5th Annual High Temperature Processing Symposium ...

4–5 February 2013

5th Annual High Temperature Processing Symposium 2013

CRICOS Provider code: 00111D

Book of Abstracts

HIGH TEMPERATURE PROCESSING SYMPOSIUM 2013 Swinburne University of Technology 4 – 5 February 2013, Melbourne, Australia

Editors M. Akbar Rhamdhani Geoffrey Brooks Abdul Khaliq

Organising Committee M. Akbar Rhamdhani Geoffrey Brooks John Grandfield Abdul Khaliq Md Saiful Islam Ben Ekman Sazzad Ahmad Mohammad Mehedi Shabnam Sabah Reiza Mukhlis Md Abdus Sattar

Published in Australia by: Faculty of Engineering and Industrial Sciences, Swinburne University of Technology ISBN 978-0-9871772-6-1 © 2013 Swinburne University of Technology Apart from fair dealing for the purpose of private study, research, criticism or review as permitted under the Copyright Act, no part may be reproduced by any process without the written permission of the publisher. Responsibility for the contents of the articles rests upon the authors and not the publisher. Data presented and conclusions drawn by the authors are for information only and not for use without independent substantiating investigations on the part of the potential user.

High Temperature Processing Symposium 2013 Swinburne University of Technology

1

HIGH TEMPERATURE PROCESSING SYMPOSIUM 2013 Swinburne University of Technology 4 – 5 February 2013, Melbourne, Australia

We wish to thank the main sponsors for their contribution to the success of this symposium


2

5th High Temperature Processing Symposium 2013 Swinburne University of Technology ATC 101, Hawthorn Campus Sponsored by CSIRO, Furnace Engineering, OneSteel, Outotec

Symposium Program Day 1 (4 February 2013) in ATC101 8.30 to 9.00

Registration in Foyer Advanced Technologies Centre (ATC)

9.00 to 9.10

Welcome by Prof George Collins (Deputy Vice-Chancellor Research and Development, Swinburne University of Technology)

Session 1

Chaired by: Dr M Akbar Rhamdhani (Swinburne)

9.10 to 9.40

01 – Keynote: Prof Aldo Steinfeld (ETH Zurich/Paul Scherrer Institute) – Solar Thermochemical Processing of Fuels and Materials 02 – Dr Rene Olivares (CSIRO) - Lithium-Sodium-Potassium Nitrate Salt for Thermal Energy Storage Thermo-Chemical Evaluation in Different Atmospheres 03 – Mr Abdul Khaliq (Swinburne) – Mechanism of VB2 Formation in Molten Aluminium 04 – Dr Stephanie Vervynckt (Umicore) – Focus on Recycling of Critical Raw Materials: An Industry’s Perspective

9.40 to 10.00 10.00 to 10.20 10.20 to 10.40 10.40 to 10.55

Coffee/Tea in ATC Foyer/ATC105

Session 2

Chaired by: Prof Geoffrey Brooks (Swinburne)

10.55 to 11.25

05 – Keynote: Prof Veena Sahajwalla (UNSW) – Recycling Endof-Life Waste Materials as Resources in EAF Steelmaking – Fundamentals of High Temperature Reactions and Industrial Implementations 06 – Ms Elien Haccuria (University of Queensland) – Recycling Lithium Ion Batteries 07 – Mr Saiful Islam (Swinburne) – Kinetics of Si Refining using Slag Treatment 08 – Prof Doug Swinbourne (RMIT) - Minor Element Distributions during Copper Flash Converter

11.25 to 11.45 11.45 to 12.05 12.05 to 12.25 12.25 to 1.15

Lunch in ATC Foyer/ATC105


3

Session 3

Chaired by: Assoc Prof Brian Monaghan (Uni of Wollongong)

1.15 to 1.45

09 – Keynote: Prof Evgueni Jak (University of Queensland) – Integrated Experimental and Modelling Research Methodology for Phase Equilibria, Thermodynamics and Viscosities of Metallurgical Slags 10 – Dr Nazmul Huda (Swinburne) – Aluminium Production Route through Carbosulfidation of Alumina utilising H2S 11 – Mr Ata Fallah Mehrjardi (University of Queensland) – Investigation of Freeze-Linings in Copper-Containing Slag Systems 12 – Mr Ross Baldock (Outotec) – Circosmelt: Outotec's Alternative Ironmaking Process

1.45 to 2.05 2.05 to 2.25 2.25 to 2.45 2.45 to 3.00


Session 4

Chaired by: Mr Warren Bruckard (CSIRO)

3.00 to 3.30

13 – Keynote: Dr Mark Pownceby (CSIRO) – Insights into the Formation of Iron Ore Sinter Bonding Phases 14 – Dr Joe Herbertson (The Crucible) – The Crucible Process 15 – Mr Hamed Abdeyazdan (University of Wollongong) – The Effect of Slag Basicity on Spinel Inclusion Wettability Panel Discussion – “What is the future of energy supply for metallurgical industries” – led by Adjunct Prof John Grandfield

3.30 to 3.50 3.50 to 4.10 4.10 to 5.00

Close of Day 1

Day 2 (5 February 2013) in ATC101 8.30 to 9.00

Registration in Foyer Advanced Technologies Centre (ATC)

Session 5 9.00 to 9.30

Chaired by: Mr Richard Simpson (Furnace Engineering) 16 – Keynote: Prof George Kaptay (University of Miskolc) Current Issues of High Temperature Thermodynamics 17 – Dr Nawshad Haque (CSIRO) – Life Cycle Based Greenhouse Gas Emission Assessment from Ferroalloy Production 18 – Dr Yvonne Durandet (Swinburne) – Special Alloy Strip Production in a Micro-Mill Environment 19 – Dr Sri Harjanto (University of Indonesia) - Carbothermic Reaction of High-Combined-Water Iron Ore and Coal Composite Pellet

9.30 to 9.50 9.50 to 10.10 10.10 to 10.30

10.30 to 10.45



4

Session 6 10.45 to 11.05 11.05 to 11.25

11.25 to 11.45 11.45 to 12.05 12.05 to 12.25

Chaired by: Mr Leo Frawley (OneSteel) 20 – Prof Geoffrey Brooks (Swinburne) – Development of Dynamic Process Models for Oxygen Steelmaking 21 – Assist/Prof Youn-Bae Kang (Pohang University of Science and Technology) – Coupled Experimental and Thermodynamic Modelling for Metallurgical Slags and Inclusions Containing Sulphur 22 – Mrs Shabnam Sabah (Swinburne) - Analysis of Waves in a Cavity and Their Significance to Splashing in Steelmaking 23 – Mr Michael Sommerville (CSIRO) – Injection of Charcoal to Slag Bath 24 – Dr Sun-Joong Kim (Tohoku University) – Innovative Process of Manganese Recovery from Steelmaking Slag by Sulphurisation

12.25 to 1.15

Lunch in ATC Foyer/ATC105


Chaired by: Dr Guangqing Zhang (University of Wollongong) 25 – Dr Zulfiadi Zulhan (Insitute of Technology Bandung) - Vacuum Treatment of Molten Steel in RH (Ruerhstal Heraeus) and VTD (Vacuum Tank Degasser): A Comparative Study 26 – Dr Luckman Muhmood (CSIRO) - Experimental Investigations on the Dynamics of Interfacial Phenomena in Synthetic Blast Furnace Slags 27 – Mr Xiang Li (University of Wollongong) - Synthesis of High Purity Silicon Carbide for Solar Silicon Production 28 – Mr Taufiq Hidayat (University of Queensland) Thermodynamic Optimization of Iron-Silicate Slag for Simulation of Copper Smelting Processes

1.35 to 1.55 1.55 to 2.15 2.15 to 2.35

2.35 to 2.50



Chaired by: Mr Jacob Wood (Outotec) 29 – Dr Ryan Cottam (Swinburne) - Thermodynamic Assessment of In-Situ Formation of Hard Phase Materials 30 – Mr Zhe Wang (University of Wollongong) – Behaviour of New Zealand Ironsand during Iron Ore Sintering 31 – Mr Imam Santoso (University of Queensland) – Phase Equilibria Studies of Cu-S and Cu-Fe-S systems 32 – Mr Abdus Sattar (Swinburne) - A Comprehensive Approach for CFD Modelling of Slag Foaming with Population Balance Modelling 33 – Mr Tijl Crivits (University of Queensland) - Investigation of Phase Equilibria in the Cu-Fe-Si-Mg-O System at Low MgO Concentrations in Equilibrium with Copper Metal CLOSING

3.10 to 3.30 3.30 to 3.50 3.50 to 4.10 4.10 to 4.30 4.30 to 4.35

Close of Symposium


5

Day 3 (6 February 2013) in ATC205 Post-Symposium Workshop 8.45 to 9.00 9.00 to 13.00

Registration in ATC205, Advanced Technologies Centre (ATC) Application of Thermodynamics to Industrial Processes – by Dr M Akbar Rhamdhani, Swinburne This workshop is specifically designed for Engineers and Applied Scientists who would like to refresh their understanding of thermodynamics and/or learn its applications to industrial processes (evaluation and optimisation processes).

