Origin and timing of spilitic alterations in volcanic rocks from Głuszyca Górna in the Intra-Sudetic Basin, Poland

The formation of spilitic assemblages (i.e. chlorite and albite) has been ubiquitously involved during the evolution of continental early-Permian volcanics from the Intra-Sudetic Basin (ISB). Based on the investigation of laccolith-type and variably-altered trachyandesite exposure in the vicinity of Głuszyca Górna (Lower Silesia, Poland), we have demonstrated that apatite fission-track dating (AFT), coupled with chlorite geothermometry, can be successfully applied to denote the timing of low-temperature alterations within volcanic rocks. The primary magmatic assemblages of the trachyandesites (i.e. augite and andesine-labradorite) have been affected by chloritization and alblitization respectively, followed by the formation of secondary titanite, celadonite, and calcite. The chlorite species have crystallized in the range of 106–170 °C, that exceeds Apatite Partial Annealing Zone (70–110 °C). The secondary, nearly pure albite (Ab ~ 99 mol.%) with weak to dark-brown cathodoluminescence replaces primary plagioclase (~ An37–50Ab47–58Or2–4) along the cleavage and/or twinning planes during Al3+–conservative reaction. The accessory apatite is marked by swallow-tail terminations indicative of rapid cooling formation conditions. It shows homogenous chemical composition, high F− content, and pink to yellow (REE3+ and Mn2+-activated, respectively) cathodoluminescence. Based on the AFT dating, the development of spilitic alterations within the early-Permian (ca 290 Ma) laccolith from Głuszyca could not only span the range of 182–161 Ma (Middle Jurassic), but also occurred prior to large-scale geological events in the ISB, such as burial under late-Mesozoic sediments, as well as tectonic inversion and exhumation. Whole-rock geochemistry of trachyandesites altered to various extent, indicates that original trace elements concentrations, except for i.e. Sr, Cs, and Ba, could be preserved during low-temperature alteration (spilitization). Meanwhile, geochemical fingerprint of the volcanics (i.e. humped-shaped mantle normalized trace element diagrams and positive Zr–Hf anomaly) points to the crustal contamination during magma evolution, combined with the mantle metasomatism in the source via subduction-derived components (i.e. fluids), as shown by i.e. low Nb/Th and Nb/LREE ratios.

www.nature.com/scientificreports/ medium-temperature (up to ca. 350 °C) process depending on the physicochemical properties of alternating fluids.
The formation of secondary albite in basaltic lavas, followed by the development of associated low-temperature (secondary) mineral assemblages such as chlorite, calcite, actinolite, prehnite, or epidote, commonly refers to so-called spilitization 13 . This process has been first described by Brongniart 14 and leads to the formation of spilites (also known as spilitic rocks). The exact nature of spilites (so-called spilite problem 13,15 ) is not entirely understood owing to the origin of albite-chlorite assemblage and corresponding source of Na + . Overall, spilitization has been reported from basic to intermediate rocks where it may originate from: (1) water-rock interactions during e.g. submarine eruptions or intrusions of magma into water-rich sediments, which are both capable of providing necessary amounts of Na + for albitization, (2) auto-hydrothermal (late-magmatic or post-magmatic) alterations related to the interactions between early-formed mineral phases with coexisting residual (deuteric) fluids, and (3) low-grade metamorphic reactions during the burial stage 13 . Nevertheless, the primary nature of spilites-linked to either formation of the specific "hydrous", Na-CO 2 -rich magma or mixing magma with hot brines-has been also proposed. However, the latter theory was nearly abandoned these days. Spilitization plays an important role in the recycling of nitrogen within the crust 16 and may induce the development of carbonatization in the presence of dissolved CO 2 17 . Furthermore, spilitization commonly obliterates original microtextural and mineralogical features of the rocks, but is also capable of overprinting their original geochemical signature, including main and trace element abundances (e.g. REE and LILE; e.g. 18 ) and isotopic ratios (i.e. Sr and Nd; see: 19 ). Hence, the isotopic homogenization during spilitic alterations has been thus used to constrain the timing of spilitization 19 .
Albitization (including spilitization) and K-metasomatism are conspicuously presented among early-Permian volcanogenic formations in Central Europe (see i.e. 20 ). The spilitic assemblages, represented by albite, chlorite, riebeckite, and/or uralite have been particularly reported from upper parts of basaltic shallow-level magmatic bodies that cover a wide area of the Intra-Sudetic Basin (ISB) in Lower Silesia Poland 21 . In general, it has been accepted that albitization (spilitization) of early-Permian volcanics from the Intra-Sudetic Basin has resulted from late-to post-volcanic alkaline emanations that affected not only volcanic rocks, but also adjacent sediments 22 . Additionally, the formation of spilitic rocks in the ISB was spatially and genetically linked to the development of agate mineralization, as shown by Powolny et al. 23 .
