- © The Mineralogical Society
Apatite samples from rare-metal mineralization were investigated by a combination of cathodoluminescence (CL) microscopy and spectroscopy, microchemical analysis and trace element analysis. Internal structures revealed by CL can be related to variations in the crystal chemistry and may sometimes reflect changes in the composition of the mineralizing fluids.
Apatite from mineralization related to alkaline rocks and carbonatites (Type 1) typically exhibits relatively homogeneous blue and lilac/violet CL colours due to activation by trace quantities of rare earth element ions (Ce3+, Eu2+, Sm3+, Dy3+ and Nd3+). These results correlate with determined trace element abundances, which show strong light rare earth element (LREE) enrichment for this type of apatite. However, a simple quantitative correlation between emission intensities of REE3+/2+ and analysed element concentrations was not found.
Apatite from P-rich altered granites, greisens, pegmatites and veins from Sn-W deposits (Type 2) shows strong Mn2+-activated yellow-greenish CL, partially with distinct oscillatory zoning. Variations in the intensity of the Mn2+-activated CL emission can be related either to varying Mn/Fe ratios (quenching of Mn activated CL by Fe) or to self-quenching effects in zones with high Mn contents (>2.0 wt.%). The REE distribution patterns of apatite reflect the specific geological position of each sample and may serve as a “tracer” for the REE behaviour within the ore system. Although the REE contents are sometimes as high as several hundred parts per million, the spectral CL measurements do not exhibit typical REE emission lines because of dominance of the Mn emission. In these samples, REE-activated luminescence is only detectable by time-resolved laser-induced luminescence spectroscopy.
Both types of apatite (Type 1 in the core and Type 2 in the rim) were found in single crystals from the Be deposit Ermakovka (Transbaikalia). This finding proves the existence of two stages of mineralization within this deposit.
Apatite (general formula Ca5[(PO4)3(OH,F,Cl)]) is an important accessory mineral in various types of igneous, sedimentary and metamorphic rocks. Due to chemical substitutions within the apatite lattice, the Ca2+ sites can be occupied by a number of cations, including but not limited to Sr2+, Mn2+, Fe2+, rare earth elements (REE2+/3+), Y3+ and Na+. The site preference of Mn2+, Sr2+ and REE3+ to either the ninefold Ca(1) or sevenfold Ca(2) sites in the apatite structure remains a point of controversy (e.g. Morozov et al., 1970; Marfunin, 1979; Hughes et al., 1991a,b; Gaft et al., 1998). Ions such as Cl− and OH− can substitute for F− in the apatite structure and P5+ can be partly replaced by Si4+, As5+, S6+ and C4+ (Elliot, 1994). Because of differences in the valence of the substituting ions, a coupled substitution is sometimes required to maintain charge balance, e.g. 2Ca2+ ↔ Na+ + REE3+ or Ca2+ + P5+ ↔ REE3+ + Si4+ (Rønsbo, 1989; Finch and Fletcher, 1992; Elliot, 1994; Coulson and Chambers, 1996; Rae et al., 1996). Another mechanism for charge balancing, sometimes causing deviations from stoichiometric composition, is a substitution in the form of 3Ca2+ ↔ 2REE3+ + □ (where □ is a site vacancy, cf. Elliot, 1994).
It was previously observed that incorporation of trace elements in accessory apatite depends on the petrology and specifics of the host rocks (e.g. Zhang et al., 1985; Fleischer and Altschuler, 1986; Boudreau and Kruger, 1990). Thus, geochemical features of apatite can be used as an important tool for mineralogical and geological interpretation (cf. Landa et al., 1983; Fleischer and Altschuler, 1986; Wenzel and Ramseyer, 1992; Pan and Fleet, 1996; Rae et al., 1996; Whitney and Olmsted, 1998; Irber, 1999; Perseil et al., 2000; Seifert et al., 2000).