13.00

Close of Workshop

Campus Map – Swinburne @ Hawthorn, Melbourne


6

KEYNOTE: PRESENTATION - 1 Solar Thermochemical Processing of Fuels and Materials Aldo Steinfeld Department of Mechanical and Process Engineering, ETH Zurich, Switzerland and Solar Technology Laboratory, Paul Scherrer Institute, Switzerland www.pre.ethz.ch

Keywords: solar, thermochemical, fuel, syngas, redox, metal, gasification, carbothermal. Solar thermochemical processes for the production of synthetic fuels and materials make use of concentrated solar radiation as the energy source of high-temperature process heat. Considered are H2O/CO2-splitting thermochemical cycles via metal oxide redox reactions, gasification processes for the thermal conversion of biomass and other carbonaceous feedstock, and carbothermic reduction processes in the extractive metallurgy. These processes inherently operate at high temperatures and utilize the entire solar spectrum, and as such provide thermodynamic favorable paths to efficient and clean production [1]. Solar Fuels — Considered is the ceria-based redox cycle for splitting H2O and CO2 [2]. A 3kW solar cavity-receiver containing a reticulated porous ceramic (RPC) foam made of pure CeO2 has been experimentally investigated [3]. The RPC was directly exposed to concentrated thermal radiation at mean solar flux concentration ratios exceeding 3,000 suns. During the endothermic reduction step, solar radiative power inputs in the range 2.8-3.8 kW and nominal reactor temperatures from 1400 to 1600°C yielded CeO2-δ with oxygen deficiency δ up to 0.042. In the subsequent exothermic oxidation step at below about 1000°C, CeO2-δ was stoichiometrically re-oxidized with CO2 to generate CO. The solar-to-fuel energy conversion efficiency, defined as the ratio of the calorific value of the fuel produced to the solar radiative energy input through the reactor’s aperture, was 1.73% average and 3.53% peak. These are the highest solar-to-fuel energy conversion efficiency values reported to date for a solar-driven device converting CO2 to CO. We also demonstrated the simultaneous splitting of H2O and CO2 for the co-production of H2 and CO (syngas), whose molar ratio H2:CO was controlled by adjusting the H2O:CO2 molar ratio. Solar Gasification — The advantages of the solar-driven gasification vis-à-vis the conventional autothermal gasification are [4]: a) It delivers higher syngas output per unit of feedstock, as no portion of the feedstock is combusted for process heat; b) It avoids contamination of syngas with combustion by-products, and consequently reduces costly downstream gas cleaning and separation requirements; c) It produces syngas with higher calorific value and lower CO2 intensity, as the energy content of the feedstock is upgraded by up to 33% through the solar energy input; d) It allows for higher gasification temperatures High Temperature Processing Symposium 2013 Swinburne University of Technology

7

(>1200°C), resulting in faster reaction kinetics and higher quality of the syngas produced with low – or without – tar content, and further enabling the processing of virtually any type of carbonaceous feedstock, resulting in higher exploitation of the available resources; e) It eliminates the need for an upstream air separation unit, as steam is the only gasifying agent, which further facilitates economic competitiveness. Within the framework of a joint ETH-PSI-Holcim R&D project, a 250 kW solar industrial pilot plant was experimentally demonstrated at the solar tower concentrating facility of the Plataforma Solar de Almería (Spain) under realistic operating conditions relevant to industrial solar concentrating systems. Residual biomass and other carbonaceous wastes such as tire chips, plastics, and industrial and sewage sludges of heterogeneous characteristics (in terms of chemical composition, particle size, morphology, moisture, volatile, ash, and fixed carbon contents) were thermochemically converted to high-quality syngas with a calorific value upgraded over that of the input feedstock. Solar Metals — The extractive metallurgical industry is characterized by its energy-intensive processes and their concomitant CO2 emissions, derived mainly from the combustion of fossil fuels for heat and electricity generation. These emissions can be significantly mitigated by applying concentrated solar energy as the source of high-temperature process heat. Considered are the solar-driven productions of Fe, Zn, Mg, Al, and Si by carbothermal reduction of their oxides. When the reducing agent is derived from a biomass source, the solar-driven carbothermal processes are CO2 neutral. The EU-project SOLZINC demonstrated the solar pilot production of Zn by carbothermal reduction of ZnO. The key component is a 300-kW solar chemical reactor. It consists of two cavities in series. The upper cavity functions as the solar absorber and contains a windowed aperture to let in concentrated solar radiation. The lower cavity functions as the reaction chamber and contains a packed-bed of the reacting ZnO and biocharcoal. Testing at a solar tower in the 1300–1500 K range yielded up to 50 kg/h of 95%-purity Zn with energy conversion efficiency (ratio of the reaction enthalpy change to the solar power input) of about 30%.

upper cavity (absorber)

300 kW concentrated solar power quartz window

carrier gas inlet

lower cavity (reaction chamber)

ZnO/C packed-bed

gaseous products

1

Using concentrated solar process heat, the carbothermal reductions of Al2O3 to Al and SiO2 to Si were examined thermodynamically and demonstrated experimentally at vacuum pressures [5]. Reducing the system pressure favors Al(g) and Si(g) formation, enabling their vacuum distillation and avoiding contamination by carbides and/or oxycarbides. Exploratory High Temperature Processing Symposium 2013 Swinburne University of Technology

8

experimentation in a solar reactor was performed with mixtures of charcoal with alumina and silica in the ranges of 1300-2000 K and 1997–2263 K, respectively, at ~3×10-3 bar by direct exposure to concentrated thermal radiation. Distilled samples contained up to 19 wt.% of Al in Al-Al2O3 mixtures and 79 wt% of Si in Si-SiO2 mixtures. For all these solar thermochemical processes, R&D work encompasses fundamental studies on thermodynamics, reaction kinetics, heat/mass transfer, and chemical reactor engineering. Solar reactor prototypes – at the 10 kW power level – are designed, fabricated, modeled, and tested in a high-flux solar furnace, further optimized for maximum solar-to-chemical energy conversion efficiency, and finally scaled-up for industrial applications – at the MW power level – using concentrating solar tower technology.

References 1. Romero M., Steinfeld A., “Concentrating Solar Thermal Power and Thermochemical Fuels”, Energy & Environmental Science, Vol. 5, pp. 9234–9245, 2012. 2. Chueh W.C., Falter C., Abbott M,, Scipio D., Furler P., Haile S.M., Steinfeld A., “HighFlux Solar-Driven Thermochemical Dissociation of CO2 and H2O using Nonstoichiometric Ceria”, Science, Vol. 330, pp. 1797-1801, 2010. 3. Furler P., Scheffe J., Gorbar M., Moes L., Vogt U., Steinfeld A., “Solar thermochemical CO2 splitting utilizing a reticulated porous ceria redox system”, Energy & Fuels, in press. 4. Piatkowski N., Wieckert C., Weimer A.W., Steinfeld A., “Solar-driven gasification of carbonaceous feedstock – A review”, Energy & Environmental Science, Vol. 4, pp. 73-82, 2011. 5. Kruesi M., Galvez M.E., Halmann M., Steinfeld A., “Solar Aluminum Production by Vacuum Carbothermal Reduction of Alumina – Thermodynamic and Experimental Analyses”, Metallurgical and Materials Transactions B, Vol. 42B, pp. 254-260, 2011.


9

PRESENTATION - 2 Lithium-Sodium-Potassium Nitrate Salt for Thermal Energy Storage Thermo-Chemical Evaluation in Different Atmospheres Rene I. Olivares CSIRO Energy Centre PO Box 330, Newcastle NSW 2300, Australia Keywords: thermal energy storage, concentrating solar power, high temperature stability

Introduction With the development of tower technologies for concentrating solar power (CSP), the possibility of reaching very high temperatures has become a reality. The recent commercially demonstrated system, a 20MWe power tower Torresol Gemasolar in Spain [1], has a molten salt thermal storage capacity of 15 hours and uses a binary solar salt of 60wt% NaNO3 40wt% KNO3 that has a relatively elevated melting point of 222oC and is available up to temperature of 565oC. It would be desirable to identify lower melting point salt that is stable at higher temperature than 565oC; this would increase the heat-to-electricity conversion efficiency. Using nitrates-based salts at higher temperature may be possible by exercising rigorous atmosphere control to delay thermal decomposition [2]. The exploration of this possibility on the low melting point ternary eutectic LiNO3-NaNO3-KNO3 was studied in this work by simultaneous differential scanning calorimetry, thermogravimetry and mass spectrometry (DSC/TG-MS). Stability and thermal decomposition of nitrates The principal mode of thermal decomposition in nitrate salts has been agreed amongst researchers [3-5] to be NO3- ↔ NO2- + ½O2. Further decomposition also takes place with the evolution of the oxides of nitrogen, particularly at higher temperatures [2,6]. In some studies [3,7,8] the decomposition temperature has been reported as the temperature at which oxygen, nitrogen or nitrous oxide is detected in the gas phase. Gordon and Campbell [7] reported that the alkali metal nitrates were observed to bubble undergoing a thermal reaction at temperatures as low as 100oC to 300oC above their melting points and the evolution of nitrous fumes observed to occur at temperatures ranging from about 200oC to 350oC above the initial bubbling reaction. The later stage of the decomposition of these nitrates was still occurring at temperatures as high as 900oC which was the upper limit of the apparatus employed [7]. Experiments at 300oC with a pure equimolar sodium potassium nitrate melt in an oxygen rich atmosphere [9] reported an oxide ion concentration (O2-) in the molten salt much higher than that which would correspond to the dissociation constant for nitrite ion (NO3-), based on the equilibrium NO3-  NO2- + O2-, demonstrating that nitrate ion (NO3-) could decompose to a measurable extent at temperatures as low as 295oC. The thermal stability of binary mixes of alkali metal nitrates as investigated by means of differential thermal analysis (DTA) and evolved gas analysis (EGA) [10] concluded that the nitrate solutions are thermally unstable at temperatures higher than 500oC. Evolved gas analysis measurements [10] indicated that the decomposition of nitrates took place independently and followed the ranking order of thermodynamic stability as determined by the reactions sequence MNO3  MNO2 + ½O2 followed by 2MNO2  M2O + NO + NO2 where M can High Temperature Processing Symposium 2013 Swinburne University of Technology

10

represent Li, Na or K respectively. On this basis the decomposition of a mixture of LiNO 3NaNO3-KNO3 would be determined by the least thermodynamically stable species, in this case LiNO3 followed by NaNO3 and then KNO3, a possibility if the solution is close to ideal. In the present work the thermo-chemical behaviour and thermal stability of the LiNO3NaNO3-KNO3 eutectic was studied from at least three different points of view: (i) the temperature at which the rapid evolution of gases, NO and/or NO2 and O2, are detected by means of mass spectrometry (MS), (ii) the temperature of the melt at which an irreversible endothermic peak of decomposition is resolved by differential scanning calorimetry (DSC), and (iii) the temperature at which rapid weight loss is observed in a thermo-gravimetric curve (TG). A combination of these three criteria is used for elucidation of the limiting temperature for operation of the salt as would apply in a TES installation. The chemistry and equilibrium reactions by which molten nitrites/nitrates may interact with the atmosphere can be found in [2,11-13]. Experimental results It was found that the stability of the LiNO3-NaNO3-KNO3 eutectic, as measured by the gases evolving from the melt, was influenced by the atmosphere. Evolution of the main gaseous species NO was detected at 325oC in an atmosphere of argon, at 425oC in an atmosphere of nitrogen, at 475oC in an atmosphere of air and at 540oC in an atmosphere of oxygen. Examples of the mode of decomposition and thermal stability of the ternary eutectic are shown in Figure 1 for an argon cover gas and in Figure 2 for an air cover gas.

