Mineralogical and Cathodoluminescence Characteristics of Ca-Rich Kutnohorite from the U¤ Rku¤ T Mn-Carbonate Mineralization, Hungary

Mineralogical Magazine, October 2007, Vol.71(5), pp.493–508

Mineralogical and cathodoluminescence characteristics of Ca-rich kutnohorite from the U¤ rku¤ t Mn-carbonate mineralization, Hungary

1, 1 2 3 1 1 4 M. POLGA´ RI *, B. BAJNO´ CZI ,V.KOVA´ CS KIS ,J.GO¨ TZE ,G.DOBOSI ,M.TO´ TH AND T. VIGH 1 Institute for Geochemical Research, Hungarian Academy of Sciences, H-1112, Budapest, Budao¨rsi u´t45, Hungary 2 Research Institute for Technical Physics and Materials Science, Hungarian Academy of Sciences, H-1525 Budapest, P.O. Box 49, Hungary 3 Department of Mineralogy, TU Bergakademie Freiberg, Brennhausgasse 14, D-09596 Freiberg, Germany 4 Manga´n Ltd., H-8409 U´ rku´t, Hungary [Received 22 November 2007; Accepted 24 March 2008]

ABSTRACT Kutnohorite with moderate and bright orange-red cathodoluminescence (CL) was studied by CL microscopy and spectroscopy. This mineral was found in fossiliferous concretions composed mainly of rhodochrosite from the Mn-carbonate mineralization at U´ rku´t, Hungary. The CL microscopy reveals that kutnohorite occurs as impregnations, layers and veinlets. X-ray diffraction, infrared spectroscopy and electron microprobe studies indicate that the luminescent kutnohorite has excess Ca (72.9À80.0 mol.% CaCO3, 16.3À20.5 mol.% MnCO3, 3.3À5.6 mol.% MgCO3 and 0.0À0.5 mol.% FeCO3). Transmission electron microscopy shows that the luminescent carbonate has a dolomite-type structure, with modulated and mosaic microstructures. The CL spectra of this Ca-rich kutnohorite have a single emission band at 630 nm that is characteristic of Mn2+ substitution in the structure. Our results provide evidence for moderate-to-bright cathodoluminescence of Mn-rich natural carbonates even at 8À10 wt.% Mn and up to 2400 ppm Fe. The self-quenching of Mn appears incomplete in the case of Ca-rich kutnohorite from U´ rku´t.

{ KEYWORDS: kutnohorite, kutnahorite , rhodochrosite, cathodoluminescence, spectroscopy, PIXE, TEM, U´ rku´t, Hungary.

Introduction about the activator elements in carbonates CATHODOLUMINESCENCE microscopy is a technique regarding their coordination and ionic charge routinely used in sedimentary petrology for the (Richter et al., 2003; Gaft et al., 2005). Much study of carbonate rocks and their diagenetic work has been done on the CL characteristics of history (Marshall, 1988; Machel, 2000). Aside the most important and abundant carbonate from the texture revealed by cathodolumines- minerals such as calcite, dolomite and aragonite. cence (CL) microscopy, the (quantitative) CL Other less common carbonate species have spectroscopy combined with trace element received little attention using either CL micro- determinations (e.g. particle-induced X-ray emis- scopy or spectroscopy. There are some studies sion À PIXE) can provide further information which deal with magnesite (Spo¨tl, 1991; El Ali et al., 1993), smithsonite (Go¨tte and Richter, 2004) and rhodochrosite (Gaft et al., 2005). 2+ 5 * E-mail: [email protected] It is generally accepted that the Mn ion (3d ) DOI: 10.1180/minmag.2007.071.5.493 and most trivalent rare-earth elements (REE) (4f) { According to the IMA CNMNC the official name is are the most important activators of extrinsic CL kutnohorite, but kutnahorite is also widely used. in carbonate minerals, while Fe2+ is a quencher of

# 2007 The Mineralogical Society M. POLGA´ RI ET AL.

CL (e.g. Marshall, 1988; Machel, 2000). Using Although it is well known that Mn is the most the Quantitative High Resolution Spectral important CL activator in carbonates, the CL analysis of CL (QHRS-CL) the lower activation method is rarely applied in the petrological limitof Mn 2+ in calcite (yellow-orange CL) and investigation of Mn-carbonate mineralization, dolomite (yellow to red CL) has been extended to due to the widely acknowledged effect of <1 ppm (Habermann et al., 1998, 1999; Gillhaus concentration-quenching or self-quenching. The et al., 2001). A dull-blue CL emission, possibly decrease in CL intensity (self-quenching) occurs caused by lattice defects and imperfections, has at large activator (Mn) concentrations, when the been reported for carbonates with no or only average space between activator ions becomes minor traces of impurities (intrinsic luminescence, smaller and instead of being emitted as e.g. Machel et al., 1991). luminescence, the excitation energy is trans- The CL spectra of calcite usually display one ferred from the absorbing activator ion to emission peak and studies have reported various neighbouring similar ions (Marshall, 1988). wavelengths from 580 to 640 nm; mostly Critical concentrations of 500À1000 ppm were 605À620 nm for this peak (El Ali et al., 1993; documented for the start of self-quenching for Richter et al., 2003). Dolomite can also show Mn2+-activated CL in calcite (Mason, 1987; single-band CL spectra at ~660 nm, but two peaks Mason and Mariano, 1990; El Ali et al., 1993; are much more characteristic. The Mn2+ ion can Habermann et al., 2000). However, comprehen- occupy Ca2+-sites (A sites) as well as Mg2+-sites sive knowledge of the luminescence behaviour (B sites) in dolomite and this phenomenon of Mn-rich (>1 wt.% Mn) carbonates (mangano- produces two overlapping broad bands in the CL calcite, kutnohorite, rhodochrosite) does not spectra (El Ali et al., 1993; Gillhaus et al., 2000, exist. The possible limit of concentration 2001). The emission peak of red luminescent extinction is not known. Walker et al. (1989) dolomites at 640À670 nm is related to Mn found that rhodochrosite with 46.0 wt.% Mn and substitution in the Mg-site, as is the case for the 1.0 wt.% Fe was virtually non-luminescent. ~655 nm peak for Mn2+ in magnesite (El Ali et However, luminescentrhodochrosite,was al., 1993; Spo¨tl, 1991) and the ~660 nm peak in reported by Gorobets et al. (1978) and Gaft et smithsonite (Go¨tte and Richter, 2004), due to the al. (1981, 2005), which seems to indicate similar ionic radii and metal-oxygen bond lengths incomplete self-quenching for Mn2+ and of their structures (Marshall, 1988). In yellow- support for the suggestion of Machel et al. orange dolomites, subtraction of the 640À670 nm (1991) that luminescence up to that of the large emission band reveals the 565À595 nm peak for Mn concentrations may occur. Mn2+ at the Ca-site (El Ali et al., 1993; Gillhaus We found a brightly to moderately luminescent et al., 2000, 2001). carbonate with 8À10 wt.% Mn in some rhodo- Habermann et al. (1998, 1999) developed a chrosite concretions containing fossil fish quantification method for Mn with QHRS-CL collected from the U´ rku´tManganese Mine, (<10 ppm) for Fe-poor calcite using the peak Hungary. Powder X-ray diffraction (XRD) height of the Mn emission band in the CL spectra revealed that this carbonate was kutnohorite. correlated with the Mn content determined by Kutnohorite is a dolomite-type double carbonate PIXE (>10 ppm Mn). The application of Mn with the general formula CaMn(CO3)2 (hexa- quantification to dolomite is complicated by the gonal, space group R3¯ (Frondel and Bauer, 1955; two overlapping bands in the emission spectra. Farkas et al., 1988). It is often cited without any Gillhaus et al. (2000) distinguished red and yellow structural or chemical characterization as Mn-rich luminescent dolomites with different relations of calcite occurring as a rare mineral in hydro- the two broad bands: using the peak-height as a thermal, metamorphosed and sedimentary criterion, a linear correlation for Mn between sequences (e.g. Ozturk and Frakes, 1995; 20À1000 ppm was detected, which corresponds to Gutzmer and Beukes, 1996, 1998; Fan et al., the same correlation found by Habermann et al. 1999; Hein et al., 1999; Krajewsky et al., 2001; (1998, 1999) for calcite. By integrating peak areas, Large et al., 2001; Burke and Kemp, 2002; rather than peak heights, a very good correlation Nyame et al., 2003). In this paper we report the between Mn content and CL intensity in Fe-poor CL microscopic and spectroscopic properties of dolomite with <3000 ppm Mn was obtained kutnohorite for the first time. The aims of our (Gillhaus et al.,2001).Similarresultswerefound study were to characterize the structure of the for other carbonates by Go¨tte and Richter (2004). luminescent carbonate and determine the possible