In this paper, we have aimed at the reconstruction of mineral-replacement reactions during low-temperature alteration (spilitization) of continental volcanic rocks from the Intra-Sudetic Basin based on variably altered samples collected from laccolith-type exposure in the vicinity of Głuszyca Górna (Lower Silesia, Poland). Particularly, we show that combination of chlorite thermometry and apatite fission-track dating (AFT) can be successfully applied to establish the timing of low-temperature hydrothermal fluid flow responsible for the formation of spilitic assemblages in volcanic rocks. Finally, the whole rock major and trace element data provided a reconstruction model of magma evolution and the effects of secondary alterations on the possible redistribution of trace elements, such as REE and HFSE.

Geological setting and field observations
The study area is situated within the Intra-Sudetic Basin and falls into NE flank of the Bohemian Massif, which in turn represents a large eastern exposure of European Variscan Belt in Central Europe. The ISB is filled with volcanogenic-sedimentary succession and exhibits a fault-bounded synclinorial structure of ca. 70 km long and 35 km wide that extends into NW-SE direction (e.g. 20,24 ). It originally appeared as an intramontane depression confined by tectonically active margins 25 . The incipient stages of the ISB evolution began during Mid-Visean 26 and were strictly related to the regression of the sea that appeared in SW Poland during the Devonian period. The Intra-Sudetic Basin is mantled by crystalline basement units of Variscan age and Late Palaeozoic sedimentary basins. According to Aramowicz et al. 27 , Sobczyk et al. 28 , and Botor et al. 29 , both volcanogenic and sedimentary rocks of the Bohemian Massif (including the Intra-Sudetic Basin) show quite complex thermo-tectonic histories imprinted by the sediment burial connected with transgression during the Cenomanian, Europe-Africa plate convergence followed by tectonic inversion and reactivation of faults and thrusts, as well as post-magmatic hydrothermal activity 30 .
Głuszyca trachyandesite exposure of early-Permian age is located ca. 20 km SE from Wałbrzych and 15 km NW from Nowa Ruda (Lower Silesia, Poland)- Fig. 1A-B. It comprises a laccolith-type magmatic body of ca. 200 m in depth and 300 m in length 31,32 that was actively exploited in the past. According to Kozłowski 31 and Awdankiewicz 32 , the volcanics from the study area belong to the so-called second volcanic cycle of the Middle Rotliegendes and mildly-alkaline suite of the Rybnica Leśna Volcanic Association, respectively. The magmatic body reveals a well-developed, horizontal-oriented system of joint planes 31 . It lies over sandstones with interlocations of shales, but appears covered by acidic igneous rocks (i.e. rhyolitic tuffs)- Fig. 2A. The middle areas of the exposure reveal the presence of shales interbedded with carbonaceous limestones (the latter has been partially affected by post-magmatic silification)-see Fig. 2B.
The emplacement of volcanics from the Głuszyca exposure occurred during the climax of Variscan Orogeny and hence reflects the peak of magmatic activity developed within the ISB 31 . This magmatic activity has developed in subsidence-related areas such as ISB and resulted in the formation of alternating basic (i.e. trachyandesites or trachybasalts) and acidic (porphyries and rhyolites) rocks. Meanwhile, clastic and chemical sediments (i.e. shales, sandstones, and limestones) separate basic and acidic magmatic rock and thus reflect the presence of relatively short intervals that lacked in magmatic activity 31 . Additionally, previous petrological studies (e.g. 21,32,33 ) have shown that early-Permian volcanism in the study area was marked by post-collisional and extension-related (continental) affinity.   Fig. 1B) of the subvolcanic laccolith-type magmatic body exposed abandoned Głuszyca quarry (after 31 ; modified via own field observations); Note the overall 3 sampling areas: Loc. 1 (lower parts of the laccolith), Loc. 2 (middle parts at the contact zone with adjacent intraeruptive sediments), and Loc. 3 (top portions of the laccolith). (B) Field photo showing the contact between trachyandesites and intra-eruptive deposits (shales with partially-silicified limestones). The cross-section was generated using CorelDrawX6 software. Petrography and mineral geochemistry. Pyroxene. Pyroxene occurs as prismatic subhedral crystals up to ca 0.3 mm in size and has been partially (i.e. Loc. 1, sample GL_01A) or almost entirely (samples GL_01B, GL_02, and GL_03) converted into deep-green massive or bundle-like aggregates of chlorites (see: Fig. 3A-B). Locally, the pervasive alteration of pyroxene has also led to the formation of other, fibrous and pale-green chlorite crystals (see: Fig. 3D). SEM-BSE observations revealed that pyroxenes are represented by both Na-rich (1) and Ca-rich (2) species. Both of them are predominately clustered together (i.e. Na-rich pyroxene is rimmed by Ca-rich pyroxene), but also occur in the form of individual crystals ( Fig. 4A-B). Meanwhile, the boundary between Na-rich and Ca-rich pyroxenes phases is quite sharp as shown in Fig. 4B. Na-rich (1)  , and corresponds to aegirine based on Q-Jd-Aeg relationship after Morimoto 34 . The presence of aegirine was further confirmed by Raman microspectroscopic measurements (Table 3, Fig. 5). Its diagnostic bands are found at 951 and 542 cm −1 and assigned to Si-O nbr and Si-O br stretching vibrations, respectively 35 . Under SEM-BSE images, aegirine often appears riddled as it shows etched cleavage planes, forming so-called chevron (V-shaped) pattern, as well as the presence of co-existing (replacive?) Mg-chlorites. It also occasionally comprises tabular calcite up to ca. 50 µm in size. The aegirine locally form intergrowths with magnetite that not only shows thin oriented trellis-type ilmenite lamellae, but also alters to rutile (or anatase)- Fig. 4A and D. Moreover, aegirine is frequently surrounded by fan-shaped aggregates of titanite followed by minor amounts of quartz.