1E-8

N2O (N2) T: 693.19 (°C)

1E-9 (O2) T: 418.29 (°C)

O2 N2 O NO

T : 895.11 (°C)

(O) T: 588.42 (°C)

1E-10

N

(N) T: 584.12 (°C)

NO2

1E-11

(NO) T: 325.09 (°C)

(NO2) T: 659.54 (°C)

25

Rapid wt loss begin, T: 602.81 (°C)

20 15

TG |sc (mg)

MS Ion Current Intensity (A)

1E-7

10

Heat : 4.2 (J/g) Peak Maximum : 94 (°C) Solid-solid transf : 88 (°C)

5

Onset, bulk of decomposition : 671 (°C)

Exo

HeatFlow |sc (mW)

0 -20 Heat : 163.4 (J/g) Peak Maximum : 129 (°C) Melting T : 121 (°C)

-40 -60 -80 100

200

300

400

500 600 Sample Temperature (°C)

700

800

900

1000

Figure 1: DSC/TG-MS analysis of thermal decomposition of lithium-sodium-potassium nitrate eutectic under a cover gas of argon The delay in the thermal decomposition as measured by the evolution of gaseous NO is clearly demonstrated in Figure 2 for the salt under a cover gas of air.


11

1E-9 (NO) T: 475.72 (°C)

1E-10

1E-11

(NO2) T: 550.41 (°C)

1E-12

24 Rapid wt loss begin, T: 601.93 (°C)

20

TG |sc (mg)

MS Ion Current Intensity (A)

1E-8

16 12 8

HeatFlow |sc (mW)

20

Heat : 4.1 (J/g) Peak Maximum : 95 (°C) Exo Solid-solid transf : 88 (°C)

4 Onset, bulk of decomposition : 625 (°C)

0 -20

Heat : 135.5 (J/g) Peak Maximum : 128 (°C) Melting T : 122 (°C)

-40 -60 100

200

300

400


700

800

900

1000

Figure 2: DSC/TG-MS analysis of thermal decomposition of lithium-sodium-potassium nitrate eutectic under a cover gas of air

Exo

40 30 20

Heat : -97.4 (J/g) Peak Maximum : 93 (°C) Onset (solidification temperature) : 98 (°C)

10

HeatFlow |sc (heating) (mW)

0 Exo

0 -10 -20

HeatFlow |c [6] (Cooling) (mW)

Cycling the melt several times between 50oC and 400oC allowed the determination of the melting and solidification points respectively. Figure 3 shows that prior to melting; the eutectic underwent endothermic (α/β) solid-solid type transformation. From an average of four cycles; theCycling (α/β)LiNaK type nitrate transformation occurred at 87oC, the point was 121oC, and eutectic between room temperature andmelting 400 Celsius o the solidification point 98 C. Under-cooling of the salt coincided with the onset of the (α/β) solid-solid transformation upon heating. The measurement of temperature is accurate to 2.7oC.

Heat : 6.9 (J/g) Peak Maximum : 94 (°C) Onset (solid-solid transf): 86 (°C)

Heat : 140.5 (J/g) Peak Maximum : 128 (°C) Onset (melting temperature) : 122 (°C)

-30 -40 -50 -60 50

100

150


300

350

400

Figure 3: Melting point and solidification point determination of lithium-sodium-potassium nitrate eutectic High Temperature Processing Symposium 2013 Swinburne University of Technology

12

Exo

1.00

20

0.98

400oC

Peak Maximum : 92 (°C) Onset (solidification temperature) : 97 (°C)

0.96

y= 0.94

2E-07x 2 -

0.0002x + 0.9997 R² = 0.9999

500oC

0.92

-20 20

Exo

10 Peak Maximum : 90 (°C)

0 Onset : 82 (°C)

Peak Maximum : 127 (°C) Onset : 120 (°C)

-10 -20 -30 100

150


0.90 0

60

120

0 -10

HeatFlow |c |sc (heating) (mW)

Fractional weight loss

10

y = -2E-05x + 1 R² = 0.999

180

300

350

400

240

Time at temperature, min 5 0

Peak Maximum : 130 (°C) Onset (endothermic): 135 (°C)

Peak Maximum : 222 (°C) Onset (endothermic): 224 (°C)

-5 -10

Peak Maximum : 90 (°C) Onset (solidification temperature): 93 (°C)

HeatFlow |c |sc (heating) (mW)

30

-15

HeatFlow |c |sc (cooling) (mW)

Exo

-20

Exo

25

Peak Maximum : 219 (°C) Onset (exothermic): 213 (°C)

20

15

10 50

Peak Maximum : 122 (°C) Onset (melting temperature) : 118 (°C) 100

150


300

350

400

Figure 4: Long term stability of salt at 400oC and 500oC respectively and DSC analysis after 4 hours Although the salt can be stabilised in an oxygen rich atmosphere to reach 540oC [2], the longterm stability of the ternary eutectic at greater than 500oC is considerably impacted, this being evident by the increased rate of salt vaporization as measured by TG analysis. The use of the salt in a closed or slightly pressurised containment arrangement is expected to minimise this problem.

References 1. R.I. Dunn, P.J. Hearps, M.N. Wright, Molten-Salt Power Towers: newly Commercial

Concentrating Solar Storage, Proceedings of the IEEE 100(2) (2012) 504-515. 2. R. Olivares, The Thermal Stability of Molten Nitrite/Nitrates Salt for Solar Thermal Energy Storage in Different Atmospheres, Solar Energy 86 (2012) 2576-2583.


13

HeatFlow |c |sc (cooling) (mW)

Long term stability At a temperature of 500oC in air, TG analysis indicated that the long term stability of the salt was limited and this was confirmed by DSC. Figure 4 shows that for a salt sample after four hours at 500oC, although the melting and solidification points were unaffected, the DSC signature of the salt was significantly different.

3. E.S. Freeman, The kinetics of the thermal decomposition of sodium nitrate and of the

4. 5. 6.

7.

8. 9. 10. 11. 12.

13.

reaction between sodium nitrite and oxygen, Journal of Physical Chemistry 60(11) (1956) 1471-1600. G.D. Sirotkin, Equilibrium in melts of the nitrates and nitrites of sodium potassium, Russian Journal of Inorganic Chemistry 4(11) (1959) 1180-1184. A. Buchler, J. Stauffer, Gaseous Alkali-Nitrogen-Oxygen and Alkali-PhosphorousOxygen Compounds, The Journal of Physical Chemistry 70 (1966) 4092-4095. R.F. Bartholomew, A study of the equilibrium KNO3(l) = KNO3(l) + 1/202(g) over the temperature range 550-750oC. The Journal of Physical Chemistry 70(11) (1966) 34423446. S. Gordon, C. Campbell, Differential thermal analysis of inorganic compounds nitrates and perchlorates of the alkali and alkaline earth groups and their subgroups, Analytical Chemistry 27(7) (1955) 1102-1109. B. Bond, P. Jacobs, The thermal decomposition of sodium nitrate, Journal of the Chemical Society A (1966) 1265-1269. R.N. Kust, J.D. Burke, Thermal decomposition in alkali metal nitrate melts, Inorganic Nuclear Chemistry Letters 6 (1970) 333-335. O. Abe, T. Utsunomiya, T. Hoshino, The thermal stability of binary alkali metal nitrates, Thermochimica Acta 78 (1984) 251-260. D.A. Nissen, D.E. Meeker, Nitrate/nitrite chemistry in sodium nitrate-potassium nitrate melts, Inorganic Chemistry 22(5) (1983) 716-721. R.W. Bradshaw, N.P. Siegel, Molten nitrate salt development for thermal energy storage in parabolic trough solar power systems, Energy Sustainability ES (2008) Jacksonville, Florida 4p. C. M. Kramer, DSC of Sodium and Potassium Nitrates and Nitrites, Thermochimica Acta 55 (1982) 11-17.


14

PRESENTATION - 3 Mechanism of VB2 Formation in Molten Aluminum A. Khaliq1, 2, M.A. Rhamdhani1, 2, G.A. Brooks1, 2, J. Grandfield1, 2, 3 1 Swinburne University of Technology, Faculty of Engineering and Industrial Sciences, Melbourne, Australia 2 CAST Cooperative Research Centre (CAST CRC), Australia 3 Grandfield Technology Pty, Ltd, Victoria, Australia Keywords: Transition metal borides, kinetics, mechanism, molten Al, boron treatment Smelter grade aluminium is used for electrical grade conductor applications. Impurities present in solution, especially transition metals such as V, Ti, Zr and Cr reduce the electrical conductivity of smelter grade aluminium [1-2]. In the cast houses, these impurities such as V, Ti, Zr and Cr are removed by the addition of Al-B (AlB12/AlB2) master alloys, called boron treatment [1-7].The thermodynamic analysis of transition metal impurities in molten aluminium [8], it was predicted that the order of impurities removal will be from Zr and Ti to V in the temperature range of 650oC to 900oC. It was further predicted that diborides of transition metals are more stable as compared with their other possible borides in the temperature range investigated [8]. Although the process for the removal of transition metal impurities is explained in literature, limited information has been reported related to the kinetics of process and the mechanism of borides formation. For the better understanding of boron treatment of molten aluminium, kinetics experiments were performed. In this paper, selected results and the mechanism of VB2 formation in molten aluminium are presented. In the kinetic experiments, removal of vanadium was investigated by the addition of boron (stoichiometric amount to form VB2) addition in the form of Al-B (AlB12) master alloy, assuming reactions given in Eq. (1) and (2). An alloy of Al-1wt%V was prepared in the resistance pot furnace and weighed ingots of Al-B master alloy were added assuming 100% recovery of boron. Samples were taken at 0, 2, 4, 6, 8, 10, 15, 30, 45 and 60 minutes interval after the addition of boron. The experiments were performed at 700 oC, 750oC and 800oC and melt was held in the clay bonded graphite crucible. Samples were analysed for vanadium in solution with aluminium using ICP-AES technique. The possible reactions during the formation of VB2 are shown in Eq. (1) & (2). The thermodynamic feasibility of transition metal borides formation has already been explained elsewhere [8]. (1) (2) Where “[ ]” indicates that elements are dissolved in solution with molten aluminium and “(s)” represents that compounds are present in solid state. The change in concentration of [V] at 750oC and SEM image of early stage reaction are shown in Figure 1. Figure 1(a) shows that the concentration of V in solution with aluminium much less than its solubility (0.61wt %) at 750oC. Vanadium out of solution was most likely present in the form of Al10V/Al7V intermetallics. These particles were settled during the melt holding period and were observed under SEM in the borides sludge collected from the bottom High Temperature Processing Symposium 2013 Swinburne University of Technology