Imaging of internal structures and alteration features in natural apatite by cathodoluminescence (CL) microscopy provides essential information on rock formation and mineralization processes, particularly fluid evolution (Knutson et al.. 1985; Finch, 1992; Wenzel and Ramseyer, 1992; Coulson and Chambers, 1996; Rae et al., 1996; Perseil et al., 2000). Apatites from different parent rocks or mineral parageneses were found to exhibit various CL and photoluminescence (PL) emissions. For example, apatite from granitic rocks and pegmatites typically exhibits yellowish to orange CL and PL, whereas apatite from carbonatites and other alkaline rocks tends to show blue, violet or lilac CL and PL (e.g. Portnov and Gorobets, 1969; Marfunin, 1979; Mariano, 1988). The characterization of luminescence colour alone, however, is of qualitative nature only and does not necessarily permit identification of all ions responsible for the luminescence colour. Spatially-resolved cathodoluminescence spectroscopy provides useful information about emission centres and their intra-crystal distribution (e.g. Rakovan and Reeder, 1996). In particular, Mn2+ and REE ions serve as effective activators in apatite. Investigations on synthetic and naturally occurring apatite have revealed characteristic emission bands (Mn2+, Eu2+, Ce3+) or sets of emission peaks (REE3+ but not Ce3+) that can be related to specific activator ions (e.g. Portnov and Gorobets, 1969; Kuznecov and Tarashchan, 1975; Marfunin, 1979; Roeder et al., 1987; Mariano, 1988; Blanc et al., 1995; Rakovan and Reeder, 1996; Mitchell et al., 1997; Blanc et al., 2000). Data concerning peak identification are summarized in Götze et al. (2001). On the other hand, as shown for synthetic samples and As-rich natural apatite, the incorporation of Fe2+ or As5+ into the apatite structure can result in significant quenching of the luminescence intensity (Filippelli and Delaney, 1993; Perseil et al., 2000).
Apatite is a common constituent in mineral assemblages from several types of ore deposits. In the present study, two types of occurrences are examined and discussed in detail: (1) apatite from mineralization in relation to alkaline rock sequences including carbonatites; and (2) apatite from P-rich altered granites, greisens, pegmatites and veins from Sn-W deposits. The purpose of the study is to reveal variations in chemical composition and related CL behaviour of apatite from these two settings. The verification of the prevailing CL activation and quenching mechanisms in the mineral was possible due to contrasting crystal chemistry of the samples investigated. Consequences for genetic interpretations based on CL investigations are discussed briefly in conclusion.
Geological setting, materials and methods
The sample locations are reported in Table 1⇓. One sample from Kovdor (Kola Peninsula, Russia) and two samples from Mushugai Khudug (Gobi desert, Mongolia) represent apatite from carbonatite complexes. Apatite from larvikite from an open pit at Svensken (Norway) is included in our study representing the occurrence in alkaline rocks. Samples from Sn-W deposits derive from Ehrenfriedersdorf (Central Erzgebirge, Germany), Sadisdorf (Eastern Erzgebirge, Germany) and Vysoky Kamen (Slavkovsky les, Czech Republic; cf. Jachovsky, 1994; René, 1998).
The Kovdor massif consists of a sequence of ultrabasic-alkaline-carbonatite rocks (Kharlamov et al., 1981). The sample investigated derives from the main magnetite-apatite ore body (Landa et al., 1983). At Mushugai, the alkaline rocks are in general acidic. Mushugai apatite was sampled from a small magnetite-apatite ore body but near the contact with the surrounding syenite. The mineral occurs as idiomorphic crystals intergrown with phlogopite. More details of the mineralogy and geology of the Mushugai massif are presented in Baskina et al.(1978), Vasil’eva et al.(1978), Ontoev et al.(1979) and Rundqvist et al.(1995).
An apatite sample with distinct colour zonation (core – yellowish-green, rim – bluish green) was collected from a Be orebody in Ermakovka (Transbaikalia, Russia). Be-F orebodies are developed as zones of metasomatic alteration within Upper Proterozoic schist, limestone and dolomite near their contact with Jurassic subalkaline granites (Novikova et al., 1994; Lykhin et al., 2001). The apatite was sampled from an altered dolomite near the largest ore body of the deposit, where it occurs in association with dolomite, phlogopite and fluorite.
