Professional Documents
Culture Documents
Volume 7
Plenary
Mineral Processing
Recycling
Posters
Authors Index
Keywords Index
Conference organized by GDMB, IIMCh, MetSoc, MMIJ, SME and TMS
Editor
GDMB
Paul-Ernst-Strae 10, D-38678 Clausthal-Zellerfeld
Internet: www.GDMB.de
Volume 1 ISBN 978-3-940276-25-4
Volume 2 ISBN 978-3-940276-26-1
Volume 3 ISBN 978-3-940276-27-8
Volume 4 ISBN 978-3-940276-28-5
Volume 5 ISBN 978-3-940276-29-2
Volume 6 ISBN 978-3-940276-30-8
Volume 7 ISBN 978-3-940276-31-5
Set (Volume 1+2+3+4+5+6+7) ISBN 978-3-940276-32-2
All rights reserved. No part of this publication may be reproduced or electronically processed,
copied or distributed without the prior consent by the editor.
The content of the papers is the sole responsibility of the authors. All papers were peer reviewed by
the corresponding members of the technical groups of the organizing societies.
Editorial staff: Dipl.-Ing. Jens Harre
Production and marketing: GDMB Informationsgesellschaft mbH
Printed by: Papierflieger
GDMB Clausthal-Zellerfeld 2010
Bibliographische Information Der Deutschen Bibliothek
Die Deutsche Bibliothek verzeichnet diese Publikation in der Deutschen Nationalbibliographie;
detaillierte bibliographische Daten sind im Internet ber http://dnb.ddb.de abrufbar.
Bibliographic information published by Die Deutsche Bibliothek
Die Deutsche Bibliothek lists this publication in the Deutsche Nationalbibliographie; detailed
bibliografic data is available in the internet at http://dnb.ddb.de.
Proceedings
Volume 7
Plenary
Mineral Processing
Recycling
Posters
Authors Index
Keywords Index
The Copper 2010-Proceedings are friendly supported by
Proceedings of Copper 2010 IV
The Organizing Society:
GDMB Society for Mining, Metallurgy, Resource and
Environmental Technology
The GDMB is a non-profit organization. Its activities focus on combining science with practical
experience in the fields of mining, engineering, tunnelling, mineral processing, extraction, recycling
and refining of metals, as well as on the manufacturing of semi and finished products. There is an
increasing emphasis on associated environmental issues. The GDMB is internationally active with a
European basis and covers a wide variety of topics from applied geology via processing to recy-
cling. These include many important areas of chemistry, especially the complex metallurgical
chemistry and, last not least, also analytical chemistry. As a consequence of their increasing impor-
tance, aspects of industrial minerals are addressed in addition to the traditional fields of metals and
alloys. In order to remain a vibrant and attractive professional society, the GDMB draws on the ex-
perience and interests of its worldwide members.
The Co-Organizing Societies and their Representatives
Institutos de Ingenieros de Minas de Chile (IIMCh)
Enrique Miranda S., Gerente IIMCh, Chile
The Metallurgical Society of the Canadian Institute of Mining, Metallurgy,
and Petroleum (MetSoc)
Jol Kapusta, Ph.D., Air Liquide Canada Inc., Canada
Dr. Phillip Mackey, Xstrata Process Support Centre, Canada
The Mining and Materials Processing Institute of Japan (MMIJ)
Dr. Takahiko Okura, The University of Tokyo, Institute of Industrial Science, Japan
Yasuo Tamura, Japan Mining Industry Association, Japan
Society for Mining, Metallurgy, and Exploration (SME)
Dr. John L. Uhrie, Newmont Mining Corporation, USA
The Minerals, Metals & Materials Society (TMS)
Dr.-Ing. Andreas Siegmund, LanMetCon, USA
Proceedings of Copper 2010 V
Conference Chairman
Dipl.-Ing. Michael Kopke Aurubis AG, Germany
Technical Programme Chair
Dipl.-Ing. Jo Rogiers Aurubis AG, Belgium
Conference Chair Assistance
Dipl.-Ing. Jrgen Zuchowski GDMB Gesellschaft fr Bergbau, Metallurgie,
Rohstoff- und Umwelttechnik e. V.
Session Chairs
1 Plenary lessons of general interest Dipl.-Ing. Norbert L. Piret,
for all conference members Piret & Stolberg Partners, Germany
2 Economics Dr. Patricio Barrios, Aurubis AG, Germany
3 Downstream Fabrication, Application Dr.-Ing. Hans Achim Kuhn,
and New Products Wieland Werke AG, Germany
4 Mineral Processing Assoc. Prof. Sadan Kelebek,
Queens University Canada
5 Pyrometallurgy David B. George, Rio Tinto, USA
6 Hydrometallurgy Dr.-Ing. Andreas Siegmund, LanMetCon, USA
7 Electrowinning and -refining Dr.-Ing. Heinrich Traulsen, Germany
Mike Murphy, Xstrata Technology, Australia
Mike Hourn, Xstrata Technology, Australia
8 Process Control, Automatization Prof. Dr. Markus Andreas Reuter,
and Optimization Outotec Ausmelt, Australia
9 Recycling Dipl.-Ing. Jrg Wallner, Austria
10 Sustainable Development / Health, Dipl.-Ing. Miguel Palacios
Safety and Environmental Control Atlantic Copper S.A., Spain
Proceedings of Copper 2010 VI
Technical Group Chairs
GDMB (region: Europe, Russia, near Orient) Dipl.-Ing. Jo Rogiers, Aurubis AG, Belgium
IIMCh (region: South America) Sergio Demetrio, IIMCh, Chile
MetSoc (region: Canada, Australia, Africa) Ass. Prof. Edouard Asselin,
University of British Columbia, Canada
MMIJ (region: Japan, China, South East Asia) Dr. Takahiko Okura,
The University of Tokyo, Japan
SME / TMS (region: USA, Mexico) Dr. Andreas Siegmund, LanMetCon, USA
Technical Group Members
Full information you will find in the internet at: www.Cu2010.GDMB.de
Short Course Organizing Committee
Dipl.-Ing. Michael Kopke (Chair), Aurubis AG, Germany
Dipl.-Ing. Miguel Palacios, Atlantic Copper S.A., Spain
Prof. Dr. mont. Peter Paschen, Austria
Dipl.-Ing. Norbert L. Piret, Piret & Stolberg Partners, Germany
Organizing Committee
Dipl.-Ing. Jrgen Zuchowski GDMB Gesellschaft fr Bergbau, Metallurgie, Rohstoff-
und Umwelttechnik e. V.
(Copper2010 Organizing Committee Chairman)
Mareike Hahn GDMB Gesellschaft fr Bergbau, Metallurgie, Rohstoff-
und Umwelttechnik e. V.
Thomas Marbach GDMB Gesellschaft fr Bergbau, Metallurgie, Rohstoff-
und Umwelttechnik e. V.
Dipl.-Ing. Jens Harre GDMB Informationsgesellschaft mbH
Mareike Mller GDMB Informationsgesellschaft mbH
Ulrich Waschki GDMB Informationsgesellschaft mbH
Proceedings of Copper 2010 VII
Preface
Copper Indicator of the progress of civilization
This is the motto for the 7
th
international copper conference, the most important copper seminar in the
world, which has been organized by the GDMB, the German based Society for Mining, Metallurgy,
Resource and Environmental Technology, together with IIMCh from Chile, MetSoc from Canada,
TMS, SME from USA and MMIJ from Japan.
The copper conferences bring together the highest level of science and technology: universities, metal
producers, manufacturing companies, suppliers and finally the people who work with copper: scien-
tists, technicians, engineers, traders and many more.
An extensive programme has been arranged for this conference and an abundance of contributions from
all over the world dealing with the different aspects of copper making and its use are registered already,
for which we gratefully thank the authors. Apart from plenary addresses, separate sessions will be held
for economics, mineral processing, pyrometallurgy, hydrometallurgy, electrowinning and -refining,
downstream fabrication and application, process control and automation, recycling and sustainable de-
velopment, environmental control, health and safety. Copper, one of the oldest metals used by mankind,
is still today one of the most important industrial metals and indispensable for modern life. It is the indi-
cator of industrialization and progress in every country. It is used everywhere, where electricity flows
and thus is still valued so highly today. The increased economic potential of newly industrialized coun-
tries, above all East Asia and China, has increased the significance of the red metal once again. More
recent technologies in production, processing and application often provide new answers to old ques-
tions. From the middle of the last century there was another innovation surge resulting in totally new
technologies, a trend still going on. This has made each copper conference into an exciting adventure. It
is positive and reassuring that particularly the high industrialized countries have become the vanguard,
not just in technical innovation, but also protection of environment and nature and preserving resources.
They repeatedly prove that ecology and economy may go hand in hand.
Many sponsors have contributed to the conferences success, for which I would like to express my
sincere thanks!
Hamburg is expecting its guests! Hamburg, the old Hanseatic city with a 1200 year long tradition, one
of the biggest and most beautiful cities in Germany, combines a wonderful mixture of industry, com-
merce, nature and culture. Not only the Copper 2010 in the Congress Centre awaits you, but rich
offerings of sightseeing and shopping in a cosmopolitan city, one of the largest harbours in Europe,
the Alster lake in the city centre, green parks and plenty of cultural events, some of which we hope to
show you in the companions programme.
We are delighted that you will join us and look forward to a highly interesting conference!
Michael Kopke
Chairman Copper 2010
Proceedings of Copper 2010 VIII
Structure of the Proceedings
Volume 1:
Downstream Fabrication, Application and New Products
Sustainable Development / Health, Safety and Environmental Control
Volume 2:
Pyrometallurgy I
Volume 3:
Pyrometallurgy II
Volume 4:
Electrowinning and -refining
Volume 5:
Hydrometallurgy
Volume 6:
Economics
Process Control, Automatization and Optimization
Volume 7:
Plenary lessons of general interest for all conference members
Mineral Processing
Recycling
Posters
Authors Index
Keywords Index
Proceedings of Copper 2010 IX
Plenary Lectures (Abstracts)
Some of the full papers will be published in World of Metallurgy ERZMETALL.
Is the Copper Industry Fit for the Future?
Dr.-Ing. Bernd Drouven, CEO, Aurubis AG, Hamburg, Germany
The copper world has already changed a great deal during recent years. But what still lies ahead of
us? What are the changes in conditions that we have to cope with and how will our solutions look?
The rise in the demand for raw materials is unrelenting. At the same time, both the primary and sec-
ondary feed materials have a specific complexity, the customers needs are becoming increasingly
differentiated and their orders are placed at increasingly short notice. Production times for metals
have to be faster and inventories minimised.
The volatility of the metal prices has increased significantly in recent years and we will probably
have to live with that in the future as well. The LME functioned well in the crisis, but the copper
price is being influenced more and more by funds.
How can the value added chain change to adapt to this? Which consequences will that have for
process technology, production planning and logistics? How will the consolidation of our industry
continue?
This keynote will go into the various effects and challenges in particular from the perspective of a
European custom smelter and fabricator and present possible approaches for solutions.
Sustainable Growth Strategy for Japanese Copper Business
Toshinori Kato, Managing Director, Mitsubishi Materials Corporation, Tokyo, Japan
Business environment, for any industry, has changed dramatically over the last decade. Copper in-
dustry is no exception and those involved are experiencing an unprecedented period of upheaval.
The landscape of the market has completely altered, with large-scale M&As taking place among
miners - creating an oligopoly situation - and a rapid expansion of smelting capacity within de-
veloping countries, driven by strong economic growth. Traditional copper smelters and fabricators
have been facing challenges, but a long-term sustainability of Japanese copper business is achiev-
able.
Firstly, an immense amount of effort has been made over the years to develop the clean and one of
the most environmentally-friendly processes in the industry. Mitsubishi Materials (MMC) particu-
larly plays a great role in establishing Japanese smelters reputation as the most energy-efficient
operations in the world. It is our strong commitment to be a leader in this expertise by sharing our
technologies with the industry.
Secondly, infrastructures for copper smelting have been utilized in developing the recycling busi-
ness. Exemplified by the operations at Onahama smelter, which has the worlds largest furnace for
Proceedings of Copper 2010 X
treatment of shredder residue, the industry has worked closely with other sectors and municipalities.
Building an effective structure to make the best use of our facilities is a key for success in this field.
It is fair to say that Japanese copper smelters, including MMC, are now regarded as an indispensa-
ble part of national environmental policy.
Finally, progression of integration in the copper fabrication sector has strengthened the industrys
capability of providing a variety of high-value-added products to end-users. Tight relationships with
the consumers have been beneficial in developing new use of copper. Evidence of a promising fu-
ture can be seen in increasing use of copper in the growing sectors such as hybrid automobiles and
renewable energy.
Copper Sulphide Smelting: Past Achievements and Current Challenges
Dr. Carlos M. Daz, Adjunct Professor, University of Toronto, Private Consultant, Toronto, Canada
In the last three decades, increased oxygen consumption in copper sulphide smelting and converting
and the implementation of computerized process control, among other factors, have led to higher
process intensity and smelter concentrate processing capacity, decreased smelting energy consump-
tion and improved SO
2
capture from process gas streams. An annual primary smelting furnace
throughput of one million tonnes of concentrate is the new industry standard. Top submerged lance
smelting has become an important processing route in recent years. Two new continuous converting
routes were commercialized in the 1990s. However, due to substantially improved converting prac-
tice and larger converters, Peirce-Smith converting still maintains its position as the dominant tech-
nology.
A major realignment of world copper smelting has taken place in the last 30 years. Spurred by rapid
economic expansion and the resulting huge increase in demand for basic materials, a number of
modern, large capacity copper smelters have been built in China, India and other Asian countries.
Only moderate growth in smelting and electrorefining capacity has taken place elsewhere. More-
over, ER cathode output in the USA has substantially decreased.
In this paper, the author examines recent technological advances and industry changes and high-
lights issues, such as energy consumption and the corresponding greenhouse gas emissions that will
become the focus of future discussion.
Energy as a Key Factor of Sustainability
Javier Targhetta, Vice President, Freeport McMoRan, Phoenix, Arizona, USA
Energy must be seen as the prime mover for development and is therefore vital for economic equi-
librium and social welfare. Energy will continue to play a key role in the coming decades being not
only an environmental challenge but also a fundamental issue in terms of the progress of humanity.
Nevertheless, harmonizing several aspects related to energy management will be of essence for the
near future and will make energy one of the greatest challenges of the 21
st
century. It is becoming
increasingly important to maintain the appropriate equilibrium between environmental issues
Proceedings of Copper 2010 XI
(global warming, the future of nuclear waste, the integration of renewable energies), coordinated
governments policies to ensure that global energy costs are kept competitive and companies finding
a balanced, rational and fair model of energy utilization based on the review of ethical business as-
pects and the establishment of social responsibility principles.
This paper develops the idea that the rules and procedures for future energy use must be based on a
reasonable equilibrium between technical aspects mainly associated to environmental issues, politi-
cal decisions made by governments and public administrations and social responsibility pro-
grammes that companies themselves must assume and implement.
The Supply and Copper Producer Response to a Growing Demand Scenario
Ricardo Alvarez, General Manager, Codelco El Teniente Division, Santiago, Chile
The prospects for medium and long-term consumption of copper are promising. A recovery pro-
jected for developed economies, starting 2010, joins the growing impact on demand generated from
the process of development and urbanization of emerging countries. The intensity of copper use will
also maintain the positive growth started a decade ago based on factors such as providing solutions
to combat global warming.
The objective of this presentation is to analyze how supply may react and adjust to envisaged de-
mand scenarios, bearing in mind some distinctive elements of the copper industry like:
- Historical supply reaction to demand and price levels
- Availability of resources and copper ore reserves, incorporating a dynamic analysis and the ef-
fects of geological exploration and possible technological changes.
- Pipeline of probable and possible projects, analysing the effects of projects in less developed and
riskier geographical locations
- Technological changes under development that may positively impact the amount of reserves
available and the competitiveness of projects.
- The growing role of scrap as a supply source for copper.
The analysis of these points will allow us to confirm the capability of supply response to growing
demand, ruling out revisited hypothesis of insufficient reserves
Implementation of Recent Global Copper Projects
Tim J. A. Smith, Vice President, Copper, SNC-Lavalin UK Limited, Croydon, Surrey, UK
Despite the recent global recession, a relatively strong copper price combined with continuing sup-
ply shortfalls continues to drive the implementation of a number of important new worldwide cop-
per projects.
Along with base metals projects in general, the scale and complexity of such projects has increased
such that multibillion dollar projects are increasingly common.
Proceedings of Copper 2010 XII
Together with expansion projects in the older traditional copper producing regions, new geographic
areas are still being opened up. These frequently require major infrastructural development, envi-
ronmental and global procurement capabilities as major components of such projects.
This plenary session address will examine and discuss the many project management challenges
and skills needed to deliver successful projects worldwide, as viewed from the perspective of one of
the worlds leading metallurgical plant engineers and constructors.
Proceedings of Copper 2010 XIII
Table of Contents Volume 7
Plenary
Copper Sulphide Smelting: Past Achievements and Current Challenges 2543
Carlos M. Daz
Sustainable Growth Strategy for Japanese Copper Business 2563
Toshinori Kato
Mineral Processing
Optimization of Copper Concentrate Bioleaching by 2575
Mixed Moderate Thermophile Bacteria
A. Ahmadi, M. Ranjbar, M. Schaffie, Z. Manafi
Industrial NSC Pressure Oxidation of Combined Copper and 2589
Molybdenum Concentrates
Corby G. Anderson, Todd S. Fayram, Larry G. Twidwell
HPGRs in Copper Ore Comminution A Technology Broke Barriers 2621
E. Burchardt, N. Patzelt, J. Knecht, R. Klymowsky
Evaluation of Copper Losses in the Slag Cleaning Circuits from 2637
Two Chilean Smelters
N. Cardona, L. Hernandez, E. Araneda, R. Parra, L. Bahamondes, R. Parada,
J. Vargas, M. Artigas
Leaching of Gangue in Technological Flotation Circuits of 2655
Polish Copper Ores
Tomasz Chmielewski, Andrzej Luszczkiewicz
Proceedings of Copper 2010 XIV
Pressure Leaching of Shale Middlings from Lubin Concentrator in 2673
Oxygenated Sulphuric Acid
Tomasz Chmielewski, Jerzy Wdka
Mine to Heap in Mantoverde Anglo American Division 2693
Manuel Daz, Cristian Salgado, Carlos Prez, Cristian Alvayai,
Leonardo Herrera, Gabriel Zrate
Predicting the Effects of Locked, Partially Locked, and 2703
Liberated Minerals in Copper Leaching
Michael L. Free, Abraham L. Jurovitzki
Predicting Leaching Solution Acid Consumption as a 2711
Function of pH in Copper Ore Leaching
Michael L. Free
Copper from Pyrite A Short History 2721
Fathi Habashi
Copper Crud Treatment, Concentration Dependent Pond Depth 2737
Adjustment for Decanter Centrifuges, DControl
Dipl.-Ing.Tore Hartmann, Dr. Ulrich Horbach, Jens Kramer
A Specific Electrode for On-line pH Measurement in 2747
SAG Cleaner Flotation Circuits
Christian Hecker C., Alejandro Gonzlez S., Alejandra Mejas J., Claudia Rodrguez F.
Redox Potential Control in Column Leaching of Chalcopyrite 2753
Naoki Hiroyoshi, Takenari Kuwazawa, Yuki Takehara, Masami Tsunekawa
Separation Characteristics of Chalcopyrite and Pyrite via 2765
Bench Scale Flotation Investigations
S. Kelebek, Z. El Jundi, S. Reeves, H. zdeniz
Proceedings of Copper 2010 XV
Bioleaching of Crude Chalcopyrite Ores by the Thermophilic 2781
Archaean Acidianus brierleyi in a Batch Reactor
Y. Konishi, N. Saitoh, M. Shuto, T. Ogi, K. Kawakita, T. Kamiya
Hybrid Flotation Newly Developed Flotation Technology for 2793
Increased Recovery Especially in the Finest Particle Fractions
W. Krieglstein, L. Grossmann
Perspectives of Copper Mining Industry Development in Poland 2807
Ph.D. Eng. Jan Kudeko, Ph.D. Jacek Pyra, Ph.D. Eng. Jerzy Sobociski
Control of Bubble Size in a Laboratory Flotation Column 2829
Miguel Maldonado, Dr. Andr Desbiens, Dr. ric Poulin, Dr. Ren del Villar,
Alberto Riquelme
Flow Process in the Aerator of the Flotation Machine 2845
Preliminary Simulations
Adam Maka, Adam Fic, Andrzej Sachajdak, Ireneusz Szczygie
Analysis of Fine Particles Behaviour in Flotation of Polish Copper Ores 2859
Ph.D. Eng. A. Potulska
HPGR versus SAG Milling Technology in Hard-Rock Mining 2873
Review and Analysis
Irshad Rana, Kris Chandrasekaran, Ken Wood
Selective Leaching of Arsenic from Copper Ores and 2883
Concentrates Containing Enargite in NaHS Media
William Tongamp, Yasushi Takasaki, Atsushi Shibayama
Copper Leaching from Molybdenite in Acidic FeCl
3
Solutions with FeCl
2
2897
Yan Zhang, Narangarav Tumen-Ulzii, Zhibao Li
Proceedings of Copper 2010 XVI
Recycling
Shifting Core Business Vision: From Copper to Polymetallics 2913
A Recycling Point of View
Juan Ignacio Barturen Zabala
Solubility of Scorodite Synthesized by Oxidation of Ferrous Ions 2923
Tetsuo Fujita, Etsuro Shibata, Takashi Nakamura
Biosolubilization of Copper from Waste Electric Cables 2935
Asst. Prof. S. Gaydardzhiev, D. Bastin, Dr. P. F. Bareel, Eng. F. Goffinet
Dowa Mining Scorodite Process
Optimization of Copper Concentrate Bioleaching
Proceedings of Copper 2010 2581
3.3 Influence of temperature
Figures 2-A and 3 show the effect of temperature on copper and iron recovery. As can be seen in
Figure 2-A, by increasing temperature from 44 to 50 C, copper recovery significantly increases in
a period of 14 days while by prolonging the time to 30 days, its influence is diminished. Further-
more, the effect of temperature on iron recovery has shown at different levels of temperature and
pH in both media (see Figure 3). It shows that increasing both temperature to 50 C and pH to 1.8 is
led to a remarkable decrease in copper recovery especially in 9K medium. On the other hand, at low
level of pH, rising temperature in 9K medium has a negative influence on iron extraction while it
hasn't influence on its value in Norris medium. Low iron extraction is attributed to the precipitation
of a part of dissolved iron as jarosite (Equation 7) on chalcopyrite and reduces copper recovery.
3
4 2 3 4 2 6
3 2 6 ( ) ( ) 8 Fe X HSO H O XFe SO OH H
+ + +
+ + + +
, (7)
where
4 3
, , X K Na NH and H O
+ + + +
= .
Moreover, by prolonging leaching time to 30 days, high ORP (high ferric concentration) and high
temperature are two suitable factors to form jarosite. This justification is confirmed by remarkable
decrease of iron recovery at higher temperature and more leaching time (30 days). The influence of
temperature on decreasing iron extraction is more significant at high level of pH (see Figure 3).
Figure 2-B shows that increasing temperature in 9K medium is led to decreasing of cells number. It
seems that essential nutrient elements co-precipitates with ferric irons (Equation 7) leading to an
unsuitable culture medium to growth and activity of bacteria. High concentration basalt salts
(9K medium) is suitable for bacteria at low level of temperature, but when temperature is risen to
50 C, iron-basalt salt precipitates is increased leading to the depletion of needed components in
medium.
Figure 2: A) Influence of temperature on copper recovery in leaching times 14 and 30 days,
B) Interaction of temperature and nutrient medium on cells number in the presence of
silver at leaching time 30 days
N
u
m
b
e
r
o
f
B
a
c
t
e
r
i
a
(
C
e
l
l
s
/
m
L
1
*
1
0
^
7
)
A
B
Ahmadi, Ranjbar, Schaffie, Manafi
Proceedings of Copper 2010 2582
Figure 3: Interaction of temperature and pH on iron recovery at A) Norris medium and
B) 9K medium in leaching time 14 days
3.4 Influence of pH
pH of a bioleaching system is one of the most important operating parameters that influences on
metal dissolution and the growth and activity of acidophilic microorganisms [22, 23].
Effect of initial pH on copper recovery can be seen in Figure 4-A. It shows that increasing pH has a
little positive effect on copper recovery in leaching time 14 days, while in the leaching time of 30
days, changing pH doesn't influence on it significantly. pH values in the bioleaching system have a
transient increase then decrease quickly. The transient increase is due to occurring acid consuming
reactions which decreases to desired value by addition 6M H
2
SO
4
.
There are two reasons for strong decrease in the pH: one of them is the activity of sulphure oxidiz-
ing bacteria to convert sulphure and sulphide species to sulphate that produces acid according to
Equation (8):
2 2 2 4
3 2
Sulphur oxidizing microorganisms
S H O O H SO + +
(8)
And the other reason is the precipitation of jarosites according to Equation (7) that is an acid pro-
ducing reaction. The probability of jarosite formation at low level of initial pH is remarkably dimin-
ished.
The excess low pH may be harmful for growth and activity of bacteria. The negative effect of very
low pH value on the microorganisms and its positive effect on chemical leaching (as a leaching
agent) may neutralize each other. Figure 4-B shows that low level of pH at 9K medium is led to
decreasing cells number. More increase of pH suits the condition for jarosite precipitation which is
led to chalcopyrite passivation.
Optimization of Copper Concentrate Bioleaching
Proceedings of Copper 2010 2583
Figure 4: A) Influence of pH on copper recovery in leaching times 14 and 30 days,
B) Interaction of pH and temperature on cells number in leaching time 30 days
3.5 Influence of nutrient medium type
Microorganisms require nutrients for their metabolism and biosynthesis [15]. To investigate the
effect of nutrient medium on chalcopyrite concentrate dissolution, experiments were carried out in
9K (high amount of basalt salts) and Norris media (diluted medium) (see Table 3).
Figure 4-A shows the influence of temperature on copper recovery at leaching times 14 and
30 days. It exhibits that at lower leaching time, increasing temperature has a positive effect on cop-
per recovery while it would be diminished at higher leaching time (30 days). On the other hand, as
can be seen in Figure 3, rising temperature at 9K medium significantly decreases iron recovery. The
decrease is attributed to the precipitation of a part of dissolved iron as jarosite caused at high levels
of temperature and nutrient medium. As mentioned before, increasing temperature and basalt salts
is led to increasing iron precipitates. Considering the insignificancy of nutrient medium on copper
recovery (the target response) and the cheapness of Norris medium respect to 9K, using Norris me-
dium is chosen as a more suitable nutrient medium for the moderate thermophile mixed culture.
3.6 Influence of silver addition
Figure 5-A shows that at lower leaching time (14 days), the addition of 30 mg/l silver ion hasn't a
significant influence on copper recovery, while by increasing leaching time to 30 days, silver addi-
tion is the most effective factor. The important role of silver on copper recovery is exhibited at high
coefficient of D variable in Equation 3. The reason for negligible effect of silver addition on copper
recovery at lower leaching time is toxic effect of silver ions on the metabolism of bacteria leading
to prolonging of lag phase. While by increasing leaching time, bacteria are adapted with silver ions,
then because of the catalytic role of silver ions in the presence of bacteria, the copper dissolution
increases. Nevertheless, as Figure 5-B shows that silver addition decreases cells number even at
longer leaching time that is caused by toxic effect of silver on bacteria. On the other hand, Figure 6
N
u
m
b
e
r
o
f
B
a
c
t
e
r
i
a
(
C
e
l
l
s
/
m
L
1
*
1
0
^
7
)
A
B
Ahmadi, Ranjbar, Schaffie, Manafi
Proceedings of Copper 2010 2584
shows that silver addition has decreased iron recovery at the high level of pH. This decrease is at-
tributed to the co-precipitation of silver and iron and formation of silver-jarosite.
It was explained that silver forms an Ag
2
S film on the mineral surface according Equation 9 and
acts as a channel to transfer electrons leading to activation of the passive layer. Ferric ions oxidizes
the Ag
2
S (Equation 10). In bioleaching systems, bacteria oxidize sulphure and ferrous ions pro-
duced in Equation 10 and accelerate the rate of reaction [12].
S Ag Fe Cu Ag CuFeS
2
2 2
2
2 4 + + +
+ + +
(9)
+ + +
+ + +
2 0 3
2
2 2 2 Fe S Ag Fe S Ag
(10)
It was also found that Ag
2
S as a cathode and chalcopyrite as an anode form a galvanic couple, as a
consequence, the dissolution of chalcopyrite increases [12].
Figure 5: Influence of silver addition on A) copper recovery in leaching times 14 and 30 days,
B) cells number in leaching time 30 days
Figure 6: Interaction of pH and silver addition on iron recovery in leaching time 30 days
N
u
m
b
e
r
o
f
B
a
c
t
e
r
i
a
(
C
e
l
l
s
/
m
L
1
*
1
0
^
7
)
A
B
Optimization of Copper Concentrate Bioleaching
Proceedings of Copper 2010 2585
4 Conclusion
In this research, the influences of four important factors in bioleaching of chalcopyrite concentrate
were investigated and the main following results were obtained:
The optimum condition obtained from modelling of the experimental results using ANOVA was:
temperature 50 C, initial pH 1.8, nutrient medium Norris and silver concentration 30 mg/l in which
its calculated copper recovery was approximately 65 %. Meanwhile, the maximums copper recov-
ery (~70 %) and iron recovery (~40 %) from experiments were obtained at an experiment (run 9)
which its ORP was lower than 420 mV during the test and silver was added, while the minimum
copper recovery was achieved at an experiment (run 3) which its ORP reached to around 700 mV.
The least iron recovery (~10 %) was obtained in experiments (runs 8 and 12) which their ORP
reached to maximum values (~700 mV). It seems that this behaviour is due to the passivation of
chalcopyrite by jarosite.
It was found that despite of significant negative effect of silver addition on the cells number, when
bacteria are adapted to silver ions, its addition is the most effective parameter to extract copper from
the concentrate.
Both increasing temperature and using 9K medium (high amount of basalt salts) especially in the
presence of silver ions are led to an intensive decrease in iron recovery that are attributed to the
formation of jarosite.
Acknowledgment
This research was a part of Electro-bioleaching Program for Iranian copper sulphide concentrates.
The financial support of the National Iranian Copper Industry Company is gratefully acknowledged.
References
[1] HABASHI, F. (1978): Chalcopyrite; its Chemistry and Metallurgy. Mac-Graw Hill, New
York.
[2] HABASHI, F. (2006): Chalcopyrite: Bioleaching versus pressure hydrometallurgy. Interna-
tional Conference of Metallurgy of the XXI Century, Almaty.
[3] AHONEN, L. & TUOVINEN, O. H. (1990): Catalytic effects of silver in the microbiological
leaching of finely ground chalcopyrite containing ore materials in shake flasks: Hydrometal-
lurgy, 24, 219-236.
[4] OLSON, G.J., BRIERLEY, J.A. & BRIERLEY, C.L. (2003): Bioleaching review part B: Pro-
gress in bioleaching: applications of microbial processes by the minerals industries. Appl Mi-
crobiol Biotechnol, 63, 249257.
Ahmadi, Ranjbar, Schaffie, Manafi
Proceedings of Copper 2010 2586
[5] NORRIS, P.R., BURTON, N.P. & FOULIS, N.A.M. (2000): Acidophiles in bioreactor min-
eral processing- Extremophiles, 4, 7176.
[6] RODRIGUEZ, Y., BALLESTER, A., BLAZQUEZ, M.L., GONZALEZ, F. & MUNOZ, J.A.
(2003): New information on the chalcopyrite bioleaching mechanism at low and high tem-
peratures. Hydrometallurgy, 71, 4756.
[7] OKIBE, N., GERICKE, M., HALLBERG, K.B. & JOHNSON, D.B. (2003): Enumeration and
characterization of acidophilic microorganisms isolated from a pilot plant stirred-tank
bioleaching operation. Applied and Environmental Microbiology, 69, 19361943.
[8] OLSON, G.J. & CLARK, T.R. (2004): Fundamentals of metal sulfide biooxidation. Mining
Engineering, 56, 4046.
[9] CANCHO, L., BLAZQUEZ, M.L., BALLESTER, A., GONZALEZ, F. & MUNOZ, J.A.
(2007): Bioleaching of a chalcopyrite concentrate with moderate thermophilic microorgan-
isms in a continuous reactor system. Hydrometallurgy, 87, 100111.
[10] BLAZQUEZ, M. L., ALVAREZ, A., BALLESTER, A., GONZALEZ, F. & MUNOZE, J.A.
(1999): Bioleaching behaviour of chalcopyrite in the presence of silver at 35 and 68C. Bio-
hydrometallurgy and environment, PATR A, 127-137.
[11] PRICE, D. W. & WARREN, G. W. (1986): The influence of silver ion on the electrochemical
response of chalcopyrite and other mineral sulfide electrodes in sulfuric acid. Hydrometal-
lurgy, 15, 303-324.
[12] BALLESTER, A., BLAZQUEZ, L.M., GONZALEZ, F. & MUNOZ, J.A. (2007): Catalytic
role of silver and other ions on the mechanism of chemical and biological leaching. In: Do-
nati, E.R. and Sand, W. (Ed.), Microbial Processing of Metal Sulfides, 77101, Springer,
Netherlands.
[13] HIROYOSHI, N., ARAI, M., MIKI, H., TSUNEKAWA, M. & HIRAJIMA, T. (2002): A new
reaction model for the catalytic effect of silver iIons on chalcopyrite leaching in sulfuric acid
solutions. Hydrometallurgy, 63, 257 267.
[14] LPEZ-JUREZ, A., GUTIRREZ-ARENAS, N. & RIVERA-SANTILLN, R.E. (2006):
Electrochemical behavior of massive chalcopyrite bioleached electrodes in presence of silver
at 35 C. Hydrometallurgy, 83, 6368.
[15] ROSSI, G. (1990): Biohydrometallurgy. McGraw-Hill, Hamburg.
[16] HIROYOSHI, N., HIROTA, M., HIRAJIMA, T. & TSUNEKAWA, M. (1997): A case of
ferrous sulfate addition enhancing chalcopyrite leaching. Hydrometallurgy 47, 37-45.
[17] HIROYOSHI, N., MIKI, H., HIRAJIMA, T. & TSUNEKAWA, M. (2000): A model for fer-
rous-promoted chalcopyrite leaching. Hydrometallurgy, 57, 3138.
Optimization of Copper Concentrate Bioleaching
Proceedings of Copper 2010 2587
[18] HIROYOSHI N., MIKI, H., HIRAJIMA, T. & TSUNEKAWA, M. (2001): Enhancement of
chalcopyrite leaching by ferrous ions in acidic ferric sulfate solutions. Hydrometallurgy, 60,
185197.
[19] PINCHES, A., MYBURGH, P.J. & MERWE, C. (2001): Process for the rapid leaching of
chalcopyrite in the absence of catalysis. Patent No: US 6,277,341 B1.
[20] THIRD, K.A., CORD-RUWISCH, R., WATLING, H. R. (2002): Control of the redox poten-
tial by oxygen limitation improves bacterial leaching of chalcopyrite. Biotechnology and Bio-
engineering 78(4), 433-441.
[21] CORDOBA, E.M., MUNOZ, J.A., BLAZQUEZ, M.L., GONZALEZ, F. & BALLESTER, A.
(2008): Leaching of chalcopyrite with ferric ion. Part III: Effect of redox potential on the sil-
ver catalyzed process. 93 (3-4), 97-105.