15

of crucible. The general observation of curve shown in Figure 1(a) also revealed that the rate of [V] removal was rapid in the beginning of reaction and slower down after 10 minutes of reaction time. This observation along with microstructural study suggested that the kinetics of VB2 formation may be controlled by two kinetic mechanisms. In the beginning, VB2 formation may be controlled by chemical reaction or liquid metal phase mass transfer and at the later stage 2 (e.g beyond 10 minutes) is most likely controlled by diffusion through the product boundary layer (VB2). The formation of product (VB2) layer has already been reported in our previous work [7] and is also shown in Figure 1(b). Similar trends of [V] removal were observed in experiments carried out at 700oC and 800oC.

(a)

(b)

Figure 5. (a) The change in [V] with time of Al-V-B alloy and (b) SEM image of Al-1wt%V alloy after the addition of Al-10wt%B (AlB12) master alloy at 750oC One of the important aspects of heterogeneous (solid-liquid) reaction is the interfacial area. Careful estimation of interfacial area will predict the kinetic of reaction close to actual situation which will be helpful in designing the chemical processes. In this particular case, AlB12 particles surface area was estimated using image analysis technique. Assuming the kinetic was controlled by the chemical reactions and follow the first order with respect to boron, the reaction rate constant was calculated for the first 10 minutes of process. The values of rate constant/mass transfer at 700oC, 750oC and 800oC were calculated to be 1.61x103 m/sec, 2.095x10-3m/sec and 2.685x10-3m/sec respectively. The activation energy was calculated to be 23.73kJ/mol. The curve fitting with Jander model as shown in Figure 2 (a) and other diffusion control models such as Ginstling-Brounshtein after 10 minutes of reaction gave a linear fit with R2 values more than 99%. This suggested that the later stage of reaction during the formation of VB2 is control of diffusion through VB2 shell. It could be observed from the preliminary kinetic analysis of [V] removal and the formation of VB2 in molten aluminium that the reaction mechanism is complex which show mixed control. The kinetics of reaction was most likely controlled by chemical reactions or liquid metal phase in the first 10 minutes of reaction and by the diffusion through product layer in the later stage during the boron treatment of molten aluminium. The detail investigation of the kinetic analysis is under progress that will conclude the rate limiting steps during the formation of VB2 during boron treatment of molten aluminium.


16

(a)

(b)

Figure 2. (a) Experimentally determined [V] conversion curves at 700oC, 750oC and 800oC and (b) The kinetics data after 10 minutes reaction were plotted using Jander model. References 1. Gauthier, G. G. 1936. The conductivity of super-purity aluminium: The influence of small metallic additions. J. Inst. Met., 59, 129-150. 2. Dean, W. A. 1967. Effects of Alloying Elements and Impurities on Properties. Aluminum, 1, 174 3. Cooper, P. S. and Kearns, M. A. 1996. Removal of transition metal impurities in aluminium melts by boron additives. Aluminium Alloys: Their Physical and Mechanical Properties, Pts 1-3, 217, 141-146. 4. Dube, G. 1983. Removal of Impurities from molten aluminium. 833068034. 5. Stiller, W. and Ingenlath, T. 1984. Industrial Boron Treatment of Aluminium Conductor Alloys and Its Influence on Grain Refinement and Electrical Conductivity. Aluminium (English Edition), 60 6. Karabay, S. and Uzman, I. 2005. Inoculation of transition elements by addition of AlB2 and AlB12 to decrease detrimental effect on the conductivity of 99.6% aluminium in CCL for manufacturing of conductor. Journal of Materials Processing Technology, 160, 174182. 7. Khaliq, A., Rhmadhani, M. A., Brooks, G. A., Grandfield, J., Mitchell, J. and Davidson, C., 2011. Analysis of Transition Metal (V, Zr) Borides Formation in Aluminium Melt. In: EMC (European Metallurgical Conference) June 26-29 Dusseldorf, Germany. 825838. 8. Khaliq, A., Rhamdhani, M. A., Brooks, G. A. and Grandfield, J. 2011. Thermodynamic analysis of Ti, Zr, V and Cr impurities in aluminum melt. In: TMS 2011, February 27 March 3, San Diego, CA. 751-756.


17

PRESENTATION - 4 Focus on Recycling of Critical Raw Materials: An Industry’s Perspective Stephanie Vervynckt, Mieke Campforts Umicore Group Research & Development, Kasteelstraat 7, 2250 Olen, Belgium Today’s transition to a low carbon economy boosts the demand of technology metals. As these metals have multiple competing applications, their supply chain is pressurized. Take for example precious metals which have show their use in catalysts for the conversion of harmful components in car exhaust gases and in industrial catalysts but are also needed in electronics such as cell phone and laptops. Because of the development of catalysts and electronics the precious metals demand has risen resulting in an increase in mining. To give some numbers, 9 times as much Ru, Pd and Ru are mined between 1980 and 2010 compared to the 80 years before. Also recycling of these products has increased as the metal prices can pay off for the processing even though in some products such as electronics only small volumes are present. This increase in metal prices is due to the potential metal scarcity that is general encountered for technology metals. Primarily the main driver is of a temporary base, namely that there is a mismatch between demand and supply due to new developments, speculation, trade barriers because of securing access to raw materials, time lag and investment risk for new mines and smelters. But in case of the technology metals, also structural scarcity needs to be taken into account. These ‘minor’ metals are only accessible as by-product of major metals due to coupled production. As a result it is not surprising that worldwide exercises are done to map the criticalness of materials and programs are setup in order to assure access to raw materials. The challenge to tackle resources scarcity is further complicated by several issues. Here four additional challenges are defined. First of all everything is connected. The high degree of linkages among resources means strong demand for one can spread to others. The production of technology metals will use energy, land, water and produce carbon dioxide. Secondly products are complex (containing multiple metals) and getting more complex with time. Moreover, element combinations that are not encountered in nature put new challenges to recycling industry as most extraction processes are developed to treat ore. This complexity is mirrored in the supply chain and the recycling process flow sheets. A third challenge is to keep materials in the cycle with the main challenge being end of life consumer goods. A last challenge is the knowledge, awareness and engagement of people on material life cycle in order to support and drive optimal closing of the loop. This presentation describes how Umicore contributes to achieve a secured supply chain through recycling and discusses which challenges are still ahead.


18

KEYNOTE: PRESENTATION - 5 Recycling End-of-Life Waste Materials as Resources in EAF Steelmaking – Fundamentals of High Temperature Reactions and Industrial Implementations M. Zaharia(*, M. Rahman(*, N. F.Yunos, R. Khanna, N. Saha-Chaudhury, P. O’Kane(*, A. Fontana(**, J. Dicker(*, C. Skidmore(* and V. Sahajwalla SMaRT Centre, University of New South Wales, NSW, 2052, Australia (* OneSteel, Sydney Steel Mill, Rooty Hill, NSW (** OneSteel, Laverton Steel Mill, VIC Keywords: waste, steelmaking, slag foaming, reduction The steel industry consumes a large amount of energy GHG emissions [1]. Rubber tires and agricultural wastes have the potential to be used in industries seeking alternative fuel and sustainable raw materials sources. Previous studies focused on recycling these materials as fuel resources, i.e. rubber in cement industry [2-3] and agricultural materials for power production [4]. The present paper focuses on investigations of carbon /slag reactions, namely slag foaming using rubber and palm shell wastes as sustainable carbon sources through quantitative estimation of the slag volume. An improved volume ratio for the rubber blend compared to coke was seen. Foaming was also improved when palm shell char was used as carbon material. Industrial implementations at OneSteel showed reductions in electrical energy and carbon consumption. These results indicate that partial replacement of coke with rubber and palm shell is efficient due to improved interactions with EAF slag. Iron and steelmaking processes use carbon as one of the main input materials. Anthracite and metallurgical coke are the conventional materials employed for reduction of iron oxides, slag foaming in EAF steelmaking. In order to address the issues of cost, availability and restrictions on greenhouse gas emissions, alternative carbon sources are required to replace, at least partially, these conventional materials. Postconsumer plastics such as HDPE (Highdensity polyethylene), end-of-life rubber tyres as well as palm shells contain both carbon and hydrogen. Meanwhile, expansion in the use of polymeric materials over the last 3 decades has also been accompanied by increasing problems over their disposal at the end of their life cycle. The utility and relatively short lifespan of plastics and rubbers have produced a massive waste problem. Conventional recycling technologies are unable to deal with the high volume of wastes produced. Agricultural waste materials derived from palm shells are among the main renewable waste sources in Malaysia and their quantities were seen to increase steadily in the last few years [5]. Approximately 4.7 million tonnes of palm shells were generated in 2006 alone, and a steady 5% increase has been seen in the last few years [6]. Palm shell waste is a true renewable material, so it does not contribute to the increase in the global CO2 concentration. The high temperature environments offer sustainable pathways for utilising chemical reactions to re-purpose waste materials as resources; such as reducing iron oxide to iron. The


