Fluorite alignments cutting metasomatic textures in trachyte feldspars from Oki-Dogo Island, Sea of Japan

Abstract

Feldspar internal textures in a pyroclastic trachyte from Oki-Dogo, Sea of Japan, were examined to expand the understanding of feldspar reactions during the cooling through magmatic to hydrothermal stages, beyond the previous information of feldspars in Oki-Dogo alkaline lava and sheet rocks, using the methods of an electron microprobe and cathodoluminescence. Two types of micron-size internal microtextures were found to coexist in individual feldspar phenocrysts: clear domain textures, formed during a high-temperature magmatic stage, and turbid microperthitic textures, formed during a low-temperature subsolidus stage. The both microtextures are products of metasomatic replacement reactions. In addition, nano-size fluorite grains are aligned across the microtextures. The fluorite occurrence records the behavior of fluorine related to feldspar reactions. The first account of metasomatic microtextures crosscut by fluorite alignments in volcanic alkali feldspars expands our knowledge of feldspar reactions during the cooling and fluorine behaviors related to them in igneous rocks and shows the significance of the careful analysis of feldspar internal microtextures.

INTRODUCTION

Feldspar microtextures record a variety of information about various geological, petrological, and mineralogical processes (e.g., Deer et al., 2001). The analyses of them in alkaline volcanic rocks from Oki-Dogo Island, Sea of Japan, have added new information on solidification processes of dry (relatively or practically anhydrous) alkaline magmas: the examined rocks are hypersthene-augite trachyte (lava) (Nakano, 1992), alkali rhyolite (lava) (Nakano and Suwa, 1995), ferro-augite trachyte (FAT; lava) (Nakano, 2021), olivine-hedenbergite trachyte from Utagi (UTT; sheet) (Nakano and Makino, 2022), and sanidine trachyte from Hei (HET; sheet) (Nakano and Makino, 2024). Complex internal microtextures comprising lamellar, wavy, patchy, and domain textures in feldspars have been observed with or without anti-rapakivi core-mantle zoning in the individual rocks. The textural and compositional variation of feldspars through the rocks is a record of the processes of magmatic replacement reactions that progressively advanced during fractional crystallization. Among them, the alignments of nano-size fluorite grains in feldspars that are valuable records of the behavior of fluorine (F) were first found in the UTT feldspars from volcanic rocks. Fluorine is the most abundant halogen element in felsic igneous rocks (Dolejs and Zajacz, 2018; Hanley and Koga, 2018), and it strongly affects magma properties and magmatic to hydrothermal cooling processes (Dolejs and Zajacz, 2018). Fluorine easily combines with Ca to crystallize fluorite or apatite, because Ca⁺² has a bond valence (strength) close and familiar to F⁻¹ and PO₄⁻³ (Dolejs and Zajacz, 2018). Fluorite is the most common halide in many fluorine-bearing minerals. Besides, fluorine is taken into major hydrous silicate minerals such as amphibole and mica, because its ionic radius (0.133 nm) is close to that of hydroxyl groups (0.140 nm) (Dolejs and Zajacz, 2018). However, we currently have limited knowledge on the relationship between the behaviors of feldspar and F (or fluorite) (Deer et al., 2001; Harlov and Aranovich, 2018).

In this study, I examined feldspar microtextures in a pyroclastic trachyte from Oki-Dogo to further understand feldspar reactions related to their formation during the cooling through magmatic to hydrothermal stages. This paper reports the novel occurrence of aligned fluorite grains that crosscut two-types of feldspar microtextures, examined using an electron microprobe analyzer (EMPA) and a cathodoluminescence (CL) method. The results develop our knowledge of internal microtextures in volcanic feldspars and behavior of F or fluorite genesis related to feldspar microtextures. The discussion provides new insights into the behaviors of alkali feldspar and F during the cooling of alkaline volcanic rocks.

STUDY AREA AND SAMPLE

Oki-Dogo Island is sub-circular and has a diameter of ∼ 18 km. It is the largest of the Oki Islands in the Sea of Japan, and it lies ∼ 70 km north of the Shimane Peninsula, southwest Japan, at 36°15′N, 133°15′E. The Oki Islands are located in an alkaline province in the Sea of Japan (Tomita, 1935). This area is part of a back-arc region related to the subduction of the Philippine Sea plate relative to the Asian continental plate (Nakamura and Uyeda, 1980).

The geology of Oki-Dogo Island has been summarized by Yamauchi et al. (2009). Upper Miocene to Pliocene alkaline volcanic rocks are distributed across the island, and three alkaline felsic volcanic activities have been distinguished for them as Oki, Tsuzurao, and Hei groups (Uchimizu, 1966). The alkaline volcanic rocks are derived from individual magmas that originated in the lower to middle crust (Uchimizu, 1966; Xu, 1988; Uto et al., 1994; Kobayashi and Sawada, 1998; Kobayashi et al., 2002). The magmas evolved by fractional crystallization from individual parental basaltic magmas. The volcanic activity of the Oki group occurred as lava eruptions at the time from late Miocene to early Pliocene approximately of 6.3-5.1 Ma (Yamauchi et al., 2009). The Tsuzurao group activity occurred as lava and pyroclastic eruptions at nearly the same stage as the Oki group activity. And, the Hei group activity occurred as sheet or dyke intrusions at early Pliocene around 5.1 Ma after the Oki and Tsuzurao groups activities (Yamauchi et al., 2009).

The Tsuzurao Group consists mainly of rhyolitic pyroclastic flow deposits associated with trachytes (Uchimizu, 1966; Xu, 1988). The bulk compositions of several Tsuzurao trachyte (TST) samples taken from different localities were provided by Uchimizu (1966) and Xu (1988). Vent breccias and feeder dikes occur in the eastern and southwestern parts of its distribution area (Sawada et al., 2008). Alkali feldspar from a trachyte fragment from the vent dike yielded a K-Ar age of 5.45 ± 0.17 Ma (Sawada et al., 2008), which is similar in age to the Oki Group and older than the Hei Group (Yamauchi et al., 2009).