Three samples (A4, A6 and A9) are from a breccia in the near exocontact of the Sauberg cupola of the Ehrenfriedersdorf Sn-W deposit (Central Erzgebirge, Germany). The breccia contains angular granite, apatite and schist fragments cemented by vein quartz (Fig. 1⇓). Another sample derives from a pipe-like topaz-albite-granite body near the top of the granite cupola of the Greifenstein area (A3) and an additional one (A5) from cassiterite-bearing mica greisen from the upper part of the Vierung granite cupola. The geology and mineralogy of the Ehrenfriedersdorf deposit is described by Hösel (1994) and Seltmann et al.(1995).
Carbon-coated polished thin-sections of all samples were investigated by a combination of polarizing microscopy and CL microscopy and spectroscopy. The CL examinations were initially performed on a ‘hot cathode’ CL microscope at 14 kV and with a current density of ~10 μA/mm2. Luminescence images were captured ‘on-line’ during CL operations by means of an adapted digital video-camera (KAPPA 961-1138 CF 20 DXC with cooling stage). The CL spectra were obtained using an EG&G digital triple-grating spectrograph with a liquid nitrogen-cooled, Si-based charge-coupled device (CCD) detector (Rieser et al., 1994). The CCD camera was attached to the CL microscope by a silica-glass fibre guide. The CL spectra were measured in the range 380 to 1200 nm using standardized conditions (wavelength calibration by a Hg lamp). To prevent any falsification of the CL spectra due to electron bombardment, all spectra were taken on non-irradiated sample spots.
In addition, panchromatic and spectral CL investigations and energy dispersive X-ray analysis (EDX) were carried out using a JEOL JSM 6400 SEM. This SEM is equipped with an Oxford Instruments MonoCL system, an additional panchromatic CL detector, a BSE detector and a Noran/Vantage EDX system with a light element Si(Li) detector. For CL investigation, the accelerating voltage was set at 20 kV and the beam current in the range 0.6–1.6 nA. The CL spectra were detected over the range 200–800 nm, with 1 nm steps and a dwell time of 1 s per step. For EDX analyses, the microscope was operated at 20 kV and 0.6 nA. Detection limits were 0.1 wt.% for Fe, 0.2 wt.% for Mn, 0.2 wt.% for La and Ce, and 0.3 wt.% for Nd.
The combination of CL measurements on an optical CL microscope and a SEM enabled us: (1) to produce true CL colour images as well as high-resolution panchromatic SEM-CL images comparing them to BSE images; (2) to get spectral information from the UV (200 nm) up to the IR (1200 nm); and (3) to realise time-resolved spectral CL measurements of all samples.
Some of the samples were also measured by laser-induced time-resolved luminescence spectroscopy. Details of the method are reported elsewhere (Gaft et al., 1998).
To obtain more information on the trace element composition of investigated apatite, separated aliquots of selected samples and differently coloured crystal zones, respectively, were investigated by inductively coupled plasma mass spectrometry (ICP-MS) analysis (see Table 2⇓). Analyses were performed using a Perkin Elmer Sciex Elan 5000 quadrupole instrument with a cross-flow nebulizer and a rhyton spray chamber (see Monecke et al., 2000a for more details of the analytical procedure).
Results and discussion
Apatite from mineralization in alkaline rocks and carbonatites
Samples from Kovdor, Mushugai and Svensken
The apatite from the carbonatite of Kovdor shows relatively homogeneous (unzoned) blue to lilac CL with inclusions of orange luminescing calcite (Fig. 2a⇓) occasionally intergrown with richterite. Under CL this apatite behaves in a similar fashion to apatites from alkaline rocks reported from elsewhere (Portnov and Gorobets, 1969; Mariano, 1988; Finch, 1992; Coulson and Chambers, 1996; Rae et al., 1996). Further examples of apatite grains from Mushugai and Svensken (Fig. 2b,c⇓) demonstrate that blue to lilac CL is a general feature of apatite from alkaline rocks. The Svensken apatite reveals primary growth zoning not visible under conventional optical microscopy.
The CL emission spectra of the apatite samples from Kovdor, Mushugai and Svensken are dominated by REE-activated luminescence (Figs 3⇓–5⇓⇓). Blue CL is related to Eu2+ emission (around 410–430 nm, cf. Mariano, 1988). The lilac tint is caused by CL activation due to Ce3+ (in the near UV around 360 nm) and Sm3+ centres luminescing in the orange to red at 600 and 645 nm (note that the spectra are not corrected for detector response so that emission in the range from 300 to 450 nm appears to be suppressed). Additional lines are assigned to Dy3+ (480 and 570 nm) and Nd3+ (in the IR at 870 to 900 nm). The Svensken and Mushugai samples show a pronounced Nd3+ emission. However, this feature does not contribute to the visible CL.