[22] NARESH-KUMAR, A. & NAGENDRAN, R. (2007): Influence of initial pH on bioleaching
of heavy metals from contaminated soil employing indigenous Acidithiobacillus thiooxidans:
Chemosphere 66, 17751781.
[23] DAS, T., AYYAPPAN, S. & CHAUDHURY, G.R. (1999): Factors affecting bioleaching ki-
netics of sulfide ores using acidophilic micro-organisms. BioMetals, 12, 110.
Proceedings of Copper 2010 2588
Proceedings of Copper 2010 2589
Industrial NSC Pressure Oxidation of Combined
Copper and Molybdenum Concentrates
Corby G. Anderson
a
, Todd S. Fayram
b
, Larry G. Twidwell
c
a
Allihies Engineering Incorporated
b
Continental Metallurgical Services
c
Montana Enviromet
P. O. Box 123
Butte, Montana, 59703, USA
Keywords: Copper, molybdenum, pressure oxidation, NSC, tails leach, flotation
Abstract
The need for copper and molybdenum continues to grow as does the need for clean efficient metal-
lurgical technologies capable of treating mixed metal concentrates. Currently, the use of hydrome-
tallurgical pressure oxidation for copper concentrate treatment is growing. Conversely, with limited
molybdenum roasting capacity, stringent industrial molybdenum concentrate roaster feed specifica-
tions, poor rhenium recoveries, and the inherent environmental issues associated with pyrometallur-
gical treatments, hydrometallurgical options are also now being pursued for molybdenum concen-
trates. Moreover, given the inherent grade, recovery and cost inefficiencies in the differential
flotation process normally employed for molybdenum concentrates produced as a by product of
copper mining, there is a growing need to directly treat combined bulk copper and molybdenum
concentrates. This would minimize molybdenum concentrate roasting limitations, specifications and
requirements while allowing simplification of and cost reductions in upstream mineral processing
circuits producing separate copper and molybdenum concentrates by differential flotation. It would
also allow more direct and efficient recovery of rhenium. Finally, hydrometallurgical technology
will also limit the need for costly final molybdenum concentrate impurity treatment circuits thereby
allowing for lower grade mixed metal molybdenum concentrates to be treated directly for a fuller
metal value realization. In summary, industrial Nitrogen Species Catalyzed (i.e. NSC) hydrometal-
lurgical pressure oxidation has many advantages over conventional pressure oxidation systems and
offers a tangible process route. This paper illustrates the fundamanetal concepts, confirmatory test-
ing, application, proposed design concepts and cost estimates for the development and industrial
implementation of this proven mode of hydrometallurgical processing.
Anderson, Fayram, Twidwell
Proceedings of Copper 2010 2590
1 Introduction
As a first step, the basics of Nitrogen Species Catalyzed (i.e. NSC) hydrometallurgical pressure oxi-
dation will be outlined. The commonly reported leach reaction of a sulfide mineral with nitric acid
in conjunction with sulfuric acid is shown below.
3MeS
(s)
+ 2HNO
3
(aq)
+ 3H
2
SO
4
(aq)
3MeSO
4
+ 3S
(s)
+ 2NO
(g)
+ 4H
2
O (1)
However, it has been postulated and confirmed that the actual reaction species is NO
+
and not NO
3
-
(Anderson 1992, Anderson 1996, Baldwin 1996, Gok 2009). The addition of or presence of NO
2
-
instead of NO
3
D80
(Exxon Mobil aliphatic solvent) diluents matrix. Prior to introducing the PLS from the vat leach to
the SX feed, the solution was filtered through 1-micron string-wound cartridge filters. The PLS was
collected in batches using 1000 L carboys. Each batch was filtered, stirred to homogeneous, sam-
pled, and then fed to the SX circuit.
The SX circuit was fed PLS solution with a copper grade averaging 2.75 g/L Cu and it generated a
raffinate bearing an average 0.038 g/L Cu during the final two days of operation (during steady-state
conditions).
The copper grade in the aqueous solution in the preceding two stages of extraction averaged 0.046
and 0.181 g/L, respectively. This equates to a calculated recovery of 93.4%, 98.3 %, and 98.6 % as
the SX feed solution was sequentially processed through the 3-stage extraction circuit.
The pH of the SX feed and raffinate solution during this same period averaged 1.74 and 1.50, re-
spectively. The copper concentration in the loaded and stripped organic averaged 6.52 and 3.17 g/L.
Some crud formation was observed in the extraction and strip mixer-settlers. During the short test-
ing period, the crud did not appear to interfere with the SX operation. The strip solution employed at
start-up was a 235 g/L sulfuric acid solution. When the strip solution reached 45 g/L copper the EW
process was initiated.
Industrial NSC Pressure Oxidation of Combined Copper and Molybdenum Concentrates
Proceedings of Copper 2010 2601
During EW operation, the copper grade in the rich and lean electrolyte averaged 47.5 and 35.4 g/L
Cu, respectively. The free acid averaged 169 and 189 g/L H
2
SO
4
, respectively. Prior to starting the
EW process, the strip solution was dosed with sufficient guar gum to achieve 10 mg/L in the strip
solution inventory. Cobalt sulfate salt was also added to the strip solution, which yielded ~25 mg/L
Co in solution.
The target electrolyte temperature and current density were 40 C and 270 A/m
2
. The actual meas-
ured temperature of the electrolyte and current density were 31.1 C and 313 A/m
2
. Three copper
plates weighing a total of 35.2 kg were harvested. The deposit was dense and the appearance re-
flected typical cathode surface morphology. The nominal purity of the metal was 99.97 %. This
could be improved with the use of different anodes and fine tuning of reagent additions to the EW
cell. Overall copper recovery from the circuit was 98 %. The basic design parameters noted above
were used in the flowsheet design along with a design review by Cytec below in Table 5 and 6.
Table 5: Vat leaching flowsheet design parameters.
Circuit configuration 2 Extract +2 Strip
Organic 30 v/o Acorga M5640
PLS 10-15 g/L Cu, pH 1.8
Lean electrolyte 25 g/L Cu, 185 g/L H
2
SO
4
O:A Ratio- Extract 1.0-1.4:1 Strip 1.5: 1
Stage efficiency Extract 95 % Strip 95 %
Anderson, Fayram, Twidwell
Proceedings of Copper 2010 2602
Table 6: Cytec vat leaching flowsheet design review.
File Org.
Flow
PLS
Flow
L.E.
Flow
PLS O:A
Ext.
O:A
Strip
M5640 Raffi-
nate
Rich E. Recovery Produc-
tion
Produc-
tion
[g/m] [g/m] [g/m] Cu pH (v/o) [g/l] [g/l] [%] [t/day] [t/year]
CAB 1,500 1,500 1,000 10.0 1.8 1.00 1.50 30 0.73 48.90 92.7 80.9 27,650
1A 1,650 1,500 1,100 10.0 1.8 1.10 1.50 30 0.60 47.82 94.0 82.9 28,068
2 1,500 1,500 1,000 11.0 1.8 1.00 1.50 30 1.09 49.86 90.1 84.9 29,558
2A 1,650 1,500 1,100 11.0 1.8 1.10 1.50 30 0.85 48.83 92.2 88.1 30,273
3 1,500 1,500 1,000 12.0 1.8 1.00 1.50 30 1.59 50.62 86.8 84.9 30,988
3A 1,650 1,500 1,100 12.0 1.8 1.10 1.50 30 1.22 49.69 89.8 88.1 32,156
4 1,500 1,500 1,000 13.0 1.8 1.00 1.50 30 2.21 51.19 83.0 80.0 32,193
4A 1,650 1,500 1,100 13.0 1.8 1.10 1.50 30 1.71 50.40 86.9 92.4 33,689
4B 1,800 1,500 1,200 13.0 1.8 1.20 1.50 30 1.36 89.5 95.2 34,748
5 1,500 1,500 1,000 14.0 1.8 1.00 1.50 30 2.93 51.60 79.1 90.5 33,033
5A 1,650 1,500 1,100 14.0 1.8 1.10 1.50 30 2.30 50.95 83.5 93.8 34,237
5B 1,800 1,500 1,200 14.0 1.8 1.20 1.50 30 1.84 50.20 86.9 99.5 36,301
5C 1,950 1,500 1,300 14.0 1.8 1.30 1.50 30 1.50 49.41 89.3 102.2 37,295
6C 1,950 1,500 1,300 15.0 1.8 1.30 1.50 30 1.98 50.03 86.8 106.5 38,852
6D 2,100 1,500 1,500 15.0 1.8 1.40 1.50 30 1.66 49.30 89.0 109.1 39,822
No significant concerns were noted. A laboratory-scale pilot plant run will be required to confirm
the operating parameters and examine the build-up of deleterious elements such as silica and crud
under a flotation/leach setting.
A relic copper cementation circuit will be set-up on the raffinate to maximize copper recovery. The
cementation process will be similar to that of Kennecotts Spedden process and used to ensure mi-
nimal copper is sent back to the tails leach and potentially lost. The cemented copper product will
be recycled back to the pressure oxidation circuit. Subsequent, testwork will be required to optimize
the cementation circuit design, recovery, and economics.
Industrial NSC Pressure Oxidation of Combined Copper and Molybdenum Concentrates
Proceedings of Copper 2010 2603
4.5 Thickening
Limited thickening or rheological testing has been completed. Only the slimes from the Vat leach-
ing tests had thickening and rheological testing. Industry averages were used to size the thickeners
and other necessary settling equipment in the flotation and leaching circuits. The following design
parameters were used in developing the thickener characteristics:
4.5.1 Thickener design criteria
Flotation Tails
Thickener U/F density 50 % solids
Flux rate t/m
2
/h 0.52
The current design uses conventional thickening and assumes a 50 % underflow density. Further
testing will be required to ensure that high compression thickening will not be required to make the
necessary 50 % underflow prior to leaching.
Post Leach
Thickener U/F density 50 % solids
Flux rate t/m
2
/h 0.52
Flocculent may be required to meet the specific design criteria noted above.
The current design uses conventional thickening and assumes a 50 % underflow density. A case
study comparing filtration with counter-current decantation (CCD) for the leach residue application
will need to be reviewed. Based on the available data, CCD typically has slightly better overall eco-
nomics, with significant performance advantage and reliability in the cleaning of the leach residues.
The process design will need to review the use of high compression thickeners to maximize under-
flow densities and minimize losses of copper through the CCD thickeners.
NSC pressure oxidation (POX) residue
Thickener U/F density 50 % solids
Flux rate t/m
2
/h 1.0
Flocculent may be required to meet the specific design criteria noted above.
The current design uses conventional thickening and assumes a 50 % underflow density. A case
study comparing thickening to filtration will be required to optimize this part of the circuit. Further
thickening and rheological testing will be required on all aspects of the operation.
Anderson, Fayram, Twidwell
Proceedings of Copper 2010 2604
4.6 Filtration
No vacuum and pressure filtration tests were carried out in conjunction with the Pregnant Feed So-
lution thickening tests. Industry averages for chalcopyrite concentrate filtration were used. The fol-
lowing design parameters were used in developing the filtration characteristics:
4.7 Filtration design criteria
Copper Concentrate (Includes concentrate thickener)
(Assumes 65 % solids, product has a P
80
of 325 mesh)
Flux rate t/m
2
/h 0.487
NSC POX Material (includes any vertical mill ground material to include leach residues)
(Assumes 65 % solids, product has a P
80
of 10 micron)
Flux rate t/m
2
/h 1.0
Further filtration test data will be required for the finalization of the flowsheet.
4.8 Mixed copper and molybdenum concentrate leaching
The sulfide flotation concentrate will be treated using industrially proven Nitrogen Species Cata-
lyzed Pressure Oxidation Leaching (NSC). A vertical grinding mill will be fed concentrate contain-
ing a nominal 29 % copper and approximately 0.9 % molybdenum and ground to approximately 10
microns. The ground concentrate is then fed to one of two pressure oxidation autoclaves using the
following leaching parameters listed in Table 7.
Table 7: NSC pressure oxidation combined concentrate leaching parameters.
Parameter Unit Value
Solids flowrate t/d 202
Slurry solids content g/l 50
Initial free sulfuirc acid g/l 25
Reactor working pressure kPag 620
Maximum temperature C 155
Total leach time min 45
Industrial NSC Pressure Oxidation of Combined Copper and Molybdenum Concentrates
Proceedings of Copper 2010 2605
From the autoclave, a Cu/Mo bearing solution is recovered from the autoclave effluent via a high-
rate thickener and polishing filtration, and then through a series of activated carbon columns for
molybdenum recovery. Then an EcoTec Acid Purification Unit (APU) is used to recover excess
sulfuric acid which is recycled back to the autoclave feed system. After molybdenum and acid re-
covery, the copper-laden leach solution is sent to a conventional SX/EW circuit and the copper re-
covered as electrowon cathode.
From industrial operating data, the NSC process results in near total oxidation (+99 %) of the sul-
fide concentrate and is estimated to convert a majority of the sulfide sulfur to sulfuric acid (60 %)
based on operating experience, and some to elemental sulfur (40 %). The elemental sulfur is col-
lected in the POX residue through thickening and filtration. The POX residue is then leached for the
sulfur content with sodium hydroxide to make a soluble sodium sulfate and a final leach residue.
The NSC POX residue leaching parameters are listed in Table 8.
Table 8: NSC POX residue leaching parameters.
Parameter Unit Value
Solids flowrate t/d 126
Slurry solids content g/l 50
Reactor working pressure kPag 620
Maximum temperature C 130
Total leach time min 45
Approximately 50 % of the POX residue will be recycled back to the NSC leach for enhanced cop-
per and molybdenum recovery and to provide inert heat sink materials to the NSC POX system.
This helps to optimize the concentrate pressure oxidation system by providing more heat capacity.
The other estimated 50 % of residue will be sent to tails. The amount recycled and sent to tails will
be optimized to maximize throughput and metal recoveries.
In the overall POX residue reaction, sodium sulfate is formed. After thickening and polishing filtra-
tion, the sodium sulfate solution is passed through a set of carbon columns to recover relic molyb-
denum. The purified sodium sulfate solution will be further processed via electrodialysis (EDU) and
evaporation to recover a 70 % sulfuric acid product and a 100-g/l sodium hydroxide product. These
will be recycled back into the process plant. Any excess sodium sulfate will be sent to a multiple
effect crystallizer, crystallized into flake sodium sulfate, filtered, dried, bagged, and sold. The relic
molybdenum recovered will be processed as identified below.
The molybdenum-laden activated carbon is atmospherically stripped with ammonium hydroxide to
make an ammonium molybdate solution. The pH of the stripped solution is then decreased to 2.0
using sulfuric acid causing ammonium molybdate to precipitate. The precipitated ammonium mo-
lybdate is then filtered with the clear solution being recycled back to the process. The filtered am-
Anderson, Fayram, Twidwell
Proceedings of Copper 2010 2606
monium molybdate solid is then calcined at 550 C, converted to molybdenum trioxide, bagged, and
sold. The ammonia driven off by the calcination process is scrubbed, collected, and reused in the
system. The stripped carbon will be recycled back to the molybdenum recovery circuit. A portion of
the carbon will also be reactivated in a conventional kiln operation.
Both the NSC concentrate and leached residue pressure oxidation processes can provide significant
heat. For efficient process, heat, or energy economy, the resultant heat will be used in the process
plant to increase the atmospheric flotation tailings leach temperatures, thereby improving copper
leaching kinetics and ultimately increasing copper recovery levels. This waste heat will also be uti-
lized in the EDU sulfuric acid recovery and sodium sulfate multiple effect evaporation system.
The production rate for the NSC concentrate pressure oxidation circuit is designed with excess ca-
pacity so outside copper and molybdenum containing concentrates can be processed through the
plant on a toll or purchased basis. Further testing will be required to ensure the optimized parame-
ters of the NSC pressure oxidation system are identified and quantified for concentrate treatment
and for final sizing of the various items associated in the metal recovery systems.
4.9 Crushing and ore storage and reclaim
Initial ore to the plant is expected to be relatively easy to treat. A small portion is highly weathered
and the remainder is highly competent with low moisture content values reported at approximately
3%. The proposed crushing circuit utilizes a C160 Nordberg Jaw Crusher that would crush run of
mine rock to a nominal 152 mm. The jaw crusher feeds a live-surge capacity stockpile which is no-
minally designed for 24 hours of live capacity, and six reclaim feeders will feed the two proposed
grinding circuits.
4.10 Grinding and classification
The two milling circuits have been designed on the basis of the use of a SAG mill and ball mill
combination. The design is based on the Asarco South Mill (Pima Mill). The ore milling circuits
have been designed using a SAG Mill Work Index of 5.58 kWh/t to 1200 microns, and a Ball Mill
Work Index of 16.5 kWh/t. The Bond Ball Mill Work Index for the ore is reported to vary from 15.4
to 17.6 kWh/t, and there is no significant difference between the sulfide and oxide material. On this
basis, the installed power for the SAG mill grinding 5.2 Mt/yr of ore is calculated at 8.5 MW.
The designed steel charge for the SAG Mill consists of a nominal ball load of 8 % by volume, with
a design load of 12 %. Abrasion indices are based on the nearby Highmont Mill which are consi-
dered moderate and have been incorporated into the operating costs. Values for the abrasion indices
vary from 0.5 to 0.7. The design cyclone overflow product size is at a P
65
of 74 micrometer for the
downstream processes of leaching and/or flotation.
Industrial NSC Pressure Oxidation of Combined Copper and Molybdenum Concentrates
Proceedings of Copper 2010 2607
4.11 Flotation
Oxide and sulfide ores will be campaigned separately with all of the ores initially treated by flota-
tion and leaching. As oxide ore is depleted, the leaching circuit will be bypassed. The flotation of
the sulfide ore is based on a conventional rougher/scavenger and two-stage cleaning circuit. Rough-
er concentrate potentially will be reground to liberate sulfide copper minerals from gangue. Testing
will be required to confirm the need for a regrind mill. The selected reagent regime is relatively
simple, using only xanthate as collector combined with a frother. The plant has been designed to
maximize recovery of sulfide mineralization.
Recovery of acid soluble copper mineralization is not critical as the flotation tailings will be acid
leached. General reagent consumption and residence times have been based largely on recent test-
work performed with flotation prior to leaching. Locked cycle bench testing has been performed on
the different ore types and locations.
4.12 Pre-leach dewatering
The pre-leach dewatering (thickening) step is designed to maintain the solution balance in the leach
SX/EW circuit. The leach circuit is designed for a 50 % underflow.
4.13 Copper leach-oxide/mixed ore only
Testwork results indicate relatively slow copper leach kinetics for the flotation tail streams. The
leach circuit design is based on a 24 hour residence time at 50 % solids. Gangue acid consumption
will be based on the mine schedule, with a total estimated acid consumption of 17 kg/t, allowing for
losses and bleed streams.
4.14 Leach residue counter current decantation (CCD)
Literature data has been used to design a CCD circuit comprising four stages of 35 m diameter stan-
dard thickeners. Testing will be required to evaluate the CCD circuit design.
4.15 Solvent extraction
The mass balance indicates a maximum PLS flowrate of approximately 617 m
3
/h. The mixer-
settlers have then been sized on the basis of residence time of two minutes in each of the two mixer
stages, and a settler specific area of 4.5 m
3
/h/m
2
. The configuration of the solvent extraction circuit
is comprised of two extract mixer/settler stages, a wash stage, and a single strip stage.
Anderson, Fayram, Twidwell
Proceedings of Copper 2010 2608
Maximum flow will occur during mining of the predominantly oxide leach cap. Flow rates to the
Pregnant Leach Solution pond will consistently drop during the first three years of the project to
approximately 143 m
3
/hr as sulfides become the predominant species mined. The solvent extraction
system is designed for maximum expected flows and copper grades.
4.16 Electrowinning
The tank house will use stainless steel cathode technology, with anodes consisting of a lead-
calcium-tin alloy. Electrode sizes will be standard for copper recovery, with a submerged area of
1.1 m
2
, and the cathodes will be pulled on a seven-day stripping cycle. An average current density of
258 (maximum 275) A/m
2
will be used and which should be readily achievable in the operating
environment, resulting in a requirement for 149 cells of 45 cathodes each. This cell size results in a
design current of 24,000 Amps divided between two rectifiers. The cells will be arranged in two
blocks of 75 cells, each with its own rectifier. Cathode stripping will typically be manual with au-
tomatic presentation of the cathodes to the stripping station, and will operate on at least two shifts
per day. Total design cathode copper production is 30,000 tonnes per year.
4.17 Concentrate handling
The concentrate handling system will consist of a Larox pressure filter and storage system. The La-
rox filter is designed for 37 m
2
of filtration capacity. Dewatered material from the Larox filter will
either be repulped and fed to the Nitrogen Species Catalyzed (NSC) concentrate leach system, or
stored and shipped to a smelter
The NSC oxidative pressure leaching system consists of a 300 HP Tower Mill to grind to a P
80
of
10 m. From there, the material will be diluted to 50 g/L with recycled acidic process solution and
pressure leached in one of two unlined Grade 316 Stainless Steel pressure leach vessels. Pressure
and temperature will be maintained accordingly for full leaching of the copper and molybdenum,
while maintaining some of the sulfur as elemental sulfur.
Upon discharge from the leach vessels, the pulp will be flashed and the leach residue thickened and
filtered. The pregnant leach solution (PLS) will pass through secondary filtration and on to molybde-
num recovery. The filtered pressure leached residue, or POX residue, will report to sulfur recovery.
4.18 Molybdenum recovery
The PLS grade will be maintained at approximately 15 gram/liter copper and 0.28 g/L molybdenum.
The PLS will be passed through a series of five 2-ton carbon columns for molybdenum recovery.
Based on industrial practices, approximately 99 % of the molybdenum is expected to be recovered
in the carbon columns. After molybdenum recovery, the PLS will pass through an industrial EcoTec
Industrial NSC Pressure Oxidation of Combined Copper and Molybdenum Concentrates
Proceedings of Copper 2010 2609
Acid Purification Unit to recover free acid from the PLS. The recovered sulfuric acid solution prod-
uct will be recycled back to the NSC pressure oxidation system. The resultant PLS free acidity will
be dropped from about 30 g/L to approximately 5 g/L. This molybdenum-free PLS solution is then
sent to the PLS pond where the copper is recovered through a conventional SX/EW circuit.
Upon loading the carbon with molybdenum, the carbon is transferred to an atmospheric strip vessel.
The pH is raised with ammonium hydroxide to strip the molybdenum from the carbon, and a con-
centrated solution of ammonium molybdate is then formed. The ammonium molybdate solution is
treated by pH adjustment to 2.0 using sulfuric acid, and then solid ammonium molybdate precipi-
tates. The ammonium molybdate slurry is then filtered in a 3.7 m
2
filter, dewatered, and fed to a
dryer heated to 550 C. The ammonium molybdate is calcined to form molybdenum trioxide. The
ammonia driven off in the calcining process is captured in a water scrubber and reused in the
process. The molybdenum trioxide product is dried, bagged and sold.
4.19 NSC POX leached residue sulfur recovery system
Due to the high cost of sulfur based products, another POX system will be implemented to recover
sulfur from the NSC leached residue. The NSC POX residue is settled in a 35-meter thickener and
the underflow is filtered in a Larox filter. The POX material is stored until ready for sulfur recovery.
During sulfur recovery, the POX residue is fed into a leach vessel and ultimately pressure leached
with sodium hydroxide and oxygen. The sulfur is converted to sodium sulfate and a sulfur free POX
residue is formed. The POX residue is thickened and approximately 50 % of this residue is recycled
back to NSC, with the remaining 50 % of the material going to final tails.
The sodium sulfate solution is filtered and polished, and then fed to a relic molybdenum recovery
circuit similar to that discussed above. The clarified and purified sodium sulfate solution is then fed
into an Electrodialysis Unit (EDU) and evaporator system. The EDU will convert approximately
50 t/d of sodium sulfate into about 25 t/d of caustic. The caustic grade will be 100 g/L sodium hy-
droxide. Also, 35 t/d of sulfuric acid will be produced at a concentration of 70 % sulfuric acid.
These by-product solutions will be recycled and utilized in the process where required for leaching,
sulfur recovery, and neutralization. Excess sodium sulfate solution from this process will also be
treated in a conventional crystallizer system and solid sodium sulfate will be produced, bagged, and
sold based on market prices and demand.
4.20 Operating and capital cost estimate
Annual production is based on mining activities operating at assumed 90 % efficiency. The pre-
production stripping numbers are also assumed and will be defined when the final geologic and
mine reserve models have been completed. Haul road distances are also assumed and will be deter-
Anderson, Fayram, Twidwell
Proceedings of Copper 2010 2610
mined once a location for the processing plant has been selected and waste rock storage areas
designated. The Development and Operational key process indicators are:
Pre-production Stripping (North Pit) 300,000 t
Pre-production Stripping (South Pit) 200,000 t
Haul Road Construction (North Pit) 2500 m
Haul Road Construction (South Pit) 2500 m
Initial Production
Ore Production 15,000 t/d
Waste Production 37,000 t/d
Concentration Production 200 t/d
Annual Production Ore 5.25 million t
Annual Production Waste 13.0 million t
The milling plan estimates capital and operating costs assume a nominal 15,000 tonnes per day min-
ing operation. The milling plan includes:
All labor, material, supply and equipment operating costs for the mill and associated concentrate
leach plant
Supervision, administration and on-site management
Benefits and employment taxes
All on-site development for start-up and production
Mill equipment and facilities purchase and installation or construction.
Engineering and construction management fees
Industrial NSC Pressure Oxidation of Combined Copper and Molybdenum Concentrates
Proceedings of Copper 2010 2611
Pre-production development, installation and construction of all equipment and facilities necessary
to operate the mill at a nominal 15,000 t/d are included. Costs associated with the following facili-
ties and operations are included. However final locations and design details are pending:
Crushing and conveyance of ore to the grinding circuit
Grinding and flotation
Flotation Tails leaching
Copper and molybdenum concentrate pressure leaching
Molybdenum trioxide recovery
Copper solvent extraction and copper cathode electrowinning
Eco Tec APU sulfuric acid recovery
Sodium sulfate production
Electrodialysis salt splitting of sodium sulfate for caustic and sulfuric acid production
Tailings facility
Basic access roads, power lines, and pipelines
Construction, installation and operation of facilities and equipment necessary for equipment
maintenance and repair, electrical system, fuel distribution, water storage and drainage, sanitation
facilities, offices, labs, storage, and equipment parts and supply storage.
The mill and metallurgical plan does not include:
Permitting and environmental assessment costs
Home office overhead
Taxes (except sales taxes and employment taxes)
Insurance
Depreciation
Off-site transportation of products
Incentive bonus premiums
Overtime labor costs
Sales expenses
Interest expenses
Start-up costs (except for working capital)
Depletion rates
Environmental costs, including reclamation
Reclamation on-going assuming a two-year start-up period
Anderson, Fayram, Twidwell
Proceedings of Copper 2010 2612
Tables 9 and 10 summarize the operating and capital cost estimates done to an accuracy of +/- 30 %.
Table 9: Operating cost estimate USD +/- 30 %.
Cost summary
Category USD $/t
Mine 1.41
Plant average 12.94
Plant Oxide 10.13
Plant Sulfide 13.91
Administration 0.54
Total operating cost 14.76
Table 10: Capital cost estimate, USD +/- 30 %.
Cost summary
Category USD $
Mine (less EPCM) 42,638,050
Mill & hydromet plant 230,770,238
Indirects 45,291,677
Contingency 39,541,957
Working capital 19,727,557
Reclamation bond 7,277,278
Total capital 381,920,231
5 Summary
This paper has demonstrated the successful application of industrial NSC pressure oxidation to
combined concentrates containing both copper and molybdenum as chalcopyrite and molybdenite
respectively. Confirmatory optimized NSC pressure leaching was successfully undertaken to leach
copper, molybdenum and rhenium. Further, as practiced industrially, selective separation and con-
centration of NSC leached rhenium and molybdenum using ion exchange and activated carbon was
successfully tested and confirmed. Finally, using this confirmatory test data along with industrial
operating data and experience, design criteria flow sheets and capital and operating costs estimates
were produced.
Industrial NSC Pressure Oxidation of Combined Copper and Molybdenum Concentrates
Proceedings of Copper 2010 2613
Acknowledgements
Thank you to Dr. Paul J. Miranda for all the analysis and mineralogical assessment for the NSC
confirmatory testing of the bulk molybdenite and chalcopyrite concentrates. In addition, as a cour-
tesy to the reader, both pertinent references and a bibliography of related work are included in the
paper.
References and bibliography
[1] Abel, E., and Schmid, H., 1928, Z. Physc. Chem. (Leipzig), 132: 56-64.
[2] Abel, E., Schmid, H. and Pollack, F. 1936, Monatsh, Chem., 69: 125-143.
[3] Ackerman, J.B., Anderson C.G., S.M. Nordwick and L.E. Krys. 1993. Hydrometallurgy at the
Sunshine Metallurgical Complex. Proceedings of the Fourth International Symposium on
Hydrometallurgy. Littleton, Colorado, SME.
[4] Anderson, C.G. and K.D. Harrison. 1990. Optimization of Nitric-Sulfuric Acid Pressure
Leaching of Silver from Refractory Sulfide Concentrate. Precious Metals 1990. International
Precious Metals Institute, Allentown, Pennsylvania, 119-151.
[5] Anderson, C. G. and Harrison, K. D., "Improvements in Plant Scale Nitric-Sulfuric Acid Pres
sure Leaching of Refractory Sulfide Concentrates", Preprint No. 91-183, AIME-SME Annual
Meeting, Denver, Colorado, 1991.
[6] Anderson, C.G., L.E. Krys, and K.D. Harrison. 1992. Treatment of Metal Bearing MineralMa-
terial. US Patent 5,096,486.
[7] Anderson, C.G., K.D. Harrison, and L.E. Krys, edited by R.K. Mishra. 1993a. Process Integra-
tion of Sodium Nitrite Oxidation and Fine Grinding in Refractory Precious Metal Concentrate
Pressure Leaching. Precious Metals 1993. International Precious Metals Institute, Allentown,
Pennsylvania, 19-45.
[8] Anderson, C.G., and L.E. Krys. 1993b. Leaching of Antimony from a Refractory Precious
Metals Concentrate. Proceedings Of The Fourth International Symposium On
Hydrometallurgy. Salt Lake City, Utah, August 1-5.
[9] Anderson, C.G. and S.M. Nordwick. 1994. The Application of Sunshine Nitrous-Sulfuric
Acid Pressure Leaching to Sulfide Materials Containing Platinum Group Metals. Precious
Metals 1994. Proceedings of the 18th Annual IPMI Conference, Vancouver, B.C., 223-234,
June.
Anderson, Fayram, Twidwell
Proceedings of Copper 2010 2614
[10] Anderson, C.G., K.D. Harrison, and L.E. Krys. 1996. Theoretical Considerations of Sodium
Nitrite Oxidation and Fine Grinding in Refractory Precious Metals Concentrate Pressure
Leaching. Minerals and Metallurgical Processing, AIME-SME, Volume 13, Number 1,
February.
[11] Anderson, C. G., Harrison, K. J., and Krys, L. E., "The Application of Sodium Nitrite Oxida-
tion and Fine Grinding In Refractory Precious Metals Concentrate Pressure Leaching", Mine-
rals and Metallurgical Processing, AIME-SME, Volume 13, Number 1, February, 1996.
[12] Anderson, C.G. 2000a. Nitrogen Species Catalyzed Pressure Leaching of Copper Ores and
Concentrates. ALTA Copper 2000, Adelaide, South Australia, October.
[13] Anderson, C.G. 2000b. The Design, Optimization and Operation of an Industrial Copper Sol-
vent Extraction and Electrowinning Circuit at a Commercial Nitrogen Species Catalyzed
Pressure Leaching Plant. ALTA SX/IX 2000, Adelaide, South Australia, October.
[14] Anderson, C.G. 2001a. The Industrial Non-Cyanide Hydrometallurgical Recovery of Silver
and Gold Utilizing Nitrogen Species Catalyzed Pressure Oxidation. Cyanide Social, Industrial
and Economic Aspects. Symposium Proceedings, TMS Annual Meeting, New Orleans, Loui-
siana, February.
[15] Anderson, C.G. 2001b. Industrial Nitrogen Species Catalyzed Pressure Leaching and Alkaline
Sulfide Gold Recovery from Refractory Gold Concentrates. Precious Metals 2001, 25th An-
nual IPMI Meeting, Tucson, Arizona. June.
[16] Anderson, C.G. April 2002a. The Chemical Analysis of Industrial Alkaline Sulfide
Hydrometallurgical Processes. The Society of Mineral Analysts and the Canadian Mineral
Analysts Annual Meeting, Spokane, Washington.
[17] Anderson, C.G. May 2002b. The Mineral Processing and Industrial Nitrogen Species
Catalyzed Pressure Leaching of Formation Capital's Cobaltite and Chalcopyrite Concentrates.
ALTA Ni/Co and Cu International Conference, Perth, W.A. Australia.
[18] Anderson, C.G. June 2002c. The Application of Industrial NSC Pressure Leaching in the Re-
cycle and Recovery of Secondary Precious and Base Metals. TMS Fall 2002, Recycling and
Waste Treatment Meeting. Lulea, Sweden.
[19] Anderson, C.G. 2003a. Alkaline Sulfide Recovery of Gold Utilizing Nitrogen Species
Catalyzed Pressure Leaching. Hydrometallurgy 2003. Vancouver, B.C., October.
[20] Anderson, C.G., and E. Rosenberg. 2003b. Single Step Separation and Recovery of Palladium
Using Nitrogen Species Catalyzed Pressure Leaching and Silica Polyamine Composites.
Hydrometallurgy 2003. Vancouver, B.C., October.
[21] Anderson, C.G. 2003c. The Application and Economics of Industrial NSC Pressure Leaching
to Copper Ores and Concentrates. COBRE 2003, Santiago, Chile, December.
[22] Anderson, C.G., E. Dahlgren and D. Stacey. 2004. Unpublished research.
Industrial NSC Pressure Oxidation of Combined Copper and Molybdenum Concentrates
Proceedings of Copper 2010 2615
[23] Anderson, C.G. and L.G. Twidwell. 2006. The Control of Iron and Arsenic in the Treatment
of Gold Bearing Enargite Concentrates, Iron Control In Hydrometallurgy, Proceedings of the
36th Annual Hydrometallurgy Meeting of CIM, October 2006, Montreal, Quebec
[24] Anderson, C.G. and L.G. Twidwell. 2007. Hydrometallurgical Processing of Gold-Bearing
Copper Enargite Concentrates, 6th COPPER/COBRE Conference, John Dutrizac
Hydrometallurgy Symposium, 37th Annual Hydrometallurgy Meeting of CIM, Toronto, Onta-
rio, August.
[25] Anderson, C.G. and L.G. Twidwell. 2008. Hydrometallurgical Processing of Gold-Bearing
Copper Enargite Concentrates, Canadian Metallurgical Quarterly, John Dutrizac
Hydrometallurgy Symposium Special Issue.
[26] Anderson, C.G. 2008. NSC Pressure Leaching: Industrial and Potential Applications,
Hydrometallurgy 2008 Proceedings of the Sixth International Symposium, SME, 858-885
[27] Awad, H.H., and Stanbury, D.M., 1993, Autooxidation of NO in aqueous solution. Int. J.
Chem. Kinet. 25: 375-381.
[28] Baldwin, S.A. and G.V. Van Weert. 1996. On the Catalysis of Ferrous Sulfate Oxidation in
Autoclaves by Nitrates and Nitrites. Hydrometallurgy, Elsevier Science, B.V., Vol. 42.