A large vent dike, about 400 m wide, consisting of host rhyolite with inclusions of trachyte and basement-derived fragments, outcrops in the middle of the Naka-dani Valley (Sawada et al., 2008). The trachyte samples used in the current study were taken from the vent dike. The elongated trachyte fragments, which are variably undulated, several to several tens of centimeters in width, and several tens of centimeters to 1 m in length, are included in the host pyroclastic rhyolite. The pyroclastic trachyte mainly consists of alkali feldspar with subordinate fayalite, hedenbergite-ferroaugite, alkali amphiboles (arfvedsonite-riebeckite), and opaque minerals (Sawada et al., 2008). The corroded porphyritic alkali feldspars, several millimeters in length, are mostly fragmented; however, there are some euhedral crystals. The trachyte matrix is glassy and heterogeneously devitrified to fine materials (Fig. 1).

Figure 1. Photomicrographs of TST phenocryst alkali feldspars. (a-1) and (b-1) plane-polarized light, (a-2) and (b-2) cross-polarized light. P-mp, patch microperthite; Px, pyroxene; Afs, alkali feldspar.

ANALYTICAL METHODS

Several thin sections were prepared by cutting the sample off along two planes perpendicular each other. Optical photomicrographs were taken using a petrographic microscope of Olympus BX50 and a Cannon camera of KISS X4 at Lake Biwa Museum, Kusatsu, Japan. Backscattered electron (BSE) images, secondary electron images, element distribution maps (EDMs), and feldspar chemical compositions were obtained using an EMPA (JXA-8800M; JEOL) at Shiga University, Otsu, Japan (Nakano et al., 2005, 2016). A focused beam was used for imaging and element mapping, and a 5-µm beam was used to determine the chemical compositions of feldspars under the conditions of an accelerating voltage of 15 kV and a probe current of 8 nA. Standards provided by JEOL were used for calibration. The compositional data were filtered according to the criteria between 98 and 102 wt% of the sum of the component oxides and 4.95-5.05 of the sum of cation numbers proportioned to O = 8 (Nakano et al., 2016). Monochromatic CL images at the wavelengths of 430 and 720 nm were obtained using an EMPA (JXA-8230; JEOL) combined with a photomultiplier (Hamamatsu) and a grating monochromator (Ritsu) at Shiga University under the conditions of a 15 kV accelerating voltage, a 50 nA probe current, a resolution of 400 × 400 pixels, and a counting time of 25-200 ms/pixel. CL spectra ranging from 300 to 900 nm were obtained using the same system. CL colors were observed using a Luminoscope at Okayama University of Science, Okayama, Japan.

RESULTS

Feldspar textures and compositions

Clear and turbid feldspars areas coexist in phenocryst alkali feldspars (Fig. 1). The clear feldspars are free of micropores, while the turbid feldspars that are irregularly and heterogeneously developed from rims toward interiors, contain many micropores, most of which are nano-size. The turbidity observed under a microscope, is caused by nano-size micropores in microperthitic alkali feldspars (Walker et al., 1995; Nakano et al., 2019). Element distribution maps clearly show that turbid feldspars are patchy microperthite (Figs. 2a-2e), and that clear feldspars free of turbidity consist of domains around and wider than 100 µm in size with undulated or wavy boundaries (Figs. 2f and 2g). Among mapped elements, Ca distribution patterns are very complex and enigmatic across the domain and microperthitic feldspars [Figs. 2(b-2)-2(g-2)]. Several elements, such as Fe, Ti, Ca, and P, which are present in the Fe-Ti oxide and apatite grains in microperthitic areas [Figs. 2(c-4) and 2(d-2)].

Figure 2. (a) EDMs of phenocryst alkali feldspars with clear domain and turbid microperthitic textures. Small fluorite (Fl) grains are elongated and aligned across two types of microtextures (domain and microperthite), as understood by viewing both Ca and F maps (b), (c) and then viewing individual Ca maps without F maps (d)-(g) (Fl align, alignment of nano-size fluorite grains). Fluorite alignments are elongated across both clear domain and microperthitic feldspars (b)-(f) and (g). Small microperthitic areas seen here and there (b)-(d), (f), and (g) are incipient of hydrothermal perthite coarsening. Fe is depleted in microperthitic feldspars but concentrated into Fe-Ti oxide grains (c-4). Apatite grains are scattered irrelevantly to fluorite alignments (d-2).

Alignments of fluorite grains elongate along two or three directions (hereafter fluorite alignments) and crosscut the above microtextures with curving and/or bending (Fig. 2). The size of individual fluorite grains is nano-size, which is shown in individually very small Ca spots in original Ca maps before the processing to the present maps in Figures 2 and 4, as in the UTT and HET cases (Nakano and Makino, 2022, 2024). Fluorite alignments are arrays of nano-size fluorite grains arranged at some small intervals. The fluorite alignments are characteristically braided. The braided occurrence of fluorite alignments in TST feldspars quite differs from non-braided alignments along the Murchison and a-axis directions developed in UTT feldspars (Nakano and Makino, 2022), but similar to that in the HET feldspars (Nakano and Makino, 2024).

The TST feldspar compositions are presented in a ternary feldspar diagram (Fig. 3) and selected compositions are listed in Table 1. In this paper, we use the abbreviations of Kf, Naf, and Caf to specifically indicate end-member feldspars, instead of Or (orthoclase), Ab (albite), and An (anorthite) usually used to date but referred to a range of compositions. The compositions are separated into the three groups of intermediate, K-rich and Na-rich feldspars (Fig. 3 and Table 1). The compositions of the domain feldspars intermediate in the Kf/Naf ratio complicatedly vary with overlapping between Ca-rich (around 4 mol% Caf) and Ca-poor (around 1-2 mol% Caf) domains (Fig. 3). In this paper, the boundary value of Caf content that separates Ca-rich and Ca-poor domains is conveniently set to be roughly 3 mol% irrelevant to Kf/Naf ratios. The microperthitic K-rich and Na-rich feldspars vary in composition: the K-rich feldspars are almost depleted in Caf; however, the Na-rich feldspars vary from Ca-rich (∼ 10 mol% Caf) to Ca-depleted (near 0 mol% Caf).

Figure 3. Compositions of TST feldspars (black closed circles) plotted on the K-feldspar (Kf)-Na-feldspar (Naf)-Ca-feldspar (Caf) ternary diagram. The abbreviations of Kf, Naf, and Caf are used to indicate end member feldspar compositions, instead of Or (orthoclase), Ab (albite), and An (anorthite) that may be referred to a range of compositions, respectively.

