1

ΝΑΤΙΟΝΑL AND KAPODISTRIAN

UNIVERSITY OF ATHENS

DEPARTMENT OF GEOLOGY AND GEOENVIRONMENT FACULTY OF ECONOMIC GEOLOGY AND GEOCHEMISTRY

MASTER OF SCIENCE ON ENVIRONMENTAL APPLIED

GEOLOGY

SPIROS OLIVOTOS

Examination Board:

Dr. Maria Economou – Eliopoulos (supervisor)

Dr. Konstantinos Papavasiliou

Dr. Ariadne Argyraki

ATHENS

JULY 2015

“THERMODYNAMIC CONTROLS ON

THE FORMATION AND STABILITY

OF PLATINUM-GROUP MINERALS”

2

ΕΘΝΙΚΟ ΚΑΙ ΚΑΠΟΔΙΣΤΡΙΑΚΟ ΠΑΝΕΠΙΣΤΗΜΙΟ ΑΘΗΝΩΝ

ΣΧΟΛΗ ΘΕΤΙΚΩΝ ΕΠΙΣΤΗΜΩΝ

ΤΜΗΜΑ ΓΕΩΛΟΓΙΑΣ ΚΑΙ ΓΕΩΠΕΡΙΒΑΛΛΟΝΤΟΣ

ΠΡΟΓΡΑΜΜΑ ΜΕΤΑΠΤΥΧΙΑΚΩΝ ΣΠΟΥΔΩΝ

ΠΡΟΓΡΑΜΜΑ ΜΕΤΑΠΤΥΧΙΑΚΩΝ ΣΠΟΥΔΩΝ

ΕΦΑΡΜΟΣΜΕΝΗ ΠΕΡΙΒΑΛΛΟΝΤΙΚΗ ΓΕΩΛΟΓΙΑ

ΜΕΤΑΠΤΥΧΙΑΚΗ ΕΡΓΑΣΙΑ ΕΙΔΙΚΕΥΣΗΣ

«ΣΧΗΜΑΤΙΣΜΟΣ ΚΑΙ ΕΥΣΤΑΘΕΙΑ ΟΡΥΚΤΩΝ ΤΩΝ ΣΤΟΙΧΕΙΩΝ ΤΗΣ

ΟΜΑΔΑΣ ΤΟΥ ΛΕΥΚΟΧΡΥΣΟΥ: ΜΙΑ ΘΕΡΜΟΔΥΝΑΜΙΚΗ

ΠΡΟΣΕΓΓΙΣΗ»

ΣΠΥΡΟΣ ΟΛΙΒΩΤΟΣ

Eξεταστική επιτροπή :

Μαρία Οικονόμου, Ομοτ. Καθηγήτρια, Ε.Κ.Π.Α (επιβλέπουσα),

Κωνσταντίνος Παπαβασιλείου, Αναπλ. Καθηγητής, Ε.Κ.Π.Α

Αριάδνη Αργυράκη, Αναπλ. Καθηγήτρια, Ε.Κ.Π.Α

ΑΘΗΝΑ

ΙΟΥΛΙΟΣ 2015

3

TABLE OF CONTENTS I. LIST OF TABLES ................................................................................................................................ 4

II. LIST OF FIGURES ............................................................................................................................... 5

III. ACKNOLEDGEMENTS ...................................................................................................................... 7

IV. ABSTRACT .......................................................................................................................................... 8

1. INTRODUCTION ............................................................................................................................... 12

1.1. GEOCHEMISTRY OF PLATINUM GROUP ELEMENTS ..................................................... 16

1.2. Mafic-Ultramafic Rocks ............................................................................................................. 17

1.2.1. Mantle Rocks ...................................................................................................................... 18

1.2.2. Komatiites ........................................................................................................................... 26

1.2.3. Ophiolites ............................................................................................................................ 29

1.3. CHROMITE DEPOSITS ............................................................................................................ 34

2. METHODOLOGY .............................................................................................................................. 35

2.1. MATERIALS AND METHODS ................................................................................................ 35

2.1.2. Skyros Island ....................................................................................................................... 35

2.1.3. Examples from the Literature ............................................................................................. 42

2.2. PLATINUM GROUP MINERALS ............................................................................................ 50

2.3. PGM IN LAYERED INTRUSIONS AND OPHIOLITE COMPLEXES .................................. 52

2.4. CALCULATING PROCESS ...................................................................................................... 63

2.4.1. Thermodynamic Laws ......................................................................................................... 64

2.4.2. Free Energy ......................................................................................................................... 68

2.5. EVALUATION OF THE METHODOLOGY ............................................................................ 71

3. CALCULATED GIBBS FREE ENERGY AND STABILITY DIAGRAMS .................................... 75

4. APPLICATION OF (ΔGf) VALUES .................................................................................................. 83

4.1. THE ORIGIN OF PGM .............................................................................................................. 83

4.2. THE STABILITY OF PGM ....................................................................................................... 86

4.3. THE POTENTIAL ENVIRONMENTAL RISK ........................................................................ 87

5. CONCLUSIONS ................................................................................................................................. 90

6. REFERENCES .................................................................................................................................... 91

4

I. LIST OF TABLES

No. Caption Page

Table 1 Selected properties of the six platinum-group elements (PGE) (British Geological Survey,

Natural Environment Research Council, Platinum, September 2009) 16

Table 2 PGE data for lherzolites, harzburgites, websterite and gabbro dyke rocks from several

orogenic peridotites 19

Table 3 a) PGE data for mantle nodules and abyssal peridotites from oceanic environments 23

b) PGE data for mantle nodules and abyssal peridotites from continental environments 24

Table 4 PGE data for peridotitic komitiites 27

Table 5 a) PGE data for Ophiolitic Complexes 30

b) PGE data for Ophiolitic Complexes from Greece 32

Table 6 Representative microprobe analyses of PGM of PGE – rich chromites from Skyros Island,

Greece 38

Table 7 Representative microprobe analyses of Laurite, Skyros Island, Greece 39

Table 8 Representative Microprobe Analyses of Sperrylite, Skyros Island, Greece 40

Table 9 Representative microprobe analyses of PGM from PGE-rich chromites of the Veria

ophiolite Complex 43

Table 10 Representative Microprobe Analyses of Pt – Fe and Os – Ir – Ru alloys in Alaska type

systems 47

Table 11 Representative Microprobe Analyses of Pt – Fe and Os – Ir – Ru alloys in Ophiolite type

systems 47

Table 12 Platinum Group Minerals (ideal formulas) 50

Table 13 A) Comparison between ΔGf values of (a) this study and (b) Jacob & Gupta (2014) for the

reaction 3Rh(s) + 2S2(g) = Rh3S4(s) 71

B) Comparison between ΔGf values of (a) this study and (b) Jacob & Gupta (2014) for the

reaction 4Rh3S4(s) + S2(g) = 6Rh2S3(s) 72

Table 14 Gibbs free energy (ΔG/kJ*mol-1) of formation for Platinum-group minerals estimated from

the literature (experimental) data 76

Table 15 Palladium and platinum in porphyry-Cu-Au-Pd-Pt deposits 88

Table 16 Precious metal content in dust and soils along roadside 89

5

II. LIST OF FIGURES

No. Caption Page

Figure 1 Periodic Table of elements, compatible (IPGE) and incompatible (PPGE) elements. 12

Figure 2 Location map showing mafic-ultramafic complexes associated with PGE mineralizations 14

Figure 3 PGE correlation of orogenic lherzolites and primitive mantle abundances 20

Figure 4 PGE correlation of orogenic harzburgites and primitive mantle abundances 21

Figure 5 PGE correlation of dyke rocks and primitive mantle abundances 22

Figure 6 PGE correlation of oceanic environment mantle nodules and abyssal peridotites with

primitive mantle abundances 25

Figure 7 PGE correlation of continental environment mantle nodules and abyssal peridotites with

primitive mantle abundances 26

Figure 8 PGE correlation of komatiites with primitive mantle abundances 28

Figure 9 a) PGE correlation of chromites in ophiolite complexes with primitive mantle

abundances 31

b) PGE correlation of chromites in ophiolite complexes from Greece with primitive

mantle abundances 33

Figure 10

Chondrite-normalized PGE-patterns (sample/C1 chondrite) for chromite ores hosted in

ophiolite complexes from the Vourinos and other large chromite deposits (a), IPGE-

enriched ores (b) and PPGE-enriched ores (after Economou-Eliopoulos et al. 2013).

34

Figure 11 Back – scattered electron image of chromite (chr) from the Skyros island. Inclusions of

chlorite (chl) and serpentine (spt) in chromite 36

Figure 12

Back – scattered electron images of strongly fragmented chromite (chr) associated with

garnet (grt) and serpentine (spr) (Fig. 12a, b), IPGM in a matrix of serpentine associated

with chromite/Fe – chromite (Fe – chr) (Fig. 12c, d, e) and alloys along cracks of chromite

and Fe – chromite (Fig. 12f) from Skyros Island. Heazlewoodite is also common (Fig.

12b).

37

Figure 13 Back – scattered electron images of heazlewoodite (Ni3S2) as inclusion in chloritic matrix

(chl) in contact with magnetite (mt). 38

Figure 14 Back – scattered electron images of laurite crystal as inclusion in chromite, Skyros Island,

Greece. 39

Figure 15

Inclusion of sperrylite (sp) in intergrowth with chalcocite (cc) in magnetite. Reflected

light. Scale bar = 15#m. (A) and Secondary electron image of sperrylite in the

serpentinized silicate matrix (B), Skyros Island, Greece

41

Figure 16 Back – scattered electron images of fragmented chromite (chr) with an abundance of

PGM grains, Veria, Greece 43

Figure 17

Back-scattered electron images of (A): A typical fine laurite inclusion within unaltered

chromite. (B): An extremely coarse PGM grain and (C): remnants of laurite and Ru-Os-

Ir-oxides within a large laurite grain, (D): Pt-Rh-irarsite between chromite grains, all

from the Veria ophiolite complex (Greece), (E): (Pt,Rh,Cu)S from the UG2 horizon and

(F): (Pd, Pt)S from the Merensky reef, Bushveld complex. Abbreviation: irs = irarsite;

45

6

chr = chromite; Fe-ch = iron chromite; Po = pyrrhotite; srp=serpentine; grt = garnet; cpx

= clinopyroxene

Figure 18 Modal distributions of PGM in the placers derived from Ural-Alaska type complexes 48

Figure 19

Microphotographs of PGM showing different paragenesis from placers related to the

Ural-Alaskan-type: (A) Pt-Fe alloy-osmium; (B) isoferroplatinum-iridium; and from

placers related to ophiolite-type ultramafic rocks: (C) osmium-iridium; (D)

isoferroplatinum-ruthenium.

49

Figure 20 – 27 Pie charts with the most dominant PGMs in worldwide regions 54

Figure 28

Comparison of the calculated Gibbs free energy (ΔG) against temperature, for two

reactions. (A) 3Rh(s) + 2S2(g) = Rh3S4(s) in this study with values obtained by previous

authors. Data for line (a) from Table 6A; data for line (b): ΔGf (J * mol-1) = -548026 +

304.5 * T (K) from Jacob & Gupta (2014) and (B) 4Rh3S4(s) + S2(g) = 6Rh2S3(s) in this

study with values obtained by previous authors. Data for line (a) from Table 6B; data for

line (b): ΔGf (J * mol-1) = -230957 + 160.03 * T (K) from Jacob & Gupta (2014)

74

Figure 29 - 31 Plot of the Free energy (ΔG) against temperature for dominant PGMs. Data from Table

14 77

Figure 32 A comparative diagram plotting the Free energy (ΔG) against temperature. Data from

Table 14 80

Figure 33 Plot of the Free energy (ΔG) against temperature for RuO2 and RuS2. Data from Table

14 81

7

III. ACKNOLEDGEMENTS

Dr. Maria Economou – Eliopoulos is gratefully acknowledged for the scientific support,

supervision and personal motivation. Her encouragement was catalytic throughout the completion

of my diploma thesis. Additionally special thanks to my professors, Dr. Ariadne Argyraki & Dr.

Konstantinos Papavassiliou who supported me and provided me with many useful comments and

indications. The University of Athens (ΚΕ 11078) is also acknowledged for the financial support

of this work.

8

IV. ABSTRACT

The investigation, by SEM/EDS, of chromite ores from Skyros island, which are

characterized by elevated platinum-group element (PGE) content, showed that apart from laurite

(RuS2) occurring as small inclusions within unaltered chromite grains, Ru-Ir-Os- minerals were

found within silicate matrix as interstitial between chromite grains. An association of laurite (Ru-

Ir-Os-sulfide) remnants with Ru-Ir-Os- alloys Ru-Ir-Os-oxides and/or hydroxides showing a wide

compositional variation and significant Fe, Ni, Cr, Sb and As contents and their association with

serpentine, chlorite and Cr-garnet is a common feature in ophiolite complexes of Greece (Veria,

Pindos) and elsewhere worldwide. They may be related to post-magmatic processes, during an

extended period of deformation, including the ductile asthenospheric mantle flow and crustal

brittle deformation. Metasomatic fluids have substantially modified the composition of primary

laurite.

Despite the great interest in the Os, Ir, Ru, Pt, Pd and Rh, and the extensive investigations

of the PGE behavior during partial melting and crystal fractionation, thermodynamic data for

PGE/PGM are very limited, due probably to required experiments at high temperatures or

pressures and very long equilibration processes. Present study is focused on the calculation of the

Gibbs free energy of formation for dominant PGM in layered intrusions and ophiolite complexes

worldwide, applying available experimental data on Gibbs free energy for PGEs. The free energy

of formation (ΔGof) was calculated as a function of ΔG = G(species) - ΔG(elements), applying the

computer program HSC Chemistry software 6.0. The evaluation of the accuracy of the method of

the (ΔGof) calculation at standard state conditions was made by the calculation of (ΔGf) of rhodium

sulfide phases. The calculated (ΔGf) for rhodium sulfides were found to be in a good agreement

with those measured in the binary system (Rh + S) as a function of temperature by previous authors

(Jacob and Gupta (2014). The diagrams of the calculated values of (ΔG) versus temperature

(Ellingham diagrams) are presented.

The estimated Gibbs free energy of formation (ΔGof) followed the order: RuO2<RuS2<

OsS2< (Pt, Pd)S< (Pd, Pt)Te2, increasing from compatible to incompatible noble metals, from

oxides to sulfides and from sulfides to tellurides. Also, the calculated Gibbs free energy of

formation (ΔGof) for the oxide RuO2 is lower compared to that for RuS2 at temperatures lower than

500oC, suggesting that the a transformation of laurite to oxides and probably hydroxides is

9

facilitated at relatively low temperature. In addition, thermodynamic data on the stability of Pt, Pd

and Rh in the environment (literature data) indicating that they can be bio-available are consistent

with the bio-accumulation of these elements derived from traffic emissions (automobile catalyst

converters) along high-way roadside.

Therefore, the calculated Gibbs free energies (ΔGof) (molecular scale) appear to be in a

good agreement with textural relationships between PGM and associated minerals (microscope,

SEM) and geological data (macroscopic scale), suggesting that a compilation of all data may

contribute to a better understanding of the PGM-forming processes.

10

ΠΕΡΙΛΗΨΗ

Η έρευνα, με SEM/EDS, μεταλλευμάτων χρωμίτη από το νησί της Σκύρου, τα οποία

χαρακτηρίζονται από αυξημένες περιεκτικότητες στοιχείων της ομάδας του λευκόχρυσου (PGE),

έδειξε ότι, εκτός από τον λωρίτη (RuS2), που παρατηρείται με τη μορφή μικρών εγκλεισμάτων

μέσα σε μη εγαλλοιωμένους κόκκους χρωμίτη, βρέθηκαν κράματα και οξείδια Ru-Ir-Os μέσα σε

πυριτική κύρια μάζα ενδιάμεσα των κόκκων του χρωμίτη. Το φαινόμενο αυτό, δηλαδή η

συνύπαρξη υπολειμμάτων λωρίτη (Ru-Ir-Os- σουλφίδιο) με κράματα Ru-Ir-Os-, και οξείδια Ru-

Ir-Os- έχει παρατηρηθεί στα οφιολιθικά συμπλεγμάτα της Βέροιας και Πίνδου (Β. Ελλάδα, αλλά

και παγκοσμίως. Τα ορυκτά αυτά χαρακτηρίζονται από ευρεία διακύμανση της χημικής σύστασης,

σημαντικά ποσοστά σε Fe, Ni, Cr, Sb και As, και συνδέονται με σερπεντινίτη, χλωρίτη και

χρωμιούχο γρανάτη. Ενδεχομένως, να σχετίζονται με μετα – μαγματικές/ δευτερογενείς

διεργασίες, κατά την διάρκεια μίας μακράς περιόδου παραμόρφωσης, συμπεριλαμβανομένης και

της πλαστικής παραμόρφωσης της ασθενόσφαιρας και της θραυσιγενούς παραμόρφωσης κατά

την τοποθέτηση του οφιολιθικού συμπλέγματος.. Παράλληλα, η επίδραση ρευστής φάσης έχει

τροποποιήσει την σύσταση των PGM.

Παρά το μεγάλο ενδιαφέρον που επικρατεί για τα Os, Ir, Ru, Pt, Pd and Rh, και τις εκτενείς

έρευνες της συμπεριφοράς των PGE κατά τη διάρκεια της μερικής τήξης και της κλασματικής

κρυστάλλωσης τα θερμοδυναμικά δεδομένα για τα PGE/PGM είναι περιορισμένα, πιθανόν λόγω

των απαιτούμενων πειραμάτων σε υψηλές θερμοκρασίες ή πιέσεις και χρονοβόρων διεργασιών

χημικής ισορροπίας. Η παρούσα μελέτη επικεντρώνεται στον υπολογισμό της ελεύθερης

ενέργειας Gibbs για την διαμόρφωση των PGM που επικρατούν σε ενστρωμμένες διεισδύσεις

και σε οφιολιθικά συμπλέγματα παγκοσμίως, εφαρμόζοντας διαθέσιμα πειραματικά δεδομένα της

ελεύθερης ενέργειας του Gibbs για τα PGE. Η ελεύθερη ενέργεια δημιουργίας αρκετών PGM

(ΔGof) υπολογίστηκε μέσω της εξίσωσης: ΔG = G(species) – ΔG(elements), κάνοντας χρήση του

υπολογιστικού προγράμματος HSC Chemistry software 6.0. Η αξιολόγηση της ακρίβειας της

μεθόδου υπολογισμού σε σταθερές συνθήκες κατάστασης έγινε με τον υπολογισμό του (ΔGf) για

τις φάσεις του σουλφιδίου του ρόδιου. Οι υπολογισμένες τιμές ελεύθερης ενέργειας για τα

σουλφιδία του ροδίου βρέθηκαν να είναι σύμφωνες με αυτές που μετρήθηκαν σε δυαδικό σύστημα

(Rh + S) από προηγούμενους συγγραφείς (Jacob and Gupta (2014). Παρουσιάζονται επίσης,

διαγράμματα από τις υπολογισμένες τιμές του (ΔG) σε σχέση με την θερμοκρασία (Ellingham

diagrams).

