Journal of Earth Science  2018, Vol. 29 Issue (5): 989-1004   PDF    
0
Natural End Member Samples of Pyrope and Grossular: A Cathodoluminescence-Microscopy and -Spectra Case Study
Hans-Peter Schertl1,2, Joana Polednia1,3, Rolf D. Neuser1, Arne P. Willner1    
1. Ruhr-University Bochum, Institute of Geology, Mineralogy and Geophysics, D-44780 Bochum, Germany;
2. College of Earth Science & Engineering, Shandong University of Science and Technology, Qingdao 266590, China;
3. Bayerisches Geoinstitut, Universität Bayreuth, D-95440 Bayreuth, Germany
ABSTRACT: Garnet is one of the most significant minerals in metamorphic rocks, that provides key information on prograde, peak-metamorphic and retrograde parts of the pressure-temperature (PT) path. Such results require a detailed knowledge of its different growth domains. For iron-poor compositions, the cathodoluminescence (CL) microscopy is an important and often overlooked method and allows to identify the internal structures of all garnet grains in one thin section within only a few seconds. The advantage of the CL-microscope is to deliver low magnification images in true color, not only of garnet but also, for instance, of other rock forming silicates, carbonates, sulfates, etc., of metamorphic, but also of sedimentary and magmatic origin, using polished thin sections. Internal structures of grossular from Mexico and pyrope from the Italian Alps were characterized and visualized by CL-microscopy. The different growth domains were additionally studied using CL-spectra and electron microprobe (EMP) analysis. Grossular shows a patchy zonation in its core while in mantle and rim zones oscillatory zoning is observed. It contains zones of anomalous birefringence, zones of orange and bluish luminescence and zones lacking luminescence. Different but low amounts of the activator elements Mn2+ and Eu2+ are responsible for the orange and bluish luminescent domains. Pyrope is also characterized by oscillatory growth zones, shows a dull luminescent core with a change of crystal morphology during growth, and displays an increase of brightness from core towards rim-the outermost rim, however, is lacking luminescence. The different luminescent zones are characterized by different amounts of Dy3+, Tb3+, Sm3+ and Sm2+ as activator elements. Because of slow diffusion rates of activators such as the REEs Sm, Dy and Tb, it can be still possible to visualize possible prograde and/or peak pressure stage growth domains of garnet, even if later high temperature events may have homogenized the major element profiles. Such domains may help to identify respective assemblages of mineral inclusions, and hence these results can represent an integral part of a detailed PT path. Thus the CL-information can be used as an important pathfinder prior to supplementary investigations, as for instance EMP, ion probe, mineral or fluid inclusion studies.
KEY WORDS: pyrope    grossular    cathodoluminescence (CL)    oscillatory zoning    REE    CL-spectra    

INTRODUCTION

Garnets are key minerals in metamorphic rocks. The knowledge of their internal growth structures and respective compositional variations is indispensable for deriving reliable pressure-temperature (PT) paths, using either classical geothermobarometers (e.g., Ferry and Spear, 1978; Råheim and Green, 1974), multivariant reaction calculations with internal consistent thermodynamic datasets (e.g., Holland and Powell, 1990, 1985; Berman, 1988), or pseudosection techniques (e.g., Caddick and Kohn, 2013; Connolly, 1990; de Capitani, 1987). Furthermore, knowledge of garnet zoning is indispensable for dating of garnet with systems like Lu/Hf and Nd/Sm (e.g., Baxter and Scherer, 2013). Today it is mandatory to use two-dimensional X-ray element maps (e.g., Gross et al., 2008; Rollinson, 2003; Schumacher et al., 1998) in order to identify possible zoning, resorption and annealing features, and to link mineral inclusions and assemblages to distinct growth zones. However, the use of the cathodoluminescence (CL) microscope to identify and evaluate different growth zones of garnet is frequently underestimated (Götze et al., 2013; Schertl et al., 2004). Such CL-studies are undertaken using polished thin sections, covered with a conductible thin film of carbon or gold. As an advantage of the CL-microscopy, growth zones of garnet (or other luminescent minerals) can be identified within seconds, and within a few minutes all garnet crystals of an entire thin section can be studied in detail. The most promising crystals, finally, can be used for further studies, as for instance electron microprobe (EMP; Schertl et al., 2015, 2004; Sobolev et al., 2007) or ionprobe (e.g., Sobolev et al., 2011; Allan and Yardley, 2007) investigations. As a disadvantage, iron suppresses the luminescence and the method does not apply for iron rich garnet compositions. Such garnets, for instance showing almandine- or andradite-dominant compositions, are lacking luminescence. Thus, for this study, iron-poor, pale green, euhedral grossular from Lake Jaco (Sierra de la Cruz), Coahuila, Mexico (Fig. 1a) and whitish-pink pyrope, which is very close to its end-member composition and which comes from ultrahigh-pressure (UHP) metamorphic rocks of the southern Dora-Maira massif (Fig. 1b; see Ferrando et al., 2009; Schertl et al., 1991; Chopin, 1984) were investigated.

Download:
Figure 1. Studied garnet samples: (a) grossular from Mexico and (b) pyrope as a constituent of a pyrope quartzite rock from Masueria, Dora-Maira massif, Italy; (c) pyrope-grossular solid solution, diamondiferous UHP-calcsilicate rock, Kochetav massif, Kazakhstan; (d) oscillatory zoned grossular from jadeitite, Motagua fault zone, Guatemala; (e) euhedral oscillatory zoned and (f) subhedral grossular of a garnet-bearing marble, Cap Corse, Corsica; (g) euhedral grossular crystals (current study) from Mexico (small dark brownish portions=adhesive); (h) pyrope of a pyrope quartzite rock (current study), Masueria, Dora-Maira massif, Italy. (a), (b) Macroscopical views; (c)-(f) cathodoluminescence (CL) images; (g) plane polarized light (PPL); (h) crossed polars (XPL).

The high impact of studies to qualitatively illustrate internal structures of garnet using the CL-microscope, has already been demonstrated on different lithological rock types, as the pyrope- quartzite UHP-rocks from the Dora Maira massif (Schertl et al., 2004), where Chopin (1984) for the first time discovered metamorphic coesite and which are studied here in a more quantitative approach, using CL-spectra. A further example refers to garnets in UHP-metamorphic calcsilicates from the Kokchetav massif, which are described to contain patchy to vein-like (Schertl et al., 2015) as well as concentric zonations (Sobolev et al., 2011) of a brownish luminescence; they essentially consist of grossular- pyrope solid solution (Fig. 1c) and can contain metamorphic diamond (e.g., Schertl and Sobolev, 2013; Sobolev and Shatsky, 1990). Recently, the first author of the present work discovered grossular-rich luminescent concentric garnet crystals (Fig. 1d) in jadeitite in serpentinite from the Motagua fault zone, Guatemala, studied by Harlow (1994) and grossular-rich oscillatory zoned or inhomogenously structured luminescent crystals in a garnet- bearing marble layer, which is structurally on top of a meta-ophiolithic rock sequence from Cap Corse, Corsica (Figs. 1e, 1f; see Chopin et al., 2008). These newly discovered growth patterns of garnet from Guatemala and Corsica were not yet published and used here for comparative reasons. Neuser et al. (1995) mentioned red luminescent spessartine-pyrope-grossular garnet from UHP rocks from Lago di Cignana and Val Clavalité (see also Reinecke, 1998).