Table 1. Representative feldspar compositions in the TST

	Featureless feldspar									Perthitic feldspar
	Domain									Na-rich				K-rich
(wt%)	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15	16	17
SiO2	65.75	65.92	65.93	66.03	65.90	66.39	66.35	66.98	66.03	66.38	66.68	68.78	68.16	64.84	64.84	0.82	65.14
Al2O3	19.08	19.29	19.03	18.71	18.95	18.92	18.86	19.02	18.93	20.22	20.25	20.19	19.56	18.09	17.48	17.09	17.77
Fe2O3	0.31	0.28	0.29	0.39	0.25	0.41	0.27	0.21	0.19	0.05	0.00	0.52	0.09	0.04	0.61	0.09	0.06
CaO	0.92	0.85	0.76	0.70	0.58	0.46	0.43	0.32	0.48	1.46	1.07	0.37	0.15	0.18	0.13	65.54	0.01
Na2O	6.62	6.53	6.34	6.26	6.10	6.78	6.07	6.10	4.74	9.87	10.37	10.82	11.22	1.31	0.86	17.94	0.64
K2O	6.71	6.89	7.29	7.50	7.75	6.59	7.72	8.06	10.13	0.56	0.29	0.41	0.23	15.86	16.02	0.08	16.47
BaO	0.00	0.00	0.00	0.01	0.05	0.01	0.00	0.00	-	0.03	0.00	0.00	-	-	-	0.00	-
P2O5	0.02	0.01	0.00	0.05	0.08	0.13	0.03	0.00	-	0.00	0.00	0.08	-	-	-	0.02	-
Total	99.41	99.77	99.63	99.64	99.67	99.69	99.72	100.70	100.50	98.57	98.66	101.18	99.40	100.32	99.93	101.58	100.09
Atomin proprtions (O = 8)
Si	2.969	2.966	2.975	2.983	2.976	2.983	2.990	2.991	2.981	2.949	2.954	2.975	2.994	2.994	3.012	0.073	3.015
Al	1.015	1.023	1.012	0.996	1.009	1.002	1.002	1.001	1.007	1.059	1.058	1.030	1.013	0.984	0.957	0.998	0.969
Fe	0.010	0.009	0.009	0.012	0.008	0.013	0.008	0.007	0.006	0.002	0.000	0.015	0.003	0.001	0.020	0.004	0.002
Ca	0.045	0.041	0.037	0.034	0.028	0.022	0.021	0.015	0.023	0.069	0.051	0.017	0.007	0.009	0.007	3.002	0.001
Na	0.579	0.570	0.554	0.549	0.535	0.591	0.530	0.529	0.414	0.850	0.891	0.907	0.955	0.118	0.077	0.969	0.057
K	0.387	0.396	0.419	0.432	0.447	0.378	0.444	0.459	0.584	0.032	0.016	0.023	0.013	0.934	0.949	0.003	0.972
Ba	0.000	0.000	0.000	0.000	0.001	0.000	0.000	0.000	-	0.001	0.000	0.000	-	-	-	0.000	-
P	0.001	0.000	0.000	0.002	0.003	0.005	0.001	0.000	-	0.000	0.000	0.003	-	-	-	0.001	-
Total	5.005	5.005	5.006	5.007	5.006	4.993	4.995	5.002	5.015	4.962	4.970	4.971	4.984	5.040	5.022	5.049	5.016
Kf-Naf-Caf ternary compositions*
Naf	57.3	56.6	54.8	54.1	53.0	59.6	53.3	52.7	40.6	89.3	93.0	95.8	97.9	11.1	7.5	6.8	5.6
Kf	38.3	39.3	41.5	42.6	44.3	38.1	44.6	45.8	57.2	3.4	1.7	2.4	1.3	88.0	91.9	92.9	94.4
Caf	4.4	4.1	3.7	3.3	2.8	2.3	2.1	1.5	2.3	7.3	5.3	1.8	0.7	0.9	0.6	0.3	0.1

^* The terms Kf (K-feldspar), Naf (Na-feldspar), and Caf (Ca-feldspar) are used to specifically indicate the end-member feldspar compositions instead of the termas, Or, Ab, and An, that are usually used but may be referred to a range of compositions in this paper.

Feldspar CL emissions

Phenocryst feldspars exhibit blue colors in clear feldspar areas under a luminoscope, but exhibit red colors in microperthitic feldspar areas (Figs. 4a and 4b). A phenocryst feldspar, shown in the middle of Figure 1b, was subjected to CL mapping: CL intensity maps (CL maps) at wavelengths of 430 nm (blue emission) [Figs. 4(d-2) and 4(e-2)] and 720 nm (red emission) [Figs. 4(d-3) and 4(e-3)]. In the CL maps at 430 nm, microperthitic areas and associated cracks appeared dark due to dull emissions, contrasting to intense emissions in adjacent domain feldspar areas. However, there is no such distinct contrast in the CL maps at 720 nm between domain feldspars with weak emissions and turbid microperthitic feldspars with dull emissions. These characteristics of CL emissions in the TST feldspars are clearly shown in their CL spectra (Fig. 4f). In Figure 4f, blue emission intensities of clear Ca-rich and Ca-poor domain feldspars and turbid microperthitic feldspars are distinctly different. On the other hand, red emissions are not so different between them as far as the data obtained by the present CL system.

Figure 4. Luminoscope images (a) and (b), EDMs (c), (d-4), and (e-4)-(e-6), BSE image (d-1), CL intensity maps measured at 430 and 720 nm (d-2), (d-3), (e-2), (e-3), and CL spectra (f). Under a luminoscope, CL colors of clear domain feldspars are blue, and those of microperthitic feldspars are red. Blue emissions from microperthitic feldspars are dull or none (d-2) and (e-2), in contrast to intense emissions from domain feldspars. However, there is no distinct difference in red emissions between domain and microperthitic feldspars (d-3) and (e-3). Fe depletion is observed in microperthitic feldspars (d) and Ti depletion is also observed there (e). The differences of CL emissions from individual feldspar phases are clearly shown in the CL spectra (f). Blue emissions are intense in domain feldspars but nearly non-luminescent in microperthitic feldspars. On the other hand, red emission intensities are not so different through all the feldspar phases.