Conversely, Ca-rich pyroxene (2) is relatively depleted in Na 2 O (~ 0.34 wt.%), but rich in CaO (~ 11.67 wt.%), with average formula of (Ca 0. 69 Table 3) and comprises two strong marker bands at 1012 and 663 cm 1 attributed to Si-O nbr and Si-O br stretching modes 35 . Augite occasionally exhibits oscillatory zonation under SEM-BSE images maintained by the variations in Fe and Mg contents. Additionally, it is accompanied by minor amounts of hornblende, as well as minute elongated crystals of baddeleyite (ZrO 2 )- Fig. 4C. The breakdown of augite into chlorite has also been noted, although in the vast majority of investigated samples the pervasive chloritization has totally wiped out original microtextural features of the rocks and, thus the detailed identification of pre-existing pyroxene is not available. Table 1. Mineralogical and micro-textural description of the four samples types (GL_01A, GL_01B, GL_2, GL_03), which has been distinguished on the basis of alteration style and degree (i.e. chloritization and albitization) and collected from various (i.e. lower-Loc. 1, middle-Loc. 2, and upper-Loc. 3) regions of Głuszyca trachyandesite laccolith. *Note the samples GL_01A and GL_01B were distinguished due to variable albitization and chloritization degree. The former were marked by partial albitization of primary plagioclases and partial chloritization of primary pyroxenes, whereas the latter revealed partial albitization and pervasive chloritization. www.nature.com/scientificreports/ Feldspar-group minerals. Primary plagioclase that has been partially preserved from pervasive albitization was recognized only in the samples collected from the lowermost parts of the laccolith (Loc. 1)-see Fig. 6A and B. It forms equigranular lath-shaped crystals (up to 0.4 mm in size) with greenish, greenish-grey, and/or bluish luminescence. The CL spectrum of primary plagioclase comprises a broad (blue) band centred at 452 nm, accompanied by the weak-intensity peak at ca. 731 nm. The former band could be attributed to either Al-O --Al centres, related to the coupled K-Si-Ba-Al substitution, or Al-O-Ti bridges, or both defects 21,36 , whereas the latter band corresponds to traces of Fe 3+ in the mineral structure. Partially and/or completely albitized plagioclase is frequently rimmed by unaltered and bright-blue to dark-blue luminescent K-feldspars (Fig. 6A). The CL spectrum of K-feldspar is fairly similar to this collected from green-luminescent plagioclase, since it consists of the strong band at ca. 469 nm (Al-O --Al centres or Ti 4+ impurity) and negligible (Fe 3+ -related) band at ca.   (Table 4). Thus, it represents either andesine or labradorite members of the feldspar group. Primary plagioclase also contains trace amounts of TiO 2 (~ 0.09 wt.%) and MgO (~ 0.09 wt.%), and K 2 O (0.49 wt.%).
The pervasive albitization of trachyandesites from Głuszyca (see Fig. 6C and D) has resulted in the presence of almost chemically pure (Ab ~ 99 mol.%- Table 4) secondary albite that forms elongated patches penetrating through host magmatic andesine-labradorite and displays weak to dark-brown cathodoluminescence ( Fig. 6C and D). The albite occurs as non-twinned crystals with "dusty" appearance in thin sections, but original twinning of primary plagioclase (andesine-labradorite) seems to be only locally preserved in completely albitized samples (Fig. 3E). The contact between secondary albite and andesine-labradorite is sharp on the micrometer scale, as shown in the BSE images (Fig. 7A). Furthermore, albitization of andesine-labradorite was accompanied by the emergence of numerous micropores ( Fig. 7A and B) and the occasional formation of minute rutile (or anatase) crystals.