19

current study describes the broad opportunities available to iron and steelmakers to utilise waste materials – ranging from polymeric materials to agricultural wastes -- as raw materials. At University of New South Wales (UNSW) a technology was pioneered that allowed polymer injection into the EAF steelmaking [7]. The focus was set on using the polymer material as a carbon replacement for coke to achieve slag foaming. Based on this in 2006 OneSteel adopted the new technique and started replacing part of the metallurgical coke with HDPE plastic. In 2007, rubber tyres were also considered and are now standard practice for the Sydney and Melbourne EAF plants [8-9]. The trials indicated that a reduction in the overall power consumption can be achieved (in the order of millions of kWh/year), therefore enhancing productivity with an associated reduction in energy [10]. This represents significant cost savings for an average EAF plant. The experimental procedure for carbon slag reactions involved: (1) reactions in a custommade horizontal furnace having the capability to reach 1550°C, (2) visual observation using a charge-coupled device (CCD) camera, (3) off-gas analysis. Details can be found elsewhere [8]. As the current study introduces more complex materials, polymers and palm shells, the reactions occurring at the slag/carbon interface are expected to be affected by the presence of an increased level of hydrocarbons that could further decompose into carbon and hydrogen bearing products. When put in contact with an iron oxide rich EAF slag, the presence of iron oxide leads to a reduction reaction depending on the reducing agents including C, CO and H 2 released at high temperature [11]. The produced gases (CO, CO2, CH4, H2O, H2) will allow the slag to foam. The rate of gas generation following the interaction of the HDPE-coke blend was seen to be the fastest, followed by the rate of gas generation from rubber-coke blends, while the lowest rate was seen when coke was the carbon material [12]. The high amount of CO and CO2 generated when HDPE and rubber replaced part of the coke could be attributed to a certain extent to the volatiles in the carbonaceous mixture. These volatiles are predominantly CH4 gas, which, at the temperature of the tests, transforms into CO and H2. Gas entrapment in the slag phase was quantified through slag volume measurements. In order to qualitatively demonstrate the slag foaming behaviour, a few representative dynamic images of the slag droplet in contact with metallurgical coke (MC) and its blends with rubber and HDPE and palm char are shown in Figure 1. The size of the slag droplet in contact with MC decreased with reaction time. HDPE blend showed significantly higher levels of gas entrapment and also rubber showed increased volume with time compared to coke In order to quantify the changes in slag volumes, attributed to formation, entrapment and release of gases, Vt/V0 was calculated and plotted as a function of time (Figure 2). Rubber/ HDPE blends and palm shell showed an increased slag volume compared to coke. HDPE blend revealed significantly higher volume ratios as a result of increased gas generation and entrapment. The volume ratio, Vt/Vo, of the EAF slag reacting with metallurgical coke records an initial value of 1.0; with time it decreased without any wide fluctuations to 0.5. This indicates a lower extent of gas entrapment by the slag. For the rubber blend, the slag volume ratios showed significantly different trends with much higher levels of droplet volumes (Figure 2a). The palm char showed fluctuations, with the drop volume High Temperature Processing Symposium 2013 Swinburne University of Technology

20

continuously growing and decreasing, however maintaining a higher volume compared to coke. Kinetics of reduction when HDPE partially replaced coke usage is reported elsewhere [11]. 0 sec

480 sec

60 sec

100% Coke

0 sec

480 sec

60 sec

HDPE Blend

0 sec

480 sec

60 sec

Rubber Blend

PET Blend

Figure 1 High temperature images of slag droplets in contact with 100% MC, HDPE Blend, Rubber Blend, and Palm Char at 1550ºC as a functionPU of Blend time [11-12] 4

4 100%MC

a) MC_Rubber Blend 2

MC_Rubber Blend 1

2

MC-HDPE Blend

3

MC-Rubber Blend 2

Volume Ratio, Vt/V0

Volume Ratio, Vt/V0

3

100%MC

b)

MC-Rubber Blend 1

100% MC 1

MC_HDPE Blend 2 100% MC 1

0

0 0

200

400

600

800

1000

0

200

400

Time, sec

600

800

1000

Time, sec

4

Volume Ratio, Vt/Vo

c)

100%MC

3 Palm Shell Palm Shell

2

100% MC 1

0 0

200

400

600

800

1000

Time, sec

Figure 2:Volume ratio of a) 100% MC - Rubber Blend, b) 100% MC -HDPE Blend, c) 100% MC - 100% Palm Shell interacting with EAF iron oxide slag as a function of time [15-16] High Temperature Processing Symposium 2013 Swinburne University of Technology

21

The results were subsequently confirmed, as of 2006 in industrial implementatios at OneSteel EAFs at Sydney Steel Mill, New South Wales, and Laverton Steel Mill, in Melbourne, Victoria. OneSteel has been using this technology as standard practice for over four years at SSM and LSM and 40,000 heats, consuming over one million recycled tyres [13-14]. This technology, referred to as, Polymer Injection Technology (PIT), was consequently patented, and introduced to third parties and is currently in use at UMC Metal, Thailand, where it was commissioned in May 2011 [14]. In all three cases, benefits derived from the implementation of this technology translated into cost and productivity benefits, including the reduction of foaming agent injectant, the reduction of power on time. A summary of the data for the trial heats clearly shows that the HDPE/rubber blend performs better than coke. Table I. Summary of benefits at SSM and LSM EAF [15-16] Injected materials Specific EE (KWh/t) Carbon (Kg/heat) FeO (%) Coke Recycled tyres HDPE

424 412.4 406

462 406 379

27.6 26.2 26.1

This study has shown that blends of rubber/HDPE with metallurgical coke and palm char could be used to partially replace some of the conventional metallurgical coke used in EAF steelmaking for its injecting carbon requirements. The laboratory work was reflected in the industrial implementations at OneSteel where improvements in the slag foaming behaviour and furnace efficiency were seen. References: World Resources Institute, July 2009 E Mokrzycki, A. U.- Bocheńczyk, Applied Energy 2003, Vol 74, Issues 1–2, pp. 95-100 M. Bluementhal, World Cement, 1992, Vol. 23, n12, pp.14-20 J. L. Easterly, M. Burnham, Biomass and Bioenergy , 1996, Vol 10, Issues 2–3, Pages 79–92 L. Chor, Y. Laveraging on Sustainability; Malaysian Palm Oil Institute: 2010. K. Mae, I. Hasegawa, N. Sakai, K. Miura, Energy & Fuels 14, 1212 (2000). V. Sahajwalla, L. Hong, and N. Saha-Chaudhury: Iron and Steel Technology: AIST magazine 2006, pp. 99–96. 8. M. Zaharia, V. Sahajwalla, R Khanna, P. Koshy and P. O’Kane, ” Carbon/slag interactions between Coke/Rubber Blends and EAF Slag at 1550°C”, ISIJ International 2009, 49(10) 9. V. Sahajwalla, M. Rahman, R. Khanna, N. Saha-Chaudhury, P. O’Kane, C. Skidmore and D. Knights: Steel Research Int. (2009), Vol. 80(8), pp. 531–539. 10. V. Sahajwalla, R. Khanna, M. Zaharia, S. Kongkarat, M. Rahman, B. C. Kim, N. Saha-Chaudhury, P. O’Kane, J. Dicker, C. Skidmore and David Knights: Iron and Steel Tech magazine, April 2009: 43(6) 11. J. R. Dankwah, P. Koshy, P. O’Kane and V. Sahajwalla, C. Skidmore, D. Knights, D. ISIJ International 2011, 51(3), 498–507 12. V. Sahajwalla, M. Zaharia, S. Kongkarat, M. Rahman, B. C. Kim, N. Saha-Chaudhury, P. O’Kane, J. Dicker, C. Skidmore and David Knights: Energy and Fuels, 2012, 26 (1), pp 58–66, DOI: 10.1021/ef201175t 13. A. Fontana, P. O’Kane, V. Sahajwalla, M.Zaharia, Steel research International, pp. 17-20, 2012 14. A. Fontana, P. O’Kane, D. O’Connell, & M. Schroer (2012), in Proceedings 2012 SEAISI Conference and Exhibition, Bali, Indonesia. 1. 2. 3. 4. 5. 6. 7.


22

PRESENTATION - 6 Recycling Lithium Ion Batteries Elien Haccuria, P. Hayes and E. Jak Pyrosearch, The University Of Queensland, Brisbane, QLD 4072, Australia. Keywords: lithium ion batteries, manganese, phase diagram, Al2O3-“MnO”-SiO2 slag system Introduction The lithium ion battery (LIB) market has grown considerably during the recent years and is expected to keep growing due to the application in the automotive sector. Safe recycling and extraction of valuable metals present in the batteries, including nickel, cobalt, copper and manganese, are important. Pyro-metallurgical processes involving a slag (molten oxide) phase, are especially being developed for that purpose [1-4]. Hence, the effect of manganese on the liquidus temperatures and thermodynamic properties of the possible Al2O3-CaO“MnO”-SiO2 slag system are important. Phase diagram Al2O3 - “MnO”- SiO2 Revision of phase equilibria in the key Al2O3-“MnO”-SiO2 pseudo-ternary slag sub-system in equilibrium with metallic alloy is the focus of the present study. More specific, the discrepancies between previous investigators Roghani et al. [5], Jung et al. [6] and Snow [7] are being resolved. Experiments and preliminary results The applied experimental technique is based on the equilibration/quenching/electron probe X-ray microanalysis (EMPA) approach. Slag samples are prepared from pure oxide powers with addition of excess Mn or a Mn-Si alloy and equilibrated on a silica crucible in an argon atmosphere. The equilibration temperatures vary between 1150 and 1250 °C and the equilibration time ranges from 0 h to 24 h. Particular attention in this initial stage of the study was paid to the investigation of the system behaviour, improvements in accuracy and above all, to the confirmation of the achievement of equilibria, this included the study of i) the effect of equilibration time, ii) the homogeneity of phases, iii) the attainment of equilibria from different directions, and iv) the analysis of the reactions taking place in samples during experiments. It was found that a Mn5Si3 intermetallic compound is formed in equilibrium with the slag at investigated conditions after more than 6 hours equilibration. If pure metallic Mn was added to the slag, the following changes in bulk composition were identified (Figure 1). The slag phase initially enriched with MnO due to the reactions between alloy and slag systems. Subsequently, the SiO2 concentration in the slag increases due to the dissolution of the substrate. Figure 1: Changes in bulk composition if pure Although initial work on the improvement of metallic Mn is added to the slag High Temperature Processing Symposium 2013 Swinburne University of Technology