494 CA-RICH KUTNOHORITE, HUNGARY

causes of CL. In addition to conventional powder Separate samples of the luminescent veinlet XRD, infrared (IR) spectroscopy, EMPA and CL and its adjacent matrix from sample H4 were spectroscopy were used to characterize emission examined by IR spectroscopy. Measurements centres and their structural position. Transmission between 400 and 4000 cmÀ1 were carried out electron microscopy (TEM) was used for the with a Bruker Equinox type Fourier transform structural characterization. infrared (FTIR) spectrometer with 2 cmÀ1 resolution. The KBr pastille method was used with ~1 mg of samples and ~200 mg KBr. Materials and methods Quantitative EMP measurements of carbonates Three concretions were collected from the U´ rku´t were performed on carbon-coated rock chips and Manganese Mine, which is situated in the Trans- thin sections using a JEOL Superprobe 733 EMP Danubian Central Range in Hungary (Fig. 1). equipped with an Oxford INCA 2000 energy- Samples H1, H3 and H4 were collected from shaft dispersive X-ray spectrometer (EDS). The analy- No. III level +186 (archive numbers of the tical conditions were 15 kV accelerating voltage, samples at the Institute for Geochemical 5 nA beam current and a counting time of 50 s. Research, Budapestare PM144/H1, PM144/H3 Calcite, dolomite, siderite and spessartine (for and PM144/H4). The geology and mineralogy of Mn) standards were used for calibration. An the U´ rku´tmanganese depositis described by alternative calibration using wollastonite, MgO, Polga´ri (2001) and Polga´ri et al. (1991, 2000, hematite and spessartine standards gave practi- 2004, 2005a,b). cally the same results. All analyses were made The brown and pinkish elliptical concretions using a fastscanning mode atdifferentmagnifica- are 30À50 cm in size (Fig. 2a). The very fine- tions to prevent carbonate decomposition under grained samples contain parts with different the electron beam and in the case of very fine colours and show coarser breccia-like micro- scale heterogeneities to produce an average value layers, signs of infiltration and microfractures for the analysed area. (Fig. 2bÀd, Table 1). Occasionally, veinlets a Cathodoluminescence examinations were few millimetres thick cut the matrix of the initially performed on polished rock chips concretions (Fig. 2d). using a Reliotron ‘cold-cathode’ microscope Bulk samples were analysed by powder XRD. operated at 5À7 kV accelerating voltage and A Philips 1730 X-ray diffractometer with a 0.5 to 0.9 mA current. Additionally, carbon- graphite crystal monochromator and Cu-Ka coated thin sections were studied using a ‘hot- radiation was used and continuous scans were cathode’ CL microscope HC1-LM (see Neuser, run from 3À70º2y at45 kV and 35 mA, witha 1995) operated at 14 kV acceleration voltage and scan speed of 0.02º2y/s. Fragments of the a currentdensityof ~10 mA/mm2. Luminescence luminescentveinletfrom sample H4 were also images were captured on-line during CL opera- analysed separately using a slower scan speed. tions by means of an adapted digital video-

FIG. 1. Location of the U´ rku´tManganese Mine in Hungary (centralEurope).

495 M. POLGA´ RI ET AL.

FIG. 2. Macroscopic appearance of the Mn-carbonate concretions. Carbonate concretion with fish fossil (a: sample H1). The very fine-grained samples contain parts different in colour (b: sample H1) and show coarser debris-like microlayers (c: sample H3) with a greater content of microfossils. Signs of infiltration, microfractures and, rarely, veinlets (d: sample H4) occur. For the chemical composition of the different parts see Table 1.

camera (KAPPA 961-1138 CF 20 DXC) with recorded with an Acton Research SP-2356 cooling stage. Cathodoluminescence spectra in digital triple-grating spectrograph with a the wavelength range 300À900 nm were Princeton Spec-10 400/B/XP CCD detector

TABLE 1. Chemical composition (wt.%) of the different parts of the concretions (data are averages of four EMP analyses made on an area 100 mm6100 mm for 100 s; for occurrence see Fig. 2).

