ΣΧΟΛΗ ΘΕΤΙΚΩΝ ΕΠΙΣΤΗΜΩΝ ΤΜΗΜΑ ΓΕΩΛΟΓΙΑΣ ΚΑΙ ΓΕΩΠΕΡΙΒΑΛΛΟΝΤΟΣ

National and Kapodistrian University of Athens, Faculty of Geology and Geoenvironment

Πτυχιακή Διπλωματική Εργασία: Black magnetite beach sands from and islands: research of minerals' origin through mineral analysis, and potential enrichment in specific trace elements through geochemical analysis.

Προπτυχιακή φοιτήτρια: Τραπατσέλη Μυρτώ Α.Μ.:1114201200092

Επιβλέπων καθηγητής:

Σταματάκης Μιχαήλ

Aθήνα, 2017

1

Diploma thesis: Black magnetite beach sands from Milos and Kimolos islands: research of minerals' origin through mineral analysis, and potential enrichment in specific trace elements through geochemical analysis.

Post-graduate Student: Trapatseli Myrto R.N.:1114201200092

Supervisor professor: Stamatakis Michael

National and Kapodistrian University of Athens, Faculty of Geology and Geoenvironment Athens, 2017

2

1.INTRODUCTION………………………………………………………………………………………………………………5

1.1 Purpose of thesis.………………………………………………………………………………………………………..5

1.2 Procedure……………………………………………………………………………………………………………………5

1.3 Tectonical setting and Volcanism of the SAVA…………………………………………………………….5

2. Milos Island……………………………………………………………………………………………….………………...9

2.1 Τectonics……………………………………………………………………………………………….……………….…10

2.2 Geology………………………………………………………………………………………….………….…………....11

2.2.1. Metamorphic basement ………………………………………………………………………………….……12

2.2.2. Neogene sedimentary sequence……………………………………………………………………………13

2.2.3. Volcanic sequence…………………………………………………………………………………………………13

2.2.3.1. Pliocene series……………………………………………………………………………………………………13

Phase I The basal pyroclastic series (Middle to Upper Pliocene)…………………………………….13

Subphase I-II Complex of domes and lava flows (Upper Pliocene)…….………………………..14

Phase II (Upper Pliocene to Lower-Middle Pleistocene)………………………………………………….14

2.2.3.2. Transitional Lower-Middle Pleistocene series…………………………………………………….14

2.2.3.3. Pleistocene series………………………………………………………………………………………………14

Phase III The pyroclastic series and lava domes (Lower Pleistocene)…………………………..….14

Phase IV The acid complex of Fyriplaka and Trachilas (Upper Pleistocene)……………..……..15

2.2.3.5. The products of phreatic activity (from Pleistocene to recent times) - The ‘’green lahar’’ formation……….…………………………………………………………………………………………………..16

2.3. Alluvial sequence…………………………….………………………………………………………………………16

2.4. Hydrothermal activity and mineralization………………………………..……………………………..19

2.4.1. Mineral Alterations and industrial minerals…………………………………………………….…..19

3. Kimolos Island…………………………………………………………………………………..……………………….21

3.1 Geology……………………………………………………………………………………………………………………21

4.Black Magnetite sands (Heavy mineral sands)…………………………………………………………….23

5. Sampling and methodology……………………………………………………………………………………….24

5.1. Sampling………………………………………………………………………………………………………………..24

5.2. Macroscopical examination……………………………………………………………………………………26

5.2. Analytical methods………………………………………………………………………………………………..28

5.2.1. XRD analysis technique……………………………………………………………………………………….28

3

5.2.2. SEM/EDS (Energy-dispersive X-ray spectroscopy) analysis technique…………………29

6. RESULTS…………………………………………………………………………………………………………………..31

6.1. XRD analysis and results………………………………………………………………………………………..31

6.2. SEM/EDS analysis and results………………………………………………………………………………..41

7.DISCUSSION………………………………………………………………………………………………………………42

8.CONCLUSIONS…………………………………………………………………………………………………………..52

Acknowledgments……………………………………………………………………………………………………….53

References…………………………………………………………………………………………………………………..54

Annex I…………………………………………………………………………………………………………………………57

4

1.INTRODUCTION 1.1 Purpose of thesis:

This thesis studies the mineralogy and geochemistry of black magnetite beach sands from Milos and Kimolos islands. The aim of this project is to propose the origin of the black sands’ heavy minerals, focusing especially on magnetite, and the potential enrichment in specific heavy metals, as it came up through geochemical analysis. Black magnetite sands typically consist of heavy minerals such as magnetite, ilmenite, titanomagnetite, monazite, hematite. Heavy minerals, especially magnetite, are found in black sands in streams, creeks, rivers, arroyos, but also in ocean shores, beaches, and bays, via erosion and concentration by high water’s action. Magnetite most commonly occurs in nature in a granular form as a sand or grit and is one of the important ores of iron. It is a common constituent of igneous-volcanic (basically mafic) and metamorphic rocks. It is found as in inclusion in granite too. The assumption of this project is that the black beach magnetite sands from the islands of Milos and Kimolos (especially in Milos case) derive from the volcanic products of intermediate composition that occur in most of the islands extent, which is andesitic rocks, either lavas or pyroclastics, especially tuffs. So, the study of the geological framework of the islands is of critical importance. The sands have been also subjected on geochemical analysis, for the study of their chemical composition and the potential enrichment in the heavy metals: Ba, Mn, Ni, Pb, Zn, Cr. It is general known that heavy mineral sands consist ore deposits, so enrichment in heavy metals of interest could lead in further research. 1.2 Procedure: The samples were gathered from seven locations across the coastline of Milos island and one specific location of Kimolos island. The sampling duration lasted between 6th and 8th of May 2016. The samples (after being subjected on a preparation procedure) were subjected on XRD (X-ray Diffraction) analysis and on SEM/EDS (Energy-dispersive X-ray spectroscopy), scanning 30-40 multiple spots and calculating their median. 1.3 Tectonical setting and Volcanism of the SAVA Milos island is situated in the SW part of the Volcanic Arc, which was created by the subduction of the African Plate under the margin of the Eurasian Plate (Fytikas et al.1986). The Aegean region, in general, is under an active subduction regime, a back-arc extension and an anticlockwise rotation along the North Anatolian Fault which is accommodated by roll-back of the slab southwards. In addition, the westward movement of Turkey relative to Europe and the continental collision between NW -Albania with the Apulia-Adriatic platform in the west govern the kinematics of the deformation. This resulted in an E-W shortening of the northern Aegean, which has been compensated by a fault block rotation and a N-S extension in the back-arc region. This area of extension is the South Aegean region. (Alfieris 2006).

5

Figure 1: Relative motion of the Greek microplate. The arrows indicate the motions along the fault lines. After Papazachos et al. 1998.

The median Aegean crystalline belt is made up of a Paleozoic basement overlain by a Mesozoic carbonate sequence, and has undergone two main metamorphic phases, the older one in late Eocene period, and the newer in lower and middle Miocene (approximately 33.2 Ma). During the subduction of the metamorphic phase in the Eocene, the orogenic sentiments of the Aegean’s area basin were metamorphosed in conditions of high pressure (P)- low temperature (T), resulting in the blue schists (glaucophane) paragenesis, and in some cases (the examples of Ios and islands, (Papanikolaou 2014)) in conditions of really high pressure-low temperatures, resulting in eclogitic phases. The area of the southern Aegean, specifically the ‘’Milos island territory’’ according to Alfieris 2006, remained submerged until the early Oligocene, when started the uprising of the metamorphic formations to the surface, during the period of lower-middle Miocene. During that period, the primary blue schists phases gradually converted into green schists phases, under low P- high T conditions (approximately 33.2 Ma) (Fytikas et al. 1976a). According to Fytikas et al. (1976a), this period is considered to date the last metamorphic event that affected the Attico-Cycladic basement, which is connected with the Eocene-Oligocene continental collision of Africa-Europe, setting the end of the compressive tectonism during the Alpine times. While the subduction was going on, there has been a compressional regime at the external part of southern Aegean and an extensional regime at the area of central Aegean. During middle-upper Miocene, the previous Alpine basement of the Attico- cycladic area has been fragmented into smaller domains: the internal domain of

6

Attica--Anatolia, which gradually evolves into the back-arc basin, and the external domain of Peloponnesus--, which gradually evolves into the orogenetic arc (Papanikolaou 2014, Alfieris 2006). At the internal domain, a proto- Cretan basin started to form approximately at the area where the Cycladic back arc molassic basin was (Papanikolaou and Dermitzakis,1981; Alfieris, 2006). According to Alfieris 2006, the development of the Cretan trough, a fore-arc basin, which separates the sedimentary outer arc (central Peloponnesus, Kythera, Crete, and ) from the present southern Aegean active volcanic arc, was of paramount importance for the development of the tectonic architecture of Milos complex (Alfieris, 2006). The northwestern-trending faults on Milos island is interpreted to have formed during the period of the opening of the Cretan basin, in response to the northeastern extensional regime which took place at the western part of it (Lyberis et al.,1981; Alfieris, 2006). The other principal extensional directions being NW-SE for the Eastern part of the basin of Crete and N-S for the Central part of it (Alfieris, 2006). In addition, alongside with the main compressional regimes in the south area of Aegean territory, there have been acting extensional tenses through the Anatolia slide fault, suppressed by the collision of the Arab micro-plate with the Eurasian and the transition of these tensions to the Microasian micro-plate, during the middle Miocene. As a result, the tenses affected the Aegean micro-plate, giving a clockwise rotation of the region to the south, having an extensional character along NE-SW directions in the back-arc area (Papanikolaou, 2014). Finally, starting from Pliocene (Serravallian) a neotectonic situation occurred, which had an extensional character along NE-SW directions, and was only shortly interrupted by compressional episodes (Angelier 1976,1979; Mascle et al.,1981; Alfieris, 2006), created tectonically bounded sedimentary plateaus and basins. According to Alfieris 2006, the variation of the tectonic regime with the extensional and compressional phases (Angelier et al. 1977; Alfieris, 2006) seems that directly influenced both, the rate at which the magma rose through the continental crust, as well as the composition of the magma reaching the surface. The Southern Aegean volcanic arc is made up of several volcanic centres located above an amphitheatre shaped seismic zone having the maximum depth of 190 Km, dipping 35-40º on the average and interpreted to represent the top of the subducting African slab, the current subducted part of which entered into the subduction regime approximately 12-15 Ma before (Alfieris, 2006; Papanikolaou, 2014), and consist of volcanic products ranging from basaltic, andesitic, dacitic to rhyolitic in composition, all of them displaying a typical calc-alkaline chemical character (Innocenti et al.,1981; Alfieris, 2006).

7

Figure 2: Sketch map of the Aegean region showing position of the Southern Aegean Volcanic Arc in relation to the Hellenic trench.

The current volcanic arc (SAVA extends from the gulf of Saronikos in the west to the western Turkey coast to the east, and the erupted lavas are late-Pliocene to Quaternary in age (Fytikas et al., 1976; Innocenti et al., 1979; Ferrara et al.,1980; Mitropoulos et al. 1986). The SAVA includes the volcanic centers of Sousaki, , Methana, , Milos, , and . The central and eastern sectors of the arc are characterized by large stratovolcanoes with caldera structures (Santorini, Kos, Nisyros) whereas the western sector is mainly characterized by small and generally monogenic in nature, eruptive centers, as in the case of Milos island, which is being studied here. The volcanic activity began in the lower Pliocene (approximately 4,7 Ma at Aegina) (Fytikas et al. 1984; Alfieris, 2006). According to Alfieris (2006), the trench geometry leads to localised deformation and segmentation of the subducting slab, but also, the deformation within the subducted slab, is restricted to segment boundaries. The coincidence of intermediate depth seismicity and volcanic activity in the area, indicates a close link between tectonic features within the subducting slab and active magmatism above it, and suggests that melt generation is concentrated in zones above those parts of the slab that are heavily deformed, and where pore fluids from dehydration reactions are released into the mantle wedge via hydraulic fracturing mechanisms. It is believed that these faults have had a major influence in controlling also, the distribution of the hydrothermalism along the arc generally, and in Milos particularly (Alfieris, 2006).

8

Figure 3: Normal faults trending about N59º and named after the corresponding volcanic centres are responsible for the distribution of the volcanic centres and the shallow, intermediate depth earthquakes along the arc. From Friedrich (2000).