DISCUSSION

Domain textures

The present domain textures are comparable to those observed in UTT and HET volcanic feldspars (Nakano and Makino, 2022, 2024). They are also comparable to domain textures observed in hypersolvus syenite alkali feldspars (Nakano et al., 1997; Nakano, 1998; Nakano et al., 2005), which are partly replaced by low-temperature microperthites. Replacement reactions in feldspars advance through diffusion-controlled processes (e.g., Petrovíc, 1973; Neusser et al., 2012; Abart et al., 2022) and/or dissolution and reprecipitation processes (e.g., Tsuchiyama and Takahashi, 1983; Johannes and Holtz, 1992; Putnis, 2002, 2009; Niedermeier et al., 2009). The former processes advance without the genesis of nano-size micropores and typically with gradual compositional changes. The latter processes advance with the genesis of many micropores that cause of microscopic turbidity (e.g., Walker et al., 1995; Putnis, 2002, 2009) and with sharp compositional gaps. Microscopic turbidity is absent in domain textures in the TST (Fig. 1). Individual compositional changes are small in and between domains in a total variation range from Or₂₅ to Or₅₇ (Fig. 3) and are enigmatically complex typically with gradual compositional changes (Fig. 2f), similarly to those in the UTT. As in the UTT, despite the difference of rock types, the origin of the domain textures is attributable to high-temperature (magmatic) diffusion-controlled replacement reactions, rather than dissolution-reprecipitation reactions.

We have reported that the replacement formation of domain textures in the Oki-Dogo trachyte feldspars was associated with the outward transport of Ca from feldspars to a melt (Ca-decrease or depletion in phenocryst feldspars) (Nakano and Makino, 2022, 2024). It is similarly considered in the present TST that the change from Ca-rich to Ca-poor domains (Figs. 3 and 5) was associated with the transport of Ca ions from feldspars toward the melt (present matrix), probably corresponding to a temperature decrease during the magma cooling. The temperatures, at which feldspar reactions occurred, were estimated by referring to isotherms generated by Wen and Nekvasil (1994) (Table 2 and Fig. 3). Calcium-rich domain feldspar compositions as primary phases crystallized from the magma, which are conveniently recognized as having ≥3 mol% Caf, are distributed approximately between two isotherms of 950 and 850 °C under 500 MPa (Table 2 and Figs. 3 and 5). On the other hand, the secondary Ca-poor compositions, which are poorer below 3 mol% in Caf content, are distributed approximately between two isotherms of 900 and 800 °C.

Figure 5. Ternary diagram showing the whole compositional variation from high-temperature (T) primary Ca-rich domain feldspars (area I indicated by a blue eclipse) to low-T hydrothermal microperthitic two feldspars (areas of III indicated by two pink eclipses) through secondary high-T Ca-poor domain feldspars (area II indicated by a green eclipse). Ca depletion from area I to area II feldspars is indicated by a short black arrow. A finer pink line (A) with arrows slightly dipped from the Kf-Naf sideline connects two III areas of microperthitic K-rich and Na-rich feldspars. A nearly horizontal pink line B with arrows below line A connects extreme compositions of microperthitic two feldspars depleted in Ca. Ca depletion from line A to line B in microperthitic feldspars is indicated by a short black arrow between two symbols, A and B.

Table 2. Temperatures (T) readable from feldspar compositions (Fig. 3), using isotherms in the Kf-Naf-Caf ternary feldspar system and solvus in the Kf-Naf binary feldspar system

Feldspar phase and composition	Estimated T		T-estimation tool
	500 MP (pressure) 100 MP
Ca-rich domain feldspar ≥Caf 3 mol%	∼ 950-850 °C	∼ 900-800 °C	Isotherms*
Ca-poor domain feldspar <Caf 3 mol%	∼ 900-800 °C	∼ 750-700 °C	Isotherms*
	(Independent from pressure)
Perthitic feldspar pair (Kf94-Naf98)	∼ 300 °C		Solvus**

^* calculated according to Wen and Nekvasil (1994), ^** Brown and Parsons (1989).

Microperthitic textures

Patch microperthites are common in plutonic alkali feldspars (e.g., Lee and Parsons, 1995; Hashimoto et al., 2005), but they are rare in volcanic alkali feldspars. Nakano (1990) reported metasomatic patch microperthites in the Koto rhyolitic pyroclastic (welded) rocks from Japan. The present textures that domain and microperthitic textures coexist are similar to those in the aforementioned hypersolvus syenite feldspars, rather than those in the Koto feldspars. The origin of the TST microperthites is attributed to low-temperature hydrothermal replacement reactions as indicated in Figure 5, evidenced by a lot of micropores and the resulting microscopic turbidity caused by dissolution-reprecipitation processes (e.g., Walker et al., 1995; Lee and Parsons, 1995; Putnis, 2002, 2009).

Low-temperature hydrothermal reactions that form plutonic microperthitic feldspars are ordinarily considered to be isochemical due to intracrystalline replacement in the Kf-Naf-Caf ternary feldspar system (e.g., Lee and Parsons, 1995; Parsons and Lee, 2009), which produces K-rich feldspar around Kf₉₀ (orthoclase-microcline) and Na-rich feldspar (albite-oligoclase). Compositions of microperthitic two feldspars eventually approach Kf and Naf as the end member feldspars during the proceeding hydrothermal reactions, respectively (e.g., Lee and Parsons, 1995; Hashimoto et al., 2005). The compositional variation of the present microperthitic two feldspars corresponds to this process, and that the feldspars depleted in Ca are recognized to be final products of hydrothermal reactions. As a result, the average (bulk) compositions of Ca-depleted K-rich and Na-rich feldspars finally become lower in Ca (or Caf) content than primary and secondary domain feldspars, as shown by a line III-B in Figure 5. Microperthitic Na-rich feldspars retaining some Ca contents are considered to be on the halfway of the hydrothermal reactions with the transport of Ca outward feldspars. In this context, the TST microperthite formation is not isochemical but metasomatic. Metasomatic reactions in plagioclase feldspars have been recognized as common with hematite formation (e.g., Engvik et al., 2008; Plümper and Putnis, 2009). It is confirmed in this paper that such metasomatic replacement reactions also occur in volcanic alkali feldspars, after the finding in the aforementioned Koto alkali feldspars.

Temperatures at which hydrothermal reactions occur are impossible to estimate according to Wen and Nekvasil (1994), in which low-temperature feldspar compositional pairs are out of calculation. Alternatively, temperatures are read from tsolvus of the Kf-Naf binary system (Brown and Parsons, 1989; Deer et al., 2001). The solvus temperature estimated using an extreme compositional pair is approximately 300 °C (Table 2).