[29] Balachandran P., Kamath, A.K. Mitra, S. Radhakrishnan and Kiram Shetty, 2003. Electrolyte
Impurity Control at Chinpada Refinery of Sterlite Industries (India) Limited, Eco-Tec SM
Technical Paper 167.
[30] Bardt. 1919-1923. Recovering Metals Contained in Metalliferous Ore, Waste Residues, and
Alloys. German Patent 353,795, Canadian Patent 233,566, Chemical Abstracts 173,154.
[31] Barth, T.R., A.T.C. Hair, and T.P. Meier, edited by J.E. Dutrizac et al. 1998. The Operation of
the HBM&S Zinc Pressure Leach Plant. Zinc and Lead Processing, The Metallurgical Society
of CIM.
[32] Beake, B.D., Constantine, J. and Modie, R. B., 1992, The kinetics and mechanism of the
reaction of nitrous acid with 4-substituted phenols in aqeous solution. J. Cem. Soc. Perkin
Trans. 2: 1653-1654.
[33] Beattie, M.J.V., R. Randsepp, and A. Ismay, edited by G.S. Dobby and S.R. Rao. 1989.
Arseno/Redox Process for Refractory Gold Ores. Proc. Intern. Symp. Processing Complex
Ores. Pergamon, Oxford, 431-439.
[34] Bjorling, C. and G.A. Koltz. 1964. Oxidizing Leaching of Sulfide Concentrates and Other
Materials Catalyzed by Nitric Acid. Proc. 7th In. Mineral Processing Cong., New York, NY.
[35] Bjorling, G. and G.A. Kolta. 1966a. Oxidizing Leach of Sulfide Concentrates and Other Mate-
rials Catalyzed by Nitric Acid. Proceedings of 7th International Mineral Processing Congress.
Gordon & Breach, New York, NY, 127-138.
Anderson, Fayram, Twidwell
Proceedings of Copper 2010 2616
[36] Bjorling, G. and G.A. Kolta. 1966b. Wet Oxidation of Iron Sulfide Concentrates Catalyzed by
Nitric Acid. J. Chemistry, U.A.R., 9(2):187-203.
[37] Bjorling, G. and P. Lesidrenski. 1968. Hydrometallurgical Production of Copper from
Activated Chalcopyrite. AIME Annual Meeting, New York, NY.
[38] Bjorling, G. and G.A. Kolta. 1969. Wet Oxidation as a Method of Utilization of Chalcopyrite,
Sphalerite, and Molybdenite. Chemistry U.A.R., 12(3):423-435.
[39] Bjorling, G., edited by D.J.I Evans and R.S. Shoemaker. 1973. Leaching of Mineral Sulfides
by Selective Oxidation at Normal Pressure. International Symposium on Hydrometallurgy.
AIME, New York, NY, 701-707.
[40] Bjorling, G. and W. Mulak. June 1976a. Kinetics of NiS Leaching in Nitric Acid Solutions.
Trans. Inst. Min. & Met. C98-C101.
[41] Bjorling, G. et al., edited by J.C. Yannopoulos and J.C. Agarwal. 1976b. A Nitric Acid Route
in Combination with Solvent Extraction for Hydrometallurgical Treatment of Chalcopyrite.
Extractive Metallurgy of Copper. Vol. 2. AIME, New York, NY, 725-737.
[42] Bosio, S., Ravella, A., Saracco, G.B., and Genon G., 1985. NOX Absorption by Ferrous Sul-
fate Solutions, Ind. Chem. Process Res. Dev., 24:149-152
[43] Brennecke, H.M. 1975. Recovery of Metal Values from Ore Concentrates. US Patent
3,888,748.
[44] Brennecke, H.M. et al. August 1981. Nitric-Sulfuric Leach Process for Recovery of Copper
from Concentrates. Min. Engineering. 1259-1266.
[45] Bunton, C.A., Stedman, G., 1959, Mechanism of the Azide-Nitrite Reaction. Part III Reaction
in [18O] water. J. Chem. Soc. 3466-3474.
[46] Butler, A.R., and Ridd, J.H., 2004, Formation of nitric oxide from nitrous acid in ischemic
tissue ands skin, Nitric Oxide, 10: 20-24.
[47] Caldon, F. 1978. Treatment of Metal Bearing Mineral Material. US Patent 4,084,961.
[48] Chmielewski, T. and Charewicz, W.A., 1984, The oxidation of ferrous iron in aqueous
sulfuric acid under oxygen pressure, Hydrometallurgy, 12:21.
[49] Daugherty, E.W., A.F., Erhard and J.M. Drobnick. 1973. Process of Recovery of Rhenium
and Molybdenum Values from Molybdenite Concentrate. U.S. Patent 3,739,057.
[50] Dresher, W.H., Wadsworth, M. E., Fassel, W. M., 1956, A Kinetic Study of the Leaching of
Molybdenite, Mining Engineering, Transactions of AIME, Littleton, Colorado, June 738-744.
[51] Edwards, C.R., edited by A.D. Zunkel. 1985. Engineering the Equity Concentrate Leach Pro-
cess. Complex Sulfides: Processing of Ores, Concentrates and By-Products. Proceedings of a
symposium sponsored by the Metallurgical Society of AIME and the CIMM, TMS-AIME Fall
Extractive Meeting, San Diego, CA, 197-219, November 10-13.
Industrial NSC Pressure Oxidation of Combined Copper and Molybdenum Concentrates
Proceedings of Copper 2010 2617
[52] Epstein, I.R.,Kustin, K. And Simoyi, R.H., 1982, Nitrous acid decomposition catalyzed by and
iron (II) complex: Tris (3,4,7,8-tetramethyl-1, 10-phenanthroline) iron)II). American Chemical
Society. 104: 712-717.
[53] Eugene, S.A, edited by A. Sutulov. 1979. Cymoly Process, in International Molybdenum
Encyclopedia, Vol. 2, Processing and Metallurgy, Antermet Publications, Santiago, Chile,
105.
[54] Fair, K.J., J.C. Schneider, and G. Van Weert, edited by R.D. Salter. 1987. Options in the
Nitrox Process. M. Proc. Intern. Symp. Gold Metallurgy. Pergamon, Toronto, 279-291.
[55] Fair, K.J. and F.J. Basa, edited by G.S. Dobby and S.R. Rao. 1989a. Treatment of Agnico
Eagle's Silver-Bearing Flotation Concentrate by the Nitrox Process. Proc. Intern. Symp. Pro-
cessing Complex Ores. Pergamon, Oxford, 411-420.
[56] Fair, K.J. and G. Van Weert, edited by B. Harris. 1989b. Optimizing the NITROX PROCESS
through Elemental Sulfur Formation. Precious Metals 1989. The International Precious Metals
Institute, Montreal, 305-317.
[57] Fayram, T. and C.G. Anderson. October 2003. The Development and Implementation of In-
dustrial Hydrometallurgical Gallium and Germanium Recovery. Hydrometallurgy 2003, Van-
couver, B.C.
[58] Fedulov, O.V. et al. 1966. Oxidation of Molybdenite by Nitric Acid Solutions. Sb. Statei
Aspir. Soiskatelei, Min. Vyssh. Sredn. Spets. Obrazov Kaz. SSR, Met. Obogashch. (2) 86-94,
Chemical Abstracts 69,3260 In.
[59] Fedulov, O.V. et al. 1967a. End Products of Nitric Acid Decomposition of Molybdenite. Sb.
Statei Aspir. Soiskatelei Min. Vyssh. Sredn. Spets. Obrazov Kaz. SSR, Met. Obogashch. (3)
170-177, Chemical Abstracts 70,89848ym.
[60] Fedulov, O.V. and V.D. Ponomarev. 1967b. Nitric Acid Decomposition Kinetics of
Molybdenite. Sb. Stratei Aspir. Soiskatelei. Min. Vyssh. Sredn. Spets. Obrazov, Kaz. SSR.
Met. Obogashch. (3), 129-139, Chemical Abstracts 70,9llSOg.
[61] Fedulov, O.V. and V.D. Ponomarev. 1967c. Oxidation Process Mechanism of Molybdenite by
Aqueous Nitric Acid Solutions. Sb. Statei Aspir. Soiskatelei Min. Vyssh. Sredn. Spets.
Obrazov. Kaz. SSR, Met. Obogashch. (3), 12208, Chemical Abstracts 70,91182v.
[62] Fossi, P., L. Gandon, C. Bozec, and J.M. Demarthe. July 1977. Refining of High-Nickel
Concentrates. CIM Bull. 188-196.
[63] Freeport McMoRan Copper and Gold, 2010, Bagdad Mine Molybdenum Concentrate Pressure
Oxidation, Personal Communication.
[64] Gok, O.S., 2009. On The Role of Low-Concentration Nitrite In Oxidative Leaching with
Oxygen, PhD Thesis, Colorado School of Mines, USA.
Anderson, Fayram, Twidwell
Proceedings of Copper 2010 2618
[65] Habashi, F. 1973a. The Action of Nitric Acid on Chalcopyrite. Trans. Soc. Min. Eng. AIME
254, 224-228.
[66] Habashi, F. 1973b. Treatment of a Low-Grade Nickel Copper Sulfide Concentrate by Nitric
Acid. Trans. Soc. Min. Eng. AIME 254, 228-230.
[67] Habashi, F., edited by B. Mishra. 1999. Nitric Acid in the Hydrometallurgy of Sulfides. EPD
Congress 1999. TMS-AIME, Warrendale, PA, 357-364.
[68] Haver, F.P., and Wong, M.M., 1972, Making Copper Without Pollution, Mining Engineering,
Littleton, Colorado, June, 52-53.
[69] Heckner, H.N., 1973, Potentiostatic switching experiments for the cathodic reduction of
nitrous acid in perchloric acid with te addition of nitric acid. Electroanalytcal Chemistry and
Intefacial Electrochemistry. 44: 9-20.
[70] Himmelblau, D.M., 1960, Solubilities of inert gases in water 0O C to near the critical point of
water, Journal of Chemical and Engineering Data, 5: 10-15.
[71] Huffman, R.E. and Davidson, N.J., 1956, Kinetics of the ferrous iron-oxygen reaction in
sulfuric acid solution. Amer. Chem. Soc. 78: 4836.
[72] Huang, H.H. 2002. StabCal Modeling Software, September.
[73] Jeffrey, M. and C.G. Anderson. 2002. A Fundamental Study of the Alkaline Sulfide Leaching
of Gold. The European Journal of Mineral Processing and Environmental Protection, October.
[74] John, C.I.A., R.C. Sathe, and V.S. Kasongamulilo. 1991. Improving Flotation Performance at
the Nchanga Concentrator of Zambia Consolidated Copper Mines Limited. Copper 91/Cobre
91, Volume 3, Mineral Processing and Process Control. Pergamon Press, New York, NY, 19-
33.
[75] Joseph, T.B. 1916. Metal Leaching Process. Canadian Patent 173,452. Chemical Abstracts
12,130.
[76] Kerfoot, D.G.E. and R.W. Stanley. 1976. Hydrometallurgical Production of Technical Grade
Molybdic Oxide from Molybdenite. U.S. Patent 3,988,418.
[77] Kingsley, G.E. 1915. Process for Treating Complex Sulfide Ores. US Patent 1,144,480.
[78] Krysa, B., B. Barlin, and D. Wittleton. May 1988. The Application of Zinc Pressure Leaching
at the Hudson Bay Mining and Smelting Co. Limited. Projects 88, Paper #8. 18th
Hydrometallurgical Meeting CIM.
[79] Kunda, W. 1984a. Hydrometallurgical Processing of Silver Concentrate. Precious Metals:
Mining, Extraction, and Processing. Proc. Joint Symp. AIME-IMS/IPMI, 397-423.
[80] Kunda, W., edited by V. Kudryk et al. 1984b. Hydrometallurgical Processing of Silver
Concentrate. Precious Metals: Mining, Extraction, and Processing, Proc. Joint Symp. AIME-
TMS/IPMI, 397-423.
Industrial NSC Pressure Oxidation of Combined Copper and Molybdenum Concentrates
Proceedings of Copper 2010 2619
[81] Lane, J.W., Bender, F.N., and Ronzio, R.A., 1972, Recovery of Molybdenum from Oxidized
Ore at Climax, Colorado, AIME Transactions Volume 252, 77-82..
[82] Le Nickel. 1968. Recovery of Nickel and Other Secondary Metals from Nickel Mattes. French
Patent 1,597,569.
[83] Lunt, R.R., D.K. Modrow and G.K. Roset. October 2003. Adaption of Dilute Mold Lime Dual
Alkali Scrubbing at Stillwater Mining Companys PGM Smelter. Hydrometallurgy 2003,
Vancouver, B.C.
[84] Mahmoud, M.H., C.G. Anderson, and C.A. Young. June 2002. Sulfuric Acid-Chloride
Leaching of Platinum, Palladium and Rhodium From Catalyst Residue. Recycling and Waste
Treatment in Mineral and Metal Processing. TMS Fall 2002 Recycling and Waste Treatment
Meeting, Lulea, Sweden.
[85] Markovits, G.Y., Schwartz, S.E., and Newman, L., 1981, Hydrolysis equilibrium of dinitrogen
trioxide in dilute acid solution. Inorganic Chemistry. 20: 445-450.
[86] Molybdenum Corporation of America (Molycorp). 1973. Solvent Extraction Process for
Recovery of Molybdenum and Rhenium from Molybdenite. U.S. Patent 3,751,555.
[87] Mulak, W. 1987. Silver Ion Catalysis in Nitric Acid Dissolution of Ni3S2. Hydrometallurgy
18, 195-205.
[88] Ouellet, R., A.E. Torma, and J.P. Bolduc. 1975. Extraction du Nickel d'un Concentre de
Pentlandite par HNO3 - H2SO4. Can. Met. Quart. 14(4):339-343.
[89] Olsen, D., 2008, Kennecott to invest $ 270 million in molybdenum processing facility, The
Enterprise.
[90] Pauling, E.. 1940, Nassmetallurgische Aufarbeitung schwer aufzubereitender Komplexerze.
Metall und Erz 35, 451-455.
[91] Park, J.Y. and Lee, Y.N. 11988. Solubility and Decomposition Kinetics of Nitrous Acid in
Aqueous Solution. J. Phys. Chem. 92: 6294-6302.
[92] Peters, E. 1976. Direct Leaching of Sulfides: Chemistry and Applications. Metall. Trans., 7B,
505.
[93] Peters, E. 1992. Hydrometallurgical Process Innovation. Hydrometallurgy 29, 431-459.
[94] Prasad, M.S. 1991. Concentration Capacity in Treating Copper-Cobalt and Copper-Zinc Ores
at Gecamines. Zaire. Mining Engineering, 43(1): 129-133.
[95] Prater, J.C., P.B. Queneau, and T.J. Hudson. 1973. Nitric Acid Route to Processing Copper
Concentrates. Trans. Soc. Min. Eng. AIME 254(2):117-122.
[96] Queneau, P.B. and Prater, J.D., 1974, Nitric acid process for recovering metal values from
sulfide ore materials containing iron sulfides. US Patent 3793429, Chemical Abstracts.
Anderson, Fayram, Twidwell
Proceedings of Copper 2010 2620
[97] Rankin, H.D. 1915. Method of Treating Metalliferous Materials and Recovering Solvents
Used. US Patent 1,150,787.
[98] Shukla, P.P., T.K. Mukherjee, and C.K. Gupta. 1978. A Nitric Acid Route for Processing a
Nickel-Copper Sulfide Concentrate. Hydrometallurgy 3. 55-56.
[99] Stumpf, A. and Berube, Y. 1973, Aqueous Oxidation of Molybdenite in Chalcopyrite
Concentrates, Transactions of AIME, Vol 254, December, 305-309.
[100] Topol, L.E., Osteryoung, R.A. and Christie, J.H. 1965. Electrochemical Studies of NO+ and
NO2+ in Concentrated H2SO4 (1965). 112: 861-864.
[101] Turney, T.A. and Wright, G.A. 1959. Nitrous Acid and Nitrosation. Chem. Rev. 59: 497-513.
[102] Van Weert, G., K.J. Fair, and J.C. Schneider. 1986a. Prochem's Nitrox Process. CIM Bull. 79,
84-85, November.
[103] Van Weert, G., K.J. Fair, and J.C. Schneider. 1986b. The Nitrox Process for Treating Gold-
Bearing Arsenopyrites. Annual Meeting TSM-AIME, Denver, Colorado.
[104] Van Weert, G. 1988a. An Update on the Nitrox Process. Randol Gold Forum, 209-210.
[105] Van Weert, G., K..J. Fair, and V.H. Aprahamian. 1988b. Design and Operating Results of the
Nitrox Process. Gold Mining 88, Society of Mining Engineers of AIME, Littleton, Colorado,
286-302.
[106] Van Weert, G. and K.J. Fair. 1989. Capital and Operating Costs of the Nitrox Process for
Auriferous Arsenopyrites. Extraction Metallurgy '89. Institution of Mining & Metallurgy,
London.
[107] Vizsolyi and E. Peters. 1980. Nitric Acid Leaching of Molybdenite Concentrates.
Hydrometallurgy 6, 103-119.
[108] Weber, F.W. 1925. Silver from Sulfides. US Patent 1,555,615. Chemical Abstracts 193,473.
[109] Westby, G.C. 1918a. Nitric Acid and Copper Ores. Metallurgical & Chemical Engineering.
18(6):290-296.
[110] Westby, G.C. 1918b. Treating Sulfide Ores of Copper, Zinc, or Other Metals. US Patent
1,244,811, Chemical Abstracts 12,131.
[111] Yurkevich, Y.N. and K.Y. Shapiro. 1967. Decomposition of Molybdenite by Nitric Acid. Met.
Vol'frama, Molibdena, Niobiya, 53-56. Chemical Abstracts 69,98546e.
[112] Zelikman, A.N., L.V. Belyaevskaya, and T.E. Prosenkova, 1969, Decomposition of
Molybdenite by Nitric Acid, lzvestiya Vysshikh Uchebnykh, Zavedenii, Tsvetnaya
Metallurgiya 12 (6) 43-48, Chemical Abstracts 72,81734d.
Proceedings of Copper 2010 2621
HPGRs in Copper Ore Comminution
A Technology Broke Barriers
E. Burchardt, N. Patzelt, J. Knecht, R. Klymowsky
Polysius AG
Graf-Galen-Strae 17
Beckum, Germany
Keywords: HPGR, grinding, minerals references, minerals applications
Abstract
The growth of High Pressure Grinding Rolls has been steady and relentless since their introduction
into the diamond industry in 1986. The fastest growth was in iron ores, but the breakthrough came
with the first installations of HPGRs in the copper industry at the Freeports Cerro Verde Mine in
Peru (2006) and Grasberg Mine in Indonesia (2007). The recent fall in commodity prices in 2008
has temporarily interrupted this growth, but has not dampened the interest in the technology.
Many of the newest copper and gold deposits are large, low-grade and hard. Large tonnages and
especially hard rock favour HPGRs. Typically HPGRs offer the lowest operating costs. However,
HPGRs are also able to reduce investment cost for very hard ores.
HPGR technology broke barriers in a two fold way. Starting as a newcomer some years ago, it has
established itself as a matured and accepted technology in the precious and base metals industry. In
addition, HPGRs have proven their capability of being the key to make some projects economical
viable at all.
This paper reviews some further projects that have come on-stream since Cerro Verde and Gras-
berg:
Boddington Gold Mines, July 2009, 35 million t/a; Anglo Platinums Mogalakwena, March 2008,
600 ktpm; Assmangs BKM Iron Ore Project, July 2008; Northam Platinums UG2 plant, May
2008.
The first commercial HPGR for heap leaching in gold started in Goldfields Tarkwa Mine in Ghana,
October 2009.
The paper includes some of the lessons learnt from operating the rolls at various plants and reviews
on-going development programmes that have been initiated to improve wear life of the tyres, reduce
energy consumption and overall operating costs.
Burchardt, Patzelt, Knecht, Klymowsky
Proceedings of Copper 2010 2622
In addition, the paper provides an outlook on how HPGR technology may further advance the
Concentrator of the Future.
There are signs of recovery in several commodities. This gives confidence that when the recovery
takes hold, HPGRs will be in a strong position to secure a leading role in mineral comminution.
1 Introduction
In the middle of the 80
th
, HPGR (High Pressure Grinding Roll) technology was firstly adopted by
the cement industry due to the high potential for energy savings. Energy requirements for grinding
cement clinker, slag and limestone were in the order of 50 % lower than those consumed in conven-
tional ball mills.
Nevertheless, the minerals industry outside of diamonds and iron ore was initially very reluctant to
take over this technology due to concerns about wear and consequently availability.
It took a lot of effort to improve the design of HPGRs to provide acceptable lifetimes of the wear
components for high wear applications and to make these units maintenance friendly enough to
reach availabilities of more than 94 %. These were the preconditions for the final acceptance of
HPGRs in the base and precious metals industry. The wear issue has been successfully resolved,
although there will be a continuous development process in the next couple of years, to further
reduce wear cost. After the first successful installations in copper and other minerals applications,
todays focus is more on issues like:
Under which conditions are HPGR circuits more economical than conventional circuits?
Are HPGRs able to make low grade and high cost deposits viable from an economical standpoint?
What are the most economical HPGR based flow sheet configurations and plant layouts?
How to match highest grinding circuit energy with lowest investment cost?
What are the optimum interfaces between secondary crushing, high pressure grinding and ball
milling in terms of transfer size and energy distribution?
What has to be done to reduce investment cost of HPGR based flow sheets and obtain a further wear
and energy cost reduction?
Which lessons can be learnt from major installations in terms of future sizing and design criteria,
optimum plant layout, as well as best operating and maintenance practices?
This paper addresses at least some of these issues, providing a basic overview on to the current
status of this technology.
HPGRs in Copper Ore Comminution
Proceedings of Copper 2010 2623
2 Grinding Circuits Incorporating HPGRs
The first serious approach of HPGR technology to the copper industry was the installation at Cyprus
Sierrita in the Middle of the 90
th
. At that time, this installation was considered as a failure despite a
lot of positive achievements [1, 2]. The first successful installations of HPGRs in the copper
industry started at the Freeports Cerro Verde Mine in Peru 2006 [3] and Grasberg Mine in
Indonesia 2007 [4].
Major applications for HPGRs in copper and other minerals are
for pebble crushing in SAG mill circuits of concentrators (greenfield, brownfield) (Figure 1),
for tertiary crushing in concentrators (greenfield) (Figure 2),
for quaternary crushing in concentrators (brownfield) (Figure 4),
for tertiary and / or quaternary crushing in heap leach operations (greenfield, brownfield).
The potential of HPGR for pebble crushing has been studied for a long time. A large variety of pos-
sible flow sheets have been investigated. Figure 1 includes two of these concepts. The most efficient
way to implement HPGRs is to treat the pebbles separately in a secondary crushing (SC) HPGR
circuit directly feeding the screen undersize product to the ball mills. HPGRs can be retrofitted effi-
ciently into existing SAG/AG mill circuits (brownfield) for capacity increases. First HPGRs for
pebble or SAG oversize crushing were applied in iron ore applications in the 90
th
. In 2008, a gold
producer [5] decided to build a SAG mill based concentrator with a HPGR for pebble crushing
(Figure 1: SAG Option 2).
Figure 1: HPGR re-grinding crushed pebbles (left); SC/HPGR circuit for pebble crushing (right)
HPGR based circuits for new concentrators (greenfield) would typically use HPGRs in closed
circuit for tertiary crushing, either with wet or dry screening. The HPGR circuit is normally fed with
a full feed in a normal closed circuit (Figure 2 left hand) or with a truncated feed in a reversed
closed circuit (Figure 2 right hand). There have been objections in the gold/copper industry
Burchardt, Patzelt, Knecht, Klymowsky
Proceedings of Copper 2010 2624
against the reversed closed circuit mainly due to wear issues although the truncation of the feed is
standard practice in diamonds. Plants in operation like Freeport Cerro Verde (copper since
2006), Anglo Platinum Mogalakwena (platinum since 2008) and Newmont - Boddington (gold
since 2009) are based on a normal closed circuit with wet screening (Figure 2 left hand). So far,
only Anglo Platinum has opted for a dry screening plant [6, 7].
Figure 2: HPGR in tertiary crushing duty normal closed circuit (left); reversed closed circuit (right)
Figure 3: HPGRs at Cerro Verde (Freeport) first concentrator in copper fully HPGR based [4]
Other operations to be commissioned in the future, like Vale Salobo, MolyMines Spinnefex
Ridge and Gindalbie Metals - Mt. Karara are based on a full feed, wet screening circuit (Figure 2
left hand).
A reversed HPGR circuit with wet screening was selected for the BKM Iron Ore Project of
Assmang which went into operation in July 2008. Major reason for this decision was the need to
minimise fines generation which are production losses in this application. Any concerns about
HPGRs in Copper Ore Comminution
Proceedings of Copper 2010 2625
accelerated wear due the missing fines in the feed have not been confirmed. On-going feed back
from this application may be the basis for a re-thinking of the wide spread objections against feed
truncation in copper/gold or any other hard rock applications - especially in low wear applications.
HPGRs have proven themselves as powerful tools for plant up-grades in existing crushing and
grinding circuits (brownfield). First successful installation as a quaternary crushing stage has been
in the Grasberg Mine of Freeport in Indonesia (gold/copper since 2007) [Figure 4].
Figure 4: HPGR application at Grasberg (Freeport) first one as a quaternary stage in copper
HPGRs at Grasberg [4] are receiving the product of a 3-stage crushing plant. Re-grinding of this
relative fine feed of minus 15 mm results in a finer ball mill feed. This provides the opportunity to
increase the plant throughput at same flotation fineness and recovery or to keep the throughput at
the same level and increase flotation fineness and consequently recovery. Both cases increase the
production value, often at lower overall operating cost.
Reduction of operating cost by replacing existing tertiary cone crushers or rod mills without any
throughput or recovery increase does not justify a HPGR retrofit typically.
A HPGR in similar application has been operated at Northam Platinum on a UG2 ore since 2008,
re-grinding a fine bi-modal feed from a crushing plant. The HPGR application allowed the
installation of an intermediate flash flotation circuit and the conversion of the original rod mill to a
ball mill. The complete re-design project has been reported as an overwhelming success [7].
In October 2009, the first commercial HPGR for a heap leach operation in gold (Goldfields
Tarkwa Mine, Ghana) has been commissioned. This installation is considered as the first full scale
industrial test in order to prove the general suitability of HPGRs for heap leaching as well as the
achievement of faster and higher recoveries of HPGR treated ores, which have been found in the lab
again and again. Positive results from this operation could also influence the SX/EW heap leach
operations in the copper industry.
Burchardt, Patzelt, Knecht, Klymowsky
Proceedings of Copper 2010 2626
3 Industrial Operating Experience
Basic design data of recent copper and minerals installations are summarised in Figure 5. Operating
experience gathered from those installations have provided a large basis for design modifications
and improvements, wear life improvements, verification of scale-up and sizing criteria as well as for
the evolution of best operating and maintenance practices. New projects as well as projects coming
on stream in the near future will be able to benefit from the experience continuously collected.
Figure 5: Basic design & operating data from recent applications on copper and other mineral ores
Major findings from these installations are summarised in the following.
In general, sizing of the early minerals HPGRs in terms of throughput, required circulating loads
and energy input in closed circuit operation as well as the lifetime predictions have been quite con-
servative initially due to a lack of industrial references in similar applications.
Typically, lifetimes of industrial HPGRs rolls have been higher than the conservatively provided
figures for warranties. HPGR wear testing (ATWAL abrasion index) gave a realistic basis for
lifetime projections. Maximum achieved lifetime has been more than 20,000 hours in a copper in-
stallation on finer feed. There has been only one case where the predicted lifetime (6000 hours) was
significantly lower than the lifetime initially reached (3000 hours). This difference was mainly at-
tributable to changes of ore properties (higher ATWAL abrasion index) compared to the design
phase and in addition to instantaneously higher than design capacities. An optimisation program on
the wear protection tyres, improving stud design and grade, resulted in an increase of the lifetime to
recently more than 5000 hours with a potential to finally exceed 6000 hours with the next design
change. Other high wear operations will benefit from these developments.
HPGRs in Copper Ore Comminution
Proceedings of Copper 2010 2627
The major driver for abrasion on HPGR rolls is the mineralogy of the ore to be treated. Typically the
mineralogy has a higher influence on the wear rate than the feed size, grinding pressure and mois-
ture content. The second dominant factor for the wear rate is the feed size or feed size to gap ratio
respectively. A feed size, significantly smaller than the gap, reduces the wear rate. The hardness or
competency of an ore has only a minor impact on the wear rate according to ATWAL abrasion test-
ing and subsequent industrial experience.
Top feed size control to HPGR circuits is essential for long roll life since this is a precondition to
minimise the risk of stud breakage. Competent ores have a higher inherent risk of breaking studs.
Excessive oversize resulted in severe stud breakage and finally resulted in surface damage in one
application. A reduction or limitation of the feed size to the HPGRS treating very competent ore
would increase tyre lifetime and is a precondition to apply stud grades with improved wear resis-
tance on those ores in the future
Roll unit change outs were conducted within less than 36 hours after establishing optimised working
procedures.
Initial problems with roll skewing in some applications were successfully resolved. In general, roll
skewing is caused by insufficient feed rates and/or segregation in the feed. In some applications, roll
skewing was mainly caused by insufficient feed rates during feed material start-ups as well as in
continuous operation resulting in feeding only to the roll centre and preventing the build up of
choke feed conditions. Non choke feed conditions could be overcome by adjusting the roll speed to
the actual available feed quantities in normal operation and a ramp up of the roll speed during feed
material start-ups.
The specific throughput rates of industrial HPGRs have been usually higher than those obtained in
batch type pilot test units, which were the basis for initial sizing. Consequently the achieved
throughput capacities of industrial units often exceeded the conservative design figures.
Circulating loads in closed circuit operation with screens were found to be on the levels as predicted
by the supplier. Initial concerns especially of engineers and consultants about supplier predicted
circulating loads seem to vanish and the tendency to oversize HPGR circuits may be brought down
to reasonable levels. A common understanding of the supplier, engineer and consultant about realis-
tic circulating loads in HPGR circuits as well as of the achievable throughput rates would provide
an opportunity to reduce investment cost.
Oversizing of HPGRs due to underestimating actually achievable capacities and overestimating re-
quired capacities created sometimes operational problems due to a lack of feed material - especially
in case of fixed speed drives. Variable speed HPGR drives should be installed for proper process
control and for the handling of any fluctuations in required throughput.
Variation of the grinding pressure turned out to be a powerful approach to either optimise a HPGR
circuit in terms of maximum throughput at lower circulating loads or maximum energy efficiency at
higher circulating loads.
Burchardt, Patzelt, Knecht, Klymowsky
Proceedings of Copper 2010 2628
Cake formation in the HPGR discharge material was found to be less critical than initially feared.
Current cake stability testing provides a good assessment of cake competency. Screen efficiency
was higher in wet than in dry screening. Moisture control of the wet screen oversize could be man-
aged even for screening at 5 to 6 mm. Efficient pulping of the screen feed ahead of the screens,
proper material distribution along the screen width as well as sufficient screening area are precondi-
tions to limit the moisture content in the screen oversize (S/O) to reasonable levels. Typically, the
moisture content of a plus 6 mm S/O is less than 4 % under proper operating conditions. These low
moisture levels may give room to a further reduction of the cut size between HPGR and ball mill
circuit.
Under good operating conditions, specific energy input for producing < 6 mm copper ore product
was lower than 3 kWh/t referred to circuit product.
HPGR performance was found to be quite insensitive against changes in feed material properties.
Even treating a variable ore body, specific throughput, circulating load, power absorption and spe-
cific energy input did not vary significantly. Ball mill feed size distribution was very consistent and
mainly determined by the mesh size of the wet screens rather than influenced by material properties
like competency. This feature allows a smooth ball mill operation at nearly constant throughput and
product fineness and subsequently a controlled operation of the flotation circuit. Throughput varia-
tions required due to operational requirements or caused by minor feed characteristic changes were
easily managed by roll speed adjustments.
SAG mill circuit performance is far more feed material dependent. Feed size distribution and ore
competency has a much larger impact on SAG mill throughput and the transfer size between the
SAG and ball mill circuit. In summary, a consistent circuit balance is much easier to achieve be-
tween a HPGR and a ball mill circuit
4 HPGR vs. SAG Mill Circuits
HPGR based circuits usually compete with traditional SAG mill circuits, although there seem to be
some recent signs of a revival of conventional 3-stage crushing circuits especially on competent
ores at smaller scale operations.
Large scale concentrators with a capacity of 100,000 tpd would require 5 to 6 secondary crushers
and about 10 to 15 tertiary crushers depending on the ore competency and targeted product fineness.
Lifetime of secondary crusher liners can be less than 30 days in high wear applications with an even
shorter lifetime to be expected in tertiary duty. This maintenance intensity besides the high capital
and operating cost usually avoids that conventional crushing circuits are considered for world class
projects.
Trade off studies on larger scale operations compare SAG and HPGR concepts in most cases. The
economical differences between a conventional SABC circuit and a HPGR based circuit are mainly
HPGRs in Copper Ore Comminution
Proceedings of Copper 2010 2629
determined by the behaviour of the applicable ore(s) in a SAG mill. Originally, SAG mill circuits
were sized on the basis of the Bond Theory using the Bond work index from test work. However,
the Bond Theory frequently failed on SAG mills in the past.
In recent years, SAG mills have been sized on either on the drop weight (DWI) test or the SPI test.
Outcome of the drop weight test is the so-called A*b parameter or the DWI of an ore.
Newmont (Veilette & Parker, 2005) [8] published a diagram covering the relation between the re-
quired SAG mill energy input and the ore competency expressed by the A*b parameter for a number
of their operating plants and projects [Figure 6]. The conclusion from this diagram is that the energy
requirements in a SAG mill may range from something like 4 kWh/t (for A*B = 80) to 16 kWh/t
(for A*b = 25) or in other words, the behaviour of individual ores could differ dramatically in SAG
mills even in one deposit.
Figure 6: Newmont database of SAG power as a function of ore competency A* b parameter [8]
A significant relation between the A*b parameter of an ore and the Bond work index (propor-
tional to the ball mill energy requirements) or breakage characteristic in a HPGR could not be
found.
Figure 7 includes a diagram (left hand side), showing the fineness vs. grinding force relation in a
pilot scale HPGR for ores with A*b parameters between 18 and 36. In fact, no significant differ-
ence in product fineness was found. This result suggests that the circulating load in a HPGR circuit
would not vary significantly even for ores within a wide range of A*b parameters or competency
respectively.
The correlation between the Bond work index and the A*b parameter of more than 30 ores is
shown in a second diagram (right hand side) included in Figure 7.
Burchardt, Patzelt, Knecht, Klymowsky
Proceedings of Copper 2010 2630
The Bond index varied only between 16 2 kWh/t even though the A*b parameters were in the
range of between 18 and 75 indicating SAG mill energy requirements between about 5 kWh/t and
more than 16 kWh/t.
Figure 7: HPGR fineness vs. grinding force (left), Bond Index vs. A*b ore competency (right)
The conclusions from these findings are that
variation of ore competency is mainly an issue in SAG mills,
ore competency is not a critical issue once ore particles are smaller than the feed size (< 3.15 mm)
in Bond testing, and
HPGRs are able to efficiently equalize variations in ore competency (The Big Equalizer).
