9HSTFMG*aejjdj+

ISBN 978-952-60-4993-9 ISBN 978-952-60-4994-6 (pdf) ISSN-L 1799-4934 ISSN 1799-4934 ISSN 1799-4942 (pdf) Aalto University School of Chemical Technology Department of Materials Science and Engineering www.aalto.fi

BUSINESS + ECONOMY ART + DESIGN + ARCHITECTURE SCIENCE + TECHNOLOGY CROSSOVER DOCTORAL DISSERTATIONS

Aalto-D

D 18

/2013

Ali B

unjaku T

he effect of mineralogy, sulphur, and reducing gases on the reducibility of saprolitic nickel ores

Aalto

Unive

rsity

Department of Materials Science and Engineering

The effect of mineralogy, sulphur, and reducing gases on the reducibility of saprolitic nickel ores

Ali Bunjaku

DOCTORAL DISSERTATIONS

Aalto University publication series DOCTORAL DISSERTATIONS 18/2013

The effect of mineralogy, sulphur, and reducing gases on the reducibility of saprolitic nickel ores

Ali Bunjaku

A doctoral dissertation completed for the degree of Doctor of Science (Technology) to be defended, with the permission of the Aalto University School of Chemical Technology, at a public examination held at the lecture hall V1 of the school on 1 February 2013 at 12.

Aalto University School of Chemical Technology Department of Materials Science and Engineering Metallurgical thermodynamics and modelling

Supervising professor Professor Pekka Taskinen Thesis advisor D.Sc.(Tech.) Marko Kekkonen Preliminary examiners Professor Rauf Hürman Eric, University of the Witwatersrand, South Africa Professor Merete Tangstad, Norwegian University of Science and Technology, Norway Opponent Professor Timo Fabritius, University of Oulu, Finland

Aalto University publication series DOCTORAL DISSERTATIONS 18/2013 © Ali Bunjaku ISBN 978-952-60-4993-9 (printed) ISBN 978-952-60-4994-6 (pdf) ISSN-L 1799-4934 ISSN 1799-4934 (printed) ISSN 1799-4942 (pdf) http://urn.fi/URN:ISBN:978-952-60-4994-6 Unigrafia Oy Helsinki 2013 Finland

Abstract Aalto University, P.O. Box 11000, FI-00076 Aalto www.aalto.fi

Author Ali Bunjaku Name of the doctoral dissertation The effect of mineralogy, sulphur, and reducing gases on the reducibility of saprolitic nickel ores Publisher Aalto University School of Chemical Technology Unit Department of Materials Science and Engineering

Series Aalto University publication series DOCTORAL DISSERTATIONS 18/2013

Field of research Metallurgy

Manuscript submitted 15 October 2012 Date of the defence 1 February 2013

Permission to publish granted (date) 20 December 2012 Language English

Monograph Article dissertation (summary + original articles)

Abstract The economically important nickel ores can be divided into two types: sulphide and laterite

ores. Because of the recent decline in sulphide nickel reserves, laterites are becoming an increasingly important source of nickel. The challenge facing the production of nickel from laterites is the energy consumption, as the process requires significantly more energy than the one involving sulphides. The high energy consumption and greenhouse gas footprints for laterite ores will be critical sustainability issues affecting the future of the nickel laterite industry. One commonly used processing method for nickel extraction via the pyrometallurgical treatment of laterite ores involves smelting to ferronickel in a rotary kiln-electric furnace process. The reduction roasting of laterite ore in a rotary kiln has a significant impact on the final ferronickel production; the higher the degree of the necessary pre-reduction of the calcine, the greater the energy saving during the smelting in the electric furnace. Better knowledge of nickel-bearing minerals in saprolite ores and their behaviour at high temperatures is important for developing the pre-reduction process in the rotary kiln.

The aim of the thesis reported here was to examine: (i) the thermal behaviour of different saprolite ores in order to determine the microstructures and phases that formed as a result of dehydroxylation and to detect the phases where nickel is predominantly abundant after dehydroxylation; (ii) the reducibility of the ores, in order to explore the relationship between the mineralogy of the ore, the phases formed, and the reducibility, as well as to define the optimal reducing conditions for the ores; (iii) the effect of sulphur on the formation of high-temperature phases during heat treatment and further on the reducibility of saprolite ores.

The results reveal that during heat treatment, three endothermic peaks at 100, 250, and 600°C, caused by the removal of free water, thermal dissociation of goethite, and dehydroxylation reactions were observed. Followed by an exothermic process at ~820°C, which is evidently associated with the crystallisation of new phases. The formation of new phases (talc-like, amorphous, olivine, and pyroxene) appears to depend on the type of minerals in the initial ore and the phases formed during heating were observed to affect the reducibility of the ores. The type of reducing gas also appears to have an impact on the reduction behaviour and metallisation of the components in saprolitic ores. Considering the results from reduction experiments, ~750°C appears to be the optimal reduction temperature for the ores, CO reducing gas is recommended in order to achieve high nickel metallisation, and finally, the addition of sulphur has a great effect on the reducibility of the ore and thus the formation of metallic particles.

Keywords Nickel laterite, mineralogy, dehydration, reduction

ISBN (printed) 978-952-60-4993-9 ISBN (pdf) 978-952-60-4994-6

ISSN-L 1799-4934 ISSN (printed) 1799-4934 ISSN (pdf) 1799-4942

Location of publisher Espoo Location of printing Helsinki Year 2013

Pages 129 urn http://urn.fi/URN:ISBN:978-952-60-4994-6

Tiivistelmä Aalto-yliopisto, PL 11000, 00076 Aalto www.aalto.fi

Tekijä Ali Bunjaku Väitöskirjan nimi Mineralogian, rikin ja pelkistyskaasujen vaikutus saproliittisten nikkelimalmien pelkistymiseen Julkaisija Aalto-yliopisto Kemian tekniikan korkeakoulu Yksikkö Materiaalitekniikan laitos

Sarja Aalto University publication series DOCTORAL DISSERTATIONS 18/2013

Tutkimusala Metallurgia

Käsikirjoituksen pvm 15.10.2012 Väitöspäivä 01.02.2013

Julkaisuluvan myöntämispäivä 20.12.2012 Kieli Englanti

Monografia Yhdistelmäväitöskirja (yhteenveto-osa + erillisartikkelit)

Tiivistelmä Nikkeliä voidaan valmistaa joko lateriittisesta tai sulfidisesta malmista. Sulfidisten

nikkelimalmiesiintymien köyhtyessä on kiinnostus kasvanut yhä enemmän lateriittisia nikkelimalmeja kohtaan. Nikkelin valmistus lateriittisista malmeista vaatii huomattavasti enemmän energiaa kuin valmistus sulfideista. Lateriittien hyödyntäminen onkin toistaiseksi ollut vähäistä, koska ei ole vielä kehitetty menetelmiä, joilla voitaisiin valmistaa taloudellisesti, energiatehokkaasti ja ympäristöystävällisesti riittävän korkealaatuista tuotetta.

Nykyiset, lateriittia käyttävät pyroprosessit perustuvat raaka-aineen kuivatukseen, esikuumennukseen ja esipelkistykseen rumpu-uunissa sekä sähköuuni-pelkistykseen. Rumpu-uunissa suoritettavalla esipelkistyksellä on tärkeä merkitys sähköuunin energiankulutuksen kannalta; mitä parempi pelkistysaste rumpu-uunissa saavutetaan, sitä vähemmän energiaa tarvitaan sähköuunissa. Tutkimalla saproliittisten nikkelimalmien käyttäytymistä korkeissa lämpötiloissa, saadaan tärkeää tietoa niiden pelkistymiseen vaikuttavista tekijöistä.

Tämän väitöskirjan tavoitteena oli tutkia: (i) lateriittisten nikkelimalmien käyttäytymistä korkeassa lämpötilassa; (ii) lateriittien pelkistymistä määrittämällä malmien mineralogian, kuumennuksen aikana muodostuvien faasien ja pelkistymisen välisiä suhteita ja selvittää lateriittisten nikkelimalmien optimaaliset pelkistysolosuhteet; sekä (iii) rikin vaikutusta lateriittisten nikkelimalmien kuumennuksessa muodostuviin faaseihin sekä pelkistymiseen.

Tulokset osoittivat, että lateriittiset nikkelimalmit ovat heterogeenisiä ja kuumennuksen seurauksena niiden vapaa vesi poistuu noin 100°C:ssa, götiitin OH-ryhmä vapautuu 250°C:ssa ja kidevesi haihtuu noin 600°C:ssa. Uusien faasien muodostuminen eksotermisena reaktiona tapahtuu noin 820°C:ssa. Lisäksi havaittiin, että uusien faasien (talkki, kiteytymätön, oliviini ja pyroksiini) muodostuminen riippuu lateriittisen nikkelimalmin mineralogiasta ja että nämä muodostuneet faasit vaikuttavat edullisesti/epäedullisesti lateriittien pelkistymiseen. Myös pelkistimen valinnalla näytti olevan vaikutusta komponenttien metallisoitumiseen. Optimaalisen pelkistyslämpötilan todettiin olevan ~750oC. Lisäksi CO:n käyttö pelkistimenä johti nikkelin parempaan pelkistymiseen, myös rikillä näytti olevan parantava vaikutus nikkelin pelkistymiseen.

Avainsanat Lateriittinen nikkelimalmi, mineralogia, dehydraatio, pelkistys

ISBN (painettu) 978-952-60-4993-9 ISBN (pdf) 978-952-60-4994-6

ISSN-L 1799-4934 ISSN (painettu) 1799-4934 ISSN (pdf) 1799-4942

Julkaisupaikka Espoo Painopaikka Helsinki Vuosi 2013

Sivumäärä 129 urn http://urn.fi/URN:ISBN:978-952-60-4994-6

PREFACE The research work of this thesis was carried out in the Laboratories of Metallurgy and

Metallurgical Thermodynamics and Modelling of the Departments of Materials Science

and Engineering, Aalto University School of Chemical Technology, during the years

2008-2011.

First of all, I would like to express my sincere appreciation and gratitude to my

supervisor, Prof. Dr. Pekka Taskinen, for his invaluable supervision, kind support,

endless patience, and continuous guidance in the preparation of this thesis.

I also wish to take this opportunity to express my special thanks to Professor Emeritus

Lauri Hoappa, Marko Kekkonen, and Michael Friman for their interest and valuable

suggestions and comments on my work. The staff of the Laboratories of Metallurgical

Thermodynamics and Modelling and Metallurgy are also acknowledged for offering help

with this work when it was needed.

Outotec (Finland) Oy and Outokumpu Oyj, which supported this investigation

financially, are greatly appreciated. My thanks also go to the Research Foundation of

Helsinki University of Technology, which helped to get this work done.

I am also grateful to the Outotec Research Centre for the XRD, mineralogical, and

chemical analyses and microanalyses.

Finally, I feel indebted to my family and my wonderful sons Arian and Arbnor, whose

encouragement, support, and existence helped me to accomplish this study. I would also

like to thank my wife Lindita for her infinite ability to motivate me.

Espoo, 3rd January 2013

Ali Bunjaku

LIST OF PUBLICATIONS The thesis is mainly based on the following scientific articles, which are referred to in

the text by their Roman numerals.

I. Bunjaku, A., Kekkonen, M. and Holappa, L. (2010). Phenomena in thermal

treatment of lateritic nickel ores up to 1300°C, The Twelfth International

Ferroalloys Congress Proceedings (ed: A. Vartiainen), pp. 641-652

(Outotec Oyj: Helsinki, Finland).

II. Bunjaku, A., Kekkonen, M., Taskinen, P. and Holappa, L. (2011). Thermal

behaviour of hydrous nickel-magnesium silicates when heating up to

750°C, Min. Process. Extract. Metall. (Trans. Inst. Min. Metall. C), 120(3),

139-146.

III. Bunjaku, A., Kekkonen, M. and Taskinen, P. (2012). Effect of mineralogy

on reducibility of calcined nickel saprolite ore by hydrogen, Min. Process.

Extract. Metall. (Trans. Inst. Min. Metall. C), 120(1), 16-22.

IV. Bunjaku, A., Kekkonen, M., Pietilä, K. and Taskinen, P. (2012a). Effect of

mineralogy and reducing agent on reduction of saprolitic nickel ores, Min.

Process. Extract. Metall. (Trans. Inst. Min. Metall. C), 121(3), 156-165.

V. Bunjaku, A., Kekkonen, M., Pekkarinen, S. and Taskinen, P. (2012b).

Effect of sulphur addition on the roasting reduction of saprolitic nickel

ores in CO/CO2 gas mixture, Min. Process. Extract. Metall. (Trans. Inst.

Min. Metall. C). The paper was accepted for publication on 31.08.2012.

AUTHOR’S CONTRIBUTION I-III. Ali Bunjaku was responsible for the research plan, together with his co-

authors, and carried out the experimental work, including the data

collection, himself. The Outotec Research Centre prepared the pellets for

the experiments and carried out the XRD and mineralogical analyses and

microanalyses of the samples. Ali Bunjaku interpreted the results together

with his co-authors and he played a major role in writing the manuscripts.

IV. Ali Bunjaku was responsible for the research plan in cooperation with his

co-authors, supervised the experimental work, and wrote the manuscript.

The Outotec Research Centre prepared the pellets for the experiments and

carried out the XRD and mineralogical analyses and microanalyses of the

samples.

V. Ali Bunjaku was responsible for the research plan, together with his co-

authors, supervised the experimental work, and wrote the manuscript. The

Outotec Research Centre prepared the pellets for the experiments and

carried out the XRD and mineralogical analyses and microanalyses of the

samples.