Fluorite alignments

Fluorine, together with other halogens, is an important element for affecting various magmatic to hydrothermal processes. To date, various experimental and theoretical approaches have been made to investigate F behaviors in magmas (e.g., Dingwell et al., 1985; Carroll and Webster, 1994; Mysen et al., 2004; Aiuppa et al., 2009; Doherty et al., 2014; Dalou et al., 2015; Dolejs and Zajacz, 2018; Cassidy et al., 2022; Feisel et al., 2022; Sharpe et al., 2022). Fluorite is the most common mineral together with apatite, taking F into its structure (Hanley and Koga, 2018). Among the various approaches, an important one is to elucidate the crystallization conditions of fluorite in various magma systems (melts, fluid, or vapor) (e.g., Scaillet and Macdonald, 2001, 2004; Li et al., 2020). Another important approach is to find records of F and fluorite behaviors in natural magmatic to hydrothermal systems (e.g., Christiansen et al., 1983; Congdon and Nash, 1991; Marshall et al., 1998; Melluso et al., 2012, 2014). However, the natural occurrence of fluorite in feldspars has rarely been reported to date: early formed fluorite in granite feldspars from Corsica (Bonin, 1986), hydrothermally replaced fluorite in pegmatite feldspars (Pivec, 1974), and alignments of nano-size fluorite grains in the Balmaceda syenite feldspars (Nakano et al., 2002). The report of fluorite grains in volcanic feldspars is limited to the finding in UTT and HET feldspars (Nakano and Makino, 2022, 2024). The present low-temperature hydrothermal fluorite alignments that cut microperthitic textures is the first finding from not only volcanic and but also plutonic feldspars, beyond the cases of the previous Oki-Dogo volcanic and Balmaceda syenite feldspars.

The TST braided fluorite alignments cut the two types of (domain and microperthitic) metasomatic feldspar textures (Fig. 1). This suggests that the TST fluorite alignments were formed by replacement reactions at lower temperatures after the two-stage formation of domain and patch microperthitic feldspars. Although Snow and Kidman (1991) already pointed out that a trace amount of F replacing O strikingly enhances diffusion of alkali ions in alkali feldspars, F contents primarily taken into O sites of feldspar structures are generally very low: Mosenfelder et al. (2015) reported that F contents in the examined feldspars are mostly below 5 ppm (wt%), with an exceptional high content of 39 ppm. In the replacement reactions, F forming fluorite may have been furnished from the matrix into the feldspars during the cooling. On the other hand, Ca ions in the primary domain feldspar were transported or leached out to the residual melt or fluid and re-distributed. The Ca enrichment in the fluid at the latest stage during the cooling could result in the re-transportation and re-precipitation of Ca together with F, leading to the present occurrence of fluorite alignments that are elongated across both the domain and microperthitic feldspars.

Feldspar CL emissions

Previous CL studies of feldspars have focused on plutonic alkali feldspars (in granites and syenites), but those of volcanic alkali feldspars have been rare (Ginibre et al., 2004; Nakano and Makino, 2022, 2024). It is distinct in the TST feldspars that the high-temperature domain feldspars are blue, but the low-temperature microperthitic feldspars are red under a luminoscope (Figs. 4a and 4b). This CL contrast being more distinct in the CL spectra (Fig. 4f) has not been reported from volcanic feldspars.

Blue CL emissions from alkali feldspars are caused by Al-O-Al defects and Ti impurities, and red CL emissions are caused by Fe⁺³ ions (e.g., Mariano, 1988; Finch and Klein, 1999; Götze et al., 2000; Lee et al., 2007; Kayama et al., 2010). Titanium ions cause blue emissions even below EMPA detection limits (Lee et al., 2007). Blue emission peaks due to Ti impurities are detected at longer wavelengths than those caused by Al-O-Al defects (Lee et al., 2007; Kayama et al., 2010). The presence of Ba⁺² in alkali feldspars may contribute to the peaks of blue emissions, too (Słaby et al., 2008). Following these studies, the broad bands of blue emissions in the CL spectra (Fig. 4f) observed in this study are attributed to Al-O-Al defects (dominant) and Ti impurities (subordinate), apart from the contribution of Ba ions. On the other hand, the broad bands of the weak red emissions are attributable to the presence of Fe⁺³ in tetrahedral sites of feldspar crystal structures (Finch and Klein, 1999; Kayama et al., 2010), although the contents of Fe₂O₃ in feldspars are generally below 0.3 wt% (Nakano et al., 2005).

Mariano (1988) described that CL emissions in alkali feldspars change notably during alkali metasomatism. Finch and Walker (1991) reported that an increase in the microporosity of alkali feldspar causes an increase in red emissions. And, Finch and Klein (1999) elucidated in syenite feldspars that hydrothermal alteration reactions cause a decrease in blue emissions due to increasing Al-ordering that erases Al-O-Al defects, and that an increase of red CL emissions with an increase of ordered Fe⁺³ in the feldspar tetrahedral sites.

The change of CL emission intensities between the feldspar phases generated during the cooling is distinct especially in the blue range and is separated into the following two-step decreasees (weakening) of blue emission intensities: the first-step decrease is recognized in the change from the primary Ca-rich domain feldspars to secondary Ca-poor domain feldspars (Fig. 4f), and the second-step one is recognized in the change from the domain feldspars to microperthitic (K-rich and Na-rich) feldspars. Blue emissions drastically decrease to a dull or non-luminescent state in the micropertitic feldspars (Fig. 4f). The second-step change is consistent with the results by Mariano (1988), Finch and Walker (1991), and Finch and Klein (1999) that reported drastic blue emission decreases during hydrothermal reactions forming microperthites, as mentioned above.

On the other hand, red emission intensities are not so changed between the feldspar phases. Thus, red emissions are relatively more intense than blue emissions in the microperthitic feldspars, and, as a result, the microperthitic feldspars emit visible red colors under a luminoscope (Figs. 4a and 4b). The intensities of red emissions, which may increase corresponding to the depletion of Fe₂O₃ as Fe³⁺ from around 0.3 wt% in the domain feldspars to around 0.05 wt% in the microperthitic feldspars (Table 1), do not so differ throughout the feldspar phases. This result may imply that, in red emissions, the decrease of Fe³⁺ content in the low-temperature microperthitic feldspars (Finch and Klein, 1999) supplements the increase of ordered Fe³⁺ in them, as far as estimated from the data obtained in the present CL measurement system.