The implications of ore competency on the specific circuit energy as well as on required equipment
sizes and numbers are summarised in Figure 8. This comparison - on a very general desktop level -
compares three different ores with 4, 10 and 16 kWh/t energy input in a SAG mill and an equivalent
HPGR circuit producing a minus 6 mm ball mill feed. Any variation of the HPGR performance
treating these different ores is ignored due to the minor influence on the results. The ball mill energy
is mainly a function of the feed and required product fineness which may significantly vary for
different ore bodies and was therefore ignored for simplification. Ball mill energy may be typically
up to 10 % higher in the HPGR case depending on the required ball mill fineness. Nevertheless,
such a correction will not change the general conclusions from Figure 8.
HPGRs in Copper Ore Comminution
Proceedings of Copper 2010 2631
Figure 8: Desktop study SAG energy for ores of different competency vs. HPGR circuit energy
The conclusions from the desktop study are that a 100,000 tpd plant
can be built with only one large SAG mill provided the ore is low competent,
would require at least 2 SAG mills for medium competent ores and up to 4 SAG mills for highly
competent ores,
would require most typically 4 HPGRs independent of the ore competency,
could be built with 3 HPGRs in case of applying the largest POLYCOM models available and/or
considering a coarser transfer size to the ball mills than 6 mm as often used,
need to compare the installation cost of 2 to 3 SAG mills against the cost for 4 HPGR circuits in
case of medium competent ores for capital expenditure comparison,
need to compare the installation cost of 4 SAG mills (38) against the cost for 4 HPGR circuits
in case of extremely competent ores for capital expenditure comparison.
In general, it is concluded that hard and competent ores at large scale definitely favour HPGR based
circuits.
Wear cost, which are a major fraction of the operating cost, will be a function of the required energy
in the SAG and ball mill and will be dependent on the abrasiveness of the ore itself. Wear cost in
HPGRs are very much ore dependent. Wear cost for a SAG vs. HPGR comparison need to be evalu-
ated on a case by case basis.
5 Future Trends for HPGRs in Copper and for Other Minerals
A need for further reduction of total operating cost and in particular of the energy consumption in
tomorrows comminution circuits is obvious. This is a result of lower grade and harder ore bodies to
be treated in the future. Restriction of CO
2
emissions to avoid global warming has become an im-
Burchardt, Patzelt, Knecht, Klymowsky
Proceedings of Copper 2010 2632
portant issue on the world wide political agenda and may result in a limitation of available energy
(CO
2
certificates) or at least in higher energy cost than today. Only highly efficient comminution
concepts may make in particular low grade/hard ore projects feasible at all, keeping capital
(capex) and operating (opex) expenditures at economical levels. Finally, availability of water is be-
coming more and more of an issue.
The introduction of HPGRs as the tertiary crushing stage in hard rock applications was a first mile-
stone to achieve energy and wear cost reductions for energy intensive ore bodies. Nevertheless,
there is the demand to utilise the potential of HPGRs to a greater extent than today.
There are three major focus areas for optimisation of HPGR Ball Mill plants are:
to reduce the investment cost of secondary crushing - HPGR ball milling plants (SC-HPGR-BM)
which are percepted to be generally higher than those of traditional SAG Pebble Crushing - Ball
Milling (SABC) circuits,
to shift more energy from the ball mills and secondary crushers to the more efficient HPGRs in
order to improve the overall energy efficiency of comminution circuits,
to further reduce the wear cost in the crushing, HPGR and milling sections.
Trade off studies often resulted in operating cost advantages of HPGR based circuits at higher capi-
tal expenditure as compared to traditional SABC circuits. In fact, this perception is not valid in gen-
eral. The economical differences of both concepts are mainly determined by the energy require-
ments of an ore in SAG mills. The required number of SAG mills as already shown above may
range from one to four 20 MW mills for a 100,000 tpd plant whereas the number of HPGRs
would be four for the majority of ores. An investment cost disadvantage of HPGR circuits is ques-
tionable in case that four large SAG mills would be required. Mayor cost driver in HPGR based
circuits is the material handling equipment and, in particular, todays practice to operate the secon-
dary crushers and HPGRs in closed circuit with screens.
Open circuit secondary crushing and multiple pass, open circuit grinding in HPGRs is sometimes
viewed as a potential approach for investment cost reductions due to expected plant simplifications
and hopefully also operating cost reductions.
On the paper, open circuit secondary crushing might be a viable option if ideal operating conditions
could be permanently guaranteed on the crushers. However, open circuit operation would not pro-
vide any protection against oversize ore entering the HPGRs circuit. Such oversize material could
pass the crushers under certain conditions, especially after gap releases in overload situations. Per-
manent as well as temporary oversize feed would reduce the lifetime of the HPGR wear protection.
For a long time there was or even still is the perception that HPGRs should be fed as coarse as pos-
sible to release the secondary crushers and ensure highest circuit efficiency. Although the lifetime of
secondary crushers wear parts may be quite short in some applications, the wear cost for the secon-
dary crushing duty is still relatively low. Wear cost typically increases disproportional in each com-
HPGRs in Copper Ore Comminution
Proceedings of Copper 2010 2633
minution step towards the final product. From this point of view it should be more economical to
install additional secondary crushing capability rather than to tantalize the tertiary HPGRs with a
large or even oversized feed.
HPGRs would definitely benefit from a finer feed size. Firstly, the lifetime of the HPGR wear pro-
tection is significantly longer with finer feed and thus reducing wear cost. Secondly, a finer feed
could become the precondition to apply the next generation of stud grades. Those stud grates are
more wear resistant (longer lifetime) but also more brittle at the same time (higher risk of stud
breakage). Earlier recommendations with regard to the top feed size to gap ratio have to be re-
defined taking into account particularly the hardness or better ductility of the ore. The risk of stud
breakage has to be mitigated by matching ore abrasiveness, ore hardness, top feed size and applica-
ble stud grades. A comprehensive understanding of these interactions is the precondition to optimise
the interface between secondary crushing and HPGRs in order to ensure the application of the best
suited wear protection design for a particular application and to minimise wear cost.
Furthermore, a frequently discussed issue is: How to shift more energy from the ball mills to the
HPGRs while simplifying the plant at the same time? A distinction is to be made between two
principal approaches despite the fact that there is wide variety of possible flow sheet configurations
which may also combine the two said principles. The first option considered for efficiency im-
provements is to produce a finer HPGR circuit product by fine screening and accepting higher circu-
lating loads. Such an approach would not change the current plant design and layout significantly.
The second option is to use HPGRs in multiple pass, open circuit mode without intermediate classi-
fication. Supporters of this concept additionally hope to achieve a simplification of the plant result-
ing in lower investment cost. Further engineering studies are required to prove this expectation.
Four typical flow sheets deviating from the standard flow sheet are shown in Figure 9.
Burchardt, Patzelt, Knecht, Klymowsky
Proceedings of Copper 2010 2634
Figure 9: Alternative HPGR based circuits
Extensive testwork has been conducted using pilot scale HPGRs and lab scale ball mills to
investigate the difference of various flow sheet configurations in terms of overall energy efficiency
including the subsequent ball mill stage.
HPGRs in Copper Ore Comminution
Proceedings of Copper 2010 2635
The general conclusions from this testwork were that
screening at finer than current industrial screen sizes of 5 to 10 mm would improve the circuit
efficiency as long as the moisture content of the wet screen oversize could be maintained at
reasonable levels,
fines removal between subsequent HPGR stages is essential for highest efficiency,
any multiple pressing without prior fines removal resulted in reduced circuit efficiencies.
These findings are in line with industrial practise outside of minerals applications. Today, the Ce-
ment Industry still has the widest experience with regard to the most efficient utilisation of HPGRs.
Two important lessons can be learnt from the various cement applications. Firstly, HPGRs are 40 to
60 % more energy efficient than normal ball mills and secondly, high efficiency separation is the
key to actually obtain these potential energy savings. Any multiple pressing without intermediate
classification reduced the efficiency of the overall grinding circuits in cement. Based on these ex-
periences is obvious that highest energy savings in minerals would also be achievable if HPGRs
could eliminate the ball mills at all as several applications on cement clinker, blast furnace slag
and limestones have proven in the past.
So far, the largest HPGR finish grinding applications in limestone are currently operated in an In-
dian cement plant. Two grinding plants, each incorporating one POLYCOM 21/16, are treating a
limestone with a Bond work index in the range of from 15 to 16 kWh/t. Each grinding plant pro-
duces between 320 and 370 tph of product at a P80 of 65 to 90 m. The energy consumption for the
HPGR, fan, separator and bucket elevators is between 10.7 and 11.5 kWh/t depending on the prod-
uct fineness.
A Bond work index of 15 to 16 kWh/t is not unusual for typical copper ores. Assuming an identical
HPGR size and same grinding and classification efficiencies, the said Indian plant would have the
potential to grind in the order of 450 tph of copper ore to flotation fineness of 150 m and up to
650 tph with a larger POLYCOM 24/18. Finish grinding with a HPGR could reduce the energy re-
quirements by in the order of 40 % compared to HPGR/wet ball mill circuits as applied today.
Definitely, there are questions and challenges around such a dry finish grinding system for mineral
applications, like moisture, wear, maintenance, etc. Nevertheless, this would be the ultimate ap-
proach to minimise energy requirements for grinding ferrous and non ferrous ores.
The evolution to expand efficient HPGR utilisation in todays crushing and milling circuits will be
an on-going development in the near future. A revolution would be to apply HPGRs for finish
grinding.
Burchardt, Patzelt, Knecht, Klymowsky
Proceedings of Copper 2010 2636
6 Summary
HPGRs have established themselves as an accepted and well matured technology which is able to
significantly reduce the operating cost of copper concentrators and other minerals applications.
The barrier I do not want to be the first using a new technology has been broken successfully by
installations such as Cerro Verde, Grasberg, Boddington and Mogalakwena. Feasibility studies of
further world class projects in Chile and Pakistan are solely based on HPGR technology.
HPGRs broke the barrier for some recent and current projects to become viable at all from an eco-
nomical standpoint, especially in the case of competent, low grade ore bodies which would require
high energy inputs in conventional SAG mills.
Nevertheless, HPGRs are a relatively young technology compared to well known conventional
crushing and grinding circuits which have been operated for decades. There is a focus on opportuni-
ties to optimise the first generation of HPGR circuits in terms of operating and capital cost. This
will be an ongoing process in the next couple of years.
Application of HPGRs in heap leach installations as well as for finish grinding to flotation fineness
could be the next milestones in the copper industry.
References
[1] Patzelt N., J. Knecht, E. Burchardt, and R. Klymowsky: Challenges for High Pressure Grind-
ing in the New Millenium; 7
th
Mill Operators Conference 2000, Kalgoorlie, WA. 47-55.
[2] Patzelt N., R. Klymowsky, E. Burchardt, and J. Knecht: High Pressure Grinding Rolls in
AG/SAG Mill Circuits. Proceedings; SAG 2001, Vol. III. 107-123.
[3] J.L. Vanderbeek, T.B. Linde, W.S. Brack, and J.O. Marsden: HPGR Implementation at Cerro
Verde; SAG 2006, IV 45 61
[4] M. Mular, John Mosher: A Pre-production Review of PT Freeport Indonesias High Pressure
Grinding Roll Project; SAG 2006, IV 62 - 79
[5] S. Dixon, B. Olson, E. Wipf: Squeezing an extra 30% of a typical SABC circuit for 4.8 kWh/t;
SME 2010
[6] C. Rule: Development of a Process Flow Sheet for the New Anglo Platinum Concentrator In-
corporating HPGR Technolgy; SAG 2006, IV 94 109
[7] C. Rule, D. Minnaar, G. Sauermann: HPGR Revolution in Platinum; Third International
Platinum Conference, The Southern African Institute of Mining and Metallurgy, 2008.
[8] B. Parker, G. Veillette: Boddington Expansion Project; Randol Conference, Perth, 2005
Proceedings of Copper 2010 2637
Evaluation of Copper Losses in the Slag Cleaning
Circuits from Two Chilean Smelters
N. Cardona, L. Hernandez, E. Araneda, R. Parra L. Bahamondes, R. Parada
Universidad de Concepcin Chagres Smelter, Anglo American
Edmundo Larenas 285 Camino Troncal s/n
Concepcin, Chile Catemu, Chile
J. Vargas, M. Artigas
Paipote Smelter, ENAMI
Camino Pblico s/n
Copiap, Chile
Keywords: Pyrometallurgy, copper losses, slag cleaning
Abstract
The slags from two cleaning processes circuits were analyzed in order to evaluate the copper losses
and identify options for its control. The first is a bath smelting unit (Teniente Converter) with an
electric slag cleaning furnace (TC-EAF) and the second is a flash smelting (Outokumpu) with a
tilting furnace (FSF-SCF). A sampling campaign was done in each circuit during the furnace opera-
tions and two sampling devices were tested. Bulk chemical analyses of the slags samples were
combined with the characterization of the solid phases using optic microscopy and electron probe
microanalysis (EMPA). The quantification of phases and distribution of particle sizes were obtained
using Qemscan
and image analysis techniques with special attention to copper containing phases.
Characteristics phases as solid matrix, fayalite, spinels and sulfides in matte droplets were iden-
tified in all the slags. Differences appear on the proportions of these phases depending of the slag
chemical composition, oxidizing/reducing condition of the process and the sample cooling. Some
distinctive characteristics of the copper losses and the presence of Fe-spinel equiaxial grains were
used to explain the behavior of the matte droplets during the settling and to estimate the reducing
condition of the melt slag.
1 Introduction
The copper losses in slag are of chemical and mechanical nature; the chemically dissolved cationic
copper (Cu
+
) is stable in the slag and depends on the matte grade (oxygen-sulphur potential ratio),
temperature and slag composition [1, 2, 4]; the mechanical losses include entrapped or floated un-
settled matte drops and metallic inclusions. The copper content in these phases may be higher or
Cardona, Hernandez, Araneda, Parra, Bahamondes, Parada, Vargas, Artigas
Proceedings of Copper 2010 2638
lower than the furnace matte grade [3] in the smelting unit. The majority of copper in industrial
slags is found as mechanically entrained matte particles with sizes varying from several millimeters
to a few microns [2], its recuperation is accomplished in the slag cleaning process.
The pyrometallurgical slag cleaning processes are based on slag reduction and settling of matte par-
ticles. It uses a reductant (coke, graphite, oil or natural gas) for minimize the content of solid mag-
netite and improve the settling of matte droplets toward the matte phase. Many factors influence the
copper recovery and the quality of the final slag: Chemical composition, temperature and opera-
tional times, the nature of the copper losses and type of reductant and reduction mechanism. The
operational control is usually based in the analysis of total copper content and the magnetite content
by the Satmagan balance in the final slag. This information is limited to establish the nature and
distribution of the copper losses in the different slags and to evaluate the effect of the slag composi-
tion in the copper recovery.
In order to evaluate the slag quality in the cleaning circuits a complete characterization of the slag
currents should be done, obtaining the final copper losses distribution that can be related to the con-
trol process parameters and chemical characteristics. The knowledge of the chemical and minera-
logical composition of the solidified samples allows an approximation of the liquid condition of the
slag from a thermodynamic description of the equilibrium between theirs components. In this per-
spective we have developed a methodology for the systematic study of copper losses based in the
bulk chemical and mineralogical characterization of phases in slag samples obtained during the tap-
ping and from the bath into the furnaces. The characterization and sizes distribution of slag phases
associated with copper losses were obtained using analysis on mirror polished surfaces. This paper
presents the results of this analysis on slags from cleaning circuit of two Chilean smelters: The Her-
nan Videla Lira (HVL) Smelter and The Chagres Smelter.
1.1 HVL Slag Cleaning Circuit (TC-EAF)
The HVL smelter of ENAMI (Empresa Nacional de Minera) is situated near Copiap city in Chile
and has 340 kt of concentrate processing capacity. The process flow sheet is showed at Figure 1. In
2008 the HVL smelter generated 184.5 kt of dump slag with 0.8 % of average copper content. A
Teniente Converter (TC) operates as a unique smelting unit, it produces highly oxidized slag (high
magnetite and dissolved copper contents) that are cleaned in an electric arc furnace (EAF) with ca-
pacity for the treatment of 720 t of slag and 120 t of solid reverts. The TC slags is tapped each 0.8-
1 h and transported by ladles to the electric arc furnace (EAF), then is transferred by launder and
charged through the furnace wall. The tapping time is 0.5 h aproximately and produces 4 ladles of
15 t each.
Evaluation of Copper Losses in Slag Cleaning Circuits from Two Chilean Smelters
Proceedings of Copper 2010 2639
Figure 1: Current flow sheet of the Hernn Videla Lira Smelter [7].
The EAF furnace operates in semi-continuous mode. It uses solid coke as reductant (12-14 kg/ton of
slag) and the reduction occur both on the surface of the carbon electrodes and at the slag/floating coke
interface. The average consumption of energy is 134 kWh/ton (slag +reverts) and the temperature of
molten phases is between 1230-1270 C. The maximum level of molten bath is 1.7 m. The copper
matte from the EAF is tapped (fourth times in turn) and returned to the TC. The final slag is tapped
from the EAF, it flows by a launder to charge 5 or 6 ladles of 6 ton each and finally is transported
toward the dump yard.
1.2 Chagres Slag Cleaning Circuit (FSF-SCF)
The Chagres smelter, part of the Anglo Base Metals Division, Anglo American Plc is situated
80 km north of Santiago in Chile and has 600 kta of concentrate processing capacity. In 2008 pro-
duced 315 kt of slag with 1.1 % Cu content. Process flow sheet is showed in Figure 2. The smelter
unit is an Outokumpu-Flash Furnace (FSF) and operates with instantaneous rate of 78-80 t/h dry
feed. The slag is tapped each 1.5 h after the matte tapping and flows through a refrigerated launder
to one of the Teniente slag cleaning furnace (SCF), tapping time varies from 0.3 to 0.5 h.
Escoria
Inyeccin
Conc. Seco
Metal
Blanco
Anodos
Inyeccin
Mezcla CPS
Plantas de Acido
Sulfrico
Secador Secador
Escorias a Botadero Escorias a Botadero
CT
CPS CPS
RAF RAF
Gases
Gases
Blister
POX POX
HELE HELE HELE
Precipitador
Electrosttico
Gases
Metal
Blanco
Circulantes Circulantes
Precipitador
Electrosttico
Gases
Recepcin Recepcin
Mezcla RAM Mezcla RAM
Gases
Polvos HE
CNU - slice
Cardona, Hernandez, Araneda, Parra, Bahamondes, Parada, Vargas, Artigas
Proceedings of Copper 2010 2640
The two slag cleaning furnaces with 90 slag capacity each operates in discontinuous mode with 1.3-
1.5 h per cycle. The converter slag is charged with ladles through the mouth before the FSF slag
was charged. The reduction step uses fuel oil (API Grade 6) at a rate of 6.6-7 l/min with an air ra-
tion of 1.6-1.7 Nm
3
air/l fuel oil through two tuyers during 0.5-0.7 h. The reduction step is followed
by the slag settling for 0.7-1 h. The final slag is skimmed through the mouth in two ladles of 33 ton
each and transported to the dump yard. The temperatures are between the 1230-1250 C and there is
a burner air-oil for thermal support. The metal is tapped each 4-5 cleaning cycles and returned to the
Pierce Smith converter in the Fe-blowing stage.
Figure 2: Current flow sheet of the Chagres Smelter [6].
RECEPCION
TOLVAS
ESCORIA
FINAL
ESCORIA
FLASH
POLVO
Concentrado Seco
R. Radiacin R. Conveccin R. Lurgi
EJE ESCORIA CONVERSION
Bomba Moller
Cobre
Blister
Gases Combustin, Quemador
y Reduccin
ANODOS
BLISTER
Camara de Mezcla
VTI Conversin
VTI Flash
C Rad
4
CPS 4
CPS 3
CPS 2
Lurgi
Horno Flash
C Rad
3
C Rad
2
HLE 2
HR 2
HR 1
Rueda
de
Moldeo
TOLVA
400
Bronces
Botadero Conchos
Fondos
Botadero Escoria Descarte
PLANTA DE
ACIDO
HLE 2
Tolva
50
Secador 1
Secador 2
Soldado +
Terceros Silice
Caldera
Recuperadora
POLVO
CPS 1
C Rad
1
Evaluation of Copper Losses in Slag Cleaning Circuits from Two Chilean Smelters
Proceedings of Copper 2010 2641
2 Experimental
2.1 Slag sampling campaigns
The slags from TC-EAF circuit were sampling during three consecutive TC tapping following by
EAF tapping in a period of 6 h. Two samples of the TC slag (1
st
and 3
rd
ladles) were taken before
the charge to the EAF and two samples (1
st
and 3
rd
ladles) of the corresponding final slag were tak-
ing at the end of the cleaning cycle. The furnaces operational conditions are showed in the Table 1.
The samples were taking with a steel ladle of 500 cm
3
and air cooled.
Table 1: TC-EAF parameters
Parameters/turn [8 h] TC EAF
Dry concentrate [t/h] 42.3
Enrichment oxygen [%] 37
Reverts charge [t] 32.3 38
Slag Temperature [C] 1200 1230
Coke charge [t] 4.3
Matte Grade [%] 74.8 71.3
The sampling campaign in the FSF-SCF circuit was done during a complete slag cleaning cycle in
the SCF #2 in a period of 1.5 h. The FSF slag was the only feed and no reduction was done. The
operational conditions are showed in the Table 2. A sample of the FSF slag was taken in the launder
with a steel ladle and cooled with water. Three samples of the SCF slag were taken with a steel bar
grooved through one tuyere in three times: at the beginning, in the middle and at the end of the set-
tling period. The ECF samples collected with the steel bar were cooled by air.
Table 2: FSF-SCF parameters
Parameters FSF SCF
Smelting rate [t/h] 76
Matte Grade [%] 60 76
Settling time [t] 60
Slag Temperature [C] 1296 1250
Maximum Bath level [m] 1.8
Bath level sampled with steel bar [m] 1.5
2.2 Methodology and experimental techniques
The samples obtained from both circuits were milled and submitted for chemical analysis and bri-
quettes were appropriated prepared for mineralogical characterization. The bulk chemical elemental
composition was determined using atomic absorption spectroscopy for Cu, Fe, Ca, Al, Mg and Zn;
gravimetric analysis for SiO
2
and LECO titration for S. The magnetite contents were obtained by
Cardona, Hernandez, Araneda, Parra, Bahamondes, Parada, Vargas, Artigas
Proceedings of Copper 2010 2642
Satmagan. Polished surfaces on briquettes of samples were examined under an Olympus CK40M
microscopy in reflected light and photomicrographs were taken.
The first step of the identification and microanalysis of phases in industrial slags consisted in the
definition of standards that will be used to identify the phases in different slag samples. The phases
were identified by X-ray powder diffractometry (DRX), analysis of energy dispersion spectrums
(EDS) and backscatter electron image (BSE). The chemical composition of the sulfides, oxides and
silicates phases was determined by electron probe microanalysis (EPMA) in a JXA JEOL JXA-
8600 Electron Probe Microanalyser with Wavelength Dispersive Detectors (WDS) using ZAF cor-
rection. The common phases in the different slags were grouping according to the chemical compo-
sition, association mode and microstructure characteristics developed during the cooling (Table 3).
The slag samples were analyzed comparing with the standards and characteristics or new found
phases didn't allow a complete match between the sample and the standards, new analysis were per-
formed.
After identifying phases, the slag phases composition and copper phases sizes distribution were
obtained means two procedures: 1) Image analysis on microphotographs with specialized metal-
lographic software (Scentis by Struers) for the TC EAF slag samples; and 2) Analysis of BSE im-
ages and EDS spectrums with Qemscan
equipment.
Cardona, Hernandez, Araneda, Parra, Bahamondes, Parada, Vargas, Artigas
Proceedings of Copper 2010 2650
Figure 9: Sizes distribution of Cu-Cu-sulphides inclusions at the end of settling period in the SCF
obtained by Qemscan with diameters higher than 20 m. Sulfides and copper in matte
inclusions at different heights: h=106 cm, b) 86 cm, c) 43 cm.
a
0
0
1
10
100
200 195 185 175 165 155 145 135 125 115 105 95 85 75 65 55 45 35 24 14
R
e
l
a
t
i
v
e
f
r
e
q
u
e
n
c
y
%
Diameter, m
5 min 30 min 60 min
Background
Cu-metal
Chalcocite/Digenite
Covellite
Cu-S-(low Fe) phase
Fe-O phases
a b
c
Colour Standard
Evaluation of Copper Losses in Slag Cleaning Circuits from Two Chilean Smelters
Proceedings of Copper 2010 2651
The sizes distribution obtained in the slag samples of the upper zone of the bath are almost similar:
All the particles are below 75 m. There are 1 % of 100-190 m droplets in the slag in the zone
near the matte furnace. Images with False color were obtained by Qemscan
.
The settling rate of matte droplets (particles with sulfides phases) can be calculated by the modified
Stokes-Law (Hadamard-Rybeczynski formula) as a first approximation, supposing spherical and
constant diameter matte particles, laminar and homogeneous flow, constant temperature and chemi-
cal composition.
( )
2
m e
e
g - d
u=
12
(1)
Where: u=settling rate [m/s],
m
=density of matte drops [kg/m
3
],
e
= density of slag [kg/m
3
],
d=diameter of matte drops [m],
e
=slag viscosity [Pas].
The viscosities were calculated by the Kv method [8] based in the chemical composition of slags
(Table 4 and 6). The values 0.18 Pas and 0.12 Pas were used for the SCF and EAF, respectively.
Cardona, Hernandez, Araneda, Parra, Bahamondes, Parada, Vargas, Artigas
Proceedings of Copper 2010 2652
Considering 60 minutes of settling in the SCF (see Table 2), can be recovered all the particles with
a diameter higher than 135 m. The normal operation is 40 min of settling, for this condition, the
calculated critical diameter is 145 m. The high SiO
2
and Al
2
O
3
contents affect the basicity of final
slag and the viscosity, also solids of Fe-Spinel and Cr-Spinel in the zone near the matte furnace may
obstruct the settling of matte drops.
In the EAF the critical diameter is 125 m for an operational residence time of 60 min. As it is
shown in Figure 5, there are particles with sizes bigger than this critical size; these particles could
be recovering with a specific strategy for tapping practice. This strategy must consider increasing
the time between tap to tap, a specific amount of liquid slag tap for each tapping and an appropriate
control of reverts addition.
The copper enrichment in matte drops from SCF slags and the reduction of magnetite contents in
zones near the matte slag interface could be explained by the interaction of FeS in matte droplets
with the magnetite. A similar phenomenon was studied by other authors [3] in FSF slags with two
matte ranges (60-63 wt. % Cu and 72-75 wt. %Cu). They explained such phenomena as an equali-
zation of the composition of low grade droplets (matte droplets proceeding from FSF) with the fur-
nace matte by the Fe
3
O
4
reduction for slag formation:
FeS (matte) + 3Fe
3
O
4
(slag) + 5SiO
2
(slag) = 5(2FeOSiO
2
)+SO
2
(2)
The FeS content in the matte droplets decreases and magnetite is reduced. The inversion tempera-
ture in standard conditions is 1167 C. The FSF matte grade coexisting with the FSF slag is
60 wt. % of copper; the sulfide phase tapped from the SCF was about 76 % Cu after 4 cycles of slag
charging and tapping. Considering this condition and the Cu-sulfide phase contents in the slag
(Table 6), it is feasible to suppose the copper contents in the matte droplets varying between 60 and
76 %.
The PSO
2
for the SCF can be determined in function of the equilibrium constant K
I
of Reaction (2)
from the FeS content and the slag composition at Temperatures of 1210-1250 C. Some primary
solids of Fe-Spinel are presents in the SCF (see the Figure 9b), then the a(Fe
3
O
4
) could be consid-
ered near unity.
( )
3 4
2
2 2
3
5
2
( )
( / )
I Fe O FeS
SO FeS
FeOSiO SiO
K a a
p K T a
a a
= =
(3)
Supposing a constant temperature (1250 C), total pressure (1 atm), the a(FeS)
matte
were calculated
by FactSage
tm
for the range of copper contents in matte droplets (60-76 % Cu) and the values are
between 0.149-0.027. The ratio a(2FeOSiO
2
)/a(SiO
2
)1.93 (activities were calculated by Fact-
Sage
TM
). The pSO
2
obtained for the SCF are in the range of 0.025-0.002 atm. Those values are ac-
cording with other thermodynamic calculations for the SCF (0.1-1 atm, 1250 C and 60 % Cu in
matte) [2].
Evaluation of Copper Losses in Slag Cleaning Circuits from Two Chilean Smelters
Proceedings of Copper 2010 2653
5 Conclusions
The copper losses in TC-EAF are mainly in the form of mechanical matte spherical inclusions
(matte droplets) constituted by Cu-sulfides and Cu-(Low Fe)-sulfides. The high grade copper matte
particles in the EAF containing a rich copper alloy phase and Cu-sulfide phases in the final slag is
explained by a insufficient settling time of the reverts that are continuously added to the furnace.
The reduction process in the EAF can be improve and major control of reverts rate addition must be
done in order to improve the copper recovery.
The nature of copper losses in slag from the FSF-SCF circuit has been clarified by sampling the
slag from different height of the bath of the SCF and doing mineral analysis of the samples to cam-
pare it with the mineral analysis of the slag charged. The sizes distribution of copper losses at dif-
ferent levels was obtained and a range for pSO
2
was determined as function of the a(FeS). In order
to improve the copper recovery, a higher settling time can be used and the chemical composition of
slag should be controlled by reduction of oxides that the increase viscosity (e.g. diminishing the
Al
2
O
3
content increase the slag basicity), an optimizing of the Fe/SiO
2
ratio and temperatures are
convenient for the control of the Fe-Spinel formation in the SCF.
Especial care should be taking with the magnetite analysis in industrial slag samples. The origin of
different sources of Fe-Spinel must be considered. Some equiaxials grains were detected in a
quenched surface portion of an industrial slag sample in this study; its an evidence of the solid
phases in the bath, that can be affect the settling of matte droplets. More slag samples with appro-
priate cooling should be analyzed, and EPMA measures of copper dissolved in the solid matrix
phase must be done to understand better the complex phenomena of the copper losses in slag.
References
[1] JALKANEN H., VEHVILAINEN J., POIJARVI J (2003). Copper in Solidified copper
smelter slag. Scandinavian Journal of Metallurgy. P 65-70.
[2] IMRIS, I., REBOLLEDO S., SANCHEZ M. , CASTRO, G. ACHURRA G. & HERNANDEZ
F. (2000). The copper losses in the slags from the el teniente Process. Canadian Metallurgical
Quarterly, Vol 39, No 3, pg 281-290.
[3] GENEVSKI, K. & STEFANOVA, V. (2008). Dispersed droplets in industrial slag melts from
flash smelting furnace. Canadian Metallurgical Quarterly, Vol 47, No 1 pp51-58.
[4] SRIDHAR, R. TOGURI, J.M. AND SIMEONOV. S. (1997). Copper Losses and Thermody-
namic considerations in Copper Smelting. Metallurgical and materials transactions B. Volume
28b. Pg.191-200.
[5] WARCZOK, A., RIVEROS, G., ECHEVERRA, P., DAZ, C.M. (2002). Factors governing
Slag Cleaning in an Electric Furnace. Canadian Metallurgical Quarterly. Vol. 41, No 4 . pp.
465-474.
Cardona, Hernandez, Araneda, Parra, Bahamondes, Parada, Vargas, Artigas
Proceedings of Copper 2010 2654
[6] PARADA, R., CHACANA, P.(2008). Chagres Evolution after Debottnecking project.
12
th
International Flash Smelting Congress. October 26-31. China.
[7] NAVARRO, V., VARGAS J. C., PARRA, R.(2007). Minor elements behavior and develop-
ment of predictive model at The Teniente Converter- Electric Furnace System in the HVL
Smelter. First Meeting on Minor Element Contaminants in Copper Metallurgy. Chile.
[8] BAZAN, V., GOI, C., CASTELL L., BRANDALEZE E. VERDEJA L. F. Y PARRA, R.
(2006). Estimacin de la viscosidad de escorias fayalticas utilizando el mtodo kv y el
mtodo experimental del plano inclinado. Revista de Metalurgia, 42(2). Pg. 84-90.
[9] SLAG ATLAS (1995). Edited by Verein Deutscher Eisenhttenleute(VDEh). Germany.
[10] HERREROS, O., QUIROZ R., MANZANO, E. BOU C., VIALS, J. (1998). Copper extrac-
tion from reverberatory and flash furnace slags by chlorine leaching. Hydrometallurgy 49. Pp
87-101.
Proceedings of Copper 2010 2655
Leaching of Gangue in Technological Flotation
Circuits of Polish Copper Ores
Tomasz Chmielewski Andrzej Luszczkiewicz
Wroclaw University of Technology Wroclaw University of Technology
Faculty of Chemistry Laboratory of Mineral Processing
Division of Chemical Metallurgy Institute of Mining Engineering
Wybrzeze Wyspianskiego 27 Wybrzeze Wyspianskiego 27
50-307 Wroclaw, Poland 50-307 Wroclaw, Poland
Keywords: Copper ores, black shales, flotation, leaching
Abstract
The by-products flotation circuit at the copper concentrator was modified by application of an addi-
tional acidic leaching operation for this hard-to-upgrade material. The operation is based on separa-
tion of a part of the carbonate-clay material, which creates difficulties in the copper ore beneficia-
tion. The main reason for this is a very fine dissemination of valuable sulphidic minerals in the
carbonate gangue matrix. According to the presented concept, the separated byproduct is leached
with sulfuric acid in stirred tanks in order to decompose the carbonate matter and to liberate sulfide
minerals. The leached material is then returned to the flotation circuit for separation of precipitated
gypsum and for cleaning of concentrate after gypsum removal.
The role of chemical modification was evaluated by comparison of commercial flotation results
when the process was performed with and without chemical leaching of the byproducts. It was con-
firmed, both on the commercial scale and in earlier laboratory tests that in order to achieve the
highest possible flotation indices (both recovery and grade) it is necessary to decompose above
70 % of carbonates in processed middlings.
From the analysis of flotation results and mineralogical examination of solids it was found that
acidic leaching results in a high liberation of metal sulfides and leads to considerable improvement
of flotation parameters. After several months of industrial application of the chemical modification
process, a remarkable increase in both flotation recovery and concentrate grade was reported. The
evaluation of the new, chemically modified flotation process, and its effect on the flotation and at-
mospheric leaching were presented in the paper.
Chmielewski, Luszczkiewicz
Proceedings of Copper 2010 2656
1 Introduction
Copper ores mined in the LGOM (Legnica-Glogow Copper Basin, SW Poland) deposit are of sedi-
mentary origin and consists of three types of rocks of different grade and different flotation proper-
ties: sandstone and carbonates which are rather of lean grade and easy-to-upgrade and black shale
which is of higher grade but difficult to upgrade. The shale fraction from the Polish copper ores is
specific and unique in terms of lithology, geochemistry, mineralogy, chemical composition as well
as technology of processing.
The shale series in the copper ores of the Foresudetic Monocline deposit, called copper-bearing
shale (Kupferschiefer), forms from 0.3 to 1.7 m thick layers. It is a non-uniform material which
comprises clay minerals, carbonates (dolomite and calcite), organic matter and dendritic compo-
nents. Dark or black color of the shale is a result of presence of carbonaceous matter [1, 2].
The shale ore is considered as a most valuable fraction among lithological layers in terms of high
copper content and considerably elevated content of base (Co, Ni, Zn, Pb) and precious metals (Ag,
Au, PGE). Tomaszewski [3] considers the shale as a natural polymetallic concentrate and esti-
mated that about 25 % of total copper and from 30 to 40 % of base metals are present in the shale.
Kijewski and Jarosz [4] estimated that from 5 to 8 % of copper resources are in the form of shale.