11

Η υπολογισμένη ελεύθερη ενέργεια δημιουργίας του Gibbs (ΔGof) για τα PGM

ακολούθησε τη σειρά: RuO2<RuS2<OsS2< (Pt, Pd)S<(Pd, Pt)Te2, αυξανόμενη από τα

ανταγωνιστικά σε μη ανταγωνιστικά ευγενή μέταλλα, από οξείδια σε σουλφίδια και από σουλφίδια

σε τελλουρίδια. Επίσης, η υπολογισμένη ελεύθερη ενέργεια Gibbs (ΔGof) για το οξείδιο RuO2

είναι χαμηλότερη σε σύγκριση με αυτήν του RuS2 σε θερμοκρασίες χαμηλότερες των 500oC,

προτείνοντας ότι ο μετασχηματισμός του λωρίτη σε οξείδια και πιθανόν σε υδροξείδια

διευκολύνεται σε σχετικά χαμηλή θερμοκρασία. Επιπλέον, θερμοδυναμικά δεδομένα σχετικά με

την σταθερότητα των Pt, Pd και Rh στο περιβάλλον (βιβλιογραφικά δεδομένα) ,τα οποία

υποδεικνύουν ότι είναι βιο – διαθέσιμα, συνάδουν με την βιο – συσσώρευση αυτών των στοιχείων,

τα οποία εκπέμπονται από καταλύτες αυτοκινήτων, κατά μήκος μεγάλων λεωφόρων.

Ως εκ τούτου, οι υπολογισμένες ελεύθερες ενέργειες Gibbs (ΔGof) (σε μοριακή κλίμακα)

φαίνεται να είναι σε καλή συμφωνία με τις ιστολογικές σχέσεις μεταξύ των PGM και των σχετικών

ορυκτά (μικροσκόπιο, SEM) και των γεωλογικών δεδομένων (μακροσκοπική κλίμακα),

προτείνοντας ότι η σύνθεση όλων αυτών των δεδομένων μπορεί να συμβάλλει σε μία καλύτερη

κατανόηση των διεργασιών δημιουργίας των ορυκτών της ομάδας του λευκοχρύσου.

12

1. INTRODUCTION

Platinum group elements (PGE) is a group consisted of 6 main metals, platinum (Pt), palladium

(Pd), iridium (Ir), osmium (Os), rhodium (Rh) and ruthenium (Ru). The chemical properties of the

elements are quite similar (Table1). Characteristically, platinum, iridium and osmium are the

densest known metals, being significantly denser than gold. Platinum and palladium are highly

resistant to heat and to corrosion, and are soft and ductile. Rhodium and iridium are more difficult

to work, while ruthenium and osmium are hard, brittle and almost unworkable. All PGE,

commonly alloyed with one another or with other metals, can act as catalysts which are exploited

in wide range of industrial applications (British Geological Survey, Natural Environment Research

Council, Platinum, September 2009) (Table 1).

Figure 1: Periodic Table of elements, compatible (IPGE) and incompatible (PPGE) elements.

PGE are among the most valuable elements in nature with strategic importance, due to their

growing use in advanced technologies and automobile catalyst converters. The Bushveld Complex

(South Africa) has the world's largest reserves of PGEs, while significant world supply of PGE is

produced from the Great Dyke (Zimbabwe) and the Stillwater Complex (USA) and other magmatic

ores derived from basaltic magmas (Lee and Tredoux, 1986; Cowden et al., 1986; Barnes and

Campbell, 1988; Cawthorn, 1999; Barnes and Maier, 1999, 2002; Cawthorn, 1999; Naldrett and

13

Duke, 1980; Campbell et al., 1983; Fig. 2). Although the PGE content in ophiolite complexes

(chromite is a major collector of PGE) is relatively low (a few hundreds of ppb) in large chromite

deposits, a significant PGE-enrichment: (a) in all PGE, (b) only in Os, Ir and Ru or (c) in Pt and/or

Pd is a common feature of disseminated chromite and/or relatively small chromite occurrences in

the uppermost parts of the mantle and/or in the lowest crust sequence (Ohnenstetter et al., 1999;

Prichard et al., 1986, 1994; Economou-Eliopoulos, 1996; Tarkian et al., 1996). Platinum-group

elements (PGE) can be classified into two subgroups: the Os-, Ir- Ru-rich or IPGE (Ir-goup)

showing compatible behavior and (Pt, Pd)-rich or PPGE (Pt-group) assemblages, showing

enrichment as a function of the differentiation degree (Barnes and Picard, 1993). The behavior of

Os, Ir, Ru, Pt, Pd and Rh during partial melting and crystal fractionation and which minerals are

collectors of PGEs have been investigated and reviewed extensively (Fig. 3) (O'Neill et al., 1995,

1997; Tredoux et al., 1995; Garuti et al., 1997, 1999a, 1999b; Ohnenstetter et al., 1999; Hattori et

al., 1991, 2004; Augé and Maurizot, 1995; Brenan and Andrews, 2001; Cabri, 2002; Weiser, 2002;

Righter et al., 2004; Mungall, 2005; Hanley, 2005; Tolstykh et al., 2005; Kaptiotis et al., 2006,

2009; Mungall and Naldrett, 2008; Prichard et al., 2008; Grammatikopoulos et al., 2009; Locmelis

et al., 2011; Page et al., 2012; Helmy et al., 2013; Mungall and Brenan, 2014).

The Platinum Group Minerals (PGM) possess some unique physical and chemical

characteristics. They generally display low reactivity. The state of sub-division of each metal is an

important factor when considering reactivity. Pt and Pd dissolve in aqua regia (AR) whereas Rh,

Ru, Ir and Os are relatively inert. Strong, alkaline oxidizing agents, however, can dissolve PGM.

All PGM dissolve in molten bases such as sodium, phosphorus, silicon, arsenic, antimony and lead

(Rao and Reddi, 2000). They have very high melting points, and are chemically inert to a wide

variety of substances. PGE are classified as noble metals (along with gold and silver) because of

their high resistance to oxidation and corrosion.

14

Figure 2: Location map showing mafic-ultramafic complexes associated with PGE mineralizations.

Crystal fractionation or partial melting processes is a major factor controlling PGE content.

Nevertheless, alteration and metamorphism play a main role in the formation of minerals.

Secondary PGE alloys, sulfides, sulfarsenides, or oxides/hydroxides are produced by alteration of

magmatic PGM precursors. PGE can be mobilized under hydrothermal or lateritic weathering

conditions. Supergene formation of PGM without magmatic indicators has previously been

reported from the platiniferous Au-Pd belt of Minas Gerais, Brazil (Aiglsperger et al., 2.014). The

PGM are also found in placers formed from the resituates of the sulphide minerals. A typical

example is the alloys of osmium and iridium known as osmiridium (syserkskite) or iridosmium

(nevyanskite).

Despite the great interest in the platinum-group elements, the thermo-chemical basis of their

geochemical properties remain unclear. Thermodynamic data for PGE/PGM are very limited,

largely due to experimental difficulties. The experimental data on the PGE solubility (Sassani and

Shock, 1990; Wood etal., 1992; Xiong and Wood, 2000) demonstrated that PGE solubility in

surface, low-T hydrothermal (< 500oC) and high-T post-cumulus aqueous fluids can be significant

15

at nature conditions. In addition, the presence of organic compounds such as humic substances

prevents the precipitation of the PGE where they occur as dissolved hydroxides species (Hanley,

2005). Recently, an overview of the formation of adsorption structures and surface oxides on Rh,

Pd, Ag, Cu and Pt surfaces (Seriani and Mittendor, 2008) and thermodynamic modeling have been

applied to provide evidence for the mobility of Pt and Pd in hydrothermal systems (Colombo et

al., 2007; Barnes and Liu, 2012).

Assuming that the required thermodynamic data (enthalpy, entropy, Gibbs free energy) of the

reactants and physical/chemical conditions (P, T) are available, then the mineral Gibbs free energy

can be estimated (Gupta, 2003; Cemic, 2005; Nordstrom, 2006). This paper focuses on (a)

reporting the main occurrences of PGM from primary or secondary formation (b) comparing two

of the greatest source of PGE in Greece (Skyros Island and Veria Ophiolite Massif) in order to

describe primary or secondary minerals (e.g. placer deposits) (c) the calculation of Gibbs free

energy of formation for the dominant PGM in PGE-bearing complexes worldwide, at standard

state conditions (ΔGof), using the current state of knowledge from experimental data on the PGE,

(d) the validation of the calculation of the Gibbs free energy (ΔGf) as a function of temperature,

(e) the application of the calculated Gibbs free energy (ΔGf) values to predict the stability of PGM

and (f) the discussion about environmental risks and hazards.

16

1.1. GEOCHEMISTRY OF PLATINUM GROUP ELEMENTS

It is well known that the platinum group elements are highly associated with sulphides in

mafic-ultramafic rock, thus the concentrations of PGE in sulphur-poor silicate rocks are typically

low. Advances in analytical technology gave us the opportunity to study the behavior of the group

further. Pd and Ir are thoroughly analyzed and extensively used for geochemical studies, yet Os

and Pt isotopic data has opened the possibility of characterizing lithospheric reservoirs with respect

to PGE (Crocket, 2002). The chemical and physical characteristics of these metals are unique.

Some of them are the low reactivity, Pt and Pd dissolve in aqua regia whereas Rh, Ru, Ir and Os

are relatively inert. All PGM dissolve in molten bases such as sodium, phosphorus, silicon, arsenic,

antimony and lead (Rao and Reddi, 2000). Os, Ir and Ru, the iridium-group PGE (IPGE) are

refractory, strongly compatible elements whereas Rh, Pt, and Pd, the palladium-group PGE,

(PPGE) are less compatible and more volatile (Barnes et al., 1985). In this chapter the

geochemistry of several PGE types will be described sententious.

Pt Pd Rh Ir Ru Os Au

Atomic

weight 195.08 106.42 102.91 192.22 101.07 190.23 196.97

Atomic

number 78 46 45 77 44 76 79

Density

(gcm-3) 21.45 12.02 12.41 22.65 12.45 22.61 19.3

Melting

point

(ºC)

1769 1554 1960 2443 2310 3050 1064

Electrical

resistivity

(micro-

ohm cm

at 0 ºC)

9.85 9.93 4.33 4.71 6.8 8.12 2.15

Hardness

(Mohs) 4-4.5 4.75 5.5 6.5 6.5 7 2.5-3

Table 1: Selected properties of the six platinum-group elements (PGE) (British Geological Survey, Natural

Environment Research Council, Platinum, September 2009).

17

1.2. Mafic-Ultramafic Rocks

PGE in mafic rocks are a great index for the geochemical characterization of the environment,

and the evolution of the fractional crystallization of the mafic magma. Gabbroic rocks have a

tremendous content of PGE, although mafic volcanic rocks do not host such great amount of PGE

mineralization (Cabri, 1981). As we mentioned above, Pd and Ir are great indicators and broadly

used on mafic rocks of geochemical analysis. That was the main obstacle for an expanded database.

Nowadays, new technological discoveries and analytical methods trigger the onset of extended

studies and the usage of all of the six elements of the group.

The platinum group elements are concentrated in high levels in the mafic and ultramafic rocks,

forming minerals include metallic alloys, tellurides, selenides, arsenides, sulpharsenides and

sulphides. Mantle rocks and peridotites contain very low levels except for sulphides. The primary

sulphides common to mafic-ultramafic rocks are those exsolved from crystallization of

monosulphide solid solution (pyrrhotite and petlandite) and intermediate solid solution

(chalcopyrite and cubanite) (Crocket, 2002). Olivine, pyroxene, plagioclase and oxides such as

magnetite and chromite could strongly host high concentrations of PGE, howbeit the concentration

mechanism is not yet totally understood. The levels of PGE are utterly connected with major

elements oxides derived from the magma. PGE are characterized as siderophile and chalcophile.

The siderophilic tendencies deduced by the high concentrations in iron meteorites, the high

proportion of PGE carried by iron-nickel metal in chondrites, the strong partition of PGE in the

metal phase in sulfide-bearing iron meteorites, the occurrences of many metallic alloys among the

platinum-group minerals (PGM) and the high proportion of metallic bonding characteristic of

many PGM. Chalcophile characteristics are summarized by high, ppm-level PGE concentrations

in association with Ni-Cu sulfide in mafic and ultramafic rocks and the occurrence of many sulfide

minerals of the PGE (Crocket, 1979). Due to their siderophile nature, PGE inherited information

of the early history of the earth’s formation. PGE occur at very low concentration levels in mantle

rocks, due to their extraction from the Primitive Earth into the metallic core. They offer great

potentiality for discussing global-scale processes such as core–mantle segregation processes and

core–mantle exchanges (Lorand et al., 2008).

18

1.2.1. Mantle Rocks

Transfer of PGE from the mantle into the crust is confined to specific geodynamic areas in

which important mantle-crust interactions occur. PGE are geochemical monitors of these

geological processes, and able to give information about the petrological nature and evolution of

the mantle source from which they were derived (Garuti et al., 1997). The rocks that carry these

information are orogenic and abyssal peridotites, mantle nodules and ophiolites.

Orogenic Peridotites

Orgenic massifs include domains of relatively unaltered rocks with a virtually tectonic

foliation. They are considered as a window to the mantle. Orogenetic peridotites may hot

harzburgites, lherzolites and a variety of dykes, veins and layers, usually of pyroxenite and gabbro

(Table 1). These dyke rocks are usually ascribed to reactions of basalt melts with mantle peridotite,

triggering processes capable of mobilizing PGE in a mantle environment. Broadly speaking,

orogenic peridotites could be included in a general aspect of sub-continental lithospheric mantle,

which may experience melt infiltration, melts-rock reaction and metamorphic events

(recrystallization) with final outcome the perturbation of primitive mantle PGE abundances

(Crocket, 2002). Orogenic lherzolites exhibit a specific platinum group element signature

characterized by slight positive deviations of Pd/Ir, Rh/Ir and Ru/Ir ratios from the canonical

chondritic model. Highly reproducible from one occurrence to another, the PPGE enrichment of

orogenic lherzolites is usually interpreted as a compositional feature of the primitive upper mantle.

PGE are strongly chalcophile elements in the present day metal - undersaturated upper mantle.

Thus they partition strongly into accessory BMS, being mostly excluded from major mantle

minerals, except perhaps from Cr-spinel (Lorand et al., 2008).

19

Lherzolites Harzburgites Dyke Rocks

Orogenic

peridotites

Ronda

(Spain)

Beni

Bousera

(Morocco)

Ivrea

Zone

(Italy)

Ronda

(Spain)

Beni

Bousera

(Morocco)

Ivrea

Zone

(Italy)

Websterite1 Gabbro2

Os 4.3 3.7 8.6 7.2 5.5 6.6 6.9 4

Ir 3.1 2.5 3.3 3.7 3.6 3.3 1.1 0.67

Ru 5.4 6.9 6.6 8.3 6.9 2.9 3.2 5.6

Rh 1.3 1.6 3.2 1.9 1.4 1.2 2.7 2.7

Pt 10 8 11 6 3.9 2.6 11 15

Pd 5.9 5.2 11 3.8 3.9 5.4 8.4 6.2

Au 8.2 3.2 16 5.8 2.9 41 52 10

Cu 19 17 31 6 10 9 164 55

Pd/Ir 2 2 3 1 1 2 8 9

Table 2: PGE data for lherzolites, harzburgites, websterite and gabbro dyke rocks from several orogenic

peridotites (selected data from, Crocket, 2002).

Table 2 presents some selected data from Spanish, Moroccan and Italian samples. The

abundances of PGE in these peridotites are a unique trademark of their primitive origin. In figure

3 is presented the variation and deviation of the PGE abundances of the orogenic lherzolites from

primitive mantle standard values. All the values were normalized by primitive mantle values that

were derived from, Palme and O’Neil, 2004. The approximation of the values are close to the

average of the primitive mantle, except from small fractionations of Pd and Pt. Ivrea Zone is an

example of deviation from the normal primitive mantle abundances (higher Os, Rh and Au).

1 Ronda, Spain - Baldissero/Balmuccia/Lanzo, Italy. 2 Ronda, Spain – Balmuccia, Italy – Beni Bousera, Morocco.

20

Figure 3: PGE correlation of orogenic lherzolites and primitive mantle abundances.

Noticing the high Pd/Cu ratio of the lherzolitie data, is resulting that these elements are

present in sulphides which originated as sulphide immiscible liquids in mafic silicate melts.

Furthermore it must be highlighted that the large difference between PGE and Au in the lherzolites

is explained by the smaller liquid/ silicate melt partition coefficient of Au relative to the PGE

(Crocket, 2002).

21

Figure 4: PGE correlation of orogenic harzburgites and primitive mantle abundances.

Harzburgites have approximately the same pattern as lherzolites. The notable difference is

the correlation of the IPGE and PPGE regions. Os and Ir are slightly higher and Pt, Pd and Cu are

lower respectively with the lherzolites’ abundances. Ru and Rh values are nearly identical. It is

well known that harburgite is the result of a higher degree of partial melting, therefore low

concentrations of Pd and Cu could be occurred due to lower retention of sulphide liquid in the

peridotite host. Another explanation is the dilution of metal concentrations in the sulphide,

resulting to larger masses of silicate melt (Crocket, 2002).

22

Figure 5: PGE correlation of dyke rocks and primitive mantle abundances

Dyke rocks are also present in the main lithologies of orogenic peridotite massifs. These

dykes are the the interaction of basalt melts with mantle peridotites. The major deviation of these

types are in Os/Ir ratio, the higher values of Ph, Pt and Pd than harzburgites and the remarkable

increase of the Au values, phenomenon explained by the exchange between mafic silicate melt and

peridotite in the mantle. In figure 6 are presented the correlations between the dyke rocks and the

signature from the primitive mantle.

Mantle Nodules and Abyssal Peridotites

The term mantle nodules (xenoliths) includes rocks connected with alkaline basalts and

kimberlite magmas. Garnet, spinel and plagioclase are the main minerals that dominate and they

equilibrate over a wide range of pressure and temperature conditions. Abyssal peridotites referred

to ultramafic rocks in oceanic environments ranging from mid-ocean ridges to island arc

subduction zones. These mantle rocks provide us further information about core–mantle

23

differentiation and the late accretionary history of the Earth. Because of their very high metal–

silicate partition coefficients over a wide P-T range (Luguet et al., 2003).

Ocean Environment

Orogenic

peridotites

Kane Fracture Zone (Atlantic

Ocean)

Oahu

(Hawaii)

East Pacific

Rise

Gakkel Ridge

(Arctic Ocean)

Os 2.9 1.4 1.6 2.8

Ir 2.69 4.4 3.2 2.7

Ru 5.3 5.3 5.5

Rh 0.89

Pt 5 7.4 5.3

Pd 2.23 7.1 5.2 3.2

Au 1 0.6

Pd/Ir 0.8 1.6 1.6 1.2

Table 3α: PGE data for mantle nodules and abyssal peridotites from oceanic environments.

Table 3 presents data from mantle nodules and abyssal peridotites around the world,

divided in oceanic (Table 3a) and continental (Table 3b) environments. The oceanic environments

include samples from Atlantic Ocean, from the ocean island basalt setting of Hawaii, East Pacific

Rise and the Arctic Ocean. The nodules from the island basalt setting are originated from lavas

represent mixing of melts derived by partial melting of pyroxenite and peridotite from the Pacific

lithospheric mantle (Lassiter et al., 2000). The samples from Kane Fracture zone and the East

Pacific Rise are lherzolites and harzburgites respectively. The Mid-Atlantic Ridge is referred as a

slow-spreading ridge, whilst East Pacific Rise is a fast-spreading one. On the other hand, Gokkel

Ridge is the slowest-spreading mid-ocean ridge in the world. Many of the specific spinel

lherzotites give supra-chondritic PGE/Ir ratios.

