectors (Fig. 9), the first being the dominant substitution (
70 and 60% for STV and LTV tourmalines, respectively). Among the minor components, uvite and feruvite together represent, on average, 8·25 mol % in STV and 4·52 mol % in LTV. The mechanisms of Ca-incorporation in these tourmalines are similar to those that operated in TQ and TPQ tourmalines (Fig. 9). In BSE images, LTV tourmalines commonly display chemical zonation (Fig. 5g). Figure 11e shows a typical two-zone pattern in which the outer zone is richer in Fe, Na and F than the inner zone, whereas Mg and Ca show an inverse relation. In addition, an oscillatory chemical zoning involving fluctuations in Fe, Mg, Al and F overlaps the two-zone pattern (Fig. 11e). The compositional variation of STV tourmalines has been evaluated in elongated blocky veins. These appear to have evolved by a crack-seal mechanism and display successive acicular tourmaline-bearing bands (Fig. 3f). The tourmaline through the bands shows significant compositional fluctuations in Al, Fe, Mg, Ti, Na and F.
Tourmaline chemistry in granitic rocks
In granitic rocks, the compositions of tourmalines plot in the schorl field with Mg/(Mg + Fe) and x/(x + Na) ratios of 0·23–0·46 and 0·29–0·52, respectively. In contrast, those of the SP orthogneiss are dravites (Mg/(Mg + Fe) = 0·55–0·63; x/(x + Na) = 0·19–0·35, with relatively high Ti (av. 0·13 apfu) (Fig. 9, Table 3). The amount of Fe as Fe3+ in tourmalines from granites appears to be negligible from crystal chemical considerations and Mössbauer data (Fig. 9). In all cases, the tourmalines have Al > 6 apfu, and the data in the Altot vs X-site vacancy diagram indicate that both the Al(NaR)–1 and AlO[R(OH)]–1 exchange vectors account for the incorporation of Al (Fig. 10). The alkali-defect substitution seems to be more important in the granites (60%) than in the orthogneiss (40%), which is consistent with the higher content of alkali-free tourmaline in granites (av. 36·42 mol %) than in orthogneisses (av. 24·80 mol %). In addition to these substitution schemes, some Tschermak substitution could be operative as the amount of Si is <6·0 apfu (Table 3). Tourmaline in both types of rock, however, has a similar olenite component (av. 12·67 and 10·48 mol %). Orthogneiss-hosted tourmalines are characterized by relatively high dravite contents (av. 33·06 mol %) with appreciable uvite (av. 5·22 mol %) and feruvite (av. 3·22 mol %), whereas those in the granites have averages of 17·64, 1·33 and 2·01 mol %, respectively. The variations in Ca for the orthogneiss tourmalines may be influenced by one or more of the substitutions CaMg–1Al–2, CaMgO[Al(OH)]–1 and CaMg2(OH)–1Al–2 (Fig. 9).
Tourmaline from granitic rocks may show a relatively weak chemical zonation, commonly with a two-zone pattern (Fig. 5g and h). The transition from the core to the outer zone is marked by a discontinuity across which Mg, Ti, Na and F increase as Al and Fe decrease (Fig. 11f). Oscillations in composition, particularly in Fe, Mg, Na and F, may be superimposed on this compositional trend.
Trace elements
Trace-element concentrations of representative tourmaline separates are presented in Table 5. Tourmalines from tourmalinites have Li, Sn, and W in the range of 59·21–178·19 ppm, 67·67–152·52 ppm and 0·33–12·63 ppm, respectively. In general, the amounts of Li and Sn tend to be higher than those reported in the literature for tourmalines from granites (Power, 1968; Neiva, 1974). In contrast, tourmalines from tourmalinites associated with metamorphic rocks in the Sierra Nevada (south-eastern Spain) have higher and lower contents of Li and Sn, respectively (Torres-Ruiz et al., 2003). Except for Rb, Cs, Zn and Ga, tourmalines from tourmalinites have higher Sr, Ba, Sc, V, Cr, Ni, Co, Y, Nb, Zr, Sn, W, Th, U and Pb, compared with the granitic tourmalines. The relatively high contents in Zr, Hf, Th, Y and REE of four tourmalines from tourmalinites (numbers 3, 4, 5 and 7, Table 5) are believed to reflect the presence of zircon, xenotime and monazite in these samples.
Tourmaline from granites displays no, or negative, Eu anomalies with REE-contents similar to the whole rock (Fig. 7). This is consistent with the formation of tourmaline from the melt, which is also in accordance with the textural evidence. It appears that REE prefer melt relative to coexisting aqueous fluids (Flynn & Burnham, 1978), so that if tourmaline crystallized from fluids, it would have a lower REE concentration. Tourmalines that derive from fluids rather than directly crystallizing from silicate melt are poorer in REE relative to corresponding melt-derived tourmaline (Jolliff et al., 1987).
Tourmalines from tourmalinites show two different chondrite-normalized REE patterns (Fig. 7): (a) relatively high REE abundances and patterns similar to those of the whole rock, probably influenced by heavy minerals; (b) low REE abundances and patterns characterized by positive Eu anomalies and flat or slightly positive HREE trends. This may be due to changes in the concentration of fluids during the tourmalinization processes. A similar feature was ascribed by Torres-Ruiz et al. (2003) to a possible contribution of Eu from the hydrothermal fluids. Relative to the whole rock, tourmalines from tourmalinites exhibit: (1) depletion of low-field-strength elements (LFSE) such as Rb, Cs and Ba that are preferentially partitioned into micas and feldspar; (2) higher Li, Sc, V, Zn and Ga, but lower Cr, Ni, Co, Sn, W, Y, and high-field-strength elements (HFSE), including Zr, Hf, Nb and Ta, the latter probably controlled by the distribution of zircon, rutile and monazite; (3) lower HREE-contents.
Most of the trace-element and REE-contents in tourmalines from veins are lower than in tourmalines from tourmalinites and granites (Table 5). It must be noted that tourmaline from veins displays REE patterns, particularly for the light REE, which are similar to those of tourmalines from tourmalinites with lower REE abundances (Fig. 7).