Table 1: Composition of lithological copper ore layers and content of copper, silver, and organic
carbon in the feed of the KGHM copper concentrators in 90s of 20
th
century and in
2004 [7, 8]
Ore component
Rudna Polkowice Sieroszowice Lubin
90s 2004 90s 2004 90s 2004
Carbonate ore, % 51.0 33 84.0 75 38.0 25
Shale ore, % 5.0 11 6.0 17 8.0 15
Sandstone ore, % 44.0 56 10.0 8 54.0 60
Cu content, % 2.05 2.23 1.81 2.03 1.36 1.28
Ag content, g/Mg 47 53 34 40 68 67
C
org
content, % 0.64 1.49 1.14 1.66 1.76 1.62
In currently applied techniques of ore exploitation, the shale fraction, which content has been in-
creasing in recent years, is a part of a feed to copper concentrators as a mixture with sandstone and
carbonate fraction. Table 1 exhibits an average lithological composition of the copper ore, which
was used as a feed in the processing plants in the 90s of 20
th
century and in 2004. This clearly em-
phasizes that the shale fraction content in the feed of the KGHM Polish Copper S.A. processing
plants has almost doubled. A remarkable increase of the black shale content (up to 27 % for the
Lubin Mine) in the mined ore is currently observed [5].
In Table 2, results of investigations of ore samples taken from one of the mining areas of the Lubin
Mine in 2004 are set together [6]. As can be seen, when the volumetric distribution of the shale in
the ore is 15 %, the distribution of copper and silver originating from this shale in the working face
Leaching of Gangue Matter in Technological Flotation Circuit of Polish Copper Ores
Proceedings of Copper 2010 2657
is over 45 %. For organic carbon (C
org
) this distribution is about 80 %. The organic carbon distribu-
tion in the ore is fully related to the shale fraction content. Organic carbon in the ore mainly origi-
nates from the shale. Therefore, under particular conditions of ore upgrading, the product enriched
in organic carbon will be of similar in character to the shale fraction.
Table 2: Lithological composition and distribution of copper and organic carbon of working face
in the Lubin Mine (region G1, Malomice I) [6]
Name
of the sample
Layers of
the thick-
ness, m
Volume-
tric dis-
tribution,
%
Cu, % Ag, g/Mg C
org
, %
Content
distri-
bution
Content
distri-
bution,
%
Content
distri-
bution
Dolomitic-clay shale 0.10 5.32 2.96 10.09 134 8.77 6.40 26.05
Clay shale 0.18 9.57 5.74 35.21 293 34.51 7.19 52.67
Boundary dolomite 0.10 5.32 0.75 2.56 42 2.75 0.73 2.97
Grey sandstone 1.50 79.79 1.02 52.14 55 53.98 0.30 18.31
Mixed ore (calculated) 1.88 100.00 1.56 100.00 81 100.00 1.31 100.00
Mixture assay 1.64 83 1.36
Chemical analyses of the ore samples of different lithological layers, collected from different min-
ing areas of KGHM Polish Copper S.A., provide a clear correlation between organic carbon (C
org
)
content and metal concentration. The correlation between metal and C
org
content for copper and
silver is shown in Figures 1 and 2. Figures 3 and 4 exhibit a correlations for cobalt and nickel, re-
spectively. Organic carbon concentration has been found a measure of shale content in the ore.
According to the claims of the KGHM technologists, a growing content of the shale fractions in the
flotation feed results in severe difficulties in flotation of the mixed ores. The flotation flowsheet
became very complex and unit operations very expensive, even though technological ore processing
systems have recently been considerably improved and modernized. The growing contribution of
the difficult-to-upgrade shale fraction is also a major reason for hard-to-accept metal losses in the
flotation tailings [7]. This material is considered as the most troublesome material in the existing
flotation circuit at the Lubin, Polkowice and Rudna concentrators due to the presence of clay and
organic matter and fine dissemination of sulfide minerals.
Chmielewski, Luszczkiewicz
Proceedings of Copper 2010 2658
y = 0.0523x
2
+ 0.2947x + 0.988
R
2
= 0.929
0
5
10
15
20
25
0 2 4 6 8 10 12 14 16 18
Content of organic carbon, %
C
o
n
t
e
n
t
o
f
C
u
,
%
.
y = 46.379x
0.6661
R
2
= 0.6279
0
50
100
150
200
250
300
350
400
450
500
0 2 4 6 8 10 12 14 16
Content of organic carbon, %
C
o
n
t
e
n
t
o
f
A
g
,
%
.
Figure 1: Cu vs. organic carbon content in Figure 2: Ag vs. organic carbon content in
shale ore samples from shale ore samples from
KGHM deposits. KGHM deposits.
0
50
100
150
200
250
300
350
0 1 2 3 4 5
Content of organic carbon, %
C
o
n
t
e
n
t
o
f
N
i
,
g
/
t
.
0
50
100
150
200
250
300
0 1 2 3 4 5
Content of organic carbon, %
C
o
n
t
e
n
t
o
f
C
o
,
g
/
t
.
Figure 3: Ni vs. organic carbon content in Figure 4: Co vs. organic carbon content in
shale ore samples from various shale ore samples from various
KGHM deposits. KGHM deposits.
Considerable difficulties to liberate effectively sulphide minerals from the shale matter appear dur-
ing ore comminution. Simultaneously, a part of clay and organic components, which are easily libe-
rated from the shale, create severe difficulties in the flotation of copper sulphide minerals. The
easy-floating fines of clay minerals, liberated during grinding, impregnated with organic matter,
Leaching of Gangue Matter in Technological Flotation Circuit of Polish Copper Ores
Proceedings of Copper 2010 2659
form slime coating on the air bubbles and sulphide minerals, hindering remarkably flotation. Un-
doubtedly, apart from very fine granulation and dissemination of sulphide minerals in the shale, this
is one the most significant reasons of low efficiency of flotation process of the feed containing the
shale fraction.
Figure 5: General balance of organic carbon in flotation circuits of Lubin Concentrator [8].
Figure 5 shows general copper and organic carbon mass balances in the flotation process at the Lu-
bin concentrator. It can be clearly found that above 70 % of organic carbon is transferred do final
tailings and the remaining 30 % becomes a concentrate component [8].
Figure 6 exhibits a simplified flowsheet for the Lubin concentrator, whereas Figure 7 illustrates a
mass balance for the 1
st
cleaning flotation [8]. According to these figures, almost 70 % of the organ-
ic carbon and some 30 % of copper circulate between 1
st
cleaning flotation and additional milling
operations. The shale by-product (middlings) from 1
st
cleaning flotation corresponds to about 60 %
of mass of the flotation plant feed. These data became a base of a novel concept of separation of a
part of the hard-to-upgrade shale fraction from the flotation circuit by means of individual
processing of this by-product (middlings) from 1
st
cleaning in a separate chemical-flotation process,
followed by oxidative leaching (atmospheric or pressure leaching) as it was shown in Figure 8 [9].
100,0
1,2 100,0
1,6 100,0
93,6 6,4
0,2 14,0 16,0 86,0
1,3 72,2 7,0 27,8
Yield, %
Cu, % Cu recovery, %
C
org
, % C
org
recovery,%
FINAL CONCENTRATE
LUBIN CONCENTRATOR
FINAL TAILINGS
FEED
(Run of Mine ore)
Chmielewski, Luszczkiewicz
Proceedings of Copper 2010 2660
Figure 6: Simplified flowsheet of first technological line of Lubin Concentrator.
Figure 7: Flowsheet and mass balance of copper and organic carbon for 1
st
cleaning flotation [8]
(as in Figure 6).
C - concentrate
M - middlings
BM - ball mill
HC - hydrocyclone
final
tailings
final concentrate
HC
M
M
C
3rd cleaning
1st cleaning
1st rougher
BM
C
C
regrind
M
C
from & to
grinding circuit
M
2nd cleaning
Rougher-scavenger
2nd rougher
100,0 25,7
100,0 3,8 75,7
100,0 6,0 95,0
40,2 10,3 59,8 15,4
60,3 5,7 45,7 39,7 2,5 30,1
27,0 4,0 25,7 73,0 7,3 69,3
Total mass balance Operational balance
Yield, %
Cu, % Cu recovery, %
C
org
, % C
org
recovery,%
Cu recovery, %
C
org
recovery,%
Yield, %
1st CLEANING FLOTATION
Feed: concentrate from
rougher flotation
Concentrate
to 2nd cleaning flotation
Middlings to rougher
flotation after cycloning
and regrinding
Leaching of Gangue Matter in Technological Flotation Circuit of Polish Copper Ores
Proceedings of Copper 2010 2661
Figure 8: General concept of separate processing of shale-enriched by-product from Lubin Con-
centrator using hydro- and bio-metallurgy [9].
2 Concept and objectives of examinations
The main purpose of the presented investigations was to analyse the concept of separation of flota-
tion by-product enriched in organic carbon (called shale product) from the flotation feed, which was
a natural mixture of three lithological fractions of the ore, and subsequent processing it by means of
chemically-assisted flotation. In general, the method of processing presented in this paper assumes
separation of the hard-to-upgrade shale fractions from the existing flotation circuits and their com-
plex hydrometallurgical processing. This concept was schematically presented in Figure 8. The im-
portant base of the presented concept is the necessity of keeping unchanged the main technological
circuit, producing concentrates for metallurgical plants.
COPPER
BASE AND
NOBLE METALS
FLOTATION
TAILINGS
Technological process of copper recovery
at the KGHM Polish Copper
Shale-enriched
by-product
Non-oxidative leaching
with H
2
SO
4
for removing
carbonate gangue
Bio- and hydrometallurgical
processes for metals recovery
Proposed and investigated process of
beneficiation of black shale fraction
SLAGS
FLOTATION
SMELTING
REFINING
CRUSHING
& GRINDING
MINING
FLOTATION GYPSUM
Chmielewski, Luszczkiewicz
Proceedings of Copper 2010 2662
Figure 9: Comparison of results of standard flotation and shale flotation using various doses of
fuel oil as a non-polar collector [10].
Luszczkiewicz and co-authors [10] analysed results of flotation experiments in terms of separation
of shale fraction from a natural mixture of three ore fractions flotation feed of Lubin concentrator.
Non-polar collector was applied for this purpose. The results of the experiments are collected in
Figure 9 as organic carbon recovery vs. copper recovery plots. The diagonal line of the diagram
represents a lack of differences in floatability of both feed components copper and organic carbon.
Above the diagonal line and on the left side of the diagram, better organic carbon upgrading was
observed. Below the diagonal line and on the right side of the graph, copper was found to better
upgradeable. The best separation of copper from organic carbon was observed when low doses
(100 g/Mg) of collector (fuel oil) were applied. The differences in the flotation of two components
are rather small.
It is hard to evaluate the process performed on the industrial scale, but separation of the shale
enriched fractions from the rest of feed is already observed, for example in 1
st
cleaning flotation in
Lubin. The tailings from this process, containing about 2-3 % of Cu and evidently elevated content
of other metals, exhibit some 7-8 % of organic carbon very similarly to the final concentrate (Ta-
ble 3). The material can be, therefore, considered as a shale concentrate and can be separated from
the flotation circuit for individual treatment, to recover all metals. Hydrometallurgical or biometal-
0
20
40
60
80
100
0 20 40 60 80 100
Recovery of copper in concentrate, Cu, %
R
e
c
o
v
e
r
y
o
f
o
r
g
a
n
i
c
c
a
r
b
o
n
i
n
c
o
n
c
e
n
t
r
a
t
e
,
%
.
lack of floatabilit y differences
St andard Xant .
Fuel oil 100g/Mg
Fuel oil 200g/Mg
Fuel oil 400g/Mg
Fuel oil 600g/Mg
Leaching of Gangue Matter in Technological Flotation Circuit of Polish Copper Ores
Proceedings of Copper 2010 2663
lurgical processes of leaching were preliminary tested for this purposes. This approach can be addi-
tionally justified by mass balances presented in Figure 7.
Table 3: Chemical composition of the Lubin Concentrator shale middlings (tailings of 1
st
cleaning)
in comparison with a processing feed (run-of-mine ore) and final concentrate produced in
2004
Material Cu, % Ag, g/t C
org
, % Ni, g/t Co, g/t Pb, % Zn, g/t Fe, % S
c,
% V, g/t Mo, g/t
Lubin flotation feed,
2004
1.28 67 1.62 65 126 0.27 0.04 0.85 1.68 141 42
Tailings from 1
st
cleaning 2.30 151 7.71 396 429 1.18 1.23 2.14 2.96 765 291
Lubin final concentrate,
2004
16.94 830 8.35 527 1287 3.12 0.54 5.79 11.68 630 269
The concept of separation of hard-to-upgrade shale fraction and its individual, hydrometallurgical
processing should result in significant simplification of the flotation process. Moreover, metals re-
covery in flotation and hydrometallurgical processes are expected to be remarkably higher (cobalt is
not recovered) with regard to existing systems.
The shale product from the Lubin Concentrator contains carbonate matter and has to be initially
subjected to non-oxidative leaching with sulphuric acid in order to totally decompose acid-
consuming components prior to the atmospheric or pressure leaching. In this paper, additional flo-
tation of leached by-product for separation of gypsum, which is being formed in leaching, was con-
sidered. This process is already applied on the industrial scale at KGHM [11].
3 Non-oxidative leaching
Non-oxidative leaching of the shale-enriched by-product from the commercial circuits was used as
a unit operation which aim was to liberate metal-bearing minerals from carbonate particles. This
kind of leaching consists of selective (without breaking sulphides) chemical calcium and magne-
sium carbonates decomposition by means of sulphuric acid, according to the reactions:
CaCO
3
+ H
2
SO
4
+ H
2
O = CaSO
4
2H
2
O + CO
2
(1)
MgCO
3
+ H
2
SO
4
= MgSO
4
+ CO
2
+ H
2
O (2)
Hydrated calcium sulfate (gypsum) precipitates as a solid reaction product, whereas water-soluble
magnesium sulfate and gaseous carbon dioxide are other reaction products. Since particles of mid-
dlings are fine, leaching of the carbonate gangue with H
2
SO
4
is very rapid and can be performed at
ambient temperatures in reactors with mechanical stirring of a simple construction.
The amount of H
2
SO
4
applied in the non-oxidative leaching operation directly corresponds to the
content of carbonates and has to be precisely controlled to maintain the final pH of the pulp at a
level enabling its direct transfer either to the flotation circuit without a need for pH correction or to
the leaching and bioleaching. Therefore, for further flotation, the amount of sulfuric acid introduced
Chmielewski, Luszczkiewicz
Proceedings of Copper 2010 2664
to the leaching operation should always be kept below the analytically determined maximum
amount of acid required for the total carbonates decomposition.
0,5
1,5
2,5
3,5
4,5
5,5
6,5
7,5
0 10 20 30 40 50 60 70 80 90
Leaching time, min.
p
H
20%
40%
50%
70%
90%
100%
1,0
1,5
2,0
2,5
3,0
3,5
4,0
4,5
5,0
5,5
6,0
0 1 2 3 4 5 6 7 8 9 10
Leaching time, min.
p
H
20%
40%
50%
70%
90%
100%
Figure 10: pH vs. leaching time plots for Figure 11: pH vs. leaching time plots for
non-oxidative leaching of Lubin initial stage of non-oxidative
shale middlings. leaching of Lubin shale
middlings.
Kinetics of non-oxidative leaching of the middlings from the Lubin Concentrator was investigated
for different degrees of carbonates decomposition varying from 20 to 100 % (Figures 10 and 11).
The key parameter determining the amount of H
2
SO
4
required for carbonates leaching is the maxi-
mum demand for acid (
max
SO H
4 2
z ), which is the mass of pure H
2
SO
4
necessary for a total decomposi-
tion of carbonates in 1 kg of dry solid feed. The
max
SO H
4 2
z parameter should be determined analytically
from laboratory tests. H
2
SO
4
is introduced to the reactor containing the shale slurry at a rate that
either assures its total utilization or only a selected grade. The maximum demand for sulphuric acid
for the examined Lubin middlings was 494 g H
2
SO
4
/kg of dry material.
The process control of non-oxidative leaching is based on pH measurement of the leached mid-
dlings suspension after introduction of required amount of acid. On the basis of kinetic results for
the Lubin middlings it was found that the process is very rapid (Figures 10 and 11) and after about
5 minutes almost entire amount of acid is already used up. The observed further pH changes (up to
about 40-60 minutes) correspond only to the saturation of the slurry with CO
2
.
Non-oxidative leaching is not only very fast but also selective. It does not cause chemical decompo-
sition of metal sulphides under non-oxidative conditions created by the carbon dioxide.
Leaching of Gangue Matter in Technological Flotation Circuit of Polish Copper Ores
Proceedings of Copper 2010 2665
The samples of middlings being the tailings from the 1
st
cleaning commercial operation at the Lubin
Concentrator before and after non-oxidative leaching with sulfuric acid were the material for expe-
riments. The total time of carbonate decomposition in the examined samples was 60 minutes.
4 Flotation of the shale middlings after non-oxidative leaching
Flotation feeds with either 50, 70 or 90 % of carbonate decomposition were used in experiments
performed according to the flow-sheet shown in Figure 12. For comparison, the flotation of raw
middling (i.e. non-leached with H
2
SO
4
) was also carried out. For the flotation experiments a Mek-
hanobr sub-aeration type laboratory flotation machine equipped with a 1 dm
3
cell was applied. pH
measured during the flotation for the non-leached material was from 7.6 to 8.3, and for the leached
samples from 5.5 to 7.0.
Figure 12: Scheme of flotation experiments.
In the standard xanthate flotation of sulfides, 50-60 g/Mg of collector (mixture of sodium
ethyl + amyl xanthate, 1:1) and 10-20 g/Mg of frother (Corflot) were used. The rougher flotation
was conducted up to the moment when the flotation froth did not contain useful minerals. The ob-
tained concentrate was subsequently subjected to the cleaning flotation where the following prod-
ucts of flotation were collected at suitable time intervals. The procedures of the flotation experi-
ments were the same, but quantities of the reagents and times intervals of the products collection
were different for various examined feeds.
Due to strong frothing properties of the suspension after leaching, the experiments were carried out
with a minimal air-flow. This caused low yield of the products and resulted in higher selectivity of
flotation.
FEED
CLEANING FLOTATION
K1 K2 ... Middlings
(cleaning tail)
Sodium
ethyl+amyl
xanthate (1:1) + frother
Tailing
( Sulfide concentrates)
ROUGHER FLOTATION
Chmielewski, Luszczkiewicz
Proceedings of Copper 2010 2666
Figure 13: Copper upgrading curves for various Figure 14: Copper upgrading curves
degree of carbonate decomposition. enlarged part of Figure 13.
Figure 15: Silver upgrading curves for various Figure 16: Upgrading curves for analyzed
degree of carbonate decomposition. components for 70 % of
carbonate decomposition.
Shale middlings samples were taken from the Lubin Concentrator commercial circuit and initially
subjected to determination of the maximum consumption of sulphuric acid for total decomposition
of carbonates [11]. Subsequently, non-oxidative leaching with controlled grade of carbonates de-
composition was conducted followed by laboratory flotation tests. Figures 13 to 16 exhibit the re-
sults of flotation experiments in the form of Fuerstenau curves, drawn according to procedure de-
scribed by Drzymala and Ahmed [12].
0
20
40
60
80
100
0 20 40 60 80 100
Recovery of copper in concentrate, %
R
e
c
o
v
e
r
y
o
f
b
a
r
r
e
n
p
a
r
t
i
n
t
a
i
l
i
n
g
,
%
.
Lack of upgrading
Standard flot at ion
50% of decomposition
70% of decomposition
90% of decomposition
Cu
0
10
20
30
40
50
50 60 70 80 90 100
Recovery of copper in concentrate, %
R
e
c
o
v
e
r
y
o
f
b
a
r
r
e
n
p
a
r
t
o
f
t
a
i
l
i
n
g
,
%
.
Lack of upgrading
Standard flot at ion
50% of decomposition
70% of decomposition
90% of decomposition
Cu
0
10
20
30
40
50
50 60 70 80 90 100
Recovery of silver in concentrat e, %
R
e
c
o
v
e
r
y
o
f
b
a
r
r
e
n
p
a
r
t
o
f
t
a
i
l
i
n
g
,
%
.
Lack of upgrading
Standard flotation
50% of decomposit ion
70% of decomposit ion
90% of decomposit ion
Ag
0
10
20
30
40
50
50 60 70 80 90 100
Recovery of component in concent rate, %
R
e
c
o
v
e
r
y
o
f
b
a
r
r
e
n
p
a
r
t
o
f
t
a
i
l
i
n
g
,
%
.
Lack of upgrading
Cu
Ag
Co
Zn
Fe
Corg
70% of carbonate
decomposition
Leaching of Gangue Matter in Technological Flotation Circuit of Polish Copper Ores
Proceedings of Copper 2010 2667
Figure 13 shows the floatability curves for copper at various carbonate decomposition grades (50,
70 and 90 %), whereas Figure 14 presents an enlarged part of the relationship for more adequate
evaluation. The floatability curves for silver are given in Figure 15 and Figure 16 shows the effect
of 70 % carbonate decomposition on the flotation results of Cu, Ag, Co, Zn, Fe and organic carbon.
It is well seen that the higher is the degree of carbonates decomposition, the better is floatability of
valuable components. From Figure 16 we can find that upgradeability of copper and organic carbon
exhibit the highest values with lowest for iron. Silver, cobalt and zinc are found to be similarly
floatable after chemical decomposition of 70 % of carbonates from the feed.
5 Atmospheric leaching in oxygenated sulphuric acid
The shale by-product from the Lubin Concentrator was subjected to atmospheric leaching after ini-
tial total non-oxidative decomposition of carbonates. Leaching was performed at various tempera-
tures (25, 60, 80 and 90
o
C) and various concentration of iron(III) (7, 15 and 30 g/l). The effect of
sulphuric acid was also examined at various temperatures (25-90
o
C) at a solid/liquid ratio from 1:5
to 1:4 and at the oxygen flow rate of 30 l/h.
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
10.0
0 60 120 180 240 300
Leaching time, min.
C
u
c
o
n
c
e
n
t
r
a
t
i
o
n
,
g
/
l
.
25oC
60oC
80oC
90oC
0, 0
1, 0
2, 0
3, 0
4, 0
5, 0
6, 0
7, 0
8, 0
9, 0
10, 0
0 60 120 180 240 300
Leaching time, min.
C
u
c
o
n
c
e
n
t
r
a
t
i
o
n
,
g
/
l
.
0
30 g/l
7 g/l
15 g/l
Figure 17: The effect of temperature on the Figure 18: The effect of Fe(III) on the
atmospheric leaching of copper atmospheric leaching of copper
from Lubin middlings in from Lubin shale middlings.
oxygenated sulphuric acid.
A minor effect of concentration of sulphuric acid was observed for metals leaching in preliminary
tests. Therefore, the H
2
SO
4
concentration was kept at the level of 50 g/dm
3
for all leaching experi-
ments. Concentration of Cu, Fe, Ni, Co and As was analysed in the pregnant leaching solutions vs.
leaching time. Ag and Pb content was determined in solid residue samples.
Chmielewski, Luszczkiewicz
Proceedings of Copper 2010 2668
Solubility of metals being in the sulphidic form requires the presence of oxidation agent: gaseous
oxygen or/and iron(III). Copper, which is present in the feed predominantly as chalcocite (Cu
2
S)
and bornite (Cu
5
FeS
4
) can be leached very rapidly by oxygenated H
2
SO
4
solution and the leaching
rate increases remarkably in the presence of iron(III).
Selected kinetic curves for atmospheric leaching are presented in Figures 17 to 20 for copper, cobalt
and nickel. Both temperature and concentration of iron(III) appeared to be a key parameters in at-
mospheric leaching. It was determined that at 90
o
C and at s/l ratio of 1:4 it was possible to reach
concentration of copper of about 9 g/l after 5 hours leaching. Immense effect of Fe(III) was also
observed for leaching cobalt and nickel.
0
10
20
30
40
50
60
70
80
90
100
110
0 60 120 180 240 300
Leaching time, min.
C
o
c
o
n
c
e
n
t
r
a
t
i
o
n
,
m
g
/
l
.
0
7
15
30
0
5
10
15
20
25
30
0 60 120 180 240 300
Leaching time, min.
N
i
c
o
n
c
e
n
t
r
a
t
i
o
n
,
m
g
/
l
.
0
7
15
30
Figure 19: The effect of Fe(III) on the Figure 20: The effect of Fe(III) on the
atmospheric leaching of cobalt atmospheric leaching of nickel
from Lubin middlings. from Lubin middlings.
Particle size analyses for the Lubin middlings indicated that the applied middlings were rather to
coarse in terms of leaching rate. Parameter d
80
exceeded 100 m even after non-oxidative leaching
with H
2
SO
4
(carbonates decomposition R
w
= 30-90 %). Additional microscopic SEM (Figure 21)
observations lead to a significant conclusion, that the most coarse particles are formed by the shale
material, which can not be chemically decomposed during treatment with acid, even in the presence
of oxygen (Figure 21D). Therefore, separation of +50/60 m particle size fraction and its additional
regrinding seems to be very desirable to facilitate further leaching.
Leaching of Gangue Matter in Technological Flotation Circuit of Polish Copper Ores
Proceedings of Copper 2010 2669
Figure 21: SEM mineralogical analyses of Lubin shale middlings before leaching (A, B, C) and
after atmospheric leaching in acidic consitions (D).
6 Conclusions
1. Shale middlings, being the tailings from the 1
st
cleaning flotation operation at the Lubin Con-
centrator, is recognized as the most troublesome product in the existing flotation circuit due to a
high amount of the difficult-to-treat carbonate and organic fraction. Fine dissemination of sul-
fide minerals in gangue minerals is the main reason of difficulties in upgrading of the shale frac-
tion. It was the major reason for choosing the middlings as the material for alternative process-
ing by means of non-oxidative and atmospheric leaching.
2. Examined middlings can be considered as a ready-to-use shale concentrate. This is due to the
high content of organic carbon (from 5 to 10 %). Moreover, the content of Cu (2-3 %) is close
to the copper concentration in a natural shale. Recovery of organic carbon, directly correlated
with quantity of shale, indicates, that most of the shale fraction is in this by-product. Shale mid-
dlings should be the principal feed for the bio- and hydro-metallurgical processes.
D
C
B
A
Chmielewski, Luszczkiewicz
Proceedings of Copper 2010 2670
3. Flotation experiments demonstrated that upgrading the middlings was nearly impossible without
a chemical pretreatment. Non-oxidative leaching of Lubin middlings with H
2
SO
4
appeared to be
useful chemical operation for selective liberation of sulphide minerals prior to flotation.
4. Partial decomposition of carbonates from 50 to 90 % remarkably improves the liberation degree
of the valuable minerals from the hydrophilic intergrowths with carbonate matter.
5. Evidently improved flotation of Cu, Ag, Zn, Co, Fe and C
org
occurs for leached materials in
comparison with direct flotation of untreated flotation feed.
6. An increase of the carbonates decomposition grade of the feed results in enhanced recovery and
content of the useful components in the concentrate. The best results of the flotation parameters
(recovery and concentrate grade) were obtained for the feed with the 90 % of carbonate decom-
position.
7. Atmospheric leaching of shale middlings after total carbonate decomposition can be effectively
used as alternative process for recovering of metal values from this troublesome feed
References
[1] Konstantynowicz-Zieliska J. 1990, Petrography and genesis of copper-bearing shales from
Foresudetic Monocline. Rudy i Metale Nieelazne. R.35, Nr 5-6, 128-133, in Polish
[2] Rydzewski A., 1996, Lithology of the deposit rocks. In: Monografia KGHM Polska Mied
S.A., A. Piestrzynski (Ed). Publ. CPBM Cuprum Sp. z O.O., Lubin, 137-141, in Polish
[3] Tomaszewski J., 1985, Problems of a rational utilization of copper-polymetallic ores from the
Foresudetic Monocline deposits. Physicochemical Problems of Mineral Processing, Nr 17,
131-141, in Polish
[4] Kijewski P., Jarosz J., Ore mineralization and form of occurrence of accompanying elements
in the copper deposit. Proceedings of the Conference: Accompanying metals in the copper
ore deposit, the state and prospects for further use", Rydzyna, May 1987, P. Kijewski (Ed),
Publ. NOT/SITG/Cuprum, Wroclaw 1987, 21-48, in Polish
[5] Kubacz N., Skorupska B., Evaluation of influence of organic carbon on concentration and
smelting. processes. Proc. VIII International Conference on Non-ferrous Ore Processing,
Wojcieszyce (Poland), May 21-23, KGHM Cuprum Wroclaw 2007, 157-166, in Polish
[6] Luszczkiewicz A., Evaluation of upgradebility of the ore with elevated content of black shale.
Report of Investigations, Archive of Laboratory of Mineral Processing, Wroclaw University
of Technology, Wroclaw, October 2004, in Polish
Leaching of Gangue Matter in Technological Flotation Circuit of Polish Copper Ores
Proceedings of Copper 2010 2671
[7] Luszczkiewicza A., Beneficiation of copper black shale from ores of Lubin-Glogow region.
Conference Proceedings on: Current Problems of Mineral Processing of Copper Ore in Pol-
and, Polkowice, November 16, 2000, Publ. Committee of Mining Polish Academy of
Sciences and KGHM Polish Copper, 137-156, in Polish
[8] Luszczkiewicz A., Separation of black shale fraction from the polish copper deposit and a
general concept of change of technology at the Lubin Concentrator. Presentation at Bioshale
Final Meeting, Orleans, November 27-29, 2007
[9] Chmielewski T., Luszczkiewicz A., Konopacka Z., Separation and concept of processing of
black shale copper ore from Lubin mine, Proc. VIII International Conference on Non-ferrous
Ore Processing, Wojcieszyce (Poland), May 21-23, KGHM Cuprum Wroclaw 2007, 171-184,
in Polish
[10] Luszczkiewicz A.., Konopacka Z., Drzymala J., Flotation of black shale of Polish copper ore
from Lubin. Proceedings: Perspectives for applying bioleaching technology to process shale-
bearing copper ores, BIOPPROCOP 06, Lubin, June 19, 2006, Publ. KGHM Cuprum Sp. z
O.O., Wroclaw 2006, 29-47, in Polish
[11] Luszczkiewicz A., Chmielewski T., Acid treatment of copper sulfide middlings and rougher
concentrates in the flotation circuit of carbonate ores. International Journal of Mineral
Processing, Vol. 88, Issue: 1-2, 2008, 45-52
[12] Drzymala J, Ahmed H.A.M., Mathematical equations for approximation of separation results
using the Fuerstenau upgrading curves. International Journal of Mineral Processing, Vol. 76,
Issue: 1-2, 2005, 55-65
Proceedings of Copper 2010 2672
Proceedings of Copper 2010 2673
Pressure Leaching of Shale Middlings from Lubin
Concentrator in Oxygenated Sulphuric Acid
Tomasz Chmielewski, Jerzy Wdka
Wroclaw University of Technology
Faculty of Chemistry, Division of Chemical Metallurgy
Wybrzeze Wyspianskiego 27
50-307 Wroclaw, Poland
Keywords: Pressure leaching, shale ore, copper
Abstract
The effect of initial temperature, concentration of sulphuric acid, and oxygen partial pressure on
acidic pressure leaching of shale middlings-tailings from 1
st
cleaning flotation of Lubin Concentra-
tor (ZWR Lubin, Poland) - have been investigated. Lubin middlings were selected for hydrometal-
lurgical treatment as a troublesome shale material exhibiting elevated content of metals (Cu, Co, Ni,
Zn, As, Ag, Pb) and of organic carbon. Leaching was performed in a stirred autoclave at tempera-
tures within the range of 100-180
o
C. Oxygen at partial pressure from 2.5 to 10.0 at was used as an
oxidizing agent. Leaching process was performed at initial sulphuric acid concentration from 20 to
50 g/dm
3
. Pressure leaching appeared to be very efficient process for recovering of Cu, Co, Fe, Zn
and As from examined shale middlings. Leaching recovery of Ni was remarkably lower. Ag and Pb
remained in the solid residue and will be recovered in separate processes.
1 Introduction
Pressure hydrometallurgy has an extensive and growing application in processing of zinc [1, 2],
nickel in particular from laterite ores [3-5], copper sulphides [6-12] and in pretreatment of refrac-
tory gold ores [13-14], in which gold is finely disseminated in a lattice of sulphidic minerals. Re-
cently, numerous investigations are undertaken to apply pressure hydrometallurgy to process bypro-
ducts and wastes of copper refining (anode slimes, non-ferrous smelter slags) [16] and for recovery
of metals from ores and raw materials of the specific properties and composition, which cause diffi-
culties in their treatment by standard methods (black shale ores, tailings). Pressure hydrometallurgy
exhibits numerous advantages, creating intensive investigation of it at laboratory and industrial
scale. The main advantages of application of pressure leaching processes for sulphidic, polymetallic
raw materials are:
Chmielewski, Wodka
Proceedings of Copper 2010 2674
- high rate of leaching reactions,
- elimination of SO
2
, other gases and dust emission,
- high selectivity of pressure leaching,
- possibility of arsenic utilization or stabilization (as a low soluble scorodite),
- total recovery of base and noble metals,
- no restriction of the scale production.
Pressure leaching can be applied in a large scale to treat millions ton per year of refractory gold
ores, or to produce several tons of nickel and cobalt from their sulphidic or lateritic resources. The
methods of hydrometallurgical treatment of copper sulphide concentrates, as an alternative to tradi-
tional smelter-refinery processes, were compared and discussed in details [17]. Pressure leaching
was recently successfully applied in industrial scale by Phelps Dodge (Freeport McMoran Copper &
Gold) for production of copper (80 kt/y) from copper sulphide concentrates at Morenci plant [18]. It
is the first in the World full-scale application of pressure leaching process for copper sulphide con-
centrates.
Sedimentary, sulphidic copper ores from Polish deposits (LGOM - Legnica - Glogow Basin, SW
Poland) exhibit the elevated content of black shale fraction which has been systematically increas-
ing, particularly in recent years [19]. The black shale ores reveal unique both advantageous and de-
trimental properties. They contain evidently more copper, base metals and noble metals than sand-
stone or carbonate fractions. However, the elevated carbonate and organic coal content as well as
dissemination of the metals-bearing minerals occurring in the black shale ore create significant dif-
ficulties in the flotation and cause remarkable and unacceptable metals losses, which has to be li-
mited for technical and economical reasons [20]. Therefore, hydrometallurgy has been declared as a
major strategy objective for coming years at KGHM.
During the comprehensive investigations on application of bio- and hydrometallurgy for alternative
processing of black shale fraction from Lubin Mine, the non-oxidative leaching, atmospheric leach-
ing in oxygenated sulphuric acid, acidic leaching under oxygen pressure and bioleaching have been
considered as optional processes for recovering of copper and base metals from shale by-product
(middlings) from Lubin Concentrator [21]. Middlings-tailings of 1
st
cleaning were separated from
the industrial flotation circuit as a feed for further alternative, hydrometallurgical treatment.
Tailings from the first cleaning flotation (middlings) of the first circuit at Lubin Concentrator (ZWR
Lubin) were primarily selected for the laboratory investigations as a shale concentrate. This by-
product can not be effectively upgraded by flotation and creates serious decline of flotation indices,
both recovery and concentrate grade, primarily at Lubin Concentrator. It appeared, that the separa-
tion of the middlings from Lubin flotation circuit and further processing by means of leaching pur-
poses is quite simple. Moreover, the middlings have an enormously advantageous composition with
regard to both atmospheric and pressure leaching. Bornite and chalcocite, the easiest leachable cop-
Pressure Leaching of Shale Middlings in Oxygenated Sulphuric Acid
Proceedings of Copper 2010 2675
per sulphides, are found to be dominating copper-bearing minerals in this by-product. Moreover, the
solid contains up to 9 % of organic carbon and some 30 % of carbonates that must be decomposed
by sulphuric acid prior to pressure leaching.