24

Table 3b: PGE data for mantle nodules and abyssal peridotites from continental environments.

The Abyssal peridotites from Gokkel Ridge provide constrains on the HSE composition of

the primitive mantle. They have consistent HSE patterns with suprachondritic Ru/Ir and Pd/Ir

ratios. Both supra-chondritic Pd/Ir ratio and supra-chondritic Ru/Ir ratio are indigenous

characteristics of the upper mantle.

Mantle nodules and mainly abyssal peridotites varying from fertile lherzolites, harzburgites

to dunites and nodules of garnet lherzolites. Lherzolites and harzburgites from West Eife, Germany

indicate that PGE can be mobile in a metasomatized mantle. Peridotites with low Pd/Ir ratio

(Pd/Ir<1) indicate contents with more volatile PGE, while the enriched harzburgites had been

3 Harzburgite from hydrous suite, Dreiser Weiher, West Eifel, Germany, Schmidt et al., 2003. 4 Spinel lherzolite nodule, Kilbourne Hole, New Mexico, United States, Crocket, 2002. 5 Garnet lherzolite nodules, South African Kimberlites, Crocket, 2002. 6 Dunite, Mugla, Turkey, Uysal et al., 2012. 7 Harzburgite from Olmani Cinder Cone, Tanzanian Craton, Crocket, 2002.

Continental Environment

Orogenic

peridotites

West Eifel

(Germany)3

Kilbourne

Hole (U.S.A)4

South African

Kimberlites5

Mugla

(Turkey)6

Tanzania

Craton7

Os 4.55 2.7 3.5 4

Ir 5.46 3.2 5.8 5 6.7

Ru 8.22 6 9.5

Rh 2.94 2

Pt 12.4 5 6.5

Pd 5.74 4.5 3.4 3 4.7

Au 0.79 0.51 0.74

Pd/Ir 3.4 1.4 0.6 0.6 0.7

25

Figure 6: PGE correlation of oceanic environment mantle nodules and abyssal peridotites with primitive mantle

abundances.

strongly affected by metasomatic processes providing the component with high PGE contents

(Schmidt et al., 2003). Australian spinel lherzolites give weakly negative slopes from from the

IPGE to the PPGE regions. Furthermore the Rh abundance are relative to primitive mantle. The

Garnet lherzolites from Africa (Matsoku and Thaba Putsoa, Lesotho) have steeper negative slope

from IPPE to PPGE and the Os and Ir elements have higher abundances relative to primitive

mantle. Peridotites from Australia and Tanzania are similar in PGE, both with sub-primitive mantle

Os and weak negative slopes from IPGE to PPGE regions (Crocket, 2002). Harzburgites from

Mugla, Turkey indicate that these rocks were depleted in varying degrees. Pt and Pd are highly

depleted during partial melting, resulting in the relative enrichment of Os and Ir in the mantle

residue. However, dunites show significant enrichments in their PGE concentrations.

26

Figure 7: PGE correlation of continental environment mantle nodules and abyssal peridotites with primitive mantle

abundances.

This observation is in accordance with the interpretation assuming that the depleted

harzburgites and dunites were formed by second-stage partial melting of cpx-harzburgites in a

supra – subduction zone environment that produced boninitic magmas. In this stage, boninitic melt

would be enriched especially in incompatible PGEs due to the addition of incompatible PGEs

derived from the subducted slab (Uysal et al., 2012).

1.2.2. Komatiites

High magnesia rocks of mafic and ultramafic composition, is the perfect description of the old

rocks called Komatiites. They vary in peridotitic textures (e.g. spinifex) and often they have

distinct coarse-grained ultramafic and mafic cumulates. Their economic interest is significant as

the abundance of PGE concentrations are higher than in other volcanic rocks. As other peridotites,

komatiites provide a window to melting of the earth’s interior (high degree of partial melting at

27

T=1700°C) (Crocket, 2002). Komatiitic and picritic magmas are associated with many of the

world's "magmatic" Ni-Cu sulphide ore deposits. Furthermore, addition, komatiite like magmas

gave rise to platiniferous mineralization in some of the world's largest mafic/ultramafic layered

intrusions, such as the Bushveld Complex (Keayes, 1995).

Komatiites that remain S-undersaturated during ascent and emplacement may preserve most

of their original PGE concentration patterns and thereby provide the only insights into the PGE

systematics of the Archaean deep mantle. PGE concentration patterns in different types of

komatiites may constrain the melting processes in the Archaean mantle. Most komatiite flows are

S-undersaturated during emplacement and much of their crystallization history, and thus offer ideal

opportunities to study the effect of the non-sulphidic phases on PGE fractionation (Maier et al.,

2003).

Table 4: PGE data for peridotitic komitiites.

8 Spinifex-textured komatiite, and flow-top breccias, Westonaria, South Africa, Maier et al., 2003. 9 Spinifex-textured komatiitic flows, Mt. Keith, Western Australia, Dowling and Hill, 1992. 10 Spinifex-textured komatiitic flows, Belingwe greenstone belt, Zimbabwe, Zhou, 1994. 11 Spinifex-textured komatiite, and flow-top breccias, Kostomushka greenstone belt, Russia, Crocket, 2002.

Komatiites

Peridotitic

Komatiites

Westonaria (South

Africa)8

Mt. Keith

(Western Australia)9

Belingwe

greenstone belt

(Zimbabwe)10

Kostomushka

greenstones

(Russia)11

Os 1.3 2.2 0.49 1.5

Ir 1.2 1.6 0.8 1.5

Ru 2 4.8 4.4 3.9

Rh 0.4 1.4 1

Pt 9.1 9.6 8.7

Pd 3.7 7.8 11 11

Au 3.2 4 1.8

Pd/Ir 3 5.2 13.8 7.3

28

Figure 8: PGE correlation of komatiites with primitive mantle abundances.

The Komatiite field is characterized by a very small range in the Ru, Rh and Pt region in

contrast to highly variable at the IPGE and PPGE ends of the field. Rh is the most poorly

represented PGE among the analyses comprising Table 4. The Pd/Ir ration is about six. In respect

of McDonough and Sun (1995) the value for primitive mantle is 1.2. No other class of mafic-

ultramafic melts considered in the present survey approach the mantle Pd/Ir ratio as closely as

komatiites. This is expected from the high degree of partial melting required to generate komatiite

melts and the quantitative dissolution of mantle sulphides in such melts (Crocket, 2002).

The alternative that low Ir-Os contents in some komatiites are inherited from the magma

generation stage rests very much on the mineralogy of mantle host phases for PGE. With some

30% to 60% melting of mantle source rock, peridotitic komatiite partial melts will extract all PGE

hosted in sulphides (Keays, 1995). Complete extraction of sulphides (petlandite, Ni-rich pyrrhotite,

PGE sulphides) to partial melts will not generate large IPGE variations in komatiite melts.

29

The PGE content of komatiites is not an obvious guide to mineralization potential in that intra-

flow variations are small with little difference apparent in flows associated with massive sulphides

and those which are not. The metals most reflective of the source region are probably the IPGE,

and the thermal regime pertinent to partial melt generation may be significant in controlling how

much of these metals are liberated to komatiite melts (Crocket, 2002).

1.2.3. Ophiolites

All six PGE have been recorded as concentrated at parts per million (ppm) in a number of

ophiolite complexes. Occurrences of PGE concentrations in ophiolites are common in podiform

chromitite where Os, Ir and Ru may be concentrated to give negative slope chondrite normalised

patterns. If sulphur saturation of the magma occurred during chromite crystallisation then Pt and

Pd + Rh will also be concentrated in the chromitite giving positive chondrite normalised patterns

(Prichard and Brough, 2009).

Mantle harzburgite, usually tectonized, is well characterized with respect to PGE. Apparently,

the extraction of modest amounts of mafic melts from primitive mantle peridotite does not

significantly fractionate the PGE. The concentration of PGE in ophiolite cumulates is highly

variable. The most consistently mineralized facies are chromite-rich dunites in which IPGE and Pt

may be concentrated with total PGE often reaching the ppm range. Pyroxenites and gabbros may

concentrate PPGE, at least locally (Crocket, 2002). Assuming that crystal fractionation or partial

melting processes rather than alteration and metamorphism is a major factor controlling PGE

content changes in the whole-rock PGE content may provide evidence for chromite mineralization

(Economou-Eliopoulos, 1996).

30

Table 5a: PGE data for Ophiolitic Complexes.

The concentrations of PGE in the chromitites vary considerably from a few ppb to several

ppm for all PGE except Rh. The massive podiform chromitites studied all contain higher

concentrations of total PGE than the three dunites with disseminated chromite. Platinum and Pd

values greater than 300 ppb are restricted to chromitite in dunite – rich areas in the upper part of

the ultramafic unit. Ruthenium values of over 300 ppb are more widespread and, although they

overlap with values from areas where there are high Pt and Pd values, they also occur throughout

the ultramafic unit. All the higher Rh values of over 100 ppb occur in samples that contain high

values of the other PGE. High Os and Ir values occur with high Ru values. Similarly, Pt and Rh

tend to be enriched together, but there is no similar mutual enrichment of Ru with Pd or Ru with

Pt.

Studies on platinum group element (PGE) concentrations in host rocks and chromite ores have

revealed generally low PGE concentrations in mantle chromitites but higher ΣPGE in some

chromite bearing transition zone dunites. Also, a corresponding PGE enrichment was observed

12 Chromite from Elekdag Ophiolite Complex, Kastamonu, Northern Turkey, Doenmez et al., 2014. 13 Ophiolitic chromites, La Cabana Ophiolite Complex, Centinela Bajo, Chile, Ganzalez-Jimenez et al., 2014. 14 Chromites, Shebenik Ophiolite Complex, Eastern Albania, Kocks et al., 2007. 15 Podiform chromites, Al’ Ays Ophiolite Complex, Saudi Arabia, Prichard et al., 2008.

Ophiolites

Chromites

Kastanomou

(Turkey)12 Continela Bajo

(Chile)13

Shebenik Complex

(Albania)14

Al’ Ays

Complex

(S. Arabia)15

Os 66 31.3 24.2 188

Ir 63 134.4 15 312.9

Ru 188 118 56.8 396.2

Rh 10 9.9 13.8 82.6

Pt 4.7 19.7 7.5 173.9

Pd 4.7 9.5 2.8 73.7

Au 0.1 0.1 0.2 0.2

Pd/Ir 66 31.3 24.2 188

31

Figure 9a: PGE correlation of chromites in ophiolite complexes with primitive mantle abundances

with IPGE>PPGE in mantle chromitites and PPGE>IPGE in the transition zone dunites (Kocks et

al., 2007). The total PGE concentration in the chromitites ranges between 82 (in high-Cr

chromitite) and 1466ppb (in high-Al chromitite).

32

Table 5b: PGE data for Ophiolitic Complexes from Greece.

The platinum group element (PGE) and gold concentrations in concentrated chromite ores are

generally low, less than 100 ppb to a few hundred ppb. In particular, the Vourinos and Othrys

ophiolite complexes, hosted the largest chromite deposits in Greece, with exception the central

part of the former, are characterized by low PGE concentrations and Pd/Ir ratios. However, some

chromitite samples exhibit an enrichment (up to 3-5 ppm): (a) only in Os, Ir and Ru; (b) only in Pt

and/or Pd; or (c) in all PGE, which seems to be local and independent of their major element

composition, e.g. Skyros island, Central Greece, Pindos complex.

The majority of chromite samples show a relatively high content in Ir, Os and Ru, while Pt and

Pd are low resulting in primitive mantle-normalized PGE patterns with a negative slope (Fig. 10b)

and a low Pd/Ir ratio.

16 Chromitites, Pindos Ophiolite Complex, Northwestern Greece, Economou-Eliopoulos and Vacondios, 1995. 17 Spinels, Vourinos Ophiolite Complex, Northwestern Greece, Cocherie et al., 1989. 18 Chromitites, Othrys Ophiolite Complex, Eastern Central Greece, Garuti et al., 1999. 19 Chromites, Veria Ophiolite Complex, Northern Greece, Tsoupas and Economou-Eliopoulos, 2008.

Ophiolites (Greece)

Chromites

Pndos Ophilite

Complex16

Vourinos

Ophiolite

Complex17

Othrys Ophiolite

Complex18

Veria Ophiolite

Complex19

Os 54.9 50.3 23.1 1628.3

Ir 64.6 39.5 8.3 1846

Ru 144.1 102 27.4 2305.8

Rh 15.90 0.9 152.5

Pt 27.9 2.3 3.2 353.3

Pd 7.90 5.7 0.2 167

Au 5.9 8.6 4

Pd/Ir 0.1 0.1 0.02 0.09

33

Figure 9b: PGE correlation of chromites in ophiolite complexes from Greece with primitive mantle abundances.

Due to the behavior of Ni, like Os, Ir and Ru, it is compatible with most early crystallizing

phases and trend to be enriched in olivine and chromite-bearing rocks. On the other hand, CuO

like Pt and Pd, is an incompatible element and, therefore, the Pd/Ir and Ni/Cu ratios have been

proposed as an evidence of the magma fractionation. Host and barren dunites, and harzburgites,

which are strongly depleted (clinopyroxes are absent or less than 1%) show low PGE Cu and S

content. Typically, the variation of all chalcophile elements within harzburgite is more or less

homogeneous, whereas dunites show a significant variability, but a correlation between Ni and Cu

and PGE content is not obvious (Economou-Eliopoulos, 1996).

34

1.3. CHROMITE DEPOSITS

Chromite is the main PGE collector ans the majority of chromite ores show a relatively

high Ir, Os and Ru contents, while Pt and Pd are low resulting in chondrite (C2)-normalized PGE

patterns with a negative slope (Fig. 10) and a low Pd/Ir ratio. More specifically, the PGE total for

most of the chromitite samples from the Vourinos complex are less than 200 ppb. In contrast, some

chromitite samples from the Pindos complex, which are Rh, Pt and Pd enriched and low Ir, Os and

Ru contents, show smooth sloped to positive slope PGE-patterns in ophiolite complexes

(Economou-Eliopoulos and Vaconndios 1995, Tarkian et al. 1996, Prichard et al. 2008;

Grammatikopoulos et al. 2009). The Skyros island in the Aegean (up to 3 ppm), the Veria (up to

25 ppm) (Tsoupas and Economou – Eliopoulos, 2008) and Vourinos (up to 3 ppm)

(Konstantopoulou and Economou-Eliopoulos 1991).

Figure 10: Chondrite-normalized PGE-patterns (sample/C1 chondrite) for chromite ores hosted in ophiolite

complexes from the Vourinos and other large chromite deposits (a), IPGE-enriched ores (b) and PPGE-enriched ores

(after Economou-Eliopoulos et al. 2013).

35

2. METHODOLOGY

2.1. MATERIALS AND METHODS

2.1.1. Preparation of the Samples and Analytical Methods

Polished sections prepared from the chromite occurrences of the Skyros and Veria areas were

examined by reflected light microscopy and scanning electron microscope. Quantitative analyses

were carried out at the University of Athens, Department of Geology, using a JEOL JSM 5600

scanning electron microscope, equipped with automated OXFORD ISIS 300 energy dis-persive

analysis system. Analytical conditions were 20 kV accelerating voltage, 0.5 nA beam current, <2

μm beam diameter and 50 second count times.

2.1.2. Skyros Island

The ultramafic massifs of the Skyros Island, belong to the Pre-Cretaceous nape. Skyros

dismembered ophiolitic massif composed by a tectonic melange and ophiolitic rocks, which are

over-thrusted on to the Pelagonian massif. The predominant petrological types are mainly

serpentinized harzburgites and to a lesser extend lherzolites and dunites. Broad mantle shear zones,

are characterized by the presence of fish - shaped serpentinite bodies, commonly containing

juxtaposed blocks with distinct compositional and micro-structural characteristics. Small (a few

thousand tons) massive and schlieren type chromite bodies are hosted within serpentinized dunites.

Mineralogy

Pentlandite, pyrrhotite, chalcopyrite, laurite and rarely native iron were found as inclusions in

chromites. The silicate matrix is entirely serpentinized and carries numerous grains of

heazlewoodite and millerite. A distinct and more complex sulphide assemblage consisting of

pentlandite, pyrrhotite, pyrite, Cu – Fe – and Cu sulphides, tetraauricupride and sperrylite is

showed in the magnetite ores. The serpentinized silicate matrix is abundant with tetraauricupride,

less frequent also small grains of sperrylite were found (Tarkian et al., 1992). The compositions

of chromitites from the entire Skyros island area fall within both high – Cr and – Al types

(Economou-Eliopoulos, 1996) (Fig. 11).

36

Figure 11: Back – scattered electron image of chromite (chr) from the Skyros island.

Inclusions of chlorite (chl) and serpentine (spt) in chromite.

Numerous polished thin – sections showed a dominance of the high – Al variety in both poor

and enriched PGE chromitites. Although the composition of chromite is commonly homogeneous,

dark gray and light colored phases can be distinguished (Fig. 12a, c) close to highly brecciated

zones. A salient feature of PGE – enriched samples is the alteration of chromite (the average

Cr/(Cr+Al) ratio of unaltered chromite is 0.56 and Mg/(Mg+Fe2+ is 64) to ferrian chromite (the

Cr/(Cr+Al) ratio ranging from 69 to 0.76 and Mg/(Mg+Fe2+) from 57 to 51 (Fig. 12c). All silicate

minerals occurring in the space between chromite grains have been entirely altered to Fe – poor

serpentine and chlorite. Ru – Ir – Os – minerals were identified within silicate matrix as interstitial

secondary minerals (Fig. 12c, d, e). Both Ru – Ir – Os alloys and sulfides contain significant Fe,

Ni and Cr amounts and a wide compositional variation (Table 6; Fig. 12d, e). In addition sulfides

contain Pt, Sb and As. Heazlewoodite (Ni3S2) occurs commonly within fractures and the

serpentine-chlorite matrix (Fig. 13a). Inclusions of native iron had also been found (Fig. 13b).

37

Figure 12: Back – scattered electron images of strongly fragmented chromite (chr) associated with garnet (grt) and

serpentine (spr) (Fig. 12a, b), IPGM in a matrix of serpentine associated with chromite/Fe – chromite (Fe – chr) (Fig.

12c, d, e) and alloys along cracks of chromite and Fe – chromite (Fig. 12f) from Skyros Island. Heazlewoodite is also

common (Fig. 12b).

38

Table 6: Representative microprobe analyses of PGM (figure 12) of PGE – rich chromites from Skyros Island, Greece.

Figure 13: Back – scattered electron image of Heazlewoodite (Ni3S2)

as inclusion in chloritic matrix (chl) in contact with magnetite (mt).

Laurite is the only PGM found in chromites, as euhedral to subhedral crystals exclusively

as single inclusions (Fig. 14). The Ir contents are moderate, whereas Os values are variable and

high (Table 7). Their compositional variation corresponds to those reported by Tarkian et al.

(1991).