The luminescence can be caused by a structural incorporation of so-called "activator-elements", as for instance Mn2+ (as spessartine component in garnet), leading to broad emission bands in the CL-spectrum (e.g., Pagel et al., 2000). Divalent manganese is well known to represent an important activator element in carbonates of sedimentary origin (Richter et al., 2003). Low amounts of Mn2+ causing different luminescence colors are frequently observed in calcite, Mg-calcite and dolomite; thus limestone (Richter et al., 2014), marble and calcsilicate rocks (Schertl and Sobolev, 2013; Schertl et al., 2004; Houzar and Leichmann, 2003) as well as carbonatite and kimberlite (Sobolev et al., 2017) are promising candidates deriving internal structures of carbonate minerals. Note that the broad emission bands (e.g., Mn, Fe, Eu) can occur at different wavelength positions in different minerals; as a result the luminescence color generated by one single element can be different. In case of carbonate minerals, the activator element Mn2+ can produce yellow (pure calcite), orange (Mg-calcite), red (dolomite) or green (aragonite) colors (Götze et al., 2013; Schertl et al., 2004); thus compositional variations of carbonates can be easily detected. On the other hand, for instance trivalent REE can also serve as important activator elements, producing narrow emission bands. Such bands-independent of the mineral species studied-occupy essentially the same wavelength position (Götze, 2002; Fig. 5a therein). It is known that the amount of REE, causing a significant luminescence, can be below the detection limit of the electron microprobe or even the micro- PIXE (Proton Induced X-ray Emission; Habermann et al., 1999). Thus, for the current studies, besides measurements using the EMP, CL-spectra of different growth zones of garnet were performed in order to identify-if present-even tiny amounts of REE. For more detailed information on the physical and crystallographic background of CL, the reader is referred to Marfunin (1979), Pagel et al. (2000) and Götze et al. (2013).

Download:
Figure 2. Studied garnet samples: (a) oscillatory zoned grossular rim zone (CL; Mexico); (b) same as (a) under XPL, note that anomalous birefringence growth zones (a dominant one is marked in yellow) not necessarily correspond with the orange luminescent growth zones; (c) orange luminescent grossular patches in the core portion of garnet; (d) pyrope (17642-1) from the Dora-Maira massif (CL, respective element distribution maps see Fig. 3); (e) pyrope (17642-2) from the Dora-Maira massif (CL); (f) pyrope (17642-3) from the Dora-Maira massif (CL). Referenced yellow points represent spots of CL-spectra; white crosses with numbers refer to EMP-analyses (Tables 1 and 2).
Download:
Figure 3. Element distribution maps of Mg, Fe, Ca and Dy of pyrope (sample 17642-1).
Download:
Figure 4. CL-spectra of (a) grossular from the core portion showing different luminescence colors (orange: number 1, blue: 2, and weak: 3 in Fig. 2c); (b) details of (a) focusing on the Eu2+ peak with higher resolution (wavelength range 350-530 nm); (c) apatite inclusion compared to grossular host; (d) grossular from the oscillatory rim portion (numbers 1-4 in Figs. 2a, 2b refer to the four spectra, top to bottom).
Download:
Figure 5. CL-spectra (core, mantle and rim) of the three pyrope crystals studied.

This paper represents a case study to document the high impact of CL-studies on garnet as one of the most important and traditionally used minerals in metamorphic rocks for geothermobarometrical applications. Note, that other silicates as the three aluminium-polymorphs kyanite, andalusite and sillimanite, but likewise also zoisite, topaz, diopside, jadeite, wollastonite, spurrite, tilleyite, prehnite, plagioclase and K-feldspar as well as different SiO2-polymorphs, as a selection of the most important rock forming silicates, are representing key candidates for similar CL-studies (e.g., Schertl et al., 2018, 2015, 2012, 2005, 2004; Takahashi et al., 2017; Götze et al., 2013; Catlos et al., 2011; Parsons et al., 2008; Barwood, 2007; Sobolev et al., 2007; Satish-Kumar et al., 2006; Harlow and Sorensen, 2005; Dudley, 1976). Mineral abbreviations are after Whitney and Evans (2010); further ones are Schor (schorlomite) and Mori (morimotoite).

1 ANALYTICAL TECHNIQUES

The CL-examinations (see Figs. 1, 2) were performed using a HC1-LM hot cathode microscope at Ruhr-University Bochum; the beam energy was 14 keV and the beam current density ~9 μA/mm2 on the sample surface. The device used allowed simultaneous investigation of polished thin sections under linear polarized light (LPL), crossed polars (XPL) and the electron beam (for details see Götze et al., 2013; Schertl et al., 2004). Photos were taken by a highly sensitive digital microscope camera (DP73, Olympus). For spectral analyses a quartz light guide was used connecting the CL-microscope to a triple grating spectrograph (Model 275, Acton Research). Spectra were obtained by an ultra-sensitive cooled CCD detector (PIXIS, Princeton Instruments) that collects the light at the exit of the spectrograph. The minimal size of the measured area is about 5 μm in diameter.

Electron microprobe analyses and element distribution maps (Fig. 3) were performed using a CAMECA SX-50 at Ruhr-University Bochum. Operating conditions were 20 kV acceleration voltage and 20 nA beam current. Elements considered for the garnet analyses were Si, Ti, Al, Mg, Fe, Mn, Ca; the calculation of garnet end-member compositions was performed after Locock (2008).

2 PETROGRAPHICAL CHARACTERISTICS OF SAMP-LES

Grossular used for the current study is cm-sized; it forms euhedral dodecahedron crystals, pale greenish in color. The sample chosen (Fig. 1g) represents an intergrowth of at least four different euhedral grossular crystals. Under the polarizing microscope it becomes obvious that it displays a concentric oscillatory zoning with isotropic growth domains, and with domains of an anomalous birefringence which is inconsistent with the cubic symmetry. While the oscillatory zoning is essentially present in the mantle and rim domains (Fig. 1g), the core contains oval patches of garnet. The grossular contains inclusions of apatite, zircon and titanite. The sample was provided by the Mineralogical Collection of the Ruhr-University Bochum and was labeled "Lake Jaco, Coahuila, Mexico" (it likely derives from a skarn deposit at Sierra de la Cruz).

The pyrope was collected at Masueria/Val Gilba, Dora-Maira Massif, Italy. The respective thin section for the current study represents a fine-grained pyrope quartzite rock (sample No. 17642; Fig. 1h); major constituents are pyrope, quartz, coesite, kyanite, phengite, and talc, whereas zircon, monazite, rutile are accessory (for further information see Ferrando et al., 2009; Schertl et al., 1991; Chopin, 1984). This type of pyrope forms crystals up to about 2 cm size, is close to its end-member composition, and typically contains inclusions of coesite, quartz, kyanite, and phengite. It has to be distinguished from pyrope megacrysts up to 25 cm in diameter with inclusions of chlorite, talc, kyanite, and ellenbergerite. While the megacrysts are known to have formed due to the so-called first pyrope forming reaction (Tc+ Ky+Chl=Py) on the prograde PT-path, small pyrope crystals are formed due to the second pyrope-forming reaction (Tc+Ky= Py+Coe) at about peak metamorphic conditions (Schertl et al., 1991). Three different pyrope grains of one thin section (17642-1, 17642-2, 17642-3) were studied in detail.

3 CATHODOLUMINESCENCE MICROSCOPY

The oval growth domains in the core of grossular show intense orange, occasionally also bluish luminescence colors with an oscillatory zonation (Fig. 2c); the interstitial portion of garnet in the core is lacking luminescence. Zircon crystals, displaying intense whitish-blue luminescence colors, typically are forming inclusions in these interstitial portions. Inclusions of apatite are bright orange in luminescence color. Close to the rim, garnet is characterized by parallel growth zones different in size and luminescence color (bright yellow, orange, reddish orange, dark bluish, or portions which are lacking luminescence; Fig. 2b). Note that grossular contains domains of an anomalous birefringence (either as part of the oval structures in the garnet core, or the oscillatory zoning in mantle and rim portions; Fig. 2c); some of these domains are characterized by an intense orange luminescence whereas others are lacking luminescence. Whereas the orange luminescent domains in the oscillatory zoning portions of grossular can be followed continuously throughout the entire crystal, the birefringent domains form disconnected zones; their patterns may change at growth sector boundaries. Numbers 1-4 (yellow dots) in Figs. 2a-2c refer to CL-spectra (Fig. 4) whereas white crosses refer to EMP-analyses (respective numbers see Table 1).

Table 1 Representative grossular analyses (wt.%; calculated on the basis of 12 oxygens following Locock, 2008). Respective No.'s see Figs. 2a-2c.