2. Milos Island Milos is located in the west-central part of the south Aegean active volcanic arc (SAVA), and geotectonically belongs to the central-southern Aegean crystalline sequence, known as the Attic- Cycladic massif or the median Aegean crystalline belt which extends eastwards into Menderes massif and westwards into the Pelagonian Massif. The island is a recently emergent 2 Ma volcano of the Pliocene, located in the Southern Aegean volcanic arc. Milos island is composed of late Pliocene submarine volcanic deposits and late Pleistocene-Holocene subaerial volcanic deposits, which overlie Miocene and Lower Pliocene carbonate sedimentary successions (Fytikas et al. 1986; Stewart and McPhie 2006). The volcanic rocks are calc-alkaline, predominantly andesites and lesser dacites. Milos has an active high-enthalpy geothermal system that vents in the shallow (,100 m) marine and terrestrial environments (Valsami-Jones et al. 2005), and the island as a whole has been hydrothermally active for around 1.5 Ma. Recent research has identified a new metallogenic environment—namely, mineralization and geothermal activity associated with emergent volcanoes. (Kilias, 2011)

9

2.1 Τectonics Four main fault trends have been recognized in Milos Island: (1) The NW-SE, which principally influenced and tectonically determined the metamorphic basement. This trend is an arc parallel extensional fault, as it is parallel to the SAVA, and it is the main marginal fault zone direction of the Neogene basins at western Cyclades and southeastern Peloponnesus (Mariolakos and Papanikolaou, 1981; Alfieris, 2006). This direction also includes the large graben of Milos gulf, the Lower Quartenary volcanic activity in Trachilas and Fyriplaka (E. Milos) and the horst of Profitis Ilias-Chondro Vouno area (W. Milos). According to Alfieris 2006, the faults of the basement which were parallel to the subduction zone were most probably formed as steeply deeping transcurrent faults, which displayed predominantly dip-slip components of displacement (Alfieris, 2006). (2) The E-W, which, according to Alfieris 2006, is considered to be the result of two chronologically different processes: an older one which took place before the starting of the volcanic activity in the island, and which is related to that series of tectonic events that affected the basement, and a later one which led to the reactivation of the previous, and which is related to the emplacement of the transgressive series, which took place during Quaternary times. This last one tectonic phase interested affected all the south Aegean Quaternary Volcanic Arc and it is characterized by big block faulting structures. This set of faults affected all the rock types at Milos. (The graben of central part of W.Milos (Ammoudaraki-Ntasifnos-Plakota-Rivari graben) also belongs to this tectonic trend.) (Alfieris, 2006). (3) the N-S, which affected both the metamorphic basement and the overlain magmatic pile. It is considered to be responsible for the ascending of the magma at shallow depths and/or at the surface (Fytikas et al.,1986) and thus responsible for the emplacement of some of the domes at western Milos along this direction. It is also considered that this fault trend has an extensional character and it is related to the EW stretching of the magmatic arc (Alfieris, 2006). The N-S horst and graben structures at Milos are of smaller extend in contradiction with that of NW-SE and E-W trend faults, and include the Zephyria graben (E. Milos), the Chalepa horst (central part of southern Milos), and a small horst that exposes basement rocks east of Profitis mountain (W. Milos) (Alfieris, 2006). (4) The NE-SW, which is approximately normal to the axis of the outer (southern) sedimentary arc. This is also the trend of certain geomorphological features in the island, comprising the volcanic centers (volcanoes, fumaroles and solfatara fields) and the epicentres of the shallow and intermediate depths of strong earthquakes (Papazachos and Panagiotopoulos,1993). Earthquake epicenter locations indicate that, the same major transcurrent structures facilitated oblique convergence, by reactivation during successive earthquakes. This trade is also generally associated with extensional / dilational features (horstails, splays and jogs) and their linking faults mainly at W. Milos (Alfieris, 2006).

10

2.2 Geology

The small archipelago of Milos, that includes the islands of , Kimolos and Polyegos, is sited in the central part of the South Aegean active volcanic arc (Fig. 1), which formed during the Pliocene as a consequence of the northward subduction of the African plate beneath the Aegean one. In this context, the Milos archipelago represents the most important volcanic area where recent magmatic activity favoured the formation of a high enthalpy geothermal field which is considered the most important in Greece (Fytikas and Marinelli, 1976; Fytikas et al. 1986).

Figure 4: The Milos group of islands. The island of Milos (fig.10) comprises four main geological units / sequences (Metamorphic basement, Neogene sedimentary sequence, Volcanic sequence and Alluvial cover) which were first distinguished by Sonder (1924) and then by Fytikas (1977), Fytikas et al. (1986). In 2005, Stewart and McPhie, gave a new approach concerning the volcanic successions, based on the five main volcano types that introduced Fytikas et al. 1986 in their study. Fytikas (1977) studied the geological framework of the island and distinguished: (i) An Alpine metamorphic basement, which is intensively eroded, characterized by a Middle Eocene high pressure/low temperature metamorphic event (45 +-5 Ma) overprinted by a high temperature/low pressure one at the boundary between Oligocene and Miocene (25 +- 5 Ma) (Fytikas, 1977); (ii) A Neogene marine sedimentary sequence of Upper Miocene- Pliocane age (Fyticas, 1977). (iii) A volcanic sequence, the product of eruptive activity which started in the Upper Pliocene and finished in the Upper Quartenary (Fyticas et al., 1976, 1986; Angelier et al., 1977) and produced both pyroclastics and lavas, the composition of which varies from rhyolites to low-Si andesites (Fyticas, 1977; Fyticas et al., 1986). The volcanic activity, which occurred under both subaerial and submarine conditions, started in the western part of the island and migrated to the eastern part in the Lower Quaternary (Fyticas et al., 1986). Subsequent alteration of these volcanics gave rise to extensive bentonite deposits in

11 the eastern part and small kaolin deposits mainly in the western part of the island. (iv) Alluvial deposits. (Kilias, 2011).

2.2.1. Metamorphic basement The island of Milos is almost totally built up of volcanic rocks and a Neogenic sedimentary cover. Limited outcrops of metasedimentary metamorphic rocks of Alpine age (pre-volcanic units) occur to the SW, S and SE part of the island (characteristic appearance site: Spathi cape (at Palaiochori) , SE Milos) . Other sites: at the east of Kalamos bay, at Paliorema site, at the west of Mourtorahi site and on the isle Prasonisi at the west Milos (Fytikas, 1977)), consisting the Alpine crystalline basement (Alfieris 2006, Fytikas and Marinelli 1976; Fytikas et al.1986).(According to Fyticas (1977), rocks similar to that of the crystalline basement of Milos where tracted to the neighbor islets of Ios, Folegandros, and .) The pre-volcanic metamorphic basement of Alpine age consists of metasediments, recrystallized in facies of greenschists and blueschists from claystones, sandstones and impure limestones (Durr et al., 1978). Fragments of metamorphic rocks occur abundantly as xenoliths in phreatic products or in surge deposits from the eruptive center of Fyriplaka (Fytikas, 1978, 1986). The metamorphic basement consists of Alpine metasedimentary rocks that include lawsonite-free jadeite eclogites, lawsonite eclogites, glaucophane schists, quartz- muscovite-chlorite and chlorite-amphibole schists (Fytikas and Marinelli 1976; Kornprobst et al.,1979; Fytikas et al. 1986; Alfieris 2006), calcareous schists with intercalated lavas and calcareous rocks, chloritic schists, and quartzites, which were metamorphosed to blueschist (64.2+- 6.5Ma) and greenschist facies (35.2+-1.0 Ma) (Kilias et al. 2000), as well as abundant mesothermal quartz veins (Alfieris, 2006). According to Fytikas (1977) all the schists were subjected under epizonic metamorphosis, which evolved through two chronological phases, one older (64.2+- 6.5Ma), which is characterized by HP/LT conditions and gave the blueschists facies, and a later one (35.2+-1.0 Ma), which is characterized by LP/HT conditions and gave the greenschists facies. Both phases of the metamorphosis that took place after the subduction of the African lithosphere under the Aegean plate are analyzed in previous chapter of this thesis. The blue schists are characterized by the existence of alkali amphibole (glaucophane mostly) of blue colour, which is usually associated with other minerals, as the sodium pyroxene jadeite and lawsonite (Fytikas, 1977). Also, the jadeite and quartz paragenesis suggests metamorphosis under very high pressure and low temperature conditions (Fytikas, 1977). The blue schists of the metamorphic basement are not in abundance, and they are usually replaced by green schists of the later LP/HT faces of the metamorphosis. The green schists include epidote, which is a mineral in

12 abundance, but we can also trace minerals such as albite, chlorite and amphiboles of the tremolite-actinolite series (Fytikas, 1977).

2.2.2. Neogene sedimentary sequence This Neogene sequence represents the oldest post-orogenic autochtonous sediments of the Cycladic area (Upper Miocene (Messinian) - Lower Pliocene) (Fytikas, 1977; Altherr, 1981; Fytikas et al., 1986) and can be described as a two parts sequence. The lowermost part (up to 30 m thick, according to Fytikas (1977) and Alfieris (2006) consists of two basal conglomerate horizons (sub aerial red bed sediments), composed mainly of metamorphic quartz pebbles from the basement, in between of which a gray limestone horizon is situated (Alfieris, 2006). Minor gypsum layers appear as outcrops locally at Provatas area (S. Milos). The upper part consists of a shallow marine carbonate sequence (limestones, only marly (Fyticas, 1977; Alfieris, 2006) and arenaceous ones at its base), slightly folded with a variable thickness up to 150 m. (Alfieris, 2006). The upper limestones at their turn are capped by a light brown sandstone and a sandy dolomite sequence which appears to have formed in an evaporitic environment. The total maximum thickness of the sequence it is considered to be about 185m. (Alfieris, 2006). From the middle part of the Upper Pliocene, the area of Milos was affected by an extensive volcanism, which was more or less continuously active until late Quaternary times (Fyticas et al. 1986)

2.2.3. Volcanic sequence Based on geochronological and field data Fytikas et al. (1986) recognised five volcanic units; modifications though from Stewart and McPhie (2005) suggest four main volcanic units (Alfieris,2006); they will be described proceeding from the lower up to the upper part. 2.2.3.1. Pliocene series The Pliocene series include acid pyroclastics followed by intrusive subvolcanics and flow dome complexes. Phase I The basal pyroclastic series (Middle to Upper Pliocene) The earliest volcanic episode (Middle to Upper Pliocene 3.5-3.0 M.a.) produced a thick (>120m) succession of felsic submarine units which crop out mainly in the SW part of the island (Fytikas et al. 1986), including pyroclastic flows (enriched with pumice), pumice flows, cinerites, tuffs, with locally subordinate intrusive subvolcanic bodies, lavas and/or hyaloclastites at the end of the sequence (Fytikas et al.,1986). Remnants of submarine tuffs are locally embedded within the chaotic flow units. Locally, the sequence ends with basic volcanic products, in particular, basaltic andesites, which

13 produced pillow lavas, pillow breccias and subordinate hyaloclastites (Fytikas et al.,1986). It is considered that this sequence (basal pyroclastic series-BPS of Fytikas et al. 1986; large submarine felsic cryptodome-pumice cone volcanoes of Stewart and McPhie 2005) has been erupted by a series of vents/volcanic centers (Alfieris, 2006; Stewart and McPhie, 2005) which today are presented partly by the sites occupied by the various subvolcanic bodies / dome complexes around the island (fig 10). Subphase I-II Complex of domes and lava flows (Upper Pliocene) Following the submarine eruptive activity, the W part of Milos was affected by a phase of intermediate to acid subvolcanic products in submarine to partially subaerial environment (Upper Pliocene 2.7 M.a.) (Alfieris, 2006), with subaerial volcanism characterized by the emplacement of several Upper Pliocene domes and lava flows which almost completely covered the previous products (Fytikas et al. 1986). The effusive activity was accompanied by explosive episodes which locally produced pyroclastic flows, breccias and nudes ardentes (Fytikas et al.1986). Due to tectonic processes and reassessments, the opening of a quasi E-W directed graben structure has been established and filled by volcanosedimentary products (Alfieris, 2006). The largest domes were emplaced along preferential NNE or NE trends (Fytikas et al.1986) Phase II (Upper Pliocene to Lower-Middle Pleistocene) During Upper Pliocene to Lower-Middle Pleistocene (2.7-1.4 M.a.), a phase of submarine volcanism was taking place, characterized by the emplacement of a complex of lava flows, plugs and domes (Fytikas et al. 1986; Alfieris,2006) and submarine effusion and intrusion of dacitic/andesitic domes/lavas (Stewart and McPhie 2005; Alfieris, 2006), developing along a system mainly of NNE-SSW faults, and subordinately along NW-SE to E-W directions (Fytikas et al. 1986; Alfieris, 2006). Along with the submarine effusive activity there have been locally explosive episodes from the domes themselves, which locally produced bedded pumiceous tuff, cones and/or pyroclastic flows and breccias on submarine and partially on subaerial environment (Alfieris, 2006). 2.2.3.2. Transitional Lower-Middle Pleistocene series According to Alfieris (2006) during the period of Lower-Middle Pleistocene and most probably around 1.4 Ma, due to a combine effect of tectonism and subvolcanic emplacement of the volcanic phase that described above, the deposition of the tuffitic (volcanosedimentary) series along an E - W direction, that initially had taken place at western Milos (subphase I-II), will continue on eastern Milos. Alfieris (2006) suggests that this evidence indicate the existence of a small number of islands at the northern part of the island during this period (corresponding to the various emplaced subvolcanic bodies), while the southern Milos was a continental area.