�

TABLE OF CONTENTS

1 INTRODUCTION ......................................................................................... 1

1.1 Lateritic nickel ores ....................................................................................................... 1

1.2 Processing of lateritic nickel ores .................................................................................3

1.3 Energy consumption in nickel production from laterite ores .................................... 6

1.4 The aim of the study ..................................................................................................... 7

2 THERMODYNAMICS OF SAPROLITIC NICKEL ORES ............................... 9

2.1 Ni-Mg-Si-O stability diagrams .................................................................................... 9

2.2 Activities of NiO in silicate solid solutions of a MgO-NiO-SiO2 system .................. 12

2.3 Isothermal phase diagram of Fe-Ni-O (LogpO2 vs Composition) and

thermodynamic system Fe3O4-FeO-NiO-Fe-Ni with CO2/CO and H2O/H2 gas

mixture ........................................................................................................................ 15

3 EXPERIMENTAL ....................................................................................... 17

4 THERMAL BEHAVIOUR OF SAPROLITIC NICKEL ORES ........................ 22

4.1 Dehydration and recrystallisation of saprolites ....................................................... 22

4.1.1 Results and discussion ................................................................................... 23

4.2 Effect of sulphur on recrystallised phases ................................................................. 27

4.2.1 Results and Discussion ................................................................................... 27

5 REDUCTION OF SAPROLITIC NICKEL ORES .......................................... 30

5.1 Results and Discussion .............................................................................................. 30

5.1.1 Isothermal Reduction ..................................................................................... 30

5.1.2 Non-isothermal Reduction ............................................................................ 34

5.1.3 Effect of additives (sulphur, carbon) on saprolitic nickel reduction ........... 36

6 CONCLUSIONS ........................................................................................ 39

7 REFERENCES ........................................................................................... 42

1

1 INTRODUCTION

Nickel is commonly present in two principal ore types – sulphide ores and oxidic laterite

ores. The reserves of nickel ores in the world are estimated to be at least 24,000 million

tons, of which laterites represent 73% (Pariser et al. 2010). Historically, production has

been dominated by sulphide ores, and laterites currently account for only ~48% of world

nickel production (Pariser et al. 2010). The principal reason for this is that the laterite

ores typically require substantially more energy and chemicals to produce nickel than

sulphide ones (Diaz et al. 2004, Mäkinen et al. 2008, Eckleman 2010). Because of the

recent decline in sulphide nickel reserves, laterites are becoming an increasingly

important source of nickel and thus the growth in nickel production in the future is

expected to come from laterite ores.

1.1 Lateritic nickel ores

Lateritic nickel ores are geographically found in tropical and subtropical regions

(Golightly 1979, Burger 1996), where there is an abundance of rainfall and decaying

vegetation which together provide acidic ground waters that are effective in the

weathering of nickel olivine-rich rocks (peridotite, pyroxenite, and dunite) which have

primary initial nickel contents. The acidic ground waters attack and dissolve

magnesium, iron, cobalt, and nickel, while silica is colloidally suspended in the solution.

Since the ground waters are not highly acidic, a portion of the iron dissolved in the

ground waters is oxidised to ferric iron and precipitates as ferric goethite. Cobalt is also

precipitated, along with a portion of iron. As the iron-depleted ground waters percolate

through the underlying soil and rock, the ground waters are partially neutralised by the

magnesia in the underlying rock and nickel, along with further amounts of iron and

cobalt, is precipitated as hydrous nickel-magnesium silicates (saprolite).

Factors of importance for the formation of well-developed laterite deposits (Fig. 1)

with parallel relative concentrations of nickel, cobalt, chromium, aluminium, and iron

include:

� the mineralogy of the peridotite and its tectonic setting;

� the climate;

� the topography;

� the geomorphic history.

No single factor dominates the formation of Ni lateries, but, combined in a dynamic

system, each may act as a major element in the processes and rates of development,

2

ultimately controlling the characteristics of any deposit. The factors that influence the

development of Ni laterites in one region may be less meaningful elsewhere (Barros et

al. 1992, Brand et al. 1998, Gleeson et al. 2004). Golightly (1979) proposed that the

formation rates of laterite profiles could be 20 m millionyears-1 under ideal conditions,

and mature profiles could form in one to five million years, with 20 to 40 m being usual.

The nickel profile in the Earth’s crust is divided into three layers or zones that depend

on depth, mineralogy, and chemistry (Fig.1); the limonite (red and yellow), transition,

and saprolite layers, respectively (Golightly 1979, Hallett 1997, O’Connor et al. 2006,

Botsis et al. 2008). The limonitic layer, which includes limonite and part of the

transition zone, is high in iron and relatively low in nickel (1 to 1.5% Ni) and magnesia

(Diaz et al. 2004). The transition layer is dominated by Ni-rich, smectites such as

nontronite and saponite, which are also commonly found in the upper saprolite (Elias et

al. 1981). The saprolite deposit section extends deeper than the limonitic zone and is

situated in the lower part of the profile. The silicates are mainly nickeloan varieties of

serpentine, talc, chlorite, and sepiolite. The nickel (2 to 2.5% Ni), magnesia, and silica

contents are relatively high and the iron content falls off (Hang et al. 1973, Springer

1974, Chen et al. 2004). These kinds of layers of nickel deposits are very important in

nickel extraction (Thurneyssen et al. 1960, Coleman et al. 1960, Diaz et al. 1988, Kerfoot

1991).

Figure 1. The Earth’s crust, appropriate processing technology and approximate chemical assay of nickel profile.

Additionally, on the basis of their iron and MgO contents, the lateritic nickel ores that

have been commercially processed by rotary kiln-electric furnace techniques are further

classified into 5 categories (Dor et al. 1979, Diaz et al. 1988 and 2004, Utigard 1994,

Zevgolis et al. 2010).

Layer name MineralogyAproximate Analysis Appropriate

processing Technology Ni Co Fe MgO

Red Limonite

HematiteFe2O3

<0.8 <0.1 >50 <0.5

Yellow Limonite

Goethite(Fe,Al,Ni)O(OH)

o.8-

1.5

0.1-

0.2

40-

50

0.5-5

Transition Nontronite/SmectiteNa0.3Fe2(Si,Al)4O10(OH)2.nH2O

1.5-2

0.02-

0.1

25-

40

5-

15

SaproliteGarnerite/Serpentine(Mg,Ni)3Si2O5(OH)4

1.8-3

10-

25

15-

35

UltramaficRock

Olivine(Mg,Ni,Fe)2SiO4,

Pyroxene(Mg,Ni,Fe)SiO3

0.3 0.01 535-

45

AcidLeach

CaronProcess

Smelting

3

� Saprolite ores which are low in Fe (< 10-12%) and high in MgO (> 25-40%).

� Limonitic ores which are low in MgO (< 5-10%), subdivided into:

– higher in Fe (> 30-40%).

– lower in Fe (< 30-40%).

� Transition ores which contain 12-15% Fe, subdivided into:

– higher in MgO (20-35%).

– lower in MgO (10-25%).

1.2 Processing of lateritic nickel ores

Nickel is extracted from laterite ores by hydro- or pyrometallurgical routes (Fig. 2),

depending on the nature of the gangue. In general, the silicate laterites with high

magnesia contents can be treated by means of reduction roasting followed by ammonia

leaching or by pyrometallurgical processes, whereas low-magnesia and high-iron

limonitic ores are usually treated by means of hydrometallurgical processes (Diaz et al.

1988). Approximately 40% of the lateritic nickel ores that are suitable for smelting are

estimated to be in the form of hydrous nickel-magnesium silicates (saprolite ores) (Diaz

et al. 1988 and 2004, Taylor 1995, Bergman 2003, INSG 2009).

Figure 2. Processing routes for nickel production from both sulphide and laterite ores.

The processing of lateritic nickel ore is demanding. Unlike nickel sulphide ores, in

which the sulphide minerals usually occur as distinct grains in the rock matrix, enabling

the ores to be concentrated easily and cheaply by the conventional dressing methods,

the complex mineralogy, heterogeneous nature, and low nickel content of the nickel

laterites not only make physical beneficiation almost impossible but also make nickel

4

extraction quite difficult (Kerfoot 1991). Additionally, the high temperatures required for

pyrometallurgical processes or the consumption of high levels of chemicals to leach the

ore for hydrometallurgical processes make the processing of laterite ores relatively

complex. Therefore, the optimisation of the metallurgical laterite processing methods

constitutes a great challenge for the nickel industry, as any additional demand for nickel

is expected to be satisfied by the mining of laterite deposits (Eckelman 2010, Mudd

2010, Pariser et al. 2010).

Nevertheless, several pyrometallurgical processes have been adopted to recover nickel

from saprolitic ores. Probably the oldest technique is that of matte smelting, in which

the ore is first sintered and then smelted with or without gypsum in a blast furnace to

form an iron-nickel matte or nickel pig iron (Guo 2009, Pietilä 2011, Yucel et al. 2012). A

second method employs rotary kiln-electric furnace techniques to produce a ferronickel

containing 20 to 45% nickel (Dor et al. 1979). The ore is first roasted and pre-reduced in

a rotary kiln and then smelted in an electric furnace with carbon to reduce the nickel

and a portion of the iron contained in the ore to a metallic state (Thurneyssen et al.

1960, Murati et al. 1986, Mehmetaj et al. 2001, Solar 2008, Zhu et al. 2008, Walker et

al. 2009). A third technique, known as the Ugine process, employs a ferro-silicon

reductant. The ore is first smelted in an electric furnace without a reductant and the

molten ore is then transferred to a ladle, thus to be reduced (Coleman et al. 1960). In

addition, the process which is called Nippon Yakin Oheyama is completely unique and

offers an adequate source of nickel for stainless steel making (Watanabe et al. 1987). The

ore is smelted in a rotary kiln without liquefaction to produce ferronickel luppen (the

Krupp-Renn process), which is magnetically separated from the gangue (Matsumori et

al. 1997, Tsuji 2012). The Caron process (Caron 1950) has also been utilised for the

production of nickel from all nickel and cobalt ores that originated from the weathering

of peridotites since 1944, and to date it has gone on-stream commercially at Nicaro,

Punta-Gorda, and BHP Billiton Yabulu. In general the process results in the satisfactory

treatment of nickel limonite ores. But the process has proven less effective when applied

to the transition layer of a nickel ore profile (Fig. 1) (Apostolidis et al. 1978, Kukura et al.

1979, De Graaf 1980). One of the industrial applications of the process is a modified

Caron process used at BHP Billiton Yabulu, Australia (Fittock 2012). In the process,

nickel laterite ores are processed through reduction roasting in a reducing atmosphere,

followed by ammoniacal leaching and the precipitation of an intermediate basic nickel

carbonate (BNC) product (Rhamdhani et al. 2008 and 2008a, Hidayat et al. 2009,

2009a, 2009b). Of these, the rotary kiln-electric furnace-based process is nowadays the

main method used for the processing of saprolitic nickel ore (Bergman 2003, Warner et

al. 2006, Norgate et al. 2011).

The majority of pyrometallurgical processes (ferronickel and matte smelting) aim at

separating nickel and cobalt from iron, silicon, and magnesium, which are the principal

elements accompanying nickel in laterite ores. For a given addition of reductant, the

recovery of nickel is the highest, followed by that of cobalt, and the recovery of iron to

5

the metal phase is the lowest. Losses of nickel to the electric furnace slag under these

conditions (for 25% Ni in the ferronickel) are quite low (0.1% Ni in the slag), but above

45% Ni in the ferronickel, slag losses increase rapidly (Gomez et al. 1979, Solar 2008). A

nickel recovery level of 87-97% was achieved with this technology in smelting (Solar

2009). Furthermore, in an electric furnace, depending on the electrical conditions in the

furnace and the characteristics of the slag (which depend on the SiO2/MgO ratio), the

temperature difference between the metal and slag is within a certain range, generally

between 100 and 200°C at the metal-slag interface (Antola 1993, Utigard 1994, Bunjaku

et al. 2006 and 2009).

The economics of the processes (pyrometallugical and Caron) are affected by the

reduction roasting, which has a significant impact on the final ferronickel or nickel

production; the higher the degree of pre-reduction of the calcine, the greater the nickel

recovery and the higher the grade of the ferronickel formed during the smelting in the

electric furnace (EF) respectively, and the lower the consumption of high levels of

chemicals to leach the reduced ore for hydrometallurgical processes (Caron process)

(Caron 1950, Rhamdhani et al. 2009b). In addition, through the pre-reduction, the ratio

of Ni to Fe can be controlled by the amount of solid reductant added to the process in

order to obtain an appreciable grade of Fe-Ni alloys with high levels of nickel recovery

(Diaz et al. 2004, Solar 2008 and 2009).

Furthermore, the problem posed by the presence of MgO and SiO2 contents is that

they significantly affect not only the degree of pre-reduction, but also the operating

temperature used in the subsequent steps of the process. In general, the greater the

quantities of these constituents are, the higher the temperature that is necessary to

produce a liquid phase medium, and the lower the pre-reduction that is obtained (Chen

et al. 2005, Taskinen et al. 2005, Reinecke et al. 2007). Additionally, the greater the

nickel metallisation achieved in the pre-reduction of laterite ores, the higher the amount

of nickel is found in the acid solution (Caron 1950, O’Connor et al. 2006).

Besides the thermal transformation of lateritic nickel ores through the reduction

roasting process in a rotary kiln, there is also a series of chemical reactions and phase

transformations (Taylor 1962, Ball et al. 1963, Brindley et al. 1975, MacKenzie et al.

1994, Chen et al. 2009). Undoubtedly, those reactions, which depend on the mineralogy

of the ore, should further affect the progress of the reduction. The reduction of nickel-

bearing hydrosilicate minerals (saprolite) is insignificant until the temperature has been

raised to a point at which the dehydration of the minerals commences. However, after

the nickel-bearing hydrosilicate minerals are recrystallised into new phases at ~820°C,

the reduction of nickel oxide becomes considerably more difficult (Hallett 1997, Li et al.

2000, Rhamdhani et al. 2008). Therefore, the mechanism for preventing the phase

transformation of the minerals, especially serpentine or other groups of silicates, in the

6

reduction roasting process is important for the improvement of the overall nickel yield

from saprolitic ores.

In the case of nickel production from nickel oxides (NiO) formed by the calcination of

an intermediate basic nickel carbonate product (Rhamdhani et al. 2008a, Hidayat et al.

2009), the vapour deposition of a chloride solution (Utigard et al. 2005), or roasted

nickel sulphide pellets (Plascenia et al. 2009), the residual oxygen and overall reduction

rate are the key parameters determining the specification and value of the final Ni

product and the production rates, respectively. The European Union’s new chemical

policy regulations (Bergeson 2003) require oxygen concentrations in the nickel product

to be less than 0.1 wt% as nickel oxide. This limitation is initiated by the need to

minimise the generation of residual NiO dust, which has been shown to be carcinogenic.