Pyrope typically is characterized by non- to low-luminescent core regions; the mantle region shows intense to less intense bluish-white to brownish-violet luminescent, oscillatory growth zones and the rim zones are lacking luminescence (Figs. 2d-2f; see also Schertl et al., 2004). In general, the intense luminescent oscillatory mantle zone yields a concentrical zoning. Occasionally, however, this mantle zone is not present throughout the whole grain and the non-luminescent rim does not always envelope the entire garnet grain. Schertl et al. (2004; Fig. 1a therein) described a change of crystal morphology during growth, observed in the core of pyrope. Numbers 1-3 in Figs. 2d-2f (yellow dots) refer to CL-spectra (Figs. 5-7). White crosses refer to EMP-analyses; the respective numbers are given in Table 2. Kyanite inclusions in pyrope demonstrate bright whitish-blue luminescence.

Table 2 Representative pyrope analyses (wt.%; calculated on the basis of 12 oxygens following Locock, 2008, No.'s see Figs. 2d-2f)
Download:
Figure 6. Details of Figs. 5a, 5c, using a higher resolution (wavelength range 460-620 nm).
Download:
Figure 7. Details of Figs. 5a, 5c, focusing on the Fe3+ broad emission band, using a higher resolution (wavelength range 620-800 nm).
4 ELECTRON MICROPROBE (EMP) ANALYSIS 4.1 Grossular

Although in general the amounts of MnO and MgO are very low (< 1 wt.%), the core of grossular is slightly higher in MnO and lower in MgO compared to the oscillatory rim (EMP analyses of grossular see Table 1). As a general rule, all luminescent areas were demonstrated to contain lower amounts of iron and titanium than the non-luminescent areas; in contrast, the amount of aluminum is higher which likely reflects the presence of substitutions (ⅰ) Al3+=Fe3+, and (ⅱ) 2Al3+=Ti4++Mg2+. The oval patches in the garnet core domain come close to the grossular end member composition (Grs93.4Adr3.3) whereas the interstitial regions lacking luminescence are generally much lower in grossular component and contain higher amounts of andradite, schorlomite-Al (Ca3Ti2[Si1Al2O4]3) and morimotoite-Mg (Ca3TiMg[SiO4]3) components (Grs65.29Adr21.8Schor-Al5.8Mor-Mg4.5). The luminescent oscillatory rim portions are also high in grossular-component (yellow-orange luminescence: Grs88.8Adr7.5; bluish luminescence: Grs86.8Adr8.1Schor-Al1.4), however not as high as in the luminescent patches observed in the core. The oscillatory rim portions lacking luminescence are low in grossular-component (Grs75.3Adr12.8Schor-Al4.4Mori-Mg3.8).

In addition, domains of grossular which are characterized by an anomalous birefringence (Fig. 2b) were studied related to their luminescence characteristics. An infrared (IR) investigation of grossular documents the presence of OH-groups (most likely as hydrogrossular-component). The amount of OH-groups, however, cannot be very high since the totals of oxides of the anomalous domains, independently whether they are luminescent or not, are around 100 wt.% (Grs78.9Adr9.8Schor-Al3.0Mori-Mg4.3). According to the CL-spectra (see respective chapter below) grossular contains small amounts of europium and the orange luminescent grossular domains are expected to contain slightly higher amounts compared to the blue, red or not very distinct grossular domains or those who are even lacking luminescence. The EMP measurements document the luminescent domains to contain 85 ppm to 105 ppm europium whereas the amount of europium in the non-luminescent domains is between 98 ppm and 102 ppm; thus, applying EMP analysis, it is not possible to quantitatively distinguish between these two domains. At average, in the core of grossular the amount of manganese in the luminescent domains is slightly higher than in the non-luminescent ones (0.2 wt.% compared to 0.15 wt.% MnO); in the oscillatory rim portions these differences are less pronounced.

4.2 Pyrope

The different luminescent and non-luminescent zones of pyrope could be chemically distinguished as documented by EMP measurements (Table 2). Compositional profiles through three pyrope grains of the thin section were performed; one pyrope grain representatively was chosen for a 2-dimensional map. In Table 2 the results of the profiles were compiled, related to the non- to low-luminescent core, the luminescent mantle and the non-luminescent rim. In general, all analyses demonstrate compositions close to pyrope endmember (Prp > 94). From the EMP measurements it becomes obvious, that the more intense the luminescence, the lower the amount of iron. Thus, the amount of pyrope end member composition is highest in the luminescent regions, reaching about 98 mol%. A negative correlation between Mg-contents and Fe- as well as Ca-contents are shown, documenting Mg=(Fe2++Ca) as the main substitution mechanism. Such a behavior becomes obvious also via the 2-dimensional element maps; Figs. 3a-3c shows the element distribution in pyrope related to Mg, Fe, and Ca, respectively. Note-with respect to the Ca-content-the sharp contact between the garnet mantle and the outermost rim zone. The Ca-rich zone does not concentrically envelope the entire garnet grain, probably due to later resorption processes. It has to be considered, that even very small changes in the concentration of Fe2+ as quencher element, have a significant influence in the intensity of luminescence. Studying pyrope from a different locality of the Dora-Maira massif (Parigi [or Case Ramello; Compagnoni et al., 1994]), Schertl et al. (2004) documented FeO-contents of 1.32 wt.% for the dull luminescent core, 1.17 wt.% for the intense luminescent mantle and 2.0 wt.% for the outermost rim lacking luminescence (CL-spectra were not performed by the authors). Figure 3d shows the element distribution map of Dy. This map was chosen since, considering the CL-spectra performed, Dy represents the most abundant lanthanide besides Sm and Tb. Only a very limited enrichment of Dy becomes visible in the lower right corner of the distribution map, where the border between garnet rim and the phengite+quartz-rich matrix is exposed. The red colored portions in pyrope, demonstrating high amounts of Dy (arrows), may trace back to inclusions of florencite or monazite.

5 CATHODOLUMINESCENCE-SPECTRA

CL-spectra on various growth domains were performed to find the source of the luminescent colors observed in grossular and pyrope, and to find possible indicators of a contribution of elements not detected by EMP (since especially the amount of REE can be below the detection limit). CL-colours can be measured by aquiring the spectral composition with a spectrograph. The advantage of CL-spectra is the objective description of CL-colours. The spectra are composed of a combination of broad band and narrow band peaks that are indicative of the luminescing material. Here, e.g., the broad band emission is attributed to lattice spacings in crystals and can be influenced by the incorporation of trace elements that do not exactly fit into the crystal lattice. Narrow bands are due to specific emissions of distinct elements and thus are a method to detect the presence of such elements. If calibrated, CL-spectroscopy is a very sensitive tool for elemental analysis at low concentrations (Habermann et al., 2000, 1996). Often a combination of broad and narrow bands is observed in CL spectra.

Different luminescence colors of grossular, either from the oval structures in the core or from the oscillatory growth zones close to the rim, essentially derive from small amounts of manganese and-to a lesser extent-also europium. Figure 4a shows the CL-spectra derived from the garnet core. The broad emission band at 589 nm is related to the orange luminescent patches and to areas in garnet where the amount of Mn2+ is high; as a result the Mn2+-related emission band dominates and the orange luminescence is very intense. On the other hand, if Mn2+ is less pronounced, and the Eu2+ band at 430 nm becomes more dominant, the resulting luminescence color is bluish. Domains with very little luminescence show lower intensities in both, Mn2+ and Eu2+. Broad emission bands are generally attributed to variations in lattice spacings due to the incorporation of activator or quencher elements. Figure 4b represents an enlarged section of Fig. 4a, documenting in detail the broad band emission of europium.