CONCLUDING REMARKS

The coexistence of high-temperature metasomatic domain and low-temperature microperthitic textures in volcanic feldspars are for the first time observed in the TST feldspars from Oki-Dogo, Sea of Japan, which provides several implications for feldspar reactions in volcanic rocks. The two-stage metasomatic reactions are together associated with Ca transport outward feldspar grains with Ca depletion in feldspars, which shows that the behavior of Ca should be notable comparably to those of K and Na in alkali feldspar reactions. The metasomatic formation of subsolidus microperthitic feldspars has been hardly known, in contrast to the enormous reports that microperthites in (plutonic) alkali feldspars are products of in situ isochemical replacement (deuteric) reactions. The occurrence of irregularly braided alignments of fluorite grains across the domain and microperthitic textures is also a novel finding through various host rocks. This finding implies that the occurrence of fluorite in feldspars is more common than has been known and advocates the further necessity of examining fluorine behaviors related to feldspars and other major anhydrous minerals.

ACKNOWLEDGMENTS

I thank Dr. K. Makino for continuing collaboration with my feldspar studies, Dr. H. Nishido for permission to use a luminoscope, and Dr. Y. Satoguchi for help to take optical photomicrographs of thin sections. I also thank Dr. M. Enami and an anonymous reviewer for critical reading and providing constructive comment to improve the manuscript greatly.