Chemistry of coexisting minerals
Garnets display a variable chemical composition, depending on the host rock (Table 7). Tourmalinite-hosted garnets have the lowest almandine (av. 45·28 mol % Alm) and the highest spessartine (av. 52·24 mol % Sps) contents, whereas garnets from metasedimentary and granitic rocks are higher in almandine (av. 63·79 and 65·22 mol % Alm, respectively) and lower in spessartine (av. 16·47 and 31·19 mol % Sps, respectively). There is also a significant difference in pyrope and grossular components. Garnets from the metasedimentary rocks have the highest pyrope (av. 9·40 mol % Prp) and grossular (av. 10·35 mol % Grs), whereas those from tourmalinites and granites show similar proportions (av. 1·94 and 2·76 mol % Prp; 0·54 and 0·83 mol % Grs, respectively). Metasediment-hosted garnets show normal chemical zoning in which Fe–Mg increases from core to rim, while Ca–Mn decreases—a pattern characteristic in garnets of the greenschist and amphibolite facies. Because garnet is known to fractionate Mn relative to other minerals, the high content of the spessartine component in the tourmalinite-hosted garnets probably is due to the low modal volume of garnet in these rocks.
|
Micas occur in variable amounts in tourmalinites of the Martinamor antiform; muscovite is dominant over biotite. Distinct chemical compositions reflect compositional dependence on the nature of the host rock (Table 4). Muscovite is less aluminous and richer in MgO, FeO and F in tourmaline-rich rocks compared with muscovite in metasedimentary rocks, granites and orthogneiss (Table 4). Tourmalinite-hosted biotites are less aluminous but richer in F than those from the metasedimentary rocks, granites and orthogneiss. Amounts of TiO2 are higher in biotites from orthogneisses than those from granites, tourmalinites and metasedimentary rocks (Table 4). Ti-contents in biotite from tourmalinites and metasedimentary rocks increase with decreasing Mg/(Mg + Fe) ratios, which seem to be controlled by crystal-chemical factors (Henry & Guidotti, 2002
). Temperatures calculated (570–590°C) for biotites from the tourmalinites and metasedimentary rocks, using the Ti-saturation surface method developed by Henry & Guidotti (2002), are consistent with the metamorphic conditions (staurolite zone) in the study area. Fe–Mg partitioning between coexisting garnet and biotite in the metasediments mostly defines a lower range of temperatures (520–560°C), using the calibrations compiled by Reche & Martinez (1996). Such differences in temperature may be due to a combination of factors: (1) specific characteristics of the thermometric methods; (2) differences in pressure because an increase in this variable decreases the Ti-content in biotite (Henry & Guidotti, 2002); (3) Mg–Fe exchange on cooling or changes in Ti by formation of rutile. The influence of pressure is expected to be negligible relative to the 3·3 kbar isobaric saturation surface of Henry & Guidotti (2002), taking into account the pressure regime (2–4 kbar) in the Martinamor area during metamorphism. Fluorine enrichment in muscovite and biotite is expressed as intercept values IV(F), as defined by Munoz (1984). Micas from tourmalinites have lower IV(F) values than those from metasediments, granites and orthogneisses (Table 4). Such values are in the range of intercept values reported in the literature for micas associated with Sn–W–Be mineralizations (see Munoz, 1984). The calculated log(fH2O/fHF) values corresponding to biotites of the tourmalinites and metasedimentary rocks are of the order of 4·0 and 4·5, respectively, for assumed temperatures of 550°C. At these temperatures, the log fugacity ratios for biotites from granites and orthogneisses would be of the order of 4·3–4·4. However, because the F exchange depends on temperature (Munoz, 1984), it is expected that biotites from granitic rocks have lower log(fH2O/fHF) values (4·0–4·2), assuming that they stopped exchanging F at higher temperatures (600–700°C) than the biotites from tourmalinites and metasediments.
Apatite is a common accessory mineral in all rock types from the Martinamor antiform. However, in tourmaline-rich rocks, fine-grained euhedral to skeletal apatite may occur in significant quantities. Overall, Ca/P values range between 1·60 and 1·65, and the Cl-contents are negligible (<0·02 wt %). Amounts of SrO, FeO and MnO in apatites from tourmaline-rich rocks are higher than those of the metasediments, granites and orthogneiss (Table 8). The F-contents of apatites are very similar in all rock types (av. 3·67–3·72 wt % F). These apatite F-contents and the IV(F) parameter for coexisting biotites in granitic rocks are consistent with peraluminous associations, but they define lower temperatures than the solidus temperature (650°C) of peraluminous silicic melts (Sallet, 2000). This may reflect a non-equilibrium relation between apatite and biotite, but, bearing in mind the presence of tourmaline in the Martinamor granite, it is more likely to be an effect of boron on the solidus temperature. In B-bearing systems, quartz and two alkali feldspars may coexist with the melt below 600°C (Dingwell et al., 1996).
|
Rutile is also a minor component in the tourmaline-rich rocks. It mostly occurs as an intergranular, fine-grained (100–300 µm), yellowish-brown to opaque phase, typically containing tiny inclusions of tourmaline, quartz and/or mica. Its composition is characterized by relatively high SnO2 (up to 2·80 wt %) and WO3 (up to 4·70 wt %) in tourmalinites compared with rutile in the metasediments (Table 9), in which these elements are negligible; manganiferous ilmenite (av. 5·06 wt % MnO) is the main Ti-bearing mineral. Nb and Ta are below the detection limit. Tiny grains of Sn- and W-free rutile included in tourmaline are also present in the tourmaline-rich rocks. The Sn- and W-free rutile is believed to be a relict detrital phase, whereas Sn- and W-bearing rutile probably records metasomatic and metamorphic processes.
|
Plagioclase is the only feldspar in the tourmaline-rich rocks, occurring in variable quantities as fine-grained, unzoned crystals, with an albite to oligoclase composition (Ab100–75). Plagioclase in metasedimentary rocks also shows a variable distribution and similar composition.
|
BORON ISOTOPES |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTINGANALYTICAL TECHNIQUESTOURMALINE OCCURRENCES AND...WHOLE-ROCK CHEMISTRYMINERAL CHEMISTRYBORON ISOTOPESDISCUSSIONSUPPLEMENTARY DATAREFERENCES |
|---|
Boron isotope analyses for representative tourmaline separates from the major occurrences are given in Table 10. Overall, the data show a wide range of
11B values, from –15·60 to +6·80. In the Martinamor granite, fine-grained tourmaline has a lower value (–15·60) than the tourmaline from the clots (–11·7), both being in the middle of the range of magmatic tourmalines (Barth, 1993; Jiang & Palmer, 1998). A tourmaline from a discordant pegmatite has a 11B value of –4·5, which falls at the heavier end of the range for pegmatitic tourmalines (Swihart & Moore, 1989; Jiang & Palmer, 1998). Tourmalines from premetamorphic tourmalinites have 11B values that vary from –4·4 to +2, falling in the central part of the range for tourmalines from metapelites (–22 to +22; Swihart & Moore, 1989). The highest 11B values were found in tourmalines from syntectonic veins (+2·60 and +6·80); tourmaline from a late-tectonic vein also has a relatively high 11B value of +1·90.