Sample code and partSiO 2 TiO2 Al2O3 FeO MnO MgO CaO Na2OK2OP2O5 H1, part1 (Fig. 2 b) 0.16 0.00 0.09 0.32 40.65 0.23 12.85 0.31 0.03 8.72 H1, part2 0.65 0.09 0.40 0.32 42.92 1.05 10.19 0.12 0.06 0.34 H1, part3 1.81 0.01 0.75 0.16 32.41 2.01 19.43 0.06 0.01 0.37 H3, part1 (Fig. 2 c) 1.22 0.04 0.31 0.24 33.82 1.76 18.49 0.08 0.03 0.19 H3, part2 29.43 0.13 2.60 2.24 23.49 12.29 2.80 0.11 0.62 0.08 H3, part3 11.25 0.08 1.01 0.55 33.36 6.36 10.27 0.13 0.22 0.13 H4, part1 (Fig. 2 d) 1.57 0.05 4.65 0.19 43.52 1.69 5.16 0.21 0.06 0.46 H4, part2 0.92 0.05 1.18 0.23 40.70 1.23 11.21 0.08 0.13 0.32 H4, part3 (veinlet) 0.17 0.03 0.11 0.20 11.21 1.25 39.89 0.05 0.00 0.00

496 CA-RICH KUTNOHORITE, HUNGARY attached to the CL microscope by a silica-glass compositions were measured using a NORAN fibre guide. The spectra were measured under EDS. standardized conditions (wavelength calibration using Hg lamp, spotwidth30 mm). To prevent any falsification of the CL spectra due to electron Results bombardment, all spectra were taken on non- Mineralogy and chemistry irradiated sample spots. Rhodochrosite (3.67À3.66 A˚ ,2.85À2.84 A˚ , For the TEM study, a fragment of the 2.39 A˚ , 2.18 A˚ , 2.00 A˚ , 1.83 A˚ , ASTM 86-0173 luminescentveinletfrom sample H4 was gently for rhodochrosite, syn) and kutnohorite crushed in ethanol and a drop of the resulting (3.80À3.79 A˚ ,2.99A˚ ,2.79A˚ ,2.46A˚ , suspension was placed onto a carbon-Cu grid. The 2.23À2.26 A˚ , 2.07 A˚ , 1.87 A˚ , 1.84 A˚ , ASTM selected area electron diffraction (SAED) patterns 19-0234 for kutnohorite, calcic) can be identified and TEM images were obtained using a Philips as the main components in the XRD patterns of CM20 TEM operating at 200 keV. The grain the bulk concretions (Fig. 3). Sharp peaks are

FIG.3.(a) XRD patterns of the bulk concretions and (b) the separated veinlet of sample H4.

497 M. POLGA´ RI ET AL.

characteristic for kutnohorite, while rhodochrosite identified as rhodochrosite with shows slight peak broadening. Some smectite and 81.2À87.5 mol.% MnCO3. The dark phase (in hydroxylapatite were also detected, the latter patches and the veinlet in sample H4) is Ca-rich mostly in the core of sample H1. The main with significantly less Mn than the light phase componentof theveinletfrom sample H4 proved (38.1À40.5 wt.% CaO, 10.2À12.5 wt.% MnO to be Ca-rich kutnohorite, beside which the most and 1.2À2.7 wt.% MgO, Table 2). This dark intense peaks of rhodochrosite also appear carbonate is kutnohorite with an excess of Ca (Fig. 3b), which is inferred to be derived from (72.9À80.0 mol.% CaCO3,16.3À20.5 mol.% the adjacent matrix of the concretion. MnCO3, 3.3À7.4 mol.% MgCO3). Lightzones Back-scattered electron (BSE) images revealed in the feather-like pattern of the kutnohorite that the bulk material of the concretions is an veinlet(Fig. 4 f) have ~1 wt.% more MnO and intimate mixture of two 10À100 mm phases 1.6 wt.% less CaO than the darker zones. The (Fig. 4). Irregularly-shaped lightand dark FeO contents of kutnohorite and rhodochrosite are patches alternate in the samples; the sizes of the 0.0À0.3 and 0.1À0.3 wt.%, respectively. patches as well as the proportion of the two Mid-IR spectra of the separated parts of sample phases change from site to site (Fig. 4aÀd). At H4 confirm the different chemical composition of lower magnification, the dark veinlet in sample the two carbonate phases (Fig. 5). The matrix of H4 exhibits an apparently homogeneous appear- the concretion is mainly composed of rhodochro- ance (Fig. 4e); however, crystals with feather-like site with vibration bands n2 at865 cm À1 and n4at compositional patterns are seen at higher magni- 725 cmÀ1. Vibration bands (n2 at872 cm À1 and fication (Fig. 4f). The lightphase in theBSE n4 at717 cm À1) for the dominant carbonate of the images is a Mn-rich carbonate (45.5À49.5 wt.% veinlet are shifted towards higher and lower MnO, Table 2) with a small Ca and Mg content frequencies, respectively, compared with rhodo- (4.1À7.0 wt.% CaO and 0.8À1.0 wt.% MgO) chrosite spectra and indicate a Ca-dominant

FIG. 4. BSE images showing the texture of the concretions. (a,b) Sample H1 with dark phase (kutnohorite) and light phase (rhodochrosite). (c) Dark kutnohorite displacing light rhodochrosite in sample H3. (d) Matrix of sample H4. (e,f) Kutnohorite veinlet in sample H4 at higher magniﬁcation showing a feather-like compositional pattern.

498 TABLE 2. Chemical composition of the carbonate phases in the concretions determined by EMP analysis (number of analyses are indicated in parentheses; the remainder were acquired by pointmeasurements).

———— Oxide (wt.%) ———— —— Atom fraction (a.p.f.u.) —— ——— Carbonate (mol.%) ——— Sample code and partCaO MnO MgO FeO Ca Mn Mg Fe CaCO 3 MnCO3 MgCO3 FeCO3 C

Kutnohorite (dark phase on BSE images, Fig. 4) A H1, part2 39.10 11.95 1.95 0.05 1.524 0.368 0.106 0.002 76.20 18.40 5.30 0.10 HUNGARY KUTNOHORITE, -RICH H1, part3 37.03 12.35 2.71 0.32 1.457 0.384 0.148 0.010 72.85 19.20 7.40 0.50 H3, part1 38.36 12.47 2.03 0.00 1.503 0.386 0.111 0.000 75.15 19.30 5.55 0.00 H4, part2 38.14 13.26 1.71 0.26 1.490 0.409 0.093 0.008 74.50 20.45 4.65 0.40 H4, part3 (veinlet)(4) 39.89 11.21 1.25 0.20 1.575 0.350 0.069 0.006 78.75 17.50 3.45 0.30