2.2.3.3. Pleistocene series

14

Since the end of the Pliocene and the beginning of Pleistocene a submarine sequence associated with rhyolitic domes developed in the eastern and northern parts of the island. These Pleistocene series include acid pyroclastics followed by flow dome complexes mainly in subaerial environment (Fytikas et all.1986; Alfieris, 2006). Phase III The pyroclastic series and lava domes (Lower Pleistocene) During Pleistocene (1.4-0.5 Ma) the E-NE parts of the present island were partially covered by shallow sea water (Fytikas et al. 1986, Alfieris, 2006) and in this environment, has been developed an intense pyroclastic activity associated with rhyolitic domes (Pyroclastic series and Lava Domes-PLD (Fytikas et al., 1986) and rhyolitic pumice-cone volcanoes / rhyolitic lavas accompanied by phreatic activity (Stewart and McPhie 2005)). The pyroclastic products mainly outcropp in the N-NE part of the island, between Puntes and Pollonia (small outcrops are also scattered in the area between Adamas and Zephyria) (Fytikas et al. 1986)) and consist of different ash flow tuff units locally intercalated by tuffites, pumice flow deposits, cinerites, and thick diatomite-rich ash deposits (Fytikas et al. 1986; Alfieris, 2006). The same pyroclastic products at the Filakopi area, at the N coast of the island, are partially covered by a formation of hyaloclastic deposits, mainly pillow lavas and fragments of pillows, of andesitic composition (Fytikas et al.1086; Alfieris, 2006) (Philakopi Hyaloclastites formation- PLHF (Fytikas et al. 1986), submarine felsic pumice breccia intruded by compositional similar cryptodomes (Stewart and McPhie 2004, 2005)). The rhyolitic domes are sited mainly in the eastern sector of the island and are often intensely hydrothermally altered and sometimes brecciated. The effusive products of the volcanic activity created the large complex of domes and acid lava flows in the southern and northern middle part of the island (Halepa, Bombarda and Plakes- HALPLA), as well as in the eastern part (Demeneghaki) (Fytikas et al.,1986), the emplacement of which probably ended the volcanic cycle (Fytikas et al. 1986; Alfieris, 2006). According to Alfieris (2006), these domes were developed mainly as extrusive subaerial domes along a N-S lineament. In the NE part of the submarine volcanic activity of andesitic composition was followed by a phase of subaerial activity, where mainly dacitic in composition domes and lava flows (Korakia, Kalogeros and Glaronisia) have been emplaced (Fytikas et al. 1986; Stewart and McPhie 2004,2005). Phase IV The acid complex of Fyriplaka and Trachilas (Upper Pleistocene) The most recent volcanic activity on Milos (Upper Pleistocene) is concentrated in two distinct zones along the eastern coast of Milos gulf, on a precise NW-SE striking structural lineament. They form the acid centers, which are characterized by subaerial explosive eruptions followed by rhyolitic lavas, of Trachilas to the north, and Firiplaka to the south, with ages of 0.38 Ma and 0.14-0.08 Ma respectively (Alfieris, 2006; Fytikas et al. 1986; Τhe rhyolitic complexes of Firiplaka and Trachilas of Stewart and McPhie 2005). This acid in composition activity has been accompanied by widespread

15 phreatic activity (the “green lahar” of Fytikas 1977 and the mass flow deposits of Stewart and McPhie 2005). Both centers show a different evolution: the northern center (Trachilas) consists of two individual structures, while the southern system (Fyriplaka) is formed by a series of cones partially superimposed one on the other. In both centers the final phases of activity are characterized by the effusion of lavas. Explosive activity which originated the main cone of Trachilas began with the deposition of pyroclastic deposits with subordinate pyroclastic flows. These deposits form a wide basal ring, but they can also be tracked on the northern edge of the neighboring dome’s slopes of Plakes. The initial surge deposits consist of blocks and lapilli and they eventually end up to lava flows, which indicates that the first phases of volcanic activity were characterized by interaction between magma and water, and progressively decreased to a merely effusive activity (Fytikas et al.1986). The Fyriplaka complex consists of an oldest part formed by a ring-tuff, remnants of which crop out to the SE of Provatas. A pyroclastic cone covered the basal ring-tuff with blocks and lapilli. The activity of this older volcanic phase probably terminated with the effusion of lavas. An extensive phase of phreatic explosions followed this first activity, originating deposits almost exclusively formed of fragments from the metamorphic basement ("green lahar") (Fytikas, 1977; Fytikas et al. 1986). The presence of a paleosoil covering the phreatic deposits gives evidence for a pause in eruptive activity. A new volcanic phase then began, being characterized by phreatomagmatic activity, from which a wide ring tuff originated, with an internal diameter about 1500 m. The pyroclastics consist mostly of surge deposits, the lowermost part of which include up to 50% of metamorphic material (Fytikas et al. 1986). The subsequent activity occurred on the western margin and in the inner part of this large cone, giving rise to a new ring tuff, which also evolved into a blocky and lapilli cone. The depression that the concretion of these two large cones created, gave a series of 2-3 smaller cones made up of blocks and lapilli. The inner cones are generally associated with lava flows that reached the sea inside the Gulf of Milos. These lava flows are mainly perlitic, with a blocky surface (Fytikas et all.1986). After the evolving and ending of each volcanic cycle with the lava flows and the setting deposits (it is mentioned by Fytikas et al.1986 that the different volcanic cycles often begin as of purely phreatic type, then evolving into phreatomagmatic activity originating surge deposits and, in later phases, fall deposits followed by lava effusion, usually marking the end of individual cycles of activity, a sequence that occurs in both situations), the water-magma interaction towards the volcanic feeding system, through the fractured rocks of the phreatic activity, gave rise to an hydromagmatic type of activity (surge deposits). Further intrusion of magma towards the surface reduced the magma-water interaction, giving rise to the transition to a purely magmatic activity. The hydromagmatic products of both volcanic complexes are more

16 abundant in the older cycles, and it is considered to reflect yet a decrease of water- magma interaction but at the scale of the single system (Fytikas et al.1986). 2.2.3.5. The products of phreatic activity (from Pleistocene to recent times) - The ‘’green lahar’’ formation Numerous craters formed by phreatic explosions are scattered throughout the eastern part of Milos, with their maximum occurrence to the north of the Zephyria plain. The craters rarely exceed 1000 m in diameter, and they often overlap one another. The lithology of the phreatic products reflects the lithology of the bedrock: in the northern occurrences, rock fragments from the pyroclastic units prevail, while in the southeastern sector elements of the metamorphic basement are dominant (Fytikas et al.1986). The so called "green lahar" (Fytikas, 1977) must be included among the products of the phreatic activity (Fytikas, 1977; Fytikas et al.1986; Alfieris 2006). The formation is regularly associated with phreatic depressions and tends to thicken close to explosion craters. Further evidence of its origin is given by the early explosive products of the Fyriplaka system, which mostly consists of metamorphic rocks, similar in lithology and grain size to those forming the "green lahar" (Fytikas et al.1986). Fytikas (1977) explains that the lahar was formed after the mudflows of ash that wiped the volcanic and phreatic products around the craters alongside with rocks and products of erosion of the bedrock (the lahar was named after the green schists of the metamorphic basement), which was exposed after a former activation of a faults system (Fytikas, 1977). This activity is of Quaternary age, since the phreatic products overlie Pliocene-Lower Pleistocene volcanic products and are covered by very recent aeolian sediments. In the southern part of the island, however, some layers of phreatic origin are overlain by the younger pyroclastic products of the Fyriplaka volcanic system and appear locally interbedded between the older and the younger cycles of the same eruptive complex. It can be assumed that the wide-spread phase of phreatic activity in the island of Milos is of Pleistocene age, probably younger than 0.2 m.y. (Fytikas et al.1986). The present formation expands scattered in a large area in the eastern half of the island. It is observed among the main villages of the island, at the N and E coasts, at the domes of Kalogeros and Korakia, at the SE coasts and at the depression of Provatas, according to Fytikas (1977). The formation is overlaid by the alluvian deposits.

2.3. Alluvial sequence According to Alfieris (2006), recent alluvial sediments that are found in the Zefiria 0graben indicatethe presence of a lagoon-like environment. The sediments consist of clay, sand and gypsum, and their total thickness is about 80 m (Fytikas et al.,1976b)

17

Figure 5: Simplified geological map of Milos island showing the distribution of the main volcanic phases and metamorphic, sedimentary units as well as the geochronological data so far available. Modified after Fytikas et al.(1986).

18

2.4. Hydrothermal activity and mineralization Although the last recorded volcanism was approximately 100000 years ago, Milos is still an active geothermal field. This is presently expressed by widespread surface manifestations in the form of fumaroles, hot springs, hot grounds, and submarine gas escapes. The tectonics of the area control the hydrothermal activity. The host rocks of the hydrothermal activity are, as mentioned above, calc-alkaline, predominantly andesites and lesser dacites (Kilias, 2001, 2011). Milos has an active high-enthalpy geothermal system that vents in the shallow (~100 m) marine and terrestrial environments (Valsami- Jones et al. 2005; Kilias, 2011), and the island as a whole has been hydrothermally active for around 1.5 Ma (Kilias, 2011). Recent research has identified a new metallogenic environment where mineralization and geothermal activity is associated with emergent volcanoes. Metalliferous mineralization occurs mainly in western Milos, being controlled by E-W trending horst structures. A high sulfidation and a low sulfidation epithermal system both occur in western Milos. The older Basal Pyroclastic Series at Profitis Ilias hosts the gold mineralization (Kilias, 2001). In present days, the hydrothermal acivity is concentrated in the E-SE part of the island (Kilias,2001 (14); Alfieris, 2006).

2.4.1. Mineral Alterations and industrial minerals Milos island hosts a great number of industrial mineral deposits. The volcanic activity on the island has left over an active geothermal field, where hot hydrothermal fluids cause the alteration of the volcanic rocks for the last 1.5 M.a., producing an important number of mineral ores and deposits. though, not all of the industrial minerals are products of hydrothermal activity. Perlite consists the most significant example. Perlite is natural volcanic glass and is formed by the sudden cooling of submarine lava on the sea floor, trapping large amounts of water into its mass. The abrupt, controlled rise of temperature causes a white mass of minuscule glass bubbles to be formed. Perlite melts and expands due to the entrapped water’s evaporation. The mineral thus acquires special properties as heat and sound insulation material, whereas at the same time it becomes extremely porous. The minerals’ main property of expanding when heated in temperatures 800o–950o C is due particularly to the evaporation of the trapped water. There are two perlite deposits on Milos island that date back to two distinctive geological periods. The older deposit is the Halepa perlite. The younger Milos perlite deposits were the result of the submarine volcanoes of Fyriplaka on the south, and Trahilas on the north. They are chemically unique, different from other Greek perlites and of superior quality to the perlite at Halepa, due to their highly glassy structure and large water amounts that make them particularly light when expanded.

19

Pozzolan, which derives from volcanic ash and pumices, is also found on Milos island while large quantities also exist on Kimolos. It is a siliceous and aluminal material Pozzolans are volcanic earths, specifically tuffs, usually of trachitic – andesitic origin, and they use to alerted to zeolites by interaction with alkaline waters. Bentonites are plastic clays derived from the in situ hydrothermal alteration of volcanic ash by the mixture of seawater with hot alkaline fluids and consist essentially from smectite. In Milos, bentonite deposits occure in the NE and E part of the island and are dated in the Upper Pleistocene. These deposits formed at the expense of pyroclastic rocks, mainly pyroclastic flows, and to lesser extend of lavas which varies in composition from andesitic to ryodacitic-ryolitic (Christidis and Duncham, 1992), in a submarine environment, and were affected by at least two hydrothermal events, which caused, at places, extensive illitization. Post-formation hydrothermal alteration may also have caused kaolinization and alunizitation apart from illitization. Also, authigenic minerals like zeolites, and/or K-felspars are present in some bentonite deposits (Christidis, 2000). Kaolin is a mixture of hydrous aluminium silisic oxides and it includes the minerals kaolinite, anoxite, halloysite and allophane. These minerals have the same chemical composition, but different optical properties and a different internal structure of their crystals, as shown under x-ray examination. Kaolinite derives due to hydrothermal alteration of aluminium silicate minerals like feldspars, via acid mixture fluids. Manganese ores derived from the hydrothermal avtivity in NW Milos. Old manganese mines exist at Cape Vani, to the North-Western edge of Milos, where also barite deposits can be seen. The rich in metals hydrothermal fluids found their way to the surface of the Volcanosedimentary rocks of the area (a volcanoclastic sandstone hosts the manganese deposit) through fractures up towards the surface, depositing along their way the metals, including this of manganese. Barite has similar origin: as the hot, metal-rich fluid moved up the open fractures, it met cold seawater. This mixing resulted in the precipitation of barium sulphate (barite) in the fractures. Such deposits can also be seen along old fractures at Triades and Komia in the Voudia area, and in former times, were mined for barite. Former studies of the silver rich baryte of Trides-Galana area (W. Milos), and through their chemical analyses, indicate that most probable has considered the presence of silver chlorine minerals because of the high concentrations of these two elements (Alfieris,2006). Baryte is a heavy, inert and stable mineral. “Barytine” in Greece, is the name given to baryte, with a low content of silver. Important ores have been found also in Kimolos. Finally, sulphur was exploited on the south coast of the island (Pliorema location). Its formation is product of the interaction between extremely acid hydrothermal fluids and the original volcanic rocks. The ascending fluids, rich in sulphuric acid, leach the host rocks, filling them with native sulphur.

20

3. Kimolos Island Kimolos island belongs to the small archipelago of Milos, that also includes the islands of Polyegos and Antimilos, and was formed during the Pliocene, under the almost same conditions that Milos was formed. 3.1 Geology

Kimolos is, as Milos island, nearly completely built up by volcanic rocks (most of which are pyroclastic (Fytikas and Vougioukalakis, 1993)); only few outcrops of metamorphic rocks and Neogene sediments are found (and only on this island), while at the same island little occurrences of granitic s.l. fragments into the pyroclastic formations testify the presence of a granitic s.l. body emplaced at shallow depths (Alfieris, 2006). In Kimolos the pro-volcanic sequences outcrops in the W of the island (Athinias Avlaki area) in two small appearances: the south one constitutes the Alpine crystallised basement and comprises of mica-quartz schists, with a thickness of 20m approximately. The rocks are tectonised, effected by the volcanic activity and the intrusions of lava (Fytikas and Vougioukalakis, 1993). The northern occurrence consists of neogenic sentiments with breccias and sandstones, with a thikness of 40m approximately. The rocks are also tectonised by the intrusion of lavas (Fytikas and Vougioukalakis, 1993). During the Lower Pliocene occurred the first intrusions of magma in the area, so a small plutonic body has been crystallised and briefly outcrops at the central part of the island. This plutonite is a typical granite, rich in plagioclase, orthoclase, quartz, biotite, hornblende, ferrum and titanite oxides (secondary minerals obderved: apatite, titanite, zircon, epidote) (Fytikas and Vougioukalakis, 1993). The volcanic activity in the Kimolos and Polyegos area (in comparison with the volcanic activity of Milos) was manifested during the Upper Pliocene and Lower Pleistocene, ranging in age between 3,5 and 0,9 Ma. Two cycles of volcanic activity are distinguished, which gave the deposits of the two ignimbrites of Kastro and Prassa, distinguishing two cycles of volcanic activity that characterized two big explosive episodes (Fytikas and Vougioukalakis, 1993). The first cycle, dating approximately between 3,5 and 2,0 Ma, comprises the lower lavas of Kimolos, the Kastro ignimbrite and the andesitic -dacitic lavas of Kimolos. The Kastro ignimbrites comprises of a complex of hydrothermally altered volcano glass flows, and it is the dominant formation of the island, exceeding the 400 m.The outcrops of the dome in the NW coasts of the island is being highly alterated, though the parental rock (which is considered to be a rhyolite (Fytikas and Vougioukalakis, 1993) is rich in feldspars, while the altered rock is rich in zeolithes, quartz and argillic minerals. The second cycle, between 2,0 and 0,9 Ma, comprises the Prassa ignimbrite, the domes of the neighbor islet of Polyegos, the andesitic pyroclastics and the Geronikola lavas οf Kimolos, the rhyolitic pyroclastics of Psathi and Mersini and, finally, the domes of Psathi, Xaplovouni and Mersini. The ignimbrite of Prassa, which

21 occurs at the eastern part of Kimolos and NW part of , comprises of pyroclastic flows of maximum thickness of 200 m. After the extrusion of Psathi, Xaplovouni and Mersini’s domes the volcanic activity was exhausted in the area of Kimolos, and Polyegos as well (Fytikas and Vougioukalakis, 1993). Α NB-SW trending tectonic lineament, along which the volcanic centres of the Milos island group were arranged, seems to continue to be active in the same tectonic trend was used as a path for the hydrothermal fluids that deposited the Μη, Pb, Ba ores. The hot springs of Kimolos are also aligned in a NE-SW trending direction. Α geothermal field wlth a probable fluid temperature between 80 and 120 oC is supposed to exist in the Kimolos area while the existence of deeper reservoirs with higher temperatures is not excluded.