|
Boron isotope fractionation in tourmaline is mainly controlled by: (1) the composition of the boron source; (2) temperature; (3) fluid/rock ratios; (4) regional metamorphism (Palmer & Slack, 1989
; Jiang, 2001). The main factor governing the behavior of boron isotopes in tourmalines during metamorphism is considered to be fluid activity (Palmer & Slack, 1989). In the Martinamor area, there is clear evidence of fluid activity and, because boron isotope fractionation depends on temperature, it is logical that prograde metamorphism will give rise to isotopic differences between premetamorphic and recrystallized tourmalines (Slack et al., 1993). Taking into account that recrystallized tourmalines in the Martinamor antiform (11B = –4·4 to +2·0) underwent metamorphic conditions consistent with the staurolite zone (
550°C calculated from biotite–garnet geothermometer), we can assume a fractionation of >3 based on the experimental data of Palmer et al. (1992) for the fractionation of 11B/10B between tourmaline and fluid. The fluids released during metamorphism of tourmalinites account for the formation of many quartz–tourmaline veins in the study area. Because the heavy isotope, 11B, is preferentially partitioned into the fluid phase (Palmer & Slack, 1989; Palmer et al., 1992), it is expected that tourmalines from veins will have higher 11B values than those of premetamorphic tourmalinites, in order to balance the fractionation effects of metamorphic recrystallization. For example, two tourmalines from syntectonic veins have 11B values of +2·6 and +6·8, compared with the nearby tourmalinites in which tourmalines have 11B of –1·9 and +2, respectively. Hence, an approximate 11B range of 0–4 for the premetamorphic tourmalinites can be inferred, which falls in the middle of the range for typical continental crust (Palmer & Slack, 1989). Reaction of tourmalinites during anatexis and migration of boron into a melt could explain the presence of tourmaline in the Martinamor granite, which has lower 11B values (–15·6 and –11·7) than the tourmalinite tourmalines (–4·4 to +2·0). Melting a tourmalinite and crystallizing tourmaline during cooling of the melt would produce similar isotopic effects to metamorphic recrystallization of tourmalinite (Slack et al., 1993). If minimal fractionation occurs between melt and tourmaline (Smith & Yardley, 1996), the isotopic composition of magmatic tourmaline should approximate the boron isotopic signature of the parental granitic magma. Depletion of the granitic tourmalines in 11B and the isotopic difference between fine- and medium-grained tourmalines (–15·6 and –11·7, respectively) suggest that 11B was incrementally removed by loss of fluids from the crystallizing magma. The effect of fluid loss on the boron isotopic composition can be evaluated by a Rayleigh process, using the expression of Taylor & Sheppard (1986):
 |
11Bi = initial magmatic 11B value, 11Bf = final magmatic 11B value, F = fraction of boron remaining in the magma, and 
= fractionation factor. Assuming that the isotopic shift between the initial magma (11B = –11·7) and the residual fraction of melt (11B = –15·6) is 4, and assuming a fractionation factor of 3–3·5 at 550–600°C (Palmer et al., 1992), 70–75% of the boron in the original melt was removed into the fluid phase. The late tourmalinite, in which tourmaline has a lower 11B value (–8) than the premetamorphic tourmaline, could represent an additional tourmalinization event caused by boron-rich fluids derived from the Martinamor granite. The isotopic difference (7) between late tourmalinite and late-stage granitic tourmaline implies relatively high fractionation factors, involving separation of fluid from melt at relatively low temperatures (350–400°C). Under these conditions, the loss of boron would have been of the order of 50%. Nonetheless, repeated episodes of phase separation may lead to large isotopic shifts in the residual melt at relatively small fractionation factors (Smith & Yardley, 1996). We are aware that these calculations are a simplification; nevertheless, the boron contents of the Martinamor granite (<0·47 wt % B2O3, Table 2) compared with the values required to attain saturation of silicic melts in tourmaline (>2 wt % B2O3, Wolf & London, 1997) are consistent with significant loss of boron during magmatic evolution of this granite. |
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTINGANALYTICAL TECHNIQUESTOURMALINE OCCURRENCES AND...WHOLE-ROCK CHEMISTRYMINERAL CHEMISTRYBORON ISOTOPESDISCUSSIONSUPPLEMENTARY DATAREFERENCES |
|---|
An important problem in the petrogenetic interpretation of the Martinamor tourmalinites is understanding the relationship of these rocks with the local granites and pegmatites. Deformation–crystallization relationships in some tourmaline-rich rocks indicate that tourmaline grew syn- to post-tectonically with respect to D2, probably linked to the evolution of the Martinamor granitic complex. However, strong evidence for an early, pre-Variscan development of other tourmalinites, which are therefore unrelated to the granites and pegmatites, is provided by field and petrographic observations. The origin of these older Cadomian tourmalinites, however, is not completely clear.
Inferences from mineral chemistry and boron isotopes
(1) Overall, the tourmaline compositions of the Martinamor antiform belong to the alkali group and most fall within the psammopelitic domain (Fig. 9), suggesting an important control by bulk-rock composition. Tourmalines from tourmalinites that plot in the granitoid field may reflect the chemistry and fluid/rock ratios of hydrothermal systems. Compared with the tourmaline in TQ and TPQ tourmalinites, tourmaline in TM tourmalinites shows the highest contents of dravite and uvite + feruvite, and the lowest contents of alkali-free tourmaline. Tourmaline in granites is characterized by low contents of uvite + feruvite with relatively high alkali-free tourmaline and olenite components (Table 3). No significant differences in composition exist between tourmalines in early, Cadomian tourmalinites and those in late, Variscan tourmalinites.
(2) Tourmaline in syntectonic veins is compositionally similar to those of late-tectonic veins and both are also similar to those from TQ and TPQ tourmalinites (Figs 9 and 10, Table 3). These veins, therefore, may have formed under closed-system conditions, assuming, in this case, that a closed system implies less than a meter scale (Cartwright et al., 1994; Oliver & Bons, 2001). An ancillary criterion in support of this is the fact that the vein-forming minerals correspond to those in the wall rocks. In contrast, relatively large quartz–tourmaline veins and less common feldspar-bearing veins are more consistent with open-system conditions, probably related to the precipitation of fluids of strongly magmatic affiliation.
(3) Compositional differences between recrystallized tourmaline grains and tourmaline porphyroclasts suggest that recrystallization took place by boundary migration mechanisms [syndeformational recrystallization of Stünitz (1998)]. Subgrains are observed in the microstructures, but it appears that progressive subgrain rotation cannot explain the compositional differences between porphyroclasts and recrystallized grains (Stünitz, 1998).