A similar picture emerges from the CL-spectra of the oscillatory rim zone of grossular; the entire wavelength range is shown in Fig. 4d. Again, the orange luminescent zone is characterized by an intense broad band emission which is related to small amounts of Mn2+. Likewise, essentially Eu2+ is in charge for the domains which display a blue or a weak red luminescence; if the amounts of Mn2+ and Eu2+ are low, the respective luminescence is not very distinct. With a decreasing intensity of the Mn2+-related emission band, the influence of europium becomes more dominant. The enlarged wavelength range (Fig. 4b), however, proves that with a higher emission band of europium, the emission band of Mn2+ is also higher (compare Fig. 4a). One exception is related to the bluish luminescent domain in the oscillatory rim that has a significantly lower Mn2+ emission band but higher Eu2+ band, compared to the domain with orange luminescence (Fig. 4d). The attempt to quantitatively distinguish between the europium content of the luminescent, compared to the non-luminescent domains by EMP failed (see chapter "electron microprobe analyses"). Figure 4c shows the positions of the Mn2+ and Eu2+ broad emission bands of bluish and orange luminescent grossular zones and their shift compared to apatite, forming an inclusion in grossular. As already mentioned in the introduction, the same activator elements that produce broad emission bands can occur at different wavelength positions in different minerals and thus also produce different luminescence colors.

Three pyrope crystals of a fine-grained pyrope quartzite rock were chosen to perform different CL-spectra (Figs. 5a-5c). Contrasting the grossular spectra studied, which are characterized by broad emission bands of Mn2+ and Eu2+, pyrope spectra in general exhibit one minor broad emission band that refers to Fe3+ and several narrow emission band which refer to different amounts of Dy3+, Sm3+, Sm2+ and Tb3+. Since narrow emission bands-independent of the mineral species studied-occur at closely the same wavelength position, comparative studies on different minerals that contain REE, allow to identify the respective activator elements incorporated in pyrope (e.g., Götze, 2002).

Major Dy3+ peaks were described to occur between 457 and 488 nm, 543 and 595 nm, at wavelength positions of 660 and 667 nm, as well as in the ranges of 750-774 and 830-965 nm (Gorobets and Rogojine, 2002; Blanc et al., 2000; Mitchell et al., 1997). Further peaks were observed to refer to Sm3+, possibly also Sm2+ (Table 2). Recent studies on Dy3+- and Dy3++Ce3+-doped garnet (YAG) by Zorenko et al. (2016; Figs. 3 and 4 therein) confirm that the narrow emission band in pyrope at about 501 nm is also related to the incorporation of Dy3+. Narrow emission bands that occur at wavelength values of 466, 475, 490, 544 and 766 nm (Gorobets and Rogojine, 2002; Blanc et al., 2000; Mitchell et al., 1997) source from Tb3+. However, not the entire range of peaks is developed; in general, the Tb3+ peaks are less intense. The broad emission band of Fe3+ is reported to be located at around 678 nm (Kempe et al., 2002) which is in agreement with the location of the respective peak in pyrope observed at 650-750 nm. It has to be considered that the lack of some emission bands may be due to the fact that garnet spectra were not always available for comparison; for that reason also small shifts of a few nm are possible (Gaft et al., 2005; Marfunin, 1979).

Three CL-spectra of each of the three pyrope crystals studied were recorded, one from the dull luminescing core, one from mantle characterized by an intense luminescence, and one from the rim nearly lacking luminescence. As a general rule, all the emission bands related to the rim domain demonstrate the lowest intensities (Fig. 5). The spectrum of pyrope 17642-1 proofs the highest intensities (and thus also the highest amounts) of Dy3+, Sm3+, Sm2+ and Tb3+ to occur in the mantle, the second highest in the core and the lowest in the rim domains. The broad emission band of Fe3+ plays only a subordinate role, above all in the rim domain. The spectra of pyropes 17642-2 and 17642-3 document high intensities for REE activators in the dull luminescent core and in the luminescent mantle. Note, however, that particularly for the wavelength range between 650 and 790 nm, the measured intensities of these two pyropes (Figs. 5b, 5c) are highest in the core and not in the luminescent mantle. It has to be considered, however, that besides the narrow emission bands related to REE, in this wavelength range also the broad emission band related to Fe3+ is located, and an interaction of both types of bands is plausible. A list of small band emission peaks of pyrope is given in Table 3. In Figs. 7a, 7b, the respective wavelength range is displayed using a higher resolution. It documents that for sample 17642-3 the intensity of the Fe3+ broad emission band in the core is higher than in the luminescent mantle, whereas contrary in sample 17642-1, the luminescent mantle exhibits the highest Fe3+ broad emission band.

Table 3 Cathodoluminescence spectra peak-positions of REE distinguished

Using a higher resolution, in Figs. 6a, 6b, the CL-spectra of samples 17642-1 and 17642-3 are compared, selecting a wavelength range between 460 and 620 nm. It is clearly shown, that, related to the spectra of sample 17642-1, the REE-peaks in the mantle are about 4-5 times higher than in the respective core. On the other hand, the spectra of sample 17642-3 document similar intensities of REE-peaks in core and mantle.

6 DISCUSSION

It could be demonstrated that grossular contains oscillatory growth zones, some of which show an anomalous birefringence, typical for a certain amount of hydrogrossular- component. However, due to electron microprobe and IR studies there is evidence only for very low amounts of hydrogrossular. Note that not the hydrogrossular component itself causes the birefringence but the orientation ordering of the (OH)--groups in the garnet structure. This dependence of orientation explains why the birefringent zones can be disconnected at growth sector boundaries (Shtukenberg et al., 2001a, b). Typically hydrogrossular-rich garnets are observed in skarn deposits or in rodingite (e.g., Li et al., 2017; Schandl and Mittweide, 2001; Leach and Rogers, 1978; Henmi et al., 1971). It should, however, be considered that other explanations as strain (McAloon and Hoffmeister, 1993) or growth dissymetrization (due to order-disorder processes of elements; see Shtukenberg et al., 2001a) can also be in charge for an anomalous birefringence. A chemical correlation between the birefringent zones and those showing luminescence was not observed. Using CL-spectrum measurements, it became obvious that the REE Dy3+, Sm3+, Sm2+ and Tb3+, which produce narrow emission bands, play a significant role as to their incorporation into pyrope of UHP-metamorphic rocks from the Dora-Maira massif. These elements, however, could not be quantitatively determined using the EMP, since their contents were below their detection limits. On the other hand, using EMP- measurements, it was possible to demonstrate that with a decreasing intensity of luminescence an increasing amount of Fe2+ was observed, because, contrasting the activator elements as REE or Mn2+, iron is known to behave as a quencher element (e.g., Götze et al., 2013; Pagel et al., 2000). The role of the activator element Mn2+, producing broad emission bands, is supported by our studies of the luminescent domains of grossular.

But what is the petrological and geological significance of such studies? First of all, distinguishing different growth stages of any metamorphic mineral is a very important issue with respect to the possible derivation of a PT-path. Especially prograde stages are very important since they document the early history of any metamorphic rock. Examples pointing out different generations of luminescent metamorphic growth domains of kyanite, andalusite, garnet, jadeite, diopsidic clinopyroxene, wollastonite, and diamond were reported by Schertl et al. (2004), Harlow et al. (2005), Schertl et al. (2012), Götze et al. (2013), and Liu et al. (2017). Garnet, however, plays a very decisive role, because it is widespread in metamorphic rocks and often used for geothermobarometrical calculations. However in granulite facies and particularly in UHT metamorphic rocks it is often a challenge to define different growth stages of garnet, since at high temperatures major elements may have changed their structural positions, producing homogeneous profiles. Whereas in general exchange of Fe2+ and Mg is rapid, Ca diffusion and transport by fluids is less dominant (e.g., Carlson, 2006). Rare earth elements such as Sm and Dy and also further minor elements which can be incorporated in the garnet structure as P, Ti, Na, etc. (e.g., Moore and Gurney, 1985; Bishop et al., 1976; Sobolev et al., 1971), however, in parts are known to reveal very slow diffusion rates (Van Orman et al., 2002; Cherniak, 1998). The aim of the current study is not to focus on details related to diffusion rates, which depends on the composition of garnet (and thus the concomitant expansion or reduction of the lattice), the effect of fluids (for instance the role of hydrogrossular component), deformation, temperature, pressure, valency of the respective element(s) and possible coupled substitutions involved, the adjacent mineral in contact to garnet, etc.-such discussions are considered elsewhere (e.g., Carlson, 2017; Ganguly, 2010; Perchuk et al., 2009; Tirone et al., 2005; Burton et al., 1995; Chakraborty and Ganguly, 1991; Lasaga et al., 1977). The main target of this case study is related to the distribution of those elements within garnet, which can be used to visualize any CL-information and thus to help to distinguish possible prograde and/or peak pressure stage growth domains. Thus, although the memory on any possible early zonation of garnet may have been erased at high temperatures, it can be still possible to define different garnet domains via CL-studies, to identify, for instance, related assemblages of mineral inclusions and to incorporate these results in a unified evolution of a detailed PT path.