Figure 6: Kimolos island modified after Fytikas et al.1986

22

4.Black Magnetite sands (Heavy mineral sands) The genuine interest of this thesis was the study of heavy mineral beach sands, and specifically black sands rich in magnetite. In general, heavy mineral sands are a class of placer ore deposits formed most usually in beach environments by concentration due to the specific gravity of the mineral grains, which are an important source of zirconium, titanium, thorium, tungsten, rare- earth elements, diamonds, garnets, and occasionally precious metals or gemstones. It is equally likely that some concentrations of heavy minerals (aside from the usual gold placers) exist within streambeds, but most are of a low grade and are relatively small. Black sands typically consist of heavy minerals such as magnetite, ilmenite, titanomagnetite, monazite, hematite. Only the magnetite (and also titanomagnetite, which exhibits much weaker magnetic properties because it is only paramagnetic rather than ferromagnetic) will be attracted to the magnet, while the hematite, titanohematite, monazite and any other non-magnetite components will largely be left behind. Dense dark sands that often include high levels of magnetite sand are commonplace in many streams, creeks, rivers and arroyos across the world. Magnetite is sometimes found in large quantities in beach sand. The magnetite particles, due to their high mass and high density, are being collecting in the same spots in streams and arroyos, carried to the shores of oceans and bays via rivers from erosion and is concentrated via wave action and currents. So, the black sand that is often found in bands and layers near or above the high-water line (and sometimes in hollows or cavitations below the high- water mark) along a number of ocean (or bay) shores and on the beaches of some lakes is often primarily magnetite black sand. (In general, the denser heavy mineral grains (magnetite, ilmenite, monazite, rutile, zircon etc.) are finally concentrated in low-energy environments, while lighter heavy mineral grains (tourmaline, kynite, etc.) can concentrate in environments of relatively higher energy (M.A. Mange and H. Maurer, 1992) Magnetite is an iron oxide with a formula of Fe3O4, and is a member of the spinel group which has the standard formula A(B)2O4. The A and B represent different metal ions that occupy specific sites in the crystal structure. In the case of magnetite, the A metal is Fe +2 and the B metal is Fe +3 (FeFe2O4). Magnetite is a heavy and dense material, with a specific gravity of 5.18 when it is pure; this is 3 to 5 times denser than most silica-rich light-colored sands, and about 5 to 6 times denser than most soils. The color of magnetite sand ranges from gray to dark gray or dark tan to almost black in color. Magnetite is usually strongly magnetic (i.e., ferromagnetic) and exhibits high magnetic susceptibility, so it responds strongly to a magnet, and it usually exhibits strong ferromagnetic properties, although a few forms found in nature may be more paramagnetic than ferromagnetic. Magnetite is widely found in a variety of natural settings as a commonplace mineral; it most commonly occurs in nature in a granular form as a sand or grit. While

23 magnetite may sometimes occur in nature in the form in larger chunks or rocks, those chunks are almost always aggregates or conglomerates of grains held together only loosely, and these chunks, with a bit of pressure, will often crumble into granular form. Magnetite is one of the most important ores of iron (magnetic iron ore) and is a common constituent of igneous (basically mafic) and metamorphic rocks. It is found as in inclusion in granite too, where it will usually be quite tightly bound in the rocky matrix. Magnetite sand tends to be more abundant in areas where rock formations and sands are relatively low in silica and are dark-colored and magnetite tends to be far less plentiful in lighter-colored, silica-rich rocks and sands. Many regions which contain lots of surface rocks or formations which are dark-colored and relatively low in silica are high in magnetite-rich black sands, as are many volcanic regions, including volcanic sand deposits in mountains and deserts and the flanks of volcanoes, but also in areas rich in alluvial deposits or the sites of ancient alluvial fans or alluvial plains.

5. Sampling and methodology 5.1. Sampling The sampling of black magnetite sands from marine costs of the islands of Milos and Kimolos, which are the subject of the present thesis, was carried out by the author of the thesis, post-graduate student Myrto Trapatseli, during the period between 4-8 of May 2016. The samples are eight (8) in number: seven (7) samples from seven different sites of Milos island (Paleochori, Fyriplaka, Adamas, Agathia, Triades, Empourios, Rivari) and one (1) sample from Kimolos island (Ellinika). The choice of the sites has been determined based on previous observed quantities of black sands. As it has mentioned above, the purpose of this thesis is the study of the mineral and chemical analysis of black magnetite sands that have been taken from different sites of Milos and Kimolos islands coastline, in order to reach in results that correlate the origin of these materials with specific lithologies if possible.

24

1. 2. 3.

4. 5. 6.

Images 1-6: Black beach sands from the locations: 1-2: Triades, 3-4: Adamantas, 5-6: Fyriplaka

The samples consist of black magnetite sand (which is the subject of interest of this thesis) and silica sand with feldspars, amphiboles, phyllosilicates and other minerals, as it was impossible for a good enough separation of the grains to be achieved. Of course, the most representative samples were gathered, in order to reach the most accurate results. The samples consist of typical granular sand material, with grain size 2mm-1/16mm. Below is presented a map with the sites where the sampling took place:

25

Map 1: Sampling sites An extra observation on the field, is that black magnetite sand was traced also in Sarakiniko, in the northern part of the island. Samples of this material has also been gathered for mineral and chemical analysis in the future.

5.2. Macroscopical examination Depending on the amount of black sand in the total of the samples, each demonstrated different colour depending on their mineralogical composition. It is obvious that the richer in magnetite and other dark coloured minerals the sample is, the darker the sand appears to be. The darker sample, which was almost black in colour, was that one gathered from Rivari (Ri) spot, where the ferrous-ferric minerals were in abundance. Also dark, but more greyish, were the samples from Triades (Tr), Agathia (Ag), and Empourios (Em) sites. The sample from Adamantas (Ky-Ad) was also darkish and browner, and the sample from Fyriplaka (Fi) was light grey because of the fine silicate sand fraction. The sample from Palaiochori (Pa) was light brown-greyish, but it had no ferrous-ferric minerals, except from a few grains that were hardly

26

Images a1-h2: Samples material from: a1,a2: Fyriplaka, b1,b2: Palaiochori, c1,c2: Empourios, d1,d2: Rivari, e1,e2: Triades, f1,f2:Agathia, g1,g2: Adamantas, h1,h2: Ellinika-Kimolos

27 gathered with the use of a magnete. In a similar way, the sample from Ellinika (El-Ki) site, οn Kimolos island, was light coloured, as the quantity of the ferrous-ferric minerals was quite small. As it concerns the grain size, the samples consist of typical granular sand material, with grain size 2mm-1/16mm. The material from Fyriplaka, Triades, Agathia, Rivari, and Empourios was quite fine grained, meanwhile the material from Adamantas, Palaichori and Ellinika demonstrated a variety in grain size. After the examination for ferrous-ferric minerals with a use of a magnet, the results confirmed their existence. The samples Ri, Em, Tr, Ag, Ky-Ad and Fi were extremely rich in magnetite (or other ferrous-ferric mineral with ferro/para-magnetic properties), especially the sample of Ri, while the samples El-Ki and Pa were poor in iron minerals.

5.2. Analytical methods 5.2.1. XRD analysis technique X-ray diffraction (XRD) is an analytical technique primarily used for phase identification of a crystalline material and can provide information on unit cell dimensions. The analyzed material is finely ground, homogenized, and average bulk composition is determined. X-ray diffraction is a common technique for the study of crystal structures and atomic spacing. X-ray diffraction is based on constructive interference of monochromatic X-rays and a crystalline sample. These X-rays are generated by a cathode ray tube, filtered to produce monochromatic radiation, collimated to concentrate, and directed toward the sample. The interaction of the incident rays with the sample produces constructive interference (and a diffracted ray) when conditions satisfy Bragg's Law, which relates the wavelength of electromagnetic radiation to the diffraction angle and the lattice spacing in a crystalline sample. (nλ=2d sin θ, θ are certain angles of incidence, d is the distance between atomic layers in a crystal, λ is the wavelength of the incident X-ray beam, n is an interger). These diffracted X-rays are then detected, processed and counted. By scanning the sample through a range of 2θangles, all possible diffraction directions of the lattice should be attained due to the random orientation of the powdered material. Conversion of the diffraction peaks to d-spacings allows identification of the mineral because each mineral has a set of unique d-spacings. Typically, this is achieved by comparison of d-spacings with standard reference patterns. All diffraction methods are based on generation of X-rays in an X-ray tube. These X- rays are directed at the sample, and the diffracted rays are collected. A key component of all diffraction is the angle between the incident and diffracted rays. Powder and single crystal diffraction vary in instrumentation beyond this.

28

X-ray diffractometers consist of three basic elements: an X-ray tube, a sample holder, and an X-ray detector. X-rays are generated in a cathode ray tube by heating a filament to produce electrons, accelerating the electrons toward a target by applying a voltage, and bombarding the target material with electrons. When electrons have sufficient energy to dislodge inner shell electrons of the target material, characteristic X-ray spectra are produced.These spectra consist of several components, the most common being Kα and Kβ. Kα consists, in part, of Kα1 and Kα2. Kα1 has a slightly shorter wavelength and twice the intensity as Kα2. The specific wavelengths are characteristic of the target material (Cu, Fe, Mo, Cr). Filtering, by foils or crystal monochrometers, is required to produce monochromatic X-rays needed for diffraction. Kα1and Kα2 are sufficiently close in wavelength such that a weighted average of the two is used. Copper is the most common target material for single-crystal diffraction, with CuKα radiation = 1.5418Å. These X-rays are collimated and directed onto the sample. As the sample and detector are rotated, the intensity of the reflected X-rays is recorded. When the geometry of the incident X-rays impinging the sample satisfies the Bragg Equation, constructive interference occurs and a peak in intensity occurs. A detector records and processes this X-ray signal and converts the signal to a count rate which is then output to a device such as a printer or computer monitor. The geometry of an X-ray diffractometer is such that the sample rotates in the path of the collimated X- ray beam at an angle θ while the X-ray detector is mounted on an arm to collect the diffracted X-rays and rotates at an angle of 2θ. The instrument used to maintain the angle and rotate the sample is termed a goniometer. For typical powder patterns, data is collected at 2θ from ~5° to 70°, angles that are preset in the X-ray scan.

5.2.2. SEM/EDS (Energy-dispersive X-ray spectroscopy) analysis technique Energy Dispersive X-Ray Spectroscopy (EDS or EDX) is a chemical microanalysis technique used in conjunction with scanning electron microscopy (SEM). The EDS technique detects x-rays emitted from the sample during bombardment by an

29 electron beam to characterize the elemental composition of the analyzed volume. Features or phases as small as 1 µm or less can be analyzed. When the sample is bombarded by the SEM's electron beam, electrons are ejected from the atoms comprising the sample's surface. The resulting electron vacancies are filled by electrons from a higher state, and an x-ray is emitted to balance the energy difference between the two electrons' states. The x-ray energy is characteristic of the element from which it was emitted. The EDS x-ray detector measures the relative abundance of emitted x-rays versus their energy. The detector is typically a lithium-drifted silicon, solid-state device. When an incident x-ray strikes the detector, it creates a charge pulse that is proportional to the energy of the x-ray. The charge pulse is converted to a voltage pulse (which remains proportional to the x-ray energy) by a charge-sensitive preamplifier. The signal is then sent to a multichannel analyzer where the pulses are sorted by voltage. The energy, as determined from the voltage measurement, for each incident x-ray is sent to a computer for display and further data evaluation. The spectrum of x-ray energy versus counts is evaluated to determine the elemental composition of the sampled volume. Interaction of an electron beam with a sample target produces a variety of emissions, including x-rays. An energy-dispersive (EDS) detector is used to separate the characteristic x-rays of different elements into an energy spectrum, and EDS system software is used to analyze the energy spectrum in order to determine the abundance of specific elements. EDS can be used to find the chemical composition of materials down to a spot size of a few microns, and to create element composition maps over a much broader raster area. Together, these capabilities provide fundamental compositional information for a wide variety of materials. Scanning electron microscopy (SEM) is a method for high-resolution imaging of surfaces. The SEM uses electrons for imaging, much as a light microscope uses visible light. The advantages of SEM over light microscopy include much higher magnification (>100,000X) and greater depth of field up to 100 times that of light microscopy. Qualitative and quantitative chemical analysis information is also obtained using an energy dispersive x-ray spectrometer (EDS) with the SEM. The SEM generates a beam of incident electrons in an electron column above the sample chamber. The electrons are produced by a thermal emission source, such as a heated tungsten filament, or by a field emission cathode. The energy of the incident electrons can be as low as 100 eV or as high as 30 keV depending on the evaluation objectives. The electrons are focused into a small beam by a series of electromagnetic lenses in the SEM column. Scanning coils near the end of the column direct and position the focused beam onto the sample surface. The electron beam is scanned in a raster pattern over the surface for imaging. The beam can also be focused at a single point or scanned along a line for x-ray analysis. The beam can be focused to a final probe diameter as small as about 10 Å.