Representative Microprobe Analyses of PGM (Figure 12)

Fig. 12d, 1 Fig. 12d, 2 Fig. 12d, 3 Fig. 12d, 4 Fig. 12e, 1 Fig. 12e, 2 Fig. 12e, 3

Os 5.88 7.96 15.49 13.05 44.8 38 80.9

Ir 6.31 5.66 18.51 27.21 7.2 13.2 2.2

Ru 53.35 48.48 38.23 30.59 37.8 41.2 10.3

Pt 1.6

Pd 2.74 4.37 3.61 4.59

Ni 0.68 1.17 1.21 1.03 6.7 4.3 3.3

Fe 0.95 0.73 0.67 0.51 1.5 1.1 0.8

Cr 0.38 0.34 0.34 0.44 0.9 1.1 0.6

As 2.11 2.76 5.81 10.01

S 26.71 27.15 14.78 11.85

Total 99.11 99.14 99.28 100.42 98.9 100.5 98.1

39

Table 7: Representative microprobe analyses of Laurite, Skyros Island, Greece

Figure 14: Back – scattered electron image of laurite crystal as inclusion in

chromite, Skyros Island, Greece (after Trakian et al., 1992)

At Skyros all investigated samples from the magnetite contain numerous irregular grains

of tetraauricupride. The mineral mainly occurs in cracks along magnetite grains or at the grain

boundaries of magnetite and serpentinized matrix. Occasionally small inclusions in magnetite were

also found. The ideal formula of the mineral is AuCu. Although, small amounts of Fe and Ag were

detected. Sperrylite is the only PGM found as inclusions in magnetite (Fig. 15). It occasionally

occurs as intergrowths with chalcocite as well as in the serpentinized silicate matrix of the

magnetite crystals. It occurs as euhedral to subhedral crystals. Small amounts of Fe but no other

elements were detected in sperrylite (Table 8).

Representative Microprobe Analyses of Laurite

1 2 3 4 5

Os 23.5 16.8 8.2 10.1 12.1

Ir 3.7 6.5 10.5 9 8.9

Ru 38.1 42.2 46 46.9 43.8

As 1.5

S 33.9 34.1 33.1 34.5 34.8

Total 99.2 99.6 99.3 100.5 99.6

40

Table 8: Representative Microprobe Analyses of Sperrylite, Skyros Island,

Greece

Early magmatic phases of laurite formed by direct crystallization from basaltic magmas

are inclusions in unaltered chromites. Also, the presence of Os-, Ir, Ru- and Rh-bearing PGM

within chromitites that occur in some ophiolite complexes have been attributed to in situ alteration

or/and re-mobilization and re-deposition of PGE during post-magmatic processes (e.g. retrograde

hydrothermal metamorphism). Thus, besides the typical laurite grains found as inclusions within

chromite and the Pt and Pd- enrichment in chromitite due to magmatic processes (Prichard et al.,

1996; Economou-Eliopoulos, 1996) the mineralogical and texture characteristics of relatively high

IPGE enrichment in chromitites may be originated to post-magmatic processes. They may cover

an extended period of deformation, including the ductile asthenospheric mantle flow and

culminated in shallow crustal brittle deformation. The composition of primary PGM, mostly

laurite, have been substantially modified by metasomatic fluids, generating Fe-Cr-Co-Ni-bearing

Os-Ir-Ru minerals (Tsoupas and Economou-Eliopoulos, 2005, 2007). Furthermore, it seems likely

that the most targeted locations for the chromitite-IPGM association are small chromite

occurrences found along shear zones within ophiolite complexes (Economou-Eliopoulos, 2009).

Representative Microprobe Analyses of Sperrylite

1 2 3

Pt 56.2 55.9 56.3

Fe 0.9 0.3

As 42.8 43.3 44

Total 99.9 99.5 100.3

41

Figure 15: Inclusion of sperrylite (sp) in intergrowth with chalcocite (cc) in magnetite.

Reflected light. Scale bar = 15#m. (A) and Secondary electron image of sperrylite in the

serpentinized silicate matrix (B), Skyros Island, Greece.

42

2.1.3. Examples from the Literature

Veria Ophiolite Massif

The Veria ophiolite massif has been thrust onto the Pelagonian Massif. The sequence is, in

turn, partly thrust over the Veria peridotites. Highly melt – depleted mantle harzburgite, and crustal

magmatic rocks, including a well-developed cumulate series (pyroxenites, gabbros),

plagiogranites (quartz – diorites, tonalities) and volcanic rocks (metabasalts, metadiabases,

metarhyolites) are constituting the main lithologic types of Veria ophiolite. The petrological,

mineralogical and geochemical characteristics of the Veria ophiolite are typical of ophiolites

associated with a SSZ geotectonic environment (Economou and Economou, 1986). Broad mantle

shear zones, several hundreds of meters in width, dominate the mantle sequence of the Veria

ophiolite. The shear zones are characterized by the presence of fish-shaped serpentinite bodies,

commonly containing juxtaposed blocks with distinct compositional and micro-structural

characteristics. Small lensshaped bodies of massive chromitite are common along zones of

shearing. The massive peridotite is mostly serpentinized. Diopside with 1.3 wt.% Cr2O3 and

andradite with 1.1 wt.% Cr2O3 are commonly associated with serpentinized olivine, whilst

orthopyroxene is wholly replaced by low-Fe bastite. Awaruite (often Co-bearing) is a common

mineral in the serpentinized peridotite. Chromite has been transformed in to porous ferrian–

chromite, with few surviving relict grains of chromite (Tsoupas and Economou-Eliopoulos, 2008)

(Fig. 16).

The compositions of chromitites from the entire Veria area fall within both high-Cr and -Al

types. Detailed investigation of numerous polished thin-sections from the area of Galaktos (Veria)

by SEM/EDS showed a dominance of the high-Cr variety in both PGE-poor and PGE-enriched

chromitites. In addition, the cores of large chromite crystals exhibit a significant trend of increasing

Fe and Cr with decreasing Al and Mg contents even on the cm-scale. Laurite (RuS2) was identified

as small inclusions within unaltered chromite grains in the chromitites (Fig. 17a). It mostly occurs,

however, as relicts containing very small irregular grains of irarsite within a matrix of Ru–Os–Ir

alloy/oxide that constitutes the greater part of extremely coarse PGM grains. Detailed investigation

43

of the larger of two large PGM grains showed that the composition of the laurite is rather constant,

whereas the Os–Ru–Ir–Fe alloys exhibit a wider compositional variation (Table 9).

Figure 16: Back – scattered electron image of fragmented chromite (chr) with an abundance of

PGM grains, Veria, Greece

TABLE 9. Representative microprobe analyses of PGM from PGE-rich chromites of the Veria

ophiolite Complex (Tsoupas and Economou-Eliopoulos, 2008).

44

Α major PGM is also irarsite. It is associated with laurite in the large PGM grain (Fig. 17a)

and as intergrowths with platarsite – hollingworthite and garnet. In addition, in the serpentine

matrix were found fine intergrowths of irarsite and garnet. The irarsite hosted within serpentine

appears to be crosscutting the garnet. Awaruite (Ni3Fe), containing small amounts of Co and Cr,

heazlewoodite (Ni3S2), pentlandite and maucherite commonly occur within fractures and the

serpentine–chlorite matrix. Although they are commonly fine-grained, heazlewoodite cores

sometimes occur within awaruite, suggesting a gradual transformation of the former to the latter.

The mineralogical characteristics and textural relationships between PGM and hosting

chromitites from the Skyros ophiolites are similar to those from Veria in terms of: (a) the IPGE

enrichment, although it is much higher in the Veria chromitites (up to 25 ppm), (b) the association

of the IPGE-enrichment with shearing zones, (c) the presence of laurite as inclusions within

chromite, and Ir-group phases interstitial to strongly fragmented chromite grains, (d) the

significant Ni, Cr, Fe and As contents in both Ru, Ir, Os-sulfides and alloys, and their wide

compositional variation, (e) the association of IPGM with Cr-garnet and low-Fe serpentine

(Economou-Eliopoulos, 2007).

Platinum-group minerals (PGM) in nature is a research topic for many authors and various

hypotheses have been expressed to account for their decoupling (Auge, 1985, 1988; Prichard et

al., 1986; Slansky et al., 1991;Auge & Legendre, 1994; Garuti& Zaccarini,1997; Garuti et al.,

1999; Ohnenstetter et al., 1999; Tarkian et al., 1996; Cabri, 2002; Weiser, 2002; Tolstykh et al.,

2005; Kaptiotis et al., 2006, 2009; Prichard et al., 2008; Grammatikopoulos et al., 2009).

Experimental data on stability limits of laurite and Ru–Os–Ir alloys as functions of temperature in

basaltic magmas suggest that they could be primary magmatic minerals, crystallizing at high

temperature from silicate magma (Brenan and Andrews, 2001). Also, secondary mobilization and

redistribution of PGE appears to be more extensive than initially was considered to be. Although

platinum-group elements are considered to be inert noble metals, primary PGM, mostly laurite,

has been transformed to Fe–Cr–Co–Ni-bearing Fe–Os–Ir–Ru alloys and oxides, as exemplified by

these in small chromite occurrences found along shear zones in the Veria ophiolite complex (Table

9; Fig. 17), weathered chromitites and a laterite cover of the Niquelândia complex (Garuti et al.,

2012).

45

Figure 17: Back-scattered electron images of (A): A typical fine laurite inclusion within

unaltered chromite. (B): An extremely coarse PGM grain and (C): remnants of laurite and

Ru-Os-Ir-oxides within a large laurite grain, (D): Pt-Rh-irarsite between chromite grains, all

from the Veria ophiolite complex (Greece), (E): (Pt,Rh,Cu)S from the UG2 horizon and (F):

(Pd, Pt)S from the Merensky reef, Bushveld complex. Abbreviation: irs = irarsite; chr =

chromite; Fe-ch = iron chromite; Po = pyrrhotite; srp=serpentine; grt = garnet; cpx =

clinopyroxene (Tsoupas and Economou-Eliopoulos, 2008).

46

Alaskan Type Intrusions Placer Deposits

It is well known that the zoned mafic-ultramafic igneous complexes of the Ural-Alaska type

are the lode sources of the most significant economic platinum-group element (PGE) placer

resources and reserves (Tolstykh et al., 2005). Large Alaskan - type complexes exposed along SE

Alaska, in the Urals and south-central British Columbia. The majority of placer mining was carried

out in Russia, particularly in the Urals. It has been concluded that the formation of big Ir-group

PGE grains can be formed at relatively high temperatures rather than a secondary origin. Laurite

(RuS2), from placer deposits, shows a variety of morphologies, including euhedral grains with

conchoidal fractures and pits, and spherical grains. Isoferroplatinum, Os-Ir-Ru alloys, laurite, and

sperrylite occur in both PGE placer deposits related to the zoned Ural-Alaskan type and ophiolite

complexes. Pt – bearing minerals are also dominant in placer deposits, whilst Os-Ir-Ru-minerals

are in a larger proportion in those derived from ophiolite complexes. Furthermore, Ural-Alaskan

type system has an Os – Ir trend and ophiolite massifs show a Ru trend, which has been attributed

to evolution of the ore-forming system of parent complexes (Economou-Eliopoulos, 2009).

PGM-bearing placer mineralogy and its economic potential are primarily defined by the nature

of the presumed lode sources. The genesis of primary magma defines the geochemical

specialization of platinum-group mineral placers and their final formation. Pt-Fe alloy placers

associated with complexes of Ural-Alaska type have been the most important type of PGE placers.

PGE-bearing placers associated with zoned Ural-Alaska complexes are dominated by Pt–Fe

alloys (Fig. 18). Among the refractory PGE minerals, either native osmium or iridium can occur.

The quantity of PGM related to a later paragenesis can also be present in a placer, e.g., sperrylite,

cooperite, stibiopalladinite, or isomertieite. The relative amounts of these minor PGE occurrences

depend on the intensity of postmagmatic hydrothermal alteration of bedrock sources. Pt–Fe alloys

from placers related to Ural-Alaska type complexes have wide ranges of Fe concentrations, which

can vary from native platinum with 3–5 wt.% Fe to Fe-rich platinum, where Fe content reaches 12

wt.%. Variations in the concentration of Fe in Pt–Fe alloys frequently can be observed within a

single placer.

47

Table 10: Representative Microprobe Analyses of Pt – Fe and Os – Ir – Ru alloys in Alaska type systems (Data

collected from Tolstykh et al., 2005)

Table 11: Representative Microprobe Analyses of Pt – Fe and Os – Ir – Ru alloys in Ophiolite type systems (Data

collected from Tolstykh et al., 2005)

20 Levtyrynyvayam River placer (Gal'moenan massif). 21 Penisty Creek placer (Gal'moenan massif). 22 Inagli River placer (Inagli massif). 23 Salmon River placer (Red Mountain, Goodnews Bay, Alaska). 24 Zolotaya River placer (Kurtucshibinsky ophiolite belt). 25 Gar'-1.2 rivers placers (Ust'- Depsky ophiolite belt). 26 Ol'khovaya River placer (Ust'- Kamchatsky ophiolite massif). 27 Suenga River placer.

Representative Microprobe Analyses of PGM-alloys in Alaska type systems

Pt-Fe20 Pt-Fe21 Pt-Fe22 Pt-Fe23 Os-Ir-Ru24 Os-Ir-Ru26 Os-Ir-Ru27

Pt 94.76 87.23 77.59 89.8 0.56 0.91 13.34

Fe 4.62 9.59 8.19 8.08 0.88

Ni 0.14

Cu 0.33 1.26 1.15 0.45

Ru 0.07 0.66 0.05 3.66 1.38

Rh 0.99 0.91 0.45 3.36

Pd 0.26 0.39

Os 0.07 0.42 99.54 70.87 17.62

Ir 1.09 10.85 21.19 62.97

Total 99.64 99.9 99.88 99.53 100.1 99.94 99.99

Representative Microprobe Analyses of PGM-alloys in Ophiolite type systems

Pt-Fe24 Pt-Fe25 Pt-Fe26 Pt-Fe27 Os-Ir-Ru28 Os-Ir-Ru30 Os-Ir-Ru31

Pt 88.32 86.1 82.12 89.44 2.12 0.17 5.03

Fe 8.8 7.85 10.11 8.95 1.76 0.11 0.2

Ni 1.43

Cu 0.41 0.95 0.45 0.27 0.14

Ru 0.09 1.05 2.53 35.29

Rh 0.42 1.52 2.27 0.45 0.06 1.88

Pd 1.19 0.29

Os 0.07 0.09 4.61 53.95 28.6

Ir 0.28 3.06 3.61 90.22 43.17 28.85

Total 99.42 99.64 100 99.49 99.96 99.92 99.85

48

Figure 18: Modal distributions of PGM in the placers derived from Ural-Alaska type complexes (Tolstykh et al., 2005).

49

Os–Ir–Ru alloys in placers connected to the Ural-Alaska type are usually inclusions in Pt–Fe

alloys, and infrequently they are as discrete grains in placer concentrates. In contrast, in the placers

related to ophiolite, all mineral species of the system Os–Ir– Ru are characteristically seen in the

mineral suite (Table 10 – 11). In the placers derived from the Ural-Alaska type complexes, there

are also two types of magmatic equilibrium paragenesis: Pt–Fe Alloy-Osmium and

Isoferroplatinum-Iridium paragenesis (Fig. 19) (Tolstykh et al., 2005).

Figure 19: Microphotographs of PGM showing different paragenesis from placers related to the Ural-

Alaskan-type: (A) Pt-Fe alloy-osmium; (B) isoferroplatinum-iridium; and from placers related to

ophiolite-type ultramafic rocks: (C) osmium-iridium; (D) isoferroplatinum-ruthenium. (images from

Tolstykh et al., 2005).

50

2.2. PLATINUM GROUP MINERALS

Platinum Group Elements are participating in the forming of minerals. Many of the compounds

are involving the presence of Sulfur (S), Selenium (Se), Bismuth (Bi), Tellurium (Te) and Arsenic

(As). Elements like Lead (Pb), Copper (Cu), Nickel (Ni), Iron (Fe), Tin (Sn) and Gold (Au) are

incompatible but playing an extensive role on the stability of PGE. Furthermore, many minerals

are consisted of native elements (Ruthenium, Osmium). Inside the minerals, or in the matrix of the

rock, alloys are present. The most abundant alloy are the ones that contain Fe, such as

Tetraferroplatinum (PtFe) and Isoferroplatinum (Pt3Fe). In the below table (Table 12) are gathered

all the known Platinum Group Minerals (PGM) recognized by the Commision on New Minerals

and Mineral Names (CNMMN), International Nineralogical Association (IMA), with few

exceptions.

TABLE 12. Platinum Group Minerals (ideal formulas) Formula Name Formula Name

Ru Ruthenium PdTe Kotulskite

RuS2 Laurite Pd3Pb2S2 Laflammeite

RuAs2 Anduoite PdNiAs Majakite

RuAsS Ruarsite Pd3Ni2As3 Menshikovite

RuAs Ruthenarsenite PdTe2 Merenskyite

Rh Rhodium Pd11(Sb, As)4 Mertieite I

Rh2S3 Bowieite Pd8(Sb, As)3 Mertieite II

RhAs Cherepanovite PdBiTe Michenerite

CuRh2S4 Cuprorhodsite (Cu, Pd)7Se5 Oostreboschite

FeRh2S4 Ferrorhodsite Pd5(Sn, As)2 Palarstanide

RhAsS Hollingworthite Pd2As Palladoarsenide

Cu3PbRh8S16 Konderite Pd2As0.8Bi0.2 Palladobismutharsenide

Rh12As7 Polkanovite (Pd, Rh)2As Palladodymite

Rh17S15 Prassoite Pd17Se15 Palladseite

(Rh, Pd)2As Rhodarsenite Pd2SnCu Paolovite

Rh3Pb2S2 Rhodplumsite Pd3Pb2S2 Plumbopalladinite

Pd Palladium PdBi Polarite, Sobolevskite

Pd8As2.5Sb0.5 Arsenopalladinite PdHg Potarite

(Pd, Hg)3As Atheneite Ag4Pd3Te4 Sopcheite

Pd3Sn Atokite Pd5Sn2Cu Stannopalladinite

Pd3SbTe4 Borovskite Pd5+xSb2-x Stibiopalladinite

Pd2SnCu Cabriite Pd8As3 Stillwaterite

PdBi2 Froodite PdSb Sudburyite

Pd11Sb2As2 Isomertieite (Pd, Cu)3Sn Taimyrite

Pd3-xTe Keithconnite IrSbS Tolovkite

(Pd, Ag)3+xTe Telargpalite Pt Platinum

Pd9Te4 Telluropalladinite PtS Cooperite

Pd3HgTe3 Temagamite (Pt, Pd)Bi3S4-x Crerarite

PdSbTe Testibiopalladite PtIn2 Dmiaoite

Pd8Hg3Se9 Tischendorfite PtCuAsS2 Daomanite

51

Pd(Bi, Pb)2 Urvantsevite PtFe0.5Ni0.5 Ferronickelplatinum

Pd16S7 Vasilite (Pt, Pd)4Sb3 Genkinite

Pd16Se2 Verbeekite PtSb2 Geversite

(Pd, Pt)3(As, Sb, Te) Vincentite PtCu Hongshiite

PdS Vysotskite Pt3Fe Isoferroplatinum

Pd3Pb Zvyagintsevite (Cu, Fe)4(Pt, Pb)4NiS8 Kharaelakhite

Os Osmium Pt5Se4 Luberoite

OsS2 Erlichmanite CuPt2S4 Malanite

OsAs2 Omeiite PtBiTe Maslovite

OsAsS Osarsite PtTe2 Moncheite

Ir Iridium PtSn Niggliite

IrBiS Changchengite PtAsS Platarsite

Ir3Fe Chengdeite Pt3Sn Rustenburgite

Ir3Te8 Gaotaiite PtAs2 Sperrylite

Cu3PbIr8S16 Inaglyite PtSb Stumpflite

IrAs2 Iridarsenite PtSe2 Sudovikovite

Ir2S3 Kashinite Pt9Cu3Sn4 Tatyanaite

IrBiTe Mayingite PtFe Tetraferroplatinum

IrTe2 Shuangfengite PtFe0.5Cu0.5 Tulameenite

Pt3In Yixunite

52

2.3. PGM IN LAYERED INTRUSIONS AND OPHIOLITE COMPLEXES

Sulfides and alloys of the IPGE may occur as small discrete inclusions within chromite,

commonly as polyphase aggregates with laurite, erlichmanite and irarsite. The Pt–Fe alloys among

other PGE mineral can be found in diverse settings ranging from orogenetic zones to layered

complexes, such as the Merensky Reef, with a low proportion of base-metal sulfides (Ohnenstetter

et al., 1999). Also, a salient feature of platinum-group minerals from nature complexes is the

presence of secondary PGE-minerals (Os–Ir–Ru–oxides) suggesting mobilization and

redistribution of PGE subsequently of their deposition stage, as exemplified by those in small

chromite occurrences located along shear zones in the Veria and Skyros ophiolite complexes (Fig.