(4) Tourmaline in tourmalinites shows different styles of chemical zoning, even in the same sample, which may involve steep compositional gradients in Al, Na, Mg, Fe, Ti and F (Figs 5 and 11). Oscillatory chemical zoning in tourmalines from tourmalinites is probably a premetamorphic feature that reflects mineral growth in open systems. Taylor & Slack (1984) argued that multiple, oscillatory zoning in tourmaline involves a high fluid–rock environment and is generally uncommon in metamorphic rocks. The early generation of tourmaline with oscillatory and sector zoning is very common in geologic settings in which there are dynamically changing hydrothermal fluid compositions (and, consequently, tourmaline compositions) (London et al., 1996; Slack, 1996). The oscillatory zoning is a possible criterion to use in support of extensive metasomatism (e.g. Yardley et al., 1991). Cellular textures, patchy zoning and embayed interior zones in tourmaline (Fig. 5) are thought to reflect invasive, reactive fluids that replaced the tourmaline with a later generation of tourmaline. Intracrystal compositional boundaries and other structural defects are relatively high-energy sites that might have favoured dissolution processes (Vance, 1965). Experimental studies indicate that tourmaline is unstable under alkaline conditions (Frondel & Collette, 1957; Morgan & London, 1989) and von Goerne et al. (2001) demonstrated that the amount of Na in tourmaline depends on both temperature and the Na-content of the coexisting aqueous fluid. At constant Na-concentration in the fluid, the Na-content in the tourmaline should decrease as the temperature increases. Consequently, as Henry et al. (2002) claimed for tourmaline partially replacing pre-existing tourmaline from a vein in the Eastern Alps, the relatively low-Na tourmaline that forms part of some replacement zones, fractures and overgrowths (Figs 5 and 11) could be due to the influx of reactive fluids of alkaline character during the metamorphism. Compositional fluctuations in tourmalines from veins may reflect both cyclicity in fluid pressure related to fault valve action (Renard et al., 2000) and feedback processes associated with chemical disequilibrium (Ortoleva et al., 1987).
(5) The relatively high contents of Li, Be, Sn and W in tourmaline separates from tourmalinites are consistent with formation in a granite-related environment. Tourmalines in tourmalinites show two different REE patterns that suggest differential fluid/rock ratios during metasomatic processes; tourmalines with low REE abundances and positive Eu anomalies are considered to reflect deposition under relatively high fluid/rock conditions in comparison with those that have high REE abundances (Fig. 7). These latter, however, might not be representative because of the influence of heavy mineral fractionations on REE patterns.
(6) Available boron isotope data are limited, but the 11B values give some insights into the origin and evolution of the tourmaline-rich rocks. The observed variations in 11B support an important recycling of boron in the study area, between the Cadomian and Variscan cycles. Additional B-isotopic data are required in order to constrain the boron source of the premetamorphic tourmalinites, and to better understand the main stages of tourmaline formation.
(7) Calculated values for log(fH2O/fHF) based on the F-content of biotite reflect equilibration with moderately F-rich fluids. In principle, the F-intercept values for biotites from tourmalinites (1·46) denote fluids richer in F than those from the granites and metasediments. However, because the halogen content of mica depends on temperature, it is possible that the difference in the log fugacity ratio accounts for different F-exchange temperatures. Indeed, if the biotites from granites stopped exchanging F at a higher temperature (e.g. 600°C) than the tourmalinite biotites, and assuming that they have not been reset by lower-temperature fluids, the log(fH2O/fHF) in both suites of biotites would have comparable values. In contrast, tourmalinites and metasediments underwent similar P–T conditions, with a difference of 0·5 in the log(fH2O/fHF) ratio. This would indicate differences in the relative halogen acid activity during the formation and evolution of the tourmalinites. The F-enrichment, together with the higher Sn- and W-contents in tourmalinites in comparison with the metasediments (Table 1), suggests that F, B, Sn and W are closely related. The F-contents of apatite coexisting with biotite in the granitic rocks suggest that melt fractions may have persisted down to relatively low temperatures, probably caused by abundant boron in the melt.
Bulk compositional constraints
In recent years, different theories have been postulated to explain tourmalinite formation, from syngenetic to epigenetic models [see Slack (1996) for an overview]. At the Martinamor antiform, one can rule out a diagenetic–metamorphic origin on the basis of textural relationships and constraints from bulk-rock boron contents. Metamorphism of clay-rich sediments with relatively high boron (2000 ppm) would yield a maximum of 4 vol. % tourmaline, which represents an insignificant amount compared with the tourmalinites in this area (up to 90 vol. % or more). Likewise, an evaporitic source for the boron seems improbable because of the lack of known evaporites or meta-evaporites in this region. In the Schist–Greywacke Complex, the occurrence of bedded tourmalinites associated with scheelite-bearing layers suggests a cogenetic relationship for B and W, which has been attributed to exhalative processes by Arribas-Rosado (1986). Although the stratiform morphology of the tourmalinite has been considered a characteristic feature of an exhalative origin (Slack, 1996), such a configuration can equally result from selective replacement of favourable layers. In the Martinamor area, it is difficult to support an exhalative model because there is no evidence for exhalites such as coticules (garnet-rich quartzites), iron formations, bedded sulfides, etc. Premetamorphic formation of the tourmalinites by hydrothermal replacement of clastic sediments at or below the sediment–water interface may be a plausible mechanism, based on the similar trends of some major and trace elements and REE patterns for the tourmalinites and psammopelitic metasediments. Moreover, the fact that both types of rocks show similar Y/Ho ratios, close to the chondritic ratio (28), is consistent with this interpretation and indicates a coherent geochemical behavior during hydrothermal and metamorphic processes. Aqueous fluids and their precipitates display non-chondritic Y/Ho ratios (Bau, 1996). However, the high Li-, Sn- and Be-contents in some of the Martinamor tourmaline-rich rocks suggest that tourmalinizing fluids had a large igneous component.
Origin of the tourmalinites
Criteria in support of a contact metasomatic origin are not individually definitive but, in combination, they offer compelling reasons to believe the tourmalinites affected by Variscan deformation and metamorphism are the result of metasomatic alteration of the metasedimentary rocks by externally derived B-rich fluids of magmatic affiliation. Besides boron, geochemical data suggest that Li, Be, F, Sn and W were the main components introduced into the surrounding country rocks, which provided a local source of the Al, Fe, Mg and Ca necessary to precipitate tourmaline from boron-rich fluids. The felsic volcanic rocks (‘porphyroids’) that occur within the metasedimentary sequence are unlikely candidates as a source for the boron because they lack tourmaline and tourmalinization effects on surrounding rocks. A more probable candidate is the San Pelayo granitoid, taking into account that it: (1) is a pre-Variscan peraluminous body, probably Cadomian; (2) crops out in the core of the antiform; (3) contains tourmaline. The fact that the amount of tourmaline in this granitoid is small does not mean that boron in the melt was necessarily scarce, as the crystallization of tourmaline may be influenced by the formation of biotite. The B2O3-content needed to attain saturation of silicic melts in tourmaline varies, depending on diverse factors. For example, whereas Bénard et al. (1985) observed that tourmaline saturation in silicic melts requires over 0·3 wt % B2O3, Wolf & London (1997) found that saturation of melt in tourmaline, with or without ferromagnesian minerals, requires in excess of 2 wt % B2O3. Although a minimum amount of B is necessary for tourmaline crystallization, it appears that low Ti-contents in the melt stabilize tourmaline with respect to biotite (Nabelek et al., 1992; Wilke et al., 2002). Tourmaline and biotite may coexist in some leucogranites, but tourmaline-bearing facies generally are distinct from biotite-bearing facies (Nabelek et al., 1992). The TiO2-content of the Martinamor granitic rocks decreases in the order: biotite-bearing SP orthogneiss, including accessory tourmaline (0·41–0·47 wt % TiO2); tourmaline-bearing granite with biotite subordinate (0·09–0·07 wt % TiO2); tourmaline-bearing granite without biotite (<0·06 wt % TiO2). This appears to indicate that the amount of TiO2 may have a significant influence on the tourmaline content in granites and, in turn, on the amount of B that can be expelled into country rocks.