Using such minor elements, Ague and Axler (2016) described clearly defined P-rich domains in garnet (in parts showing oscillatory zoning) of HP granulites from the Saxonian Erzgebirge, which experienced 1 000-1 050 ℃. The author demonstrated that Na and Ti-in a more diffuse way- mimic the zoning pattern of P. The Mg-distribution, however, was shown to be very homogeneous and even Ca, which diffuses more slowly than Mg and Fe2+, did not demonstrate a sharp zonation (e.g., Chernoff and Carlson, 1999). Thus conventional geothermobarometry based on Mg-Fe2+ exchange reactions involving garnet may lead to erroneous results, if applied to HT and particularly UHT metamorphic rocks. The study of such trace element and REE distribution maps, and their interpretation in combination with mineral inclusion studies, can help to define distinct PT stages, even if they occur during an early subduction process or if they, for instance, represent an older metamorphic event of a polymetamorphic rock. Such a combined study however has to be undertaken in a very careful way, since the former major element zonation does not necessarily need to coincide with the former zonation derived from minor or trace/REE elements.

The internal growth structures of garnet visualized by CL-studies have also significant thermodynamic/kinetic implications. The large amount of roundish orange luminescent patches in the core domain of grossular may refer to very early nucleation stages of garnet growth. It is known from the classical nucleation theory (CNT) that the scenario of crystal growth starts with a stage of a multitude of nucleation seeds, which can only begin to crystallize when a critical nucleus size is reached at an oversaturation stage (e.g., Teng, 2013). This of course takes place at an atomistic level. However, the cores of the roundish and early grossular patches may represent such seed areas and may have started to grow/crystallize when their growth rate exceeded their nucleation rate. The electron microprobe data would support such a scenario, since in general, when garnet starts to grow in a metamorphic rock, these zones are known to remove manganese from the local surrounding in order to incorporate it into the garnet structure and the average amount of MnO in the core patches (0.22 wt.%, Ø 11 analyses) is higher (not much but distinctively) than in the rim portions (0.14 wt.%, Ø 29 analyses) or even in the luminescent rim portions (0.17 wt.%, Ø 7 analyses).

As a further advantage of the CL-microscopy, an entire thin section can be studied within minutes and just those garnets (or other luminescent minerals) can be chosen for further studies which are promising or interesting with respect to their zonation or internal structures. One problem that often occurs can be easily prevented by using the CL-technique: if for instance further EMP- or ion probe studies are required, it is often not easy to locate a garnet crystal in thin section, which is cut through the middle of the grain and thus provides core, mantle as well as rim zones for succeeding measurements. If a luminescent pyrope, as those studied, is cut only through the mantle region, it then would only show an oscillatory core and a non- to dull luminescent rim and the interpretation of such sections would lead to erroneous results.

The high resolution of the CL images represents an advantage not to be underestimated. The oscillatory zoning of the pyrope mantle zone, for instance, becomes easily visible using CL-microscope. In Fig. 3 which shows microprobe X-ray element distribution maps performed with a step width of 4 μm, an oscillatory zoning does not become visible. At most the Ca-distribution map (Fig. 3c) shows a very small indication of this zonation. The measuring time of such an image, however, depending on the resolution chosen, takes 3-10 hours. The clear indication of the oscillatory zoning of pyrope displayed using the CL-technique, can be demonstrated within seconds. The knowledge of such kind of zonations, especially if the resolution is very high as shown by the studied samples and if difficult to visualize using X-ray maps, is of invaluable significance for kinetical as well as for geochronological questions. The sharp oscillatory zones very likely represent growth features which result from distinctively different, although small, amounts of REEs, visualized by CL. Further studies would need to be performed in order to figure out if the oscillatory zonation is, for instance, caused by external processes as cycling changes of the fluid composition (or even of pressure/ temperature conditions) or by kinetic reasons which supported the growth of pyrope (see e.g., discussion by Kohn, 2004). In any event, the pyrope is interpreted to result from metasomatic formation (e.g., Ferrando et al., 2009; Schertl et al., 2008; Compagnoni and Hirajima, 2001) and fluid played a dominant role in their formation. Thus it is possible that within a fluid saturated system, different elements which exhibit different rates of mass transport towards the crystal surface, caused the zonation observed. On the other hand the oscillatory zones indicate slow rates of intracrystalline diffusion of the respective REEs involved, preventing the zoning patterns from a later modification. Possible concomitant options to gain from such data derived from garnet, refer to the usage of REEs for geothermobarometrical analysis (e.g., Schmidt et al., 2011; Scherer et al., 2000; Ganguly et al., 1998) or even as pathfinder of ore deposits (e.g., Park et al., 2017; Zhai et al., 2014).

7 CONCLUSIONS

Our study documents a couple of advantages for petrological and crystallographic applications using CL-studies. Using polished thin sections, internal structures of luminescent minerals (and even of synthetic materials) can be identified within seconds. Customarily, prior to zircon age dating by SHRIMP or laser ablation techniques, different growth zones are distinguished using the CL-detector of an EMP that delivers black and white images with high magnification. By contrast, the advantage of the CL-microscope is its low magnification and the fact that the images recorded, are in true color. This technique allows a number of typical rock forming silicates but also, for instance, carbonate and sulfate minerals to study with respect to their structural characteristics. Such a knowledge has a substantial impact not only on metamorphic rocks which allows a more precise derivation of PT-paths. This method is also advantageous regarding sedimentary processes, distinguishing, for instance, growth stages of different carbonate cements, and it can also help to contribute to a better understanding of crystallization stages in any process igneous rocks are involved. In the current study, using CL-microscope and -spectra, different growth domains of grossular could be distinguished, such as orange luminescent patchy cores with interstitial portions lacking luminescence, as well as oscillatory rim portions (also containing orange luminescent portions and portions lacking luminescence), which in parts show an anomalous birefringence. Pyrope displays non- to low-luminescent cores, luminescent oscillatory zoned mantle, and non-luminescent rim domains. While the CL-spectra of grossular document Mn2+ and Eu2+ as activator elements, pyrope was shown to contain small amounts of Dy3+, Sm3+, Sm2+ and Tb3+. Of special value are combined studies, using the CL- information as a pathfinder prior to supplementary investigations, as for instance EMP, ion probe, or mineral inclusion studies. Regarding a possible routine application, it should however be emphasized, that garnets with high or average Fe-contents may not be luminescent. On the other hand, CL can make very detailed structures visible in any luminescent mineral which may provide further solution to kinetic issues.

In essence, the CL-method is mainly descriptive, however, using a spectrometer unit, a chemical characterization of minerals is possible and even low amounts of REE, which can be below the detection limit of EMP or PIXE, can be identified. It is an easily applied and inexpensive "small scale method", however with the strength to help to reconstruct large scale geological processes.

ACKNOWLEDGMENTS

This paper is dedicated to the celebration of Prof. Zhendong Youʼs 90th birthday. We are grateful to two anonymous reviewers who helped to improve the paper, and editor Dr. Yanru Song for handling and editing. The final publication is available at Springer via https://doi.org/10.1007/s12583-018-0842-0.