17), in placer deposits (Tolstykh et al., 2005; in the oxidized zone of the Great Dyke of Zimbabwe

layered intrusion (Oberthur, et al., 2013), in laterite deposits (Bowles, 1986; Aiglsperger et al.,

2014) and elsewhere.

Platinum-group minerals (PGM) can be classified into two subgroups: the more Os-, Ir- and

Ru-rich or IPGE (Ir-goup) showing compatible behavior (partition or distribution coefficient

between solid and magma ≥ 1) and Pt, Pd-rich or PPGE (Pt-group) assemblages, showing

incompatible behavior (partition or distribution coefficient between solid and magma ≤ 1) and

hence enrichment as a function of the differentiation degree (Barnes and Picard, 1993). Although

laurite (RuS2), erlichmanite (OsS2) and irarsite (IrAsS) have been identified as small inclusions

within chromite grains in layered intrutions, they are mostly dominated in ophiolite complexes

(Talkington et al.,1984; Auge, 1985; Prichard and Tarkian, 1988; Tarkian et al, 1992; Cabri et al.,

1996; Melcher et al, 1997; Garuti et al., 1997, 1999a, 1999b; Ohnenstetter et al., 1999; Hattori et

al., 1991, 2004; Auge and Maurizot, 1995; Cabri, 2000). Alloys of the IPGE may occur as small

discrete inclusions within chromite, commonly as polyphase aggregates with laurite, erlichmanite

and irarsite. The Pt–Fe alloys among other PGE mineral can be found in diverse settings ranging

from orogenetic zones to layered complexes, such as the Merensky Reef, with a low proportion of

base-metal sulfides (Ohnenstetter et al., 1999).

Platinum and Pd-tellurides, such as moncheite (PtTe2), merenskyite (PdTe2), melonite

((Ni,Pd)Te2) and kotulskite ((Pd,Ni)Te), and bismuth-tellurides ((Pt,Pd)(TeBi)2) associated with

Fe–Ni–Cu sulfide deposits and porphyry-Cu-Au systems are common in nature (Figs. 20 – 23).

Thus, a high proportion of major platinum-group minerals laurite [(Ru,Os,Ir)S2] and

53

rutheniridosmine (Ru–Ir–Os alloy) is found in large chromite deposits related to ophiolite

complexes, whereas Pt, Pd and Rh-bearing PGM are the most abundant PGM in layered intrusions

and small chromite occurrences related to ophiolites occurrences (Figs. 24 – 26). Furthermore a

decoupling of (Pt,Pd)-sulfides and (Pt,Pd)-tellurides was recorded Fe–Ni–Cu sulfide deposits

while (Pd,Pt)-tellurides are the only PGM identified in porphyry-Cu-Au deposits.

54

Proportion of PGM in layered intrusions

Figure 20: Pie chart illustrating the proportion of platinum-group minerals in the Merensky Reef, dominated by Pd-

Pt tellurides and UG2, dominated by Pt-Pd sulfides, of the Bushveld Complex, South Africa. Data was taken from

Goldel et al., 2007; Penberthy et al., 2000 respectively.

55

Figure 21: Pie chart illustrating the proportion of platinum-group minerals in the Platreef and Rustenburg Layered

Suite, dominated by Pd-Pt(±Bi) tellurides, of the Bushveld Complex, South Africa. Data was taken from Hollwell and

MacDonald, 2007 and Cawthorn et al., 2002 respectively.

56

Figure 22: Pie chart illustrating the average proportion of platinum-group minerals in the Bushveld Complex, South

Africa and the proportion of platinum-group minerals in the J-M Reef in the Stillwater Igneous Complex. The average

values were collected from specific localities (Amandelbult mine, Union mine, Bafokeng-Rasimone mine, Impala

mine, Lebowa mine and from UG2 chromite layer, Merensky Reef and Platreef) around the Bushveld Complex. The

dominating platinum-group minerals are Pd-Pt(±Bi) sulfides. Data taken from Goldel et al., 2007; Penberthy et al.,

2000; Hollwell and MacDonald, 2007; Cawthorn et al., 2002 and Zientek et al., 2002 respectively.

57

Figure 23: Pie chart illustrating the proportion of platinum-group minerals in the Main Sulphide Zone (Great Dyke,

Zimbabwe) and McCreedy East Deposit (Sudbury Igneous Complex, Canada), dominated by Pd-Pt(±Bi) tellurides.

Data from Oberthur, 2002 and Dare et al. 2011.

58

Proportion of PGM in Mesozoic ophiolite complexes

Figure 24: Pie chart illustrating the proportion of platinum-group minerals in the Santa Elena Ophiolite Complex

(Costa Rica) and Al’ Ays Ophiolite Complex (Saudi Arabia), dominated by laurite and PGE Ni-Cu alloys respectively.

Data was taken from Zaccarini et al., 2010 and Prichard et al., 2008.

59

Figure 25: Pie chart illustrating the proportion of platinum-group minerals in the Bulqiza and Tropoja Massifs

(Albanian Ophiolite Complex), dominated by laurite. Data was taken from Ohnenstetter et al., 1991.

60

Figure 26: Pie chart illustrating the proportion of platinum-group minerals in the Pindos Ophiolite Complex (Greece)

and Faryab Ophiolite Complex (Iran), dominated by Pt-Fe alloys and laurite respectively. Data was taken from

Prichard et al., 2008 and Rajabzadeh and Moosavinasab, 2013.

61

Average proportion of PGM in layered intrusions and ophiolite complexes

Figure 27: Pie chart illustrating the average proportion of platinum-group minerals in layered intrusions and ophiolite

complexes from around the world, dominated by Pd-Pt(±Bi) tellurides and laurite respectively. Data are result of the

average values of the above mentioned pie charts.

62

Recently, elevated Pd and Pt contents, reaching values over 5 ppm were determined in

certain porphyry Cu-Au deposits. British Columbia, Colorado and Late Cretaceous to Miocene

porphyry Cu deposits, extending from Romania, through Serbia and Bulgaria to Greece are the

most important porphyry intrusions. The Pd-telluride, merenskyite occurs mostly as inclusions

or/and at margins of chalcopyrite and bornite or forms intergrowths with Pd-Pt-Bi- and Ag-

tellurides, and are considered to be encouraging for Pd and Pt as by-products, and porphyry

deposits a good target for Pd & Pt exploration (Economou-Eliopoulos, 2005).

In a comparison of the PGM distribution on the basis of the pie charts (Fig. 27) it is clear

that average proportion of laurite (6%) in layered intrusions is much lower than in ophiolite

complexes (43%), reaching up to 75% in the Tropoja ophiolite, Albania (Fig. 25). (Pt-Pd±Rh±Ir)-

sulfides are mostly associated with layered intrusions and they comprise about 67% in the UG2

horizon of the Bushveld complex (Fig. 20) which is comparable to the J-M Reef of Stillwater (Fig.

22), while the proportion of laurite is 6% (Fig. 27). Tellurides are stable in a wide variety of

geological environments. In the Merensky reef of the Bushveld Complex, Pt and Pd-tellurides

comprise about 60%of the modal abundance of PGM while the proportion of laurite is 5% only

(Fig. 20). However, in the Pindos ophiolite complex the proportion of laurite is 1% , Os-Ir-Ru-Rh

alloys 18%, (Pt-Fe±Ni±Cu±Au)-alloys is 30% and (Pt-Pd±Cu±Pd)-selenides is 16% (Fig. 26).

Such a PGM enrichment comparable to that in layered intrusions is a common feature of relatively

small chromite occurrences, which are located mostly at the uppermost parts of the mantle and/or

the lowest parts of the crust sequence (Prichard et al., 1994; Economou-Eliopoulos et al., 1996).

In the Sudbury complex, Pd is an important by product, and more than 90% of Pd occurs as

bismuth-telluride minerals (Fig. 23).

63

2.4. CALCULATING PROCESS

Minerals have a Gibbs free energy of formation (∆Gf) value, which describes the amount

of energy that is released or consumed when a phase is created from other phases, chemical

thermodynamics which can be used to predict what minerals will be formed under different

conditions. It is well known that the ΔGf of a mineral varies with changes in pressure (P),

temperature (T) and mineral composition (X) as well as the more stable position is one of lower

energy. Literature data on well studied mafic-ultramafic complexes worldwide (Fig. 2) were used

(estimation of the pie charts) for the dominated PGE-minerals (Fig. 20 – 27), illustrating the

dominant (PGM) in major PGE-bearing complexes. The pie charts were calculated with the

average amount of mineral grains per section. Also, the computer software, HSC Chemistry

software 6.0, including databases and various modules providing different types of

chemical/thermodynamic calculation, was used to calculate thermodynamic parameters. This

program was applied to calculate free energy (Tables 13 and 14) and develop diagrams of free

energy versus temperature (Figs 29 – 33). The HSC Chemistry software enables the user to

simulate chemical reactions and processes on the thermo-chemical basis, giving you the

opportunity to calculate chemical equilibrium between pure substances and the ideal solutions (and

to some extent, non-ideal solutions). In order to understand the use of the software, thermodynamic

laws and principles need to be analyzed.

Earth is a system of various disparate components. It contains more than one phase,

(solids/minerals, liquids, gases) that all together contribute to the formation of our planet. Every

procedure and reaction that takes place can be treated as an independent system, according to the

scale of the processes being considered. The rest of the environment is called the surrounding. For

example, chemical reactions between minerals, as a metamorphic event unfolds, are systems under

observation. These thermodynamic systems are being controlled by boundaries (walls) that define

their material and energy state. In case of these walls enable the material transport (mass,

elements), the system can be considered as an “open” system (metasomatic process). To the

contrary, if the material movement is blocked up, the system is called “closed”. The third kind of

system that can be described is the “isolated” system, which does not allow any energy nor matter

transfer. We assume that our system has no constraints in respect of the energy transfer, such as

64

temperature and pressure variations, so “isolated” systems are not observable in nature. In

mineralogy systems are often subdivided in different parts, which then belong to different types.

To describe the chemical composition of any system, we must define the chemical entities

that are required, which called components. The number of the components that participate in a

system depends ultimately upon the nature of the thermodynamic problem. A system consisting

of two components is referred as binary. If the components are more, the system is referred as

ternary, quaternary etc. System properties, such as pressure (P), volume (V), and temperature (T)

are called state functions. The value of a state function depends only on the state of the system

and not on the way in which the system came to be in that state. A change in a state function

describes a difference between the two states. It is independent of the process or pathway by which

the change occurs. The simplest approach of the term Energy is the one that describes “the capacity

to do work and/or transfer heat”. Heat, light, electricity, sound chemical or gravitational potential

and nuclear matter are some energy forms that can produce work.

2.4.1. Thermodynamic Laws

First Law of Thermodynamics

To understand the phenomena of thermodynamics and its applications, we have to

introduce specified principles and parameters, which can be defined by the so called

thermodynamic laws. The basic relationship that triggers this research study is the law of the

conservation of energy. According to this law the sum of kinetic and potential energy of a system

is constant (only in frictionless systems). If friction is present in the system, kinetic energy

decreases and heat is produced. We are all familiar with the relationship of the dissipated dynamic

energy and the produced heat, which indicates that the two types of energy (work, W and heat, Q)

are equivalent. Based on that fact we can formulate the first thermodynamic law: “If a closed

system, in the course of a thermodynamic process, moves from state A into state B, the sum of the

absorbed heat and performed work equals the change in a function of state, termed the internal

energy U.”

∆U = Q + W (1)

65

where, ΔU is the difference between internal energies of the system in its terminal and initial states,

Q is the heat absorbed by the system from the surrounding and W is the work done by the

surrounding on the system, which is the same as the work absorbed by the system. Thus,

differentiating this equation:

dU = δQ + δW (2)

Moreover, work can be expressed as related to the change of volume (P-V) or as various other

forms (δω). So:

δW = - PdV + δω (3)

As a result of this formula we deduce that:

dU = δQ – PdV + δω (4)

In the case of an isobaric volume change, the integration of the equation between states A and B

yields:

𝑈𝐵 − 𝑈𝐴 = 𝛥𝑄 − 𝑃(𝑉𝐵 − 𝑉𝐴) (5)

or re-arranged:

𝑈𝐵 − 𝑈𝐴 = 𝛥𝑄 − 𝑃(𝑉𝐵 − 𝑉𝐴) (6)

If we assume that the pressure in our system is constant, we can easily understand that the

variation of the term H depends only on the change of heat, ΔQ. This term is called enthalpy H, it

is measured on kJ.mol–1 and its definition reads:

H = U + PV (7)

Differentiating this formula, we assume that:

dH = dU + PdV + VdP (8a)

dH = δQ + VdP (8b)

66

Internal energy is directly involved with temperature and volume [U = f(T,V)], while

enthalpy depends on temperature and pressure [H = f(T,P)]. In specified variables we can define

the term of the heat capacity C. According to the system we can obtain cv, as the heat capacity of

a single phase or of a given system at constant volume:

𝑐𝑣 = (𝑑𝑄

𝑑𝑇)

𝑉= (

𝜕𝑈

𝜕𝑇)

𝑉 (9a)

and cp, as the heat capacity of a single phase or of a system at constant pressure:

𝑐𝑝 = (𝑑𝑄

𝑑𝑇)

𝑃= (

𝜕𝐻

𝜕𝑇)

𝑃 (9b)

Dividing either quantity by the number of moles yields the molar heat capacity which is an

intensive property. It is given as: 𝐶𝑣 =𝑐𝑣

𝑛 , 𝐶𝑝 =

𝑐𝑝

𝑛

So, we can now express the relationship between enthalpy and its dependence on temperature.

Using the definition of the heat capacity, cp, the change in enthalpy, dH, with change in

temperature, dT, at constant pressure is given by:

𝑑𝐻 = 𝑐𝑝𝑑𝑇 (10)

For one mole of a pure phase, the enthalpy, H, and heat capacity, cp, must be replaced by the molar

quantities H and Cp, respectively.

The Second Law of Thermodynamics

To continue with the definition of the thermodynamic laws, we have to mention that all

natural systems tend to proceed only in one direction. The target is to achieve the lowest energy

state (gravitational energy, internal energy, heat etc.). However, spontaneous reactions that take

place in a thermodynamic system cannot be foreseen. The internal energy and the enthalpy can

either increase or decrease relatively. This specific “movement” of a reaction depends on some

definite values. The search of the definition of this function leads us to the formation of a new

term, called entropy, S and its magnitude is J.mol–1K–1. This property was first discovered by

Rudolf Clausius (1822–1888). Entropy is central to the second law of thermodynamics. The

entropy measures the tendency for spontaneous change. Combined with the functions defined by

the first law, the entropy provides information about the spontaneous direction of chemical

67

reactions. It is also a property which reflects the degree of disorder in a system. The more

disordered the system, the higher its entropy. We use the term of “disorder” to describe the

positional state of atoms, whether the atoms are free to move or have fixed positions or vibrate

about an average position (usually due to thermal motion) etc.

Mathematically, entropy is a term that depends on the change of heat in respect with

temperature. We observe entropy in reversible and irreversible thermodynamic systems. In nature

all systems are irreversible. As reversible systems we define processes that proceeds in such

infinitesimally small steps that the system remains at equilibrium all the time. Reversible processes

can be closely approached but never realized completely. Irreversible natural systems proceed

from a non-equilibrated to an equilibrated condition. This path could only be disturbed by external

parameters that leave changes in the surroundings. Expressed in words, “the entropy change

associated with a change of state of a system is given by the heat absorbed by the system divided

by its temperature when the same change of state is brought about reversibly”:

𝑑𝑆 = (𝛿𝑄

𝑇)

𝑟𝑒𝑣 (11)

According to this statement, δQ is an inexact differential. However, δQ can be transformed

into an exact differential by multiply it with an integrating factor, 1/T. The subscript 'rev' means

that the relation holds only for reversible processes. It is also known that in isolated and reversible

systems entropy tend to be zero (dSrev = 0). Respectively, in an irreversible isolated system,

entropy either remains stable or the change is positive, it can never decrease. (dSirrev ≥ 0). So if a

closed system exists, in which the heat exchange take place between the system and the

surroundings, entropy should always be increased:

𝑑𝑆𝑡𝑜𝑡 = 𝑑𝑆𝑟𝑒𝑣(𝑠𝑦𝑠𝑡) + 𝑑𝑆𝑖𝑟𝑟𝑒𝑣(𝑠𝑢𝑟𝑟) ≥ 0 (12)

Combining the first and the second thermodynamic law for reversible processes, two new

equations are yielded:

dU = TdS – PdV (13a)

dH = TdS + VdP (13b)

68

2.4.2. Free Energy

These were the two classic thermodynamic laws on its simplicity. Despite of all these terms,

that enhance our explanation of the thermodynamic functions, we have to introduce some state

variables, which help us to understand and define better the spontaneous character of the

thermodynamic systems. As we mentioned above, every spontaneous reaction in nature has a

tendency to minimize its energy state. The equilibrium condition, between minerals, reflects the

minimum state of energy. These state functions referred as the thermodynamic potentials and

derive from the initial functions of the classic thermodynamic laws. By mathematical approaches

one can easily obtain, that for every irreversible process yields:

𝑑𝑆𝑠𝑦𝑠𝑡 −𝛿𝑄𝑟𝑒𝑣

𝑇> 0 (14a)

and for every reversible:

𝑑𝑆𝑠𝑦𝑠𝑡 −𝛿𝑄𝑟𝑒𝑣

𝑇= 0 (14b)

Under isochoric conditions one has δQrev = dU. So:

dU – TdS > 0 (15)

or

dU – TdS = 0 (16)

In the case of an isobaric process, δQrev can be replaced by dH and one obtains that:

dH – TdS > 0 (17a)

and

dH – TdS = 0 (17b)

These parameters could be expressed as the Helmholtz free energy (F) and Gibbs free energy (G),

respectively.

dF = dU – TdS (18)

69

or integrating the equation:

F = U – TS (19)

G = H – TS (20)

In that case, we can obtain weather or not a process is potentially spontaneous, ignoring

any changes in the surroundings. The Gibbs free energy is a measurable property of the phase

which is dependent on temperature and pressure, the composition of the component and its crystal

structure. Helmholtz free energy (F) exists at constant volume and Gibbs free energy (G) at

constant pressure. If dF and dG is smaller than zero (dF, dG<0), a process is potentially

spontaneous. If they are equal to zero (dF, dG=0), then the system has achieved thermodynamic

equilibrium. And in case they are greater than zero, a process does not proceed in the given

direction.