Our proposed model for the origin and evolution of the tourmalinites in the Martinamor area can be summarized as follows
(1) Formation of tourmalinites by boron metasomatism of psammopelitic rocks. The boron was probably derived from magmatic fluids related to emplacement of the Cadomian granitoid of San Pelayo, which escaped from the crystallizing melt to interact with surrounding country rocks.
(2) Folding, mylonitization and recrystallization of the tourmalinites as a result of the Variscan regional metamorphism and deformation. These processes, in addition to causing replacement and development of overgrowths on tourmaline, gave rise to multiple tourmaline ± quartz veins, which represent important sinks of boron derived from the Cadomian tourmalinites.
(3) Formation of boron-rich melts by partial melting of high-grade metamorphic peraluminous rocks during the Variscan orogeny, the Cadomian tourmalinites being the main source of boron in the granitic magma. The emplacement of Variscan granites and pegmatites led to new tourmalinization processes. Early to late growth of tourmaline is believed to have occurred during crystallization of the Martinamor granite.
(4) Fluids exsolved during the final crystallization of tourmaline-bearing melts and expelled into the wall rocks yielded a new stage of tourmalinization (Variscan tourmalinites), as well as the formation of quartz +tourmaline veins. Albitization of psammitic schists and quartzites, and the development of albite veins were concurrent with these late tourmalinization processes.
This sequence of events suggests a multistage recycling of boron in the Martinamor antiform from the Cadomian to Variscan orogenic cycles, which accounts for the presence of tourmaline in a variety of metamorphic, hydrothermal, granitic and pegmatitic rocks. Boron isotopic compositions of tourmalines from these environments depend on several factors (source region, metamorphic grade, magma degassing, etc.) but, overall, the 11B values indicate that redistribution of boron through time was linked to recycling of crustal material. This model is in accord with the idea that peraluminous leucogranites are the only magmatic rocks that form entirely by crustal anatexis, without influx of mantle-derived material (Patiño Douce, 1999). In the Martinamor region, boron may have been incorporated into peraluminous granitic melts by partial melting of local B-rich metamorphic basement with loss of B-bearing hydrothermal fluids. The recycling of pre-Paleozoic crust, including unusually boron-rich rocks, may have played an important role in the generation of B-rich granitic melts in other areas of the Iberian massif. Granites that have elevated B-contents and contain magmatic tourmaline require an unusual B-enrichment in their source regions (Leeman and Sisson, 1996).
|
SUPPLEMENTARY DATA |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTINGANALYTICAL TECHNIQUESTOURMALINE OCCURRENCES AND...WHOLE-ROCK CHEMISTRYMINERAL CHEMISTRYBORON ISOTOPESDISCUSSIONSUPPLEMENTARY DATAREFERENCES |
|---|
Microprobe data for tourmaline and coexisting minerals are available at Journal of Petrology online (http://www.petrology.oupjournals.org).
|
ACKNOWLEDGEMENTS |
|---|
Support for this study was provided by the Spanish CICYT (project BTE 2002-01920). John Slack offered constructive comments and suggestions on a draft version of this manuscript. The authors wish to thank Benito Ábalos from the Geodynamics Department of the Basque Country University for helpful comments on microtectonic aspects. Reviews of Darrell Henry, Martin Smith, an anonymous reviewer and the Executive Editor, Marjorie Wilson, greatly improved the manuscript.
* Corresponding author. E-mail: npppepea{at}lg.ehu.es
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTINGANALYTICAL TECHNIQUESTOURMALINE OCCURRENCES AND...WHOLE-ROCK CHEMISTRYMINERAL CHEMISTRYBORON ISOTOPESDISCUSSIONSUPPLEMENTARY DATAREFERENCES |
|---|
Ague, J. J. (1991). Evidence for major mass transfer and volume strain during regional metamorphism of pelites. Geology 9, 855–858.
Argast, S. & Donnelly, T. W. (1987). The chemical discrimination of clastic sedimentary components. Journal of Sedimentary Petrology 57, 813–823.
Arribas-Rosado, A. (1986). The significance of tourmaline in stratabound tungsten deposits in Spain. Tungsten deposits, Toulouse, France. Abstract p. 33.
Ayers, J. C. & Watson, E. B. (1993). Rutile solubility and mobility in supercritical aqueous fluids. Contributions to Mineralogy and Petrology 114, 321–330.[CrossRef][Web of Science]
Barker, A. J. (1990). Introduction to Metamorphic Textures and Microstructures. New York: Chapman and Hall, 162 pp.
Barth, S. (1993). Boron isotope variations in nature: a synthesis. Geologische Rundschau 82, 640–651.
Bau, M. (1991). Rare-earth element mobility during hydrothermal and metamorphic fluid–rock interaction and the significance of the oxidation state of europium. Chemical Geology 93, 219–230.[CrossRef][Web of Science]
Bau, M. (1996). Controls on the fractionation of isovalent trace elements in magmatic and aqueous systems: evidence from Y/Ho, Zr/Hf, and lanthanide tetrad effect. Contributions to Mineralogy and Petrology 123, 323–333.[CrossRef][Web of Science]
Bebout, G. E., Ryan, J. G. & Leeman, W. P. (1993). B–Be systematics in subduction-related metamorphic rocks: characterization of the subducted component. Geochimica et Cosmochimica Acta 57, 2227–2237.[CrossRef][Web of Science]
Bell, T. H. & Rubenach, M. J. (1980). Crenulation cleavage development: evidence for progressive, bulk inhomogeneous shortening from ‘millipede’ microstructures in the Robertson River Metamorphics. Tectonophysics 68, T9–T15.[CrossRef][Web of Science]
Bénard, F., Moutou, P. & Pichavant, M. (1985). Phase relations of tourmaline leucogranites and the significance of tourmaline in silicic magmas. Journal of Geology 93, 271–291.[Web of Science]
Bons, P. D. (2000). The formation of veins and their microstructures. In: Jessell, M. W. & Urai, J. L. (eds) Stress, Strain and Structure: a Volume in Honour of W.D. Means, Vol. 2. Journal of the Virtual Explorer 2, 12.
Caggianelli, A., Fiore, S., Mongelli, G. & Salvemini, A. (1992). REE distribution in the clay fraction of pelites from the southern Apennines, Italy. Chemical Geology 99, 253–263.[CrossRef][Web of Science]
Cartwright, I., Power, W. L., Oliver, N. H. S., Valenta, R. K. & McLatchie, G. S. (1994). Fluid migration and vein formation during deformation and greenschist facies metamorphism at Ormiston gorge, central Australia. Journal of Metamorphic Geology 12, 373–386.[Web of Science]
Díez Balda, M. A. (1986). El Complejo Esquisto-Grauváquico, las Series Paleozoicas y la estructura Hercínica al Sur de Salamanca. Universidad de Salamanca, 162 pp.