30 the incident electron beam is scanned in a raster pattern across the sample's surface. The emitted electrons are detected for each position in the scanned area by an electron detector. The intensity of the emitted electron signal is displayed as brightness on a display monitor and/or in a digital image file. By synchronizing the position in the image scan to that of the scan of the incident electron beam, the display represents the morphology of the sample surface area. Magnification of the image is the ratio of the image display size to the sample area scanned by the electron beam. The incident electrons cause electrons to be emitted from the sample due to elastic and inelastic scattering events within the sample’s surface and near-surface material. High-energy electrons that are ejected by an elastic collision of an incident electron, typically with a sample atom’s nucleus, are referred to as backscattered electrons. The energy of backscattered electrons will be comparable to that of the incident electrons. Emitted lower-energy electrons resulting from inelastic scattering are called secondary electrons. Secondary electrons can be formed by collisions with the nucleus where substantial energy loss occurs or by the ejection of loosely bound electrons from the sample atoms. The energy of secondary electrons is typically 50 eV or less. Two electron detector types are predominantly used for SEM imaging. Scintillator type detectors (Everhart-Thornley) are used for secondary electron imaging. This detector is charged with a positive voltage to attract electrons to the detector for improved signal to noise ratio. Detectors for backscattered electrons can be scintillator types or a solid-state detector. The SEM column and sample chamber are at a moderate vacuum to allow the electrons to travel freely from the electron beam source to the sample and then to the detectors. High-resolution imaging is done with the chamber at higher vacuum, typically from 10-5 to 10-7 Torr. Imaging of nonconductive, volatile, and vacuum- sensitive samples can be performed at higher pressures.

6. Results

6.1. XRD analysis and results The XRD analysis of the 8 sand samples took place in the laboratory of Greek institute of Geology and Mineral Exploration. Initially, the sand samples were prepared into slides suitable for the X-ray Diffractioner. Subsequently, the mineral analysis was carried out by the diffractioner, and the results were evaluated on the EVA program. The analysis gave the following:

31

70

60

50

40

00-003-0793 (D) - Ilmenite - - FeTiO3 Y: 2.96 % - d x by: 1. - WL: 1.5406 - 0 -

00-002-0028 (D) - Chlorite - (Mg,Fe)5(Al,Si)5O10(OH)8 - Y: 1.37 % - d x by: 1. - WL: 1.5406 - 0 -

00-003-0024 (D) - Biotite - (K,H)2(Mg,Fe)2(Al,Fe)2(SiO4)3 - Y: 2.39 % - d x by: 1. - WL: 1.5406 - 0 -

00-005-0448 (D) - Barite - BaSO4 - Y: 3.02 % - d x by: 1. - WL: 1.5406 - 0 - I/Ic 2.6 PDF -

00-019-0002 (I) - Orthoclase, barian - (K,Ba,Na)(Si,Al)4O8 - Y: 3.92 % - d x by: 1. - WL: 1.5406 - 0 - I/Ic

2-Theta - Scale - 2-Theta

30

20

10

00-003-0058 (D) - Kaolinite - Al2Si2O5(OH)4 - Y: 1.97 % - d x by: 1. - WL: 1.5406 - 0 -

00-007-0322 (D) - Magnetite - FeO·Fe2O3 00-007-0322 - (D) - Y: Magnetite8.04 % - - d FeO·Fe2O3 x by: 1. - WL: 1.5406 - 0 -

00-034-0517 (D) - Dolomite, ferroan - Ca(Mg,Fe)(CO3)2 - Y: 19.53 % - d x by: 1. - WL: 1.5406 - 0 -

00-005-0586 (*) - Calcite, syn - - CaCO3 Y: 13.12 % - d x by: 1. - WL: 1.5406 - 0 - I/Ic 2. PDF -

00-010-0393 (*) - Albite, disordered - Na(Si3Al)O8 - Y: 12.34 % - d x by: 1. - WL: 1.5406 - 0 - I/Ic User 1.

00-046-1045 (*) - syn Quartz, - SiO2 - Y: 94.47 % - d x by: 1. - WL: 1.5406 - 0 - I/Ic 3.4 PDF -

Operations: Import

File: XRD 2014_ _C0518 FI.raw - locked 2Th/Th 2014_ Type: _C0518 - File: FI.raw Start:XRD 3.010 ° - End: 69.990 ° - Step: 0.020 ° -

4

0

100

200

300

400

500

600

700

800

900

1000

1100

1200

1300

1400

1500

1600

1700

1800 1900

Lin (Counts) Lin Figure 7: XRD analysis data from the sample of Fyriplaka beach; sample code: FI

32

70

60

50

40

00-002-0028 (D) - Chlorite - (Mg,Fe)5(Al,Si)5O10(OH)8 - Y: 0.85 % - d x by: 1. - WL: 1.5406 - 0 -

2-Theta - Scale - 2-Theta

30

20

10

00-010-0359 (D) - Andesine, low - 0.62NaAlSi2O8·0.38CaAl2Si2O8 - Y: 00-010-0359 0.60(D) - - 0.62NaAlSi2O8·0.38CaAl2Si2O8 % Andesine,- d x by: low 1. - WL: 1.5406

00-024-0072 (D) - Hematite - Fe2O3 - Y: 0.99 % - d x by: 1. - WL: 1.5406 - 0 - I/Ic User 1. -

00-007-0032 (D) - Muscovite 2M1, syn - KAl2Si3AlO10(OH)2 - Y: 3.70 % - d x by: 1. - WL: 1.5406 - 0 -

00-003-0058 (D) - Kaolinite - Al2Si2O5(OH)4 - Y: 1.84 % - d x by: 1. - WL: 1.5406 - 0 -

00-041-1480 (I) - Albite, calcian, ordered - (Na,Ca)Al(Si,Al)3O8 - Y: 1.01 % - d x by: 1. - WL: 1.5406 - 0 -

00-033-1161 (D) - syn Quartz, - SiO2 - Y: 78.80 % - d x by: 1. - WL: 1.5406 - 0 - I/Ic 3.6PDF -

Operations: Import

File: XRD 2014_C0521 PA.raw - locked2014_C0521 2Th/Th Type: - File: PA.raw Start:XRD 3.010 ° - End: 69.990 ° - Step: 0.020 ° -

4

0

1000

2000

3000

4000

5000

6000

7000 8000

Lin (Counts) Lin Figure 8: XRD analysis data from the sample of Palaiochori beach; sample code: PA

33

70

60

50

40

00-012-0243 (D) - Clinochlore - - Mg-Fe-Fe-Al-Si-O-OH Y: 4.74 % - d x by: 1. - WL: 1.5406 - Orthorhombi

00-003-0778 (D) - Ilmenite - - FeTiO3 Y: 10.07 % - d x by: 1. - WL: 1.5406 - Rhombo.R.axes -

00-003-0788 (D) - Magnesite - MgCO3 - Y: 50.00 % - d x by: 1. - WL: 1.5406 - Rhombo.R.axes -

00-020-0481 (I) - Magnesiohornblende - (Ca,Na)2.26(Mg,Fe,Al)5.15(Si,Al)8O22(OH)2 - Y: 50.00 % - d x

00-026-0919 (C) - Halite, potassian, syn - K0.4Na0.6Cl - Y: 50.00 % - d x by: 1. - WL: 1.5406 - Cubic -

2-Theta - Scale - 2-Theta

30

20

10

00-020-0572 (D) - Albite, disordered - NaAlSi3O8 - Y: 50.00 % - d x by: 1. - WL: 1.5406 - Triclinic -

00-024-0072 (D) - Hematite - Fe2O3 - Y: 6.47 % - d x by: 1. - WL: 1.5406 - 0 - I/Ic User 1. -

00-019-0002 (I) - Orthoclase, barian - (K,Ba,Na)(Si,Al)4O8 - Y: 2.85 % - d x by: 1. - WL: 1.5406 - 0 - I/Ic

00-002-0623 (D) - Calcite - CaCO3/CaO·CO2 - Y: 00-002-0623 2.39(D) - % Calcite- d - x by: CaCO3/CaO·CO2 1. - WL: 1.5406 - 0 -

00-001-1111 (D) - Magnetite - Fe3O4 - Y: 11.43 % - d x by: 1. - WL: 1.5406 - 0 -

00-046-1045 (*) - syn Quartz, - SiO2 - Y: 15.18 % - d x by: 1. - WL: 1.5406 - 0 - I/Ic 3.4 PDF -

Operations: Import

File: XRD 2014_C0522 EM.raw - locked 2Th/Th Type: 2014_C0522 EM.raw - Start: File: XRD 3.010 ° - End: 69.990 ° - Step: 0.020 ° - S

4

0

100

200

300

400

500

600

700

800

900

1000

1100

1200

1300

1400

1500 1600

Lin (Counts) Lin Figure 9: XRD analysis data from the sample of Empourios beach; sample code: EM.

34

70

60

50

40

00-045-1371 (I) - Magnesiohornblende, ferroan - Ca2(Mg,Fe+2)4Al(Si7Al)O22(OH,F)2 - Y: 14.56 % - d x

00-003-0778 (D) - Ilmenite - - FeTiO3 Y: 21.47 % - d x by: 1. - WL: 1.5406 - Rhombo.R.axes -

00-041-1480 (I) - Albite, calcian, ordered - (Na,Ca)Al(Si,Al)3O8 - Y: 50.00 % - d x by: 1. - WL: 1.5406 - Tr

00-025-1376 (D) - Magnetite - (Fe,Mg)(Al,Cr,Fe,Ti)2O4 - Y: 50.00 % - d x by: 1. - WL: 1.5406 - Cubic -

2-Theta - Scale - 2-Theta

30

20

10

00-003-0788 (D) - Magnesite - MgCO3 - Y: 11.54 % - d x by: 1. - WL: 1.5406 - Rhombo.R.axes -

00-010-0359 (D) - Andesine, low - 0.62NaAlSi2O8·0.38CaAl2Si2O8 - Y: 00-010-0359 6.22(D) - - 0.62NaAlSi2O8·0.38CaAl2Si2O8 % Andesine,- d x by: low 1. - WL: 1.5406

00-041-1366 (I) - Actinolite - Ca2(Mg,Fe)5Si8O22(OH)2 - Y: 20.38 % - d x by: 1. - WL: 1.5406 - Monoclin

00-003-0427 (D) - Quartz SiO2 - Y: 3.01 % - d x by: 1. - WL: 1.5406 - - Hexagonal

00-024-0072 (D) - Hematite - Fe2O3 - Y: 21.40 % - d x by: 1. - WL: 1.5406 - Rhombo.R.axes - I/Ic User 1

00-003-0863 (D) - Magnetite - Fe3O4 - Y: 44.66 % - d x by: 1. - WL: 1.5406 -

Operations: Import

File: XRD 2014_C0520 RI.raw - locked 2Th/Th Type: 2014_C0520 - RI.raw Start: File: XRD 3.010 ° - End: 69.990 ° - Step: 0.020 ° - St

4

0

100

200

300

400

500

600

700 800

Lin (Counts) Lin Figure 10: XRD analysis data from the sample of Rivari beach; sample code: RI.

35

70

60

50

40

00-003-0789 (D) - Ilmenite - (Fe,Mg)TiO3 - Y: 5.24 % - d x by: 1. - WL: 1.5406 - Rhombo.R.axes -

00-005-0622 (D) - Dolomite - CaMg(CO3)2/CaO·MgO·2CO2 - Y: 2.75 % - 00-005-0622 d (D) - x by: 1.Dolomite - - WL:CaMg(CO3)2/CaO·MgO·2CO2 1.5406 - 0 -

00-003-0788 (D) - Magnesite - MgCO3 - Y: 4.63 % - d x by: 1. - WL: 1.5406 - 0 -

00-024-0020 (D) - Barite - BaSO4 - Y: 5.01 % - d x by: 1. - WL: 1.5406 - 0 -

00-046-1415 (*) - Barite, plumbian - (Ba,Pb)SO4 - Y: 7.69 % - d x by: 1. - WL: 1.5406 - 0 -

2-Theta - Scale - 2-Theta

30

20

10

00-033-0664 (*) - Hematite, syn - Fe2O3 - Y: 3.78 % - d x by: 1. - WL: 1.5406 - 0 - I/Ic 2.4 PDF -

00-019-0629 (*) - Magnetite, syn - - FeFe2O4 Y: 7.10 % - d x by: 1. - WL: 1.5406 - 0 - I/Ic 4.9 PDF -

00-046-1045 (*) - syn Quartz, - SiO2 - Y: 30.53 % - d x by: 1. - WL: 1.5406 - 0 - I/Ic 3.4 PDF -

00-045-1371 (I) - Magnesiohornblende, ferroan - Ca2(Mg,Fe+2)4Al(Si7Al)O22(OH,F)2 - Y: 19.24 % - d x

00-041-1486 (*) - Anorthite, ordered - CaAl2Si2O8 - Y: 32.97 % - d x by: 1. - WL: 1.5406 - 0 - I/Ic 0. PDF

00-020-0572 (D) - Albite, disordered - NaAlSi3O8 - Y: 21.44 % - d x by: 1. - WL: 1.5406 - 0 -

Operations: Import

File: XRD 2014_C0523 TR.raw - locked2Th/Th Type: - Start: 2014_C0523 TR.raw File: 3.010XRD ° - End: 69.990 ° - Step: 0.020 ° - St

4

0

100

200

300

400

500

600

700

800

900

1000

1100

1200

1300

1400

1500

1600

1700

1800

1900

2000 2100

Lin (Counts) Lin Figure 11: XRD analysis data from the sample of Triades beach; sample code: TR