Fluorapatite. Fluorapatite forms euhedral or subhedral acicular crystals showing quite large variations in size (from several µm up to ca. 0.4 mm). It exhibits pinkish and/or greyish-blue luminescence (Fig. 8A), which is well-visible in prismatic sections. In contrast, tiny apatite crystals, as well as those situated perpendicular to the c-axis in thin sections display yellowish CL colours. A discrete yellow-luminescent glow has also developed along the margins of particular larger crystals. CL observations revealed that fluorapatite locally exhibits skeletal hollow-type fabrics characterized by the preservation of crystal outlines (Fig. 8A), with their interior infilled by other phases such as secondary albite. Additionally, few crystals revealed the presence of H-shaped (swallowtail) terminations (Fig. 8B). The CL spectrum of pink-luminescent fluorapatite is marked by strong lines at 377, 424, and 456 nm. These bands correspond to the trace amounts of REE 3+ within apatite structure 37 including Ce 3+ and Tb 3+ , as well as Eu 2+ , respectively. Additionally, the low-intensity band located at 574 nm is likely related to the traces of Mn 2+37 . The spectrum of yellow-luminescent apatite is similar to the one recorded for   Table 3. Representative composition (EMPA) of greenish-gray luminescent primary plagioclase and secondary (weak-luminescent to dark-brown luminescent) replacive albite (based on 8 oxygen atoms). *After i.e. 35 ; nb -non-bridging modes, br -bridging modes; s and w correspond to strong and weak intensity diagnostic Raman bands. www.nature.com/scientificreports/ apatite with pink CL colour, as its strongest (REE 3+ -related) lines were observed at 377 (Ce 3+ -activator) and 439 (Tb 3+ -activator), along with 464 nm (Eu 2+ -activator). However, the Mn 2+ -related band found at 584 nm displays quite higher intensity relative to the corresponding line from pink-luminescent apatite. Fluorapatite displays lack of chemical zonation in BSE images (Fig. 8B). The compositional variations of apatite from various parts of trachyandesite exposure are presented in Table 5 (sample GL_02). Meanwhile, the fluorapatite shows a strong predominance of F over Cl, as F/Cl ratios are within the range of 8.04-19.95. The content of OH (calculated from stoichiometry, i.e. assuming F + OH + Cl = 2) is very low and varies between 0 and0.10 wt. %, while SO 3 is below detection limits.
Chlorite. Under plane-polarized light, three types of chlorite-group species were distinguished in volcanics: (1) replacive (after pyroxene-group minerals) dark-green aggregates exhibiting massive and bundle-like fabrics, (2) replacive (after pyroxene-group minerals) pale-green fibrous crystals, as well as (3) non-replacive, vugs-filling (colourless) crystals embedded within celadonite. Their compositions are presented in Table 6.    www.nature.com/scientificreports/ Chlorite geothermometry. The temperature of chlorite crystallization has been assessed using five independent empirical geothermometers ( Table 7). The temperatures obtained using these methods depend directly on the chemical composition of chlorites and take into account the contents of Si 40 , Al IV41 , as well as combined Al IV content and Fe/Fe + Mg ratios [42][43][44] . The semi-empirical (graphical) approach proposed by Bourdelle and Cathelineau 45 , has been also applied for comparison. This geothermometer assumes chlorite + quartz to be in equilibrium and utilizes the activities of chlorite end-members. As a result, it was found that the formation temperature of replacive chlorites, i.e. both massive dark-green (1) and fibrous pale-green (2) varieties could be in the range of 106-170 °C (Table 7). Overall, the lowest temperature values (~ 107 °C) were provided according to the formula proposed by Kavalieris et al. 40 . The low formation temperatures of chlorites seem to be also supported by quite high octahedral vacancy (up to 0.40 a.p.f.u.), which tends to decrease along with increasing formation temperature 46 43 , and 125-157 °C 42 for replacive chlorites of (1) and (2) type. Additionally, the local presence of high alkalis (Na + Ca + K) in dark-green massive replacive chlorites (1) is typical of low-temperature chlorites found in basaltic rocks and may account for i.e. the mixedlayer chlorite-smectite phases 47 . The distribution of formation temperatures is roughly unimodal as shown in Fig. 10. The results obtained from empirical (chemical) geothermometers are also consistent with T-R 2+ -Si plot 45 as the analytical points mostly fall close to the 150 °C isotherm (Fig. 11) for the fibrous pale-green type of chlorite (2). Otherwise, massive dark-green chlorite (1) plots between 150 and 200 °C isotherm. Since the reliability of the semi-empirical geothermometer is partially maintained by the knowledge of Fe 2+/ Fe 3+ ratios, these results possibly reflect the presence of some amounts of Fe 3+ in chlorite structure and corresponding overestimation of R 2+ (sum of divalent cations) parameter. The non-replacive chlorite (3) associated with celadonite is believed to crystallize at roughly similar temperature conditions (Table 7; Fig. 10). Its average formation temperatures