Pressure leaching in oxygenated H
2
SO
4
solution, at elevated temperatures, can be considered as a
mostly recommended way of copper manufacturing. Sulphuric acid is produced at KGHM smelters
(about 650 kt) as a by-product or rather troublesome waste during the processing of copper sulphide
concentrate and is considered as the most suitable, cheap and easy-accessible leaching agent.
The effect of initial temperature, sulphuric acid concentration and oxygen partial pressure on pres-
sure leaching of shale fraction of copper ore has been investigated to evaluate the leaching ability of
the shale fraction (middlings) separated as tailing of 1
st
cleaning from Lubin Concentrator. Pressure
leaching examinations were performed in the temperature range of 100-180
o
C using oxygen as an
oxidizing agent.
2 Experimental
2.1 Lubin middlings characterization
Selection of black shale feed and experimental determination of conditions for effective recovering
of copper and other valuable metals (Fe, Ni, Co, Pb, Au, PGM) from shale-containing materials
(ore, middlings, shale concentrates) was the general task at the starting point of the BIOSHALE
research project, co-financed by European Commission within the range of VII Frame Pro-
gramme [22]. The research program was initially focused on the geological shale samples collected
for preliminary laboratory examinations (chemical and mineralogical analyses, non-oxidative, at-
mospheric and acid pressure leaching). After the extended period of experimental work, the shale
middlings-tailings from the 1
st
cleaning flotation at Lubin Concentrator (ZWR Lubin) were finally
selected and accepted by a technical, managing, and engineering staff at KGHM as the only shale
containing solid for hydrometallurgical laboratory tests and for prospective full scale alternative
processing. Simultaneously, this material was unanimously recognized as the most troublesome
solid in existing flotation circuit at Lubin Concentrator due to the high content of shale fraction
(clay and organic matter) and fine dissemination of sulfide minerals. It was fully considered as a
shale concentrate, which can be separated from the flotation circuit for hydrometallurgical treat-
ment. Initial samples were collected of numerous shale ore fractions from various LGOM (Poland)
deposits.
Chemical analyses exhibited varying metal content in the shale. For further laboratory studies the
process alternatives were selected including chemical pretreatment with H
2
SO
4
for flotation, non-
oxidative leaching for carbonates decomposition, atmospheric leaching and pressure leaching inves-
tigations as BIOSHALE project WP4 tasks. The laboratory tests undoubtedly confirmed that the
assumptions made in the first stage of the work program were correct.
Chmielewski, Wodka
Proceedings of Copper 2010 2676
Acidic leaching of shale middlings appeared to be very effective from mineralogical and technical
viewpoint. Favorable mineralogical composition of Lubin middlings (dominating content of copper
in the form of chalcocite and bornite) and easy access to sulfuric acid, being in fact a waste ma-
terial from copper smelters, makes this approach very attractive for future, necessary technological
alterations [23-27].
One of the main reasons, that black shale fraction became a serious problem in flotation of Polish
copper ores is a very fine mineralization of metal-bearing sulfides and their dissemination in hydro-
philic carbonate and in organic matter. It is impossible to liberate these minerals only by means of
grinding. Consequently, a great amount of metal values remain in flotation tailings and leads to
hardly accepted metals loss. Very complex and expensive alterations in flotation and grinding cir-
cuits at Lubin Concentrator appeared to be ineffective in terms of flotation indices. Both the copper
and silver recovery and concentrate grade have been reported to decrease remarkably.
Table 1: Content of shale fraction (in percent) in Polish copper ores mined from various LGOM
(Legnica Glogow Copper Basin) deposits [19].
Mine - Deposit
Year
2002 2004 2006
Lubin 18.8 15.0 27.0
Polkowice-Sieroszowice 11.3 13.3 11.2
Rudna 2.2 3.9 4.0
An additional unfavorable effect of increasing shale content has been lately observed during the
flash smelting at Gogw II smelter. The growing content of shale fraction (Table 1) and the in-
crease in organic carbon content exceeding 9 % for Lubin and Polkowice in the concentrate
smelter feed, results in a notable diminishing of the process efficiency [19]. Therefore, separation of
shale fraction from the flotation circuits (e.g. Lubin middlings) and its individual, hydrometallurgic-
al processing, postulated within the frame of BIOSHALE project, can be accepted as a key alterna-
tive for existing technologies. Such alteration can result in the following advantages:
Integration of new technology with existing processes.
Simplification of existing flotation circuits and enhancement of flotation efficiency after separa-
tion of shale fraction.
Increase of metals recovery in total (flotation concentrate + hydrometallurgy) remarkable de-
crease of metals loses and environmental impact.
Stabilization of organic carbon content in flotation concentrates on the proper level, accepted for
flash smelting process.
Effective utilization of an excess of sulfuric acid a by-product or waste from smelting.
It seems to be rather obvious that due to the observed apparent decline of copper ores quality and
upgradeability, particularly from Lubin and Polkowice deposits (Table 1), present beneficiation
Pressure Leaching of Shale Middlings in Oxygenated Sulphuric Acid
Proceedings of Copper 2010 2677
technologies are not able to guarantee the acceptable metals recovery and concentrate grade without
significant technology alteration and urgent introduction of new alternative concepts. This appeared
to be urgent for Polish copper industry. Acidic pressure leaching can be considered as an alternative
process.
Shale middlings from Lubin Concentrator were collected and characterized in details in D.4.2. Deli-
verables within BIOSHALE European project, on the basis of 1
st
sampling campaign in 2006 and
2
nd
sampling campaign in 2007. Middlings from 1
st
and 2
nd
sampling campaigns were used as a feed
for pressure leaching. It is well seen from Table 2, that chemical composition of material examined
in 2007 was very similar to those used in previous investigations. Copper content was about 2.60 %,
organic carbon was 6.30 % lower than in 2006 campaign. The content of accompanying metals
was observed to be almost identical, except of zinc. The comparison of chemical analyses of Lubin
middlings samples from 2006 and 2007 evidently shows, that examined feed is fairly stable in terms
of its chemical composition.
Table 2: Chemical characterization of Lubin shale middlings
(1
st
and 2
nd
sampling campaigns 2006 and 2007).
CONTENT
Cu, % Fe, % Ni, g/t Co, g/t Pb, % As, %
1
st
campaign 2.72 1.76 374 572 1.51 0.090
2
nd
campaign 2.60 1.89 328 613 - 0.085
CONTENT
Ag, g/t Zn
,
g/t S
c,
% Sso
4
, % C
total,
% C
org
, %
1
st
campaign 190 1200 2.95 1.45 14.30 8.96
2
nd
campaign 168 740 2.62 2.47 10.4 6.30
Figure 1 exhibits the mineralogical composition of Lubin middlings and compares it with composi-
tion of Lubin final concentrate, currently produced as a feed for smelting. This analysis was made
by BRGM (Orleans, France) laboratories and became a very important justification in discussion on
considered technology alterations. The mineralogical composition of flotation concentrate appeared
to be very similar to the composition of the ore. Chalcocite (Cu
2
S) and bornite (Cu
5
FeS
4
), easiest to
leach copper sulfides, are dominating in the material. The content of chalcopyrite (CuFeS
2
), most
refractory copper sulfide, was only about 20.6 %.
In contrary to concentrate, mineralogical composition of shale middlings appeared to be enormously
favorable for hydrometallurgical treatment. Chalcocite was reported to be dominating (90.5 %) with
some content of bornite (9 %) (Figure 1). Chalcopyrite is a negligible component in this by-product.
Such a mineralogical composition can be considered as an ideal for hydrometallurgical or biometal-
lurgical treatment. The observed mineralogical composition of shale middlings exhibits evidently,
Chmielewski, Wodka
Proceedings of Copper 2010 2678
that beneficial mineralogical segregation of minerals takes place during the 1
st
cleaning flotation
process at Lubin Concentrator.
Figure 1: Mineralogical composition of Lubin final concentrate and Lubin shale middlings
(BRGM data).
It is very obvious that shale fraction has to be processed not only as a copper-bearing raw material
but mainly as a polymetallic one. Separation of this fraction from flotation circuit and its processing
for recovering of Cu and other metals is therefore fully justified.
Particle size analyses for Lubin middlings (Figure 2) indicate that applied middlings was rather to
coarse in terms of leaching rate. Parameter d
80
exceeded 100 m even after non-oxidative leaching
with H
2
SO
4
(carbonates decomposition, R
w
= 30-90 %).
Figure 2: Particle size analyses of Lubin middlings raw and subjected to the non-oxidative acidic
leaching at various degree of carbonate decomposition, R
w
(within 30-90 %).
20
30
40
50
60
70
80
90
100
10 100 1000
Particle size, m
C
u
m
u
l
a
t
i
v
e
r
e
c
o
v
e
r
y
,
%
.
0
Rw - 30%
Rw - 50%
Rw - 70%
Rw - 90%
Pressure Leaching of Shale Middlings in Oxygenated Sulphuric Acid
Proceedings of Copper 2010 2679
Additional microscopic SEM (Figure 3) observations lead to the significant conclusion, that most
coarse particles are formed by shale material, which can not be chemically decomposed during
treatment with acid, even in the presence of oxygen (Figure 3D). Therefore, separation of particle
fraction above 50-60 m and its additional regrinding seem to be very desirable to facilitate further
leaching.
Figure 3: SEM mineralogical analyses of Lubin shale middlings before leaching (A, B, C) and
after atmospheric leaching in acidic conditions (D).
2.2 Pressure leaching procedure
The effect of temperature, oxygen partial pressure and sulphuric acid concentration on the kinetics
and efficiency of pressure leaching of Lubin middlings have been investigated. All experiments
were carried out at temperatures between 100 and 180
o
C and under oxygen pressure within the
range of 2.5-10 atm. The pressure leaching was always preceded with non-oxidative acidic pre-
treatment of the feed in order to totally decompose the acid-consuming componets, predominantly
carbonates.
A
B
D
C
Chmielewski, Wodka
Proceedings of Copper 2010 2680
The experiments were performed in 2.0 dm
3
autoclave having the stirrer and sampling pipe made of
teflon. The following experimental procedure was used: A teflon beaker containing 1.0 dm
3
of solu-
tion of the required concentration of sulphuric acid and the required amount of the middlings was
introduced into the autoclave. The liquid to solid phase ratio (s/l) = 1:10 was kept in all experi-
ments. The stirring was switched on when the non-oxidative decomposition of carbonates with
H
2
SO
4
started prior to the oxidative leaching. After carbonates decomposition (ca. 60 min.) auto-
clave lid was installed and the slurry was purged two times with nitrogen. The heating was switched
on and as the temperature rose to 100
o
C the nitrogen was removed from the autoclave.
After the temperature reached the require level the solution zero sample was drawn off for chemi-
cal analyses. Subsequently, the mixture of 50 % of oxygen and 50 % of nitrogen was introduced into
the autoclave to establish the required oxygen partial pressure. During the experiments samples of
the solution were taken periodically to determine metals concentration (Cu, Zn, Ni, Co, As) by
AAS.
Solid samples of middlings before and after each pressure leaching test were examined by minera-
logical SEM microscopy to evaluate the composition of the solid and to assess the effectiveness of
applied leaching parameters range.
3 Results and discussion
3.1 Effect of temperature on pressure leaching of the shale middlings
Pressure leaching of Lubin middlings was examined at elevated temperatures from 100 to 180
o
C
while concentrations of Cu, Fe, Ni, Co and As were analysed in the liquid samples taken during the
leaching. Experimental results are shown in Figures 4 for Cu, Fe, and Co, respectively. The ob-
served effect of temperature on the leaching of metals was quite complex and the following obser-
vations have been done:
Leaching of copper in oxygenated sulphuric acid at 100
o
C required about 120 min. of activation
time. This was most likely due to the formation of H
2
S observed at the initial stage of the process.
Remarkable acceleration of the process was subsequently detected after 120 min. of activation.
Unexpectedly, at 120
o
C no copper leaching was detected. This might be most likely explained as a
hindering effect produced by the presence of elemental sulphur at its melting point.
At temperatures exceeding 140
o
C the leaching appeared to be very rapid, although some decrease
of Cu concentration was observed at 160 and 180
o
C versus 140
o
C. Co-precipitation of iron com-
pounds or sorption on organic matter are the only explanation of observed effects.
The rate of leaching of Fe from Lubin middlings increases with temperature up to 120
o
C, then iron
concentration slightly drops as the result of precipitation of iron(III) oxide (Fe
2
O
3
) or goethite
(FeOOH). The appearance of soluble iron in leaching solution results from leaching of Cu-Fe sul-
phides, mainly bornite Cu
5
FeS
4
and chalcopyrite CuFeS
2
.
Pressure Leaching of Shale Middlings in Oxygenated Sulphuric Acid
Proceedings of Copper 2010 2681
0
500
1000
1500
2000
2500
3000
0 30 60 90 120 150 180 210
Leaching time, min
C
u
c
o
n
c
e
n
t
r
a
t
i
o
n
,
m
g
/
d
m
3
.
100
120
140
160
180
Cu
800
1000
1200
1400
1600
1800
2000
2200
2400
2600
0 30 60 90 120 150 180 210
Leaching time, min
F
e
c
o
n
c
e
n
t
r
a
t
i
o
n
,
m
g
/
d
m
3
.
100
120
140
160
180
Fe
0
10
20
30
40
50
0 30 60 90 120 150 180 210
Leaching time, min
C
o
c
o
n
c
e
n
t
r
a
t
i
o
n
,
m
g
/
d
m 3
100
120
140
160
180
Co
Figure 4: Effect of temperature on the pressure
leaching of copper, iron and cobalt
from Lubin middlings.
Oxygen partial pressure: 0.5 MPa,
concentration of H
2
SO
4
: 5 %,
solid to liquid ratio, s/l: 1:10.
Pressure leaching of cobalt was practically not observed at temperatures 100 and 120
o
C with detected
concentration of cobalt about 5 mg/dm
3
. Remarkable acceleration of leaching of Co was recorded at
temperatures above 140
o
C and reported concentration of cobalt increased to 40-50 mg/dm
3
.
Quite complex kinetic curves for nickel (not presented here) indicated that at 120
o
C the highest
leaching recovery was observed with the final Ni concentration of about 180 mg /dm
3
. At tempera-
tures exceeding 140
o
C the leaching rate of nickel decreases, which are hard to explain at this stage
Chmielewski, Wodka
Proceedings of Copper 2010 2682
of investigation and require detailed analysis of leaching residues, particularly determination of the
mineralogical forms of metals and gangue.
The concentration vs. leaching time relationship for leaching of arsenic was similar to those of co-
balt. Pressure leaching of arsenic was nearly not observed at temperatures 100 and 120
o
C when
detected concentration of As was about 10 mg/dm
3
. Significant increase in leaching rate of As was
observed at temperatures above 140
o
C while concentration of arsenic increased to 70-90 mg/dm
3
.
From results of pressure leaching it is well seen, that very effective recovery of metals from Lubin
middlings can be observed for experiments performed at temperatures exceeding 140
o
C. At 140
o
C
observed were the highest Cu, Co and As recoveries. It was also detected that at temperatures ex-
ceeding 180
o
C iron control became possible as a result of precipitation of FeOOH or Fe
2
O
3
. Arsen-
ic, a harmful contaminant, will require a removal or stabilisation prior to recovering process of cop-
per, nickel and cobalt. Solvent extraction or precipitation of crystalline scorodite FeAsO
4
2H
2
O at
temperatures above 180
o
C can be taken into consideration.
The relationship presented in Figure 4 are characterized by rapid copper, iron and cobalt leaching
during the initial 60 minutes. Subsequently, a slower increase in metals concentration is observed.
The first, 30 to 60 minutes, very rapid step of leaching, most likely due to initial fast leaching of the
easiest leachable copper minerals bornite and chalcocite, dominating in the shale fraction from
Lubin Concentrator. The second, slower step, is possibly due to leaching of chalcopyrite and
covellite the most refractory copper minerals.
The observed concentration of iron(II) ions in the solution before introduction of oxygen is about
1.0 g/dm
3
. It suggests that iron(III) partly occurred as an oxide in the shale dissolves in acidic condi-
tions. Maximum recovery of copper (97 %) and cobalt (80 %) was observed at 140
o
C and above
90 % for Fe at 160
o
C. However, in the investigated range of temperatures nickel did not leach well.
Nickel leaching recovery rarely exceeded 30-40 % in most experiments. Neither lead no silver was
apparently detected in the solution.
3.2 Effect of initial sulphuric acid concentration on Cu, Fe, and Co
leaching
The effect of sulphuric acid concentration on the metals leaching was investigated in the range of 20
to 50 g/dm
3
at temperature of 140
o
C and oxygen partial pressure 5.0 MPa. Figure 4 presents con-
centration leaching time relationships for copper and cobalt at different sulphuric acid concentra-
tions. Copper doesnt leach before introduction of oxygen and course of copper leaching resembles
those observed during investigations of the effect of temperature.
Pressure Leaching of Shale Middlings in Oxygenated Sulphuric Acid
Proceedings of Copper 2010 2683
0
500
1000
1500
2000
2500
3000
0 30 60 90 120 150 180 210
Leaching time, min. .
C
o
p
p
e
r
c
o
n
c
e
n
t
r
a
t
i
o
n
,
m
g
/
d
m
3
.
2%
3%
4%
5%
0
10
20
30
40
50
60
0 30 60 90 120 150 180 210
Leaching time, min.
C
o
b
a
l
t
c
o
n
c
e
n
t
r
a
t
i
o
n
,
m
g
/
d
m
3
.
2%
3%
4%
5%
Figure 5: Effect of sulphuric acid concentration on copper and cobalt pressure leaching.
(Temperature: 140
o
C, oxygen partial pressure: 5.0 atm, liquid-solid phase ratio: 10:1,
rate of mixing: 400 rpm).
Figure 6: Effect of sulphuric acid concentration on metals recovery by pressure leaching from
Lubin middlings.
Only for concentration of sulphuric acid of 30 g/dm
3
the highest leaching rate and the maximum
concentration of copper in the solution were observed. There are two steps of pressure leaching. The
ferric ion starts to leach sulphidic minerals before the oxygen introduction. There were also ob-
served some differences in ferric concentration depending on sulphuric acid concentration.
0
10
20
30
40
50
60
70
80
90
100
2,0 2,5 3,0 3,5 4,0 4,5 5,0
L
e
a
c
h
i
n
g
r
e
c
o
v
e
r
y
,
%
.
H
2
SO
4
concentration, %
Cu
Fe
Ni
Co
As
Chmielewski, Wodka
Proceedings of Copper 2010 2684
For 20 g/dm
3
of H
2
SO
4
the concentration of ferric ions in solutions was 680 mg/dm
3
and for
50 g/dm
3
of H
2
SO
4
the concentration of Fe(III) increased to 940 mg/dm
3
.
Generally, sulphuric acid concentration was found to have an effect on middlings ferric leaching.
For concentration of sulphuric acid below 20 g/dm
3
the precipitation of ferric compounds have oc-
curred after 60 minutes of leaching. There were also two leaching steps for 30, 40, and 50 g/dm
3
concentrations of sulphuric acid, similarly to copper leaching. The highest concentration of ferric
ions in solution has been reached for sulphuric acid exceeding 30 g/dm
3
. Kinetic curves observed
for Cu and Co leaching at acid concentrations above 30 g/dm
3
were almost identical.
There was no cobalt leaching detected before introduction of oxygen. The course of cobalt leaching
resembles the copper leaching and there were observed two steps of it. The first step with the high
rate of cobalt dissolution and the second one with evidently slow cobalt dissolution rate.
In general, a little effect of sulphuric acid concentration on cobalt and copper leaching has been
found from H
2
SO
4
concentrations above 3 % (Figure 7). A 97 % of leaching recovery of copper and
80 % of leaching recovery of cobalt recovery was reached regardless to sulphuric acid concentra-
tion. The highest iron recovery occurred for 30 g/dm
3
of H
2
SO
4
. Presented results suggest that high
recovery of copper with simultaneous partial precipitation of ferric compounds is possible. How-
ever, it was evident, that at the lowest H
2
SO
4
concentration (2 %) the precipitation of Fe(III) from
the solution was observed, most likely as FeOOH and Fe
2
O
3
and the concentration of Fe in solution
decreased to about 1.3 g/dm
3
. This precipitation of Fe was not detected at higher acid concentration
(3-5 %). Iron control in the leaching solution at elevated temperature can be useful as a method of
purification from the excesses of Fe prior the recovering of metals. His effect will be further inves-
tigated in details.
3.3 Effect of oxygen partial pressure on Cu, Fe and Co leaching
The effect of oxygen partial pressure was investigated in the range from 2.5 to 10 MPa. The leach-
ing temperature (140
o
C), sulphuric acid concentration (50 g/dm
3
) and liquid/solid ratio ratio (10:1)
were kept constant. Figure 7 presents concentration leaching time relationships for copper and
cobalt. Two steps of leaching courses were found, similarly to the other experiments. Effect of oxy-
gen partial pressure on leaching recovery of Cu, Fe, Ni, Co and As were collected in Figure 7. Lead
and silver were not solubilised during pressure leaching.
On the basis of results presented in Figure 8 it can be seen that pressure leaching is a very efficient
process. Oxygen partial pressure has a noticeable effect on copper, cobalt, arsenic and ferric solubi-
lisation. The highest recovery of metals was detected at oxygen partial pressure of 7.5 MPa. Solubi-
lisation of nickel was highly reduced most likely due to the dissemination of Ni in pyritic phase,
according to mineralogical examinations by BRGM, Orleans [28].
Pressure Leaching of Shale Middlings in Oxygenated Sulphuric Acid
Proceedings of Copper 2010 2685
0
500
1000
1500
2000
2500
3000
3500
0 30 60 90 120 150 180 210
Leaching time, min.
C
u
c
o
n
c
e
n
t
r
a
t
i
o
n
,
m
g
/
d
m
3
.
10 atm
7,5 atm
5 atm
2,5 atm
0
10
20
30
40
50
60
0 30 60 90 120 150 180 210
Leaching time, min
C
o
c
o
n
c
e
n
t
r
a
t
i
o
n
,
m
g
/
d
m
3
.
10 atm
7,5 atm
5 atm
2,5 atm
Figure 7: Effect of oxygen partial pressure on Cu and Co leaching rate for the pressure leaching of
Lubin middlings. Temperature: 140
o
C, concentration of H
2
SO
4
: 5 %, liquid/solid ratio: 10:1.
Figure 8: Effect of oxygen partial pressure on metals recovery from Lubin middlings.
Temperature: 140
o
C, H
2
SO
4
: 50 g/l, s/l: 1:10.
Table 3 summarizes the pressure leaching experiments for Cu, Co, Ni, As and Zn on the basis of
solid residue analysis. According to the presented data, only nickel leaching recovery was observed
to be below 40 %. Copper (98.9 %), cobalt (82.9 %) and zinc (98.2 %) can be easily leached out
0
10
20
30
40
50
60
70
80
90
100
2 3 4 5 6 7 8 9 10
L
e
a
c
h
i
n
g
r
e
c
o
v
e
r
y
,
%
.
Oxygen partial pressure, atm .
Cu
Fe
Ni
Co
As
Chmielewski, Wodka
Proceedings of Copper 2010 2686
from the examined Lubin middlings. Pressure leaching can be therefore considered as an efficient,
potential alternative for hydrometallurgical processing of Lubin shale middlings.
Table 3: Recovery of metals in pressure leaching of Lubin middlings in oxygenated sulphuric acid.
Sample
Tempe-
rature,
o
C
oxygen par-
tial pressure,
atm
H
2
SO
4
concen-
tration
,
%
Leaching recovery, %
Cu Co Fe Ni As Zn
1.
100 5 5 68.8 8.1 6.6 6.2 8.1 87.5
2.
120 5 5 38.1 33.6 12.6 27.2 30.7 88.3
3.
140 5 5 98.8 80.3 70.3 28.2 50.3 97.5
4.
140 10 5 98.7 81.7 77.3 22.7 41.9 97.4
5.
140 7,5 5 98.9 82.5 83.6 36.2 50.5 97.4
6.
140 2,5 5 97.9 77.5 82.6 30.4 48.5 96.8
7.
140 5 2 98.2 82.9 38.5 39.3 44.0 96.7
8.
140 5 3 98.4 81.2 70.0 33.2 50.4 97.4
9.
140 5 4 98.6 79.3 79.5 32.8 51.5 97.2
10.
160 5 5 83.4 71.7 86.4 25.5 48.4 98.1
11.
180 5 5 89.8 78.6 69.4 29.1 45.8 98.2
12.
140 5 5 98.8 81.2 61.2 31.3 47.1 97.2
13.
140 5 5 98.3 80.2 78.5 26.8 52.8 96.6
Solid residue after pressure leaching was examined by SEM (Figure 9). It can be seen that un-
leached metal-bearing minerals (mostly covellite, chalcopyrite, pyrite and galena) are finely disse-
minated in shale organic matter (Figures 9B, C and D), which was practically not decomposed nei-
ther during non-oxidative nor during pressure leaching. To liberate these particles for further metals
recovering required is either further intensification of pressure leaching or additional milling to re-
duce particle size of unleached solid grains prior the leaching.
Some unleached minerals (CuS, CuFeS
2
) remain in the solid as quite coarse, spongy structure re-
quiring either longer leaching time or more intensive leaching parameters (Figure 9 A). Reducing of
particle size in the leaching feed is also strongly recommended.
Pressure Leaching of Shale Middlings in Oxygenated Sulphuric Acid
Proceedings of Copper 2010 2687
3.4 Characterization of pressure leaching residue by SEM examinations
Figure 9: SEM pictures of pressure leached samples of Lubin middlings exhibiting fine dissemi-
nation of metals-bearing minerals in shale matter.
4 Conclusions
Shale-containing by-product (middlings)-tailings of 1
st
cleaning from flotation circuits at Lubin
Concentrator, which exhibited remarkably elevated content of organic carbon, can be efficiently
processed hydrometallurgically by non-oxidative leaching followed by pressure leaching. High car-
bonate content, a specific and unique feature of Polish copper ores, and fine dissemination of metal-
bearing minerals (predominantly sulphides) unquestionably require a chemical pretreatment of the
shale feed with H
2
SO
4
prior the atmospheric and pressure leaching in oxygenated solutions of sul-
phuric acid.
Non-oxidative leaching of Lubin middlings (Cu 2.7 %, Pb 1.52 %, Ni 374 ppm, Co 572 ppm,
Ag 170 ppm) is a very rapid, selective, and relatively simply-controlled process. Selective liberation
of metal sulphides improves their flotatability while total decomposition of carbonates makes further
pressure leaching more efficient.
B
C
D
A
Chmielewski, Wodka
Proceedings of Copper 2010 2688
Pressure leaching of Lubin middlings revealed, that the process is remarkably fast and efficient for
recovering of Cu, Fe, Co and As with much lower recovery of Ni. To accomplish maximum recovery
of metal, pressure leaching has to be conducted at temperatures above 140
o
C and under oxygen pres-
sure exceeding 5 atm. From pressure leaching experiments performed at temperatures 120-180
o
C,
under oxygen pressure 2.5-10 atm, at H
2
SO
4
concentrations 2-5 %, and at solid/liquid ratio 1:10 it re-
sults that about 96-97 % of Cu, 96 % of Fe 96 % of As, and 82 % of Co can be recovered after about
2 hours of leaching. Much lower leachability of nickel (30-40 %) can be explained in terms of its dis-
semination in pyrite. It was also observed that at temperatures exceeding 160
o
C precipitation of Fe as
goethite or hematite commences, which can be applied as an iron control process for solution purifica-
tion.
Pressure leaching with oxidized H
2
SO
4
solutions was not efficient to leach of Ag, Pb, and precious
metals. The solid residue after leaching must be either upgraded by flotation or subjected to leaching
in chloride solutions to recover remaining metals.
Acknowledgements
This work was carried out in the frame of Bioshale (European project contract NMP2-CT-2004
505710). Author acknowledges the financial support given to this project by the European Commis-
sion under the Sixth Framework Programme for Research and Development. We also wish to thank
our various partners on the project for their contributions to the work reported in this paper.
References
[1] COLLINS M.J., STIKMA J., BUBAN K.R., MASTERS J.M., (2000): Pressure acid leacing
of zinc and copper concentrates by Dynatec, EPD Congress, Ed. P.R Taylor, TMS, Nashville,
TN: USA, 2000, pp. 597-605.
[2] JANKOLA W.A., (1995): Zinc pressure leaching at Cominco. The principles and practice of
leaching. Proc. Int. Symp.: Principles and Practice of Leaching., The 25th Annual Hydrome-
tal-lurgical Meeting of the Metallurgical Society of CIM, Winnipeg, Canada, October 15-18,
1995, Ed. W.C. Cooper and D. B. Dreisinger, pp. 63-71.
[3] RUBISOV. D.H., PAPANGELAKIS V.G., (2000): Sulphuric acid pressure leaching of later-
ites speciation and prediction of metals solubilities at temperature, Hydrometallurgy, 58,
pp. 13 28.
[4] RUBISOV D.H., KROWINKIEL J.M., PAPANGELAKIS V.G., (2000): Sulphuric acid pres-
sure leaching of laterites - universal kinetics of nickel dissolution for limonites and limonit-
ic/saprolitic blends, Hydrometallurgy, 58, pp. 1-11.
Pressure Leaching of Shale Middlings in Oxygenated Sulphuric Acid
Proceedings of Copper 2010 2689
[5] NICE R.W., (2004): The Western Australian nickel laterites project What we have learned?,
Pressure Hydrometallurgy 2004, Proc. 34
th
Annual Hydrometallurgy Meeting, Banff, Alberta,
Canada, Ed. By M.J. Collins and V.G. Papangelakis, 155-180.
[6] KING J.A. AND DREISINGER D.B, (1995): Autoclaving of Copper Concentrates, Proc. Int.
Conf. Copper-Cobre 95, ed. Cooper W.C. et al., The Metallurgical Society of CIM, Montreal,
1995, 511-532.
[7] JONES D.L. AND HESTRIN J., (1998): CESL Process for Copper Sulphides: Operation of
the Demonstration Plant, The Alta 1998 Copper Sulphides Symposium, Brisbane, Australia,
October 19, 1998.
[8] BARTA L.A., BUBAN K.R., STIKSMA J. AND COLLINS M. J., (1999): Pressure Leaching
of Chalcopyrite Concentrates by Dynatec, Proc. Int. Conf. Copper Cobre 99, Ed. Young
S.K et al., vol. IV, 167 180.
[9] JONES D.L. AND HESTRIN J., (1998): CESL Process for Copper Sulphides: Operation of
the Demonstration Plant, The Alta 1998 Copper Sulphides Symposium, Brisbane, Australia,
October 19, 1998.
[10] DREISINGER D.B., STEYEL J. D.T., SOLE K.C., GNOIMSKI J., DEMPSEY P., (2003):
The Anglo-American Corporation/University of British Columbia (AAC/UBC) Chalcopiryte
process: An integrated pilot plant evaluation, Int. Copper-Cobre 2003 Conf., Santiago,
Chile, Nov. 30 Dec. 3, Volume VI Hydrometallurgy of Copper (Book 1), 2003, 223 - 238.
[11] ANDERSON C.G., (2003): The application and economics of NSC catalyzed pressure leach-
ing of copper ores and concentrates., Copper-Cobre 2003 Conf., Santiago, Chile, Nov. 30
Dec. 3, vol. VI Hydrometallurgy of Copper (Book 1), 2003, 289 306.
[12] FLEMING C.A., FERRON C.J., DREISINGER D.B., O KANE., (2000): A Process for the
Simultaneous Leaching and Recovery of Gold, Platinum Group Metals and Base Metals From
Ores and Concentrates, EPD Congress 2000, The Minerals, Metals and Materials Soci-
ety(TMS), 2000 TMS Annual Meeting, March 12 -1 6, Ed.: Taylor P.R, 419 431.
[13] THOMAS G.K., (1991): Alkaline and Acidic Autoclaving of Refractory Gold Ores, J. Metals,
43(2), pp. 16-19.
[14] THOMAS G.K., WILLIAMS R.A., (2000): Alkaline and Acid Autoclaves at Barrick Gold, A
Review, EPD Congress 2000, Ed.: P.R., Taylor, p. 433 449.
[15] MARSDEN J.O., IAIN HOUSE C., (2006): The Chemistry of Gold Extraction, 2nd Edition,
Soc. Min. Met. Expl. Inc.
Chmielewski, Wodka
Proceedings of Copper 2010 2690
[16] CURLOOK W., PAPANGELAKIS V.G., BAGHALHA M., (2004): Pressure acid leaching of
non-ferrous smelter slags for the recovery their base metal values, Pressure Hydrometallurgy
2004, Proc. 34
th
Annual Hydrometallurgy Meeting of CIM, October 23-27, 2004, Banf, Cana-
da, (Collins M.J., and Papangelakis, Eds.), MetSoc, pp. 823- 838.
[17] RAMACHANDRAN V., LAKSHMANAN V.I., KONDOS P.D., (2007): Hydrometallurgy of
copper sulphide concentrates: An update, Proc. 6
th
Int. Conf. Copper-Cobre 2007, Toronto,
Canada, (Eds.: Riveros P.A., Dixon D.G., Dreisinger D.B.) vol. IV (book 1), Can. Inst. Min.
Metal. Petrol., pp. 101 128.
[18] MARSDEN J.O., WILMOT J.C., (2007): Sulphate-based process flowsheet options for hydro-
metallurgical treatment of copper sulphide concentrates, Proc. 6
th
Int. Conf. Copper-Cobre
2007, Toronto, Canada, (Eds.: Riveros P.A., Dixon D.G., Dreisinger D.B.) vol. IV (book 1),
Can. Inst. Min. Metal. Petrol., pp. 77 100.
[19] KUBACZ N., SKORUPSKA B., (2007): Evaluation of the effect of organic carbon on flota-
tion and smelting processes. Conf. Proc.: VIII International Conference On Non-ferrous Ore
Processing, ICNOP07, 21-23 May 2007, Wojcieszyce, Poland, KGHM, IMN, pp. 157-166 (in
Polish).
[20] KONIECZNY A., KASINSKA PILUT E., PILUT R., (2009): Technological and technical
problems in mineral processing of Polish copper ores at KGHM Polska Miedz SA, Proceed-
ings: IX International Conference on Nonferrous Ore Processing, 18-20 May 2009, Ladek
Zdroj Trzebieszowice (Poland), KGHM Cuprum, pp. 11-21 (in Polish).
[21] CHMIELEWSKI T., CHAREWICZ W., (2006): Hydrometallurgical processing of shale by-
products from beneficiation circuits of Lubin Concentrator, In: Perspectives for applying bo-
leaching Technology to process shale-bearing copper ores, BIOPROCOP06, Lubin 2006,
KGHM Cuprum, Wroclaw 2006, 125-145 (in Polish).
[22] DHUGUES P., NORRIS P.R., HALLBERG K.B., SNCHEZ F., LANGWALDT J.,
GROTOWSKI A., CHMIELEWSKI T., T. GROUDEV T., (2008): Bioshale consortium, Bio-
shale FP6 European project: Exploiting black shale ores using biotechnologies?, Minerals En-
gineering 21, pp. 111120.