36

70

60

50

40

00-003-0781 (D) - Ilmenite - - FeTiO3 Y: 4.05 % - d x by: 1. - WL: 1.5406 - 0 -

00-010-0393 (*) - Albite, disordered - Na(Si3Al)O8 - Y: 68.75 % - d x by: 1. - WL: 1.5406 - Triclinic -

00-020-0481 (I) - Magnesiohornblende - (Ca,Na)2.26(Mg,Fe,Al)5.15(Si,Al)8O22(OH)2 - Y: 27.08 % - d x

2-Theta - Scale - 2-Theta

30

20

10

00-024-0020 (D) - Barite - BaSO4 - Y: 15.58 % - d x by: 1. - WL: 1.5406 - 0 -

00-033-0664 (*) - Hematite, syn - Fe2O3 - Y: 21.78 % - d x by: 1. - WL: 1.5406 - 0 - I/Ic 2.4 PDF -

00-001-0527 (D) - Kaolinite - Al2Si2O5(OH)4 - Y: 6.71 % - d x by: 1. - WL: 1.5406 - 0 -

00-007-0032 (D) - Muscovite 2M1, syn - KAl2Si3AlO10(OH)2 - Y: 12.30 % - d x by: 1. - WL: 1.5406 - 0 -

00-001-1111 (D) - Magnetite - Fe3O4 - Y: 12.04 % - d x by: 1. - WL: 1.5406 - 0 -

00-046-1045 (*) - syn Quartz, - SiO2 - Y: 52.35 % - d x by: 1. - WL: 1.5406 - 0 - I/Ic 3.4 PDF -

Operations: Import

File: XRD 2014_C0524 AG.raw - locked2Th/Th Type: 2014_C0524 - AG.raw Start: File: XRD 3.010 ° - End: 69.990 ° - Step: 0.020 ° - S

4

0

100

200

300

400

500

600

700

800

900

1000

1100

1200

1300

1400

1500

1600

1700

1800

1900 2000

Lin (Counts) Lin Figure 12: XRD analysis data from the sample of Agathia beach; sample code: AG.

37

70

60

50

40

00-003-0778 (D) - Ilmenite - - FeTiO3 Y: 15.78 % - d x by: 1. - WL: 1.5406 - Rhombo.R.axes -

00-041-1366 (I) - Actinolite - Ca2(Mg,Fe)5Si8O22(OH)2 - Y: 19.37 % - d x by: 1. - WL: 1.5406 - 0 -

00-036-0426 (*) - Dolomite - CaMg(CO3)2 - Y: 50.00 % - d x by: 1. - WL: 1.5406 - Rhombo.R.axes -

2-Theta - Scale - 2-Theta

30

20

10

00-008-0479 (I) - Magnesite, syn - MgCO3 - Y: 50.00 % - d x by: 1. - WL: 1.5406 - Rhombo.R.axes -

00-020-0572 (D) - Albite, disordered - NaAlSi3O8 - Y: 50.00 % - d x by: 1. - WL: 1.5406 - Triclinic -

00-019-0629 (*) - Magnetite, syn - - FeFe2O4 Y: 30.64 % - d x by: 1. - WL: 1.5406 - 0 - I/Ic 4.9 PDF -

00-002-0056 (D) - Illite - KAl2Si3AlO10(OH)2 - Y: 2.11 % - d x by: 1. - WL: 1.5406 - 0 -

00-001-1229 (D) - Barite - BaSO4 - Y: 4.46 % - d x by: 1. - WL: 1.5406 - 0 -

00-046-1045 (*) - syn Quartz, - SiO2 - Y: 19.97 % - d x by: 1. - WL: 1.5406 - 0 - I/Ic 3.4 PDF -

Operations: Import

File: XRD 2014_C0525 2014_C0525 File: - locked XRD 2Th/Th Type: K - Y Start:AD.raw 3.010 ° - End: 69.990 ° - Step: 0.0

4

0

100

200

300

400

500

600

700

800

900

1000

1100

1200

1300

1400

1500

1600

1700

1800

1900

2000

2100

2200

2300

2400 2500 Lin (Counts) Lin

Figure 13: XRD analysis data from the sample of Adamantas beach; sample code: KY-AD

38

70

60

50

40

00-020-0528 (C) - Anorthite, sodian, ordered - (Ca,Na)(Al,Si)2Si2O8 - Y: 37.09 % - d x by: 1. - WL: 1.540

2-Theta - Scale - 2-Theta

30

20

10

00-010-0359 (D) - Andesine, low - 0.62NaAlSi2O8·0.38CaAl2Si2O8 - Y: 00-010-0359 16.26(D) - - 0.62NaAlSi2O8·0.38CaAl2Si2O8 % Andesine,- d x by: low 1. - WL: 1.540

00-010-0393 (*) - Albite, disordered - Na(Si3Al)O8 - Y: 91.02 % - d x by: 1. - WL: 1.5406 - Triclinic -

00-020-0481 (I) - Magnesiohornblende - (Ca,Na)2.26(Mg,Fe,Al)5.15(Si,Al)8O22(OH)2 - Y: 23.26 % - d x

00-007-0322 (D) - Magnetite - FeO·Fe2O3 00-007-0322 - (D) - Y: Magnetite1.34 % - - d FeO·Fe2O3 x by: 1. - WL: 1.5406 - 0 -

00-010-0353 (D) - Sanidine, high, syn - KAlSi3O8 - Y: 11.70 % - d x by: 1. - WL: 1.5406 - 0 -

00-046-1045 (*) - syn Quartz, - SiO2 - Y: 81.23 % - d x by: 1. - WL: 1.5406 - 0 - I/Ic 3.4 PDF -

Operations: Import

File: XRD 2014_C0519 2014_C0519 - locked2Th/Th Type: EL- File: KI.raw - XRD Start: 3.010 ° - End: 69.990 ° - Step: 0.020 °

4

0

100

200

300

400

500

600

700

800

900

1000

1100

1200

1300

1400

1500

1600

1700

1800

1900

2000

2100

2200 2300

Lin (Counts) Lin Figure 14: XRD analysis data from the sample of Ellinika beach,Kimolos; sample code: KI-EL

39

The XRD results are shown in Table 1 below. The analysis’ results gave a representing view on the mineral composition of the sands and confirmed the existence of iron heavy minerals as occured after the macroscopical analysis. In Fyriplaka’s sample, the minerals detected were: Quartz, Albite, Calcite, Dolomite, Magnetite, Kaolinite, Orthoclase, Barite, Biotite and Chlorite. In Palaiochori’s sample the traced minerals were Quartz, Albite, Kaolinite, Muscovite, Hematite, Andesine, and Chlorite. The XRD analysis didn’t give any ferrous-ferric minerals, though in the examination with a magnet, a few grains were separated from the sample, which indicates the maintenance of ferrous-ferric minerals in an extremely low percentage (<5%). The analysis from the sample from Empourios site has shown: Quartz, Magnetite, Calcite, Orthoclase, Hematite, Albite, Halite, Magnesiohornblende, Ilmenite, Clinochlore, but also Magnesite. In the neighbouring area of Rivari, the major minerals that were traced in the sample are: Magnetite, Hematite, Quartz, Actinolite, Andesine, Magnesite, Albite, Ilmenite, and also Magnesiohornblende. Triades’ sample gave the following: Albite, Anorthite, Magnesiohornblende, Quartz, Magnetite, Hematite, Barite, Dolomite, Magnesite and Ilmenite. The sand sample from Agathia was rich in Quartz, Magnetite, Muscovite, Kaolinite, Hematite, Barite, Magnesiohornblende, Ilmenite and Albite. In Adamantas site, the sand sample was rich in Quartz, Magnetite, Barite, Ilmenite, Illite, Albite, Magnesite, Dolomite and Actinolite. From Kimolos island, the analysis of the sample from the Ellinika beach gave the minerals results as mentioned below: Quartz, Sanidine, Magnetite, Magnesiohornblende, Albite, Andesine and Anorthite. It is worth mentioning that all the sand samples were tested macroscopically with the use of a magnet, in order to confirm the existence of ferrous-ferric minerals. The results were positive about their existence, even though the miniscule quantities of some samples (in particularly, Palaiochori’s sample had an extremely small quantity of ferrous-ferric minerals. Only a few grains in the whole sample-were gathered by the magnet. The XRD analysis though gave no results).

40

Minerals Samples FI PA EM RI TR AG KY-AD KI-EL QUARTZ X X X X X X X X CALCITE X X DOLOMITE X X X ALBITE X X X X X X X X ORTHOCLASE X X ANORTHITE X X ANDESINE X X X SANIDINE X CHLORITE X X CLINOCHLORE X HALITE X ILLITE X ACTINOLITE X X MAGNESIOHORNBLENDE X X X X X BIOTITE X MUSCOVITE X X MAGNESITE X X X X BARITE X X X X ILMENITE X X X X X X MAGNETITE X X X X X X X HEMATITE X X X X X KAOLINITE X X X

Table 1: Mineral occurrences in samples- XRD results

6.2. SEM/EDS analysis and results The SEM/EDS (Energy-dispersive X-ray spectroscopy) analysis of the 8 samples, took place in the facilities of the S&B Industrial minerals-Imerys S.A. in Athens. The chemical analysis of the samples occurred after scanning 30-40 multiple spots in each and calculating their median.

41

Sample Na2O% SiO2% MgO% K2O% TiO2% Ba% Ni% Mn% Pb% Zn% Cr% B-16378/01 Fi 2.62 33.78 5.47 0.24 2.83 0.66 0.02 - - - - B-16378/08 Pa 1.96 69.74 2.26 1.20 0.20 - - 0.1 - - - B-16378/02 Em 4.90 28.36 3.59 0.10 8.67 2.31 - - - - - B-16378/06 Ri 2.63 7.87 2.66 - 8.83 2.18 - - - - - B-16378/04 Tr 5.07 37.26 8.20 0.26 1.11 8.16 - 0.03 0.18 - - B-16378/05 Ag 6.02 31.30 2.10 0.51 5.349 8.07 - - - - 0.09 B-16378/07 Ky-Ad 4.32 29.19 4.55 0.60 7.36 3.64 - 0.05 - - - B-16378/03 El-Ki 6.19 60.76 2.06 3.60 0.06 - 0.02 - - - - Table 2: Geochemical data from SEM/EDS analysis

The samples were examined about five main oxide elements (FeO2 was exluded from the analysis’ results, as its presence was obvious after the macroscopic examination of the samples with a magnete), and six trace elements (Ba, Ni, Mn, Pb, Zn, Cr) in order to study the potential enrichment of the sands’ minerals in these specific heavy metals.

7. DISCUSSION

The mineral composition of the sands depicts the general mineralogy of the petrological types of each area. It is obvious that we cannot fully separate the whole black sand fraction from the total sample, as the magnetite and titanomagnetite could easily be separated from it, with a use of a magnet, but other heavy minerals, which may be also important for the study, could not. In similar way, the silicate and feldspar fraction of the samples was almost impossible to be fully removed from the total sample. The original approach had included the study of potential heavy minerals in the black sand. Μagnetite, ilmenite, hematite and barite were finally found. The study focuses to magnetite especially, in order to examine the potential origin that relates the black sand with a specific petrological type. Also, considering the weathering and eroding conditions as well as the topography of the island, it is possible for us to track the course of the eroded material, and conclude about the most potential petrological type that gave origin to the black sands. Magnetite, as has been mentioned above, is commonly found in igneous rocks of usually mafic composition, but also in metamorphic rocks or granites. In Cyclades, in general, heavy mineral beach sands have also been found, and in particulary in , , , , Naxos, as It is mentioned in the study of Papadopoulos et al. 2016. In the study, the beach sands have been correlated with the ploutonic-granitic rocks of the Atticocycladic zone, which occur on the islands mentioned above, due to their heavy magnetic and heavy non-magnetic fractions, and the REE content of the sands and the eroded material from the parental rocks. In the

42 case of Kimolos, a similar origin of the black sands could be possible, as the intrusive granitic body occurs in outcrops at the central part of the island, though the area from where the sample was taken (Ellinika beach), is covered by the formation of the Upper rhyolitic volcanoclastics. On the contrary though, the case of Milos differs: the island is almost in whole covered mostly by felsic and intermidiate volcanic (specifically rhyolites and dacites with subordinate andesites which overlay basal pyroclastics with basaltic andesites) and sedimentary products, (with the exception of the small appearance of the metamorphic basement), but there is no occurrence of plutonic intrusions. So, considering the locations in Milos where sampling took place (areas with lavas, domes, pyroclastics and tuffs), the hypothesis of this research indicate that the rich in magnetite black sands derived from the weathering of volcanic rocks. The results from the mineralogical analysis has shown large quantities of magnetite in the black beach sands, so the best possible assumption indicates that especially the ferrous-ferric minerals come from the andesitic products of the volcanic periods around and between 2.7-1.4 Ma (Upper Pliocene - Lower-Middle Pleistocene), as well from the products of the earliest volcanic circle of Middle to Upper Pliocene (3.5-3.0 M.a.). These andesitic products cover almost more than the half of the island. According to Fytikas and Vougioukalakis (2003), magnetite phenocrysts are contained to volcanic products of andesitic composition. Considering the sites from where the samples where gathered, it is quite easy to make the correlation, depending on geological field data and mapping after other researchers (including Fytikas et al.,1986, Stewart and McPhie,2005, and Alfieris,2006).