were estimated in the range of 120-168 °C according to particular equations based on Al IV content and/or Fe/Fe + Mg Table 4. Representative composition (EMPA) of fluorapatite from the samples with variable alteration degrees (based on 26 oxygens following the procedure proposed by 85 . The crystals oriented with their c-axes perpendicular to the incident electron beam were selected for the analyses to reduce the accumulation of halogens close to the mineral surface (see e.g. 86 ). However, a slight and local excess of halogens (> 2.0 a.p.f.u.) was occasionally presented. The areas showing both yellowish and pinkish CL colours have been investigated individually, but they turned to present a similar distribution of particular elements. *Note that Fe 2 O 3 was calculated from charge balance. www.nature.com/scientificreports/ ratios. Following T-R 2+ -Si plot 45 , the formation temperature of non-replacive chlorites spanned the range of 125-175 °C (Fig. 11). Conversely, notably lower values (~ 107 °C) were revealed by Si-based geothermometer after Kavalieris et al. 40 .   Tables 8 and   9. The samples have passed the chi-squared test (χ 2 ) that confirms the homogeneity of the ages. Central ages obtained from AFT dating span from 182 ± 43.2 to 161.0 ± 32.7 Ma and thus roughly correspond to the Middle-Jurassic period (Fig. 12). Confidence intervals for single-grain ages are provided in Supplementary Data 2 and 3 for the full data set The halogen contents of fluorapatite crystals (F, Cl, and OH) were constant according to EMPA results (Table 5)     Nonetheless, all AFT ages are younger than the igneous emplacement of the magmatic body, which occurred during the early-Permian period (ca. 290 Ma, although the exact age remains unknown due to lack of detailed geochronological studies and absence of i.e. zircon, which could be useful to obtain crystallization age). Confined track lengths were measured for the samples GL_02 and GL_03 from Loc. 2 and 3 (n = 52 and n = 27, respectively)-see Table 9. The confined track-length distribution of the spontaneous tracks could be described as fairly unimodal (Fig. 13), with mean measured confined tracks lengths (MTL) of 12.23 ± 0.26 μm (GL_02) and 12.40 ± 0.29 μm (GL_03). Additionally, the standard deviation (SD) of MTL is low for grains found in both samples (GL_02-1.90 μm; GL_03-1.49 μm). Mean Dpar values, found in the range of 2.0-2.12, are typical of fluorapatite, which is marked by a relatively high susceptibility to annealing 48 .  Table 6. Representative composition (EMPA) of replacive chlorite (after aegirine and/or augite) and chlorites associated with celadonite (based on 14 oxygen atoms). Note: Al IV = 4-Si IV a.p.f.u.; ΣT: tetrahedral occupancy (Si + Al IV ); ΣOct: octahedral occupancy(Al VI + Fe + Mg + Mn); □: octahedral vacancy; R 2+ : sum of divalent cations (Fe + Mg + Mn) assuming all iron to be ferric (Fe 2+ ).   Fig. 14B. The curves are characterized by negative troughs for Nb and Ta and corresponding enrichment in LREE, as well as in U and Th. Otherwise, LILE (large ion lithophile elements) such as Cs, Rb, Ba, Sr, and K show remarkable variations in the mantle-normalized patterns. The strongly-altered samples from upper parts of the laccolith (i.e. GL_03) are relatively depleted in Sr and Ba, but enriched in Cs, Rb, and K relative to the samples from lower parts of the laccolith (with lower alteration degree-i.e. GL_01A-B). Additionally, the sample GL-01A from the lowermost part of the laccolith shows notably lower U contents. All samples are marked by slightly positive Zr and Hf spikes relative to HREE, coupled with notable negative anomalies for Ti and P.

Discussion
The origin and nature of spilitic assemblages. The formation of spilitic assemblages in Głuszyca trachyandesite laccolith involved chloritization of primary pyroxenes (i.e. augite) and albitization of primary plagioclases (andesine-labradorite). These alterations were followed by the concomitant formation of celadonite and  www.nature.com/scientificreports/ vug-filling chlorites, as well as iron oxides (after i.e. primary magnetite/ilmenite) in the uppermost parts of the laccolith. It is also noteworthy, that the abundance of calcite indicates high activity of CO 2 during metasomatic processes, whereas the formation of secondary titanite could be triggered by the release of Ca 2+ and Ti 4+ during the breakdown of pyroxene and magnetite/ilmenite, following such reaction as: Although spilitic alterations have strongly overprinted the original mineralogical features of the trachyandesites, the samples from lower parts of the laccoliths still provide some information on primary magmatic assemblages. Particularly, these samples contain unique association of sodium-and calcium-bearing pyroxenes (i.e. aegirine mantled by augite-see Fig. 4B). Although augite is quite common in volcanogenic rocks from the ISB basin where it has been reported from poorly-spilitized samples 21 , the occurrence of aegirine has not been reported elsewhere in that study area. However, it remains ambiguous whether aegirine represents primary (I) or secondary (II) phase hosted by volcanics from Głuszyca trachyandesites. Firstly, direct