1.3 Energy consumption in nickel production from laterite ores The challenge facing the production of nickel from laterites is the energy consumption,

as this requires significantly more energy per unit than the use of sulphides (2-3 times)

(Dasher 1976, Nriagu et al. 1980, Blanco et al. 1981, Eckleman 2010). Some reasons for

the higher energy requirements for laterites include these: laterites cannot be

significantly upgraded or concentrated before processing, and hence virtually all of the

ore goes to the energy-intensive metal extraction stage, e.g. all of the ore goes through

the rotary kiln-electric furnace for ferronickel production; in the high-pressure acid

leaching (HPAL) process the whole of the ore goes through the leaching stage;

differences in ore grade – laterites generally have lower grades than sulphides and hence

more ore must be processed to produce the same amount of nickel metal; the sulphur in

sulphides acts as a fuel source which is not present in laterites (Norgate et al. 2011).

Moreover, as the higher-grade reserves of nickel ores are progressively depleted, the

grades of the ore coming from the mines will gradually become poorer. This reduction in

the grades will have a dramatic effect on the energy consumption (Fig. 3), and further on

the accompanying greenhouse emissions during metal production (Norgate et al. 2011).

In addition to the grades of the laterites, the energy source used to generate the

electricity consumed in the nickel production process also influences the overall costs of

the process and the environmental impact.

7

Figure 3. Primary energy consumption for production of one ton of nickel product (according to data from Norgate et al. 2011).

Energy is consumed for different purposes, depending on the process. The

pyrometallurgical processing of laterite ores has high fossil and electrical energy

requirements, as laterites have a high water content (free moisture and combined water)

and high levels of magnesia, which leads to a high melting temperature. For

pyrometallurgical processes, fossil fuel is consumed both as a source of heat and as a

reductant. Hydrometallurgical processing requires energy to heat the acid solvent and to

pressurise the chambers (high-pressure acid leaching, HPAL).

The mining, mineral processing, and metal production sector, like other industrial

sectors, has come under increased pressure to reduce the amounts of energy it consumes

and greenhouse gases it emits. Given the anticipated increase in nickel metal production

from laterite ores, as mentioned earlier, these higher energy and greenhouse gas

footprints for laterite ores will be critical sustainability issues affecting the future of the

nickel laterite industry (Norgate et al. 2011).

1.4 The aim of the study In conventional pyrometallurgical processing, the laterite ore is dried, calcined, and

usually pre-reduced in a rotary kiln before being smelted in an electric furnace in the

presence of carbon.

The electric smelting step is the obvious major consumer of energy. The reduction

roasting of laterite ore (saprolite) in a rotary kiln has a significant impact on the final

ferronickel production; the higher the degree of the necessary pre-reduction of the

0

200

400

600

800

1000

1200

1400

1600

0 0.5 1 1.5 2 2.5 3

Em

bo

die

d e

ne

rgy

[GJ

/t N

i]

Laterite ore grade [%Ni]

CaronHPALFerronickelHeapleachAL

8

calcine, the greater the energy saving during the smelting in the electric furnace.

Therefore, better knowledge of nickel-bearing minerals in saprolite ores and their

behaviour at high temperatures is important for developing the pre-reduction process in

a rotary kiln.

The experimental work was divided into two parts. It started with studying the thermal

behaviour of saprolitic nickel ores and then continued with the reduction of the same

ores.

The main objectives of this work were to examine the following.

� The thermal behaviour of different saprolite ores in order to determine the

microstructures and phases that formed as a result of dehydroxylation and to

detect the phases where nickel is predominantly abundant after dehydroxylation

(Papers I and II).

� The reducibility of the same ores in order to explore the relationship between the

mineralogy of the ore, the phases that formed, and reducibility, as well as to

define the optimal reducing conditions for the ores that were investigated

(Papers III and IV).

� The effect of additives (sulphur) on the formation of high-temperature phases

during heat treatment and further on the reducibility of saprolite ores (Paper V).

9

2 THERMODYNAMICS OF SAPROLITIC NICKEL ORES

Although the overall reduction reaction of nickel oxide (NiO) with reducing gas appears

to be a simple reaction (Crowe et al. 2003, Rhamdhani et al. 2008, Plascenia et al.

2009), the major challenge in the reduction of saprolitic nickel ores is ensuring that the

nickel is in reducible form (Papers III and IV, De Graaf 1979, Nath et al. 1995, Chen et

al. 2009). In an actual pyrometallurgical extraction process, ferronickel alloys, rather

than metallic nickel, are formed in spite of their different standard Gibbs free energies of

formation (Szekely et al. 1971, Dalvi et al. 1970 and 1976,).

Therefore, knowledge of the thermodynamics of the MgO-FeO-NiO-SiO2 system is of

great value in understanding, for example, the selective reduction of nickel from

hydrous magnesium-nickel silicates (Campbell et al. 1968, Shirane et al. 1980,

Mukhopadhyay et al. 1995, Hallett 1997).

2.1 Ni-Mg-Si-O stability diagrams The thermal transformations of hydrous nickel-bearing magnesium silicates are of

interest both in themselves and for their potential usefulness as a further method for

studying the nature of the initial minerals (Hang et al. 1973).

According to the literature (Papers I and II, Ball et al. 1963, Brindley et al. 1965 and

1975, Faust 1966, Mackenzie 1970, Hang et al. 1973, MacKenzie et al. 1994), during

roasting the minerals of saprolitic nickel ore dehydrate, dehydroxylate, and crystallise to

new phases at specific temperatures. Consequently, the nickel would be bonded in newly

crystallised phases and thus the reduction of nickel oxide becomes more difficult

(Papers III and IV, Kawahara et al. 1988, Utigard et al. 1993, Nath et al. 1995, Li et al.

2000, Rhamdhani et al. 2009a, Harris et al. 2009). Some silicates, however,

dehydroxylate and recrystallise simultaneously (Kukura et al. 1979, MacKenzie et al.

1994, Gürtekin et al. 2006). Therefore, the mechanism for preventing the phase

transformation of minerals, especially serpentine and other groups of silicates, during

the reduction roasting process is important for the improvement of the overall nickel

yield from saprolitic ores. Sulphur is an additive that is considered to be beneficial in

increasing metal extraction (Paper V, Siemens et al. 1976, Kukura et al. 1979, Valix et al.

2002a, Pekkarinen 2011).

1

w

si

o

th

c

c

m

in

p

1

S

p

p

in

(D

a

Fti

0

Serpentine min

widespread and

ilicates (Burger

olivine (Mg,Fe,Ni

he final lattice

rystallographica

In addition to s

ompositions) ar

minerals to recry

nitial minerals.

pyroxene as the m

970, Hang 1973)

Therefore, deta

Si-O system, incl

phases present in

process but also

nterest in the Ni

Davies et al. 200

and 6, respectivel

Figure 4. Isotheie-lines drawn in

nerals where F

often occur as

1996, Botsis e

i)Si2O4, and pyro

e is related to

ally equivalent, or

erpentine miner

re also extracted

ystallisation temp

Thus, after recr

main product; o

).

ailed thermodyna

luding the inform

n the initial ore,

for achieving t

-Mg-Si-O system

02) and the Mto

ly.

ermal section ofn the two-phase e

e and Ni can

s modified prod

et al. 2008). Th

oxene (Mg,Fe,Ni

o that of the

rientational relat

rals, nickeliferous

d for nickel. Th

peratures depen

rystallisation, mi

therwise olivine

amic information

mation on the n

, is vital not onl

the maximum n

m at 800, 850, an

x database (Gisb

f the ternary systequilibria.

to some extent

ducts of olivine

he principal rec

i)SiO3 are formed

original mater

tionships (Taylor

s clay minerals (

he phases forme

d on the concen

inerals with a lo

is formed (Ball

n on the phase e

nickel distributio

y for control of

nickel yield. Thu

nd 900°C, using

by et al. 2007), a

tem MgO-NiO-S

t replace the M

e and magnesiu

crystallisation pr

d topotactically,

rial by one or

r 1962).

(nickel-bearing ta

ed by heating th

ntration of nickel

ow nickel conte

et al. 1963, Mac

equilibria in the N

n between the m

the reduction ro

us, the compou

the MTDATA so

are shown in Fig

SiO2 at 800°C w

Mg are

um-rich

roduct,

so that

r more

alc-like

he clay

l in the

nt give

ckenzie

Ni-Mg-

mineral

oasting

unds of

oftware

gs. 4, 5,

with the

11

Figure 5. Isothermal section of the ternary system MgO-NiO-SiO2 at 850°C with the tie-lines drawn in the two-phase equilibria (Paper IV).

Figure 6. Isothermal section of the ternary system MgO-NiO-SiO2 at 1000°C with the tie-lines drawn in the two-phase equilibria.

On the basis of Figs. 4-6, a partial isothermal section of a ternary MgO-NiO-SiO2

system at 800, 850, and 1000°C indicates that the solubility of NiSiO3 in the pyroxene

12

solid solution depends on the temperature. Added to this, at low concentrations of

Ni2SiO4 in olivine, cristobalite reacts with the olivine to form orthopyroxene, indicating

the thermal stability of the protopyroxene solid solution expanding the orthopyroxene

phase field of low NiO solubility to olivine+protopyroxene above ~825°C. In addition to

a partial metasilicate solid solution above ~825°C, there is a complete solid solution

between Mg2SiO4 and Ni2SiO4.

2.2 Activities of NiO in silicate solid solutions of a MgO-NiO-SiO2 system Metal losses during extractive metallurgical processes are inevitable (Kerfoot 1991,

Daenuwy et al. 1997, Rigopoulos et al. 2004, Solar 2009). However, in the

pyrometallurgical processing of saprolitic nickel ore, the major portion of the nickel loss

during smelting is not only due to the transfer of nickel from the alloy phase to the slag

phase but also to low NiO activity in the phase formed in the previous reduction roasting

stage. Therefore, the thermodynamic behaviour of the components in the oxide solid

solution is important for the development of an understanding of the solid state

reduction of nickel from saprolite minerals.

According to the literature (Mukhopadhyay et al. 1995, Shirane et al. 1980, Kiukkola &

Wagner 1957), the NiO activity in ortho- and metasilicate solid solutions, which are the

main components of saprolite after dehydroxylation, can be determined by a galvanic

cell incorporating a solid oxide electrolyte. The measurements of the e.m.f:s

(electromotive force) of high-temperature electrochemical cells incorporating

electrolytes have provided accurate and valuable thermodynamic data concerning many

solid solution systems.

Additionally, the NiO activity in the oxide solid solution of the system within the

temperature range of interest is defined by substituting e.m.f values into Equation (1).

(1))/()(ln RTnFEaNiO ��

Here NiOa is the activity of NiO in the oxide solid solution, n is a unit mol of electrons

which participate in the oxygen reaction in the cell, F is the Faraday constant

(96485.3365 [C/mol]), E is the electromotive force determined by the galvanic cell, R

is the gas constant (8.3144621 [J/molK]), and T is the experimental temperature.

A number of researchers (Campbell et al. 1968, Dalvi et al. 1976, Shirane et al. 1980,

Mukhopadhyay et al. 1995) have reported that the activity of NiO in the spinel solution

deviates greatly towards the negative side of Raoult’s law from an ideal solid solution.

Referring to Fig. 7, the calculated activity curves of NiO also deviate towards the

negative side of Raoult’s law in an ideal solid solution. It appears that the system is

converted

ai=Ni, cor

orthosilica

temperatu

Figure 7.calculeted

However

of calculat

some degr

al. 1968,

decompos

atmospher

Moreove

areas, in

superimpo

as the stan

into an ideal so

rresponding to

ate, in principle

ures if it is to be e

. Activity of NiOd by the Mtox dat

r, contrary to sap

ted NiO activitie

ree toward the po

Shirane et al. 1

ses in metasilicat

re.

er, in order to sh

Fig. 9 the iso

osed on the Gibb

ndard state.

olid solution, and

Raoult’s law w

e, needs not onl

effectively reduce

O as a function otabase (Gisby et a

prolitic nickel or

es in metasilicate

ositive side of Ra

1980). Therefore

te as a result of h

how the variabil

oactivity contour

bs triangle of Fe-

d approaches, to

with increasing

ly strong reduct

ed.

of composition inal. 2007).

re of the orthosil

e solid solution,

aoult’s law from

e, the amount o

heating will be re

lity of nickel ox

rs of NiO, as

-O-MgO-NiO-SiO

o some degree, a

temperature. T

tion conditions

n the system Mg

licate-forming ty

as shown in Fig

an ideal solution

of saprolitic nick

educed with a sl

ide activity in d

calculated by M

O2 at 800°C, wit

13

a straight line,

Therefore, Ni-

but also high

g2SiO4-Ni2SiO4

ype, the curves

g. 8, deviate to

n (Campbell et

kel ore which

ight reduction

different phase

MTDATA, are

th halite phase

14

F

FNsi

4

Figure 8. Activit

Figure 9. IsotheNiO-SiO2 at 800�ilicates as termin

ty of NiO as a fun

ermal quasitern�C and an oxygenal phases.

nction of compos

ary section of then pressure PO2=

sition in the syst

he five-compone 2.12782 104 [Pa

ems MgSiO3-“Ni

ent system Fe-Oa] with the olivin

iSiO3”.

O-MgO-ne-type

15

2.3 Isothermal phase diagram of Fe-Ni-O (LogpO2 vs Composition) and thermodynamic system Fe3O4-FeO-NiO-Fe-Ni with CO2/CO and H2O/H2 gas mixture In addition to serpentine, goethite ((Fe,Ni)OOH) is often present in saprolitic nickel

ores. Its calcination in a reducing gas is transformed to magnetite [spinel or (Fe,Ni)3O4]

and further to halite [Ni-bearing wustite or (Fe,Ni)O] solid solutions (Paper I, O’Connor

2006, Rhandhani et al. 2009). Therefore, knowledge of the thermodynamic behaviour of

the Fe-Ni-O system is important for understanding the selective reduction of nickel from

mixed iron-nickel oxides. Hence, the Fe-Ni-O isopleths at various temperatures against

the Fe-Ni ratio are depicted in Fig. 10.

Figure 10. The isopleth of the Fe-O-Ni system showing the equilibrium phases over an oxygen pressure range of PO2=104 to 10-15 Pa (0.1-10-9 atm) at 800-1000°C (Paper III).

As one can see, nickel oxide will be reduced more readily than iron oxide at a fixed PO2.

Furthermore, for the coexistence of halite and an FeNi alloy shown in Fig. 10, the oxygen

pressure increases with increasing nickel content, extending from the Fe/FeO

equilibrium to the (Fe,Ni)O-spinel-alloy equilibrium.