[23] CHMIELEWSKI T., (2007): Non-oxidative Leaching of Black Shale Copper Ore from Lubin
Mine, Physicochemical Problems of Mineral Processing, 41, pp. 323-348.
[24] CHMIELEWSKI T., LUSZCZKIEWICZ A., KONOPACKA Z., (2007): Separation and Con-
cept of Processing of Black Shale Copper Ore from Lubin Mine, Proc. VIII International Con-
ference on Non-ferrous Ore Processing, Wojcieszyce (Poland), May 21-23, KGHM Cuprum,
Wroclaw, pp. 171-184 (in Polish).
Pressure Leaching of Shale Middlings in Oxygenated Sulphuric Acid
Proceedings of Copper 2010 2691
[25] KONOPACKA Z., LUSZCZKIEWICZ A., CHMIELEWSKI T., (2007): Effect of non-
oxidative leaching on flotation efficiency of Lubin Concentrator Middlings. Physicochem.
Probl. Miner. Process., 41, pp. 275-289.
[26] CHMIELEWSKI T., (2007): Atmospheric leaching of shale by-product from Lubin Concen-
tra-tor., Physicochem. Probl. Min. Process., 41, pp. 337-348.
[27] WODKA J., CHMIELEWSKI T., ZIOLKOWSKI B., (2007): Pressure leaching of shale ore in
oxygenated sulfuric acid, Physicochemical Problems of Mineral Processing, 41, pp. 349-364.
[28] AUGE T., et al., (2007): BIOSHALE, Deliverable D.5.2., Product behavior during the hydro-
and bioprocesess (T12-30), March 2007.
Proceedings of Copper 2010 2692
Proceedings of Copper 2010 2693
Mine to Heap in Mantoverde Anglo American
Division
Manuel Daz, Cristian Salgado, Carlos Prez, Cristian Alvayai, Leonardo Herrera
Gabriel Zrate
Anglo American Chile
VicePresidency of Metallurgy
Pedro de Valdivia 291
Santiago, Chile
Keywords: Hydrometallurgy, copper, heap leaching, acid consumption
Abstract
Mantoverde is located in Chile, at 900 m above sea level, 53 km SE of Chaaral city. It currently
processes two ore types: heap ore (75 million ton, 0.73 % total copper, 0.61 % acid soluble copper
and 1.23 % calcium carbonate) and dump ore (40 million ton, 0.48 % total copper, 0.37 % acid so-
luble copper and 2.24 % calcium carbonate). In 2008, copper production was 62,500 tons.
During 2007, the following problems were experienced in heap leaching and solvent extraction:
Lower copper recovery, being the predicted recovery lower than the current one.
Higher concentration of aluminum, iron, magnesium and manganese in the pregnant leach solu-
tion and as result higher solution viscosity affecting the leach and solvent extraction operation
significantly.
Higher acid consumption, being the predicted one lower than the current acid consumption and
higher calcium carbonate content.
A multidisciplinary team was formed to investigate the reasons and to set up a working program in
order to optimize these performance indicators. The main metallurgical improvements were:
An increase in the acid soluble copper recovery by 3.2 points and a decrease in acid consumption
by 20 %.
A decrease in leaching solution viscosity by 20 %.
The evaluation of the operational parameters and the heap results obtained, after the recommenda-
tions were implemented, are discussed in this paper.
Daz, Salgado, Prez, Alvayai, Herrera, Zrate
Proceedings of Copper 2010 2694
1 Introduction
The Mantoverde mine is located in the Chaaral province, Region III, in Northern Chile, 1000 km
north of Santiago and 40 km from the coast. The altitude of the mine is 900 m above sea level. Man-
toverde was designed to produce 42,130 ton of copper per year at a treatment rate of 5.4 million ton
of ore per year by using a heap leach-SX-EW process. Mantoverde uses a two stage leach approach,
where fresh ore is sprinkled with an intermediate solution (ILS) to produce PLS in a first stage and
then is irrigated with raffinate solution to produce ILS in a second stage. A dump leach process was
scheduled to be started 5 years after the plant start-up, at a treatment rate of 2 million ton of margin-
al ore per year.
Mantoverde (MVN) was commissioned in December 1995. Copper production in 1996 was
40,539 ton and it has steadily been increased to reach 62,500 ton in 2008, due to the start-up of a
similar size orebody, called Mantoverde Sur (MVS) and of another smaller orebody called Manto
Ruso (MR). In 2002, heap leach operation was changed from a permanent pad to a dynamic pad.
Forecast copper production for 2010 is 62,000 ton for an ore treatment rate of 9.6 million ton. A
first dump leach operation was started in 1999 with a copper production of 1663 ton and a second
dump leach operation was started early in 2002. Copper production from this source will increase
over time to as much as 10,000 ton. The mine statistics for the first years of operation is shown in
Table 1.
Table 1: Mantoverde production statistics.
Year Throughput (ktpy) Production (tpy)
Heap Dump Heap Dump
1996 5.436 ---------- 40.539 ----------
1997 6.156 ---------- 47.627 ----------
1998 6.214 ---------- 48.032 ----------
1999 6.367 1.891 50.364 1.663
2000 7.144 3.233 49.456 4.152
2001 8.001 4.135 50.365 5.205
2002 8.397 4.573 51.448 5.841
2003 9.001 6.048 53.250 6.707
2004 9.017 7.028 52.016 8.095
2005 9.439 3.625 56.435 5.565
2006 9.502 4.880 53.713 6.608
2007 9.281 5.511 54.951 6.050
2008 9.557 4.300 56.618 5.883
Mine to Heap in Mantoverde Anglo American Division
Proceedings of Copper 2010 2695
During 2007, the following problems were experienced in heap leaching and solvent extraction:
Lower copper recovery, being the predicted recovery lower than the current one.
Higher concentration of aluminum, iron, magnesium and manganese in the pregnant leach solu-
tion and as result, an increase in the solution viscosity affecting the leach and solvent extraction
operation significantly.
Higher acid consumption, being the predicted one lower than the current acid consumption and
higher ore calcium carbonate content.
A team was formed to investigate the reasons and to set up a working program in order to optimize
these performance indicators. For that purpose, the operating data of 74 heaps was reviewed and a
column test program was carried out. Results were as follows.
2 Analysis of Operating Data
The operating data of 74 heaps, covering several years of operation, was reviewed and the effect on
copper extraction of variables such as total copper (Cut), acid soluble copper (Cusol) and CaCO
3
grades, heap height, net acid consumption, acid dosage in curing stage, particle size, flow rate, acid
concentration in leaching solutions and ore feed distribution was assessed.
It was concluded that heap height was one of the most important variables affecting negatively the
copper extraction, as it can be seen in Figure 1. Acid addition in the curing stage was another impor-
tant variable.
Figure 1: Cusol recovery and heap height. Average values.
70,0
72,0
74,0
76,0
78,0
80,0
82,0
84,0
86,0
88,0
90,0
5,5 6,0 6,5 7,0 7,5 8,0
Height (m)
C
u
s
o
l
R
e
c
o
v
e
r
y
(
%
)
Daz, Salgado, Prez, Alvayai, Herrera, Zrate
Proceedings of Copper 2010 2696
It was also found that heap height and CaCO
3
grade were the most important variables increasing
acid consumption. Accordingly, it was recommended to decrease gradually heap height from 7.5 m
to 5 m and to optimize acid addition distribution by reducing acid dosage in curing stage and in-
creasing proportionally acid concentration in leaching solutions [1].
3 Column Test Work
A number of column tests have been conducted using ore samples taken from the MVN, MVS, MR
and a new ore body called Kuroki, and ILS and raffinate solution samples. The main goal of this
program was to update copper recovery equations for different Cusol and CaCO
3
grades and to op-
timize acid addition management.
The results of this work were reported elsewhere [2]. It has been demonstrated that a 50 % decrease
in the acid added to the curing stage, combined with an increase in the acid concentration of ILS and
raffinate solution, had a significant reduction in acid consumption without affecting negatively the
copper extraction nor the leach cycle, as it can be seen for example in the following figures, where
column 31 was cured and leached in the standard way while column 33 was cured with 50 % less
acid and leached with solutions having 50 % more acid, as compared to standard solutions.
Figure 2: Copper extraction kinetics of columns 31 and 33.
0,0
10,0
20,0
30,0
40,0
50,0
60,0
70,0
80,0
90,0
100,0
0,0 0,5 1,0 1,5 2,0 2,5 3,0 3,5
m3/t ore
C
u
s
o
l
R
e
c
o
v
e
r
y
(
%
)
Col 31
Col 33
Mine to Heap in Mantoverde Anglo American Division
Proceedings of Copper 2010 2697
As a consequence, it was recommended to decrease gradually the acid addition to the curing step
from 90 % to 70 % and then to 60 %, being 50 % the final target, and to adjust the acid concentra-
tion of ILS and raffinate solution from 3 to 8 g/l and from 13 to 15 g/l respectively.
Figure 3: Cusol extraction and net acid consumption of columns 31 and 33.
0,0
10,0
20,0
30,0
40,0
50,0
60,0
70,0
80,0
90,0
100,0
0,0 10,0 20,0 30,0 40,0 50,0 60,0
Net Acid Consumption (kg/t ore)
C
u
s
o
l
R
e
c
o
v
e
r
y
(
%
)
Col 31
Col 33
Daz, Salgado, Prez, Alvayai, Herrera, Zrate
Proceedings of Copper 2010 2698
4 Commercial Heap Results
The heap height was reduced from 7.5 to 6.5 m by the end of August 2007, resulting in a Cusol re-
covery increase as it is shown in Figure 4, where predicted and actual recoveries are graphed.
Figure 4: Predicted and actual heap Cusol recovery.
Concerning acid addition, it was established in the design criteria that 80 % of the net acid consump-
tion was going to be added in the curing stage with the remaining 20 % through the leach solutions,
namely intermediate leach solution (ILS) and raffinate solution. At Mantoverde, the acid consump-
tion is related to the ore calcium carbonate content which is measured every three hours and the acid
addition to the curing stage is adjusted accordingly. Since the start up and until 2007, the acid was
added following this methodology. Acid addition in the curing stage was increased to 90 % and dur-
ing some period of time it reached 100 %.
The recommendations on acid addition were implemented during the first semester of 2008 at the
heap leach operation with excellent results, at it is shown in Figures 5 and 6. In the first one, the net
acid consumption, actual and estimated, and calcium carbonate grades are graphed for years 2001
through 2007, while in the second one is done for year 2008, from January through August.
It can be seen that the actual net acid consumption started to be higher than the estimated one in
2004 and that this trend has been reversed since June 2008, once the recommendations were fully
implemented. The reduction in net acid consumption was on average 5.8 kg/t, about 20 %, equiva-
70,0
72,0
74,0
76,0
78,0
80,0
82,0
84,0
86,0
88,0
90,0
Oct-06 Abr-07 Nov-07 Jun-08 Dic-08 Jul-09
Month
C
u
s
o
l
R
e
c
o
v
e
r
y
(
%
)
Predicted
Actual
Mine to Heap in Mantoverde Anglo American Division
Proceedings of Copper 2010 2699
lent to US$ 11.4 million as a projected annual figure. The Cusol recovery was not affected, as it can
be seen in the previous figure.
Figure 5: Net Acid Consumption and Calcium Carbonate Grades for 2001-2007.
Figure 6: Net Acid Consumption and Calcium Carbonate Grades for 2008.
0
5
10
15
20
25
30
35
40
2000 2001 2002 2003 2004 2005 2006 2007 2008
Year
N
e
t
A
c
i
d
C
o
n
s
u
m
p
t
i
o
n
(
k
g
/
t
o
r
e
)
0
0,2
0,4
0,6
0,8
1
1,2
1,4
1,6
Actual
Equation
CaCO3
0
5
10
15
20
25
30
35
40
Nov-07 Feb-08 Jun-08 Sep-08 Dic-08
Month
N
e
t
A
c
i
d
C
o
n
s
u
m
p
t
i
o
n
(
k
g
/
t
o
r
e
)
0
0,2
0,4
0,6
0,8
1
1,2
1,4
1,6
Predicted
Actual
CaCO3
Daz, Salgado, Prez, Alvayai, Herrera, Zrate
Proceedings of Copper 2010 2700
The modification in the acid addition strategy resulted not only in a significant reduction in
acid consumption but also in PLS viscosity, as it is shown in Figure 7. Viscosity reduction was
around 20 %.
Figure 7: PLS viscosity.
Due to the success obtained with these modifications, the application to other leach processes such
as dump and vat leaching is being evaluated.
5 Conclusions and Recommendations
Mantoverde is a heap leach-SX-EW plant that was commissioned in December 1995 at a production
rate of 42,130 ton of copper per year and 5.4 million ton of ore per year. Forecast copper production
for 2010 is 62,000 ton for an ore treatment rate of 9.6 million ton.
During 2007, a number of problems were experienced in heap leaching and solvent extraction,
among them a decrease in copper recovery and an increase in acid consumption, impurities concen-
tration and PLS viscosity. A team was formed to investigate the reasons and to set up a working
program in order to optimize these performance indicators. For that purpose, the operating data of
74 heaps was reviewed and a column test program was carried out.
It was concluded that heap height and acid addition in the curing stage were the most important va-
riables affecting negatively the copper extraction. It was also found that heap height, CaCO
3
grade
0,0
1,0
2,0
3,0
4,0
5,0
6,0
7,0
Nov-07 Feb-08 Jun-08 Sep-08 Dic-08 Mar-09 Jul-09
Month
P
L
S
v
i
s
c
o
s
i
t
y
(
c
p
)
Mine to Heap in Mantoverde Anglo American Division
Proceedings of Copper 2010 2701
and acid addition strategy were the most important variables increasing acid consumption. Accor-
dingly, it was recommended to decrease gradually heap height from 7.5 m to 5 m and to optimize
acid addition distribution by reducing acid dosage in curing stage and increasing proportionally acid
concentration in leaching solutions.
The heap height was reduced from 7.5 to 6.5 m by the end of August 2007, resulting in a Cusol re-
covery increase of 3.2 points, while the recommendations on acid addition were implemented during
the first semester of 2008 at the heap leach operation with an average reduction in net acid consump-
tion of 5.8 kg/t, about 20 %, equivalent to US$ 11.4 million, as a projected annual figure. The modifi-
cation in the acid addition strategy resulted not only in a significant reduction in acid consumption
but also in PLS viscosity by 20 %.
Due to the success obtained with these modifications, the application to other leach processes such
as dump and vat leaching is being evaluated.
Acknowledgements
The authors wish to thank the General Management of Anglo American Chile for the permission to
publish this paper. The technical contributions to the work made by a number of process engineers
at Mantoverde Division are also acknowledged.
References
[1] G. Zrate. Anlisis Operacin Pilas Mantoverde. Julio 2007. Interim Report.
[2] C. Salgado, C. Prez, C. Alvayai, G. Zrate, Acid management in heap leaching. Are we doing
right? HydroCopper 2009, E. Domic, J. Casas Eds. Gecamin, Santiago, Chile. pp. 47-56.
Proceedings of Copper 2010 2702
Proceedings of Copper 2010 2703
Predicting the Effects of Locked, Partially Locked,
and Liberated Minerals in Copper Leaching
Michael L. Free, Abraham L. Jurovitzki
University of Utah
Department of Metallurgical Engineering
Salt Lake City, 84112, USA
Keywords: Liberation, heap leach modeling
Abstract
Valuable mineral grains are associated with gangue material in an ore. The association of the valua-
ble mineral grains with the gangue material can be characterized as liberated, partially-locked, or
fully-locked. Valuable mineral grain leaching kinetics depends on the association with the gangue.
However, traditional heap leaching modeling typically assumes all valuable mineral grains are fully-
locked within the gangue. This paper presents modeling results obtained by including the effects of
liberated, partially-locked, and fully-locked valuable mineral grains on leaching.
1 Introduction
Accurate modeling of heap leaching requires accurate quantification of the rate and magnitude of
resulting valuable mineral dissolution. However, within an ore particle, valuable mineral grains are
interspersed with gangue particles. The valuable mineral grain can occupy three different states at
various ore sizes. The valuable mineral grains can be fully locked in the interior of the ore, partially
locked within the ore, or completely liberated from the ore as depicted in Figure 1 [1].
Free, Jurovitzki
Proceedings of Copper 2010 2704
Figure 1: Illustration of the three valuable mineral particle states and their relationship with an ore
particle.
Valuable mineral particles that are locked inside of the host rock particles require pore diffusion of
reactant in order for leaching to occur. In contrast, liberated particles have excellent access to reac-
tant in surrounding solution. Partially-locked valuable mineral particles have partial access to reac-
tant in the environment. Each type of relationship between the host rock and valuable mineral par-
ticles requires a different leaching model. The ability to quantify these relationships provides the
foundation for this modelling approach, which is described in greater detail in other references
[1-2].
1.1 Estimating host rock and valuable mineral particle associations
The estimated probability of liberation can be calculated using the expression [2]:
)) exp( 1 )( exp(
*
* *
*
* *
.
hrp
hrp vmp
vmp
hrp vmp
est lib
r
r r
r
r r
P
= (1)
P
lib.est
is the estimated probability of particle liberation, r
*
vmp
is the reference radius of valuable min-
eral particles, and r
*
hrp
is the reference radius of host rock particles. The probability of valuable
mineral that is partially-locked can be determined using the mathematical expressions [2]:
) ... ]...( ) ( 1 ))][ exp( 1 )( exp( 1 [
* * 3
*
* *
*
* *
*
* *
. vmp hrp
hrp
vmp hro
hrp
hrp vmp
vmp
hrp vmp
est Locked Partially
r r for
r
r r
r
r r
r
r r
P >
(2)
) ... ))]...( exp( 1 )( exp( 1 [
* *
*
* *
*
* *
. vmp hrp
hrp
hrp vmp
vmp
hrp vmp
est Locked Partially
r r for
r
r r
r
r r
P <
(3)
Liberated valuable
mineral particle
Partially-locked valuable
mineral particle
Fully-locked valuable
mineral particle
Predicting the Effect of Locked, Partially Locked, and Liberated Minerals
Proceedings of Copper 2010 2705
Correspondingly, the probability of fully-locked valuable mineral particles is [2]:
) ... ]...( ) ))][( exp( 1 )( exp( 1 [
* * 3
*
* *
*
* *
*
* *
. vmp hrp
hrp
vmp hro
hrp
hrp vmp
vmp
hrp vmp
est Locked Fully
r r for
r
r r
r
r r
r
r r
P >
(4)
) ... ...( 0
* *
. vmp hrp est Locked Fully
r r for P < =
(5)
1.2 Liberated Particle Leaching
Liberated particle leaching can be described mathematically by the expression:
] ) 1 ( 1 [ ) (
y
lib vmp lib lib vmp
r k r t = (6)
k
lib
is a reaction constant (sec/cm), is the fraction reacted, and y is 1/3,1/2, or 2/3 for reaction con-
trol, rapid flow/fine particle, and slow flow/larger particle leaching conditions, respectively. The
subscript lib is added to indicate these terms apply to the liberated fraction of valuable mineral par-
ticles. This equation assumes leaching without product layer formation. If a porous product layer
forms during leaching, a pore diffusion/shrinking core model can be applied:
] ) 1 ( 1 [ ) (
3 / 2
, , 3
2
2
,
, PD lib PD lib
eff x
vmp lib PD
PD lib vmp
D C
r k
r t = (7)
C
x
is the concentration of reactant, D
eff
is the the effective diffusivity, and k
PD,lib
is a constant.
1.3 Partially-Locked Particle Leaching
A useful model for evaluating partially-locked valuable mineral particle leaching is [2]:
] ) 1 ( 1 [ ]
)
3
4
(
4
[ ) (
3 / 2
y
lib vmp lib lib vmp
r k r t
(8)
If a porous product film forms during leaching, the pore diffusion model (equation 7) can be simi-
larly applied using the liberated particle leaching case by multiplying by 4/(4/3)
2/3
[2].
1.4 Fully-Locked Particle Leaching
A common model for diffusion controlled leaching through a porous reaction product layer or
shrinking core is:
] ) 1 ( 1 [ ) (
3 / 2
, , 3
2
2
,
, PD lock PD lock
eff x
hrp lock PD
PD lib vmp
D C
r k
r t = (9)
Free, Jurovitzki
Proceedings of Copper 2010 2706
This equation is generally the same as given for the liberated particle leaching for product layer
formation. However, in this case the host rock particle diameter is used. Other constants are also
different due to the general change in application.
1.5 Ore Leaching
Leaching of an ore that consists of liberated, partially-locked, and fully-locked valuable mineral par-
ticles is the sum of the leaching of the valuable mineral grain distributed in these three categories.
Thus, overall ore leaching is determined by summing the contributions of leaching in each category
that are weighted by their relative proportions. The overall leaching was determined using an Excel
spreadsheet and the accompanying goal seek function. The goal seek function was used to find the
fraction reacted that corresponded with a specified time. The resulting fraction reacted values were
weighted based on the probability of occurrence and summed to determine the overall fraction at a
specified time. The resulting data was used to construct overall fraction reacted versus time plots.
2 Method
2.1 Ore sample and reagents
Copper ore was dry screened into appropriate size classes. Subsequently, samples were prepared for
hydrometallurgical processing by being split into desired quantities.
Leaching solutions were prepared with reagent grade sulfuric acid (H
2
SO
4
, MW ~98.08, assay
95.0 -98.0 %) which, was obtained from EMD Chemicals. In the case of chalcopyrite ore leaching
10 g/l of Ferric sulfate (Fe(SO
4
)
3
, nH
2
O MW ~399.88, purity 73.0 %) was added, which was ob-
tained from Mallinckrodt Chemicals and 15 g/l of Sodium chloride (NaCl, MW ~58.44, purity
99.0 %), which was obtained from Mallinckrodt Chemicals, were added.
Chemical analysis solutions consisted of Hydrochloric acid (HCl, MW ~36.46, assay 36.5-38 %)
which, was obtained from EMD Chemicals, and Nitric acid (HNO
3
, MW ~63.01, purity 68.0 to
70.0 %) which, was obtained from Mallinckrodt Chemicals.
All of these chemicals were used without any further purification. All of solutions were prepared
using ASTM Type I water.
2.2 Column leaching experiments
Leaching was performed in saturated flow columns packed with glass beads at the bottom and a
layer of copper ore on top. The two layers were separated first by a rigid permeable plastic separator
and on top of that was a flexible permeable plastic barrier. The leaching solution was contained
within a large Erlenmeyer flask. The leaching solution was pumped from the large Erlenmeyer flask
Predicting the Effect of Locked, Partially Locked, and Liberated Minerals
Proceedings of Copper 2010 2707
via Masterflex pump to a connection at the bottom of the enclosed column and exited the top back
to the beaker containing the original solution via another Masterflex tube. The solution with the dis-
solved metal ions was continuously recirculated at a flow rate of 1-2 ml/sec. A pH of 1.5 was main-
tained via an AccuTipH glass body rugged blub electrode probe which was connected to an Eutech
Instruments pH 2000 series pH/ORP controller, which in turn activated an acid control pump (Mas-
terflex pump) that was connected to an acid reservoir burette of 0.5-1 M Sulfuric acid solution de-
pending on the test.
Five milliliter samples were withdrawn from the Erlenmeyer flask solution reservoir at regular in-
tervals. One milliliter of each of these samples was diluted up to a volume of 100 milliliters with
two percent Nitric acid to safeguard against possible precipitation of the metal ions.
2.3 Sample testing and chemical analysis
The dilute samples obtained during column leaching were analyzed via Inductively Coupled Plasma
Mass Spectrometry for dissolved copper content.
Chemical analysis tests of the various ore sizes were conducted to obtain a baseline for total copper
and iron content within the various size gradients. A split sample of each ore was comminuted in a
small ball mill under dry cascading conditions until it was a fine powder. Subsequently, a one gram
sample of the fine particle was further crushed via mortar and pestle for 3 minutes. 0.5 grams of this
powder was weighed out and added to a beaker containing twenty milliliters of aqua regia for diges-
tion. The aqua regia consisted of fifteen milliliters of Hydrochloric acid and five milliliters of Nitric
acid. The temperature was raised to 125 degrees Celsius on a hot plate and digested for twenty mi-
nutes. Afterwards, the beaker of digested ore was removed from the hot plate and allowed to cool
for fifteen minutes. After, the cooling period the ore was vacuum filter clear of all remaining par-
ticles using P4 Fisher filter paper. The strongly acidic filtrate was first diluted with 200 milliliters of
ATSM Type I water. One milliliter of this diluted solution was withdrawn and diluted further with
one hundred milliliters of ATSM Type I water. These chemical analysis samples were analyzed for
total iron and copper via inductively coupled plasma mass spectrometry. This process was carried
out for leached and unleached ore to verify the amount of copper that was being leached in the col-
umn leachings. Analysis of digested ore samples before and after leaching were in agreement (with-
in 20 %) with ICP evaluations of the relevant leaching solutions.
3 Results and discussion
Test results from copper oxide ore column leaching are presented in Figure 2 with the associated
model fit shown with the measured data. The data in Figure 1 show the model provides a very good
estimate of the copper leaching performance.
Free, Jurovitzki
Proceedings of Copper 2010 2708
Figure 2: Comparison of measured and calculated fraction reacted and time for copper oxide ore
with an average size of 0.4 cm. The model parameters were r
*
vmp
= 0.008 cm,
k
lib
= 1.0710
4
sec/cm, k
PD
,
lock
/D
eff
= 4.610
9
cmsec/mol.
Results from chalcopyrite ore leaching are presented in Figure 3. The measured leaching results pre-
sented in Figure 3 are fit reasonably well by the model.
Figure 3: Comparison of measured and calculated fraction reacted and time for chalcopyrite ore
with an average size of 0.284 cm. The model parameters were r
*
vmp
= 0.0045 cm,
k
lib
= 5.7610
5
sec/cm, k
PD
,
lock
/D
eff
= 3.210
9
cmsec/mol.
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
0.18
0.2
0 500000 1000000
F
r
a
c
t
i
o
n
R
e
a
c
t
e
d
Time (sec)
Model Measured
0
0.02
0.04
0.06
0.08
0.1
0.12
0 500000 1000000 1500000
F
r
a
c
t
i
o
n
R
e
a
c
t
e
d
Time (sec)
Model
Measured
Predicting the Effect of Locked, Partially Locked, and Liberated Minerals
Proceedings of Copper 2010 2709
Leaching results from the mixed oxide/sulphide ore are presented in Figure 3. These results are sim-
ilar to those presented for the other types of ore. However, the model does not fit the mixed ore
leaching data as well.
Figure 4: Comparison of measured and calculated fraction reacted and time for copper
oxide/sulfide ore with an average size of 0.284 cm. The model parameters were
r
*
vmp
= 0.008 cm, k
lib
= 1.9210
11
sec/cm, k
PD
,
lock
/D
eff
= 6.410
9
cmsec/mol.
Although the model fits the copper leaching data from three very different ore samples, there are
several issues that need to be addressed to provide more confidence in this modelling approach. The
ore that was used has not been analyzed by a mineralogical assessment to evaluate the extent to
which other minerals may affect the modeling results, which have assumed a single mineral or a
group of similar minerals is responsible for acid consumption. Thus, further work is needed, and the
work presented in this paper represents a work in progress in early stages of evaluation.
4 Summary
The data presented in this paper show that a relatively new leaching modeling approach that ac-
counts for the relationship between host rock and valuable mineral particles can be used to predict
leaching performance. The new modeling approach will help to fill an important need to more accu-
rately model leaching of valuable entities from heterogeneous materials. Additional mineralogy as-
sessments before and after leaching are needed to further evaluate the performance of the model.
Thus, further work is needed, and the work presented in this paper represents a work in progress in
early stages of evaluation.
0
0.05
0.1
0.15
0.2
0.25
0 1000000 2000000 3000000
F
r
a
c
t
i
o
n
R
e
a
c
t
e
d
Time (sec)
Model Measured
Free, Jurovitzki
Proceedings of Copper 2010 2710
Acknowledgements
The laboratory work of Abraham Jurovitzki, Ravindra Bhide, and Taylor Bird who performed the
experiments needed to provide the data in this paper are gratefully acknowledged along with partial
funding by the U. S. Department of Energy Center for Advanced Separation Technology and the
University of Utah Research Opportunities Program for undergraduate students.
List of symbols
r
*
particle reference size at which 62.3 % of the material passes as undersize when s = 1,
s index of the Weibull or Rosin-Rammler distribution function
P probability or fraction of particles smaller than size
r radius of particle
k constant
fraction reacted,
y coefficient
C concentration of reactant
D
diffusivity
subscripts
vmp valuable mineral particle
hrp host rock particle
x reactant
Eff effective
PD pore diffusion
Lib liberated
Lock indicates locked mineral particle
References
[1] FREE, M. (2008). Canadian Metallurgical Quarterly, 47(3), 277-284.
[2] FREE, M. (2010). Predicting Leaching Solution Acid Consumption as a Function of pH in
Copper Ore Leaching, 7
th
International Copper 2010 Cobre 2010 Conference at Hamburg,
Germany, Proceedings of Copper 2010, Volume 7, pp. 2711-2719.
Proceedings of Copper 2010 2711
Predicting Leaching Solution Acid Consumption
as a Function of pH in Copper Ore Leaching
Michael L. Free
University of Utah
Department of Metallurgical Engineering
Salt Lake City, 84112, USA
Keywords: Heap leach, acid consumption, modeling
Abstract
pH control and acid consumption are critical issues in many copper leaching operations. A mod-
erately low pH is a prerequisite to effective copper leaching. The quantity of acid needed to achieve
desired pH levels is one of the most important parameters in determining leaching costs. Thus, the
capability to accurately predict leaching solution pH and acid consumption during copper ore leach-
ing is important to the evaluation of leaching performance and cost. This paper will present experi-
mental and modeling results for acid consumption as a function of pH for copper leaching from
chalcopyrite ore in sodium chloride media.
1 Introduction
Acid consumption during copper leaching is a function of acid consuming minerals in the ore and
their leaching properties as well as their association with the host rock. Common acid consuming
minerals that are associated with copper ores include copper minerals such as chrysocolla, mala-
chite, azurite, tenorite, atacamite, and brochantite as well as gangue minerals such as goethite, limo-
nite, biotite, chlorite, and various feldspar minerals. Selected reactions include [1]:
CuSiO
3
2H
2
O + 12H
+
= 6Cu
2+
+ 8H
4
SiO
4
(1)
CuO + 2H
+
= Cu
2+
+ H
2
O (2)
CuSO
4
3Cu(OH)
2
+ 6H
+
= 4Cu
2+
+ 6H
2
O (3)
KAlSi
3
O
8
+ 4H
+
= K
+
+ Al
3+
+ 3H
4
SiO
4
(4)
(H,K)
2
(Mg,Fe)
2
Al
2
(SiO
4
)
3
+ 10 H
+
= 2K
+
+ 2Al
3+
+ 3H
4
SiO
4
+ 2(Fe,Mg)
2+
(5)
Fe
2
O
3
+ 6H
+
= 2Fe
3+
+ 3H
2
O (6)
FeO(OH) + 3H
+
= Fe
3+
+ H
2
O (7)
Free
Proceedings of Copper 2010 2712
Although the stoichiometry of these reactions is fixed, the actual consumption of acid by each of
these minerals is related to mineral size and distribution as well as by the formation of passive lay-
ers that can restrict access to acid consuming minerals.
As the acid consuming minerals dissolve, by-products can form that can alter the acid balance. Not-
able examples include jarosite and alunite formation [1]:
3Fe
3+
+ K
+
+2SO
4
2-
+ 6H
2
O = KFe
3
(SO
4
)
2
(OH)
6
+ 6 H
+
(8)
3H
2
O + K
+
+ 2SO
4
2-
+ 3Al
3+
= KAl
3
(OH)
6
(SO
4
)
2
+ 6 H
+
(9)
Each acid consuming mineral will eventually come to equilibrium with a surrounding solution envi-
ronment, although the time needed to reach equilibrium may exceed a practical time period. Often, a
near equilibrium condition can exist.
Reaction kinetics often temporarily overshadow the influence of equilibrium thermodynamics. In
some cases the dissolution of acid consuming minerals is very slow due to low rate constants or the
formation of secondary reaction products that passivate the mineral. Acid consuming minerals are
often found as locked, partially-locked, and liberated mineral grains. Consequently, their dissolution
behavior in acid follows several kinetic models. The rate of dissolution is also influenced by the
accumulation of ions in the leaching solution.
1.1 Estimating Liberated, Partially-Locked, and Locked Valuable Mineral
Grains
The fraction of particles that are liberated can be estimated using the size distributions of the host
rock particles and valuable mineral grains. Although a variety of size distribution functions can be
used for this purpose, the Rosin-Rammler distribution will be used here to demonstrate the process.
The Rosin-Rammler cumulative passing size distribution function is expressed as [2]:
s
r
r
r P
=
*
exp 1 ) ( (10)
in which s is an experimentally determined constant, r
*
is the particle reference size at which 62.3 %
of the material passes as undersize when s = 1, and P is the probability or fraction of particles
smaller than size r. As r becomes very large, the probability of undersize approaches one. The frac-
tion of liberated particles can be estimated using the size distribution functions for the valuable
mineral grains or particles and the host rock particles. Valuable mineral particles can be categorized
as liberated when they are larger than host rock particles and cannot, therefore, be incorporated in-
side of host rock particles. When size distributions for valuable mineral particles and host rock par-
ticles are considered, the liberated fraction of valuable mineral grains is determined by comparing
small size ranges of these distributions. If the small size range of valuable mineral particles under
examination is larger than the size range of host rock particles selected, that combination is consi-
Predicting Leaching Solution Acid Consumption as a Function of pH
Proceedings of Copper 2010 2713
dered to have liberated valuable mineral particles. Accordingly, the probability of that combination
of size ranges is used to appropriately weight that combination. This process is completed by com-
paring each small size range of valuable mineral particles to each host rock particle size range to
determine the overall fraction of valuable mineral particles that are considered to be liberated. This
process is described in more detail elsewhere [3].
The fraction of liberated particles can be estimated in a simple way using the geometric mean size
between the associated distributions. The estimated probability of liberation using the Rosin-
Rammler function assuming s is unity is [3]:
)) exp( 1 )( exp(
*
* *
*
* *
.
hrp
hrp vmp
vmp
hrp vmp
est lib
r
r r
r
r r
P
= (11)
P
lib.est
is the estimated probability of particle liberation, r
*
vmp
is the reference radius of valuable min-
eral particles, and r
*
hrp
is the reference radius of host rock particles. Particles that are not liberated
are either locked or partially locked and their corresponding joint probability is the remainder of the
probability sum of unity (1- P
lib.est
). The fraction of valuable mineral particles that are partially
locked can be determined using a similar analysis in which different small size ranges are compared
over the entire distributions and a specific formula to determine locking is used as described else-
where. [3] A simplified use of this approach which uses the estimated probability of liberation (Eq-
uation (11)) with the particle locking estimate can be used to estimate the probability of valuable
mineral that is partially-locked [3]:
) ... ]...( ) ( 1 ))][ exp( 1 )( exp( 1 [
* * 3
*
* *
*
* *
*
* *
. vmp hrp
hrp
vmp hro
hrp
hrp vmp
vmp
hrp vmp
est Locked Partially
r r for
r
r r
r
r r
r
r r
P >
(12)
) ... ))]...( exp( 1 )( exp( 1 [
* *
*
* *
*
* *
. vmp hrp
hrp
hrp vmp
vmp
hrp vmp
est Locked Partially
r r for
r
r r
r
r r
P <
(13)
Correspondingly, the fraction of valuable mineral particles that are fully locked in the host rock par-
ticles can be determined as [3]:
) ... ]...( ) ))][( exp( 1 )( exp( 1 [
* * 3
*
* *
*
* *
*
* *
. vmp hrp
hrp
vmp hro
hrp
hrp vmp
vmp
hrp vmp
est Locked Fully
r r for
r
r r
r
r r
r
r r
P >
(14)
) ... ...( 0
* *
. vmp hrp est Locked Fully
r r for P < =
(15)
These new simplified formulas were evaluated by comparing the associated results for valuable
mineral exposure measured by Miller et al. [4], and the results are presented in Figure 1. The results
in Figure 1 indicate that the predicted exposure (liberated and partially-locked valuable mineral par-
ticles) is very similar to the measured exposure obtain using X-ray tomography at most sizes eva-
luated, although the fit is best in the middle and small host rock particle sizes.