crystallization from magma (I) is not supported by geochemical (meta-to peraluminous) character of the samples and associated mineral assemblages. Conversely, (II) secondary replacive or even authigenic low-temperature (see i.e. 49 ) nature   Table 9. Apatite fission-track length and Dpar (mean etch-figure diameter parallel to the c axis) data for the samples collected from various areas of the Głuszyca trachyandesite laccolith. Note the sample from Loc. 1 was excluded due to insufficient amounts of confined tracks. nCT-number of measured confined tracks; CT mean-mean confined track length; SD-standard deviation; CT skew-skewness of distribution relative to the mean value (measure of asymmetry of the distribution); n Dpar-number of etch pit diameters measured; Dpar mean: mean etch pit diameter; Dpar SD-standard deviation of etch pit diameters; Dpar skew: skewness of etch pits. www.nature.com/scientificreports/ does not conform to microtextural observation, i.e. the presence of augite overgrowing on aegirine, as well as the possible chloritization of aegirine. Moreover, aegirine occurs as relatively large prismatic crystal found exclusively in less altered rocks; no sodic pyroxene occurs in any strongly-albitized samples. Thus, its presence cannot be explicitly related to the formation of other Na-bearing minerals (i.e. secondary albite) as a result of Na metasomatism. Alternatively, aegirine crystals may be interpreted as xenoliths, although further investigations are necessary to support such a type of scenario. The formation of secondary albite is common in spilitic rocks worldwide and has been widely reported in volcanic rocks from the ISB, including the samples investigated in the following study. The mineralogical and microtextural characteristics of secondary replacive albite, characterized by weak to dark-brown cathodoluminescence and almost pure chemical composition with Ab ~ 99 mol.%, gives a clue for its formation conditions in the Głuszyca trachyandesites. These features are typical of authigenic and/or diagenetically-altered feldspars (i.e. 50 ) and correspond to the ordering of crystal lattice and/or vanishing of structural defects within secondary albite during low-temperature fluid-mineral interactions (i.e. 51 ). The vanishing CL intensity within secondary albite might be also related to the incorporation of structural water (i.e. OHgroups) 52 . Meanwhile, the albitization is mainly developed in the interior of primary plagioclase laths (andesine-labradorite), and thus was likely   www.nature.com/scientificreports/ enhanced by i.e. the presence of polysynthetic twinning in primary andesine-labradorite or cleavage planes and other internal fractures (cf. 1 ). Based on EMPA and microtextural data, the transition from primary plagioclase to secondary albite could follow simplified, allochemical, and Al-conservative reaction: Alternatively, albitization could also reflect the constant-volume reaction that assumes some quantities of Al 3+ to be released during mineral replacement reaction: The reaction (2) seems to be more relevant in this case as the secondary albite in Głuszyca trachyandesites is devoid of Al-bearing phases such as mica and/or epidote unless they are represented by nano-scale inclusions, which cannot be visible under BSE images. Such phases have been, however, frequently recognized as cogenetic inclusion found within secondary albite from albitized granitic rocks (e.g. 2 ).
The continental character of Głuszyca trachynadesites (and other volcanics from the ISB) precludes the role of magma-water interactions during the formation of secondary albite typical of spilitic rocks. Thus, magmaticrelated Na-rich fluids could serve as a source of Na + and Si 4+ (cf. 15 ), both elements necessary for albitization of andesine-labradorite (see reactions II and III). The development and extent of albitization might be possibly related to such factors as: the geochemical character of trachynadesite-forming magma (i.e. enrichment in alkalis), high activity of CO 2 , sodium-rich character of primary plagioclase (i.e. andesine) that require relatively less amounts of Na + in alternating fluids, and/or subvolcanic rather than stricte extrusive character of volcanic rocks from the ISB. Nevertheless, it is also possible that some amount of Na + could be derived from chloritization of Na-bearing pyroxene (aegirine) since replacive Mg-chlorite represents Na-free species that could incorporate only such elements as Al, Si, and Fe during the breakdown of primary pyroxene. Thus, Na released during the breakdown of aegirine could be further fixed with secondary albite.
Constraints on the timing of spilitization. The fluorapatite separated for AFT dating represents a latemagmatic phase as evidenced by its size variations (from several µm up to ca. 0.4 mm), coupled with the occurrence of hollow-type tubes and swallow-type (H-shaped) terminations within particular crystals, visible during both OM-CL and SEM-BSE observations. These fabrics were generally reported from minerals that have been subjected to quenching (rapid cooling) and may reflect rapidly-increased growth rate during magma eruption 53 .