The mutual stability of substances (elements, compounds, ions, etc.) cannot be compared

using enthalpy (H), entropy (S), heat capacity (Cp) or temperature (T) individually. It must be done

using the Gibbs free energy defined by the above equation (20). This reaction obtains Gibbs free

energy values at selected temperatures. The enthalpy and entropy values are collected from

different databases (HSC 6.0 Thermo-chemical Data Source on PGE; Barin and Knacke., 1973;

Barin et al., 1977; Barin, 1989; 1993; Dean, 1992; Gevantman, 1981; Bailey et al., 1982). The

thermochemical data (H, S, Cp) of species have been measured experimentally using many

different methods and various calculation procedures. Particular data have been collected from

many different sources and cross-checked with several different thermochemical dependencies.

Errors due to different measurement methods also have be taken into account. The value of ΔG is

the change of the Gibbs free energy of a substance. It is also defined as the difference between free

energy (G) of the products and reactants in a reaction. Using equation (21), the change in free

energy is calculated for a unique chemical substance, as in the NBS tables of chemical

thermodynamic properties (Bailey et al., 1982).

ΔG = G(species) – ΣG(elements) (21)

70

Using the Reaction Equations mode, chemical compounds and reactions were analyzed, in order

to calculate the change of the Gibbs free energy (ΔG). Calibrating a chemical reaction, ΔG is

calculated by equation (22), when the chemical reaction follows equation (24):

ΔG = G(Products) – G(reactants) (22)

ΣG(Products) - ΣG(Reactants) = (cGC + dGD + …) – (aGA + bGB + …) (23)

aA + bB + … = cC + dD + … (24)

71

2.5. EVALUATION OF THE METHODOLOGY

Assuming the existence of metallic PGE clusters rather than PGE-oxygen bond (Tredoux

et al., 1995; Helmy et al. 2013) the Gibbs free energy values were calculated (Table 14). To

evaluate the accuracy of the method of calculation of the Gibbs free energy of formation, at

standard state conditions, the (ΔGf) for the rhodium sulfide phases Rh3S4 and Rh2S3 were

calculated using the HSC program. The Reaction Equations mode was used to estimate the free

energy of the reactions (25) and (26) through the equation (22), combining data from the HSC

Main database.

3Rh(s) + 2S2(g) = Rh3S4 (25)

4 Rh3S4(s) + S2(g) = 6Rh2S3 (26)

These values were compared to those obtained using a solid-state electrochemical

technique by Jacob and Gupta (2014). The calculated values of ΔGf for these rhodium sulfides

(Fig. 28) were found to be in a good agreement with those of temperature in the range from (925

to 1275) K (Table 13). Near to 1250 K there is excellent agreement between the measured and

calculated values. At lower temperatures the calculated values fall below the measured values and

this discrepancy increases with decreasing temperature. The reason of such a discrepancy is not

obvious, but it is probably due to many experimental difficulties and uncertainties, since current

database has been collected from many different measurement methods.

TABLE 13Α. Comparison between ΔGf values of (a) this study and (b) Jacob & Gupta (2014) for the reaction

3Rh(s) + 2S2(g) = Rh3S4(s)

T (K)

ΔG(a) (kJ) ΔG(b) (kJ) |ΔG(b)-ΔG(a)|

900 -289.9 -274.0 15.9

950 -272.6 -258.8 13.8

1000 -255.5 -243.5 12.0

1050 -238.4 -228.3 10.1

1100 -221.4 -213.1 8.3

1150 -204.5 -197.9 6.6

1200 -187.6 -182.6 5.0

1250 -170.9 -167.4 3.5

1300 -154.2 -152.2 2.0

72

TABLE 13B. Comparison between ΔGf values of (a) this study and (b) Jacob & Gupta (2014) for the reaction

4Rh3S4(s) + S2(g) = 6Rh2S3(s)

T (K)

ΔG(a) (kJ) ΔG(b) (kJ) |ΔG(b)-ΔG(a)|

900 -96.0 -86.9 9.1

950 -86.6 -78.9 7.7

1000 -77.2 -70.9 6.3

1050 -67.8 -62.9 4.9

1100 -58.4 -54.9 3.5

1150 -49.0 -46.9 2.1

1200 -39.5 -38.9 0.6

1250 -30.0 -30.9 0.9

1300 -20.5 -22.9 2.4

Since the question of the original existence of bonds between metallic PGEs and sulfur,

oxygen or other ligands remains unclear (Ohnensteter et al., 1991, 1999; Brenan and Andrews,

2001; Bockrath et al., 2004; Helmy et al., 2013) the calculated Gibbs free energy values were

applied to the PGE mineralization processes, assuming:

1) The existence of original bonds between metallic PGEs and oxygen, as oxide species (MO).

According to the general reaction:

Mmelt + x/4O2 = MOx/2.melt (27)

where x is the valence state of the dissolved metal species. The platinum-group elements

(PGE) tend to exist in the metallic state or bonded with sulfur, oxygen or other ligands, but the

mechanism of their formation remains unclear.

Although the speciation of PGE in sulfur-bearing silicate melts is unknown, a possible reaction

for the formation of PGE sulfides takes the form:

MSx/2.melt + x/4O2 = MOx/2.melt + x/4S2 (28)

Equation (28) indicates that PGE sulfides will be produced as a result of decreased fO2.

This process involves the selective uptake of elements that exist in more than one oxidation state,

73

producing a reduced zone near the solid-melt interface. (Finningan et al., 2008). In the case of

valence value x = +4:

Ru+4 + O2 = RuO2.melt (29)

and

RuO2 + S2 = RuS2 + O2 (30)

2) The presence of PGEs as metallic clusters

If the PGEs are present as metallic clusters (Tredoux et al., 1995; Helmy et al. 2013), the

calculated Gibbs free energy of formation at the level of 1250oC is –110 kJ/mole (Table 14).

Therefore, the Gibbs free energy values were calculated, assuming the existence of metallic PGE

clusters rather than PGE-oxygen bonds. The minerals were treated as one chemical substance via

equation (21).

74

Figure 28: Comparison of the calculated Gibbs free energy (ΔG) against temperature, for two reactions.

(A) 3Rh(s) + 2S2(g) = Rh3S4(s) in this study with values obtained by previous authors. Data for line (a) from Table

12A; data for line (b): ΔGf (J * mol-1) = -548026 + 304.5 * T (K) from Jacob & Gupta (2014) and

(B) 4Rh3S4(s) + S2(g) = 6Rh2S3(s) in this study with values obtained by previous authors. Data for line (a) from Table

12B; data for line (b): ΔGf (J * mol-1) = -230957 + 160.03 * T (K) from Jacob & Gupta (2014).

75

3. CALCULATED GIBBS FREE ENERGY AND

STABILITY DIAGRAMS

The calculated changes of Free Energy of formation (ΔGf) for common platinum-group

minerals are negative (Table 14). These negative values mean that these minerals are more stable

than the separate elements at room temperature and pressure, and that the reaction or process is

spontaneous. Some of the energy produced will be given off as heat; some will contribute to

entropy.

The plots of ∆G versus temperature shown are known as Ellingham diagrams (Ives, 1960;

Warn and Peters, 1996) shown (Figs 29 – 32). The position of the line for a given reaction on the

Ellingham diagram shows the stability of the PGM as a function of temperature and the relative

stability of different PGM. The majority of the lines slope upwards, because sulfides, arsenides

and tellurides are all solid phases (Warn and Peters, 1996). The calculated Gibbs free energy of

formation for the RuS2 is -110 kJ/mole at a magmatic temperature range between 1200 and

1300oC, which is lower compared to that for the OsS2, corresponding to -50 kJ/mole at the same

magmatic temperature (Fig. 26A). Also, the (ΔGf) for the Rh2S3 is much lower (-170 kJ/mole

than the Ir2S3 (-5 kJ/mole), other end member of this solid solution series (Fig. 29B). The (ΔGf)

for (Pd, Pt)Te2 is higher than that for (Pt, Pd)-sulfides (Table 14; Figs. 30A). The tellurides (Pd,

Pt)Te2 and IrTe2, as well as OsAs2 and PtAs2 are characterized by positive (ΔGf) values at a

magmatic high temperature range (Figs. 30A and 32). Thus, reactions closer to the top of the

diagram (IrTe2, PtTe2, OsAs2) appear to be unstable, whilst toward the bottom of the diagram the

PGM become progressively more stable.

Since ∆H and ∆S are essentially constant with temperature unless a phase change occurs, the

free energy versus temperature plot can be drawn as a series of straight lines, where ∆S is the slope

and ∆H is the y-intercept (Ives, 1960; Warn and Peters, 1996). With respect to the recorded

difference between the diagrams showing smooth curves (Figs 29A, 29B, 30B and Fig. 31) and

diagram for tellurides (PtTe2, PdTe2 and IrTe2 (Fig 30B), showing an abrupt change of direction

at 450°C (Fig. 30B), it may be related with the phase changes, as is exemplified by a phase change

at 449°C for compositions between PtTe2 and Te in the related phase diagram (Kim and Chao,

1990; Okimoto, 1992). Phase changes are reflected in entropy changes and in turn in the calculated

(ΔGf) values (Table 14; Fig. 30B).

76

TABLE 14. Gibbs free energy (ΔGof/kJ*mol-1) of formation for Platinum-group minerals estimated from the

literature (experimental) data and canculated values ofΔG as a faction of the temperature.

T (°C) RuS2 OsS2 IrS2 IrTe2 PtTe2 PdTe2 Ir2S3 Rh2S3 PtS PdS OsAs2 PtAs2 RuO2

25 -204.1 -135.1 -123.9 -68 -52.9 -60.6 -196.4 -252.4 -76.1 -66.7 -75.76 -193.4 -252.7

50 -202.9 -134 -123.2 -67.7 -52.4 -60.3 -195.2 -251.5 -75.7 -66.4 -75.6 -191.3 -248.3

100 -200.5 -131.8 -121.6 -66.5 -51.4 -59.9 -192.8 -249.8 -74.7 -65.6 -74.9 -186.7 -239.5

150 -197.7 -129.1 -119.6 -64.9 -50.5 -59.4 -189.8 -247.3 -73.5 -64.7 -73.7 -181.5 -230.9

200 -194.5 -126.2 -117.2 -62.8 -49.4 -58.9 -186.2 -244.3 -72.1 -63.6 -72 -175.9 -222.3

250 -191.1 -123 -114.7 -60.3 -48.4 -58.4 -182.4 -241.1 -70.6 -62.4 -70 -169.9 -213.9

300 -187.5 -119.7 -112 -57.4 -47.4 -57.9 -178.4 -237.6 -69 -61.1 -67.5 -163.5 -205.5

350 -183.9 -116.2 -109.3 -54.2 -46.3 -57.5 -174.2 -234 -67.5 -59.8 -64.7 -156.7 -197.2

400 -180.1 -112.7 -106.5 -50.6 -45.2 -57 -170 -230.3 -65.8 -58.4 -61.6 -149.6 -189.1

449.51 - - - -46.7 -44.1 -56.6 - - - - - - -

450 -176.3 -109.2 -103.6 -46.7 -44 -56.5 -165.6 -226.5 -64.1 -57 -58.2 -142.2 -180.9

500 -172.5 -105.6 -100.7 -40 -40.4 -53.6 -161.2 -222.6 -62.4 -55.5 -54.3 -134.6 -172.9

550 -168.6 -101.9 -97.8 -33.1 -36.7 -50.7 -156.7 -218.6 -60.7 -54.1 -50.6 -126.7 -165

600 -164.6 -98.2 -94.8 -25.8 -32.8 -47.6 -152.2 -214.6 -59 -52.7 -46.4 -118.4 -157.1

626.85 - - - - - -46 - - - -51.9 - - -

650 -160.6 -94.4 -91.7 -18.2 -29 -44.6 -147.7 -210.5 -57 -51.2 -42 -110 -149.3

700 -156.6 -90.7 -88.7 -10.4 -25 -41.5 -143.1 -206.3 -55.4 -49.7 -37.3 -101.3 -141.5

750 -152.5 -86.9 -85.6 -2.3 -21 -38.4 -138.4 -202.1 -53.6 -48.2 -32.4 -92.4 -133.8

800 -148.4 -83 -82.5 6 -17 -35.3 -133.8 -197.9 -51.9 -46.8 -27.3 -83.3 -126.2

850 -144.2 -79.1 -79.4 14.5 -12.9 -32.2 -129.1 -193.6 -50 -45.2 -20.6 -72.5 -118.6

876.85 - - - 19.2 -10.7 -30.6 - - - - - - -

900 -140 -75.3 -76.3 23.3 -8.7 -29.1 -124.4 -189.3 -48.3 -43.8 -12.8 -60.8 -111.1

926.85 - - -74.6 28.1 - - -121.9 -187 - - - - -

950 -135.8 -71.3 -73.1 32.2 -4.5 -26 -119.7 -185 -46.5 -42.3 -4.9 -48.8 -103.6

1000 -131.6 -67.4 -70 41.4 -0.3 -22.9 -115 -180.7 -44.7 -40.8 -3.1 -36.7 -96.2

1026.85 -129.3 - - - 2 - - - -43.7 - - -30.1 -92.3

1050 -127.3 -63.5 -66.8 50.7 3.9 -19.8 -110.2 -176.3 -42.9 -39.3 11.4 -24.4 -88.9

1100 -123.1 -59.5 -63.6 60.2 8.3 -16.7 -105.5 -171.9 -41.1 -37.8 19.8 -11.9 -81.5

1150 -118.8 -55.5 -60.4 69.9 12.6 -13.6 -100.7 -167.4 -39.3 -36.3 28.4 0.7 -74.3

1200 -114.4 -51.5 -57.2 79.7 17 -10.5 -96 -163 -37.5 -34.8 37.1 13.5 -67

1250 -110.1 -47.5 -54 89.8 21.4 -7.4 -91.2 -158.5 -35.7 -33.3 46 26.5 -59.9

1300 -105.7 -43.5 -50.8 99.9 25.8 -4.3 -86.4 -154 -34 -31.8 55 39.6 -52.7

77

Figure 29: Plot of the Free energy (ΔG) against temperature for (A): RuS2 (laurite), OsS2 (erlichmanite) and IrTe2

(shuangfengite) and (B): Rh2S3 (bowieite) and Ir2S3 (kashinite). Data from Table 14.

78

Figure 30: Plot of the Free energy (ΔG) against temperature for (A): PtTe2 (moncheite), PdTe2 (merenkyite) and IrTe2

(shuangfengite), and (B) PtS (cooperite) and PdS (braggite). Data from Table 14.

79

Figure 31: Plot of the Free energy (ΔG) against temperature for PtAs2 (sperrylite) and OsAs2 (Omeiite). Data from

Table 14.

80

Figure 32: A comparative diagram plotting the Free energy (ΔG) against temperature. Data from Table 14.

81

In addition to the Gibbs free energy of formation (ΔGof) for the RuS2 the (ΔGo

f) for the oxide

RuO2 was calculated as well (Table 14; Fig. 33). It is remarkable that at temperatures lower than

500oC the (ΔGf) values for the RuO2 are more negative compared to those for RuS2, suggesting

that the transformation of laurite to oxides and probably hydroxides is facilitated at relatively low

temperature.

Figure 33: Plot of the Free energy (ΔG) against temperature for RuO2 and RuS2. Data from Table 14.

82

The behavior of Os, Ir, Ru, Pt, Pd and Rh during partial melting and crystal fractionation and

the debate as to which mineral or minerals are collectors of Os, Ir and Ru have been investigated

and reviewed extensively (Locmelis et al., 2011; Page et al., 2012 and references here in).

Experimental data on stability limits of laurite and Ru–Os–Ir alloys, as functions of temperature

in basaltic magmas suggest that they could be primary magmatic minerals, crystallizing at high

temperature from silicate magma (Brenan and Andrews, 2001). Tredoux et al. (1995) proposed a

model whereby the PGE are present in the melt as metallic clusters of 50-100 atoms. In a S-rich

environment, the PGE clusters and associated ligands are included by sulfide melt. In S-poor

environments, the clusters remain in the melt until they coalesce to form alloys. The alloys may

then act as nucleation sites for chromite and olivine, thereby becoming included in early

crystallizing phases. The PGE most likely to form clusters are the heavy PGE (Os, Ir, Pt) and,

therefore, it might be predicted that alloys of these metals would form first. Furthermore, Helmy

et al. (2013) emphasized that if all highly siderophile elements are associated with nanophases in

undersaturated melts, the distribution of the noble metals between silicate, sulphide and metal

melts may be controlled by the surface properties of nano-associations, more than by the chemical

properties of the elements.

83

4. APPLICATION OF (ΔGf) VALUES

4.1. THE ORIGIN OF PGM

The calculated Gibbs free energy values (Tables 12 and 14) may be applied on the PGE/PGM

data from the investigation of their occurrence in nature, contributing to the better understanding

of the knowledge provided by geochemical data, mineral chemistry and structure relationships.

The calculated Gibbs free energy values can be applied to the substitution of Os (and Ir)

into the RuS2 lattice sites, and may contribute to the better understanding the knowledge provided

by geochemical data, mineral chemistry and structure relationships (microscopic scale) as they are

obtained from the investigation of PGE/PGM in nature. More specifically, although laurite in

chromitites associated with ophiolite complexes are characterized by the general formula

(Ru>Os>Ir)S2, in the Othrys complex and other ophiolites of the Balkan peninsula Os-rich

(Os>Ru>Ir)S2 members, reaching the composition of erlichmanite are common (Garuti et al.,

1999). Thus, the calculated lower Gibbs energy for the RuS2 (-110 kJ/mole ) at a magmatic

temperature (1250oC), compared to that for the OsS2, (-50 kJ/mole, at the same magmatic

temperature, Fig. 26A) is consistent with the Os-Ru substitution, during crystal fractionation, and

the metal-sulfide equilibrium curves for PGE indicating that at given T, the formation of RuS2

requires fS2 lower than that for OsS2 (Toulmin and Barton, 1964; Barin, 1995). In addition, the

lowest Gibbs free energy for the Rh2S3 phase (Fig. 29) is in agreement with its occurrence in an

intergrowth with laurite among detrital PGM identified in rivers draining the Bushveld complex

(Oberthür et al., 2014), suggesting the preference of Rh to a solid state (compatible behavior) and

its precipitation at high temperature. On the basis of comprehensive studies (Qian et al., 2013) it

has been demonstrated that the phase IrTe2 exhibits a structural phase transition from trigonal to

triclinic and various mechanisms, including the reduction of kinetic energy by the crystal field

effects. Apart from the complex structural characteristics of the IrTe2 phase a salient feature is its

highest (ΔGf) values (Table 14; Figs. 30A and 32), suggesting that it may be spontaneous only at

a range of low temperatures.

The formation of stable high-temperature phases between As and Pt or Rh, including

sperrylite (PtAs2), hollingworthite (Rh, Pt, Pd)AsS, and platarsite (Pt, Rh, Ru)AsS is well known

(Prichard et al., 1986, 1994. 2008; Tarkian et al., 1996; Tsoupas and Economou-Eliopoulos, 2008).