However, corresponding to the various oxygen partial pressures and temperatures at

which the halite and an FeNi alloy coexist, the concentration of nickel in the FeNi alloy

and nickel oxide in halite phase increases with increasing oxygen pressure at any

temperature between 800 and 1000°C (Paper III).

Commercially, the pre-reduction of the nickel ores is achieved by utilising either

hydrogen or carbon monoxide as the reducing gas. Extensive studies on reduction have

appeared in the literature because of its practical importance in the reduction of nickel

16

ores (Papers III and IV, Antola 1993, Zevgolis et al. 2009) and when defining a model

reaction for the reduction mechanisms (Szekely et al. 1971, Nath et al. 1995, Plascenia et

al. 2009).

Thermodynamically, pure iron and nickel oxide should be reduced to metallic form

under the conditions used in this investigation. However, referring to the diagram using

the HSC chemistry 7.08 software demonstrated in Fig. 11, a rise in temperature would

enlarge the area of metallic nickel and wustite, which would result in less risk of the

wustite being reduced into metallic iron, thus making the reduction of NiO into metallic

nickel easier.

Figure 11. Predominance areas of different phases in the Fe-Ni-O system as a function of log PCO2/PCO, log PH2O/PH2 and temperature (Paper IV).

However, in contrast to the almost easy reduction of pure NiO (Crowe et al. 2003,

Plascenia et al. 2009), Shirane et al. (1980) suggested that the reduction of NiO from

silicates (serpentine and clay minerals) is thermodynamically and kinetically difficult as

a result of the very low activity of NiO in the silicate matrix.

-1

-0.5

0

0.5

1

1.5

2

2.5

3

-1

-0.5

0

0.5

1

1.5

2

2.5

3

550 650 750 850 950lo

g p

CO

2/p

CO

log

pH

2O

/pH

2

Temperature ºC

NiO+CO(g)=Ni+CO2(g) Fe3O4+CO(g)=3FeO+CO2(g)

FeO+CO(g)= Fe+CO2(g) NiO+H2(g)=Ni+H2O(g)

Fe3O4+H2(g)=3FeO+H2O(g) FeO+H2(g)=Fe+H2O(g)

17

3 EXPERIMENTAL

In order to gain better knowledge of the mineralogy of the nickel saprolites and reaction

behaviours of their minerals at high temperatures, which are important for advancing

the pre-reduction process, saprolite ores from Brazil (Mirabela deposit) and Colombia

(Colombia-1 and Colombia-2, Cerro Matoso S.A. (CMSA) deposit) were examined. The

chemical composition, moisture, accessible oxygen, and main phases detected in the

ores are presented in Table 1 and Table 2, respectively.

The ores include nickel hydrous magnesium silicates

(serpentine/vermiculite/chlorite), which are green minerals as a result of their NiO

content. Minerals in this group contain Mg and Si as major cations, with a little Fe, Al,

and Ni (Chen et al. 2004). The Colombian ores consist of serpentine and chlorite group

minerals (Gomez et al. 1979, Gleeson et al. 2004), whereas the Brazilian ore is a mixture

of clay and serpentine (Colin et al. 1990, Vieira Coelhoa et al. 2000). Their total Loss on

Ignition (LOI) contents were measured by heating the ores up to 1100°C under dry

flowing nitrogen. 1100°C was taken as the upper limit for determining the LOI, since the

dehydroxylation of magnesium hydrosilicates occurs at higher temperatures than 900°C

(Papers I and II).

Table 1 Chemical analyses, moisture, and accessible oxygen content of the saprolitic ores used in the experiments (% wt) (Paper I).

Component Colombia-1 Colombia-2 Mirabela Ni 0.76 2.3 2.6 Fe total 7.4 12.4 12.5 Fe 2+ 1.0 2.5 0.73 SiO2 44.9 44.9 49.6 Al2O3 0.77 1.7 1.4 MgO 26.6 14.7 17.2 CaO 0.55 0.36 0.03 Cr2O3 0.20 0.70 1.8 Mn n.a n.a 0.14 Co 0.02 0.04 0.05 Cu - - <0.01 S 0.23 0.46 0.04 C - - 0.13 Loss On Ignition 13.4 9.4 10.7 SiO2/MgO 1.69 3.05 2.88 Accessible oxygen (%) 3.248 5.607 5.987

18

Table 2 Main phases present in the saprolitic ores (Paper I). Phases Colombia-1 Colombia-2 Mirabela

Clinochrysotile,Mg3Si2O5(OH)4

x x

Antigorite,Mg3Si2O5(OH)4

x

Lizardite, (Mg,Al)3(Si,Al)2O5(OH)4

x x

Clinochlore,(Mg,Al)6(Si,Al)4O10(OH)8

x x

Quartz, SiO2 x x x

Enstatite, MgSiO3 x

Vermiculite,Mg3Si4O10(OH)2

x

Nepouite,(Ni,Mg)3Si2O5(OH)4

x x

Iron silicate, Fe2SiO4

x

Grimaldite,CrO(OH)

x

Muscovite,KAl3Si3O10(OH)2

x

Hematite,Fe2O3

x

Goethite, FeOOH

x x

Forsterite ferroan, (Mg0.6 Fe0.4)2SiO4

x

In the experiments, the ores were first heated up to 1300°C to investigate the changes

that might happen upon heating (Paper I), and secondly, in order to avoid the

exothermic recrystallisation reaction which has been established as occurring around

~820°C, the ores were heated up to 750°C (Paper II). On the other hand, the

temperature of 850°C, which was also rather close to the temperature of the subsequent

reduction experiments, is purposely selected in order to ascertain the temperature of the

thermal transformation of the minerals.

In order to meet the experimental requirements, the air-dried ores were crushed in a

laboratory ball mill (Planetary Mono Mill PULVERISETTE 6), screened in a sieve with a

500-µm mesh size, and thermally treated in a Simultaneous Thermal Analyser (Netzsch

STA 449C Jupiter) (DSC-TG), which allows the measurement of mass changes and

thermal effects at the same time and provides information on phase transformation. The

thermally treated ores were then characterised with an X-ray diffractometer (Bruker

AXS DFocus) (XRD) and a scanning electron microscope-energy dispersive

spectroscope (FEG-SEM Jeol JSM 700F and Oxford INCA-Energy and Wave) (SEM-

EDS) (Paper I and II).

The saprolitic nickel ores (Colombia-1, Columbia-2, Mirabela), blended with various

concentrations of sulphur in the form of pyrite, were subjected into a Simultaneous

19

Thermal Analyser (Netzsch STA 449 C Jupiter) (DSC-TG) in order to examine the effect

of the sulphur on the exo-and endothermic reactions. In a typical experiment, the ores

were heated up under dry flowing argon (purity 99.999%) from room temperature to

1300°C by using a constant heating rate of 10 °Cmin-1 in the DSC-TG. After heating, the

sample was cooled down to room temperature under an argon atmosphere in order to

prevent rehydration and oxidation (Fe+2�Fe+3). The sample was subsequently examined

by means of the XRD and SEM-EDS technique for mineralogical and morphological

determinations (Paper V).

The saprolitic nickel ores, with and without pyrite, used in this study were reduced in

various reducing gases at different temperatures in order to determine the relationship

between the reduction degree of removable oxygen, Rt, and time. Furthermore, the

relationship between the ore mineralogy and reducibility is explored as well (Paper IV).

The air-dried ores were mixed with pyrite for ore-additive mixtures exclusively,

crushed in a laboratory ball mill (Planetary Mono Mill PULVERISETTE 6), and

pelletised with a laboratory pelletising disc (Erweka AR 401; outside diameter 400 mm,

inside height 165 mm, inclination and revolutions per minute can be changed), using

bentonite as a binder, and then dried at 105°C for 18h in order to meet the requirements

of the reduction experiments. Further, the diameter of the pellets was restricted to ~12

mm and the reducing gas flow rate was maintained at 1 dm3min-1 in order to reduce the

effect of mass transfer, which hampers the homogeneous reducing of the ores (Papers

III and IV).

The reduction experiments were accomplished in a thermobalance furnace (Fig. 12),

which allows continuous measurement of mass change. In a typical experiment, a pellet

of approximately 1.5 g and 12 mm in diameter was placed into a platinum basket

hanging from the balance. The furnace was heated up at a constant rate (4 °Cmin-1)

under an N2 atmosphere to the experimental temperature, where the sample was

retained until no weight loss was detected. After that, the atmosphere was changed to

H2/N2 (72%:28%) or CO/CO2 (72%:28%) and reduction started. The preceding

calcination step ensures that the mass loss during reduction represents the loss of

oxygen resulting from the reduction of the iron and nickel oxides. The purities of the N2,

H2, CO, and CO2 gases were 99.5%, 99.9%, 99%, and 99.7%, respectively. The weight

loss of the sample was continuously recorded during the experiment by a METTLER

TOLEDO AB104-S balance. A Pt-10 (wt-%) Rh/Pt thermocouple was installed just below

the pellet to record the temperature of the reaction zone. Subsequent to reduction, the

pellet was cooled down to room temperature in flowing nitrogen. The pellet was then

examined by means of SEM-EDS and XRD for microstructure and phase

characterisation, and by bromine-methanol leaching using the inductively coupled

plasma spectrophotometry (Thermo Scientific Model iCAP 6000) (ICP) technique to

determine the individual metallic elements in the reduced pellet (Papers III, IV, and V).

20

In conventional bromine-methanol leaching using the ICP technique, the reduced

pellets were leached out in a bromine-methanol solution. After that, the individual

metals contained in the extraction liquid were determined with the ICP technique

(Kawahara et al. 1988, Li & Coley 2000).

Figure 12. Schematic diagram of the experimental apparatus: 1. microbalance; 2. box furnace; 3. pellet; 4. weight loss and temperature recorder; 5. thermocouple; 6. furnace temperature controller; 7. flowmeters (Paper IV).

The degree of removable oxygen as Rt, after time t relative to the nickel, iron, and

cobalt in H2/N2 and CO/CO2 is calculated with Equations 2 and 3, since the preceding

calcination step at the corresponding temperature ensured that the mass loss during

reduction represented the loss of accessible oxygen as a result of the reduction of the

iron, nickel, and cobalt oxides (Paper III).

(2)/100*)( OXtOt mmmR ��

Where

(3)OCo

ONi

OFeOx mmmm ���

where Om is the weight of the pellet after moisture removal, tm is the observed

weight of the pellet at time t during reduction, Oxm is the amount of removable oxygen

after moisture removal, OFem is oxygen combined with iron, O

Nim is oxygen combined

with nickel, and OCom is oxygen combined with cobalt.

21

Contrary to the experimental measurements, the oxygen content, oxO , of the ores was

calculated stoichiometrically by the ratio � sox

soxor MeOMe %/%* , where orMe is the

analysed metal concentration in the ore and soxMe% , s

oxO% are the stoichiometric metal

and oxygen concentrations in the oxide, respectively.

22

4 THERMAL BEHAVIOUR OF SAPROLITIC NICKEL ORES

Saprolitic or predominantly silicate ores exhibit relatively lower degrees of weathering

from their precursor rocks and in most cases are still undergoing “laterisation”

(Golightly 1979, Brand et al. 1998). They contain residual amounts of precursor

minerals, and the ores essentially occur in situ. An important group of nickel-bearing

minerals in saprolite ore bodies is the garnierites, which is a general name for all

hydrous nickel-magnesium silicate minerals such as chlorite, clay, and serpentine

silicates (Hang et al. 1973, Brindley et al. 1974, Springer 1974). Apparently, these

components of garnierite usually occur as lizardite (Mg, Fe, Ni, Co)3Si2O5(OH)4,

antigorite (Mg, Fe, Ni, Co)3Si2O5(OH)4, and chrysotile (Mg, Fe, Ni, Co)3Si2O5(OH)4 and,

with a high percentage of nickel, as nepouite (Mg, Fe, Ni)3Si2O5(OH)4 in serpentine,

while vermiculite (Mg, Fe, Ni, Co)3Si4O10(OH)2 and talc (Mg, Fe, Ni, Co)3Si4O10(OH)2

occur as clay, whereas clinochlore (Mg, Fe, Ni, Co, Al)6(Si, Al)4O10(OH)8 is in the

chloride group (Faust 1966, Brindley et al. 1975, Colin et al. 1990, Vieira Coelhoa et al.

2000, Gleeson et al. 2004). Garnierites are related to the fine-grained nature,

inhomogeneous character, poor crystalline order, and intimate mixtures of two or more

components.

4.1 Dehydration and recrystallisation of saprolites

Minerals of saprolitic nickel ores are formed as a result of the hydration of olivine,

pyroxene, and other Mg-rich silicates containing nickel under suitable conditions, the

process of so-called sepentinisation. Completely serpentinised peridotites contain

significant amounts of chemical bonding water, ~13%, which will be released as a result

of heating in a traditional pyrometallurgical nickel production process (Papers I and II,

Valix et al. 2002, Zevgolis et al. 2010). Therefore, the thermal reactions of such minerals

and the related mineralogical changes are very important for the determination of nickel

recovery. Added to this, dehydroxylation processes and the transformation of nickel-

magnesium hydrosilicates into the high-temperature products result in a redistribution

of nickel. Particularly, the reduction of nickel oxide becomes more difficult when the

nickel is bonded in newly crystallised phases (Shirane et al. 1980, Kawahara et al. 1988,

Hallett 1997).

23

In general, it has been suggested that the dehydroxylation of nickel-magnesium

hydrosilicates proceeds by the formation of a disordered phase, (Ball et al. 1963,

Brindley et al. 1965 and 1975, Harris et al. 2009) attributed to a multistage sequence

(MacKenzie et al. 1994) and it should proceed by reaction (4).

� � � phaseDisorderedSerpentine

(4)18,,,9,,,9 27234523 OHOSiCoNiFeMgOHOSiCoNiFeMg �

However, the overall stoichiometry above ~820°C of magnesium hydrosilicates, when

nickel, cobalt, and iron are present, suggests that olivine and/or pyroxene solid solution

phases should form according to the following reaction (Paper I, Hang and Brindley

1973, Brindley and Wan 1975, HSC chemistry 7.08):

� � � � PyroxeneOlivineSerpentine

(5)2,,,,,,,,, 23424523 OHSiOCoNiFeMgSiOCoNiFeMgOHOSiCoNiFeMg ��

4.1.1 Results and discussion

The DSC curves obtained at temperatures of 850 and 1300°C (Figs. 13 and 14) exhibit

three endothermic peaks and one exothermic peak. The low-temperature range up to

100°C causes the removal of free water, while 250°C brings about the dissociation of

goethite mineral, if it is present in ore, and the temperature range between 550 and

650°C the loss of crystalline water, which is often called the dehydroxylation reaction.