In general, CL behaviour of investigated apatite is in accordance with the trace element characteristics of the samples. As a rule, a strong enrichment of the light REE (LREE) including Ce, Nd, Sm and Eu, acting as emission centres in CL, can be found in apatite from alkaline magmatic complexes (see Table 2⇑ and Fig. 6⇓ as well as Landa et al., 1983 for Kovdor apatite; cf. Vasil’eva et al., 1978 and Kovalenko et al., 1982 for Mushugai apatite). High La, Ce and Nd concentrations of up to several wt.%, were detected in the samples from Kovdor, Svensken and Mushugai by microanalysis. This is in accordance with findings for other locations (Vasil’eva et al., 1978; Rønsbo, 1989; Rae et al., 1996). Relatively high Na contents found in LREE-rich apatite from Kovdor, Mushugai, Svensken and Ermakovka by ICP-MS (Table 2⇑) and EDX (up to 1.0 wt.% Na2O) indicate that Na compensation is one important mechanism for charge balancing according to the scheme 2Ca2+ ↔ Na+ + REE3+ (Rønsbo, 1989). Strong enrichment of LREE, particularly of La and Ce, is in accordance with the REE distribution patterns of the host rocks (cf. Kempe and Dandar, 1995). At the same time, the Mn content in apatite is relatively low and ranges from 10 to several 100 ppm (Table 2⇑; Vasil’eva et al., 1978). These values are within the minimum concentration level of Mn detectable as Mn2+ by CL spectroscopy in Mn-doped synthetic apatite (Filippelli and Delaney, 1993). Another typical characteristic of the apatite from alkaline rocks is a high to very high Sr content (up to several wt.%; Landa et al., 1983; Rae et al., 1996). This is confirmed by our observations (0.7 up to 3.6 wt.% SrO detected by EDX, cf. also ICP-MS data in Table 2⇑). Moreover, we found that fluorite from such an environment is also strongly enriched in Sr (e.g. Kempe et al., 1999). In accordance with other authors, we can find no correlation between the Sr content and the CL properties of the apatite.
The core and the rim of the Ermakovka apatite crystals exhibit quite different CL behaviour. The yellowish-green coloured core exhibits blue to bluish-violet CL similar to apatite samples from alkaline rock complexes described above. The CL microscopy as well as BSE imaging reveal irregular internal structures due to polycrystalline growth with oscillatory growth zoning (Fig. 2d⇑ and Fig. 7⇓). By contrast, the monocrystalline, bluish-green rim exhibits greenish-yellow CL as found for apatite from Sn-W deposits (see below) with distinct oscillatory growth zoning. It was not possible to take CL micrographs by SEM-CL imaging due to strong phosporescence from the rim of the apatite grain. In our experience, phosphorescence is typical of Mn2+ centres in Ca minerals (calcite, fluorite, apatite). At the core-rim boundary and in the rim zone itself, intergrowths with red luminescing calcite are common. In the outer rim, a zone with extremely weak CL intensity was observed, which forms a bright zone in the BSE image (Fig. 7⇓).
The CL spectra of the differently luminescing zones (spots 1–3 in Fig. 2d⇑) reveal a change from mainly Ce3+, Eu2+, Sm3+, Dy3+ (and Nd3+) activated blue to lilac CL in the inner crystal core to mainly Mn2+ (565 nm) and minor REE3+ activated greenish-yellow CL in the outer zone (Fig. 8⇓).
Trace element analysis emphasizes the differences between the core and the rim of the Ermakovka apatite crystals. The yellowish core shows REE distribution patterns similar to the Kovdor apatite, whereas the bluish-green rim is depleted in REE and strongly enriched in heavy REE (HREE) as well as in Y (Fig. 6⇑ and Table 2⇑). The Y/Ho ratio of the rim (169) clearly exceeds the values of the CHARAC box defined by Bau (1996) and the values for the core (14) and the Kovdor apatite (21), which are even below the chondritic value of 28. However, both parts of the crystal are enriched in Sr, the rim (2.9 wt.% Sr) even more than the core (0.8 wt.%). Moreover, as in the case of Kovdor apatite, Ermakovka apatite is rich in Th (Th 0.04 wt.% in the core and 0.7 wt.% in the rim, with Th≫U).