499 H4, part3 (veinlet)(10) 39.51 10.26 1.34 0.24 1.591 0.326 0.075 0.007 79.55 16.30 3.75 0.35 H4, part3 (veinlet)(7) 40.51 10.51 1.21 0.18 1.600 0.328 0.066 0.005 80.00 16.40 3.30 0.25 H4, matrix (3) 37.95 12.38 1.43 0.31 1.519 0.392 0.080 0.010 75.95 19.60 4.00 0.50 Average 38.81 11.80 1.70 0.20 1.532 0.368 0.094 0.006 76.62 18.39 4.68 0.30 Rhodochrosite (light phase on BSE images, Fig. 4) H1, part2 4.16 49.55 0.88 0.27 0.186 1.750 0.055 0.009 9.30 87.50 2.75 0.45 H1, part3 4.13 49.48 0.96 0.22 0.184 1.748 0.060 0.008 9.20 87.40 3.00 0.40 H3, part1 4.92 48.84 0.64 0.05 0.221 1.737 0.040 0.002 11.05 86.85 2.00 0.10 H4, part2 7.04 45.54 0.91 0.06 0.317 1.623 0.057 0.002 15.85 81.15 2.85 0.10 H4, matrix (4) 5.38 45.53 0.84 0.26 0.254 1.681 0.055 0.009 12.70 84.05 2.75 0.45 Average 5.13 47.69 0.85 0.17 0.232 1.708 0.053 0.006 11.52 85.39 2.67 0.30 M. POLGA´ RI ET AL.

FIG. 5. IR absorbance spectra for sample H4. The vibrational bands n3, n2 and n4, characteristic of carbonates, are À1 indicated between 1800 and 500 cm . The veinlet contains a Ca-Mn carbonate (Ca-kutnohorite) with some MnCO3 (rhodochrosite), while its adjacent matrix is mainly composed of rhodochrosite.

carbonate, according to Bo¨ttcher et al. (1992, counts. The spectra of the areas with moderate CL 1993). have lower intensities than that of the brightly luminescent areas. The veinlets with moderate CL show higher spectral intensity than their bright Cathodoluminescence microscopy and spectroscopy margin. Our samples show moderate and bright orange-red According to the mineralogical (XRD) and luminescence in the form of layers, inﬁltration-like textural (EMPA) studies the moderately and textures and veinlets (Fig. 6). In the matrix of the brightly luminescent carbonate phase with a concretions, areas with moderate and bright CL 630 nm emission peak is kutnohorite. The non- can be observed alongside non-luminescentareas luminescent phase in the matrix is rhodochrosite. (Fig. 6b,d,f). Veinlets in sample H4 exhibit moderate red-brownish CL (Fig. 6f,h). The crys- TEM study tals in the veinlets do not show zoning, only slight inhomogeneity, but their rims have brighter orange A small fragmentof theluminescentveinletfrom CL. The borders of the veinlets show an outer part concretion H4 was investigated by TEM in order with moderate red-brown CL and an inner part to determine whether or not the excess Ca with bright orange-red CL (Fig. 6f,h). measured by EMPA with respect to a The CL spectra of the luminescent phases, in stoichiometric CaMn(CO3)2 is related to sub- both the matrix and veinlets of the concretions, microscopic domains of calcite that cannot be show one emission peak at629 À636 nm (Fig. 7), resolved by EMPA. Both monocrystalline parts without any visible shoulder. No difference in the (grains) and polycrystalline aggregates were position of this peak is observed in the spectra of identiﬁed within the analysed sample. moderately and brightly luminescent areas and A 110*-c* SAED pattern in Fig. 8a shows a only the peak intensities vary, at 3500À15000 5.4 A˚ periodicity along c*, which indicates that

FIG.6(facing page). Polarized light and ‘hot-cathode’ CL micrographs of the Mn-carbonate concretions. (a,b) Layered material with CL of different intensity in sample H1. (c,d) Moderately and brightly luminescent areas in sample H3. (e,f) Homogeneously luminescent carbonate veinlet and partly luminescent matrix in sample H4 (m: moderate CL, b: bright CL). (g,h) Luminescent carbonate veinlet with brighter rim in sample H4.

500 CA-RICH KUTNOHORITE, HUNGARY

501 M. POLGA´ RI ET AL.

FIG. 7. CL spectra of the luminescent carbonate phases (matrix and veinlet) in the concretions. All the experimental spectra are uniformly characterized by one emission band with a maximum at ~630 nm.

the grain has a dolomite-type double-carbonate of a calcite-type single-carbonate structure. structure. This supports the XRD results. The However, the missing diffraction rings with d values of polycrystalline aggregates measured respect to the dolomite structure (at 5.4 A˚ , from SAED patterns (Fig. 8b) are close to those 4.1 A˚ , 2.75 A˚ and 2.6 A˚ , JCPDS #84-1291)

FIG. 8. Images of the luminescent veinlet in sample H4. (a) 110*-c* SAED pattern of a dolomite-type double- carbonate grain. (b) SAED pattern of a polycrystalline aggregate. (c) Bright-field image of a grain with single crystal-type SAED pattern. The strongly changing contrast on a 10 nm scale is due either to bending of the individual domains or a high defectdensity.( d) Bright-field and (e) dark-field images of a polycrystalline aggregate. The dark- field image was made with the diffraction arc at 2.9 A˚ corresponding to the (104), the most intense, reflection. The bright field image shows the bent, mosaic microstructure and the corresponding dark-field image shows the sizes and distribution of domains within the aggregate.