The dehydroxylation reaction is followed by the crystallisation of a new phase (olivine

and pyroxene) at ~820°C. In addition to the temperatures of the aforementioned

reactions, there is a temperature interval (650-820°C) in which the recrystallisation of

minerals into new phases is not initiated but an intermediate phase following the

reaction (4) is formed (Paper II).

24

Figure 13. DSC curves for Mirabela, Colombia-1, and Colombia-2 when heated up to 1300°C (Paper I).

Figure 14. DSC curves for Mirabela, Colombia-1, and Colombia-2 when heated up to 850°C.

The TG curves obtained at a temperature of 1300°C (Fig. 15) show the weight loss

(H2O and 2H2O) arising from 2(OH) and 4(OH) in the mineral formula for clay and

serpentine, respectively. The shape of the weight loss curves depends on the dehydration

energy of the compounds. Additionally, the occurrence of iron, aluminium, chromium,

and earth-alkaline elements in the ores might induce a change in the thermal behaviour

and in the shape of the TG curves (Fig. 15) of the silicates. According to Fig. 15, the

additional water is lost slowly over the range 250-550°C, followed by a rapid loss of

-0.52

-0.42

-0.32

-0.22

-0.12

-0.02

0.08

0.18

0.28

0.38

0 200 400 600 800

DSC /(mW/mg)

Temperature / ºCColombia-1 Colombia-2 Mirabela

exo

25

crystalline water in the temperature range 550-700°C. The removal of crystalline water

is related to a weight loss of about 8% resulting from the disintegration of the crystalline

lattice of the minerals. The exothermic reaction involves no mass loss; however, there is

a small mass loss with the reaction at the TG curve (Paper I). The likely reason for this is

the kinetic conditions under which the hydroxyl moves outwards from the minerals or

the stability of some hydrous nickel-magnesium silicates at high temperature.

Figure 15. TG curves for Mirabela, Colombia-1, and Colombia-2 when heated up to 1300°C (Paper I).

The phase analyses of the samples, when heated up to 750°C, show the formation of

disordered phases, apart from the Colombia-1 sample, which demonstrates the

formation of olivine and pyroxene phases (Paper II). The simultaneous dehydroxylation,

closely followed by the recrystallisation, of magnesium silicate to olivine

(Mg,Fe,Ni)2SiO4 and pyroxene (Mg,Fe,Ni)SiO3 by reaction (5) is attributed to this kind

of behaviour. Antigorite, which is found in Colombia-1 ore, has been found to enhance

the formation of both phases (Paper II, Valix 2002, Gürtekin 2006).

On the other hand, the X-ray diffraction results of the Colombia-2 sample (Fig. 17)

induced by heating at 850°C do not, however, support the formation of either pyroxene

nor olivine phases; instead, trevorite (NiFe2O4), aluminium oxide solid solution, and

some disordered phases appear. Therefore, when the saprolite ore bulk mineral is

composed of lizardite, clinochrysotile, and clinochlore, the high-temperature phase

formation temperature is above 850°C. Conversely, both the pyroxene and olivine

phases formed by heating up to 850°C in Colombia-1 (Fig. 16), as well as the pyroxene

and magnesioferrite (MgFe2O4) phases resulting from heating up to 850°C in Mirabela

(Fig. 18), confirm the presence of exothermic peaks in the DSC diagram.

26

It is supposed that the simultaneous dehydroxylation and recystallisation of antigorite

in Colombia-1, and the different mineralogy of Mirabela (clay minerals) caused the

formation of the corresponding phases. Therefore, the mineralogy of the ore has an

influence on its thermal behaviour and further it is also assumed that it controls the

optimal reduction temperature (Paper III and IV).

Figure 16. X-ray diffractograms for Colombia-1 ore after heat treatment at up to 850°C in argon.

Figure 17. X-ray diffractograms for Colombia-2 ore after heat treatment at up to 850°C in argon.

27

Figure 18. X-ray diffractograms for Mirabela ore after heat treatment at up to 850°C in argon.

4.2 Effect of sulphur on recrystallised phases The reduction efficiency of nickel associated with the hydrated iron oxide minerals

(goethite) has historically been much higher than for nickel associated with magnesium

hydrosilicates (De Graaf 1979, Hallett 1997, Valix et al. 2002a, O’Connor et al. 2006,

Chen et al. 2009, Harris et al. 2009, Zevgolis et al. 2010a). The mineralogical character

of the hydrosilicate minerals is responsible for this difficulty in reduction (Paper I and

II, Kawahara et al. 1988, Utigard et al. 1993, Chen et al. 2004, Rhamdhani et al. 2009,

Pietilä 2011). Further, there is at least one serpentine mineral, namely antigorite, which

is confirmed as undergoing simultaneous dehydration and recrystallisation to olivine

(Paper II, Li et al. 2000, Harris et al. 2009).

To our knowledge, the additives have an effect on the elimination or suppression of

recrystallised phases, which is important for the subsequent reduction process. Sulphur

is an additive that is considered not only to preclude the formation of a high-

temperature phase but also to increase metal extractions, particularly from the more

refractory ores (Paper V, Siemens et al. 1976, Kukura et al. 1979, De Graaf 1980, Valix et

al. 2002a, Pekkarinen 2011).

4.2.1 Results and Discussion

The DSC results of three different additive-bearing lateritic nickel ores heated under an

Ar atmosphere from room temperature up to 1300°C are shown in Fig. 19.

28

Figure 19. DSC curves for Colombia-1, Colombia-2, and Mirabela blended with pyrite when heated up to 1300°C in an argon atmosphere (Paper V).

The DSC results (Fig. 19) of the lateritic nickel ores blended with sulphur disclosed the

fact that the exothermic peaks underwent a change in size. In addition, the dehydration

and recrystallisation of minerals is also observed to begin earlier when sulphur is added

to the ores. The capability of sulphur to react with the available oxygen at relatively low

temperatures during heat treatment, inducing the dehydration to commence at a lower

temperature than usual, is attributed to the reduced densification of the pellet particles,

which makes it easier for the evaporated crystalline water from silicates to diffuse

outward (Paper V).

Additionally, subsequent to the heat treatment, the samples were subjected to an X-

ray powder diffraction analysis in order to determine the effect of sulphur on the

mineralogical changes of the ores.

In the case of the Mirabela ore (Fig. 20), in which the sulphur concentration was

higher compared to the Colombian ores, neither olivine nor pyroxene phases were

formed when it was heated up to 1300°C. Thus preventing the formation of olivine and

pyroxene phases appears to depend on the amount of sulphur added to the ore. The DSC

analysis illustrated in Fig. 19 confirms the suppression of the exothermic peak by

sulphur (Paper V, Pekkarinen 2011).

Additionally, the mineralogy of the ore might also control the addition of the sulphur-

bearing additive and thus affect the recrystallisation and, further on, the formation of

the new phases.

29

Figure 20. Powder XRD pattern of minerals of Mirabela ore blended with pyrite induced by heat treatment at 1300°C in argon (Paper V).

30

5 REDUCTION OF SAPROLITIC NICKEL ORES

Generally, in the saprolites, both magnesia and nickel oxide form a solid solution with

silica, and the temperature for initiating the decomposition of the crystal of nickel

magnesium silicates is relatively high (Papers III and IV, Hang and Brindley 1973,

Burger 1996, Li and Coley 2000). Consequently, the nickel oxide that separates from the

disintegrated nickel magnesia silicate crystal is then susceptible to reduction.

Commercially, the reduction of saprolitic nickel ores is achieved by utilising either

carbon monoxide or hydrogen as the reducing gas. Since the nickel content of the ore

deposits is generally relatively small, the reduction process produces only a relatively

small amount of metallic nickel, which is very finely distributed. The metallic nickel and

iron particles might subsequently be recovered as a nickel-iron alloy by magnetic

concentration since the metallic nickel and iron are strongly magnetic, whereas the iron

oxide and other constituents of the tailings are only very weakly magnetic at best

(Watanabe et al. 1987, Zhu et al. 2008).

5.1 Results and Discussion 5.1.1 Isothermal Reduction

According to the experimental results, excluding the reduction results of Colombia-1

(Fig. 21), the removable oxygen degree Rt of the samples as a function of time in a fixed

H2/N2 gas mixture (72%:28%) at different temperatures increases with increasing

temperature (Paper III).

The presence of antigorite in Colombia-1 was found to enhance the dehydroxylation

and recrystallisation simultaneously (Paper II). The phases thus formed were identified

as being poorly reducible. Therefore, this indicates that the initial chemical and

mineralogical compositions of a saprolitic nickel ore are critical to the degree and the

optimal temperature of reduction, respectively (Paper III).

31

Figure 21. Effect of temperatures on the reduction of Colombia-1 by means of an H2/N2

(72%:28%) gas mixture (Paper III).

The results of reduction at different temperatures in a fixed CO/CO2 (72%:28%) gas

mixture showed a lower degree of removable oxygen compared to that in a fixed H2/N2

mixture (Fig. 22).

Figure 22. Comparison of Colombia-1 reduction in H2/N2 (72%:28%) and CO/CO2

(72%:28%) at 750 and 900°C (Paper IV).

Table 3, which shows the total degrees of metallisation (iron and nickel) of the reduced

laterite samples in fixed H2/N2 and CO/CO2 mixtures at different temperatures for 90

minutes, also confirms the higher reducibility of ores in H2/N2. Moreover, it can be seen

0

10

20

30

40

50

60

70

80

90

100

0 2000 4000 6000 8000

De

gre

e o

f re

mo

va

ble

ox

yge

n [

%]

Time[s]

Colombia-1

600°C 750°C 900°C

0

10

20

30

40

50

60

70

80

90

100

0 2000 4000 6000

De

gre

e o

f re

mo

va

ble

ox

yge

n [

%]

Time [s]

Colombia-1

750°C_CO 750°C_H2900°C_CO 900°C_H2

32

that the Mirabela ore has been reduced the best at both temperatures and in both the

reducing gases, with the exception of Colombia-2, which was reduced at 900°C in

CO/CO2. The degree of metallisation is determined as the (Fe°+Ni°)/(Fetot+Nitot) ratio,

where Me° is the analytical metallic concentration and Metot is the total concentration of

the Me in the ore.

Table 3. Metallisation degrees of the ores reduced in CO/CO2 (72%:28%) and H2/N2

(72%:28%) at different temperatures (Paper IV).

Moreover, according to the metallic iron and nickel percentages in the present study

(Fig. 23), nickel is more susceptible in CO/CO2 than iron, whereas the H2/N2 reducing

gas mixture effectively reduces iron oxides (Papers III and IV). Therefore, in order to

achieve the maximum extent of reduction and nickel metallisation, the properties of the

various reducing gases should be taken into consideration.

Figure 23. Degrees of metallisation of metallic elements in H2/N2 (72%:28%) and CO/CO2 (72%:28%) at 6oo, 750, and 900°C (Paper IV).

In addition to the benefits of various reducing gases in fixed ratios, a decrease in the

CO/CO2 ratio for a given temperature will give a higher recovery of Ni. Thus, it is

recommended to preserve the CO/CO2 ratio below 1 in order to obtain an appreciable

nickel recovery. Experimental results at 750°C and 900°C with a number of reducing gas

mixtures are shown in Fig. 24.

0

5

10

15

20

25

30

35

40

45

50

Co

lom

bia

-1H

2 6

00

°C

Co

lom

bia

-1H

2 7

50

°C

Co

lom

bia

-1H

2 9

00

°C

Co

lom

bia

-2H

2 6

00

°C

Co

lom

bia

-2H

2 7

50

°C

Co

lom

bia

-2H

2 9

00

°C

Mir

ab

ela

H2

60

0°C

Mir

ab

ela

H2

75

0°C

Mir

ab

ela

H2

90

0°C

Co

lom

bia

-1C

O 7

50

°C

Co

lom

bia

-1C

O 9

00

°C

Co

lom

bia

-2C

O 7

50

°C

Co

lom

bia

-2C

O 9

00

°C

Mir

ab

ela

CO

75

0°C

Mir

ab

ela

CO

90

0°C

Pe

rce

nt

Me

tall

ic [

%]

Fe Ni

Metallisation Degree % Temperature °C Colombia-1 Colombia-2 Mirabela

H2/N2 CO/CO2 H2/N2 CO/CO2 H2/N2 CO/CO2

750 13.12 1.69 15.26 3.46 39.08 9.99 900 6.82 1.13 15.41 4.10 38.23 2.58

33

Figure 24. Degrees of metallisation of metallic elements with different CO/CO2 ratios at 750 and 900°C.

Scanning electron micrographs of reduced samples were recorded to demonstrate

their morphological characteristics (Fig. 25). The images of SEM-EDS analysis

demonstrated in Fig. 25 a indicate the formation of ferronickel particles in the pellets.

Furthermore, the sizes of the ferronickel particles are different and are considered to be

dependent on the reducing temperature and chemistry of the ore (CaO, MgO/SiO2 ratio,

and Al2O3). Subsequently, they will grow within the softened or partially molten

magnesia silicate slag at the operating temperatures (>900°C) in the rotary kiln

(Rhamdhani et al. 2009, Kobayashi et al. 2011). Additionally, the micrographs of the

reduced pellets imply that the majority of the dark grey particles are Mg-Fe and Mg-Fe-

Al silicates. Furthermore, unreduced nickel is mainly associated with Mg-Fe-Al and Mg-

Fe silicates (Fig. 24 b, c) (Papers III, IV, and V, Chen et al. 2004, Rhamdhani et al.

2009).

0

5

10

15

20

25

Co

lom

bia

-272

%:2

8%

CO

/CO

275

0°C

Co

lom

bia

-22

2%

:78

% C

O/C

O2

750

°C

Co

lom

bia

-272

%:2

8%

CO

/CO

29

00

°C

Co

lom

bia

-23

6%

:64

% C

O/C

O2

9

00

°C

Mir

ab

ela

72%

:28

% C

O/C

O2

750

°C

Mir

ab

ela

22

%:7

8%

CO

/CO

275

0°C

Mir

ab

ela

72%

:28

% C

O/C

O2

90

0°C

Mir

ab

ela

36

%:6

4%

CO

/CO

29

00

°C

Pe

rce

nt

Me

tall

ic [

%]

Fe Ni

34

Figure 25. Microstructure of reduced samples in CO/CO2 (72%:28%): a) Colombia-1 at 750°C; b) Colombia-2 at 900°C; c) Mirabela at 900°C post-carbon monoxide reduction (Paper IV).