23

accuracy is still in progress, preliminary results of the liquidus in the tridymite and rhodonite primary phase fields can be analysed. Figure 2 presents some preliminary results obtained from the samples equilibrated for 0.5, 1, 2 and 24 hours. A range of local areas with different liquid phase compositions were found in the samples equilibrated for 0.5 and 1 hour. If local equilibrium is achieved between liquid, solid (Tephroite Mn2SiO4 and Rhodonite MnSiO3), and alloy system, then the points may represent equilibrium. This assumption is yet to be confirmed, however consistency between results from 2, 6 and 24 hours samples are an initial indication for that. These measurements are in agreement with Roghani [5] (Figure 2). Note that these results are preliminary and cannot be taken as final equilibrium results. Further work is going on to verify these preliminary findings. L + Tr

Tr R 1200 °C

Tr A

L + R+ Tr

R L + R+ Te

Te

L+R

R

Figure 2: Preliminary liquid phase composition measurements in local areas of incompletely equilibrated samples at 1200 °C in comparison with the 1200 °C liquidus isotherms by Roghani et al. [5] and Jung et al. [6]. Phases present: R) rhodonite, Tr) tridymite, Te) tephroite, A) alloy

Several other issues are also under investigation. For example, a tridymite ring is formed around the alloy particles in samples equilibrated for two hours or more, resulting in the seclusion of the alloy from the liquid (Figure 3). The mechanism for this ring formation and measures to avoid it, are being investigated. The research into such reactions is essential to ensure accuracy and reliability of the final results. R A

A

Tr Tr

R

1h

Figure 3: Alloy enclosed by a tridymite Phases present: R) rhodonite, Tr) tridymite, A) alloy


A

6h

ring

24 h

after

longer

equilibration

times

24

Future plans Further study is necessary to further improve the accuracy of the results. Additional research will be focused on resolving the discrepancies in the Al2O3-“MnO”-SiO2 pseudo-ternary slag sub-system. The work is planned to be expended to the quaternary system containing CaO. References 1. 2.

3. 4. 5. 6.

7.

Maschler, T., et al., Development of a recycling process for Li-ion batteries. Journal of Power Sources, 2012. 207: p. 173-182. Dewulf, J., et al., Recycling rechargeable lithium ion batteries: Critical analysis of natural resource savings. Resources, Conservation and Recycling, 2010. 54: p. 229234. Xu, J., et al., A review of processes and technologies for the recycling of lithium-ion secondary batteries. Journal of Power Sources, 2008. 177: p. 512-527. Umicore. Umicore Battery Recycling. 2012; Available from: http://www.batteryrecycling.umicore.com/UBR/process/. Roghani, G., E. Jak, and P. Hayes, Phase equilibrium studies in the "MnO"-Al2O3SiO2 system. Metallurgical and materials transactions B, 2002. 33B: p. 827-38. Jung, I., et al., Thermodynamic evaluatioin and optimization of the MnO-Al2O3 and MnO-Al2O3-SiO2 systms and applicatioins to inclusion engineering. Metallurgical and materials transactions B, 2004. 35B: p. 259-268. Snow, R.B., Equilibrium relationschips on the liquidus surface in part of the MnOAl2O3-SiO2 system. Journal of The American Ceramic Society, 1943. 26(1): p. 11-20.


25

PRESENTATION - 7 Kinetics of Silicon Refining using Slag Treatment Md Saiful Islam, M. Akbar Rhamdhani, Geoffrey A. Brooks Faculty of Engineering and Industrial Sciences, Swinburne University of Technology, Victoria, Australia [email protected] Keywords: silicon; solar grade silicon; metallurgical grade silicon; kinetics Silicon, an important semiconducting material and alloying element in metallurgy and chemistry, is one the most abundant elements in the earth as oxides and silicates. The rapid growth of solar cell demand is creating a shortage of solar-grade silicon feedstock. Expensive high-purity scrap silicon (99.9999999% Si) is mainly used as the raw material to produce solar-grade silicon (SOG-Si) (99.9999% Si). Many researchers have reported that relatively inexpensive metallurgical grade silicon (MG-Si) (98-99% Si) can be used as an alternative raw material. Of all the impurities present in MG-Si, boron and phosphorus are usually the most difficult to remove. Slag refining is one of the few metallurgical methods for the efficient removal of boron from silicon. In order to produce silicon for photovoltaic applications, the relationship between the slag composition and the mass transfer rate of boron from silicon to slag is of great importance. The slag treatment on MG-Si for SOG-Si production is based on the principle of liquid-liquid extraction similar to those applied in the steel industry. Impurities with a higher oxygen affinity, in comparison with silicon, oxidise and pass into the slag. The thermodynamics of the oxidation removal of boron by the slag treatment have been of interest to researchers, and numerous studies have been carried out on slag refining [1-2]. In the case of boron oxidation, the equilibrium reaction is based on Equation 1. From Equation 1, in order to remove boron from molten silicon by the slag treatment, it needs to maintain high oxygen potential, which is restricted by Si/SiO2 equilibrium at the interface between molten silicon and slag, and simultaneously decreases the activity of borate in the slag. Because borate is an acidic oxide, its basicity should be kept high with basic oxides (CaO or Na2O). In this method, liquid silicon is treated with CaO-SiO2, CaO-SiO2-CaF2, CaO-SiO2-Al2O3, CaO-SiO2-Al2O3-MgO and other molten slags [3].

(1) During refining, the rate of boron removal into the slag where the refining takes place is of great importance. It is therefore useful to assess the removal rate of boron with the equilibrium boron distribution. A list of selected studies on kinetics of boron removal from silicon and mass transfer coefficient values are shown in Table 1. The relationship between slag composition and optimal refining efficiency is still not well known, so a better understanding about the relationship between slag composition and mass transfer can improve the basic understanding of the refining process.


26

Table 1: Selected kinetics experiments for boron removal Investigators (Krystad, Tang et al. 2012) [4]

(Nishimoto and Morita 2011) [5]

Initial Slag & Alloy Composition

Experimental Conditions

SiO2-CaO CaO/SiO2=1 Si with 250 ppm B

Electric resistance furnace Temperature=1873K;atm=Ar Equilibrium time=3hr Slag/Silicon mass ratio=1 & 2

SiO2-CaO-20%MgO CaO/SiO2=1 Si with 250 ppm B

Electric resistance furnace Temperature=1873K;atm=Ar Equilibrium time=3hr Slag/Silicon mass ratio=2 Electric resistance furnace Temperature=1873K;atm=Ar Equilibrium time=3hr Slag/Silicon mass ratio=2.5

SiO2-CaO CaO/SiO2=1.22 Si with 300 ppm B

Mass Transfer Coefficient (m/s)

Rate controlling step

2.1 x 10-6 (S/M=1) 1.2 x 10-6 (S/M=2)

--

3.2 x 10-6

--

1.4 x 10-6

Mass transport in slag

The kinetics of boron removal from liquid silicon during slag refining have been investigated in this research by means of several small-scale experimental studies at temperatures of 14500C to 16500C. Slag and silicon, in batch weights of 7 g, were heated together in an Alumina crucible placed in a resistance-heated tube furnace. The slags were produced from powdered SiO2, CaO, and Al2O3. Experiments were carried out at slag-to-silicon ratios of 1.5, 2 and 2.5, where the silicon initially contained approximately 350 ppm boron. From the results of kinetics of boron removal from molten silicon through slag (CaO-SiO2, CaO-SiO210%Al2O3 & CaO-SiO2-15%Al2O3) was investigated.

Figure 1: Changes in boron contents in silicon during the reaction with CaO-SiO2 slag Figure 1 shows the change in boron contents during the experiments for CaO-SiO2 slags. The experimental results show that the rate of boron removal was very high at the beginning of the reaction; this rate dropped to zero after 120 min of the reaction, depending on the ratio of the amount of silicon to slag. In this study, the rates of boron removal were found to increase with a decrease in the initial weights of the silicon for both types of slags. The data shown in Figure 1 can be used to analyze the reaction kinetics by considering several kinetics models. The data plotted using models assuming slag and silicon mass transfer controlled. The results in Figure 2 show that all the data can be fitted into a straight line for silicon phase mass High Temperature Processing Symposium 2013 Swinburne University of Technology

27

transfer controlled. Therefore, it can be argued that the overall kinetics was controlled by mass transfer of B in silicon phase. Currently, further investigations are being carried out to confirm and evaluate the detailed mechanism.

Figure 2: Integrated rate plots obtained using the kinetic equation of (a) silicon mass-transfer control and (b) slag mass-transfer control for CaO-SiO2 slag

References 1. Suzuki, K., T. Sugiyama, et al. (1990). "Thermodynamics for removal of boron from metallurgical silicon by flux treatment." Journal of the Japan Institute of Metals 54(2): 168-172. 2. Johnston, M. D. and M. Barati (2010). "Distribution of impurity elements in slag-silicon equilibria for oxidative refining of metallurgical silicon for solar cell applications." Solar Energy Materials and Solar Cells 94(12): 2085-2090. 3. Teixeira, L. and K. Morita (2009). "Removal of Boron from Molten Silicon Using CaO– SiO2 Based Slags." ISIJ international 49(6): 783-787. 4. Krystad, E., K. Tang, et al. (2012). "The Kinetics of Boron Transfer in Slag Refining of Silicon." JOM Journal of the Minerals, Metals and Materials Society: 1-5. 5. Nishimoto, H. and K. Morita (2011). "The Rate of Boron Elimination from Molten Silicon by Slag and Cl2 gas Treatment." Supplemental Proceedings: Materials Processing and Energy Materials, Volume 1: 701-708.