REFERENCES

•  Abart,  R.,  Petrishcheva,  E.,  Habler,  G.,  Sutter,  C., et al (2022) Evolution of chemically induced cracks in alkali feldspar: thermodynamic analysis. Physics and Chemistry of Minerals, 49, 14.
•  Aiuppa,  A.,  Baker,  D.R. and  Webster,  J.D. (2009) Halogens in volcanic systems. Chemical Geology, 263, 1-18.
• Bonin, B. (1986) Ring complex granites and anorogenic magmatism (English translation). pp. 189, North Oxford Academic Publishers, London.
•  Brown,  W.L. and  Parsons,  I. (1989) Alkali feldspars: ordering rates, phase transformations and behaviour diagrams for igneous rocks. Mineralogical Magazine, 53, 25-42.
• Carroll, M.R. and Webster, J.D. (1994) Solubilities of sulphur, noble gases, nitrogen, chlorine, and fluorine in magmas. In Volatiles in Magmas (Carroll, M.R. and Holloway, J.R. Eds.). Reviews in Mineralogy, 30, Mineralogical Society of America, 231-227.
•  Cassidy,  M.,  Ivenson,  A.A.,  Humphreys,  M.C.S.,  Mather,  T.A., et al (2022) Experimentally derived F, Cl, and Br fluid/melt partitioning of intermediate to silisic melts in shallow magmatic systems. American Mineralogist, 107, 1825-1839.
•  Christiansen,  E.H.,  Burt,  D.M.,  Sheridan,  M.F. and  Wilson,  R.T. (1983) The petrogenesis of topaz rhyolites from the western United States. Contributions to Mineralogy and Petrology, 83, 16-30.
•  Congdon,  R.A. and  Nash,  W.P. (1991) Eruptive pegmatite magma: Rhyolite of the Honeycomb Hills, Utha. American Mineralogist, 76, 1261-1278.
•  Dalou,  C.L.,  Le Losq,  C. and  Mysen,  B.O. (2015) Solubility and solution mechanisms of chlorine and fluorine in aluminosilicate melts at high pressure and high temperature. American Mineralogist, 100, 2272-2283.
• Deer, W.A., Howie, R.A. and Zussman, J. (2001) Framework Silicates Feldspars (2nd Edition). pp. 972, Rock-forming minerals Vol.4A, Geological Society of London, Bath.
•  Dingwell,  D.B.,  Scarfe,  C.M. and  Cronin,  D.J. (1985) The effect of fluorine on viscosities in the system Na₂O-Al₂O₃-SiO₂: implications for phonolites, trachytes and rhyolites. American Mineralogist, 70, 80-87.
•  Doherty,  A.L.,  Webster,  J.D.,  Goldoff,  B.A. and  Piccoli,  P.M. (2014) Partitioning behavior of chlorine and fluorine in felsic melt-fluid-apatite systems at 50 MPa and 850-950°C. Chemical Geology, 384, 94-111.
• Dolejs, D. and Zajacz, Z. (2018) Halogens in silisic magmas and their hydrothermal systems. In The role of Halogens in Terrestrial and Experiment Geochemical Processes: Surface, Crust, and Mantle (Harlov, D.E. and Aranovich, L. Eds.). pp. 1030, Springer International, Cham. Switzerland, 431-543.
•  Engvik,  A.N.,  Putnis,  A.,  Fitz Gerald,  J.D. and  Austroheim,  H. (2008) Albitization of granitic rocks: The mechanism of replacement of oligoclase by albite. Canadian Mineralogist, 46, 1401-1415.
•  Feisel,  Y.,  Castro,  J.M.,  Helo,  C. and  Dingwell,  D.B. (2022) The effect of halogens (F, Cl) on the near-liquidus crystallinity of a hydrous trachytic melt. American Mineralogist, 107, 1007-1017.
•  Finch,  A. and  Walker (1991) Cathodoluminescence and microporosity in alkali feldspars from the Bla Mane So Perthosite, South Greenland. Mineralogical Magazine, 55, 583-589.
•  Finch,  A. and  Klein,  J. (1999) The causes and petrological significance of cathodoluminescence emission from alkali feldspar. Contributions to Mineralogy and Petrology, 135, 234-243.
•  Ginibre,  G.,  Wörner,  G. and  Krontz,  A. (2004) Structure and dynamics of the Laacher See magma chamber (Eifel, Germany) from major and trace elements zoning in sanidine: a cathodoluminescence and electron microprobe study. Journal of Petrology, 45, 2197-2223.
• Götze, J., Krbetschek, M.R., Harbermann, D. and Wold, D. (2000) High-resolution cathodoluminescence of feldspar minerals. In Cathodoluminescence in Geosciences (Pagel, M., Barvin, V., Blanc, P. and Ohnenstetter, D. Eds.). pp. 514, Springer-Verlag, Berlin Heidelberg New York, 245-270.
• Hanley, J.J. and Koga, K.T. (2018) Halogens in terrestrial and cosmic geochemical systems: Abundances, geochemical behaviors, and analytical methods. In The Role of Halogens in Terrestrial and Experimental Geochemical Processes: Surface, crust, and mantle (Harlov, D.E. and Aranovich, L. Eds.). pp. 1030, Springer International, Cham, Switzerland, 21-121.
• Harlov, D.E. and Aranovich, L. (2018) The Role of Halogens in Terrestrial and Experimental Geochemical Processes: Surface, Crust, and Mantle. pp. 1030, Springer International, Cham, Switzerland.
•  Hashimoto,  K.,  Akai,  J. and  Nakano,  S. (2005) Microtextures of alkali feldspar in the Tanakami granite, southwest Japan and their formation processes. Japanese Magazine of Mineralogical and Petrological Sciences, 34, 1-14 (in Japanese with English abstract).
•  Johannes,  W. and  Holtz,  F. (1992) Melting of plagioclase in granite and related systems: compositions of coexisting phases and kinetic observations. Transactions of Royal Society Edinburgh: Earth Science, 83, 417-422.
•  Kayama,  M.,  Nakano,  S. and  Nishido,  H. (2010) Characteristics of emission centers in alkali feldspar: a new approach by using cathodoluminescence spectral deconvolution. American Mineralogist, 95, 1783-1795.
•  Kobayashi,  S. and  Sawada,  Y. (1998) Origin of the Latest Miocene alkaline rocks from Oki-Dogo island, SW-Japan. Journal of Petrology, Mineralogy and Econmic Geology, 93, 162-181 (in Japanese with English Abstract).
•  Kobayashi,  S.,  Sawada,  Y. and  Yoshida,  T. (2002) Magma plumbing systems of the Latest Miocene Oki alkaline volcanic group, Oki-Dogo island, SW Japan, based on the geology and petrology. Japanese Magazine of Mineralogical and Petrological Sciences, 31, 137-161 (in Japanese with English Abstract).
•  Lee,  M.A. and  Parsons,  I. (1995) Dislocation formation and albitization in alkali feldspars from the Shap granite. Mineralogical Magazine, 82, 557-570.
•  Lee,  M.A.,  Parsons,  I.,  Edwards,  P.R. and  Martin,  R.W. (2007) Identification of cathodoluminescence activators in zoned alkali feldspars by hyperspectral imaging and electron-probe microanalysis. American Mineralogist, 92, 243-253.
•  Li,  X.,  Zhang,  C.,  Wang,  L.,  Behrens,  H. and  Holtz,  F. (2020) Experiments on the saturation of fluorite in magmatic systems: Implications for maximum F concentration and fluorine-cation bonding in silicate melt. Journal of Earth Sciences, 31, 456-467.
• Mariano, A.N. (1988) Some further application of cathodoluminescence. In Cathodoluminescence in Geological Materials (Marshall, D. Ed.). pp. 146, Urwin Hyman Boston, USA, 94-123.
•  Marshall,  A.S.,  Hinton,  R.W. and  Macdonald,  R. (1998) Phenocrystic fluorite in peralkaline rhyolites, Olkaria, Kenya Rift Valley. Mineralogical Magazine, 62, 477-486.
•  Melluso,  L.,  de’Gennaro,  R.,  Fedele,  L.,  Franciosi,  L. and  Morra,  V. (2012) Evidence of crystallization in residual, Cl-F-rich agpaitic, trachyphonolitic magmas and primitive Mg-rich basalt-trachyphonolite interaction in the lava domes of the Phlegrean Field (Italy). Geological Magazine, 149, 532-550.
•  Melluso,  L.,  Morra,  V.,  Guarino,  V.,  de’Gennaro,  R., et al (2014) The crystallization of shoshonite to peralkaline trachyphonolitic magmas in a H₂O-Cl-F-rich environment at Ischia (Italy), with implications for the feeder system of the Campania Plain volcanoes. Lithos, 210, 242-259.