Variations in luminescence intensities
More detailed microchemical investigation of the core and the rim of the Ermakovka apatite crystals reveals that oscillatory growth zoning visible in CL and BSE images may be related to variations in crystal chemistry. The Sr content is more or less constant in the core (0.65–0.75 wt.% Sr) and in the rim (2.6–2.9 wt.% Sr). Within the core, the contrast in CL and BSE correlates well with the REE content: bright luminescing zones (darker in BSE) are depleted in La and Ce, but enriched in Nd compared to darker luminescing zones (bright in BSE). The Mn was below the detection limit of the EDX. The variations in CL intensities may be explained by relative Eu and Sm enrichment in the Nd-rich zones with related enhanced luminescence intensities. Otherwise, luminescence quenching due to higher concentrations of non-luminescent La centres in the La- and Ce-rich zones may also lead to decreasing integral CL intensities.
The Mn is strongly enriched in the rim (0.3–0.6 wt.% Mn). Accordingly, the rim shows strong Mn2+ emission. However, the highest Mn content (1.8 wt.% Mn) was found in the weak luminescing zone. It may be possible to relate the high average Mn content (1.4 wt.%) found by ICP-MS for the whole rim area to this zone. Another characteristic feature of this zone is the very high Th content (>0.39 wt.%). This fact may explain low luminescence efficiency in the Mn-rich zone possibly related to strong lattice damage or to charge conversion due to high radioactive irradiation dosages.
Comparison of CL emission spectra of the apatites from Ermakovka (cores) and Kovdor reveals that both spectra are dominated by Sm3+, Dy3+ and Eu2+ but the sample from Kovdor shows only very slight Ce3+- and Nd3+-activated CL emission in contrast to the core of the Ermakovka sample. This is remarkable insofar as the REE distribution patterns and the concentrations of Ce and Nd in the two apatite samples are very close to each other. Accordingly, the occurrence or absence and the relative intensity of a specific REE emission peak in the CL spectrum of apatite cannot be used for quantitative estimations of the relative abundance of these REE. Results recently published by Barbarand and Pagel (2001) suggest that there is no simple correlation between contents of the individual REE and the emission intensities of the related centres. Mitchell et al.(1997) found that the CL spectra of natural apatite result from a subtle interplay of at least three factors: the relative concentration of the individual REE, its relative efficiency in activating luminescence and the presence or absence of elements that may act as sensitizers or quenchers. Furthermore, differences in CL behaviour can also be caused by the crystallographic orientation of apatite in thin sections (Murray and Oreskes, 1997; Barbarand and Pagel, 2001) and by charge conversion due to radioactive irradiation. In the case of the apatite samples from Ermakovka and Kovdor, the different contents of Mn, which probably act as a sensitizer of REE-activated CL emission (Portnov and Gorobets, 1969; Marfunin, 1979), yield a possible explanation for the variability in luminescence behaviour.
Apatite in altered granitic rocks from Sn-W deposits
Apatite from the Ehrenfriedersdorf, Sadisdorf and Krasno Sn deposits shows yellowish-green CL independent of the geological environment (granite, greisen, pegmatite, breccia or ore vein). Cathodoluminescence microscopy of the Ehrenfriedersdorf samples reveals internal structures of the single apatite grains, which are not discernible by polarizing microscopy. In most of the investigated samples, distinct oscillatory growth zoning is visible and several grains show additional sector zoning (Fig. 2e⇑–h). Similar internal structures and CL colours were also reported for apatite crystals from the Panasqueira W-Sn deposit, Portugal (Knutson et al., 1985) and found for apatite from greisenized granite from Vysoky Kamen during our study. The internal structure of the Sadisdorf apatite is more complicated, showing primary growth zoning as well as patchy regions of secondary alteration (Fig. 2h⇑).
As in the Ermakovka sample, it was not possible to image internal structures in Ehrenfriedersdorf samples by SEM-CL due to the strong phosphorescence. By contrast, the defocused electron beam of the CL microscope produces excellent CL micrographs of the Mn2+-activated apatite luminescence.