502 CA-RICH KUTNOHORITE, HUNGARY

have low intensities (<2%). This observation may elevated Ca content (>60 mol.% CaCO3) be explained by an incomplete random orientation compared to its standard composition was also of the polycrystalline aggregates. Bright- and described from La˚ngban (Sweden; Gabrielson and dark-field images of individual grains and the Sundius, 1966), Levane, Upper Valdarno (Italy; polycrystalline aggregates show an irregularly Bini and Menchetti, 1985), Fujikura (Japan; undulating and mosaic microstructure Tanida and Kitamura, 1982) and Kremikovtsi (Fig. 8cÀe). The patches of strong contrast may (Bulgaria; Vassileva et al., 2003) (Table 3, arise from strain (bending) of individual domains Fig. 9). While XRD indicates one phase in the or, alternatively, be due to a high defect density. U´ rku´t kutnohorite, EMPA data revealed micro- The domain size is a few tens of nanometres. scale heterogeneity in the veinlet type, which The chemical compositions of the individual seems to be a common feature in kutnohorites crystals and polycrystalline aggregates were also (Zˇ ak and Povondra, 1981; Barber and Khan, measured (EDS). All measurements give a very 1987). similar composition, with no significant differ- Two structural models for excess-Ca dolomite ence between the grains and polycrystalline have been proposed. Reeder (2000) found excess- aggregates. The cation proportions (40 at.% Ca, Ca dolomites with a mixed proportion of Ca and 8À10 at.% Mn, 2 at.% Mg) are consistent with the Mg in both structural sites, while Drits et al. EMPA data. (2005) described dolomites with Ca in the A positions and mixed Ca and Mg cations in the B layers. Both types show ‘c’-type superstructure Discussion reflections in SAED due to ordering in the basal The luminescent carbonate in the U´ rku´tMn- layers (Reeder, 1992). The SAED patterns of the carbonate concretions is kutnohorite, character- U´ rku´t kutnohorite veinlet, however, show only ized by a considerable excess of Ca. The ‘a’- and ‘b’-type reflections characteristic of the compositional variation of kutnohorites extends dolomite structure. No ‘c’- or ‘d’-type reflections, from Fe-Mg-rich varieties to Ca-rich varieties the latter resulting from periodic stacking defects (see Fig. 9, data in Table 3). In nature, Ca- (Reeder, 1992), were observed. Modulations, deficient(Frondel and Bauer, 1955) and Ca- striations and ribbon microstructures, which excessive (Gabrielson and Sundius, 1966) non- appear in, and seem to be typical of, Ca-rich stoichiometric kutnohorites are known to occur dolomite in pre-Holocene sediments (e.g. Reeder, (Reeder, 1983, Table 3). Kutnohorite with 1992) and which also occur in kutnohorite from the type locality (Barber and Khan, 1987), were not detected. U´ rku´t kutnohorite is a heterogeneous material with mosaic microstructure in which the mis-orientation of the sub-microscopic domains or crystallites is sometimes very slight, which resembles the Holocene dolomite and high-Mg calcite reported by Reeder (1981, 1992). The Mn2+ ion has an ionic radius between those of Ca2+ and Mg2+ (Reeder, 1983). As a resultit can, theoretically, occupy both Ca (A) and Mg (B) sites in the dolomite structure. The Mn substitutes in the A site and has a strong preference for the B site of dolomite (e.g. Lumsen and Lloyd, 1984). However, such ordering may be true only for small Mn concentrations and may not apply to kutnohorite (Peacor et al., 1987). Large concentrations of Mn may only have a slight tendency to FIG. 9. Chemical composition of kutnohorite from U´ rku´t order and so highly-disordered kutnohorite can and from other occurrences on the CaMn(CO3)2- also form (Peacor et al., 1987). In nature, CaMg(CO3)2-CaFe(CO3)2 ternary diagram (modified disordered as well as ordered kutnohorite can after Vassileva et al., 2003). Legend À filled circle: exist; Peacor et al. (1987) reported almost U´ rku´t, Hungary; open circles: other occurrences. For completely disordered and largely ordered kutno- localities and data, see Table 3. horites with similar (near stoichiometric)

503 TABLE 3. Chemical composition (mol.%) of kutnohorite from a range of occurrences (after Vassileva at el., 2003).

Sample no. CaCO3 MnCO3 MgCO3 FeCO3 Locality References 1 54.46 22.94 19.34 3.26 Ribnitsa, Bulgaria Vassileva et al. (2003) 2 53.03 26.20 14.64 6.13 Ribnitsa, Bulgaria Kolkovski et al. (1980) 3 51.13 20.12 19.01 9.74 Ruen, Bulgaria Dragov (1965)

4* 49.00 23.55 13.60 13.85 Martinovo, Bulgaria Dragov and Neykov (1991) POLGA M. 5 51.50 42.00 5.77 0.73 Franklin, New Jersey, USA Frondel and Bauer (1955) 6 51.00 35.85 10.12 3.03 Kutna Hora, Czech Republic Trdlicˇka (1963)

504 7 51.22 23.73 12.42 12.63 Kutna Hora, Czech Republic Trdlicˇka (1963) ´

8 47.45 34.75 11.30 6.50 Chvaletice, Czech Republic Zˇ ak and Povondra (1981) RI

9 48.53 24.90 24.81 1.76 Ryuˆjima, Japan Tsusue (1967) AL. ET 10 74.00 19.00 7.00 À La˚ngban, Sweden Gabrielson and Sundius (1966) 11* 66.95 26.35 6.50 0.20 Levane, Upper Valdarno, Italy Bini and Menchetti (1985) 12 64.03 31.06 4.12 0.79 Fujikura, Japan Tanida and Kitamura (1982) 13 68.86 25.37 5.36 0.41 Kremikovtsi, Bulgaria Vassileva et al. (2003) 14* 49.00 49.60 1.10 0.30 Bald Knob, North Carolina, USA Peacor et al. (1987) 15* 48.00 47.70 1.30 1.20 Sterling, New Jersey, USA Peacor et al. (1987) 16* 60.70 35.10 2.50 0.80 Sterling, New Jersey, USA Peacor et al. (1987) 17* 76.62 18.39 4.68 0.30 U´ rku´t, Hungary Recent study