28

PRESENTATION - 8 Minor Element Distributions during Copper Flash Converting Douglas R. Swinbourne RMIT University, Latrobe Street, Melbourne, VIC, 3122, Australia

Keywords: Modelling, Copper, Flash converting, Minor elements, Distributions. Peirce-Smith converting of copper matte is simple and reliable, but has many welldocumented disadvantages[1]. They have largely been overcome by the advent of continuous converting technologies such as the Mitsubishi bath converting process and the Kennecott flash converting process[2]. Both processes use lime (CaO) rather than silica as a flux and so produce calcium ferrite slag. Flash converting was pioneered at the KUCC smelter in Utah, USA, and has been described by Newman et al.[3] The control of minor elements is an important issue for all copper smelters. Kaur et al.[4] reported on minor element deportments (lead, arsenic, bismuth, cadmium and molybdenum ) at KUCC. HSC Chemistry for Windows v.7.1 was used to predict the distributions of these minor elements and the results were compared to published data. The first requirement is a reliable set of operating data. The most complete data set is for 2001 operations[2] The masses of dusts consumed/produced are given but not their compositions. The copper mass balance is near to closing but the iron balance is not. Kaur et al.[4] provided the most recent set of data but contains no information on dust nor the quantity of lime (CaO) flux. Again the copper and iron balances do not close. Neither data set gave minor element concentrations in matte so the data of Asteljoki and Kyto[5] was assumed, although the actual amounts are not essential when calculating fractional distributions. The input masses based on the operating data sets and the assumed activity coefficients, all taken from the literature, are given in the table below. No data for CdO was found so it was assumed to be the as that of ZnO. Where activity coefficients are not given, they are unity. Table 1. Species, the masses [ ] and activity coefficients ( ) used in the HSC model. gas matte slag copper solid O2(g) CuS0.5 [855 kg] CaO (1) [23 kg] Cu (1) FeO1.33 N2(g) FeS [122 kg] FeO1.33 (1.7) CuS0.5 (26) S2(g) PbS [14 kg] FeO (35) CuO0.5 (20) SO2(g) AsS1.5 [6 kg] CuO0.5 (3) Fe (10) SO3(g) BiS1.5 [2 kg] PbO (2) Pb (5) Pb(g) CdS [1 kg] AsO1.5 (0.2) As (0.005) PbO(g) BiO1.5 (0.8) Bi (2.4) PbS(g) CdO (2) Cd (0.73) AsO(g) Bi(g) Cd(g)


29

The mass of oxygen required to yield blister copper containing 0.2 wt% S was determined as 258kg/tonne matte, then all other process parameters were calculated at that level of oxygen addition. The copper content of the slag as a function of the sulphur content of the blister copper is shown below and the operating point for the KUCC converter is very close to the calculated equilibrium value. The figure shows why aiming for low sulphur contents in blister copper, which would lessen the load on the anode refining furnaces, carries a significant penalty. The distribution of lead by species is given to the left. More lead would report to the slag and less to the blister copper if the oxygen addition was increased past the target level of 258 kg, but the change in the proportion reporting to the waste gas would not change much. It can be seen that PbO(g) and Pb(g) are present in similar proportions at the target oxygen addition. The distribution of arsenic by species is given below. It matches the industrial data well, although the amount of arsenic being volatilised is underestimated. This could be because in practice the flash furnace is an open system i.e. gas is passing through the furnace and this would increase the amount of volatile species leaving the furnace. The very small amount of volatile arsenic as AsO(g) is striking because arsenic is widely regarded as a volatile element. The reasons for this are that AsO1.5 has a low activity coefficient in calcium ferrite slag due to the basic nature of this slag, and because arsenic has a very low activity coefficient in copper. About one half of all arsenic entering the converter reports to the blister copper. Poor arsenic elimination is a feature of continuous converting processes, as opposed to the Peirce-Smith converter where significant arsenic elimination occurs during the first stage slagging blow before any copper forms.


30

The fractional distributions of minor elements between offtake gas, slag and blister copper at an oxygen addition of 258 kg are given in the Table below, together with the data from Kaur et al.[4]. The agreement is generally excellent.

Table 2. Comparison of fractional distributions with published data for the flash converter Element

Phase

Mass fraction (%) Published This work 33 33 57 57 10 10

lead

blister slag gas

arsenic

blister slag gas

52 42 6

60 37 3

bismuth

blister slag gas

73 14 13

68 25 7

cadmium

blister slag gas

8 92

4 9 87

The success of this thermodynamic model indicates that the Kennecott flash converter can be regarded as approximating an equilibrium reactor and that minor element distributions appear to be controlled mostly by thermodynamic factors.

References 1. K.J. Richards, D.G. George & L.K. Bailey: Advances in Sulfide Smelting, Proc. Fall Meeting of TMS-AIME, San Francisco, Nov. 6-9, 1983, 489-498. 2. W.G. Davenport, M. King, M. Schlesinger & A.K. Biswas: “Extractive Metallurgy of Copper”, Pergamon, Great Britain, 4th edition, 2002. 3. C.J. Newman, D.N. Collins & A.J. Weddick: Copper 99-Cobre 99 International Conference, Phoenix, Arizona, 1999, vol. 5, The Met. Soc. of CIM, 29-45. 4. R. Kaur, C. Nexhip, M. Wilson & D. George-Kennedy: Proceedings of Copper 2010, Hamburg, GDMB, 2010, vol.6, 2415-2432. 5. J.A. Asteljoki & S.M.I Kytö: TMS Paper A86-57, The Minerals, Metals & Materials Society, Warrendale, PA, 1986.


31

KEYNOTE: PRESENTATION - 9 Integrated Experimental and Modelling Research Methodology for Phase Equilibria, Thermodynamics and Viscosities of Metallurgical Slags Evgueni Jak PYROSEARCH, The University of Queensland, Brisbane, Queensland, 4072, Australia. [email protected] Keywords: metallurgical slags, phase equilibria, thermodynamic modelling, liquidus. Coupled experimental and modelling studies are combined into an integrated research program on phase equilibria, thermodynamics and viscosities of the metallurgical slag systems. Key issues derived from experiences in continuing development and application of both experimental and thermodynamic modelling research are outlined. Particular emphasis is given to the details of the research methodologies, analysis of reasons for uncertainties and the ways to continuously improve the accuracy of both studies. The ways how the advanced research tools can be implemented into industrial operations are presented. Experimental part of the phase equilibria study involves high temperature equilibration in controlled gas atmospheres, rapid quenching and direct measurement of equilibrium phases with electron probe X-ray microanalysis (EPMA). Thermodynamic modelling undertaken using computer package FactSage with the quasi-chemical model for the liquid slag phase is closely integrated with the parallel experimental research. Experiments are planned to provide specific data for thermodynamic model development as well as for pseudo-ternary liquidus diagrams which can be used directly by process operators. Thermodynamic assessments are used to identify priorities for experiments. Experimental and modelling studies are combined into an integrated research program contributing to and enhancing outcomes of each other and of the overall program. The continuous development of experimental methodologies has brought significant advances. Importantly, these novel approaches enable measurements to be made in systems that could not previously be characterised, for example, due to uncontrollable reactions with container materials or changes in bulk composition due to vapour phase reactions. The approach, however, requires particular attention to ensure accurate information is obtained. An ongoing dedicated program of improving accuracy of all possible elements of the research revealed a number of possible sources of uncertainties and the ways developed to mitigate those shortcomings are systematically summarised in this paper. The thermodynamic modelling has progressed significantly, and achieved a level of prediction of phase equilibria and thermodynamics of complex multi-component multi-phase systems with improved accuracy. The adequate description of the systems however requires a combination of various types of data and still demands continuous further development. The outcomes of both experimental and modelling studies are applied to assist in improvements of the industrial metallurgical operations. High certainty of the predictions of the behaviour of complex industrial processes provides a strong basis for optimisation of operations. The stage of implementation of the outcomes of the laboratory experimental and theoretical modelling, however, is frequently overlooked, but requires high level of research expertise to establish the actual conditions in the real industrial process and relate them to the advanced laboratory and theoretical research tools. High Temperature Processing Symposium 2013 Swinburne University of Technology

32

PRESENTATION - 10 Aluminium Production Route through Carbosulfidation of Alumina utilising H2S Nazmul Huda1, M. Akbar Rhamdhani1, G.A. Brooks1, B. J. Monaghan2, L. Prentice3 1

HTP Research Group, Swinburne University of Technology, VIC 3122, Australia 2 PYRO Research Group, University of Wollongong, NSW 2522, Australia 3 CSIRO Process Science and Engineering, VIC 3169, Australia

Keywords: Aluminum, carbosulfidation, H2S Indirect carbothermal reduction of alumina for the production of aluminum utilizes different reducing agents to convert alumina into intermediate aluminum compounds. In the present study, the carbosulfidation route for aluminum production utilizing H2S(g) as the reductant and sulfur source has been investigated, in particular the formation of Al2S3 in the first step of the process. The results of the thermodynamic analysis predicted that conversion of Al2O3(s) to Al2S3(l) significantly increases above 1400°C at 1 atmosphere pressure. Experimental investigations were carried out at temperatures of 1100 to 1500°C using dilute H2S(g) gas in argon. The reaction products were analyzed using scanning electron microscopy (SEM), energy dispersive X-ray spectroscopy (EDS), X-ray diffraction (XRD), inductively-coupled plasma absorption emission spectroscopy (ICP-AES) and chemical filtration. The X-ray diffraction results confirmed the presence of Al2S3(s). Percentage of conversion from Al2O3 to Al2S3 was found to be over 80% at 1500°C. Equilibrium Calculations of Al2O3-C-H2S Reaction Systems The equilibrium calculations were carried out using FactSage 6.1 thermodynamic package. The equilibrium calculations for Al2O3-C-H2S system were carried at temperatures 1000°C to 2000°C at different pressures. For all equilibrium calculations, 3 moles of C and 3 moles of H2S were considered for 1 mole of Al2O3. Figure 1 shows equilibrium calculation of Al2O3+3C+3H2S for temperature range of 1000 to 2000°C at 1 atm pressure. Figure 1(b) show that significant amounts of gases are produced with majority of H 2(g) and CO(g) at higher temperatures. Al2S3 is predicted to be the main intermediate aluminum compound when H2S is reacted with Al2O3 and C at 1000 to 2000°C at 1 atmospheric pressure. Formation of Al2S3 is predicted to be very low at 1100 to 1300°C at 1 atm pressure (0.1012 mol Al2S3/ mol Al2O3) and predicted to increase with increasing temperature to 1800°C. Formation of CO is predicted to be lower at 1100°C (0.035 mol/mol Al2O3) and significantly increases with increasing temperature (2.6 mol/mol Al2O3 at 1800°C). Along with CO and other gases significant amount of H2(g) gas is also predicted to form at 1100°C (1.37 mol/ mol Al2O3). This content of H2(g) was predicted to increase to 2.62 mol/mol Al2O3 when temperature is at 1800°C.