REFERENCES CITED
Ague, J. J., Axler, J. A., 2016. Interface Coupled Dissolution-Reprecipitation in Garnet from Subducted Granulites and Ultrahigh-Pressure Rocks Revealed by Phosphorous, Sodium, and Titanium Zonation. American Mineralogist, 101(7): 1696-1699. DOI:10.2138/am-2016-5707
Allan, M. M., Yardley, B. W. D., 2007. Tracking Meteoric Infiltration into a Magmatic-Hydrothermal System:A Cathodoluminescence, Oxygen Isotope and Trace Element Study of Quartz from Mt. Leyshon, Australia. Chemical Geology, 240(3/4): 343-360. DOI:10.1016/j.chemgeo.2007.03.004
Barwood, H., 2007. Digital Near-Infrared (NIR) Cathodoluminescence (CL) Imaging and Image Processing. American Mineralogist, 92(2/3): 261-266. DOI:10.2138/am.2007.2186
Baxter, E. F., Scherer, E. E., 2013. Garnet Geochronology:Timekeeper of Tectonometamorphic Processes. Elements, 9(6): 433-438. DOI:10.2113/gselements.9.6.433
Berman, R. G., 1988. Internally-Consistent Thermodynamic Data for Minerals in the System Na2O-K2O-CaO-MgO-FeO-Fe2O3-Al2O3-SiO2-TiO2-H2O-CO2. Journal of Petrology, 29(2): 445-522. DOI:10.1093/petrology/29.2.445
Bishop, F. C., Smith, J. V., Dawson, J. B., 1976. Na, P, Ti and Coordination of Si in Garnet from Peridotite and Eclogite Xenoliths. Nature, 260(5553): 696-697. DOI:10.1038/260696a0
Blanc, P., Baumer, A., Cesbron, F., et al.., 2000. Systematic Cathodoluminescence Spectral Analysis of Synthetic Doped Minerals: Anhydrite, Apatite, Calcite, Fluorite, Scheelite and Zircon. In: Pagel, M., Barbin, V., Blanc, P., et al., eds., Cathodoluminescence in Geosciences. Springer, Berlin, Heidelberg, New York. 127-160
Burton, K. W., Kohn, M. J., Cohen, A. S., et al., 1995. The Relative Diffusion of Pb, Nd, Sr and O in Garnet. Earth and Planetary Science Letters, 133(1/2): 199-211. DOI:10.1016/0012-821x(95)00067-m
Caddick, M. J., Kohn, M. J., 2013. Garnet:Witness to the Evolution of Destructive Plate Boundaries. Elements, 9(6): 427-432. DOI:10.2113/gselements.9.6.427
Carlson, W. D., 2006. Rates of Fe, Mg, Mn, and Ca Diffusion in Garnet. American Mineralogist, 91(1): 1-11. DOI:10.2138/am.2006.2043
Carlson, W. D., 2017. Multicomponent Diffusion in Aluminosilicate Garnet:Coupling Effects due to Charge Compensation. International Geology Review, 28: 1-15. DOI:10.1080/00206814.2016.1189855
Catlos, E. J., Baker, C. B., Sorensen, S. S., et al., 2011. Linking Microcracks and Mineral Zoning of Detachment-Exhumed Granites to Their Tectonomagmatic History:Evidence from the Salihli and Turgutlu Plutons in Western Turkey (Menderes Massif). Journal of Structural Geology, 33(5): 951-969. DOI:10.1016/j.jsg.2011.02.005
Chakraborty, S., Ganguly, J., 1991. Compositional Zoning and Cation Diffusion in Garnets. In: Ganguly, J., ed., Diffusion, Atomic Ordering, and Mass Transport. Selected Topics in Geochemistry. Advances in Physical Chemistry Volume 8. Springer, New York. 566
Cherniak, D. J., 1998. Rare Earth Element and Gallium Diffusion in Yttrium Aluminum Garnet. Physics and Chemistry of Minerals, 26(2): 156-163. DOI:10.1007/s002690050172
Chernoff, C. B., Carlson, W. D., 1999. Trace Element Zoning as a Record of Chemical Disequilibrium during Garnet Growth. Geology, 27(6): 555-558. DOI:10.1130/0091-7613(1999)027<0555:tezaar>2.3.co;2
Chopin, C., 1984. Coesite and Pure Pyrope in High-Grade Blueschists of the Western Alps:A First Record and some Consequences. Contributions to Mineralogy and Petrology, 86(2): 107-118. DOI:10.1007/BF00381838
Chopin, C., Beyssac, O., Bernard, S., et al., 2008. Aragonite-Grossular Intergrowths in Eclogite-Facies Marble, Alpine Corsica. European Journal of Mineralogy, 20(5): 857-865. DOI:10.1127/0935-1221/2008/0020-1892
Compagnoni, R., Hirajima, T., 2001. Superzoned Garnets in the Coesite-Bearing Brossasco-Isasca Unit, Dora-Maira Massif, Western Alps, and the Origin of the Whiteschists. Lithos, 57(4): 219-236. DOI:10.1016/s0024-4937(01)00041-x
Compagnoni, R., Messiga, B., Castelli, D. 1994. High Pressure Metamorphism in the Western Alps. Guide-Book to the Field Excursion of the 16th Meeting of the IMA, September 10-15, 2004, Pisa. 148
Connolly, J. A. D., 1990. Multivariable Phase Diagrams:An Algorithm Based on Generalized Thermodynamics. American Journal of Science, 290(6): 666-718. DOI:10.2475/ajs.290.6.666
de Capitani, C., Brown, T. H., 1987. The Computation of Chemical Equilibrium in Complex Systems Containing Non-Ideal Solutions. Geochimica et Cosmochimica Acta, 51(10): 2639-2652. DOI:10.1016/0016-7037(87)90145-1
Dudley, R. J., 1976. The Use of Cathodoluminescence in the Identification of Soil Minerals. Journal of Soil Science, 27(4): 487-494. DOI:10.1111/j.1365-2389.1976.tb02018.x
Ferrando, S., Frezzotti, M. L., Petrelli, M., et al., 2009. Metasomatism of Continental Crust during Subduction:The UHP Whiteschists from the Southern Dora-Maira Massif (Italian Western Alps). Journal of Metamorphic Geology, 27(9): 739-756. DOI:10.1111/j.1525-1314.2009.00837.x
Ferry, J. M., Spear, F. S., 1978. Experimental Calibration of the Partitioning of Fe and Mg between Biotite and Garnet. Contributions to Mineralogy and Petrology, 66(2): 113-117. DOI:10.1007/bf00372150
Gaft, M., Reisfeld, R., Panczer, G., 2005. Modern Luminescence Spectroscopy of Minerals and Materials. Springer, Berlin, New York. 356
Ganguly, J., 2010. Cation Diffusion Kinetics in Aluminosilicate Garnets and Geological Applications. Reviews in Mineralogy and Geochemistry, 72(1): 559-601. DOI:10.2138/rmg.2010.72.12
Ganguly, J., Tirone, M., Hervig, R. L., 1998. Diffusion Kinetics of Samarium and Neodymium in Garnet, and a Method for Determining Cooling Rates of Rocks. Science, 281(5378): 805-807. DOI:10.1126/science.281.5378.805
Gorobets, B. S., Rogojine, A. A., 2002. Luminescent Spectra of Minerals: Reference-Book. RPC VIMS. 302
Götze, J., 2002. Potential of Cathodoluminescence (CL) Microscopy and Spectroscopy for the Analysis of Minerals and Materials. Analytical and Bioanalytical Chemistry, 374(4): 703-708. DOI:10.1007/s00216-002-1461-1
Götze, J., Schertl, H. P., Neuser, R. D., et al., 2013. Optical Microscope-Cathodoluminescence (OM-CL) Imaging as a Powerful Tool to Reveal Internal Textures of Minerals. Mineralogy and Petrology, 107(3): 373-392. DOI:10.1007/s00710-012-0256-0
Gross, J., Burchard, M., Schertl, H. P., et al., 2008. Common High-Pressure Metamorphic History of Eclogite Lenses and Surrounding Metasediments:A Case Study of Calc-Silicate Reaction Zones (Erzgebirge, Germany). European Journal of Mineralogy, 20(5): 757-775. DOI:10.1127/0935-1221/2008/0020-1861