* EMP data CA-RICH KUTNOHORITE, HUNGARY chemical compositions. The large Ca excess in the Mn2+-activated emission of the A (Ca) site of the U´ rku´t kutnohorite suggests that some Ca dolomite or, indeed, of calcite (CaÀO distance: occupies the B site; however, the site preference 2.36 A˚ ). Furthermore, due to the probable mixed of Mn is notclear (only B site or both A and B Ca-Mn composition, its CL band should have a sites). shorter wavelength than that of any luminescent The CL spectra of non-stoichiometric dolomite pure Mn-bearing carbonate (rhodochrosite with characteristically have two peaks. El Ali et al. MnÀO distance: 2.19 A˚ ). This is supported by our (1993) documented a small variability of KD preliminary CL data from rhodochrosite with dull 2+ 2+ values (KD =MnMg site/MnCa site) between 2.7 and orange luminescence in other U´ rku´tsamples, 5.4 compared with those of stoichiometric which has an emission band maximum at dolomites (KD = 1.8À24.0). This observation ~635À640 nm. However, itremains undecided suggests that if excess Ca is incorporated into the as to whether or not the CL spectra of the structure, it is possible for significant Mn2+ to kutnohorite have a single emission band or, due to substitute at the Ca site, i.e. there is significant a quite similar Ca-Mn occupation of A and B cation disorder. According to the dolomite-type positions, it is composed of two peaks with nearly structure of kutnohorite, a composite lumines- identical wavelengths. In the latter case it is not cence band consisting of emission bands of Mn2+ possible to distinguish the two emission bands and in A and B positions, respectively, can be it is also difficult to determine at which site Mn expected. However, the CL spectra of U´ rku´t causes luminescence or shows self-quenching. kutnohorite consistently have a single emission For self-quenching, the absolute concentration band at~630 nm, withdifferences observed only of the activator elements, and therefore the in peak intensities. Similar spectra with a single distance between Mn2+ ions, seems to be a critical peak at ~630 nm have been reported for dolomite factor. Marfunin (1979) pointed out that the with calcite domains and Mg-calcite with maximum CL intensity often occurs at ~0.1À1% magnesite clusters (Habermann et al., 1996, concentration level of the activator ion and at 2000; Habermann, 2002). Our XRD and TEM greater concentrations the self-quenching is domi- studies show that U´ rku´t kutnohorite is monophase nant. Machel et al. (1991) suggested that bright and its CL spectra more closely resemble those of Mn2+-activated CL extends up to ~5 wt.% Mn red luminescentdolomites,where only one peak (log Mn = 4.7). At greater Mn concentrations, dull appears due to Mn substituting for Mg. luminescence exists due to self-quenching, prob- The position of the Mn2+-activated emission ably up to the rhodochrosite composition (48 wt.% band in the CL spectra of kutnohorite can be Mn, log Mn = 5.68). This suggestion is strongly explained by Crystal Field Theory. As the Mn2+ supported by the excess-Ca kutnohorite from ions (3d5) are very sensitive to their immediate U´ rku´t, which shows moderate-bright cathodo- environment, changes in the average bond length luminescence at8.2 À10.3 wt.% Mn concentration between oxygen and the metal ions cause (log Mn = 4.9À5.0, Fig. 10). Our data are close to wavelength shifts in the luminescence emission the luminescent manganocalcite (11.5 wt.% Mn, of carbonates. There is a linear correlation 0.5 wt.% Fe) described by Walker et al. (1989) between the wavelength of the luminescence (Fig. 10). The Fe content (up to 2400 ppm, log Fe and the metal-oxygen (MÀO) distance in the = 3.4), determined in the U´ rku´tkutnohorite, could thermoluminescence and ionoluminescence also partially influence its luminescent character spectra of carbonates (Calderoń et al., 1996; with some Fe-quenching. Calvo de Castillo et al., 2006). The same happens In summary, the moderate- and bright-orange in cathodoluminescence: the CL emission shifts to luminescent carbonate in the Mn-carbonate shorter wavelengths as the bond-length increases. concretions from U´ rku´t is identified as kutnohorite In an ordered stoichiometric dolomite the with considerable Ca excess (average composi- ˚ M(Mg)ÀO distance in B position is 2.084 A tion: Ca1.53Mn0.37Mg0.09Fe0.01), which appears in resulting in a ~660 nm peak for the Mn2+- the form of layers, infiltrations and veinlets. The activated CL, whereas the M(Ca)ÀO distance in A TEM results show that the luminescent carbonate position is 2.38 A˚ causing a luminescence band at is a single-phase mineral with dolomite-like ~575 nm (Walker et al., 1989; El Ali et al., 1993; structure and an undulating-mosaic microstruc- Calderoń et al., 1996). ture. The CL spectra of U´ rku´t kutnohorite show a The emission band in the CL spectra of the single-peak pattern with a maximum at ~630 nm U´ rku´t kutnohorite has a longer wavelength than due to Mn2+ activation. U´ rku´t kutnohorite is a

505 M. POLGA´ RI ET AL.

References Barber, D.J. and Khan, M.R. (1987) Composition- induced microstructures in rhombohedral carbonates. Mineralogical Magazine, 51,71À86. Bini, B. and Menchetti, S. (1985) Kutnohorite from Levane Upper Valdarno (Italy). Periodico di Mineralogia, 54,61À66. Bo¨ttcher, M.E., Gehlken, P.L. and Usdowski, E. (1992) Infrared spectroscopic investigations of the calcite- rhodochrosite and parts of the calcite-magnesite mineral series. Contributions to Mineralogy and Petrology, 109, 304À306. Bo¨ttcher, M.E., Gehlken, P.L., Usdowski, E. and Reppke, V. (1993) An infrared spectroscopic study of natural and synthetic carbonates from the quaternary system CaCO3-MgCO3-FeCO3-MnCO3. Zeitschrift der Deutschen Geologischen Gesellschaft, 144, 478À484. Burke, I.T. and Kemp, A.E.S. (2002) Microfabric analysis of Mn-carbonate laminae deposition and Mn-sulfide formation in the Gotland Deep, Baltic Sea. Geochimica et Cosmochimica Acta, 66, 1589À1600. Calderoń, T., Townsend, P.D., Beneitez, P., Garcia- Guinea, J., Millań, A., Rendell, H.M., Tookey, A., Urbina, M. and Wood, R.A. (1996) Crystal field effects on the thermoluminescence of manganese in FIG. 10. CL behaviour of carbonates as a function of Mn carbonate lattices. Radiation Measurements, 26, vs. Fe concentration (ppm on logarithmic scale) 719À731. (modified after Machel et al., 1991). The U´ rku´t Calvo de Castillo, H., Ruvalcaba Sil, J.L., A´ lvarez, kutnohorite samples support the theory that intensive M.A., Beneitez, P., Millań, M.A. and Calderoń, T. luminescence can exist up to 10 wt.% Mn and 2400 ppm (2006) Relationship between ionoluminescence Fe concentrations (log Mn = 5, log Fe = 3.4). Legend À emission and bond distance (MÀO) in carbonates. open circles: kutnohorite in concretion H1; filled circles: Nuclear Instruments and Methods in Physics kutnohorite in concretion H4; grey square: mangano- Research, B249, 217À220. calcite with 11.5 wt.% Mn and 0.5 wt.% Fe (Walker et Dragov, P. (1965) Mineralogical and geochemical study al., 1989). of the Osogovo Pb-Zn deposits. Works on the Geology of Bulgaria, Series Geochemistry, natural example which supports the earlier Mineralogy and Petrography, 5, 209À265 (in assumption that Mn does not show complete Bulgarian with German abstract). self-quenching in carbonates at ~8À10 wt.% Mn Dragov, P. and Neykov, H. (1991) Carbonate petrology ˇ concentrations, with up to 2400 ppm Fe. of the Ciprovci ore zone. Geologica Balcanica, 21, 69À98. Drits, V.A., McCarty, D.K., Sakharov, B. and Milliken, Acknowledgements K.L. (2005) New insight into structural and compositional variability in some ancient excess We thank E. Vass (Institute of Chemistry, Eo¨tvo¨s Ca-dolomite. The Canadian Mineralogist, 43, L. University, Budapest) for the IR measure- 1255À1290. ments. Use of the Reliotron ‘cold-cathode’ CL El Ali, A., Barbin, V., Calas, G., Cervelle, B., Ramseyer, microscope was possible with the financial help of K. and Bouroulec, J. (1993) Mn2+-activated lumines- the Hungarian National Research Fund (OTKA) cence in dolomite, calcite and magnesite: quantita- to the Institute for Geochemical Research (M tive determination of manganese and site distribution 045792). M. Polga´ri acknowledges her grant by EPR and CL spectroscopy. Chemical Geology, OMFB-01680/2006. B. Bajnoćzi acknowledges 104, 189À202. the financial support of OTKA (post-doctoral Fan, D., Hein, J.R. and Ye, J. (1999) Ordovician reef- grantD 048631). hosted Jiaodingshan Mn-Co deposit and Dawashan