Spectral analysis of the CL emission shows only one intensive band centred at 565 nm (2.19 eV, MHW 70 nm) assigned to the emission of Mn2+ centres (Fig. 9⇓). The CL intensity of the 565 nm emission band decreases strongly with time. Saturation was reached after 1000–1200 s. In contrast, the emission of the REE3+ centres shows stable intensities under electron irradiation. A possible explanation of the emission behaviour of Mn2+ is a process of thermoluminescence decay or a decay of sensitizing centres. Intensity decrease of Mn2+-activated luminescence in apatite under the condition of a focused electron beam in a SEM was investigated by Barbarand and Pagel (2001).
The luminescence behaviour of apatite from Sn-W deposits is in general agreement with the trace element contents found. The Mn concentrations are enhanced while the REE contents (especially of the LREE) are generally low, although some of the apatite samples from Ehrenfriedersdorf have REE contents of some hundreds of ppm. However, with the exception of sample A9 (apatite breccia) spectral CL measurements did not detect the characteristic REE emission peaks. Only in sample A9, was a slight CL emission in the UV at 365 nm detected, which can probably be related to Ce3+ (Blanc et al., 2000). Obviously, the REE-activated CL emission is hidden by the dominant Mn2+ luminescence.
Applying time-resolved laser-induced spectroscopy it is possible to determine REE luminescence, which has been hidden by other strong emission bands. The differentiation of the specific luminescence centres results from the different delay times, which is in the range of nanoseconds for water-organic complexes, microseconds for rare earth elements and milliseconds for transition metal ions (Gaft et al., 1998). Studies of the Ehrenfriedersdorf apatite by time-resolved laser-induced luminescence reveal that in some cases Eu2+ and Dy3+ activation is present, which was only detectable using different excitation wavelengths and delay times, respectively (Fig. 10⇓). The Dy3+ activation correlates with the relatively high HREE contents in the samples.
Time-resolved luminescence measurements allow a comparison of Eu2+ emission (f-d transition) from all apatite samples investigated. The band position (λmax) shifts from 410 nm in the apatite from Kovdor to 445 nm in the Ehrenfriedersdorf apatite (Fig. 11⇓). Blanc et al. (1996) reported a variable position of the Eu2+-activated CL emission band in apatite depending upon the type and composition of the apatite (hydroxy-,chlor-or fluorapatite). However, our results are not in full agreement with their findings, although the trend is the same as pre dicted (Kovdor apatite is low-F, Ehrenfriedersdorf apatite F-rich and Ermakovka has an intermediate position). Additional investigation is required to resolve the dependence of Eu2+ emission on the crystal chemistry of apatite.
Although undetectable by conventional CL spectroscopy, chemically determined REE contents in the apatite samples show a clear correlation with their geological environment (Fig. 12⇓). As we found for fluorite samples from Ehrenfriedersdorf and other Sn deposits (Goldstein et al., 1995; Kempe and Goldstein, 1997, cf. Fig. 12⇓), Eu anomalies in the apatite from Ehrenfriedersdorf change from strongly negative to positive from the endo- to the exocontact of the granite. Similar results were also reported by Raimbault (1988) for apatite from Sn-W deposits in the Massif Central (France). Interestingly, all samples investigated by Knutson et al.(1985) from the Panasqueira W deposit (Portugal) derived from exocontact ore veins and show distinct positive Eu anomalies. As in the case of fluorite, the variability of Eu contents in apatite may be explained either by differences in incorporation of Eu2+ and Eu3+ during rapid changes in the redox conditions or by phase separation processes.
The REE distribution pattern of apatite from an endocontact pegmatite investigated by Trinkler and Martin (1998) shows a distinct tetrad effect (i.e. a split of chondrite-normalized REE patterns into four rounded segments). The enrichment of the HREE leads to very similar REE distribution patterns as for the enclosing granite (Seltmann et al., 1995; Kempe et al., unpubl. data). Inside of the pipe-like topaz-albite granite bodies near the contact with host rocks, the Eu anomaly disappears and REEs are generally depleted, but the tetrad effect and HREE enrichment are still significant (see apatite A5, and fluorite (1) in Fig. 12⇑). In the near exocontact, the HREEs are strongly enriched, while the tetrad effect is weak. Low LREE contents may be interpreted as indicative of only slight admixing of crustal material to the mineral-forming fluid (cf. Seifert and Kempe, 1994; Gavrilenko et al., 1997). Both apatite and fluorite show distinct positive Eu anomalies. Interestingly, all of the apatite samples analysed are Sr-rich with the exception of the unaltered pegmatite sample, while fluorite is generally found to be low in Sr.