According to Gleadow et al. 54 , AFT dating of magmatic rocks does not always indicate the age of their igneous emplacement and/or early-stage crystallization, since the closure temperature of apatite falls within the range 70-110 °C (so-called Apatite Partial Annealing Zone; APAZ). The absolute ("magmatic") ages may be, however, obtained in specific cases, i.e. assuming that particular rocks cooled quickly, being remained close to the surface, and were subsequently unaffected by any remarkable thermal disturbances. The obtained AFT results (161-182 Ma) are significantly younger than the emplacement of the magmatic body, which occurred during Middle Rotliegendes (~ 299-271 Ma) and should be considered as apparent ages that do not reflect the timing of apatite crystallization, but were possibly overprinted by a younger thermal event. These ages cannot be explained by i.e. prolonged residence above APAZ, burial metamorphism, and/or compositional variations of apatite (i.e. variable Cl content). Meanwhile, several lines of evidence indicate that AFT results (161-182 Ma) can indicate the timing of ubiquitous alterations (spilitization), including chloritization of pyroxenes (augite and aegirine) combined with albitization of primary plagioclases (andesine-labradorite). Firstly, (1) the crystallization temperatures obtained from chlorite thermometry (106-170 °C in the case of both replacive and vug-filling chlorites) indicate that alternating fluids were capable of resetting the AFT system, since APAZ falls into ca. 70-110 °C.  www.nature.com/scientificreports/ basement and magmatic-sedimentary succession in the Intra-Sudetic Basin 28,29 . Overall, two major large-scale geological events have strongly imprinted the low-temperature cooling history of the Intra-Sudetic Basin infill: (1) burial under Mesozoic sediments, followed by Europe-Africa-Iberia plate convergence  Ma) that triggered the inversion of the Intra-Sudetic Basin, and subsequent reactivation of faults and thrusts 57,58 as well as (2) Paleogene-Neogene reheating associated with the widespread magmatic activity, which has been triggered by the opening of Eger rift 59 . Therefore, the AFT results of both magmatic and sedimentary rocks found within the Intra-Sudetic Basin are often ambiguous and may lead to various tectonomagmatic implications (cf. 27,60 ). Meanwhile, there are still some contradictions whether the partial reset of AFT ages in the Intra-Sudetic Basin was linked to the transgression of a shallow-level sea during the Cenomanian. This transgression resulted in the deposition of large piles of sediments up to ca. 900 m in thickness that could be responsible for complete and/ or partial reheating of sedimentary and/or magmatic rocks above apatite partial annealing zone (70-110 °C). According to Danišík et al. 60 and Sobczyk et al. 28 , and Botor et al. 29 , the sediments related to the Cenomanian transgression had covered almost the entire area of the Intra-Sudetic Basin and resulted in the pervasive reset of AFT ages for older deposits of Carboniferous and/or Permian age. Otherwise, such authors as i.e. Skoček and Valečka 61 , Biernacka and Józefiak 62 , and Uličný 63 postulated the theory of "Western and Eastern Sudetic Islands". According to this model, the Sudetes area could be rather exposed as an archipelago of islands (paleo-highs) emerging above the sea level and hence burial during the Cenomanian transgression had not necessarily affected some older deposits. However, to the best of our knowledge, the exact paleographic reconstruction of exact position, as well as extent of these island is still quite vague and ambiguous. The results of AFT dating obtained for trachyandesites from Głuszyca (161-182 Ma) indicate that these rocks have not preserved the record of largescale geological events in the Intra-Sudetic Basin such as tectonic inversion and exhumation. These ages are also notably older than the transgression of the Cenomanian sea (~ 95 Ma), and hence seem to support the "Island scenario" in the Intra-Sudetic Basin, indicating no significant influence of Late-Mesozoic sediment deposition on the cooling history of the volcanic rocks. Thus, further investigations are necessary to eventually re-evaluate the low-temperature history of the ISB that involves the presence of emerged landmasses ("Sudetic Islands") during the Cenomanian.
Magma evolution and mobility of trace elements during low-temperature alterations. The samples have probably undergone some fractional crystallization prior to their emplacement, as shown by i.e. low contents of transition elements (e.g. Cr and Ni, Sc, V), and low MgO (< 4.5 wt.%), and low Mg# (0.35-0.50), that resulted from an early-stage separation of mafic minerals such as olivine. Negative Eu anomaly in chondritenormalized patterns (Fig. 14A) can be ascribed to the fractionation of primary (Ca-bearing) plagioclase, whereas negative Ti and P anomalies in the mantle-normalized diagrams (Fig. 14B) may be at least partially linked to fractionation of titanite and apatite, respectively. LREE-enriched rare-earth patterns for the samples are typical of continental basalts 64 . Whereas, high LREE/Nb (i.e. La/Nb ratios of ca. 2.2) and LREE/Ta (i.e. La/Ta ratios of ca. 39.9), along with the corresponding Nb-Ta troughs in the mantle-normalized patterns (Fig. 14B) suggest the involvement of subduction-modified mantle source in the petrogenesis of the trachyandesites from Głuszyca. Nb/Th ratios found within the range of 2.6-3.1 are also typical of arc-related environments 65 . The hydrous melting is also evidenced by the occurrence of sparse amounts of amphibole group species within rock matrix. Otherwise, Nb/U ratios (6.3-12.8) are quite similar to those reported for continental crust (6-10; see: 66 ). Low [Ta/U] PM (< 0.4), enrichment in Zr (up to 502.8 ppm) and Hf (up to 12.2 ppm), as well as a bit humped-shaped mantle-normalized diagrams (Fig. 14B) also indicate crustal affinities (see e.g. 67,68 ). Hence, it may be inferred that crustal contamination could have been strongly involved during the magma evolution. This conclusion is further supported by the presence of interstitial magmatic quartz within rock matrix. Moreover, the occurrence of pink-luminescent apatite (REE 3+ -activated), followed by yellow-luminescent (REE 3+ -and Mn 2+ -activated) outermost domains in particular crystals, indicate changes in melt chemistry (i.e. alkaline-acidic) during magma evolution 69 . The presence of mixed subduction-related and crustal affinities is further supported by various tectonic-setting discrimination diagrams, where the samples show either arc-related (due to i.e. modification of the source via subduction-derived components such as fluids) or within-plate affinity 70-73 -see Fig. 15A-D.