506 CA-RICH KUTNOHORITE, HUNGARY

Mn deposit, Sichuan Province, China. Ore Geology Habermann, D., Meijer, J., Neuser, R.D., Richter, D.K., Reviews, 15, 135À151. Rolfs, C. and Stephan, A. (1999) Micro-PIXE and Farkas, L., Bolzenius, B.H., Scha¨fer, W. and Will, G. quantitative cathodoluminescence spectroscopy: (1988) The crystal structure of kutnohorite combined high resolution trace element analyses in CaMn(CO3)2. Neues Jahrbuch fu¨r Mineralogie minerals. Nuclear Instruments and Methods in Monatshafte, 12, 539À546. Physics Research, B150, 470À477. Frondel, C. and Bauer, L. (1955) Kutnahorite: a Habermann, D., Neuser, R.D. and Richter, D.K. (2000) manganese dolomite: CaMn(CO3)2. American Quantitative high resolution spectral analysis of Mineralogist, 40, 748À760. Mn2+ in sedimentary calcite. Pp. 331À358 in: Gabrielson, O. and Sundius, N. (1966) Ca-rich Cathodoluminescence in Geosciences (M. Pagel, V. kutnahorite from La˚ngban, Sweden. Arkiv fo¨r Barbin, P. Blanc and D. Ohnenstetter, editors). Mineralogi och Geologi, 4, 287À289. Springer Verlag, Berlin. Gaft, M.L., Gorobets, B.S., Naumova, I.S., Mironova, Hein, J.R., Fan, D., Ye, J., Liu, T. and Yeh, H.-W. N.A. and Grinvald, G.A. (1981) Link of luminescent (1999) Composition and origin of Early Cambrian properties with crystal-chemical peculiarities in Tiantaishan phosphorite-Mn carbonate ores, Shaanxi manganese minerals. Mineralogicheskij Zurnal, 2, Province, China. Ore Geology Reviews, 15,95À134. 80À89 (in Russian). Kolkovski, B., Bogdanov, K. and Petrov, S. (1980) Gaft, M., Reisfeld, R. and Panczer, G. (2005) Modern Mineralogy, geochemistry and genetic features of the Luminescence Spectroscopy of Minerals and deposits along Goliam Palas-Ribnica fault, Madan Materials. Springer Verlag, Berlin. ore field. Annual of the Sofia University ‘St.K. Gillhaus, A., Habermann, D., Meijer, J. and Richter, Ohridski’, Faculty of Geology and Geography, Vol. D.K. (2000) Cathodoluminescence spectroscopy and 1 À Geology, 74,97À139 (in Bulgarian with English micro-PIXE: combined high resolution Mn-analyses abstract). in dolomites À Firstresults. Nuclear Instruments and Krajewsky, K.P., Lefeld, J. and Lacka, B. (2001) Early Methods in Physics Research, B161À163, 842À845. diagenetic process in the formation of carbonate- Gillhaus, A., Richter, D.K., Meijer, J., Neuser, R.D. and hosted Mn ore deposit (Lower Jurassic, Tatra Stephan, A. (2001) Quantitative high resolution Mountains) as indicated from its carbon isotopic cathodoluminescence spectroscopy of diagenetic record. Bulletin of the Polish Academy of Sciences, and hydrothermal dolomites. Sedimentary Geology, Earth Sciences, 49,13À29. 140, 191À199. Large, R.R., Allen, R.L., Blake, M.D. and Herrmann, W. Gorobets, B., Gaft, M. and Laverova, L. (1978) (2001) Hydrothermal alteration and volatile element Photoluminescence of manganese minerals. Journal halos for the Rosebery K lens volcanic-hosted of Applied Spectroscopy, 28, 750À752 (in Russian). massive sulphide deposit, western Tasmania. Go¨tte, T. and Richter, D.K. (2004) Quantitative high- Economic Geology, 96, 1055À1072. resolution cathodoluminescence spectroscopy of Lumsen, D.N. and Lloyd, R.V. (1984) Mn(II) partition- smithsonite. Mineralogical Magazine, 68, 199À207. ing between calcium and magnesium sites in studies Gutzmer, J. and Beukes, N.J. (1996) Mineral paragen- of dolomite origin. Geochimica et Cosmochimica esis of the Kalahari manganese field, South Africa. Acta, 48, 1861À1865. Ore Geology Reviews, 11, 405À428. Machel, H.G. (2000) Application of cathodolumines- Gutzmer, J. and Beukes, N.J. (1998) The manganese cence to carbonate diagenesis. Pp. 271À301 in: formation of the Neoproterozoic Pengana Group, Cathodoluminescence in Geosciences (M. Pagel, V. India, revision of an enigma. Economic Geology, 93, Barbin, P. Blanc and D. Ohnenstetter, editors). 1091À1102. Springer Verlag, Berlin. Habermann, D. (2002) Quantitative cathodolumines- Machel, H.G., Mason, R.A., Mariano, A.N. and Mucci, cence (CL) spectroscopy of minerals: possibilities A. (1991) Causes and emission of luminescence in and limitations. Mineralogy and Petrology, 76, calcite and dolomite. Pp. 9À25 in: Luminescence 247À259. Microscopy and Spectroscopy: Qualitative and Habermann, D., Neuser, R.D. and Richter, D.K. (1996) Quantitative Applications (C.E. Barker and O.C. Hochauflo¨sende Spektralanalyse der Kopp, editors). Society of Sedimentary Geologists, Kathodolumineszenz (KL) von Dolomit und Calcit: ShortCourse no. 25. Beispiele der Mn- und SEE-aktivierten KL in Marfunin, A.S. (1979) Spectroscopy, Luminescence and Karbonatsedimenten. Zentralblatt fu¨r Geologie und Radiation Centres in Minerals. Springer Verlag, Palaöntologie Teil I, 1995, 145À157. Berlin. Habermann, D., Neuser, R.D. and Richter, D.K. (1998) Marshall, D.J. (1988) Cathodoluminescence of Low limitof Mn 2+-activated cathodoluminescence of Geological Materials. Unwin-Hyman, Boston. calcite: state of art. Sedimentary Geology, 116,13À24. Mason, R.A. (1987) Ion microprobe analysis of trace