33

a) Predicted condensed phases b) Predicted gaseous phases Figure 1: Predicted equilibrium phases in the Al2O3+3C+3H2S system at T = 1000°C to 2000°C, at 1 atm pressure: a) condensed phases, b) gaseous phases Experimental results Experimental investigation on carbosulfidation of Al2O3(s) by using C(s) and dilute H2S(g) (5% H2S – 95% Ar) at different temperatures (1100 to 1600 °C) and reaction duration were carried out using a horizontal tube resistance-furnace (Nabertherm RHTV 200-600). A schematic diagram of the experimental setup is shown in Figure 2.

Figure 2: A schematic diagram of the experimental set up using a horizontal tube furnace Figure 3 shows the comparison of XRD pattern of the samples after experiments at 1400°C for three different times (3, 6 and 9 hours). Al2O3 and Al2S3 peaks are marked by “1” and “2”, respectively. As shown in Figure 3, significant aluminum sulfide (Al2S3) was detected after 6 and 9 hours of reaction. This is indicated by the higher and sharper Al 2S3 peaks at 6 and 9 hours compared to those from at 3 hours. Al2O3 peaks are still present, indicated that some Al2O3 remains and unreacted in the samples. However, it can also be seen clearly that there is a gradual decrease of the intensity with increasing reaction time.


34

Figure 3: X-ray diffraction pattern of the samples after 3, 6, and 9 hours experiments at 1400 °C. (1 = corundum (Al2O3), 2 = aluminum sulfide (Al2S3) and 3 = Graphite (C)) The percentage of conversion from Al2O3 to Al2S3 was determined by chemical dissolution and filtration. As pure Al2S3 completely dissolves in hydrochloric acid (HCl), a portion of the experimental samples were dissolved in HCl (36% w/w aqueous solution) and the solution was then filtered out. The amount of mass that dissolves in HCl represents the formed Al 2S3 while the residues are the unreacted Al2O3 and C. From the filtration results, the percent of conversion ( ) of Al2O3 to Al2S3 was calculated using following equation:

The details of calculated conversion from selected experiments are shown in Table I. The highest conversion was found for experiment at 1500°C and 9 hours duration. The conversion showed an increasing trend with respect to time and temperature. Table I: The conversion of Al2O3 to Al2S3 from selected samples at 1400°C and 1500°C Temperature Duration (°C) (hours) 1400 1500

6 9 6 9

Weight of Sample (g) 0.2012 0.2051 0.2186 0.2060

% of Conversion ( ) 75.4 77 78.9 81.6

In summary, the results, from XRD, SEM, EDS, ICP and conversion calculation, indicate that it is possible to form high amount of Al2S3 from Al2O3 using C and H2S gas in the range of conditions studied. The results also suggest that the conversion to Al2S3 increases with increasing temperature and duration of experiments.


35

PRESENTATION - 11 Investigation of Freeze-linings in Copper-Containing Slag Systems Ata Fallah Mehrjardi, Peter C. Hayes, Evgueni Jak PYROSEARCH, The University of Queensland, Brisbane, Australia

Keywords: Freeze-lining, slag, copper production, deposit Ways of increasing productivity have been always a challenging issue for pyrometallurgical processes. Increasing the throughput of the reactor through higher temperature and vigorous agitation in bath are possible options for enhancing the kinetics of reactions in the multiphase system processes. On the other hand, these measures also lead to rapid degradation of the refractory and premature shutdown of the reactor for relining, imposing additional costs on processes in the form of planned and unplanned maintenance. An alternative solution to this problem is the formation of a slag freeze lining rather than direct contact of refractory layers with the hot bath. Slag freeze-linings are increasingly used in industrial pyrometallurgical processes to ensure furnace integrity is maintained in aggressive high temperature environments 1-4). Most previous studies of freeze-linings have analysed the formation of slag deposits based solely on heat transfer models 5, 6). The focus of the present research is to determine the impact of slag chemistry and local process conditions on the microstructures, thickness, stability and heat transfer characteristics of the frozen deposit. To gain a full understanding of freeze lining formation and the effect of experimental variables on the stability, thickness and heat transfer characteristics of the freeze lining two particular approaches have been adopted. First, a 1-D heat transfer model in cylindrical coordinates was developed to approximately evaluate the deposit thickness and interpolate temperature distribution in the freeze lining as a function of key process variables such as coolant flow rate, thermal conductivity, bath convection and superheat. Second, experimental studies of the freeze lining formation and kinetics at steady state condition were undertaken including cold finger and supporting experiments. The cold finger was immersed into a synthetic slag bath heated by an induction furnace. The temperature profile across the deposit and bath was measured. A Cu-Fe-Si-Al-O slag was selected for study; the liquidus temperature for the slag in equilibrium with metallic copper has been experimentally determined to be approximately 1140◦C and the primary phase under these conditions has been found to be delafossite. The phase assemblages and microstructures of the deposits formed in the cold finger experiments differ significantly from those expected from equilibrium considerations. The freeze-lining deposits have been found in general to consist of several different layers. Starting from the cold wall these layers consist of glass; glass with microcrystalline precipitates; multiphase sub-liquidus material containing delafossite and cuprite crystal phase assemblages and highsilica metastable liquid that was separated from the bulk liquid (closed crystalline layers); phase assemblages containing delafossite and cuprite crystals and a high-silica liquid phase that is connected to the bulk liquid (open crystalline layers), and the outer layer containing a complex mixture of liquid and solid phases.


36

It has been previously widely assumed that the interface between the stationary frozen layer and the molten bath at steady-state consists of the primary phase in contact with the bulk liquid at the liquidus temperature, Tliquidus. It has been shown in the present laboratory-based studies through the use a of cold finger technique that, at steady-state and in selected ranges of process conditions and bath compositions, the phase assemblage present at the deposit/liquid interface is not that of the primary phase alone. The microstructural observations clearly demonstrate that the temperature of the deposit/liquid bath interface, T f, can be lower than the liquidus temperature of the bulk liquid, Tliquidus. These observations point to a significant change in proposed mechanism and behaviour of the systems. Acknowledgements The authors would like to thank the Australian Research Council Linkage program, Rio Tinto Kennecott Utah Copper Corp., Xstrata Technology, Xstrata Copper, BHP Billiton Olympic Dam Operation and Outotec Finland Oy for their financial support. Appreciations are extended to all PYROSEARCH and CMM staff at the University of Queensland. References 1. M. Campforts, B. Blanpain, and P. Wollants, "The importance of slag engineering in freeze-fining applications". Metall. Mater. Trans. B., 2009. 40B: p. 643-655. 2. M. Campforts, et al., "Freeze-lining formation of a synthetic lead slag: Part I.microstructure formation". Metall. Mater. Trans. B., 2009. 40B: p. 619-631. 3. M. Campforts, et al., "Freeze-lining formation of a synthetic lead slag: Part II. thermal history". Metall. Mater. Trans. B., 2009. 40B: p. 632-642. 4. M. Campforts, et al., "On the microstructure of a freeze lining of an industrial nonferrous slag". Metall. Mater. Trans. B., 2007. 38B: p. 841-851. 5. K. Verscheure, et al., "Continuous fuming of zinc-bearing residues: Part II. The submerged-plasma zinc-fuming process". Metall. Mater. Trans. B., 2007. 38B: p. 2133. 6. F. Guevara and G. Irons, "Simulation of Slag Freeze Layer Formation: Part II: Numerical Model ". Metall. Mater. Trans. B., 2011. 42: p. 664-676.


37

PRESENTATION - 12 CIRCOSMELT: Outotec’s Alternative Ironmaking Process Ross Baldock Outotec Pty Ltd Melbourne Australia

Current high iron ore prices have resulted in increased interest in alternative iron sources, which cannot be processed through conventional routes such as blast furnace or shaft furnace respectively rotary kiln based direct reduction. Outotec has been developing circulating fluidized bed (CFB) processes for more than 50 years. One such CFB application is the coal based direct reduction process - Circofer®. Since Ausmelt joined the Outotec group in 2010, further development of the Circosmelt process, combining the Circofer prereduction with the coal based AusIron® smelting reduction has taken place. The Circosmelt process uses fines, thus avoiding agglomeration, such as sintering or pelletizing, and can utilize a wide range of bituminous and sub bituminous coals. Test work in Outotec’s pilot reduction and smelting plants with different raw materials and a variety of coals has been performed. This presentation will give an overview of the Circosmelt process principles. Furthermore results achieved during pilot plant campaigns with different raw materials will be discussed.


38

KEYNOTE: PRESENTATION - 13 Insights into the Formation of Iron Ore Sinter Bonding Phases Mark I. Pownceby1 and Nathan A.S. Webster1,2 1 CSIRO Process Science and Engineering, Box 312, Clayton South, VIC, 3169, Australia 2 Australian Nuclear Science and Technology Organisation, Locked Bag 2001, Kirrawee DC, NSW, 2232, Australia Keywords: Iron ore sintering, ‘SFCA’, phase equilibria, in situ X-ray diffraction. During the iron ore sintering process, iron ore fines (