Díez Balda, M. A., Vegas, R. & González Lodeiro, F. (1990). Central Iberian Zone: structure. In: Dallmeyer, R. D. & Martínez García, E. (eds) Pre-Mesozoic Geology of Iberia. Berlin: Springer-Verlag, pp. 172–188.
Díez Balda, M. A., Martínez Catalán, J. R. & Ayarza, P. (1995). Syn-collisional extensional collapse parallel to the orogenic trend in a domain of steep tectonics: the Salamanca Detachment Zone (Central Iberian Zone, Spain). Journal of Structural Geology 17, 163–182.[CrossRef][Web of Science]
Dingwell, D. B., Pichavant, M. & Holtz, F. (1996). Experimental studies of boron in granitic melts. In: Grew, E. S. & Anovitz, L. M. (eds) Boron: Mineralogy, Petrology and Geochemistry. Mineralogical Society of America, Reviews in Mineralogy 33, 331–385.
Duong, P. K., Vache, R., Arribas, A. & Thadeu, D. (1980). Yacimientos de tungsteno y metales asociados en Europa occidental. Boletín Geológico y Minero XCI–II, 391–416.
Dyar, M. D., Taylor, M. E., Lutz, T. M., Francis, C. A., Guidotti, C. V. & Wise, M. (1998). Inclusive chemical characterization of tourmaline: Mössbauer study of Fe valence and site occupancy. American Mineralogist 83, 848–864.[Abstract]
Flynn, R. T. & Burnham, C. W. (1978). An experimental determination of the rare-earth partition coefficients between a chloride-containing vapor phase and silicate melts. Geochimica et Cosmochimica Acta 42, 685–701.[CrossRef][Web of Science]
Frondel, C. & Collette, R. L. (1957). Synthesis of tourmaline by reaction of mineral grains with NaCl–H3BO3 solution, and its implications in rock metamorphism. American Mineralogist 42, 754–758.[Web of Science]
Frost, B. R., Barnes, C. G., Collins, W. J., Arculus, R. J., Ellis, D. J. & Frost, C. D. (2001). A geochemical classification for granitic rocks. Journal of Petrology 42, 2033–2048.
Galibert, F. (1984). Géochimie et géochronologie du complexe granitique de l'antiforme de Morille (Salamanca, Espagne). Unpublished Rapport de stage de DEA, Lab. de Géochimie Isotopique, Université de Montpellier.
García de Figuerola, L. C., Franco González, P. & Castro Dorado, A. (1983). Características petrológicas del complejo laminar pegmatoide (Serie de Alamo) de las provincias de Salamanca y Avila. Studia Geologica Salmanticensia XIX, 33–77.
Gieré, R. (1990). Hydrothermal mobility of Ti, Zr, and REE: examples from the Bergell and Adamello contact aureoles (Italy). Terra Nova 2, 60–67.
Gonzalo, F. J., Saavedra, J., García Sánchez, A., Pellitero, E., Arribas, A. & Rodríguez, S. (1975). Las rocas graníticas de la antiforma de Martinamor (Salamanca). Actas II Reunión Ibero-Americana de Geología Económica 4, 227–246.
Govindaraju, K., Potts, P. J., Webb, P. C. & Watson, J. S. (1994). Report on Whin Sill dolerite WS–S from England and Pitscurie microgabbro PM–S from Scotland: assessment by one hundred and four international laboratories. Geostandards Newsletter 18, 211–300.[CrossRef][Web of Science]
Häussinger, H., Okrusch, M. & Scheepers, D. (1993). Geochemistry of premetamorphic hydrothermal alteration of metasedimentary rocks associated with the Gorob Massive Sulfide Prospect, Damara Orogen, Namibia. Economic Geology 88, 72–90.
Hawthorne, F. C. & Henry, D. J. (1999). Classification of the minerals of the tourmaline group. European Journal of Mineralogy 11, 201–215.
Henry, D. J. & Dutrow, B. L. (1996). Metamorphic tourmaline and its petrologic applications. In Grew, E. S. & Anovitz, L. M. (eds) Boron: Mineralogy, Petrology and Geochemistry. Mineralogical Society of America, Reviews in Mineralogy 33, 503–557.
Henry, D. J. & Guidotti, C. V. (1985). Tourmaline as a petrogenetic indicator mineral: an example from the staurolite-grade metapelites of NW Maine. American Mineralogist 70, 1–15.[Abstract]
Henry, D. J. & Guidotti, C. V. (2002). Titanium in biotite from metapelitic rocks: temperature effects, crystal-chemical controls, and petrologic applications. American Mineralogist 87, 375–382.
Henry, D. J., Dutrow, B. L. & Selverstone, J. (2002). Compositional asymmetry in replacement tourmaline: an example from the Tauern Window, Eastern Alps. Geological Materials Research 4, 1–18.
Jiang, S.-Y. (2001). Boron isotope geochemistry of hydrothermal ore deposits in China: a preliminary study. Physics and Chemistry of the Earth (A) 26, 851–858.[CrossRef]
Jiang, S.-Y. & Palmer, M. R. (1998). Boron isotope systematics of tourmaline from granites and pegmatites: a synthesis. European Journal of Mineralogy 10, 1253–1265.
Jiang, S.-Y., Yang, J.-H., Novak, M. & Selway, J. (2003). Chemical and boron isotopic compositions of tourmaline from the Lavicky leucogranite, Czech Republic. Geochemical Journal 37, 545–556.[Web of Science]
Jongh de, W. K. (1973). X-ray fluorescence analysis applying theoretical matriz corrections. Stainless steel. X-Ray Spectrometry 2, 151–158.
Jolliff, B. L., Papike, J. J. & Laul, J. C. (1987). Mineral recorders of pegmatite internal evolution: REE contents of tourmaline from the Bob Ingersoll pegmatite, South Dakota. Geochimica et Cosmochimica Acta 51, 2225–2232.[CrossRef][Web of Science]
Kerrick, D. M. (1990). Aluminium metasomatism. In Ribbe, P. H. (ed.) The Al2SiO5 Polymorphs. Mineralogical Society of America, Reviews in Mineralogy 22, 311–352.
Leeman, W. P. & Sisson, V. B. (1996). Geochemistry of boron and its implications for crustal and mantle processes. In: Grew, E. S. & Anovitz, L. M. (eds) Boron: Mineralogy, Petrology and Geochemistry. Mineralogical Society of America, Reviews in Mineralogy 33, 645–707.
Linares, E., Pellitero, E. & Saavedra, J. (1987). Primeras edades radiométricas en el área estanno-wolframífera de Morille–Martinamor (Centro-Oeste de España). Boletín Geológico y Minero XCVIII–V, 640–646.