•  Mosenfelder,  J.L.,  Rossman,  G.R. and  Johnson,  E.A. (2015) Hydrous species in feldspars: A reassessment based on FTIR and SIMS. American Mineralogist, 100, 1209-1221.
•  Mysen,  B.O.,  Cody,  G.D. and  Smith,  A. (2004) Solubility mechanisms of fluorine in peralkaline and meta-aluminous silicate glasses and in melts to magmatic temperatures. Geochimica et Cosmochimica Acta, 68, 2745-2769.
•  Nakamura,  K. and  Uyeda,  S. (1980) Stress gradient in arc-back-arc regions and plate subduction. Journal of Geophysical Research, 85, 6419-6428.
•  Nakano,  S. (1990) Behavior of alkali feldspar during hydrothermal alteration of the Koto Ryolites, Southwest Japan. Earth Science (Chikyu Kagaku), 44, 61-76.
•  Nakano,  S. (1992) Internal textures and chemical compositions of anti-rapakivi mantled feldspars from Oki-Dogo island, Japan. Mineralogy and Petrology, 46, 123-135.
•  Nakano,  S. (1998) Calcium distribution patterns in alkali feldspar in a quartz syenite from Oki-Dozen, southwest Japan. Mineralogy and Petrology, 63, 35-48.
•  Nakano,  S. (2021) Magmatic resorption reactions in anti-rapakivi feldspars: A case study of ferro-augite trachyte feldspars from Oki-Dogo, Japan. Lithos, 398-399, 106288.
•  Nakano,  S. and  Suwa,  K. (1995) Mantled feldspars in alkali rhyolite from Oki-Dogo island, Japan. Mineralogy and Petrology, 54, 277-289.
•  Nakano,  S.,  Hosokawa,  E. and  Akai,  J. (1997) Calcium distribution in alkali feldspar of a quartz syenite from Cape Ashizuri, southwest Japan. Mineralogical Journal, 19, 75-86.
•  Nakano,  S.,  Akai,  J. and  Sugaki,  A. (2002) Fluorite particles inducing butterfly aggregates of incipient microperthite in alkali feldspar from a syenite, the Patagonian Andes, southern Chile. American Mineralogist, 87, 1377-1383.
•  Nakano,  S.,  Akai,  J. and  Shimobayashi,  N. (2005) Contrasting Fe-Ca distributions and related microtextures in syenite alkali feldspar from the Patagonian Andes, Chile. Mineralogical Magazine, 69, 523-537.
•  Nakano,  S.,  Kojima,  S.,  Makino,  K.,  Kayama,  M., et al (2016) Cryptoperthitic and replacive intergrowths with iridescence in monzonitic rocks from Cerro Colorado, northern Chile. European Journal of Mineralogy, 28, 355-374.
•  Nakano,  S.,  Makino,  K.,  Yoshida,  I.,  Maniwa,  K., et al (2019) Combined influences of iron-oxides and micropores in reddish coloration of alkali feldspars in granitic rocks. Journal of the Geological Society of Japan, 125, 759-773.
•  Nakano,  S. and  Makino,  K. (2022) Feldspar crystallization and the behavior of fluorine in the Utagi trachyte, Oki-Dogo Island, Japan. Lithos, 418-419, 106676.
•  Nakano,  S. and  Makino,  K. (2024) Magmatic processes forming replacement textures with fluorite alignments in feldspars in an evolved trachyte from Oki-Dogo Island, Sea of Japan. Island Arc, 33, e12509.
•  Neusser,  G.,  Abart,  R.,  Fischer,  F.D.,  Harlov,  D. and  Norberg,  N. (2012) Experimental Na/K exchange between alkali feldspar and NaCl-KCl salt melt: chemically induced fracturing and element partitioning. Contributions to Mineralogy and Petrology, 164, 341-358.
•  Niedermeier,  D.R.D.,  Putnis,  A.,  Geisler,  T.,  Golla-Schindler,  U. and  Putnis,  C.V. (2009) The mechanism of cation and oxygen isotope exchange in alkali feldspars under hydrothermal conditions. Contributions to Mineralogy and Petrology, 157, 65-76.
•  Parsons,  I. and  Lee,  M.R. (2009) Mutual replacement reactions in alkali feldspars I: microtextures and mechanism. Contributions to Mineralogy and Petrology, 157, 641-661.
•  Petrovíc,  R. (1973) The effect of coherency stress on the mechanism of the reaction albite + K⁺ = K-feldspar + Na⁺ and on the mechanical state of the resulting feldspar. Contributions to Mineralogy and Petrology, 41, 151-170.
•  Pivec,  E. (1974) Psuedographic potash feldspar-fluorite intergrowths from Krupka near Teplice-spa (Krusne Hory MTS), Czechoslovakia. Neues Jahrbuch fur Mineralogy, Monatshefte, 10, 447-453.
•  Plümper,  O. and  Putnis,  A. (2009) The complex hydrothermal history of granitic rocks: Multiple feldspar replacement reactions under subsolidus conditions. Journal of Petrology, 50, 967-987.
•  Putnis,  A. (2002) Mineral replacement reactions: from macroscopic observations to microscopic mechanism. Mineralogical Magazine, 66, 689-708.
• Putnis, A. (2009) Mineral replacement reactions. In Themodynamics & Kinetics of Water-Rock Interaction (Oelkers, E.H. and Schott, J. Eds.). pp. 569, Reviews in Mineralogy and Geochemistry, 70, Mineralogical Society of America, Washington, D.C., 87-124.
•  Sawada,  Y.,  Tome,  A.,  Kitagawa,  H.,  Kobayashi,  S., et al (2008) Geology and petrography of vent breccias of the Latest Miocene Tsuzurao Formation, and alkali feldspar K-Ar age from a trachyte essential fragment, Oki-Dogo Island, SW Japan. Geoscience Report Shimane University, 27, 11-20 (in Japanese with English abstract).
•  Scaillet,  B. and  Macdonald,  R. (2001) Phase relations of peralkaline silisic magmas and petrogenetic implications. Journal of Petrology, 42, 825-845.
•  Scaillet,  B. and  Macdonald,  R. (2004) Fluorite stability in silicic magmas. Contributions to Mineralogy and Petrology, 147, 319-329.
•  Sharpe,  M.X.,  Barker,  S.J.,  Rooyakkers,  S.M.,  Wilson,  C.J.N., et al (2022) A sulfur and halogen budget for the large magmatic system beneath Taupo volcano. Contributions to Mineralogy and Petrology, 177, 95.
•  Słaby,  E.,  Götze,  J.,  Worner,  G.,  Simon,  K., et al (2008) K-feldspar phenocrysts in microgranular enclaves: A cathodoluminescence study of crystal growth as a marker of magma mingling dynamics. Lithos, 105, 85-97.
•  Snow,  E. and  Kidman,  S. (1991) Effect of fluorine on solid-state alkali interdiffusion rates in feldspar. Nature, 349, 231-233.
•  Tomita,  T. (1935) On the chemical compositions of the Cenozoic alkaline suite of the circum-Japan Sea region. Journal of Shanghai Science Institute, Section II, 1, 227-306.
•  Tsuchiyama,  A. and  Takahashi,  E. (1983) Melting kinetics of plagioclase feldspar. Contributions to Mineralogy and Petrology, 84, 345-355.
•  Uchimizu,  M. (1966) Geology and petrology of alkali rocks from Dogo, the Oki islands. Journal of the Faculty of Science, University of Tokyo, Section 2, 12, 85-159.
•  Uto,  K.,  Takahashi,  E.,  Nakamura,  E. and  Kaneoka,  I. (1994) Geochronology of alkali volcanism in Oki-Dogo island, southwest Japan: Geochemical evolution of basalts related to the opening of the Japan Sea. Geochemical Journal, 28, 431-449.
•  Walker,  F.D.L.,  Lee,  M.R. and  Parsons,  I. (1995) Micropores and micropermeable texture in alkali feldspars: geochemical and geophysical implications. Mineralogical Magazine, 59, 505-534.
•  Wen,  S. and  Nekvasil,  H. (1994) SOLVCALC: an interactive graphics program package for calculating the ternary feldspar solvus and for two-feldspar geothermometry. Computers & Geosciences, 20, 1025-1040.
•  Xu,  H. (1988) Petrology and geochemistry of the alkali rocks from Dogo, Oki islands, Shimane Prefecture, southwestern Japan. Science Report of Tohoku University, Series III, 17, 1-106.
• Yamauchi, S., Sawada, Y., Takasu, A., Komuro, H., et al. (2009) Geology of the Saigo District. Quadrangle 1:50,000, Geological Survey of Japan AIST, pp. 121 (in Japanese with English Abstract).