5.1.2 Non-isothermal Reduction

The gaseous pre-reduction of ore in a rotary kiln is a critical stage in the

pyrometallurgical process for nickel alloy production, since a good selection of the pre-

reduction will produce a more valuable nickel alloy with a higher nickel grade and

reduce the processing costs of the subsequent smelting stage.

In a conventional rotary kiln, the ore travels counter-currently to the flow of hot gas

that is mostly produced by the combustion of coal. While it is travelling, the temperature

of the ore within the kiln rises from the feed end to the discharge end and it is

continuously in contact with the reducing gas in the kiln. Subsequently, the drying,

dehydration, and reduction of the ore and the agglomeration of the reduced metal take

place.

The saprolites which are used in the study were first calcined at 750°C for 60 min, and

then reduced at rising temperatures from 750 to 900°C in a fixed-ratio (72%:28%)

CO/CO2 reducing gas mixture for 30 min in order to simulate the reduction roasting

condition that prevails in a rotary kiln (Pietilä 2011).

The weight loss of the samples as a function of temperature is illustrated in Fig. 26.

According to Fig. 26, the initial rate of weight loss of the samples was very fast up to 800

°C. Thereafter, it decreases rapidly with an increase in the temperature. However, the

steepest descent in the rate weight loss was observed in Colombia-1 above 800°C. The

olivine or pyroxene phases that occurred in Colombia-1 at 750°C (with the presence of

antigorite) are ascribed to delayed Ni and Fe reduction when they are locked up in those

phases (Papers III and IV).

35

Figure 26. The weight loss of Colombia-1, Colombia-2, and Mirabela reduced in a CO/CO2 (72%:28%) gas mixture at a rising temperature as a function of time (Pietilä 2011).

In addition to the properties of the various reducing gases, which have an impact on

the progress of the reduction and generation of metallic elements, the introduction of

the reducing gas prior or subsequent to the calcination of the ore might also initiate the

formation of new phases which are detrimental to nickel or/and iron reduction (De

Graaf 1979, Li et al. 2000, Rhamdhani et al. 2009, Valix et al. 2002a).

Therefore, taking for granted the stepwise gaseous reduction of hematite (Fe2O3) to

metallic iron (Fe°), as well as the reduction of nickel silicates [(Mg, Ni, Fe)3Si2O5(OH)4]

into metallic nickel (Ni°), and considering the results of the metallic percentage in

Fig.27, the optimal nickel recovery can be controlled by launching the reducing gas. It

was detected that introducing the reducing gas after the roasting of the saprolite ores at

a constant temperature resulted in the highest nickel recovery. Therefore, the reducing

conditions are predicted to promote the formation of the high-temperature phases

which are detrimental to nickel and iron reduction.

0

10

20

30

40

50

60

70

80

90

750 800 850 900 950

We

igh

t lo

ss [

%]

Temperature [°C]

Colombia-1, CO Colombia-2, CO Mirabela, CO

36

Figure 27. Degrees of metallisation of metallic elements when the ores are being reduced in CO/CO2 (72%:28%) at 750, 900, and 750 �900°C (Pietilä 2011).

5.1.3 Effect of additives (sulphur, carbon) on saprolitic nickel reduction

The major challenge with the reduction of saprolitic nickel ores is ensuring that the

nickel is in reducible form. Valix (2002a) performed in situ XRD experiments and

proposed that the sulphur aids the suppression of the olivine phase, allowing the nickel

to be in the active stage, where it can then be reduced. Kukura et al. (1979), De Graaf

(1980), and Pekkarinen (2011) also reported that additives have an effect on the

elimination or suppression of such recrystallised phases, which is important for the

subsequent reduction process.

In the saprolites, the situation regarding gas-nickel oxide solid reduction is probably

complicated by the presence of two types of reducible oxides, namely iron oxides and

cobalt oxide, in the silicate matrix, in addition to a high magnesia (MgO) and silica

(SiO2) content. However, iron oxides will predominantly affect the thermal

transformation, migration of ions, and concentration of reducing gas upon reduction,

because there is usually a higher content of them in nickel saprolites compared to other

possible reducible oxides.

Hence, the effect of sulphur on the reducibility of the additive-bearing saprolitic nickel

ores was examined in different reducing gases with fixed ratios at the temperatures of

750 and 900°C, i.e. below and above the recrystallisation temperature. The metallic

nickel, iron, and cobalt contents in the sulphur-bearing samples reduced in a fixed

H2/N2 (72%:28%) and CO/CO2 (72%:28%) gas mixture at different temperatures are

reported in Fig. 28.

0

2

4

6

8

10

12

14

Co

lom

bia

-1C

O 7

50

°C

Co

lom

bia

-1C

O 9

00

°C

Co

lom

bia

-2C

O 7

50

°C

Co

lom

bia

-2C

O 9

00

°C

Mir

ab

ela

CO

75

0°C

Mir

ab

ela

CO

90

0°C

Co

lom

bia

-1C

O 7

50

-90

0°C

Co

lom

bia

-2C

O 7

50

-90

0°C

Mir

ab

ela

CO

75

0-9

00

°C

Pe

rce

nt

Me

tall

ic [

%]

Fe Ni

37

Figure 28. The metallic percentages of Ni, Fe, and Co in the sulphur-containing samples reduced in fixed CO/CO2 (72%:28%) and H2/N2 (72%:28%) mixtures at 750 and 900°C (Paper V and Pekkarinen 2011).

It can be seen that nickel is reduced more effectively than iron and cobalt in CO/CO2 at

both temperatures and in every sample. Further, it can be observed that nickel is more

susceptible than iron in CO/CO2, and the H2/N2 reducing gas mixture effectively reduces

iron oxides. It can also be noticed that the formation of metallic iron is affected by the

reduction temperature. The iron with the lowest metallic percentage was obtained in the

reduction tests at the highest temperature.

The useful effect of sulphur on the nickel metallisation can merely be identified when

the results from the experiments with samples containing sulphur (Fig. 28) are

compared to those for pellets without the addition of sulphur (Figs. 23, 24, and 27).

Therefore, with the nickel metallisation being taken for granted, the degradation of

silicates at the dehydroxylation temperature, where the minerals of the metals are free

from silicates, is regarded as improving the readiness of the nickel oxide to react with

the available sulphur and thus makes it more ready for reduction. Added to this, the

discontinuation of the densification of the reaction zone as a result of the segregated

sulphur to the surface of the particles and grain boundaries will also increase the nickel

metallisation when the temperature increases to 900°C. Hence, regardless of the type of

the high temperature phases, sulphur could suppress the densification of those phases

and will allow easy access of fresh reducing gas to the unreacted core of the phases in

which the nickel is locked up (Paper V).

Additionally, as sulphur does not have an effect on the formation of magnesioferrite

(MgFe2O4) in Colombia-1 and Mirabela at 900°C (Paper V), a less reducible solid

solution, (Mg,Ni)Fe2O4, will be formed. The lower metallisation of the nickel in Mirabela

0

10

20

30

40

50

60

70

Co

lom

bia

-1F

eS

2,

CO

,75

0°C

Co

lom

bia

-1F

eS

2,

H2

,75

0°C

Co

lom

bia

-1F

eS

2,C

O, 9

00

°C

Co

lom

bia

-1F

eS

2,H

2,9

00

°C

Co

lom

bia

-2F

eS

2,C

O,7

50

°C

Co

lom

bia

-2F

eS

2,H

2,7

50

°C

Co

lom

bia

-2F

eS

2,C

O,9

00

°C

Co

lom

bia

-2F

eS

2,H

2,9

00

°C

Mir

ab

ela

Fe

S2

,CO

,75

0°C

Mir

ab

ela

Fe

S2

,H2

,75

0°C

Mir

ab

ela

Fe

S2

,CO

,90

0°C

Mir

ab

ela

Fe

S2

,H2

,90

0°C

Pe

rce

nt

Me

tall

ic [

%]

Fe Ni Co

38

and Colombia-1, as well as its increase in Colombia-2 at 900°C (Fig. 28), confirms this

presumption. The recrystallisation of magnesioferrite is proved to be a consequence of

the dehydroxylation process (Paper II).

Further, according to the chemical analyses of reduced samples (Fig. 28), the

reducibility of the cobalt is observed to have been controlled by iron metallisation and

the reduction temperatures. Therefore, it increases along with a decreasing reduction

temperature and with increasing iron metallisation.

The micrographs of reducing pellets in Fig. 29 show that nucleation and

recrystallisation have already taken place, and the majority of the particles are grey and

white. Among other findings, the white particles are observed to contain more metallic

nickel and iron, and chiefly to have round forms. The point analysis of particles,

however, show that the white particles are nucleated at the grain boundaries of silicates,

and their nickel content varies from one particle to another and even within the same

particle. Additionally, the SEM-EDS analyses imply that the added sulphur-bearing

additive was captured by the reduced pellet (Fig. 29) (Paper V). Therefore, as the

sulphation reactions are thermodynamically favoured by relatively low temperatures

(Papazoglou & Rankin 2003), it is possible that sulphates essentially prevent sulphur

from vapourisation at high temperatures. However, the reaction from oxide to sulphate

is possible if only the sulphidic sulphur reacts with oxides or oxygen to form SO3; thus

the SO3 that is formed will be absorbed onto the oxide surfaces, which react instantly,

forming sulphate.

Figure 29. Microstructure of reduced sulphur-bearing samples in CO/CO2 (72%:28%): a) Colombia-1 at 900°C; b) Colombia-2 at 900°C; c) Mirabela at 900°C (Paper V).

39

6. CONCLUSIONS

Saprolitic nickel ores are complex mixtures of minerals such as serpentine, clay, chlorite,

and goethite. They have a heterogeneous phase structure with varying contents of the

major elements, such as Mg, Si, Fe, Ni, Al, Co, Mn, and Cr. The assay of the ores used in

this study varies from one particle to another and from mineral to mineral within the

same sample, in conformity with the range of the serpentine-clay-chloride-goethite

mixtures. Most of the nickel is present in the crystal lattices of serpentine minerals.

Additionally, nickel occurs with cobalt, manganese, aluminium, chromites (spinel), and

iron oxides in the goethite lattice too.

The saprolites heated in argon or in air up to 1300°C showed three endothermic

processes at 100, 250, and 550-650°C, caused by the removal of free water, the

dissociation of goethite, and by dehydroxylation reactions, respectively. The

endothermic reactions are followed by an exothermic reaction at ~820°C which is

associated with the crystallisation of new phases. Quartz, enstatite, olivine and

maghemite (spinel) have been identified by XRD as a result of the disintegration of

minerals. The experimental results indicate that the compositions of the final phases are

heavily dependent on the proportions of serpentine-like and clay-like layers in the initial

ore.

As a result of heat treatment in argon or in air at 750°C, the DSC and XRD

observations showed that the microstructures of the ore were disrupted, and part of

their crystalline water was evaporated, which brought about the formation of cracks and

the redistribution of the nickel in the particles. The nickel was observed mainly in Fe-

Mg-Al and Mg-Fe silicates, but high nickel contents were observed to be concentrated

within Cr-spinels and Mn-oxide particles. Additionally, the formation of new phases,

namely talc-like, amorphous, forsterite, and enstatite, after heating at 750°C was noticed

to depend on the type of initial minerals of the saprolite ore. The experimental results

clearly revealed that the presence of antigorite in Colombia-1 induces the formation of

forsterite and enstatite before the recrystallisation temperature; otherwise, the laterites

containing no antigorite (Colombia-2, Mirabela) developed intermediate phases.

Additionally, 850°C, which was also rather close to the temperature of the subsequent

reduction experiments, was purposely selected in order to ascertain the temperature of

thermal transformation of the minerals. The XRD results showed that the temperaure at

which the formation of the high-temperature phase takes place is above 850°C when the

saprolite ore bulk mineral is composed of lizardite, clinochrysotile, and clinochlore.

40

Additionally, the DSC results indicated that the exothermic process at ~820°C

undergoes a change in size when sulphur is added. Particularly, in the case of the

Mirabela ore, the sulphur almost completely suppressed the exothermic reaction and

induced the dehydration and recrystallisation of the minerals to commence earlier.

The reduction kinetics of the saprolitic nickel ores were shown to depend on

temperature, time, mineralogy, and the concentration and variety of the reducing gas. In

particular, the initial chemical and mineralogical composition of the saprolitic ore

proved to be a critical parameter for the progress of the reduction. Hydrous nickel-

magnesium silicates (Mirabela) which occur in the form of vermiculite (clay mineral)

proved easier to reduce, since vermiculite recrystallises to pyroxene phases

(Mg,Fe,Ni)SiO3. However, reducing the degree of removable oxygen in the Colombia-1

ore (antigorite) from 750 to 900°C in H2/N2 and CO/CO2 is attributed to the formation

of olivine (Mg,Ni,Fe)2SiO4, which is difficult to reduce. Moreover, the observations

proved that the reducing gas promotes the formation of new phases which are

detrimental to nickel and/or iron reduction. Therefore, the properties of various

reducing gases have an impact on the progress of the reduction and generation of

metallic elements.

Concerning the different chemical and mineralogical compositions of saprolites and

the various properties of the reducing gas, their degree of reduction is not expected to be

the same in different reducing gases under similar reducing conditions. Therefore, to

achieve the highest metallisation for nickel, the current results of the reduction

experiments recommend using CO as the reducing gas.

Generally, in the reduction experiments, the sulphur proved to improve the

reducibility of nickel saprolitic ores throughout the entire temperature range that was

studied. Consequently, for temperatures between the dehydroxylation and

recrystallisation reactions, sulphur reacted with both cobalt and especially with nickel

and thus made them more reducible when the minerals of the metals were free of

silicates. At temperatures above recrystallisation, the sulphur that was already

segregated to the surfaces of the grains and particles improved the nickel metallisation.

It suppressed the densification of the reaction zone and facilitated the access of fresh

reducing gas to the unreacted core of the phases in which the nickel is locked up.