Figure 15: Geological map modified after Fytikas et al.1986

43

As it has beein mentioned before, the earliest volcanic episode (Middle to Upper Pliocene 3.5-3.0 M.a.) produced a thick succession of felsic submarine units, which crop out mainly in the W of the island, especially in the SW part (Alfieris, 2006), including pyroclastic and pumice flows, cinerites, tuffs, with locally subordinate intrusive subvolcanic bodies, lavas and/or hyaloclastites at the end of the sequence, with remnants of submarine tuffs. Locally, the sequence ends with basaltic andesites, which produced pillow lavas, pillow breccias and subordinate hyaloclastites according to Fytikas et al. (1986). After that submarine eruptive activity, during Upper Pliocene (~ 2.7 M.a.) the W part of Milos was affected by a phase of intermediate to acid subvolcanic products in submarine to partially subaerial environment. This volcanic succession is overlaid by the sequence of the Upper Pliocene- Lower-Middle Pleistocene volcanic circle. During this period (2.7-1.4 M.a.), a phase of submarine volcanism was taking place, characterized by the emplacement of a complex of lava flows, plugs and domes (Fytikas et al. 1986; Alfieris,2006) and submarine effusion and intrusion of dacitic/andesitic domes and lavas (Stewart and McPhie 2005; Alfieris, 2006). During the period of Lower-Middle Pleistocene and most probably around 1.4 Ma, due to a combine effect of tectonism and subvolcanic emplacement of the volcanic phase that described above, the deposition of the tuffitic series initially took place at western Milos and continued on eastern Milos (Alfieris, 2006). During Pleistocene (1.4-0.5 Ma),a new volcanic circle of rhyolitic composition developed in the E-NE parts of the island developed, giving pyroclastic products which outcrop in the N-NE part of the island, between Puntes and Pollonia (including the tuffs of Sarakiniko), with smaller outcrops scattered in the area between Adamas and Zephyria, that they consist of different ash flow tuff units locally intercalated by tuffites, pumice flow deposits, cinerites, and thick diatomite-rich ash outcrop. At the N coast of the island though (at the Filakopi area in particular), these pyroclastics are partially covered by a formation of hyaloclastic deposits, mainly pillow lavas and fragments of pillows, of andesitic composition, according to Fytikas et al. (1986); Alfieris (2006). The volcanic episodes that followed in the period of Lower Pleistocene, with pyroclastics of rhyolitic and dacitic composition in the E and NE sector of the island, where extended alterations occure, as well as the acid complexes of Trachilas (0.38 Ma) and Fyriplaka (0.14-0.08 Ma) in Upper Pleistocene, are not supposed to give any heavy minerals of interest. Taking into consideration that the magnetite derives from volcanic products of andesitic (or more basal) composition, it is essential to relate the sites where the samples were gathered, with those neighboring petrological types and the topography, in order to confirm the presence of the black sands to the sea shores, through possible routes.

44

Based on the modified by Fytikas et al.1986 geological map of Milos island, the western part of Milos consists of the sequences described below: The volcanic succession of the Basal pyroclastic series of Middle-Upper Pliocene (~3.5-3.0 Ma) occurs mostly in the SW part of the island. These volcanic products seem to consist part of the main volume of Profitis Ilias mountain and Chondro Vouno. Οccurrences of the Complex of lava flows and domes (~2.4-1.4 Ma) and the Volcanosedimentary series with tuffs and tuffites of the Upper Pliocene are also visible in the area. Το the north, the Complex of lava flows and domes of the Upper Pliocene period is dominant. So, andesitic rocks are the major type of rock that occur in the west part of the island.

Figure 16: Western Milos topographic map

Considering the topography of the area, it would easy to assume that weathered black material is possible to come from these andesitic volcanic formations to the nearest

45 shores of interest, that are, in this case, the Empourios (Em) and Rivari (Ri) areas in the east, and Triades (Tr) and Agathia (Ag) areas in the west. Magnetite is already traced in andesites, and since is a common mineral in basic rocks, it is also possible that the basaltic andesites gave it origin. Also, the ilmenite, which usually accompanies the magnetite, it is most possible to come from the erosion of the Basal pyroclastic series. Ilmenite is a common mineral in mafic-basic igneous rocks, as they are basalts, peridotites etc., or metamorphic rocks of analogous composition. So, at least in the case of the four locations in the west part of Milos, the basaltic andesites of the basal pyroclastic series seem to be the most possible origin for the magnetite+ilmenite mineral fraction of the black sand. Meanwhile, the andesites and tuffitic material of the other sequences may as well feed the black beach depositions with magnetite. Further research must be done in the future relating the tuffitic origin of the iron minerals, since similar material has been observed in pozzolan tuffs in the SW part of the island (Ksylokeratia area).

Image 7: Pozzolanic tuffs with black material (the black material is pointed in red shape)

Another heavy mineral also found in the black sands of Tr and Ag samples is barite. The high ratios of Ba element confirm the quantities. The area of Triades-Galana have been studied in the past for the deposits of silver-rich barite. As it has been mentioned before, barite deposits occur in veins after hydrothermal action in the general area of Triades. Furthermore, in Agathia site, the barite can be transefered from the neighboring area of cape Vani, where barite ores occur as well as manganese ores, but also from the general area of Triades-Galana, by wave action and currents. Another possible explanation is the potential authigenic formation of barite after hydrothermal activity in the area.

46

In the eastern part of the island, two more locations gave samples rich in the heavy minerals magnetite and ilmenite: Adamantas (Ky-Ad), inside the gulf of Milos, and Fyriplaka (Fi) in the south coasts of the island. According to the mapping by Fytikas et al.1986, the eastern part of Milos consists of the Lava domes of Upper Pliocene-Lower- Middle Pleistocene, the Pyroclastic series of the same period and the Volcanosedimentary of M. Pleistocene, the Acid Complexes of Trachilas and Fyriplaka (~1.4 M.a.) and the products of the freatic activity during M.-Up. Pleistocene (the green lahar formation). The sequence of dacitic/andesitic Lava domes of Upper Pliocene -Lower-MIddle Pleistocene in the N-NE and the andesitic hyaloclastic deposits in the N of the island, may consist the formations that provide the magnetite fraction of the black sands in the eastern part of Milos, especially for the location of Adamantas, considering the topography, but also in Sarakiniko site, where black sand was found in cavities in the tuffs on the high level of the water. (Sarakiniko’s tuffs compose part of the Pyroclastic series of acid composition of Lower Pleistocene, so they don’t consist a strong indication for the tuffitic origin of the magnetite. Though further research is needed to be done in the future). Ilmenite, which has also been found with the magnetite in Adamantas sample, must have derived from the same formations, even though it would be expected to be eroded from basal material. Black magnetite sand material has also been found in the area in old arroyos that connect northern territories with the shores in the general Adamantas area. Barite has also been found in the samples Ky-Ad and Fi. Though the hydothermal activity that gave origin to this mineral has been concentrated and traced in W. Milos, barite also occur in Voudia area, in NE milos, but the topography does not favor the transfer by land. In Fyriplaka, the existence of the traced ilmenite and magnetite is a bit difficult to be explained, though the most possible senarios of their origin is the Basal Pyroclastic series of western Milos, as occurrences of this sequence outcrop in the area and coast of Provatas in the South, so eroded material could be transferred via wave action to the coast of Fyriplaka. Palaiochori’s sample (Pa) gave a few grains of iron minerals after macroscopical analysis with a magnet, though the XRD analysis had no results about magnetite. These iron minerals could be magnetite, but it could also be maghemite, an also iron oxide, polymorph of hematite, formed after weathering or oxidation of iron spinels like magnetite (hematite has been traced in the XRD analysis in this sample). Possible parent rock for the potential magnetite in Palaiochori may be some small scattered occurrences of lava domes of Upper Pliocene -Lower-Middle Pleistocene in the eastern section of the island. In general, black magnetite sands involve both minerals magnetite and hematite, among others. It is comprehensible that hematite, also like maghemite, found as part

47 of the heavy minerals in the black sands, might has derived from the oxidation of some ferrous-ferric mineral, in that case magnetite.

The high ratios of TiO2 in the samples, especially those of Empourios (Em) and Rivari (Ri), confirm the existence of ferrous-titanite oxides, so does for ilmenite, but it also indicates that magnetite is titaniferous. In Kimolos’ sample (El-Ki), only magnetite of the heavy minerals was found. The area Ellinika is covered by the formation of Upper Ryolitic Lavas, while the andesitic lavas of the island occur in a small area northern than the Ellinika site, and in the central- eastern central part of the island. Though nowhere in the bibliography is referred about magnetite or other heavy minerals contained in the kimolian andesites, the volcanic circles that produced them is almost the same chronologically and similar to those of Milos (it is reminded that the first cycle of Kimolos volcanism, is dated approximately between 3.5 and 2.0 M.a., comprises the lower lavas of Kimolos, the Kastro ignimbrite and the andesitic -dacitic lavas of Kimolos, while the second cycle, between 2.0 and 0.9 M.a., comprises the Prassa ignimbrite, the domes of the neighbor islet of Polyegos, the andesitic pyroclastics and the Geronikola lavas οf Kimolos, the rhyolitic pyroclastics of Psathi and Mersini and, finally, the domes of Psathi, Xaplovouni and Mersini). So, the andesitic products of the period between 3.5-0.9 M.a. can be considered as the equivalent of these in Milos, so the magnetite material must have the same origin, meaning andesitic lavas, tuffs etc. On the contrary though, the plutonic intrusion of Lower Pliocene, which is a typical granite, rich in ferrum and titanite oxides. So, an also plausible potential concerning the origin of the magnetite in the sand sample from Kimolos, is the kimolian granite. Despite heavy minerals, the sands were also rich in felsic minerals, which constituted the light-coloured fraction of the sands that couldn’t be separated from the total sample. Quartz and felspars are in abundance in almost every sample, since the major petrological compositions of the volcanic rocks on both islands vary from ryolitic to andesitic. Alkali felspars and plagioclases (sanidine, orthoclase, albite, anorthite) exists in every sample. Albite occur in all locations, and the ratios of Na2O confirm that, while the sample with the higher ratio in K2O is that of Kimolos (El-Ki), where sanidine was found. Orthoclase has been found in the Fi and Em samples, and anorthite in Tr and El-Ki samples. The mineral analysis gave minerals from the chlorites and amphiboles groups. The traced amphiboles are hornblende and actinolite. Hornblende is a common mineral in igneous and metamorphic rocks, like andesites (which cover great part of the island) and schists (which also occur as rock fragments from the metamorphic basement in the ‘’green lahar’’ formation). Magnesiohornblende was detected in the Tr, Ag, Ri, Em and El-Ki samples, even though, the existence of ferrohornblende (the other edge in the series ferrohornblende-magnesiohornblende) could also be possible, after the great percentage of Fe in the samples. Actinolite is a dominant amphibole in

48 greenschists, which occur in the ‘’green lahar’’, and was traced in the samples from Rivari and Adamantas. The ‘’green lahar’’ formation expands scattered in a large area in the central and eastern half of the island. As it was mentioned previously, it is observed among the main villages of the island (including Adamantas, at the N and E coasts, at the domes of Kalogeros and Korakia, at the SE coasts (and at Palaiochori general area) and in a small occurrence in the Fyriplaka area and at the depression of Provatas in southern Milos (Fytikas, 1977). The locations where the formation is spread, can be seen more accurately in the modified by Fytikas et al. 1986 geological map of Milos. Chlorites were also found in Fi, Pa and Em. The chlorite from Empourios sample (Em) has been identified as clinochlore (ferrum-magnesium chlorite). Chlorites may as well derive from some schists’ erosion, from the lahar formation, or can be found in the composition of the lavas, as an alteration product of biotite. Biotite has also been traced in the sample from Fyriplaka (Fi), possibly derived from the ryolites in the area, and muscovite in the samples Pa (from Palaiochori) and Ag (from Agathia). Palaiochori’s muscovite is probably derived from schists’ material from lahar’s formation which occurs to the general area. Calcite and dolomite which occure in the samples Fi, Em, Tr, Ky-Ad (in particulary, calcite in Fyriplaka and Empourios’ samples, dolomite also in Fyriplaka and Adamantas). Limestones and dolomites consist part of the Neogenic sequence in the S of the island (southern from Koutsounoraxi, and in both Kipos and Provatas locations), though the calcite may be in many cases authigenic, due to the appropriate conditions. Magnesite has also been traced in samples Em, Ri, Tr and Ky-Ad. Since there is no ultramafic rocks (serpentinites) on the island, the existence of magnesite is quite hard to explain, though it might be the product of some kind of metasomatism of the Neogenic sequence’s dolomites, after volcanism or hydrothermal activity in the area of these sediments. Though, more study needs to be done. From argillic-clay minerals, illite and kaolinite were also traced. Both minerals occur as altered products of aluminium silicate minerals (muscovite and feldspars) after hydrothermal activity. Illite was traced in Adamantas sample and kaolinite in the samples from Fyriplaka, Palaiochori and Agathia. Finally, halite was traced in the sample from Empourios.

49

As for the chemical analysis for trace elements, six (6) heavy minerals were studied in this thesis: Ba, Ni, Mn, Pb, Zn, Cr. The potential enrichment in those metals may indicate the need for further research in the future about ore deposits and metallogenic environments.

Sample Na2O% SiO2% MgO% K2O% TiO2% Ba% Ni% Mn% Pb% Zn% Cr% B-16378/01 Fi 2.62 33.78 5.47 0.24 2.83 0.66 0.02 - - - - B-16378/08 Pa 1.96 69.74 2.26 1.20 0.20 - - 0.1 - - - B-16378/02 Em 4.90 28.36 3.59 0.10 8.67 2.31 - - - - - B-16378/06 Ri 2.63 7.87 2.66 - 8.83 2.18 - - - - - B-16378/04 Tr 5.07 37.26 8.20 0.26 1.11 8.16 - 0.03 0.18 - - B-16378/05 Ag 6.02 31.30 2.10 0.51 5.349 8.07 - - - - 0.09 B-16378/07 Ky-Ad 4.32 29.19 4.55 0.60 7.36 3.64 - 0.05 - - - B-16378/03 El-Ki 6.19 60.76 2.06 3.60 0.06 - 0.02 - - - - Table 3: Geochemical results - comparison of main and trace elements

Barium (Ba) has the higher ratios in the majority of the samples (only the samples from Palaiochori and Kimolos gave no results for Ba). The sands from Triades and Agathia demonstrates the higher ratios in content (~ 8.7-8.8%), while those ratios of Adamantas, Empourios and Rivari sands range in similar rates, lower (~2-3.5%). The lower ratio is that of Fyriplaka’s sample (0.66%). In general, the detected Ba in the samples is explained due to the presence of barite. Nickel (Ni) demonstrates the same ratio in content in both samples of Fyriplaka and Ellinika in Kimolos (0.02%). Manganese (Mn) exhibited low percentages in the samples from Triades (0.03%) and Adamantas (0.05%), though Palaiochori’s sample gave the higher ratio (0.1%). Chromium (Cr) was detected only in the sample Ag (Agathia), with a percentage of 0.09%. Zinc (Zn) wasn’t detected in any of the samples. Even though zinc usually forms binary alloys with elements traced by the geochemical analysis such iron, nickel or magnesium, zero concentrations were finally found.