Habermann, D., Götze, J., Neuser, R. D., et al., 2000. Quantitative high Resolution Spectral Analysis of Mn2+ in Sedimentary Calcite. In: Pagel, M., Barbin, V., Blanc, P., et al., eds., Cathodoluminescence in Geosciences, Springer, Berlin Heidelberg, New York. 331-358
Habermann, D., Meijer, J., Neuser, R. D., et al., 1999. Micro-PIXE and Quantitative Cathodoluminescence Spectroscopy:Combined High Resolution Trace Element Analyses in Minerals. Nuclear Instruments and Methods in Physics Research Section B:Beam Interactions with Materials and Atoms, 150(1/2/3/4): 470-477. DOI:10.1016/s0168-583x(98)00926-4
Habermann, D., Neuser, R. D., Richter, D. K., 1996. REE-Activated Cathodoluminescence of Calcite and Dolomite:High-Resolution Spectrometric Analysis of CL Emission (HRS-CL). Sedimentary Geology, 101(1/2): 1-7. DOI:10.1016/0037-0738(95)00086-0
Harlow, G. E., 1994. Jadeitites, Albitites and Related Rocks from the Motagua Fault Zone, Guatemala. Journal of Metamorphic Geology, 12(1): 49-68. DOI:10.1111/j.1525-1314.1994.tb00003.x
Harlow, G. E., Sorensen, S. S., 2005. Jade (Nephrite and Jadeitite) and Serpentinite:Metasomatic Connections. International Geology Review, 47(2): 113-146. DOI:10.2747/0020-6814.47.2.113
Henmi, K., Kusachi, I., Numano, T., 1971. Contact Minerals from Kushiro, Hiroshima Prefecture. Journal of the Mineralogical Society of Japan, 10(3): 160-169. DOI:10.2465/gkk1952.10.160
Holland, T. J. B., Powell, R., 1985. An Internally Consistent Thermodynamic Dataset with Uncertainties and Correlations:2. Data and Results. Journal of Metamorphic Geology, 3(4): 343-370. DOI:10.1111/j.1525-1314.1985.tb00325.x
Holland, T. J. B., Powell, R., 1990. An Enlarged and Updated Internally Consistent Thermodynamic Dataset with Uncertainties and Correlations:The System K2O-Na2O-CaO-MgO-MnO-FeO-Fe2O3-Al2O3-TiO2-SiO2-C-H2-O2. Journal of Metamorphic Geology, 8(1): 89-124. DOI:10.1111/j.1525-1314.1990.tb00458.x
Houzar, S., Leichmann, J., 2003. Application of Cathodoluminescence to the Study of Metamorphic Textures in Marbles from the Eastern Part of the Bohemian Massif. Bulletin of Geosciences, 78: 241-250.
Kempe, U., Plötze, M., Brachmann, A., et al., 2002. Stabilisation of Divalent Rare Earth Elements in Natural Fluorite. Mineralogy and Petrology, 76: 213-234. DOI:10.1007/s007100200042
Kohn, M. J., 2004. Oscillatory-and Sector-Zoned Garnets Record Cyclic (?) Rapid Thrusting in Central Nepal. Geochemistry, Geophysics, Geosystems, 5(12): Q12014. DOI:10.1029/2004gc000737
Lasaga, A. C., Richardson, S. M., Holland, H. D., 1977. The Mathematics of Cation Diffusion and Exchange between Silicate Minerals during Retrograde Metamorphism. In: Saxena, S. K., Bhattacharji, S. D., eds., Energetics of Geodynamic Process. Springer, New York. 353-388
Leach, T. M., Rodgers, K. A., 1978. Metasomatism in the Wairere Serpentinite, King Country, New Zealand. Mineralogical Magazine, 42(321): 45-62. DOI:10.1180/minmag.1978.042.321.06
Li, X. P., Duan, W. Y., Zhao, L. Q., et al., 2017. Rodingites from the Xigaze Ophiolite, Southern Tibet-New Insights into the Processes of Rodingitization. European Journal of Mineralogy, 29(5): 821-837. DOI:10.1127/ejm/2017/0029-2633
Liu, P. L., Massonne, H. J., Jin, Z. M., et al., 2017. Diopside, Apatite, and Rutile in an Ultrahigh Pressure Impure Marble from the Dabie Shan, Eastern China:A Record of Eclogite-Facies Metasomatism during Exhumation. Chemical Geology, 466: 123-139. DOI:10.1016/j.chemgeo.2017.06.001
Locock, A. J., 2008. An Excel Spreadsheet to Recast Analyses of Garnet into End-Member Components, and a Synopsis of the Crystal Chemistry of Natural Silicate Garnets. Computers & Geosciences, 34(12): 1769-1780. DOI:10.1016/j.cageo.2007.12.013
Marfunin, A. S., 1979. Spectroscopy, Luminescence and Radiation Centres in Minerals. Springer, Berlin, Heidelberg, New York. 352
McAloon, B. P., Hofmeister, A. M., 1993. Single-Crystal Absorption and Reflection Infrared Spectroscopy of Birefringent Grossular-Andradite Garnets. American Mineralogist, 78: 957-967.
Mitchell, R. H., Xiong, J., Mariano, A. N., et al., 1997. Rare-Earth Element-Activated Cathodoluminescence in Apatite. Canadian Mineralogist, 35: 979-998.
Moore, R. O., Gurney, J. J., 1985. Pyroxene Solid Solution in Garnets Included in Diamond. Nature, 318(6046): 553-555. DOI:10.1038/318553a0
Neuser, R. D., Reinecke, T., Schertl, H.-P., 1995. Low Temperature Cathodoluminescence of Selected Minerals from High Pressure Metamorphic Rocks. Bochumer Geologische und Geotechnische Arbeiten, 44: 119-123.
Pagel, M., Barbin, V., Blanc, P., et al., 2000. Cathodoluminescence in Geosciences. Springer, Berlin. 514
Park, C., Choi, W., Kim, H., et al., 2017. Oscillatory Zoning in Skarn Garnet:Implications for Tungsten Ore Exploration. Ore Geology Reviews, 89: 1006-1018. DOI:10.1016/j.oregeorev.2017.08.003
Parsons, I., Steele, D. A., Lee, M. R., et al., 2008. Titanium as a Cathodoluminescence Activator in Alkali Feldspars. American Mineralogist, 93(5/6): 875-879. DOI:10.2138/am.2008.2711
Perchuk, A. L., Burchard, M., Schertl, H. P., et al., 2009. Diffusion of Divalent Cations in Garnet:Multi-Couple Experiments. Contributions to Mineralogy and Petrology, 157(5): 573-592. DOI:10.1007/s00410-008-0353-6
Råheim, A., Green, D. H., 1974. Experimental Determination of the Temperature and Pressure Dependence of the Fe-Mg Partition Coefficient for Coexisting Garnet and Clinopyroxene. Contributions to Mineralogy and Petrology, 48(3): 179-203. DOI:10.1007/bf00383355
Reinecke, T., 1998. Prograde High-to Ultrahigh-Pressure Metamorphism and Exhumation of Oceanic Sediments at Lago Di Cignana, Zermatt-Saas Zone, Western Alps. Lithos, 42(3/4): 147-189. DOI:10.1016/s0024-4937(97)00041-8
Richter, D. K., Götte, T., Götze, J., et al., 2003. Progress in Application of Cathodoluminescence (CL) in Sedimentary Petrology. Mineralogy and Petrology, 79(3/4): 127-166. DOI:10.1007/s00710-003-0237-4
Richter, D. K., Heinrich, F., Geske, A., et al., 2014. First Description of Phanerozoic Radiaxial Fibrous Dolomite. Sedimentary Geology, 304: 1-10. DOI:10.1016/j.sedgeo.2014.02.002
Rollinson, H., 2003. Metamorphic History Suggested by Garnet-Growth Chronologies in the Isua Greenstone Belt, West Greenland. Precambrian Research, 126(3/4): 181-196. DOI:10.1016/s0301-9268(03)00094-9