84

Furthermore, the calculated positive (ΔGf) values for OsAs2 and PtAs2, at a magmatic high

temperature range, in contrast to the negative values, at lower temperature (Table 14; Fig. 31)

indicate that these formation reactions becomes spontaneous at lower temperature. Thus, the

presented (ΔGf) values are consistent with the common occurrence of sperrylite (PtAs2) in

chromitites as an interstitial phase between chromite grains (Prichard et al., 1986, 1994. 2008;

Tarkian et al., 1996).

The dominance of (Pt, Pd)S in the UG2 (Bushveld) and J-M horizons (Stillwater),

compared to the Merensky Reef (Bushveld) where (Pt, Pd)-tellurides are dominant, seem to be

consistent with the calculated lower Gibbs energy for the former than the latter (Table 14; Fig. 30)

and therefore the sequence of their deposition. Platinum and Pd-tellurides, such as moncheite

(PtTe2), merenskyite (PdTe2), melonite [(Ni, Pd)Te2] and kotulskite [(Pd, Ni)Te], and bismuth-

tellurides [(Pt, Pd)(Te, Bi)2] associated with Fe–Ni–Cu sulfide deposits have been described in

certain porphyry-Cu-Au±Pd±Pt systems (Eliopoulos and Economou-Eliopoulos, 1991; Tarkian

and Stribrny, 1999; Economou-Eliopoulos, 2005; Augé et al., 2005). Assuming that the elevated

values of the (Pd+Pt) content and extremely low Cr contents (<1 ppm) in high Cu–Pd–Pt-grade

ores from the Balkan Peninsula suggest their genesis from more evolved mineralized fluids, the

calculated free energy values of formation for (Pd, Pt)-tellurides (Table 14) seem to be in a good

agreement with the above minerals hosting the precious metals. In addition, the common

occurrence of the (Pd, Pt±Bi)Te2, merenskyite, the main PGE mineral in porphyry Cu–Au–Pd–Pt

deposits, at the peripheral parts of chalcopyrite (Economou-Eliopoulos, 2005) seems to be

consistent with their free energy values. More specifically, the lower free energy of formation

(ΔGf) values for chalcopyrite (-369.6 kJ/mole, after Robie et al., 1994) compared to that for

merenskyite (-70 kJ/mole) suggest its earlier deposition of the former.

In addition, the nucleation process by which clusters of a new phase form a parent phase is

a universal phenomenon in both nature and technology, and there is interest in developing

quantitative accurate theoretical tools such as thermodynamics and kinetics to address the

nucleation. The thermodynamic approach on the nanometer size is considered to provide insight

into the metastable phase generation in microphase growth from the viewpoint of thermodynamics.

Although metastable structures can have sufficiently long lifetimes to be observed on experimental

scales, it has been emphasized that clusters collide to form a single aggregate, which subsequently

85

undergoes re-arrangement processes (Wang and Yang, 2005). Thus, the application of the nano-

thermodynamic analysis on the proposed model whereby the PGE are present in the melt as

metallic clusters (Tredoux et al., 1995; Helmy et al., 2013) may be a method to better

understanding the metastable phase formations on nanoscale.

86

4.2. THE STABILITY OF PGM

Besides the unaltered IPGM enclosed in chromite, IPGM-sulfides interstitial to the chromite

grains are commonly transformed into IPGE-alloys, as is exemplified by extremely large IPGM

crystals within massive chromite ores located along a shear zone from Greece. Laurite (RuS2) with

variable Os and Ir contents occurs as remnants surrounding by Ru–Os–Ir-oxides (Fig. 15). The

Os–Ru–Ir–Fe alloys contain small amounts of Cr, Co±Mn, followed by irarsite (Tsoupas and

Economou-Eliopoulos, 2008). The apparent epigenetic desulphurisation and partially oxidation of

primary laurite crystals has been affected by strong brittle deformation (Fig. 17). The lower values

of the calculated Gibbs free energy of formation (ΔGof) for the oxide RuO2 compared to the (ΔGo

f)

for RuS2 at temperatures lower than 500oC (Table 14; Fig. 33), suggest a transformation of laurite

to oxides and probably hydroxides at relatively low temperature. They are in a good agreement

with the literature data indicating such a replacement association, which has been interpreted as in

situ desulfurization products of the PGM sulfides, at relatively low (350 – 500oC) temperature

(Stockman and Hlava, 1984; Tarkian and Prichard, 1987; Garuti and Zaccarini, 1997; Melcher et

al., 1997; Tsoupas and Economou-Eliopoulos, 2008).

The Great Dyke of Zimbabwe constitutes the world’s second largest reserve of platinum group

elements (PGE) after the Bushveld Complex, South Africa, near-surface oxidized MSZ ores have

a large potential (resources are in the range of 160 – 250 Mt). (Oberthur et al., 2013). Platinum

occurs mainly in the form of discrete platinum group mineral (PGM) grains (mainly

bismuthotellurides, sulphides, and arsenides), whereas approximately 80 per cent of the Pd (and

some Rh) is hosted in pentlandite. The base metal sulphides are destroyed, partly releasing their

base metal and PGE contents, and are replaced by iron oxides or hydroxides. Unspecified amounts

of the PGE may be in the form of (Pt/Pd)-oxides or hydroxides, or in iron-hydroxides, Mn–Co-

hydroxides, and in secondary silicates, the problematic processing of oxidized MSZ ores has been

attributed to their complex nature and polymodal distribution of the PGE (Oberthur, et al., 2013).

87

4.3. THE POTENTIAL ENVIRONMENTAL RISK

The increasing demand for mineral resources and fossil fuels continues to grow worldwide,

the contamination due to mining is a topic systematic geochemical studies for many authors. This

has led to an improved understanding of cause and effect to human health and ecosystems and for

new environmental legislation and strategies for remediation. The mining and smelting operations

for PGEs may affect the workers and local population, creating environmental contamination. In

S. Africa, local communities may be particularly exposed to the effects of contamination arising

from mining and the legacy of mining that has left thousands of sites in Africa contaminated by

tailings and slag material. The environmental impact of a tailings facility, located in Steelpoort,

Mpumalanga, South Africa, operates since 2008 and is associated with the mining of the UG2 reef

(Critical Zone of the Rustenburg Suite, Bushveld Complex) has been studied. It has been

concluded that at aquifer level, the nearby river an accumulation of the major cations and anions

as well as trace heavy metals may take place in the aquifer at the thrust fault boundary (Huisamen

and van Rooy, 2012).

In addition, traffic emissions and their impact on urban air quality, health, and atmospheric

processes has been the subject of increasing interest in recent years (Ely et al., 2001; Dahlheimer

et al., 2007). The increased use of these noble metals in automobile catalyst converters, has led to

their release into the environment and biological accumulation on roadsides, since 1974 in USA

and 1993 in European countries (Zereini et al., 1998; Palacios et al., 2000; Ely et al., 2001; Jarvis

et al., 2001; Riga-Karandinos et al., 2006; Croy et al. 2008; Economou-Eliopoulos and Sfendoni,

2010). Catalytic converters are used to treat automobile emissions: reduction of nitrogen oxides to

nitrogen and oxygen (2NOx → xO2 + N2), oxidation of carbon monoxide (toxic) to harmless

carbon dioxide (2CO + O2 → 2CO2) and oxidation of carcinogenic hydrocarbons (HC) to carbon

dioxide and water (2CxHy + (2x+y/2)O2 → 2xCO2 + yH2O). There are indications that

environmentally formed Pd species are more soluble and hence more mobile in the environment

than Rh and Pt (Dahlheimer et al., 2007). A good positive relationship between both Pd and Pt

contents in plants with their corresponding contents in soils (Economou-Eliopoulos and Sfendoni,

2010) is consistent with their solubility in soils (Wood, 1990, 2005; Wood et al., 1994; Rauch and

Morrison, 2000). More specifically, the accumulation factor [(metal content in plant × 100)/metal

content in soil] for Pt and Pd is 3.1 and 4.4%, respectively, suggesting that bio-accumulation of

88

the studied metals followed the order Pd>Pt (Economou-Eliopoulos and Sfendoni, 2010). In

addition, there is evidence showing significant redistribution of PGE in laterites (Bowles 1986,

Bowles et al. 1994; Aiglsperger et al., 2015). It has been suggested that organic compounds (humic

acids) could well provide a potential means of PGE transport in solution (Bowles et al., 1994).

Data collected from representative dust samples from Katehaki, Messoghion, Iera odos

roads and the National highway road Athens – Thessaloniki of Greece demonstrated that catalytic

converter attrition is responsible for elevated abundances and that concentrations of PGEs

increased with traffic density, reaching values up to 2070 ppb Pt and 1980 ppb Pd in dust. Gold,

ranging between 14 and 990 ppb (average 310) in dust samples and from 27 and 160 ppb Au

(average 95) in soils (Table 16). The PGE level along the highways of Greece (Table 15) that may

be washed by rain and concentrated into local water systems, is comparable to that given for many

other countries (Hodge & Stallard, 1986; Ely et al., 2001; Dahlhaimer et al., 2007).

Although there are limited thermodynamic data available for aqueous organic or

hydroxide complexes involving the PGE, recently, Colombo et al., (2008) used the thermodynamic

computer model, HSC to predict the interactions of Pt, Pd and Rh with different inorganic ligands

and estimated the thermodynamic stability of these species in the environment. They presented

Eh–pH diagrams and indicated that Pt, Pd and Rh can form complexes with all of the inorganic

ions studied, suggesting that they are capable of mobilizing the PGEs as aqueous complexes that

can be bio-available to plants/crops (food chain).

Table 15: Palladium and platinum in porphyry-Cu-Au-Pd-Pt deposits.

89

Table 16: Precious metal content in dust and soils along roadside.

90

5. CONCLUSIONS

Although there are uncertainties in heat capacity and enthalpy data for PGM, and the application

of thermodynamic experiments are limited, application of thermodynamics to the study of nature

systems can be compared with phase relations observed in rocks associated with platinum-group

minerals to predict their occurrence and stability. The preliminary results presented for the Gibbs

free energy of the PGM allow us to present the following conclusions:

1. The evaluation of the accuracy of the method of calculation of the Gibbs free energy of

formation (ΔGf) using the HSC program was made by comparison of the (ΔGf) values for the

rhodium sulfide phases Rh3S4 and Rh2S3 to those obtained using a solid-state electrochemical

technique by Jacob and Gupta (2014) and they were found to be in a good agreement.

2. The calculated free energy of formation (ΔGf) were negative values, indicating that these

minerals are more stable than the separate elements at standard conditions (the process is

spontaneous).

3. The calculated Gibbs free energy of formation (ΔGof) in the: RuO2<RuS2< OsS2< (Pt, Pd)S<

(Pd, Pt)Te2 is increasing from compatible to incompatible noble metals, from oxides to sulfides

and from sulfides to tellurides.

4. The lower values of the calculated Gibbs free energy of formation (ΔGof) for the oxide RuO2

compared to that for RuS2 at temperatures lower than 500oC suggest a transformation of laurite

to oxides and probably hydroxides at relatively low temperature.

5. The calculated positive (ΔGf) values for OsAs2 and PtAs2, at a magmatic high temperature

range, and the negative values at lower temperature may indicate that their formation becomes

spontaneous at lower temperature.

6. Samples from Skyros Island indicate two generation of laurite (primitive and secondary),

controlled by texture and geotectonic procedures.

7. Literature data on the thermodynamic stability of Pt, Pd and Rh in the environment indicating

that they can form complexes with inorganic ions and are bio-available, are consistent with the

bio-accumulation factors in plants for these noble metals, derived from traffic emissions

(automobile catalyst converters) along high-way roadside.

91

6. REFERENCES

Aiglsperger, T., Proenza, J.A., Zaccarini, F., Lewis, J.F., Garuti, G., Labrador, M., Longo, F.,

2015. Platinum group minerals (PGM) in the Falcondo Ni-laterite deposit, Loma Caribe

peridotite (Dominican Republic). Mineralium Deposita, 50 (1), 105 – 123.

Augé, T., Legendre, O., 1994. Platinum-group element oxides from the Pirogues ophiolitic

mineralization, New Caledonia; origin and significance. Economic Geology, 89 (7), 1454 –

1468.

Augé, T., Maurizot, P., 1995. Stratiform and alluvial platinum mineralization in the New

Caledonia ophiolite complex. Canadian Mineralogist, 33, 1023 – 1045.

Augé, T., 1985. Platinum-group mineral inclusions in ophiolitic chromitite from the Vourinos

complex, Greece. Canadian Mineralogist, 23 (2), 163 – 171.

Augé, T., 1988. Platinum-group minerals in the Tiébaghi and Vourinos ophiolitic complexes:

genetic implications. Canadian Mineralogist, 26 (1), 177 – 192.

Augé, T., Petrunov, R. And Bailly, L. (2005): On the mineralization of the PGE mineralization in

the Elastite porphyry Cu–Au deposit, Bulgaria: Comparison with the Baula-Nuasahi Complex,

India, and other alkaline PGE-rich porphyries. Canadian Mineralogist, 43, 1355–1372.

Bailey, S.M., Churney, K.L., Nuttall, R.L., 1982. The NBS Tables of Chemical Thermodynamic

Properties, Selected Values for Inorganic and C1 and C2 Organic Substances in SI Units,

Journal of Physical AND Chemical Reference Data, 11, Suppl. No. 2.

Barin, I., 1989. Thermochemical Data of Pure Substances, VCH Verlags Gesellschaft, Weinheim.

Barin, I., 1993. Thermochemical Data of Pure Substances, Part I, VCH Verlags Gesellschaft,

Weinheim.

Barin, I., 1993. Thermochemical Data of Pure Substances, Part II, VCH Verlags Gesellschaft,

Weinheim.

Barin, I., Knacke, O., 1973. Thermochemical properties of inorganic substances, Springer-Verlag,

Berlin, 921.

92

Barin, I., 1989. Thermochemical Data of Pure Substances. VCH, Weinheim, Germany.

Barin, I., 1995. Thermochemical Data of Pure Substances. 3rd edition, VCH Publishers, New

York.

Barin, I., Knacke, O., Kubaschewski, O., 1977. Thermochemical properties of inorganic

substances, Supplement, Springer-Verlag, Berlin, 861.

Barnes, S.J., Campbell, I.H., 1988: Role of late magmatic fluids in Merensky-type platinum

deposits: A discussion. Geology, 16, 488 – 491.

Barnes, S.J., Picard, C.P., 1993. The behaviour of platinum-group elements during partial melting,

crystal fractionation, and sulphide segregation: an example from the Cape Smith Fold Belt,

northern Quebec. Geochimica et Cosmochimica Acta, 57 (1), 79 – 87.

Barnes, S.J., Liu, W., 2012. Pt and Pd mobility in hydrothermal fluids: Evidence from komatiites

and from thermodynamic modeling. Ore Geology Reviews, 44, 49 – 58.

Barnes, S.J., Maier, W.D., 2002. Platinum-group Elements and Microstructures of Normal

Merensky Reef from Impala Platinum Mines, Bushveld Complex. Journal of Petrology, 43 (1),

103 – 128.

Bockrath, C., Ballhaus, C., Holzheid, A., 2004. Stabilities of laurite RuS2 and monosulfide liquid

solution at magmatic temperature. Chemical Geology, 208, 265 – 271.

Bockrath, C., Holzheid, A., Ballhaus C., 2002. Stability of laurite under controlled oxygen and

sulfur activities at magmatic temperature – an experimental study. 9th International Platinum

Symposium, IPS Abstracts.

Bowles, J.F.W., Gize, A.P., Vaughan, D.J., Norris, S.J., 1994. Development of platinum-group

minerals in laterites-initial comparison of organic and inorganic controls. Transactions of the

Institution of Mining and Metallurgy, 103, B53 – B56.

Bowles, J.F.W., 1986. The development of platinum-group minerals in laterites. Economic

Geology, 81, 1278 – 1285.

Brenan, J.M., Andrews, D., 2001. High-temperature stability of laurite and Ru–Os–Ir alloy and

their role in PGE fractionation in mafic magmas. Canadian Mineralogist, 39 (2), 341 – 360.

93

Cabri, J.L., Laflamme, J.H.G., 1976. The mineralogy of the platinum-group elements from some

copper-nickel deposits of the Sudbury area, Ontario. Economic Geology, 71 (7), 1159 – 1195.

Cabri, L.J., 2002. The Geology, Geochemistry, Mineralogy and Mineral Beneficiation of

Platinum-Group Elements. CIM Special, 54, Canadian Institute of Mining, Metalurgy and

Petroleum, Quebec, Canada.

Campbell, I.H., Naldrett, A.J., Barnes, S.J., 1983. Experimentally determined sulfide melt-silicate

melt partition coefficients for iridium and palladium. Chemical Geology, 117, 361 – 377.

Cawthorn, R.G., 1999. Platinum-group element mineralization in the Bushveld Complex; a critical

reassessment of geochemical models. South African Journal of Geology, 102, 268 – 281.

Cawthorn, R.G., Merkle, R.K.W., Viljoen, M.J., 2002. Platinum-Group Element Deposits in

Bushveld Complex, South Africa. In The Geology, Geochemistry, Mineralogy and Mineral

Beneficiation of Platinum-Group Elements, (L.J. Cabri, Ed.) 54, 389 – 429.

Cemic, L., 2005. Thermodynamics in mineral sciences: An Introduction. Springer-Verlag, Berlin,

Heidelberg.

Colombo, C., Monhemius, A.J., Plant, J.A., 2008. The estimation of the bioavailabilities of

platinum, palladium AND rhodium in vehicle exhaust catalysts and road dusts using a

physiologically based extraction test. Science of the Total Environment, 389 (1), 46 – 51.

Colombo, C., Oated, C.J., Monhemius, A.J., Plant, J.A., 2008. Complexation of platinum,

palladium and rhodium with inorganic ligands in the environment. Geochemistry: Exploration,

Environment, Analysis, 8, 91 – 101.

Cowden, A., Donaldson, M.J., Naldrett, A.J., Campbell, I.H., 1986. Platinum-group elements and

gold in the komatiite-hosted Fe-Ni-Cu sulfide deposits at Kambalda, Western Australia.

Economic Geologist, 81, 1226 – 1235.

Croy, J.R., Mostafa, S., Hickman, L., Heinrich, H., Cueny, B.R., 2008: Bimetallic Pt-Metal

catalysts for the decomposition of methanol: Effect of secondary metal on the oxidation state,

activity, and selectivity of Pt. Applied Catalysis A: General, 350 (2), 207 – 216.

94

Dahlheimer, S.R., Neal, C.R., Fein J.B., 2007. Potential Mobilization of Platinum-Group Elements

by Siderophores in Surface Environments. Environmental Science and Technology, 41 (3), 870

– 875.

Dean, J.A., 1992. Lange's handbook of Chemistry, Thermodynamic Properties, McGraw-Hill,

New York.

Dey, S., Jain, V.D., 2004. Platinum group metal chalcogenides. Their syntheses and applications

in catalysis and materials science. Platinum Metals Rev., 48 (1), 16-29.

Economou-Eliopoulos, M., 1996. Platinum-group element distribution in chromite ores from

ophiolite complexes: implications for their exploration. Ore Geology Reviews, 11 (6), 363 –

381.

Economou-Eliopoulos, M., 2005. Platinum-group element potential of porphyry deposits.

Mineralogical Association of Canada, Short Course 35, 203 – 245.