The mobility of trace elements in basaltic rocks during low-grade metamorphism such as spilitization is quite ambiguous. According to Herrmann et al. 74 , the trace element concentration of REEs and HFSEs is resistant to low-temperature alterations and remains constant in both weakly-and strongly-altered rock samples. Conversely, Hellman et al. (1979) 18 suggested these elements are prone to i.e. spilitization, and thus cannot be properly used to i.e. discriminate tectonic setting of altered basaltic rocks. REE patterns for the trachyandesites from Głuszyca are consistent and seem to preserve the original LREE/HREE ratios in both weakly-and strongly-altered samples (cf. Fig. 14A). The mantle-normalized diagrams are also sub-parallel, but there are also some discrepancies in terms of some LILE (i.e. Cs and Sr), as well as U contents in samples of variable alteration degree- Fig. 14B. Firstly, the samples containing pyroxene relicts and partially-albitized plagioclase (i.e. sample GL_01A, Loc. 1, lower parts of the laccolith) are relatively rich in Sr, but depleted in Cs relative to the samples from upper parts of the laccolith that show increasing alteration degree (i.e. albitization and chloritization). These changes could be explained by leaching of Sr and Ba during plagioclase (and partially K-feldspar) alteration (i.e. albitization), whereas the enrichment is Cs could be possibly explained by the weathering of feldspars followed by the formation of hydrous phases (e.g. sericite). Depletion in Ba within the sample from upper parts of the laccolith could be attributed to either alteration/weathering or fractional crystallization of K-feldspars. Finally, variations in U contents (depleted in samples from lowermost parts of the laccolith) is linked to the variable amounts of apatite rather that secondary alterations effect, as evidenced by well-developed positive correlation between P 2 O 5 and U (R 2 = 0.88). www.nature.com/scientificreports/   Table 10). Normalization values are after 84   For semi-empirical estimation of formation temperatures, graphical geothermometer introduced by Bourdelle and Cathelineau 45 has also been applied. Additionally, the chlorites containing ∑(Ca + Na + K) > 0.5 wt.% have been discarded from geothermometric calculations due to the possibility of contamination (i.e. remnants of pyroxenes or submicroscopic inclusions of other phases such as calcite) and/or the presence of interstratified chlorite-smectite phases. The application of thermodynamic approach 75 was excluded due to high (> 3 apfu) Si content among particular chlorite species.
Raman micro-spectroscopy (RS). Raman spectra of pyroxene-group minerals were recorded using Thermo Scientific DXR Raman microscope equipped with 10 × , 50 × and 100 × magnification objectives. The system operated in a confocal mode and worked in a backscatter geometry. The measurement were conducted on polished thin section using a 532-nm laser with 10mW laser power and exposure time of 10 s. The laser focus diameter was 1-2 μm. The spectra were corrected for background by the sextic polynomial method using Omnic software. The identification of mineral phases was supported by CrystalSleuth (http:// rruff. info/) and OMNIC software.
Apatite fission-track dating (AFT). Apatite fission-track thermochronology by the external detector method 76 was carried out at the Institute of Geological Sciences, Polish Academy of Sciences in Krakow (Poland). The external detector method and the ζ age calibration approach were used to determine the fissiontrack ages 77,78 . Polished grain mounts were etched for 20 s in 5 N HNO 3 at 21 °C. The standard glass CN5 was used as a dosimeter to monitor the neutron flux. Thin flakes of low-U muscovite were used as external detectors. Samples together with age standards (Fish Canyon, Durango, and Mount Dromedary apatites) and CN5 standard glass dosimeters were irradiated with thermal neutron nominal flux of 9 × 10 15 n/cm 2 at the Oregon State University TRIGA reactor in the USA. After the irradiation, the muscovite external detectors were etched for about 45 min in 40% HF to reveal the induced tracks. The spontaneous and induced tracks were counted by optical microscopy at 1250 × magnification using a NIKON Eclipse E-600, equipped with motorised stage, digitising tablet and drawing tube controlled by program FTStage 4.04. Data analyses and age calculations based on a Zeta value for CN5 ζ CN5 of 348.18 ± 6.52 were calculated using program Trackkey 4.2 and further presented in the Tables 8 and 9, as well as Figs. 12 and 13. All quoted AFT ages are "central ages" 79 , and the spread of single grain ages was assessed using the dispersion of the central age and chi-square test 80 . Additional calculations were performed using Binomfit software 81 to evaluate the reliability of obtained ages. Single-crystal ages have been further included in Supplementary Data 2 (Binomfit software results) and Supplementary Data 3 (comparison between Trackkey and Binomfit software, with error bars calculated at given coincidence interval). In all analysed samples about 20 apatite grains were selected for analyses. Only clean, defect-and inclusion-free grains were selected for track counting. The etch pit diameter (Dpar) was used to check annealing kinetics and the composition of apatites. At least four etch pits 82 per single analysed grain have been measured. The crystals chosen for confined track measurements had a well-polished surface, parallel to the c-axis. For each sample, as many confined track lengths as possible were measured 83 .