London, D., Morgan, G. B., VI, Babb, H. A. & Loomis, J. L. (1993). Behavior and effects of phosphorus in the system Na2O–K2O–Al2O3–SiO2–P2O5–H2O at 200 MPa (H2O). Contributions to Mineralogy and Petrology 113, 450–465.[CrossRef][Web of Science]
London, D., Morgan, G. B., VI & Wolf, M. B. (1996). Boron in granitic rocks and their contact aureoles. In: Grew, E. S. & Anovitz, L. M. (eds) Boron: Mineralogy, Petrology and Geochemistry. Mineralogical Society of America, Reviews in Mineralogy 33, 299–330.
Manning, D. A. C. (1981). The effect of fluorine on liquidus phase relationships in the system Qz–Ab–Or with excess water at 1 kbar. Contributions to Mineralogy and Petrology 76, 206–215.[CrossRef][Web of Science]
Martínez-García, E. & Nicolau, J. (1973). Los terrenos infraordovícicos de la antiforma de Martinamor (Salamanca). Boletín Geológico y Minero LXXXIV–VI, 407–418.
McDonough, W. F. & Sun, S.-S. (1995). The composition of the Earth. Chemical Geology 120, 223–253.[CrossRef][Web of Science]
McLennan, S. M. (1989). Rare earth elements in sedimentary rocks: influence of provenance and sedimentary processes. In: Lipin, B. R. & McKay, G. A. (eds) Geochemistry and Mineralogy of Rare Earth Elements. Mineralogical Society of America, Reviews in Mineralogy 21, 169–200.
Medaris L. G., Jr, Fournelle, J. H. & Henry, D. J. (2003). Tourmaline-bearing quartz veins in the Baraboo quartzite, Wisconsin: occurrence and significance of foitite and ‘oxy-foitite’. Canadian Mineralogist 41, 749–758.
Moine, B., Ramambazafy, A., Rakotondrazafy, M., Ravolomiandrinarivo, B., Cuney, M. & de Parseval, P. (1998). The role of fluor-rich fluids in the formation of the thorianite and sapphire deposits from SE Madagascar. Mineralogical Magazine 62, 999–1000.
Moran, A. E., Sisson, V. B. & Leeman, W. P. (1992). Boron depletion during progressive metamorphism: implications for subduction processes. Earth and Planetary Science Letters 111, 331–349.[CrossRef][Web of Science]
Morgan, G. B., VI & London, D. (1989). Experimental reactions of amphibolite with boron-bearing aqueous fluids at 200 MPa: implications for tourmaline stability and partial melting in mafic rocks. Contributions to Mineralogy and Petrology 102, 281–297.[CrossRef][Web of Science]
Munoz, J. L. (1984). F–OH and Cl–OH exchange in micas with applications to hydrothermal ore deposits. In: Bailey, S. W. (ed.) Micas. Mineralogical Society of America, Reviews in Mineralogy 13, 469–493.
Nabelek, P., Russ-Nabelek, C. & Denison, J. R. (1992). The generation and crystallization conditions of the Proterozoic Harney Peak leucogranite, Black Hills, South Dakota, USA: petrologic and geochemical constraints. Contributions to Mineralogy and Petrology 110, 173–191.[CrossRef][Web of Science]
Neiva, A. M. R. (1974). Geochemistry of tourmaline (schorlite) from granites, aplites and pegmatites from northern Portugal. Geochimica et Cosmochimica Acta 38, 1307–1317.[CrossRef][Web of Science]
Oliver, N. H. S. & Bons, P. D. (2001). Mechanisms of fluid flow and fluid–rock interaction in fossil metamorphic hydrothermal systems inferred from vein–wall-rock patterns, geometry and microstructure. Geofluids 1, 137–162.[CrossRef]
Ortoleva, P., Merino, E., Moore, C. & Chadam, J. (1987). Geochemical self-organisation I: reaction–transport feedbacks and modelling approach. American Journal of Science 287, 979–1007.
Palmer, M. R. & Slack, J. F. (1989). Boron isotopic composition of tourmaline from massive sulfide deposits and tourmalinites. Contributions to Mineralogy and Petrology 103, 434–451.[CrossRef][Web of Science]
Palmer, M. R., London, D., Morgan, G. B., VI & Babb, H. A. (1992). Experimental determination of fractionation of 11B/10B between tourmaline and aqueous vapour: a temperature and pressure-dependent isotopic system. Chemical Geology 101, 123–129.[CrossRef][Web of Science]
Passchier, C. W. & Speck, P. J. H. R. (1994). The kinematic interpretation of obliquely-transected porphyroblasts: an example from the Trois Seigneurs massif, France. Journal of Structural Geology 16, 971–984.[CrossRef][Web of Science]
Patiño Douce, A. E. (1999). What do experiments tell us about the relative contributions of crust and mantle to the origin of granitic magmas? In Castro, A., Fernández, C. & Vigneresse, J. L. (eds) Understanding Granites. Integrating New and Classical Techniques. Geological Society, London, Special Publication 168, 55–75.
Pesquera, A. & Velasco, F. (1997). Mineralogy, geochemistry and geological significance of tourmaline-rich rocks from the Paleozoic Cinco Villas massif (western Pyrenees, Spain). Contributions to Mineralogy and Petrology 129, 53–74.[CrossRef][Web of Science]
Pollard, P. J., Pichavant, M. & Charoy, B. (1987). Contrasting evolution of fluorine- and boron-rich tin systems. Mineralium Deposita 22, 315–321.[Web of Science]
Pouchou, J. L. & Pichoir, F. (1985). ‘PAP’ 
(pz) procedure for improved quantitative microanalysis. In: Armstrong, J. T. (ed.) Microbean Analysis. San Francisco: San Francisco Press, p. 104.
Power, G. M. (1968). Chemical variation in tourmalines from south-west England. Mineralogical Magazine 36, 1078–1089.
Ramsay, J. G. (1980). The crack-seal mechanism of rock deformation. Nature 284, 135–139.[CrossRef]
Reche, J. & Martinez, F. J. (1996). GPT: an Excel spreadsheet for thermobarometric calculations in metapelitic rocks. Computers and Geosciences 22, 775–784.[CrossRef]
Renard, F., Gratier, J. P. & Jamtveit, B. (2000). Kinetics of crack-sealing, intergranular pressure solution, and compaction around active faults. Journal of Structural Geology 22, 1395–1407.[CrossRef][Web of Science]
Saavedra, J., Pellitero, E., García Sánchez, A. & Fernández Turiel, J. L. (1984). Rasgos petrográficos y geoquímicos de los granitoides y rocas metamórficas del area de Morille—Martinamor (Salamanca, Centro Oeste de España). Estudios Geológicos 40, 15–21.