Luminescence quenching by Fe2+(?)
Variations in luminescence intensity between different growth zones in apatite from Ehrenfriedersdorf can be related to changes in the intensity of the Mn2+ emission band (Fig. 13⇓). Mn2+ luminescence in apatite is well known and Mariano (1988) found that low concentrations of Mn (~100 ppm) can serve as an activator for visible CL in apatite. Filipelli and Delaney (1993) reported that synthetic samples with <40 ppm Mn can be considered non-luminescent, based on visual criteria.
Microanalyses perpendicular to the visible CL zoning were performed on one of the apatite samples from Ehrenfriedersdorf (Fig. 14⇓). Only Mn and Fe were found in significant quantities in addition to Ca, P and F. Although the absolute content of Mn increases from the rim to the core, there is no direct effect on the CL intensity. As can be seen in Fig. 14⇓, there is also no direct correlation between CL intensity and the concentrations of Fe. In contrast, a strong correlation exists between the CL intensity and the Mn/Fe ratio. Zones with the highest Mn/Fe ratios exhibit the brightest CL emission, whereas low Mn/Fe ratios (even in zones with high absolute Mn contents) result in dull or no visible CL. While Mn2+ acts as a strong CL activator, Fe2+ is known as an effective quencher of luminescence (although it was not possible to prove directly the existence of Fe2+ in our samples). Therefore, a quantitative estimation of the Mn concentration in apatite using the intensity of the Mn2+-activated emission band at 565 nm, as it was proposed for calcite (Habermann et al., 2000), is not possible. Our result is in agreement with the findings by Filipelli and Delaney (1993) for synthetic apatite, although the concentrations of Mn and Fe in our samples are much higher (0.2–0.8 wt.% instead of concentrations <0.2 wt.% for Mn and <0.8 wt.% for Fe)
Concentration quenching of Mn2+ luminescence
The apatite from the Vysoky Kamen granite demonstrates another case of luminescence quenching. The only element detected here by EDX analysis besides Ca, P and F was Mn. There is a clear correlation of the luminescence intensity with the Mn content (Fig. 15⇓). All zones with Mn concentrations ≫2.0 wt.% show dull luminescence, while zones with 1.0–2.0 wt.% Mn luminesce brightly. The effect of luminescence quenching with increasing concentration is known as concentration quenching and was also found, e.g. for REE centres in scheelite (Kempe et al., 1991). According to Filipelli and Delaney (1993), Mn2+ luminescence intensity saturates at Mn concentrations of ~500 ppm.
Complex relations between trace element contents and CL intensities
Although the CL emission of Mn2+ and REE2+/3+ is clearly related to Mn and REE incorporation in apatite, there is no simple correlation between the contents of these elements and the related emission intensities. Our results suggest possible quenching and sensitizing of REE-activated luminescence by La3+ and Mn2+ centres, respectively.
Likewise, the emission of Mn2+ is possibly influenced by charge conversion and lattice damage due to radioactive irradiation. As demonstrated above, in some cases Mn2+-activated CL is quenched by Fe2+ and self-quenching mechanisms (high Mn2+ content). Additional factors influencing the CL emission intensities in apatite are discussed in the literature (Perseil et al., 2000; Barbarand and Pagel, 2001).
It can be concluded from our results that there are at least two types of apatite in the two types of mineralization investigated. These may be distinguished easily by their trace element characteristics and CL behaviour.
Apatite from alkaline rocks and carbonatite is typically extremely rich in LREE and Sr. In the samples investigated, Th was also found to be prominent. Chemical characteristics are in accordance with the composition of the enclosing rocks and the general CL behaviour. Blue to lilac/violet luminescence may be assigned to high contents of Ce3+, Eu2+ and Sm3+ emission centres.