Eliopoulos, D.G., Economou-Eliopoulos, M., 1991. Platinum-group element and gold contents in

the Skouries porphyry-copper deposit, Chalkidiki Peninsula, northern Greece. Economic

Geologist, 86,740–749.

Economou-Eliopoulos, M., Sfendoni, T., 2010. Environmental impact of Pt, Pd, Rh and Au from

catalytic converters along roadsides: the case of Attica, Greece. XIX Congress of the Carpathian

Balkan Geological Association, Thessaloniki, Greece, 100, 47-54.

Ely, J., Neal, C., Kulpa, C., Schneegurt, M., Seidler, J. Andjain, J., 2001. Implications of Platinum-

Group element accumulation along US roads from catalytic-convertor attrition. Environmental

Science AND Technology, 35, 3816 –3822.

Finnigan, G.F., Brenan, J.M., Mungall, J.E., Mcdonaugh, W.F., 2008. Experiments and models

bearing on the role of chromite as a collector of platinum group minerals by local reduction,

Journal of Petrology, 49 (9), 1647 – 1665.

Garuti, G., Zaccarini, F., 1997. In-situ alteration of platinum-group minerals at low temperature:

evidence from serpentinized and weathered chromitites of the Vourinos complex (Greece).

Canadian Mineralogist, 35, 611 – 626.

95

Garuti, G., Zaccarini F., Proenza, J.A., Thalhammer, O.A.R., Angeli, N., 2012. Platinum-group

minerals in chromitites of the Niquelândia layered intrusion (Central Goias, Brazil): their

magmatic origin and low temperature reworking during serpentinization AND lateritic

weathering. Minerals, 2, 365 – 384.

Garuti, G., Zaccarini, F., Economou-Eliopoulos, M., 1999a. Paragenesis and composition of laurite

from chromitites of Othrys (Greece): implications for Os-Ru fractionation in ophiolitic upper

mantle of the Balkan peninsula. Mineralium Deposita, 34, 312 – 319.

Garuti, G., Zaccarini, F., Moloshag, V., Alimov, V., 1999b. Platinum-group minerals as indicator

of sulfur fugacity in ophiolitic upper mantle: an example from chromitites of the Ray–Iz

ultramafic complex (Polar Urals, Russia). Canadian Mineralogist, 37, 1099 – 1115.

Gevantman, L.H., 1981. Physical Properties Data for Rock Salt. National Bureau of standards.

Godel, B., Barnes, S.J., Maier, W.D., 2007. Platinum-Group Elements in Sulphide Minerals,

Platinum-Group Minerals, and Whole-Rocks of the Merensky Reef (Bushveld Complex, South

Africa): Implications for the Formation of the Reef. Journal of Petrology, 48 (8), 1569 – 1604.

Grammatikopoulos, T.A., Kapsiotis, A., Zaccarini, F., Tsikouras, B., Hatzipanagiotou, K., Garuti,

G., 2007. Investigation of platinum-group minerals (PGM) from Pindos chromitites (Greece)

using hydro-separation concentrates, Minerals Engineering, 20, 1170 - 1178.

Gupta, C.K., 2003. Chemical Metallurgy: Principles AND Practice, John Wiley and Sons,

Weinheim.

Hanley, J.J., 2005. The aqueous geochemistry of the platinum-group elements (PGE) in surficial,

low-T hydrothermal and high-T magmatic hydrothermal environments. In Exploration for

Platinum-group Element Deposits (J.E. Mungall, ed.). Mineral. Assoc. Canada, Short Course

35, 35 – 56.

Hattori, K.H., Cabri, L.J., Hart, S.R., 1991. Os-isotope study of PGM nuggets associated with

Freetown Layered Complex, Sierra Leone. Contributions in Mineralogy and Petrology, 109, 10

– 18.

96

Hattori, K.H., Cabri, L.J., Johanson, B., Zientek, M.L., 2004. Origin of placer laurite from Borneo:

Se and As contents, and S isotopic compositions, Mineralogical Magazine, 68, 353 – 368.

Helmy, H.M., Ballhaus, C., Berndt, J., Bockrath, C., Wohlgemuth-Ueberwasser, C., 2007.

Formation of Pt, Pd and Ni tellurides: experiments in sulfide–telluride systems. Contributions

in Mineralogy and Petrology, 153, 577 – 591.

Helmy, H.M., Ballhaus, C., Fonseca, R.O.C., Wirth, R., Nagel, T., Tredoux, M., 2013. Noble metal

nanoclusters and nanoparticles precede mineral formation in magmatic sulfide melts. Nature

Communications, 4, 2405.

Holwell, D.A., McDonald, I., 2006: Petrology, geochemistry and the mechanisms determining the

distribution of platinum-group element and base metal sulphide mineralisation in the Platreef

at Overysel, northern Bushveld Complex, South Africa. Mineralium Deposita, 41, 575 – 598.

Huisamen, A., van Rooy, J.L., 2012. Contaminant mobilisation by fluid-rock interaction and

related transport mechanisms in platinum tailings. Environmental and health impacts of mining

in Africa, Proceedings of the Annual Workshop IGCP/SIDA No. 594, Windhoek, Namibia,

July 5th - 6th, 25 – 27.

Ives, D.J.G., 1960. Principles of the Extraction of Metals, Monograph 3, Royal Institute of

Chemistry.

Jacob, K.T. And Gupta, P., 2014. Gibbs free energy of formation of rhodium sulfides. J. Chem.

Thermon., 70, 39 – 45.

Jarvis, K.E., Parry, S.J. And Piper, J.M., 2001. Temporal and Spatial Studies of Autocatalyst-

Derived Platinum, Rhodium, and Palladium and Selected Vehicle-Derived Trace Elements in

the Environment. Environmental Science and Technology, 35 (6), 1031 – 1036.

Kapsiotis, A., Grammatikopoulos, T.A., Tsikouras, B., Hatzipanagiotou K., Zaccarini, F., Garuti,

G., 2009: Chromian Spinel composition and Platinum-group element mineralogy of

Chromitites from the Milia area, Pindos ophiolite complex, Greece. Canadian Mineralogist, 47

(5), 1037 – 1056.

Kapsiotis, A., Grammatikopoulos, T.A., Zaccarini, F., Tsikouras, B., Garuti, G., Hatzipanagiotou

K., 2006. Platinum-group mineral characterisation in concentrates from low-grade PGE

97

Chromitites from the Vourinos ophiolite complex, northern Greece. Applied Earth Science, 115

(2), 49 – 57.

Kim, W.S., Chao, G.Y., 1990: Phase relations in the system Pt-Sb-Te. Canadian Mineralogist, 28,

675 – 685.

Lee, C.A., Tredoux, M., 1986. Platinum-group element abundances in the lower and the lower

critical zones of the eastern Bushveld Complex. Economic Geologist, 81, 1087 – 1095.

Locmelis, M., Pearson, N.J., Barnes, S.J., Fiorentini, M.L., 2011. Ruthenium in Komatiitic

Chromite. Geochimica et Cosmochimica Acta, 75, 3645 – 3661.

Maier, W.D., Barnes, S.J., 1999. Platinum-Group Elements in Silicate Rocks of the Lower, Critical

and Main Zones at Union Section, Western Bushveld Complex. Journal of Petrology, 40 (11),

1647 – 1671.

Melcher F., Grum W., Simon G., Thalhammer T.V., Stumpfl E.F., 1997. Petrogenesis of the giant

ophiolitic chromite deposits of Kempirsai, Kazakhstan: a study of solid and fluid inclusions in

chromite. Journal of Petrology, 38, 1419 – 1458.

Mungall, J. E., 2005. Magmatic geochemistry of the platinum-group elements In Exploration for

Platinum-Group Element Deposits, (ed. J. E. Mungall). Min. Assoc. Can. Short Course 35, 1 –

34.

Mungall, J.E., Brenan, J.M., 2014. Partitioning of platinum-group elements and Au between

sulfide liquid and basalt and the origins of mantle-crust fractionation of the chalcophile

elements. Geochimica et Cosmochimica Acta, 125, 265 – 289.

Mungall, J.E., Naldrett, A. J., 2008. Ore Deposits of the Platinum-Group Elements. Elements, 4,

253 – 258.

Naldrett, A.J., Duke, J.M., 1980. Platinum Metals Magmatic Sulfide Ores. Science, 208 (4451),

1417 – 1424.

Nordstrom D.K., Munoz J.L., 2006. Geochemical Thermodynamics. Blackburn Press.

98

Oberthür, T., 2002. Platinum-Group Element Mineralization of the Great Dyke, Zimbabwe. In The

Geology, Geochemistry, Mineralogy and Mineral Beneficiation of Platinum-Group Elements,

(L.J. Cabr, Ed), 54, 483 – 506.

Oberthür, T., Melcher, F., Gast, L., Wöhrl, C., Lodzia, J., 2004. Detrital platinum-group minerals

in rivers draining the eastern Bushveld complex, South Africa. Canadian Mineralogist 42, 563-

582.

Oberthür, T., Melcher, F., Buchholz, P., Locmelis, M., 2013. The oxidized ores of the Main

Sulphide Zone, Great Dyke, Zimbabwe: Turning resources into minable reserves – mineralogy

is the key. The Journal of the Southern African Institute of Mining and Metallurgy, 113, 191 –

201.

Ohnenstetter M., Johan Z., Cocherie A., Fouillac A.M., Guerrot C., Ohnenstetter D., Chaussidon

M., Rouer O., Makovicky E., Makovicky M., Rose-Hansen J., Karup-Moeller S., Vaughan D.,

Turner G., Pattrick R.A.D., Giże A.P., Lyon I., McDonald I., 1999. New exploration methods

for platinum and rhodium deposits poor in base-metal sulphides – NEXTRIM. Applied Earth

Science (Trans Inst. Min. Metall. Section B), 108, B119 – B150.

Ohnenstetter, M., Karaj, N., Neziraj, A., Johan, Z., Cina, A., 1991. Le potential platinifere des

ophiolites: mineralizations elements du groupe du platine (PGE) dans les massifs de Tropojaet

Bulqiza, Albanie. Comptes Rendus de l'Académie des Sciences, 313 (2), 201 – 208.

Okamoto, H., 1992. The Pd-Te System (Palladium-Tellurium). Journal of Phase Equilibria and

Diffusion, 34 (1), 72 – 73.

O'neill, H.St.C., Nell, J., 1997. Gibbs free energies of formation of RuO2, IrO2, and OsO2: A high-

temperature electrochemical and calorimetric study. Geochimica et Cosmochimica Acta, 61,

5279 – 5293.

O'neill, H.St.C., Dingwell, D.B., Borisov, A., Spettel, B., Palme, H., 1995. Experimental

petrochemistry of some highly siderophile elements at high temperatures, and some

implications for core formation and the mantle's early history. Chemical Geology, 120, 255 –

273.

99

Page, P., Barnes, S.J., Bedard, J.H., Zientek, M.L., 2012. In situ determination of Os, Ir, and Ru in

chromites formed from komatiite, tholeiite and boninite magmas: Implications for chromite

control of Os, Ir and Ru during partial melting and crystal fractionation. Chemical Geology,

302, 3 – 15.

Palacios, M.A., Gomez, M.M.,Moldovan, M.,Morrisong., Rauch, S., Mcleod, C., Ma, R., Laserna,

J., Lucena, P.,Caroli, S., Alimonti, A., Petrucci, F., Bocca, B., Schramel, P.,Lustig, S., Zischka,

M.,Wass, U.,Stenbom, B., Luna, M., Saenz, J.C., Santamaria, J., Torrens, J.M., 2000. Platinum-

group elements: quantification in collected exhaust fumes and studies of catalyst surfaces. The

Science of the Total Environment, 256, 1 – 15.

Penberthy, C.J., Oosthuyzen, E.J., Merkle, R.K.W., 2000. The recovery of platinum-group

elements from the UG-2 chromitite, Bushveld Complex – a mineralogical perspective.

Mineralogy and Petrology, 68, 213 – 222.

Prichard, H.M., Tarkian M., 1988. Platinum and palladium minerals from two PGE-rich localities

in the Shetland ophiolite complex. Canadian Mineralogist, 26, 979 – 990.

Prichard, H.M., Ixer, R.A.F., Lord, R.A., Maynard, J.B., Williams, N., 1994. Assemblages of

platinum-group minerals and sulfides in silicate lithologies and chromite-rich rocks within the

Shetland ophiolite. Canadian Mineralogist, 32 (2), 271 – 294.

Prichard, H.M., Neary, C.R., Potts, P.J. (1984): Platinum and gold in the Shetland Ophiolite.

Mining Journal, 23 (7772), 77.

Prichard, H.M., Neary, C.R., Potts, P.J., 1986. Platinum group minerals in the Shetland ophiolite.

In Metalogeny of Basic AND Ultrabasic Rocks, (M.J. Gallagher, R.A. Ixer, C.R. Neary and

H.M. Prichard, eds), 395 – 414.

Prichard, H.M., Neary, C.R., Fisher, P.C., O’hara, M.J., 2008. PGE-rich Podiform Chromitites in

the Al ‘Ays Ophiolite Complex, Saudi Arabia: An Example of Critical Mantle Melting to

Extract and Concentrate PGE. Economic Geology, 103 (7), 1507 – 1529.

Qian, T., Miao, H., Wang, Z.J., Liu, X., Shi, X.,Huang, Y.B., Zhang, P., Xu, N., Richard, P., Shi,

M., Upton, M.H., Hill, J.P., Xu, G., Dai, X., Fang, Z., Lei, H.C., Petrovic, C., Fang, A.F.,Wang,

100

N.L., Ding, H., 2013. Structural phase transition induced by van Hove singularity in 5d

transition metal compound IrTe2. arXiv preprint arXiv:1311.4946.

Rauch, S., Morrison, G.M., 2000. Routes for bioaccunmulation and transormation of platinum in

the urban environment. Anthropogenic Platinum-Group Element Emissions, 2, 85 – 93.

Riga-Karandinos, A.N., Saitanis, C.J., Arapis, G., 2006. First study of anthropogenic platinum

group elements in roadside top-soils in Athens, Greece. Water, Air, and Soil Pollution, 172, 3

– 20.

Righter, K., Campbell, A.J., Humayun, M., Hervig, R. L., 2004. Partitioning of Ru, Rh, Pd, Re, Ir

and Au between Cr-bearing Spinel, Olivine, Pyroxene and Silicate melts. Geochimica et

Cosmochimica Acta, 68 (4), 867–880.

Robie, R.A., Seal, R.R., Hemingway, B.S., 1994. Heat capacity and entropy of bornite (Cu5FeS4)

between 6 and 760 K and the thermodynamic properties of phases in the system Cu-Fe-S.

Canadian Mineralogist, 32, 945 – 956.

Sassani, D.C., Shock, E.L., 1990. Speciation and solubility of Palladium in aqueous magmatic-

hydrothermal solutions. Geology, 18, 925 – 928.

Seriani, N., Mittendorfer, F., 2008. Platinum-group and noble metals under oxidizing conditions.

Journal of Physics: Condensed Matter, 20 (18), 184023.

Slansky, E., Barron, L. M., Suppel, D., Johan, Z., Ohnenstetter M., 1991. Platinum mineralization

in the Alaskan-type intrusive complexes near Fifield, N.S.W., Australia. Part 2. Platinum-group

minerals in placer deposits at Fifield. Mineralogy and Petrology, 43 (3), 161 – 180.

Tarkian M., Economou-Eliopoulos M., Eliopoulos D.G., 1992. Platinum-group mineralsand

tetraauricupride in ophiolitic rocks of Skyros island, Greece. Mineralogy and Petrology, 47, 55

– 66.

Tarkian, M., Stribrny, B., 1999. Platinum-group elements in porphyry copper deposits: A

reconnaissance study. Mineralogy and Petrology, 65, 161–183.

101

Tarkian, M., Economou-Eliopoulos, M., Sambanis, G., 1996. Platinum group minerals in

chromitites from the Pindos ophiolite complex, Greece. Neues Jahrbuch fur Mineralogie

Monatshefte, 145 – 160.

Tolstykh, N.D., Sidorov, E.G., Krivenko, A.P., 2005. Platinum-group element placers associated

with Ural-Alaska type complexes. Exploration for Platinum-Group Element Deposits,

Mineralogical Association, 35, 113 – 143.

Toulmin, P., Barton, P.B., 1964. A thermodynamic study of Pyrite and Pyrrhotite. Geochimica et

Cosmochimica Acta, 23, 64l – 671.

Tretjakow, J.D., Schmalzried, H., 1965: Berichte der Bunsengesellschaft für physikalische

Chemie, 69, 396.

Tsoupas, G, Economou-Eliopoulos, M., 2008. High PGE contents and extremely abundant PGE-

minerals hosted in chromitites from the Veria ophiolite complex, northern Greece. Ore Geology

Reviews, 33 (1), 3 – 19.

Wang, C.X. And Yang, G.W., 2005. Thermodynamics of metastable phase nucleation at the

nanoscale. Materials Science and Engineering, 49, 157–202.

Warn, J.R.W., Peters, A.P.H., 1996. Concise Chemical Thermodynamics, Second Edition, Taylor

and Francis e-Library.

Weiser, T.W., 2002. Platinum-group minerals (PGM) in placer deposits. In The Geology,

Geochemistry, Mineralogy and Mineral Beneficiation of Platinum-Group Elements (L.J. Cabri,

Ed.), 721 – 756.

Wood, S.A., 1990. The interaction of dissolved platinum with fulvic acid and simple organic acid

analogues in aqueous solutions. Canadian Mineralogist, 28, 665 – 673.

Wood, S.A., 2005. The effect of organic ligands on themobility of the PGE in soils and natural

waters: Implications for exploration and the environment. Goldschmidt Conference Abstracts

2005, Moscow, Idaho, USA.

102

Wood, S.A., Mountain, B.W., Pan, P., 1992. The aqueous geochemistry of Platinum, Palladium

and Gold: recent experimental constraints and a re-evaluation of theoretical predictions.

Canadian Mineralogist, 30, 955 – 982.

Wood, S.A., Tait, C.D., Vlassopoulos, D., Janecky, D. R., 1994. Solubility and spectroscopic

studies of the interaction of Palladium with simple carboxylic acids and fulvic acid at low

Temperature. Geochimica et Cosmochimca Acta, 58 (2), 625 – 637.

Xiong, Y., Wood, S.A., 2000. Experimental quantification of hydrothermal solubility of platinum-

group elements with special reference to porphyry copper environments. Mineralogy and

Petrology, 68, 1 – 28.

Zaccarini, F., Bakker, R.J., Garuti, G., Aiglsperger, T., Thalhammer, O.A.R., Campos, L., Proenza,

J.A., Lewis, J.F., 2010. Platinum group minerals in chromitite bodies of the Santa Elena Nappe,

Costa Rica: mineralogical characterization by electron microprobe and Raman-spectroscopy.

Boletín de la Sociedad Geológica Mexicana, 62 (1), 161 – 171.

Zereini, F., Dirksen, F., Skerstupp, B. Andurban, H., 1998. Sources of anthropogenic Platinum-

group elements (PGE): Automotive catalysts versus PGE-processing industries. Environmental

Science and Pollution Research, 5, 223–230.

Zientek, M.L., Cooper, R.W., Corson, S.R., Geraghty, E.P., 2002. Platinum-Group Element

Mineralization in the Stillwater Complex, Montana. In The Geology, Geochemistry,

Mineralogy and Mineral Beneficiation of Platinum-Group Elements, (L.J. Cabri, Ed.), 54, 459

– 481.