Sallet, R. (2000). Fluorine as a tool in the petrogenesis of quartz-bearing magmatic associations: application of an improved F–OH biotite–apatite thermometer grid. Lithos 50, 241–253.[CrossRef][Web of Science]
Shaw, D. M. (1996). Similarities and constrasts in lunar and terrestrial boron geochemistry. In: Grew, E. S. & Anovitz, L. M. (eds) Boron: Mineralogy, Petrology and Geochemistry. Mineralogical Society of America, Reviews in Mineralogy 33, 745–769.
Slack, J. F. (1982). Tourmaline in the Appalachian–Caledonian massive sulphide deposits and its exploration significance. Trans. Inst. Min. Metall. 91 (B), 81–89.
Slack, J. F. (1996). Tourmaline associations with hydrothermal ore deposits. In: Grew, E. S. & Anovitz, L. M. (eds) Boron: Mineralogy, Petrology and Geochemistry. Mineralogical Society of America, Reviews in Mineralogy 33, 559–643.
Slack, J. F., Palmer, M. R., Stevens, B. P. J. & Barnes, R. G. (1993). Origin and significance of tourmaline-rich rocks in the Broken Hill district, Australia. Economic Geology 88, 505–541.
Smith, M. P. & Yardley, W. D. (1996). The boron isotopic composition of tourmaline as a guide to fluid processes in the southwestern England orefield: an ion microprobe study. Geochimica et Cosmochimica Acta 60, 1415–1427.[CrossRef][Web of Science]
Spiess, R., Peruzzo, L., Prior, D. J. & Wheeler, J. (2001). Development of garnet porphyroblasts by multiple nucleation, coalescence and boundary misorientation-driven rotations. Journal of Metamorphic Geology 19, 269–290.[Web of Science]
Stünitz, H. (1998). Syndeformational recrystallization: dynamic or compositionally induced? Contributions to Mineralogy and Petrology 131, 219–236.[CrossRef][Web of Science]
Swihart, G. H. & Moore, P. B. (1989). A reconnaissance of the boron isotopic composition of tourmaline. Geochimica et Cosmochimica Acta 53, 911–916.[CrossRef][Web of Science]
Taylor, B. E. & Slack, J. F. (1984). Tourmalines from Appalachian–Caledonian massive sulfide deposits: textural, chemical, and isotopic relationships. Economic Geology 79, 1703–1726.
Taylor, H. P., Jr & Sheppard, S. M. F. (1986). Igneous rocks: I. Processes of isotope fractionation and isotope systematics. In: Valley, J. W., Taylor, H. P., Jr & O'Neil, J. R. (eds) Stable Isotopes in High Temperature Geological Processes. Mineralogical Society of America, Reviews in Mineralogy 16, 227–272.
Torres-Ruiz, J., Pesquera, A., Gil-Crespo, P. P. & Velilla, N. (2003). Origin and petrogenetic implications of tourmaline-rich rocks in the Sierra Nevada (Betic Cordillera, southeastern Spain). Chemical Geology 197, 55–86.[CrossRef][Web of Science]
Tracy, R. J., Rye, D. M., Hewitt, D. A. & Schiffries, C. M. (1983). Petrologic and stable-isotopic studies of fluid–rock interactions, south-central Connecticut: I. The role of infiltration in producing reaction assemblages in impure marbles. American Journal of Science 283-A, 589–616.
Vance, J. A. (1965). Zoning in igneous plagioclase: patchy zoning. Journal of Geology 73, 636–651.
Volfinger, M. & Robert, J.-L. (1980). Structural control of the distribution of trace elements between silicates and hydrothermal solutions. Geochimica et Cosmochimica Acta 44, 1455–1461.[CrossRef][Web of Science]
Von Goerne, G., Franz, G. & Robert, J.-L. (1999). Upper thermal stability of tourmaline + quartz in the system MgO–Al2O3–SiO2–B2O3–H2O and Na2O–MgO–Al2O3–SiO2–B2O3–H2O–HCl in hydrothermal solutions and siliceous melts. Canadian Mineralogist 37, 1025–1039.
Von Goerne, G., Franz, G. & Heinrich, W. (2001). Synthesis of tourmaline solid solutions in the system Na2O–MgO–Al2O3–SiO2–B2O3–H2O–HCl and the distribution of Na between tourmaline and fluid at 300 to 700°C and 200 MPa. Contributions to Mineralogy and Petrology 141, 160–173.[Web of Science]
Vorbach, A. (1989). Experimental examinations on the stability of synthetic tourmaline in temperatures from 250 to 750°C and pressures up to 4 kb. Neues Jahrbuch für Mineralogie, Abhandlungen 161, 69–83.
Walther, J. V. (1997). Experimental determination and interpretation of the solubility of corundum in H2O between 350 and 600°C from 0·5 to 2·2 kbar. Geochimica et Cosmochimica Acta 61, 4955–4964.[CrossRef][Web of Science]
Weisbrod, A., Polak, C. & Roy, D. (1986). Experimental study of tourmaline solubility in the system Na–Mg–Al–Si–B–OH: applications to the boron content of natural hydrothermal fluids and tourmalinization processes. International Symposium on Experimental Mineralogy, Abstract 1, 140–141.
Werding, G. & Schreyer, W. (1996). Experimental studies on borosilicates and selected borates. In: Grew, E. S. & Anovitz, L. M. (eds) Boron: Mineralogy, Petrology and Geochemistry. Mineralogical Society of America, Reviews in Mineralogy 33, 117–163.
Wilke, M., Nabelek, P. I. & Glascock, M. D. (2002). B and Li in Proterozoic metapelites from the Black Hills, U.S.A.: implications for the origin of leucocratic magmas. American Mineralogist 87, 491–500.
Wolf, M. B. & London, D. (1997). Boron in granitic magmas: stability of tourmaline in equilibrium with biotite and cordierite. Contributions to Mineralogy and Petrology 130, 12–30.[CrossRef][Web of Science]
Yardley, B. W. D. & Botrell, S. H. (1992). Silica mobility and fluid movement during metamorphism of the Connemara schists, Ireland. Journal of Metamorphic Geology 10, 453–464.[Web of Science]
Yardley, B. W. D., Rochelle, C. A., Barnicoat, A. C. & Lloyd, G. E. (1991). Oscillatory zoning in metamorphic minerals: an indicator of infiltration metasomatism. Mineralogical Magazine 55, 357–365.[Web of Science]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  A. Pesquera, J. Torres-Ruiz, P. P. Gil-Crespo, and E. Roda-Robles Multistage boron metasomatism in the Alamo Complex (Central Iberian Zone, Spain): Evidence from field relations, petrography, and 40Ar/39Ar tourmaline dating American Mineralogist, October 1, 2009; 94(10): 1468 - 1478. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Pesquera, F. Torres, P. Gil-Crespo, and J. Torres-Ruiz TOURCOMP: A program for estimating end-member proportions in tourmalines Mineralogical Magazine, March 18, 2009; 72(5): 1021 - 1034. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||