Intensive Mn2+ emission of apatite from Sn-W deposits is in agreement with general high Mn concentrations and relatively low REE concentrations as well as with the HREE enrichment.
The REE distribution patterns of apatite (as well as of fluorite) from Sn-W deposits may serve as a ‘tracer’ for REE behaviour in such ore systems. These patterns vary in accordance with variations in the geological environment. The high U/Th ratios found in such apatite are also in agreement with the general trend of U enrichment in Sn deposits (Seifert and Kempe, 1994; Morozov et al., 1996).
Cathodoluminescence microscopy yields important information on the primary internal growth structures in the apatite pointing to magmatic and hydrothermal formation of the crystals. Nevertheless, secondary metasomatic alteration as in the case of the Sadisdorf sample may also be found. Similar phenomena were described by Coulson and Chambers (1996) for metasomatically altered apatite from the North Qôroq Centre, South Greenland.
The Ermakovka sample provides an interesting case. Here, both types of apatite (‘carbonatite’ or ‘alkaline’ type core and ‘Sn-W’ type rim) may be found in a single crystal. This result confirms our earlier conclusions about the ‘mixed’ character of Nb-Zr-REE-Be mineralization (Kempe and Dandar, 1995) as well as of the time relations between the earlier ‘carbonatite-like’ and the later ‘Sn-W-like’ stages (Kempe et al., 1999). The Y/Ho ratios of the rim clearly indicate its hydrothermal formation. Similar cases with two types of PL and CL (violet and yellowish) in single apatite crystals were previously described for apatite from altered granitoids of the Mountain Altai/Russia by Nikitina et al. (1966) and the Meissen massif, Germany by Wenzel and Ramseyer (1992), respectively.
Cathodoluminescence microscopy reveals internal structures such as growth zoning, alteration, or different generations of apatite. Spectral analysis of the CL emission yields first evidence concerning the distribution of trace elements (in particular rare earth elements and Mn). Moreover, the results of the present investigation show that observed CL colours of apatite and associated CL spectra may reflect specific genetic environments.
The typical blue and lilac/violet CL of apatite from carbonatites is mainly caused by CL activation due to Ce3+, Eu2+ and Sm3+. The investigated apatite from a Norwegian larvikite exhibits a similar luminescence behaviour. Apatites from such rocks are enriched in LREE and Sr and depleted in Mn. By contrast, the apatite samples from hydrothermal veins with Sn-W mineralization have dominant Mn2+ activation causing yellowish CL with distinct oscillatory zoning. This type of apatite has significant lower REE contents although the HREE are enriched. The REE distribution patterns and especially Eu anomalies are sensitive to the geological environment. Samples from the Ehrenfriedersdorf Sn deposit are not only enriched in Mn, but also in U.
As shown for the Ermakovka Be deposit, single apatite crystals may contain both types of apatite – one similar to apatite from carbonatite, the other with properties characteristic of apatite from Sn-W deposits, reflecting the genetic evolution in the deposit.
In general, a simple quantitative correlation between absolute trace element contents and the CL emission intensities does not exist because of certain processes of luminescence sensitizing and quenching. Quenching of Mn2+-activated CL emission by Fe2+(?) as well as concentration quenching were proved for Mn-rich apatite with relatively low REE contents. Therefore, detailed information concerning the principal trace element geochemistry and REE distribution patterns can be obtained by direct trace-element analysis only. In consequence, CL investigation and trace element studies are effective only as complementary techniques.
The authors would like to thank H. Hildebrandt, W. Klemm, G. Bombach, M. Plötze and P. Dulski for their help in trace element analysis of apatite and fluorite samples. M. Gaft, G. Panczer, D. Wolf and T. Gruner are thanked for the laser-induced luminescence measurements on apatite. U. Kempe is grateful to D. Wolf, S. Dandar, M. Rene and T. Gruner for their assistance with sampling at Mushugai Khudug, Ermakovka, Vysokij Kamen and Ehrenfriedersdorf. The authors are grateful for the helpful comments of two anonymous referees, which helped to improve the manuscript and to S.A.T. Redfern for his editorial efforts.
- Manuscript received 18 May 2001.
- Modified version received 20 November 2001.