507 M. POLGA´ RI ET AL.

elements in calcite with an application to the origin through carbonate dissolution. Sedimentary cathodoluminescence zonation of limestone cements Geology, 177,87À96. from the Lower Carboniferous of South Wales, U.K. Reeder, R.J. (1981) Electron optical investigation of Chemical Geology, 64, 209À224. sedimentary dolomites. Contributions to Mineralogy Mason, R.A. and Mariano, A.N. (1990) and Petrology, 76, 148À157. Cathodoluminescence activation in manganese-bear- Reeder, R.J. (1983) Crystal chemistry of the rhombohe- ing and rare earth-bearing synthetic calcites. dral carbonates. Pp. 1À47 in: Carbonates: Chemical Geology, 88, 191À206. Mineralogy and Chemistry (R.J. Reeder, editor). Neuser, R.D. (1995) A new high-intensity cathodolumi- Reviews in Mineralogy, 11, Mineralogical Society of nescence microscope and its application to weakly America, Washington D.C. luminescing minerals. Bochumer Geologische und Reeder, R.J. (1992) Carbonates: growth and alteration Geotechnische Arbeiten, 44, 116À118. microstructures. Pp. 380À424 in: Minerals and Nyame, F.K., Beukes, N.J., Kase, K. and Yamamoto, M. Reactions at the Atomic Scale: Transmission (2003) Compositional variations in manganese Electron Microscopy (P.R. Buseck, editor). carbonate micronodules from the Lower Reviews in Mineralogy, 27, Mineralogical Society Proterozoic Nsuta deposit, Ghana: product of of America, Washington D.C. authigenic precipitation or postformational diagen- Reeder, R.J. (2000) Constraints on cation order in esis? Sedimentary Geology, 154, 159À175. calcium-rich sedimentary dolomite. Aquatic Ozturk, H. and Frakes, L.A. (1995) Sedimentation and Geochemistry, 6, 213À226. diagenesis of an Oligocene manganese depositin a Richter, D.K., Go¨tte, Th., Go¨tze, J. and Neuser, R.D. shallow sub-basin of the Paratethys: Thrace Basin, (2003) Progress in application of cathodolumines- Turkey. Ore Geology Reviews, 10, 117À132. cence (CL) in sedimentary petrology. Mineralogy Peacor, D., Essene, E. and Gaines, A. (1987) Petrologic and Petrology, 79, 127À166. and crystal-chemical implications of cation order- Spo¨tl, C. (1991) Cathodoluminescence of magnesite: disorder in kutnahorite [CaMn(CO ) ]. American 3 2 examples from the Alps. Geology, 19,52À55. Mineralogist, 72, 319À328. Tanida, K. and Kitamura, T. (1982) Mineralogy and Polga´ri, M. (2001) Contribution of volcanic material? À thermal transformation of kutnahorite from Fujikura A new aspect of the genesis of the black shale-hosted mine, Iwate Prefecture, with the subsolidus relation Jurassic Mn-carbonate ore formation, U´ rku´tBasin, of system CaO-manganese oxide at 1100ºC and Hungary. Acta Geologica Hungarica, 44, 419À438. 1400ºC in air. Journal of the Japanese Association of Polga´ri, M., Okita, P.M. and Hein, J.R. (1991) Stable isotope evidence for the origin of the U´ rku´t Mineralogists, Petrologists and Economic manganese ore deposit, Hungary. Journal of Geologists, 77, 227À234. Sedimentary Petrology, 61, 384À393. Trdlicˇka, Z. (1963) Mineralogick vyzkum cˇeskch Polga´ri, M., Szabo´, Z. and Szederkeńyi, T. (2000) kutnohoritu. Sbornik na´rodnı´ho muzea v Praze, 19, Manganese Ores in Hungary.In Commemoration of 163À174. Professor Gyula Grasselly. Szeged Regional Tsusue, A. (1967) Magnesian kutnahorite from Ryuˆjima Committee of the Hungarian Academy of Sciences, mine, Japan. American Mineralogist, 52, Juha´sz Publishing House, Szeged, Hungary. 1751À1761. Polga´ri, M., Szabo´-Drubina, M. and Szabo´, Z. (2004) Vassileva, M., Dobrev, S. and Damyanov, Z. (2003) Theoretical model for Jurassic manganese miner- Comparative characteristics of endogenic kutnahor- alization in U´ rku´t, Hungary. Bulletin of Geosciences, ite from Ribnitsa deposit and exogenic kutnahorite 79,53À61. from Kremikovtsi deposit. Annual of the University Polga´ri, M., Philippe, M., Szabo´-Drubina, M. and To´th, of Mining and Geology ‘St.Ivan Rilski’, Part I, M. (2005a) Manganese-impregnated wood from a Geology and Geophysics, 46, 195À200. Toarcian manganese ore deposit, Epleńy Mine, Walker, G., Abumere, O.E. and Kamaluddin, B. (1989) 2+ Bakony Mts, Transdanubia, Hungary. Neues Luminescence spectroscopy of Mn centres in rock- Jahrbuch fu¨r Geologie und Palaöntologie À forming carbonates. Mineralogical Magazine, 53, Monatshefte, 2005/3, 175À192. 201À211. Polga´ri, M., Szabo´, Z., Szabo´-Drubina, M., Hein, R.J. Zˇ ak, L. and Povondra, P. (1981) Kutnohorite from the and Yeh, H-W. (2005b) A porous silica rock Chvaletice pyrite and manganese deposit, East (‘tripoli’) in the footwall of the Jurassic U´ rku´t Bohemia. Tschermaks Mineralogische und manganese deposit, Hungary: composition and Petrographische Mitteilungen, 28,55À63.

508