Moreover, regarding the sulphur, the formation of sulphates at an early stage of heating

is essential to prevent it from vapourisation at high temperatures. Additionally, the

beneficial effect of sulphur on the degree of metallisation can only be seen when

comparing the results from the experiments with sulphur-containing samples to those

with pellets without sulphur. Considering the results from the current reduction

experiments, the optimal reduction temperature for ores is ~750°C. The higher sulphur

content also resulted in a higher degree of metallisation.

41

A complex series of phases was observed after the reduction roasting of saprolites. The

formation of ferronickel alloy particles after reduction was proved by SEM-EDS analysis.

42

7. REFERENCES

Antola, O. (1993). Reduction of nickel oxide by CO and H2 containing gases. D.Sc thesis, 135, Leoben University, Austria.

Apostolidis, C. I., and Distin, P. A. (1978). The kinetics of the sulphuric acid leaching of nickel and magnesium from reduction roasted serpentine. Hydrometallurgy, 3(2), 181-196.

Ball, M. C. and Taylor, H. F. W. (1963). The dehydration of chrysotile in air and under hydrothermal conditions. Mineralogical Magazine, 33, 467-482.

Barros, S. M., Trescases, J. J., and Jose Melfi, A. (1992). Lateritic nickel deposits of Brazil. Mineralium Deposita, 27, 137-146.

Bergman, R. A. (2003). Nickel production from low-iron laterite ores: Process descriptions. The Canadian Mining and Metallurgical Bulletin, 96, 127-138

Bergeson, L. L. (2003). EU’s new chemical policy will hurt U.S. competitiveness. Chem. Process. Mag., Oct. 9, 2003, http://www.chemicalprocessing.com/articles/2003/143.html?page=prin (accessed September 19, 2007).

Blanco, L. J. and Holliday, M. B. (1981). Energy Swings at Nickel. Journal of Metals, Annual Review of Extractive Metallurgy April, 50-51.

Botsis, N., Van Bronswijk, W. and Watling, H. (2008). Nickel laterite characterisation. ALTA 2008 Nickel-Cobalt, Copper & Uranium International Conference Proceedings Jun, 1-16.

Brand, N.W., Butt, C.R.M., and Elias, M. (1998). Nickel laterites: classification and features. AGSO Journal of Australian Geology & Geaphysics 17(4), 81-88.

Brindley, G. W. and Wan, H. M. (1975). Compositions, Structures and Thermal Behaviour of Nickel Containing Minerals in the Lizardite-Nepouite Series. The American Mineralogist 60(9-10), 863-871.

Brindley, G.W., and Maksimovic, Z. (1974). The nature and nomenclature of hydrous nickel-containing silicates. Clay Minerals 10(4), 271-277.

Brindley, G. W. and Hayami, M. S. (1965). Mechanism of formation of forsterite and enstatite from serpentine. Mineralogical Magazine 35(269), 189-195.

Bunjaku, A., Kekkonen, M. and Holappa, L. (2006). On production of ferronickel. Aalto University School of chemical Technology publications in Department of Materials and Engineering, TKK-MT-185, pp 1-16. Espoo, Finland.

Bunjaku, A., Kekkonen, M. and Holappa, L. (2009). Thermodynamic properties of nickel laterite ores. Aalto University School of chemical Technology publications in Department of Materials and Engineering, TKK-MT-208, pp 1-25. Espoo, Finland.

Bunjaku, A., Kekkonen, M. and Holappa, L. (2010). Phenomena in thermal treatment of lateritic nickel ores up to 1300°C, The twelfth International Ferroalloys Congress Proceedings (ed: A. Vartiainen) pp 641-652 (Outotec Oyj: Helsinki, Finland).

Bunjaku, A., Kekkonen, M., Taskinen, P. and Holappa, L. (2011). Thermal behaviour of hydrous nickel-magnesium silicates when heating up to 750°C. Min. Process. Extract. Metall. (Trans. Inst. Min. Metall. C) 120(3), 139-146.

43

Bunjaku, A., Kekkonen, M. and Taskinen, P. (2012). Effect of mineralogy on reducibility of calcined nickel saprolite ore by hydrogen. Min. Process. Extract. Metall. (Trans. Inst. Min. Metall. C) 121(1), 16-22.

Bunjaku, A., Kekkonen, M., Pietilä, K. and Taskinen, P. (2012a). Effect of mineralogy and reducing agent on reduction of saprolitic nickel ores, Min. Process. Extract. Metall. (Trans. Inst. Min. Metall. C), 121(3), 156-165.

Bunjaku, A., Kekkonen, M., Pekkarinen, S. and Taskinen, P. (2012b). Effect of sulphur addition on the roasting reduction of saprolitic nickel ores in CO/CO2 gas mixture. Min. Process. Extract. Metall. (Trans. Inst. Min. Metall. C). Paper is accepted for publication on 31.08.2012.

Burger, P. A. (1996). Origins and Characteristics of Lateritic Nickel Deposits. In Proceedings Nickel’ 96 Mineral to Market congress (eds: Grimsey, E. S., and Neuss, I.) Publication Series no 6/96, pp 179-183, (The Australasian Institute of Mining and Metallurgy).

Campbell, F. E., and Roeder, P. (1968). The stability of olivine and pyroxene in the Ni-Mg-Si-O system. The American Mineralogist 53(1-2), 257-268.

Caron, M. H. (1950). Fundamental and practical factors in ammonia leaching of nickel and cobalt ores. Journal of Metals 188, 67-90.

Chen, T. T., Dutrizac, J. E., Krause, E. and Osborne, R. (2004). Mineralogical characterization of nickel laterites from New Caledonia and Indonesia. International Laterite Nickel Symposium,(eds: W.P. Imrie, D.M. Lane) TMS (The Minerals, Metals& Materials Society), pp 79-99.

Chen, Sh., Jak, E., and Hayes, P. C. (2005). Phase Equilibria in the Cristobalite, Tridymite and Pyroxene Primary Phase Fiwlds in the MgO-“FeO”-SiO2 System in Equilibrium with Metalic Iron. ISIJ International, 45(6), 791-797.

Chen, J., Jak E., Hayes, P. C. (2009). Factors affecting nickel extraction from the reduction roasting of saprolite ore in the Caron process. In Proceedings Pyrometallurgy of Nickel and Cobalt International Symposium, (eds: J.Liu, J. Peacey, M. Barati, S. Kashani-Nejad, B. Davis), pp 449-461 (Institute of Mining, Metallurgy and Petroleum: Canada).

Colin, F., Nahon, D., Trescases, J. J., and Melfi, A. J. (1990). Lateritic weathering of pyroxenites at Niquelandia, Goias, Brazil: The supergene behavior of nickel, Economic Geology 85(5), 1010-1023.

Coleman, E. E. and Vedensky, D. N. (1960). Ferro-nickel production in Oregon. Journal of Metals12 (1), 197-201.

Crowe, B. J. C. and Utigard, T. A. (2003). Kinetics of Carbon Monoxide Reduction of Oxidized Calcines. Canadian Metallurgical Quarterly, 42(4), 447-454.

Daenuwy, A., Dalvi, A. D. (1997). Development of reduction kiln design and operation at P. T. Inco (Indonesia). Proceedings of the Nickel-Cobalt 97 International Symposium Vol. III, pp 93-113, (ed. C. Diaz, I. Holubec, C. G. Tan). The Metallurgical Society of CIM.

Dalvi, A. D. and Smeltzer, W. W. (1970). Thermodynamics of the Iron-Nickel-Oxygen System at 1000°C. Journal of the Electrochemical Society 117(II), 1431-1436.

Dalvi, A. D., and Sridhar, R. (1976). Thermodynamics of the Fe-Ni-O and Fe-Ni system at 1065 K to 1380 K. Canadian Metallurgical Quarterly 15(4), 249-357.

Dasher, J. (1976). The energy picture in nickel production, Mining Magazine May, 379-389.

Davies, R. H., Dinsdale, A. T., Gisby, J. A,, Robinson, J. A. J., and Martin, S. M. (2002). MTDATA-Thermodynamic and Phase Equilibrium Software from the National Physical Laboratory. Calphad 26(2), 229-271.

De Graaf, .J.E. (1979). The treatment of lateritic nickel ores- A further study of the Caron process and other possible improvements, part I. Effect of reduction conditions. Hyrdrometallurgy 5(1), 47-65.

De Graaf, .J.E. (1980). The treatment of lateritic nickel ores- A further study of the Caron process and other possible improvements, part II. Leaching studies. Hyrdrometallurgy, 5(2-3), 255-271.

44

Diaz, C. M., Landolt, C. A., Vahed, A., Warner A. E. M., and Taylor, J. C. (1988). A review of nickel pyrometallurgical operations. In proceedings Extractive metallurgy of Nickel & Cobalt symposium(ed. G. P. Tyroler and C. A Landolt), pp 211-239 (A Publication of the Metallurgical Society, Inc.,117th TMS Annual Meeting: Phoenix, Arizona).

Diaz, C. M., Landolt,C. A., Dalvi, A. D., Bacon, W. G., and Osborne, R. C. (2004). The past and the future of nickel laterite. Prospectors and Developers Association of Canada International Convention, Trade Show & Investors Exchange, publications, pp 1-27 (PDAC: Canada).

Dor, A. A., and Skretting, H. (1979). The production of Ferronickel by the rotary kiln-electric furnace process. International laterite symposium, (eds: Evans, D. J. I., Shoemaker, R. S., and Veltman, H.), published by Society of Mining Engineers of AIME February 19-21, pp. 459-490.

Eckleman, E. J. (2010). Facility-level energy and greenhouse gas life-cycle assessment of global nickel industry. Resources, Conservation and Recycling 54(3), 256-266.

Elias, M., Donaldson, M. J., and Giorgetta, N. (1981). Geology, mineralogy, and chemistry of lateritic nickel-cobalt deposits near Kalgoorlie, Western Austalia. Economic Geology 76(6), 1775-1783.

Faust, G.T. (1966). The hydrous nickel-magnesium silicates – the garnierite group. Am. Mineral51(3-4), 279-298.

Fittock, J. E. (2012). Nickel and cobalt refining at QNI Pty Ltd Yabulu Queensland. AUSIMM Monograph 19, 2rd Ed, pp 1-47.

Gisby, J., Dinsdale, A. & Taskinen, P.(2007). Predictive thermochemistry of oxide and sulfide systems. Proc. E m c 2007, Düsseldorf, June 11-14, 2007, DGBM, Clausthal-Zellerfeld, (Vol. 4), pp 1721-1736.

Gleeson, S. A., Herrington, R. J., Durango, J., Velasquez, C. A., and Koll, G. (2004). The mineralogy and geochemistry of Cerro Matoso S.A Ni laterite deposite, Montelibane, Colombia. Economic Geology 99(6), 1197-1213.

Golightly, J. P. (1979). Nickeliferous laterites: A general description. International laterite symposium, (eds: Evans, D. J. I., Shoemaker, R. S., and Veltman, H.), published by Society of Mining Engineers of AIME, February 19-21, pp. 3-119, New Orleans, Louisiana, USA.

Gomez, R., Ogryzlo, C. T., Dor, A. A. (1979). The Cerro Matoso nickel project. International laterite symposium, (eds: Evans, D. J. I., Shoemaker, R. S., and Veltman, H.), published by Society of Mining Engineers of AIME, February 19-21, pp. 412-458.

Guo, X.J. (2009). Chines nickel industry- Projects, production and technology. In Proceedings Pyrometallurgy of Nickel and Cobalt International Symposium, (eds: J.Liu, J. Peacey, M. Barati, S. Kashani-Nejad, B. Davis), pp 3-21, (Institute of Mining, Metallurgy and Petroleum: Canada).

Gürtekin, G. and Albayrak, M. (2006). The reaction of antigorite: A XRD, DTA-TG work. Mineral Res. Exp., Bull. 133, 41-49.

Hallett, C. J. (1997). A thermodynamic analysis of solid state reduction of nickel from laterite minerals. In Proceedings Nickel-Cobolt 97 International Symposium (eds: C. A. Levac, R. A. Berryman), (II) pp 299-312 (The Metallurgical Society of CIM: Canada).

Hang, P. T., and Brindley, G. W. (1973). The nature of garnierites-I, II, III. Clays and clay minerals, 21(1), 27-57.

Harris, C. T., Peacey, J. G., Pickles, C. A. (2009). Thermal Upgrading of Nickeliferous Laterites- A Review. In Proceedings Pyrometallurgy of Nickel and Cobalt International Symposium, (eds: J.Liu, J. Peacey, M. Barati, S. Kashani-Nejad, B. Davis), pp 51-75 (Institute of Mining, Metallurgy and Petroleum: Canada).

Hidayat, T., Rhamdhani, M. A., Jak, E., and Hayes, P. C. (2009). Investigation of nickel product structure developed during the gaseous reduction of solid nickel oxide. Metallurgical and Materials Transactions B (Process Metallurgy and Materials Processing Science), 40(2), B462-B473.

45

Hidayat, T., Rhamdhani, M. A., Jak, E., and Hayes, P. C. (2009a). On the relationships between the kinetics and mechanisms of gaseous hydrogen reduction of solid nickel oxide. Metallurgical and Materials Transactions B (Process Metallurgy and Materials Processing Science), 40(4), B474-B489.

Hidayat, T., Rhamdhani, M. A., Jak, E., and Hayes, P. C. (2009b). The kinetics of reduction of dense synthetic nickel oxide in H2-N2 and H2-H2O atmospheres. Metallurgical and Materials Transactions B (Process Metallurgy and Materials Processing Science), 40(1), B1-B16.

Kawahara, M., Toguri, J. M., and Bergman, R. A. (1988). Reducibility of laterite ores. Metallurgical and Materials Transactions B (Process Metallurgy and Materials Processing Science), 19(2), B181-B186.

Kerfoot, D. G. E. (1991). Nickel. Ullmann’s Encyclopedia of Industrial chemistry, A17(5), 158-219 (publisher: VCH Verlag), Weinheim, Germany.

Kiukkola, K., and Wagner, C. (1957). Measurements on galvanic cells involving solid electrolytes. Journal of the Electrochemical Society, 104(6), 379-387.

Kobayashi, Y., Todoroki, H. and Tsuji, H. 2011. Melting Behavior of Siliceous Nickel Ore in a Rotary Kiln to Produce Ferronickel Alloys, ISIJ International, 51(1), 35-40.

Kukura, M. E., Stevens L. G. and Auck, Y. T. (1979). Development of the UOP process for oxide silicate ores of nickel and cobalt. International laterite symposium, (eds: Evans, D. J. I., Shoemaker, R. S., and Veltman, H.), published by Society of Mining Engineers of AIME, February 19-21, pp 527-569, New Orleans, Louisiana, USA.