50

Diagram 1: Trace elements-heavy metals concetrations

Barium demonstrates the higher ratios in the the samples of Triades, Agathia and Adamantas due to the presence of barite in the sands. The area of Triades-Galana have been studied in the past for the deposits of silver-rich barite, as mentioned before. In the cases where barite was not found, but the ratios in Ba still are considerable (Empourios, Rivari and Fyriplaka samples), this element can be found in combination with other elements or even as the second most common phase after barite, which is witherite (barium carbonate). Manganese is usually co-existencial with barite in the case of Milos. Is already known that Cape Vani’s manganese-oxide and barite deposit is hosted in the Volcanosedimentary series of L.-M. Pleistocene on the NW Milos island, and occurs in a 1-km-long marine rift basin that was developed in a footwall of an andesitic lava dome (Kilias, 2011). The manganese mineralization in Vani occurs in a variety of deposit types such as Mn-oxides stratiform to stratabound sheets and lenses, crusts and melanges of white chimneys, with textures and zonations similar to the hydrothermal activity in Palaiochori bay in SW Milos (Valsami-Jones, 2005, Kilias et al.2007), as well as barite-Mn-oxides beds, lences and open-space fillings of barite veins. In Triades, the barite deposits that occur in veins after hydrothermal action in the area are maybe of similar occurrence with the hydrothermal activity in Cape Vani. Otherwise, the existence of manganese is due to transferred with the heavy mineral’s

51 material from the general area of NW Milos. The scenario of transferred manganese material could possibly explain the occurrence of the metal in Adamantas’ sample. But in the case of Palaiochori bay, hydrothermal activity seems to be the more plausible interpretation for the existence and the ratios in manganese. An active hydrothermal field producing sulfides in the bay area and a barite ore in the SE of the island could consist indication of possible manganese metallogenesis in small quantities. Chromium and nickel are commonly bound with iron and manganese oxides. Due to the non-existence of chromites on the island, chromium might as well occur in alloys with the iron (and potentially manganese) oxides that the area (Agathia site) hosts. Nickel in a similar way can be bound within iron oxides, as it substitutes other transition metals within ferromagnesian minerals, the most usual being olivine, nickel- ferrous varieties of amphibole, biotite, pyroxene and spinels, because of the similarity in size and charge of Ni with Fe2+. Also, Ni as being a chalcophile-like element, it has preference for sulfide phases. Due to the hydrothermal and metallogenic environments on the islands, alloys with iron, arsenic and silver are also possible. Pyrite, chalcopyrtite and galena are probably the most possible bearing minerals, since the mineralization in Profitis Ilias comprises mostly of these minerals (Kilias et al. 2001, Alfieris, 2006), though in Fyriplaka site, in southern Milos (and also on Kimolos island) none of these minerals where traced in the mineralogical analysis. Same situation must be this of the detected lead (Pb) in the sample from Triades. Lead is normally found in sulfides, and the most common lead-bearing mineral is galena, which constitutes dominant phase in Triades-Galana mineralization (Alfieris, 2006) though the sample from Triades gave no results for galena. Due to the two main oxidation states +4 and +2 that lead demonstrates, lead could replace iron or barite cations. Since barite has been traced in Triades’ sample, it is most possible for lead to consist part of barite’s plexus, composing plumbean barite. In any case, more research must be done in order to examine every possible indication and conclude about the occurrence or not of new heavy metal ores or deposits.

8. Conclusions

After the mineralogical and geochemical analysis of the black magnetite sands, the large quantities of iron minerals, especially magnetite, originate from volcanic products of andesitic composition. These products consist of lavas, pyroclastics, tuffs etc. In Milos case, andesitic products of the volcanic periods around and between 2.7- 1.4 Ma (Upper Pliocene - Lower-Middle Pleistocene), as well as the earliest volcanic circle of Middle to Upper Pliocene (3.5-3.0 M.a.), with the characteristic occurrence of basaltic andesites, may consist the origin for the ferrous-ferric minerals. In Kimolos case, the magnetite material may also derive from the andesitic products of the two

52 distinguished volcanic circles between 3.5-0.9 M.a., or the granitic intrusion that occurs scattered in the central island. As it concerns the andesitic material, is yet to be ascertained whether magnetite derives from lavas, tuffs or other formation of andesitic composition, since magnetite phenocrysts could be contained in all of them petrological types. So, further research needs to be done. As for the geochemical analysis, the main elements gave results in agreement with the calc-alkaline geochemistry of the volcanism in Milos general area. The high ratios in TiO2 implies the existence of titanomagnetite, and ilmenite as well, in the black sands. The results from the analysis for the six heavy trace elements (Ba, Ni, Mn, Pb, Cr, Zn) relates the occurrence and ratios in the samples with the already existed mineralization and hydrothermal activity of the islands, especially for barium and manganese. Despite the totally absence of Zn, the rest of the metals occur in low or higher ratios in the samples (with higher percentages those of Ba, and lower of Mn, Ni, Pb), due to their presence as alloys and replacements in the mineral plexus of ferrous-ferric and other heavy minerals. So, the enrichment of black beach sands in heavy metals needs more of research in the future, in order to ascertain possible new areas of metallogenesis and mineralization in the islands of Milos and Kimolos.

Aknowlegments:

This thesis has been also supported by Ph.D. Dimitris Alfieris, Resources Development Geologist at Imerys Filtration - Performance Additives for Metallurgy Division, F&PA Business Group, Rania Margariti, General Lab Athens Supervisor of IMERYS INDUSTRIAL MINERALS GREECE S.A. (Athens), and Dr. George Oikonomou, Geologist, Director of MIneralogy - Petrography Dpt. Institute of Geological and Mineral Exploration - Insitute of Gelogical and Mineral Exploration ( I.G.M.E.).

53

REFERENCES:

Alfieris D.(2006) Geological, geochemical and mineralogical studies of shallow submarine epithermal mineralization in an emergent volcanic edifice, at Milos island (Western side), Greece, Dissertation, Zur Erlangung des Doktorgradesder Naturwissenschaften im Department Geowissenschaften der Universität Hamburg, p.2-116 Angelier, J., Cantagrel, J.M. and Vilminot, J.C., 1977. Neotectonique cassante et volcanisme plio-quaternaire dans l'arc 6gden interne: L'Ile de Milos (Grece). Bull. Soc. G4ol. Fr., 19: 119--121.

Christidis G., Dunham A. C., (1991) Compositional variations in smectites: Part I. Alteration of intermediate volcanic rocks. A study from Milos island, Greece, Department of Geology, University of Leicester, University Road, Leicester LEI 7RH, UK, p. 1-6 Christidis G., Dunham A. C., (1992) Compositional variations in smectites: Part iI. Alteration of acidic precursors, a case study from Milos island, Greece, Department of Geology, University of Leicester, University Road, Leicester LEI 7RH, UK, Clay minerals (1997) 32, p.253-263 Christidis G., (2000) Geochemical correlation of bentonites from Milos Island, Aegean, Greece, Clay minerals (2001) 36, p. 295-306 Durr, St., Altherr, R., Keller, J., Okrusch, M. and Seidel, E., 1978. The median Aegeancrystalline belt. Stratigraphy, structure, metamorphism, magmatism. In: H. Closs, D. Roeder and K. Schmidt (Editors), Alps, Apennines, Hellenides. IUGG Sci. Rep., 38: 455--477. Fytikas, M., Giuliani, O., Innocenti, F., Marinelli, G., Mazzuoli, R. (1976a) “Geochronological data on recent magmatism of the ”. Tectonophysics, 31, p.29-34. Fytikas M., Kouris D., Marinelli G., and Surcin J. (1976b) : Preliminary geological data from the first two productive geothermal wells drilled at the island of Milos. In : proceedings of Inter. Congress on thermal waters, geothermal energy and volcanism of the Mediterranean area. Athens Oct. 1976. Expanded abstracts p. 511 - 515. Fytikas M. (1977) Geological and geothermal study of Milos island (Greece). Phd thesis,University of Thessaloniki. Geological and geophysical research vol. XVIII, No 1,(IGME) p.1-228.

54

Fytikas, M., Innocenti, F., Kolios, N., Manetti, P., Mazzuoli, R., Poli, G., Rita, F. & Villari,L. (1986) -Volcanology and petrology of volcanic products from the issland of Milos and neighbouring islets. J.Volcan. Geoth. Res. 28, p.297-317. Fytikas, M. and Marinelli, G., (1976). Geology and geothermics of the Island of Milos (Greece). Proc. Geothermal Energy, Vol. 1,516--524. Fytikas M. and Vougioukalakis G.(1993) Volcanic structure and evolution of Kimolos and Polyegos (Milos island group). Bull. Geol. Soc. Greece, vol. XXVIII/2, p. 221- 237. Fytikas M. and Vougioukalakis G. (2005) The South Aegean Active Volcanic Arc: Present Knowledge and Future Perspectives, Elsevier (2005), Developments in Volcanology, 7, p. 2-85 Innocenti F., Manetti P., Peccerillo A., Poli G. (1981) South Aegean Volcanic Arc: Geochemical Variations and Geotectonic Implications. Bull. Volcanol., Vol. 44- 3, p.377-391. Kilias S.P., Naden J., Cheliotis I., Shepherd T.J., Constandinidou H., Crossing J. and Simos I. (2001) Epithermal gold mineralisation in the active Aegean volcanic arc: the Profois Ilias deposit, Milos island, Greece, Mineralium Deposita, 36, p.32-44. Kilias S.P., (2011) Microbial mat-related structures in the Quaternary Cape Vani manganese-oxide-(barite) deposit , NW Milos island, Greece, Department of Economic Geology and Geochemistry, Faculty of Geology and Geoenvironment, National and Kapodistrian University of Athens, Athens, 15784, Greece, p. 1-3 Lyberis N., Angelier J., Huchon P. Le Pichon X., Renard V. (1981) La mer de Crete : Extension et Subsidence. In Intern. Symp. On the Hell. Arc and Trench (H.E.A.T.), Athens-Greece, vol.1, p. 364-382. Mange M.A. and Maurer H., (1992) Heavy Minerals in Color, London: Chapman and Hall, p.16-49 Mariolakos I.D. and Papanikolaou D.J.(1981) The Neogene basins of the Aegean arc from the paleogeographic and the geodynamic point of view, p.383-399

Mitropoulos P., Tarney J., Saunders A.D., Marsh N.G., Petrogenesis of Cenozoic Volcanic rocks from the Aegean island Arc, (1986), Journal of Volcanology and Geothermal Research, 32 (1987) 177-193, p. 177-180 Papadopoulos A.,, Koroneos A., Christofides G., Papadopoulou L., I. Tzifas I. ,S. Stoulos (2016)Assessment of gamma radiation exposure of beach sands in highly touristic areas associated with plutonic rocks of the Atticocycladic zone (Greece), Journal of Environmental Radioactivity 162-163 (2016) 235e243, p.235-237

55

Papanikolaou D.J. and Dermitzakis M.D. (1981) Major changes from the last stage of the Hellenides to the actual Hellenic arc and trench system. In : Intern. Symp. on the Hell. Arc and Trench (H.E.A.T.), Athens-Greece, vol.2, p. 57-73. Papanikolaou D.J. (2014) Geology of Greece, Athens University, Dept. of Geol., Section of Tectonic Geology, 250 pp. Papazachos B.C., Panagiotopoulos D.G. (1993) Normal faults associated with volcanic activity and deep rupture zones in the southern Aegean volcanic arc. Bull. Geol. Soc. Greece, vol. XXVIII/3, p. 243-252. Stewart A.L. & McPhie J. (2003) Setting of epithermal Au deposits in a modern volcanic island arc setting, Milos, Greece: Implications for mineral exploration. In: Eliopoulos D.et al. (eds) Mineral Exploration and Sustainable Development, 1, p.533-536, Millpress, Rotterdam. Stewart A.L. & McPhie J. (2004) Facies architecture of the submarine to subaerial volcanic succession on Milos, Greece. 17 AGC Hobart- Tasmania. Abstracts, in p. 293. Stewart A.L. & McPhie J. (2005) Facies architecture and Late Pliocene-Pleistocene evolution of a felsic volcanic island, Milos, Greece. Bull. Volcanol., 24p. Valsami-Jones E., Baltatzis E., Bailey E.H., Boyce A.J., Alexander J.L., Magganas A., Anderson L., Waldron S., Ragnarsdottir K.V. (2005) The geochemistry of fluids from an active shallow submarine hydrothermal system: Milos island, Hellenic Volcanic Arc. J. Volc. Geother. Res 148, p.130-151.

Internet sources: http://serc.carleton.edu/research_education/geochemsheets/techniques/XRD.html https://serc.carleton.edu/research_education/geochemsheets/techniques/SEM.htl https://serc.carleton.edu/research_education/geochemsheets/eds.html http://ormuswater.vpinf.com/magnetite-sand-1.html https://www.mindat.org/ http://www.rsc.org/periodic-table/element/

56

Annex I:

Geological map of Milos island (source: IGME)

57