Satish-Kumar, M., Mori, H., Kusachi, I., et al., 2006. Cathodoluminescence Microscopy of High-Temperature Skarn Minerals from Fuka Contact Aureole, Okayama, Japan. Geoscience Reports, Shizuoka University, 33: 21-28.
Schandl, E. S., Mittwede, S. K., 2001. Evolution of the Acipayam (Denizli, Turkey) Rodingites. International Geology Review, 43(7): 611-623. DOI:10.1080/00206810109465036
Scherer, E. E., Cameron, K. L., Blichert-Toft, J., 2000. Lu-Hf Garnet Geochronology:Closure Temperature Relative to the Sm-Nd System and the Effects of Trace Mineral Inclusions. Geochimica et Cosmochimica Acta, 64(19): 3413-3432. DOI:10.1016/s0016-7037(00)00440-3
Schertl, H.-P., Maresch, W. V., Knippenberg, S., et al., 2018. Petrography, Mineralogy and Geochemistry of Jadeite-Rich Artefacts from the Playa Grande Excavation Site, Northern Hispaniola: Evaluation of Local Provenance from the Río San Juan Complex. In: Zhang, L. F., Zhang, Z. M., Schertl, H.-P., et al., eds., HP-UHP Metamorphism and the Tectonic Evolution of Orogenic Belts. Geological Society, London, Special Publications. https: //doi.org/10.1144/sp474.3
Schertl, H.-P., Medenbach, O., Neuser, R. D., 2005. UHP-Metamorphic Rocks from Dora Maira, Western Alps:Cathodoluminescence of Silica and Twinning of Coesite. Russian Geology and Geophysics, 46: 1327-1332.
Schertl, H.-P., Maresch, W. V., Stanek, K. P., et al., 2012. New Occurrences of Jadeitite, Jadeite Quartzite and Jadeite-Lawsonite Quartzite in the Dominican Republic, Hispaniola:Petrological and Geochronological Overview. European Journal of Mineralogy, 24(2): 199-216. DOI:10.1127/0935-1221/2012/0024-2201
Schertl, H.-P., Neuser, R. D., Logvinova, A. M., et al., 2015. Cathodoluminescence Microscopy of the Kokchetav Ultrahigh-Pressure Calcsilicate Rocks:What can We Learn from Silicates, Carbon-Hosting Minerals, and Diamond?. Russian Geology and Geophysics, 56(1/2): 100-112. DOI:10.1016/j.rgg.2015.01.006
Schertl, H.-P., Neuser, R. D., Sobolev, N. V., et al., 2004. UHP-Metamorphic Rocks from Dora Maira/Western Alps and Kokchetav/Kazakhstan:New Insights Using Cathodoluminescence Petrography. European Journal of Mineralogy, 16(1): 49-57. DOI:10.1127/0935-1221/2004/0016-0049
Schertl, H.-P., Schreyer, W., 2008. Geochemistry of Coesite-Bearing "Pyrope Quartzite" and Related Rocks from the Dora-Maira Massif, Western Alps. European Journal of Mineralogy, 20(5): 791-809. DOI:10.1127/0935-1221/2008/0020-1862
Schertl, H.-P., Schreyer, W., Chopin, C., 1991. The Pyrope-Coesite Rocks and Their Country Rocks at Parigi, Dora Maira Massif, Western Alps:Detailed Petrography, Mineral Chemistry and PT-Path. Contributions to Mineralogy and Petrology, 108(1/2): 1-21. DOI:10.1007/bf00307322
Schertl, H.-P., Sobolev, N. V., 2013. The Kokchetav Massif, Kazakhstan:"Type Locality" of Diamond-Bearing UHP Metamorphic Rocks. Journal of Asian Earth Sciences, 63: 5-38. DOI:10.1016/j.jseaes.2012.10.032
Schmidt, A., Mezger, K., O'Brien, P. J., 2011. The Time of Eclogite Formation in the Ultrahigh Pressure Rocks of the Sulu Terrane. Lithos, 125(1/2): 743-756. DOI:10.1016/j.lithos.2011.04.004
Schumacher, R., Rötzler, K., Maresch, W. V., 1998. Subtle Oscillatory Zoning in Garnet from Regional Metamorphic Phyllites and Mica Schists, Western Erzgebirge, Germany. Canadian Mineralogist, 37: 381-402.
Shtukenberg, A. G., Punin, Y. O., Frank-Kamenetskaya, O. V., et al., 2001a. On the Origin of Anomalous Birefringence in Grandite Garnets. Mineralogical Magazine, 65(3): 445-459. DOI:10.1180/002646101300119538
Shtukenberg, A. G., Punin, Y. O., Haegele, E., et al., 2001b. On the Origin of Inhomogeneity of Anomalous Birefringence in Mixed Crystals:An Example of Alums. Physics and Chemistry of Minerals, 28(9): 665-674. DOI:10.1007/s002690100185
Sobolev, N. V. Jr., Lavrent'ev, J. G., 1971. Isomorphic Sodium Admixture in Garnets Formed at High Pressures. Contributions to Mineralogy and Petrology, 31(1): 1-12. DOI:10.1007/bf00373387
Sobolev, N. V., Shatsky, V. S., 1990. Diamond Inclusions in Garnets from Metamorphic Rocks:A New Environment for Diamond Formation. Nature, 343(6260): 742-746. DOI:10.1038/343742a0
Sobolev, N. V., Schertl, H. P., Neuser, R. D., et al., 2007. Relict Unusually Low Iron Pyrope-Grossular Garnets in UHPM Calc-Silicate Rocks of the Kokchetav Massif, Kazakhstan. International Geology Review, 49(8): 717-731. DOI:10.2747/0020-6814.49.8.717
Sobolev, N. V., Schertl, H. P., Valley, J. W., et al., 2011. Oxygen Isotope Variations of Garnets and Clinopyroxenes in a Layered Diamondiferous Calcsilicate Rock from Kokchetav Massif, Kazakhstan:A Window into the Geochemical Nature of Deeply Subducted UHPM Rocks. Contributions to Mineralogy and Petrology, 162(5): 1079-1092. DOI:10.1007/s00410-011-0641-4
Sobolev, N. V., Schertl, H.-P., Neuser, R. D., et al., 2017. Formation and Evolution of Hypabyssal Kimberlites from the Siberian Craton:Part 1-New Insights from Cathodoluminescence of the Carbonates. Journal of Asian Earth Sciences, 145: 670-678. DOI:10.1016/j.jseaes.2017.06.009
Takahashi, N., Tsujimori, T., Kayama, M, et al., 2017. Cathodoluminescence Petrography of P-Type Jadeitites from the New Idria Serpentinite Body, California. Journal of Mineralogical and Petrological Sciences, 112(5): 291-299. DOI:10.2465/jmps.170403
Teng, H. H., 2013. How Ions and Molecules Organize to Form Crystals. Elements, 9(3): 189-194. DOI:10.2113/gselements.9.3.189
Tirone, M., Ganguly, J., Dohmen, R., et al., 2005. Rare Earth Diffusion Kinetics in Garnet:Experimental Studies and Applications. Geochimica et Cosmochimica Acta, 69(9): 2385-2398. DOI:10.1016/j.gca.2004.09.025
Van Orman, J. A., Grove, T. L., Shimizu, N., et al., 2002. Rare Earth Element Diffusion in a Natural Pyrope Single Crystal at 2.8 GPa. Contributions to Mineralogy and Petrology, 142(4): 416-424. DOI:10.1007/s004100100304
Whitney, D. L., Evans, B. W., 2010. Abbreviations for Names of Rock-Forming Minerals. American Mineralogist, 95(1): 185-187. DOI:10.2138/am.2010.3371
Zhai, D.-G., Liu, J.-J., Zhang, H.-Y., et al., 2014. Origin of Oscillatory Zoned Garnets from the Xieertala Fe-Zn Skarn Deposit, Northern China:In situ LA-ICP-MS Evidence. Lithos, 190/191: 279-291. DOI:10.1016/j.lithos.2013.12.017
Zorenko, Y., Gorbenko, V., Zorenko, T., et al., 2016. Luminescent and Scintillation Properties of YAG:Dy and YAG:Dy, Ce Single Crystalline Films. Radiation Measurements, 90: 308-313. DOI:10.1016/j.radmeas.2016.02.026