Li, S., and Coley, K. S. (2000). Kinetics and Mechanism of Reduction of Laterite Ore High in Serpentine, in proceedings of J. M. Toguri Symposium- Fundamentals of Metallurgical Processing (eds: G. Kaiura, C. Pickles, T. Utigard and A. Vahed), pp 179-192 (Institute of Mining, Metallurgy and Petroleum: Canada).

Mackenzie, R. C. (1970).Simple Phyllosilicates Based on Gibbsite- and Brucite-like sheets. Differential thermal analysis ( ed: R. C. Mackenzie), Academic Press London and New York, Vol. 1(18), pp 499-537.

MacKenzie, K. J. D., and Meinhold, R. H. (1994). Thermal reactions of chrysotile revisited: 29Si and 25Mg MAS NMR study. American Mineralogist, 79(1-2), 43-50.

Matsumori, T., Ishizuka, T., and Matsuda, T. (1997). Smelting process for direct production of ferronickel suitable for sinless steel making. Proceedings of the Nickel-Cobalt 97 International Symposium Vol. III, pp 141-150, (eds. C. Diaz, I. Holubec, C. G. Tan). The Metallurgical Society of CIM.

Mehmetaj, B., and Boom, R. (2001). Ferronickel production in Kosova-past performance and new opportunities. Steel research, 72(11+12), 428-433.

Mukhopadhyay, S., and Jacob, K. T. (1995). Tie Lines and Activities in the System NiO-MgO-SiO2 at 1373K. Journal of Phase Equilibria, 16(3), 243-253.

Mudd, G. M. (2010). Global trends and environmental issues in nickel mining: Sulfides versus laterites. Ore Geology Reviews, 38(1-2), 9-26

Murati, A. N., Oniscin, P. B., Ystebo, R., and Vucurovic, D. (1986). Production of ferronickel by electrorduction of lateritic nickel ore at the ferronickel smelter, Glogovac, Yugoslavia. Inproceedings of Nickel Metallurgy 25th Annual Conference of Metallurgists Vol. I, Extraction and refining of nickel, pp 356-366, (eds. E. Ozberk, and S: W. Marcuson). The Metallurgical Society of CIM.

Mäkinen, T., and Taskinen, P. (2008). State of the art in nickel smelting: direct Outokumpu nickel technology. Min. Process. Extract. Metall. (Trans. Inst. Min. Metall. C), 117(2), 86-94.

Nath, N. K., Chakraborti, N., and Shekhar, R. (1995). Selective reduction of Indian nickeliferrous ore: single pellet experiments. Scandinavian Journal of Metallurgy (Processes and materials engineering), 24(3), 121-138.

46

Norgate, T., Jahanshahi, S. (2011). Assessing the energy and greenhouse gas footprints of nickel laterite processing. Minerals Engineering, 24(7), 698-707.

Nriagu, J. O. (1980). Global cycle and properties of nickel. Nickel in the Environment (ed: J. O. Nriagu), pp 833 (Environmental science and technology Wiley-Interscience publication: New York, USA).

O’Connor, F., Cheung, W. H., and Valix, M. (2006). Reduction roasting of limonite ores: effect of dehydraxylation, International Journal of Mineral Processing, 80(2-4), 88-99.

Papazoglou, D. and Rankin, W. J. (2003). Direct sulfation of nickel laterite ores using sulphur dioxide rich gases. In processdings of A. Yazawa International Symposium- Metallurgical and Materials Processing: Principles and Technologies (eds: F. Kongoli, K. Itagaki, Ch. Yamauchi and H., Y. Sohn), pp 775-767, San Diego, California USA.

Pariser, H. H. and Pariser, G. C. (2010). Changes in the Ferro Chrome & Ferro Nickel Markets. The twelfth International Ferroalloys Congress Proceedings (ed: A. Vartiainen), 641-652 (Outotec Oyj: Helsinki, Finland).

Pekkarinen, S. (2011). The uses of lateritic nickel ores and their calcination. M.Sc thesis, in finnish,Aalto University, School of Chemical Technology, Finland.

Pietilä, G. K. (2011). Reduction of nickel saprolite ore in CO/CO2 –atmosphere. M.Sc thesis, in Finnish, Aalto University, School of Chemical Technology, Finland.

Plascenia, G., Utigard, T. (2009). The reduction of Tokyo and Sinter 75 nickel oxides with hydrogen. Chemical Engineering Science, 64(17), 3879-3888.

Rhamdhani, M. A., Hayes, P. C., and Jak, E. (2008). Basic Nickel Carbonate: Part I. Microstructure and phase changes during oxidation and reduction processes. Metallurgical and Materials Transactions B (Process Metallurgy and Materials Processing Science), 39(2), B218-B229.

Rhamdhani, M. A., Hayes, P. C., and Jak, E. (2008a). Basic Nickel Carbonate: Part II. Microstructure evolution during industrial nickel production from basic nickel carbonate. Metallurgical and Materials Transactions B (Process Metallurgy and Materials Processing Science), 39(2), B234-B245.

Rhamdhani, M. A., Hayes, P. C. and Jak, E. (2009). Nickel laterite Part 1- microstructure and phase characterizations during reduction roasting and leaching. Min. Process. Extract. Metall. (Trans. Inst. Min. Metall. C), 118(3), 129-145.

Rhamdhani, M. A., Hayes, P. C., and Jak, E. (2009a). Nickel laterite part 2- thermodynamic analysis of phase transformation occurring during reduction roasting. Min. Process. Extract. Metall. (Trans. Inst. Min. Metall. C), 118(3), 146-155.

Rhamdhani, M. A., Hidayat, T., Jak, E., and Hayes, P. C. (2009b). Advances in research on nickel production through the Caron process. In Proceedings of E m c 2009, Innsbruck, Austria, June 28-July1, 2009, DGBM, Clausthal-Zellerfeld, (Vol. 3), pp 899-912.

Reinecke, I. J., and Lagendijk, H. (2007). A twin-cathode DC arc smelting test at mintek to demonstrate the feasibility of smelting FeNi from calcine prepared from siliceous laterite ores from Kazakhstan for oriel resources PLC. The eleventh International Ferroalloys Congress, Innovations in the Ferro Alloy Industri, Vol. II, pp 781-797 (The Indian Ferro Alloy Producers’ Association, India).

Rigopoulos, E., Kostika, I-P., Zevgolis, E. N., Halikia, I. (2004). Contribution to the Recycling of Rotary Kiln Dust in Nickeliferous Laterite Reduction. Reductive Behavior of the Dust. InInternational Laterite Nickel Simposium (ed: Imrie, W. P., Lane, D. M.) pp 479-492 (TMS: Charlotte, North Carolina, USA).

Shirane, Y., Kuwasaki, N. and Mitsune, Y. (1980). Activities of NiO in Solid Solutions of Ternary System MgO-NiO-SiO2. Australia/Japan Extractive Metallurgy Symposium (eds: technical papers sub-committee, Davey, T. R. A., Andrews, B. S., Ketteridge, I. B., Lean, J. B., Willis, G. M.) pp 409-418 (The Australian Institute of Mining and Metallurgy: Sydney).

Siemens, R. E. and Good, P. C. (1976). Reduction of Laterite Ores. US Patent, 3,985,556.

47

Solar, M. Y. (2008). Selection of optimum ferronickel grade for smelting nickel laterites. CIM Bulletin, paper 6, 101(1107), 1-8.

Solar, M. Y. (2009). Mechanical slag losses in laterite smelting- nickel. In Proceedings of Pyrometallurgy of Nickel and Cobalt 2009 International Symposium, (eds: Liu, J., Peacey, J., Barati, M., Kashani-Nejad, S., Davis, B.), pp 277-292 (The Metallurgical Society of the Canadian Institute of Mining, Metallurgy and Petroleum).

Springer, G. (1974). Compositional and structural variations in garnierites. The Canadian Mineralogist, 12(6), 381-388.

Szekely, J. and Evans, J. W. (1971). Studies in Gas-Solid Reactions: Part II. An Experimental Study of Nickel Oxide Reduction with Hydrogen. Metallurgical and Materials Transactions B (Process Metallurgy and Materials Processing Science), 2(6), 1699-1710

Taskinen, P., Dinsdal, A., and Gisby, J. (2005). Industrial slag chemistry: A case stady of computational thermodynamics. Scandinavian Journal of Metallurgy, 34(2), 100-107.

Taylor, A. (1995). Laterites – has the time finally come?, Mining Magazine, March, 167- 170.

Taylor, H.F.W. (1962). Homogeneous and inhomogeneous mechanisms in the dehydroxylation of minerals. Clay Minerals Bulletin, 5(28), 45-55.

Thurneyssen, C., Szczeniowski, G. J. and Michel, F. (1960). Ferro-nickel smelting in New Caledonia, Journal of Metals, 12 (1), 202-205.

Tsuji, H. (2012). Behavior of Reduction and Growth of Metal in Smelting of Saprolite Ni-ore in Rotary Kiln for Production of Ferro-nickel Alloy. ISIJ International, 52(6), 1000-1009

Utigard, T. and Bergman, R. A. (1993). Gaseous Reduction of Laterite Ores. Metallurgical and Materials Transactions B (Process Metallurgy and Materials Processing Science), 24(2), B271-B275.

Utigard, T. (1994) An analysis of slag stratification in nickel laterite smelting furnaces due to composition and temperature gradients. Metallurgical and Materials Transactions B (Process Metallurgy and Materials Processing Science), 25(4), B491-B494.

Utigard, T., Wu, M., Plascenia, G. (2005). Reduction kinetics of Goro nickel oxides using hydrogen, Chemical Engineering Science, 60(7), 3879-3888

Valix, M., Cheung, W. H. (2002). Study of phase transformation of laterite ores at high temperature, Minerals Engineering, 15(8), 607-612.

Valix, M. and Cheung, W. H. (2002a). Effect of Sulfur on the Mineral Phases of Laterite Ores at High Temperature Reduction. Minerals Engineering, 15(7), 523-530.

Vieira Coelhoa, A. C., Poncelet, G.,and Ladrièrec, J. (2000). Nickel, iron-containing clay minerals from Niquelândia deposit, Brazil: 1. Characterization. Applied Clay Science, 17(3-4), 163-181.

Walker, C., Kashani-Nejad, S., Dalvi, A. D., Voermann, N., Candy, I. M., Wasmund, B. (2009). Nickel laterite rotary kiln-electric furnace plant of the future. In Proceedings of Pyrometallurgy of Nickel and Cobalt 2009 International Symposium, (eds: Liu, J., Peacey, J., Barati, M., Kashani-Nejad, S., Davis, B.), pp 33-50 (The Metallurgical Society of the Canadian Institute of Mining, Metallurgy and Petroleum).

Warner, A. E. M., Diaz, C. M., Dalvi, A. D., Mackey, P. J., and Tarasov, A. V. (2006). JOM World Nonferrous Smelter Survey, Part III: Nickel: Laterite. JOM Journal of the Minerals, Metals and Materials Society, 58(4),11-20.

Watanabe, T., Ono, S., Arai, H. and Matsumori, T. (1987). Direct Reduction of Garnierite Ore for Production of Ferro-Nickel with a Rotary Kiln at Nippon Yakin Kogyo Co., Oheyama Works. InProceedings of the international seminar on laterite (eds: Oguar, y., and Doi, I.,), special issue of International Journal of Mineral Processing 19, 173-187.

48

Yucel, O., Turan, A., Yildirim, H. (2012). Investigation of pyrometallurgical nickel pig iron (NPI) production process from lateritic ores, 3rd International Symposium on High-Temperature Metallurgical Processing, (eds: Jiang, T., Hwang, J.Y., Masset, P., Yucel, O., Padilla, R., and Zhou, G.), pp 17-23, TMS (The Minerals, Metals & Materials Society)

Zevgolis, E. N., Zografidis. Ch., Halikia, I., Devlin, E. (2009). Roasting Reduction Study of Greek Nickeliferous Laterites. In Proceedings of EPD Congress (ed: Howard, S. M.) pp 493-500 (TMS: San Francisco).

Zevgolis, E. N., Zografidis, Ch., Perraki, Th., Deviln, E. (2010). Phase transformations of nickeliferous laterites during preheating and reduction with carbon monoxide. Journal of Thermal Analysis and Calorimetry, April, 100(1), 133–139.

Zevgolis, E. N., Zografidis, C., and Halikia, I. (2010a). The reducibility of the Greek nickelifrous laterites: a review, Mineral Processing and Extractive Metallurgy (Transaction of the Institutions of Mining and Metallurgy: Section C), 119(1), 9-17.

Zhu, D. Q., Cui, Y., Pan, J., Li, H. Q., Qiu, G. Z. (2008). Investigation on Manufacture of Ferronickel from Low Grade Nickel Laterite by Prereduction-magnetic Separation-smelting. In Proceedings of XXIV International Mineral Processing Congress (eds: Wang Dian Duo, Sun Chuan Yao, Wang Fu Liang, Zhang Li Cheng, Han Long), pp 1622-1633 (Science Press: Beijing).

http://www.npl.co.uk/advanced- materials/measurements-techniques/modelling/mtdata

HSC chemistry 7.08; Copyright © Outokumpu Research Oy, Pori, Finland, http://www.outotec.com/pages/Page.aspx?id=21783&epslanguage=EN

INSG, 2009. What is Nickel, International Nickel Study Group (INSG), www.insg.org/whatnickel.aspx, accessed.

9HSTFMG*aejjdj+

ISBN 978-952-60-4993-9 ISBN 978-952-60-4994-6 (pdf) ISSN-L 1799-4934 ISSN 1799-4934 ISSN 1799-4942 (pdf) Aalto University School of Chemical Technology Department of Materials Science and Engineering www.aalto.fi

BUSINESS + ECONOMY ART + DESIGN + ARCHITECTURE SCIENCE + TECHNOLOGY CROSSOVER DOCTORAL DISSERTATIONS

Aalto-D

D 18

/2013

Ali B

unjaku T

he effect of mineralogy, sulphur, and reducing gases on the reducibility of saprolitic nickel ores

Aalto

Unive

rsity

Department of Materials Science and Engineering

The effect of mineralogy, sulphur, and reducing gases on the reducibility of saprolitic nickel ores

Ali Bunjaku

DOCTORAL DISSERTATIONS