Free
Proceedings of Copper 2010 2714
Figure 1: Comparison of measured [4] and calculated valuable mineral exposure using the sum of
Equations (11) and (12) based on data from reference 4.
1.2 Liberated Particle Leaching
Liberated particle leaching can involve a variety of scenarios that each require a different kinetic
leaching model. Several leaching models for liberated particles can be described mathematically by
the expression:
] ) 1 ( 1 [ ) (
y
lib vmp lib lib vmp
r k r t = (16)
k
lib
is a reaction constant (sec/cm), is the fraction reacted, and y is 1/3,1/2, or 2/3 for reaction con-
trol, rapid flow/fine particle, and slow flow/larger particle leaching conditions, respectively. The
subscript lib is added to indicate these terms apply to the liberated fraction of valuable mineral par-
ticles. This equation assumes the particle dissolves without forming a product layer. If a porous
product layer forms while leaching takes place, the traditional pore diffusion/shrinking core kinetic
model can be applied:
] ) 1 ( 1 [ ) (
3 / 2
, , 3
2
2
,
, PD lib PD lib
eff x
vmp lib PD
PD lib vmp
D C
r k
r t = (17)
C
x
is the concentration of reactant, D
eff
is the the effective diffusivity, and k
PD,lib
is a constant.
1.3 Partially-Locked Particle Leaching
Leaching of partially-locked mineral particles can be described in different ways, some of which are
not simple, and none of which are rigorous for ore leaching. Valuable mineral particles exposed to
0
10
20
30
40
50
60
70
80
90
100
0 10 20 30
V
a
l
u
a
b
l
e
M
i
n
e
r
a
l
E
x
p
o
s
u
r
e
(
%
)
Size (mm)
Simplified Model
Miller et al. (2003)
Predicting Leaching Solution Acid Consumption as a Function of pH
Proceedings of Copper 2010 2715
the surface will leach similarly to liberated particles except that only a portion of the particle is di-
rectly exposed to the solution. A simplified approach to evaluate partially-locked particle leaching
considers the partially locked particles as cubes with one face exposed to solution. The ratio of ex-
posed leaching area of a fully exposed sphere to one side of a cube of equal mass is 4/(4/3)
2/3
.
Use of this ratio allows for a simplified conversion of liberated particle leaching to partially-locked
particle leaching provided that the particle dissolves and does not form a leaching product layer.
Thus, the associated general leaching model is:
] ) 1 ( 1 [ ]
)
3
4
(
4
[ ) (
3 / 2
y
lib vmp lib lib vmp
r k r t
(18)
In scenarios involving the formation of a product film, the pore diffusion model can be similarly
applied from the liberated particle leaching case with the same conversion factor.
1.4 Fully-Locked Particle Leaching
A common model for diffusion controlled leaching through a porous reaction product layer or
shrinking core is:
] ) 1 ( 1 [ ) (
3 / 2
, , 3
2
2
,
, PD lock PD lock
eff x
hrp lock PD
PD lib vmp
D C
r k
r t = (19)
This equation is the same as given for the liberated particle leaching through a porous product layer
except that the radius of host rock particle is used instead of the valuable mineral particle radius,
and the other constants are specific to the fully-locked leaching scenario.
1.5 Ore Leaching
Leaching of an ore that consists of liberated, partially-locked, and fully-locked valuable mineral
particles is the sum of the leaching of the valuable mineral grain distributed in these three catego-
ries. Thus, overall ore leaching is determined by summing the contributions of leaching in each cat-
egory that are weighted by their relative proportions. The overall leaching was determined using an
Excel
spreadsheet and the accompanying goal seek function. The goal seek function was used to
find the fraction reacted that corresponded with a specified time. The resulting fraction reacted val-
ues were weighted based on the probability of occurrence and summed to determine the overall frac-
tion at a specified time. The resulting data was used to construct overall fraction reacted versus time
plots.
Free
Proceedings of Copper 2010 2716
2 Experiments
Leaching experiments were performed using columns approximately 10 inches in height and 2 inch-
es in inner diameter. Monosize fractions of chalcopyrite ore (150-500 grams) were used together
with a solution volume approximately twice the weight of the ore sample. Solution was maintained
at specified pH levels using a pH controller and an acid pump using a reservoir of 1 M sulfuric acid.
Reagent grade chemicals and ASTM Type I water were used in the experiments. All column leach-
ing experiments were performed by flowing solution up from the bottom of the column in saturated
flow at a rate of approximately 100 ml/min. All experiments were performed at room temperature.
Acid consumption was measured based on acid reservoir depletion as a function of time. Note that
because monosize fractions of particles were used and the valuable mineral particle size is less than
the host rock particle size, liberated particles were presumed to have passed through the screens
during sample preparation and were therefore not included in this analysis.
3 Results and discussion
3.1 Effect of pH
Leaching of an ore that consists of liberated, partially-locked, and fully-locked valuable mineral par-
ticles is the sum of the leaching of the valuable mineral grain distributed in these three categories.
Thus, overall ore leaching is determined by summing the contributions of leaching in each category.
Figure 2 shows the fit of the model to measured data at pH 1.5 and pH 2.5. The fit of the model data to
the measured data is reasonable but not excellent. Because the ore contains a variety of minerals and
the model fit assumes one predominant mineral, the fit of the data was not expected to be excellent.
Figure 2: Comparison of measured and calculated fraction reacted and time for ore samples
leached at pH 1.5 and 2.5 as indicated. The ore particles were -14 + 20 Mesh.
The model parameters were r
*
vmp
= 0.004 cm, k
lib
= 3.8410
6
sec/cm,
k
PD
,
lock
/D
eff
= 3.210
9
cmsec/mol.
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0.45
0.5
0 500000 1000000
F
r
a
c
t
i
o
n
R
e
a
c
t
e
d
Time (sec)
Model pH 1.5
Measured pH 1.5
Model pH 2.5
Measured pH 2.5
Predicting Leaching Solution Acid Consumption as a Function of pH
Proceedings of Copper 2010 2717
3.2 Effect of size
The ability of the model to predict the effect of size is presented in Figure 3. Figure 3 shows the
model generally fits the data. The fit of the data to the smaller size fraction is significantly better
than the fit for the larger size fraction.
Figure 3: Comparison of measured and calculated fraction reacted and time for ore samples of the
average size indicated in microns (m). The ore particles were 14 + 20 Mesh.
The model parameters were r
*
vmp
= 0.004 cm, k
lib
= 1.3510
6
sec/cm,
k
PD
,
lock
/D
eff
= 110
9
cmsec/mol.
Although the model fits the data generally, there are several issues that need to be addressed to pro-
vide more confidence in this modeling approach. The ore that was used has not been analyzed by a
mineralogical assessment to evaluate the extent to which other minerals may affect the results which
have assumed a single mineral or a group of similar minerals is responsible for acid consumption.
Thus, further work is needed, and the work presented in this paper is a work in progress in its early
stages of evaluation.
4 Summary
The model discussed in this paper that can be used to classify the association of acid consuming par-
ticles into liberated, partially-locked, and fully-locked particles relative to host-rock particles provides
a valuable framework for evaluating leaching. The model provides new equations that can be used to
easily estimate the fraction of liberated, partially-locked, and fully-locked valuable mineral particles.
The information from the model can be used to weight the relative influence of leaching models, one
of which is new, to determine overall leaching performance. Results indicate this approach shows po-
tential for improving acid consumption predictions for leaching operations. Additional mineralogy
assessments before and after leaching are needed to further evaluate the performance of the model.
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 500000 1000000
F
r
a
c
t
i
o
n
R
e
a
c
t
e
d
Time (sec)
Model 820 microns
Measured 820 microns
Model 1420 microns
Measured 1420 microns
Free
Proceedings of Copper 2010 2718
Acknowledgements
The laboratory work of Abraham Jurovitzki, Alysha Bhide, and Prashant Saraswat who performed
the experiments needed to provide the data in this paper are gratefully acknowledged along with
partial funding by the U. S. Department of Energy Center for Advanced Separation Technology and
the University of Utah Research Opportunities Program for undergraduate students.
List of symbols
r
*
particle reference size at which 62.3 % of the material passes as undersize when s = 1,
s index of the Weibull or Rosin-Rammler distribution function
P probability or fraction of particles smaller than size
r radius of particle
k constant
fraction reacted
y coefficient
C concentration of reactant
D
diffusivity
subscripts
vmp valuable mineral particle
hrp host rock particle
x reactant
Eff effective
PD pore diffusion
Lib liberated
Lock indicates locked mineral particle
Predicting Leaching Solution Acid Consumption as a Function of pH
Proceedings of Copper 2010 2719
References
[1] JANSEN, M. & TAYLOR, A. (2003). Overview of gangue mineralogy issues in oxide copper
Heap leaching,
http://www.altamet.com.au/Technical%20Papers%20and%20Articles/ALTA%20Copper/
Overview%20of%20Gangue%20Mineralogy.pdf, Accessed October 16, 2009.
[2] KELLY, E. & SPOTTISWOOD, D. (1982). Introduction to Mineral Processing, John Wiley,
New York, 26.
[3] FREE, M. (2008). Canadian Metallurgical Quarterly, 47(3), 277-284.
[4] MILLER, J., LIN, C., GARCIA, C., & ARIAS, H. (2003). International Journal of Mineral
Processing, 2003, 72, pp. 331340.
Proceedings of Copper 2010 2720
Proceedings of Copper 2010 2721
Copper from Pyrite A Short History
Fathi Habashi
Laval University
Department of Mining, Metallurgical, and Materials Engineering
Quebec City, G1V 0A6, Canada
Keywords: Pyrite, roasting, leaching, extraction, history
Abstract
Pyrite, known since antiquity, had at one time a great strategic importance because it was the main
raw material for making elemental sulfur for gunpowder manufacture and for making SO
2
for sul-
furic acid production. Pyrite-bearing ores usually contains appreciable amounts of copper that was
recovered by a variety of methods, now obsolete. A historical review will be given for the RioTinto,
Orkla, and Duisburger Kupferhtte processes. However, when other sources of cheap elemental
sulfur became available, pyrite not only lost its importance but also became a nuisance for the me-
tallurgical industry because of problems associated with its disposal.
1 Introduction
Pyrite (Figure 1) was mentioned by the Greek and Roman writers. It is the most common sulfide
mineral and is widely associated with other metal sulfide deposits. It had at one time a great
strategic importance because it contains approximately 53 % sulfur and was the main raw material
for making elemental sulfur for gun powder and for making SO
2
for sulfuric acid manufacture. In
the sixteenth century, pyrite was heap leached in the Harz mountains in Germany and in Ro Tinto
mines in Spain to recover its copper content and in the nineteenth century pyrometallurgical
processes were developed.
Habashi
Proceedings of Copper 2010 2722
Figure 1: Naturally-occurring pyrite cubes Figure 2: French translation of Johann
Friedrich Henckels book on py-
rite published in Paris in 1760
The importance of the mineral can be judged from a book published in 1725 in Leipzig by Johann
Friedrich Henckel (1679-1744) on the mineralogy of sulfide minerals entitled Pyritologia, oder
Kiess-Historie. The book was translated in English in 1757 and in French in 1760 (Figure 2). Full
title: Pyritologia, oder Kiess-Historie (Leipzig 1725) English translation: Pyritologia or a History
of the Pyrites, the principal body in the Mineral Kingdom, in which are considered its names, spe-
cies beds, and origin; its iron, copper, unmetallic earth, sulphur, arsenic, silver, gold, original par-
ticles, vitriol, and use in smelting. In 1907 the French chemist P. Turchot published another com-
prehensive book on pyrite.
2 Copper from Pyrite by Heap Leaching
In these operations, the pyrite-bearing ores were piled in the open air and left for months to the ac-
tion of rain and air whereby oxidation and dissolution of copper took place. A solution containing
copper sulfate was drained from the heap and collected in a basin. Metallic copper was then precipi-
tated from this solution by scrap iron, a process that became known as cementation process. Mi-
croorganisms were certainly active in catalyzing the leaching process but this became known only in
the 1960s. When pyrite was fully leached it was loaded on trucks and shipped to sulfuric acid manu-
facturers.
3 Copper from Pyrite Cinder
Pyrite was a major source for generating SO
2
needed for sulfuric acid manufacture:
2 FeS
2
+
11
/
2
O
2
4 SO
2
+ Fe
2
O
3
(1)
A variety of roasters were specially developed for this purpose. The first design in 1850 by the Brit-
ish inventor Alexander Parkes (1813-1890) (Figures 3 and 6). A later design by the American
Copper from Pyrite A Short History
Proceedings of Copper 2010 2723
engineer John Brown Francis Herreshoff (1851-1932) (Figure 4) is shown in Figure 7. Multi-stage
roasters were later replaced by the more efficient fluidized bed reactors invented in Germany in
1922 by Fritz Winkler (1888-1950) (Figures 5 and 8).
Figure 3: Alexander Parkes
(1813-1890)
Figure 4: John Herreshoff
(1851-1932)
Figure 5: Fritz Winkler
(1888-1950)
Table 1 shows typical analysis of cinder. A method was patented in England in 1844 by W. Long-
maid to purify the cinder for shipping to the steel industry and at the same time to recover the non-
ferrous metals present. It was first applied by William Henderson of Scotland in 1859. Lower cop-
per content in the pyrite especially since World War I compelled the firm to extract other products
from pyrite. In the process developed, known as Longmaid-Henderson process, the pyrite cinder
was roasted with sodium chloride then leached with water to recover the nonferrous metal chlorides.
The technology was adapted in Germany for over a century at the Duisburger Kupferhtte in Duis-
burg from 1876 to 1982 (Figure 9).
Figure 6: The first pyrite
roaster, 1850
Figure 7: Herreshoff
multi-hearth
furnace
Figure 8: Fluidized bed reactors for
roasting pyrite concen-
trates
Habashi
Proceedings of Copper 2010 2724
Table 1: Typical analysis of pyrite cinder
% ppm
Iron 54-58 Cobalt 300-1500
Gangue 6-10 Nickel 10-1500
Sulfur 2.5-4 Manganese 300-3000
Copper 0.8-1.5 Silver 25-50
Zinc 2.0-3.2 Gold 0.5-1.5
Lead 0.3-0.5 Cadmium 40-100
Thallium 15-45
Indium 1-50
Figure 9: Duisburger Kupferhtte plant in Duisburg, Germany operated from 1876 to 1982 to
process pyrite cinder
In this plant, pyrite was imported from all over the world by the company, sold to acid manufactur-
ers in Germany on the agreement that the iron oxide resulting from roasting, called cinder, is
shipped back to Duisburg for further treatment to recover nonferrous metals, precious metals, and
metallic iron (Figure 10).
Copper from Pyrite A Short History
Proceedings of Copper 2010 2725
Figure 10: Importing pyrite from worldwide suppliers to Duisburger Kupferhtte plant
The pyrite cinder was mixed with NaCl and heated continuously in a multiple hearth furnace at
800 C to transform nonferrous metals into water-soluble chlorides (Figure 11). Each batch requires
about 2 days for leaching in vats (Figure 12). Copper was precipitated from solution in two steps
(Figure 13):
Habashi
Proceedings of Copper 2010 2726
Figure 11: Salt roasting of pyrite cinder Figure 12: Leaching plant for purple ore
Cu + CuCl
2
Cu
2
Cl
2(ppt)
(2)
Cu
2+
+ Fe Cu + Fe
2+
(3)
Copper for the first step was obtained from the second step. It is interesting to note that the blue
CuCl
2
solution turns pink after reduction and precipitation of Cu
2
Cl
2
due to the presence of Co
2+
in
the remaining solution. Also the residual Cu
+
is colorless. Cuprous chloride recovered is then
treated with calcium hydroxide to precipitate copper (I) oxide which was reduced with coal in a fur-
nace to black copper:
Cu
2
Cl
2
+ Ca(OH)
2
Cu
2
O + CaCl
2
+ H
2
O (4)
2Cu
2
O + C 4Cu + CO
2
(5)
The black copper was cast into anodes and refined electrolytically; the precious metals were col-
lected in the anodic slimes. The solution obtained after cementation is evaporated under vacuum to
recover Na
2
SO
4
10H
2
O (Figure 14). In 1975, the production of black copper was abandoned in fa-
vor of a hydro-electowinning process: cuprous oxide was leached in recycle acid and the CuSO
4
solution obtained was electrolyzed to produce copper cathodes.
Copper from Pyrite A Short History
Proceedings of Copper 2010 2727
Figure 13: Cementation plant
Habashi
Proceedings of Copper 2010 2728
Figure 14: Sodium sulfate plant
Copper from Pyrite A Short History
Proceedings of Copper 2010 2729
The residue, called purple ore, now a high-grade iron ore (6163 % Fe), is sintered (Figure 15)
and delivered to the blast furnace (Figure 16). Since lead and silver form chlorides during roasting
which are insoluble in the leaching step, they remained in the purple ore. When the sintered purple
ore was charged in the blast furnace, lead-silver alloy is formed. Being insoluble in iron and has a
higher density it settles at the bottom of the hearth. The furnace was provided with an opening be-
low the iron notch to tap the lead-silver alloy once a week. Figure 17 gives a general flowsheet for
the recovery processes. Some plant production data are given in Table 2.
Table 2: Data on Duisburger Kupferhtte plant
Imported 3 million tonnes of pyrite annually 60,000 tonnes Zn
Processed 2 million tonnes of cinder - 50 tonnes Ag
Employed 4100 people - 70 tonnes Cd
Produced annually: - 10 tonnes Tl
- 1.2 million tonnes pig iron - Minor amounts of Co, Au, In, Pt
- 6000 tonnes Pb - 170,000 tonnes sodium sulfate
- 24,000 tonnes Cu Consumed 200,000 tonnes NaCl
Figure 15: Sintering plant
Habashi
Proceedings of Copper 2010 2730
Figure 16: Blast furnace plant producing pig iron and lead-silver alloy
Copper from Pyrite A Short History
Proceedings of Copper 2010 2731
Figure 17: General flowsheet of operations at Duisburger Kupferhtte
4 Pyrite and the steel industry
Noting the success of Duisburger Kupferhtte, producers of sulfuric acid in Europe, Japan, and
USA became interested to market their finely ground iron oxide produced by roasting pyrite to the
steel industry. Although iron oxide pelletization was invented in Sweden in 1912, it was not intro-
duced in the iron ore industry until thirty years later. This took place when the electric arc furnace
was introduced in the steel industry and the need arose for a palletized feed. This gave an incentive
to the sulfuric acid manufacturers to upgrade their pyrite cinder for the steel industry by removing
copper and other nonferrous metals. As a result, more processes were developed to deal with this
problem.
Kowa-Seiko Process. This is a Japanese process developed at Kitakyushu in which the cinder is
mixed with calcium chloride, pelletized, then heated in a rotary kiln at 1100 C to volatilize non-
ferrous metal chlorides. These are scrubbed in water from the exit gases and the solution treated for
metal recovery. The process was also used in Portugal and in other countries.
Bethlehem Steel process. At the plant at Sparrows Point in Maryland, cobalt from the cinder was
recovered by a process based on a careful temperature control during the roasting of pyrite. If the
temperature is kept at 550 C, cobalt in the pyrite will be converted to sulfate and therefore can be
leached directly from the cinder with water. The hot pyrite cinder was quenched with water to give a
Habashi
Proceedings of Copper 2010 2732
slurry containing 68 % solids. When the solids are filtered off, the solution contains 2025 g/L Co;
it is processed further for metal recovery. In the cinder, the Fe:Co ratio is 50:1; in solution it is 1:1.
This plant supplied the only domestic source of cobalt in USA.
Outokumpu process. In Finland at the Outokumpu plant in Pori a similar process is in operation in
1979; the sulfated pyrite cinder contains 0.8-0.9 % Co and other nonferrous metals. It is leached
with water to get a solution at pH 1.5 analyzing 20 g/L Co, 6-8 Ni, 7-8 Cu, 10-12 Zn, and trace
amounts of iron, which is treated for metal recovery.
5 Copper from Pyrite by Smelting
Pyrite smelting was developed to melt massive sulfide ore to form matte, and at the same time to
recover the excess sulfur in the elemental form. It was first successfully operated in 1928 by Orkla
Grube in Norway (Figure 13). Similar operations were in Sweden, Portugal, Spain, and Russia. Py-
rite containing about 2 % Cu is mixed with coke, quartz, and limestone and heated in a blast furnace
(Figure 14). In the upper part of the furnace, one atom of sulfur in pyrite is distilled as elemental
sulfur. In the oxidizing zone, FeS formed is oxidized to ferrous oxide and SO
2
. In the middle part of
the furnace, the reduction zone, SO
2
is reduced by coke to elemental sulfur which is volatilized as
vapor. The reactions taking place can be represented by the following equations:
Upper zone: FeS
2
FeS +
1
/
2
S
2
(6)
Oxidation zone: FeS
+
3
/
2
O
2
FeO + SO
2
(7)
Middle zone: SO
2
+ C CO
2
+
1
/
2
S
2
(8)
Carbon disulfide and carbon oxysulfide are formed in the furnace; they are converted to elemental
sulfur on catalytic beds. The matte produced contains 6-8 % Cu and is usually re-smelted with coke,
silica, and limestone to 40 % Cu.
Copper from Pyrite A Short History
Proceedings of Copper 2010 2733
Figure 18: Location map of Orkla plant Figure 19: Orkla process for treating pyrite
concentrates to recover copper and
elemental sulfur
6 The Decline in the Pyrite Industry
During the Napoleonic wars, Spanish pyrite entered into competition with Sicilian sulfur in many
markets. The production of pyrite, however, declined gradually towards the end of the nineteenth
century after the discovery of an economic method for the recovery of sulfur from the sulfur domes
in the Gulf of Mexico. There was also a great rise in petroleum refining activities after World War II
that resulted in large amounts of sulfur-containing refinery gases and the need to recover this sulfur
to avoid polluting the environment with SO
2
. Also the availability of large volumes of natural gas
containing hydrogen sulfide at Lacq in southern France in 1950s and in Alberta, Canada in 1970s
contributed to the decline in pyrite demand.
Elemental sulfur replaced pyrite as a source for SO
2
for sulfuric acid manufacture because of the
purity of the gas generated and the elimination of dust recovery equipment in the plant. Sulfur be-
came available by the following processes:
Sulfur deposits discovered in the Gulf of Mexico were exploited economically since 1895 by
Herman Frasch (1851-1914) (Figure 20) using superheated water to melt the sulfur and float it to
the surface by compressed air (Figures 21 and 22). The process applies only when sulfur is stratified
between impervious rock formation which was not the case for Sicily. The success of Frasch
process ruined the Sicilian sulfur industry.
Habashi
Proceedings of Copper 2010 2734
Figure 20: Herman Frasch
(1851-1914)
Figure 21: Extracting sulfur
by Frasch process
Figure 22: Storage of
Frasch sulfur
H
2
S-containing natural gas or petroleum refining gases not only became a source of elemental sul-
fur but recovering this sulfur solved an environmental problem since the gases were burned generat-
ing large amounts of SO
2
. Hydrogen sulfide must first be separated from the gases by an absorption
desorption process (then oxidized by a controlled amount of oxygen at 400
o
C on aluminum oxide
or bauxite bed using Claus reaction discovered in 1883 by the German chemist Carl Friedrich Claus:
H
2
S + O
2
S + H
2
O (9)
As soon as these new sources of elemental sulfur became available, pyrite roasters for sulfuric acid
manufacture were dismantled and replaced by sulfur burners and the decline in pyrite production
started to decline rapidly. Acid plants based on pyrite roasting were expensive because it included
bulky equipment for dust separation.
Suggested Readings
[1] Anonymous, Upgrading Pyrite Cinders for Iron and Steel Production. Montedison Chlorina-
tion Process Removes and Recovers Nonferrous Metals and Arsenic, Sulphur 106, 52-57
(1973) May/June [The British Sulphur Corporation]
[2] W. Greiling, 75 Jahre Duisburger Kupferhtte, 1876-1951, Chemiker-Zeitung p. 577 (1951)
[3] E. Guccione, The Recovery of Elemental Sulfur from Pyrite in Finland, Chemical Engineer-
ing (New York), February 1966
[4] F. Habashi, Researches on Copper. History & Metallurgy, Mtallurgie Extractive Qubec,
Qubec City 2009. Distributed by Laval University Bookstore Zone
Copper from Pyrite A Short History
Proceedings of Copper 2010 2735
[5] F. Habashi, The Recovery of Elemental Sulfur from Sulfide Ores. Proceedings of XXXVI
Congrs international de chimie industrielle, Brussel, Belgium, 1966 (pub. 1967). Ind.
Chim. Belge 32 (Special Issue) (Part 2), 250-258 (1967). Also available as Bulletin 59 (1967)
from Montana Bureau of Mines & Geology, Butte, Montana
[6] F. Habashi, The Recovery Empire Built on Fools Gold, Eng. & Ming. J. 170 (12), 5964
(1969)
[7] W. Haynes, Brimstone. The Stone That Burns, Van Nostrand, Princeton, New Jersey 1959
[8] H.Kudelka, R.M. Dobbener, and N.L. Piret, Copper Electrowinning at Duisburger Kup-
ferhtte, CIM Bulletin 186-197 (1977) August
[9] I. Maeshiro, Recovery of Valuable Metals from Black Ore, pp. 315 324 in Proceedings of
the Eleventh Commonwealth Mining and Metallurgical Congress, Hong Kong, published by
the Institution of Mining and Metallurgy, London 1978
[10] H.J. Nowacki, Die Aufbereitung kupferhaltiger Schlmme und Fllprodukte im Schachto-
fen, Z. Erzbergbau u. Metallhttenw. 22(1), 22 28 (1969)
[11] C.Nuez et al., A Non Integral Process for the Recovery pf Gold, Silver, Copper and Zinc
Values Contained in Spanish Pyrite Cinders, pp. 328 337 in XV Congrs International de
Minralurgie, volume 3 (1985)
[12] W. Teworte, Duisburger Kupferhtte, company brochure published in Duisburg, W. Germany
1957
[13] E. Wiklund, Die Nutzung von Pyrit als Grundlage fr die Produktions von Schwefelsure
und Eisenoxyd, Aufbreitungs- Technik (6),285-289 (1977)
Proceedings of Copper 2010 2736
Proceedings of Copper 2010 2737
Copper Crud Treatment, Concentration
Dependent Pond Depth Adjustment for Decanter
Centrifuges, DControl
Dipl.-Ing.Tore Hartmann, Dr. Ulrich Horbach, Jens Kramer
GEA Westfalia Separator Process GmbH
Werner-Habig-Strae 1
D-59302 Oelde, Germany
Keywords: Hyrdometallurgy, copper, centrifuge, crud treatment
Abstract
In the hydrometal process, the presence of the so-called crud is a constant challenge in solvent
extraction. Crud is a stable emulsion which forms along the interface between the aqueous and or-
ganic phase. The spread is influenced by the following parameters: first, wind blows the dust and
impurities into the open sedimentation tanks. Second, the undissolved solids such as sand trans-
ported in the PLS cause problems in conjunction with incorrect agitator design.
The crud fraction can decisively impact the efficiency of the hydrometal extraction because the
phase interface can constitute a large fraction and the sedimentation tanks cannot react flexibly to it.
In the downstream process of the series-connected sedimentation tanks, they are thus all contami-
nated with crud. At the same time, the necessary mass transport is significantly impeded at the phase
boundary between organic phase and aqueous phase due to the crud formation.
The transfer of the organic phase into the electrolysis cell can result in a burnout of the cathode.
The carry-over of this electrolyte into the extraction can cause problems with the pH regulation. The
carry-over of the organic components into the raffinate also leads to contamination of the leaching
circuit.
The continuous treatment of the crud with a 3-phase decanter centrifuge is extremely effective in
combating this. This technology splits all three phases from each other and they are consequently
continuously separated. All subsequent process steps exhibit a stable, uniform effectiveness. The
main advantage for the customer is that fluctuations in the process are eliminated and the organic
phase can be recirculated back to extraction. The recycling of the solvent alone justifies the invest-
ment in less than 6 months as examples from South Africa and Chile show.
Hartmann, Horbach, Kramer
Proceedings of Copper 2010 2738
To operate solvent recovery in daily operation fully automated online at the optimum limit, the de-
canter centrifuge is equipped with a concentration-dependent pond adjustment called DControl
. By
this means, we can guarantee our customers the maximum possible solvent recovery and minimise
the solvent costs in this process step. This system is presented in detail in this paper.
1 Introduction
The number of applications for continuous centrifugation in the hydrometal process is rising con-
tinuously. This development is characterized by a number of clear customer demands in which de-
canters are clearly superior to other competing technologies.
This customer demand relating to productivity, process specifications, reliability and profitability
can be achieved and even surpassed. The demands of different production facilities involved in the
hydrometal process always vary because the respective composition of the ore requires a unique
process sequence. Consequently, application-specific demands are realized in close cooperation
with the customer. Preventive service concepts are a basic pre-condition for constant plant availabil-
ity which ensures the customer the leading position in hydrometallurgy.
Figure 1: Process schematic of crud formation
Copper Crud Treatment with 3 Phase Centrifuge
Proceedings of Copper 2010 2739
One example of this is the recovery of copper in an SX installation. Discharged insoluble solids lead
to crud formation.
Figure 2: Phase distribution in copper recovery
This can lead to a situation where aqueous components are transported into the recovery process
during recycling or the valuable organic phase. This can result in increased chloride contents when
the ores contain Cl
or if seawater is used for the process. This, in turn, can lead to pitting on the
higher storage tank or on the stainless steel agitator.
For this reason, the crud phase cannot be recycled back into the process. Disposal is, however, like-
wise uneconomic because on the one hand the organic phase is a valuable operating fluid and, on
the other hand, dissolved hydrometals are to be recovered in the aqueous phase. The crud phase
must therefore be split up into the three phases solids, aqueous and organic phase.
Previously, different chemical processes were used for this with which the emulsion was split.
However, it has been demonstrated that continuous centrifugation is the most effective and eco-
nomical technique.
The crud from a SX installation is discharged into the crud tank. From there it is conveyed into the
centrifuge which separates the aqueous components and the solids from the organic phase. The liq-
uid and solids are recycled into the raffinate basin, the organic material is recycled back into the SX
Hartmann, Horbach, Kramer
Proceedings of Copper 2010 2740
installation. The main purpose of this operating mode is to relieve the installation from the emulsion
and to recover the expensive organic solution.
To a certain extent, the formation of crud depends on how far the particles exhibit hydrophobic
properties.
2 Centrifugal technology in the three-phase decanter
During centrifugation, centrifugal force is produced through the rotation of the bowl which is nor-
mally a multiple of the force of gravity. This is expressed in terms of the so-called g-force which
represents the ratio between centrifugal acceleration and acceleration due to gravity. A relative ve-
locity is applied to the heavy components (particles, drops) of the feed suspension as a function of
the g-force.
The decanter is generally used to separate solids from a liquid (two phases). Three-phase systems
can also be processed with a special version. In addition to separating the solids, the liquid is sepa-
rated into a light and a heavy phase.
Figure 3: Design of a three-phase decanter
Figure 3 is a schematic representation of the design of a three-phase decanter. The separation and
dewatering of the suspended particles, which generally exhibit the highest density of the individual
components, is analogue to other decanter centrifuges by sedimentation on the inner bowl wall. The
two liquid phases are separated simultaneously under the action of centrifugal force. The fluids
build up layers in the bowl in accordance with their density. The phases are separated independently
Copper Crud Treatment with 3 Phase Centrifuge
Proceedings of Copper 2010 2741
from one another at the cylindrical end of the bowl. The respective phases are either discharged un-
der gravity over a weir or under pressure by means of a centripetal pump. In the case of the variant
illustrated in Figure 1, the heavy phase discharges under gravity, the light phase by means of a cen-
tripetal pump [1].
In the system presented here, the overflow diameter of the aqueous phase is firmly defined and can-
not be altered online. The diameter of the organic phase corresponds to the pond depth and is nar-
rower than the overflow diameter of the aqueous phase due to the difference in density between the
two liquids. Between the two layers is a boundary layer at the liquid-side end of the bowl, the so-
called separating zone. The position of the separating zone within the sedimentation pond depends
on the liquids to be processed and the overflow diameters. This is illustrated in Figure 4 where the
position of the separating zone is plotted over the pond depth. The parameter is the overflow diame-
ter of the heavy phase.
Figure 4: Position of the separating zone as a function of the pond depth
To realize a separating process that is as effective as possible, the position of the separating zone
must be adapted to the given phase distribution in the feed suspension which may be subject to tem-
poral variations depending on the raw materials and the process management. A system is therefore
desirable that permits the online displacement of the separating zone as a function of a suitable con-
trol parameter.
Hartmann, Horbach, Kramer
Proceedings of Copper 2010 2742
3 Density dependence of the organic phase
In the process under consideration here, the organic phase is primarily a valuable substance which
should be recycled into the process without impunities if possible. A potential contamination origi-
nates largely from the aqueous phase. There is a density difference between these two liquids. A
contaminated light phase accordingly has a higher density compared to the pure organic liquid. This
is graphically represented in Figure 4.
Figure 5: Density of the contaminated organic phase
The density of the valuable substance therefore serves as an indicator for the contamination of the
organic phase.
4 System description
The system illustrated here is illustrated in Figure 6. The decanter is configured as a three-phase
decanter as described above. The aqueous phase is discharged via the bowl outer diameter into a
catcher and flows off via the decanter frame. The organic phase is discharged under pressure by
means of a centripetal pump.
Copper Crud Treatment with 3 Phase Centrifuge
Proceedings of Copper 2010 2743
Figure 6: Schematic illustration of the concentration-dependent pond adjustment
In addition to the standard pressure gauge, a density measuring device is also integrated in this dis-
charge. In the control unit the measured density signal is evaluated in accordance with Figure 5 to
determine contamination through the aqueous phase and is compared with a limit value.
By way of example, Figure 7 shows a possible phase distribution in the decanter bowl. The organic
phase (yellow) accounts for a very small proportion so that there is a short retention time in the cen-
trifugal field. As a result, heightened contamination of the organic phase can occur which is de-
tected with the density sensor. If the admissible limit value is exceeded, the valve is actuated ac-
cordingly resulting in a higher discharge pressure. As the discharge pressure rises, the immersion
depth of the centripetal pump in the pond increases, which means that the pond depth likewise in-
creases. In accordance with the relation described above, the separating zone is displaced outwards.
This state is illustrated in Figure 8.
Hartmann, Horbach, Kramer
Proceedings of Copper 2010 2744
Figure 7: Position of the separating zone without increased discharge pressure
Figure 8: Position of the separating zone with increased discharge pressure
As a result of the displacement of the separating zone outwards, the volume of the organic phase in
the bowl increases and the retention time in the centrifugal field is longer. The result is an im-
provement in the degree of purity of the organic phase due to the externally induced increase of the
discharge pressure.
Copper Crud Treatment with 3 Phase Centrifuge
Proceedings of Copper 2010 2745
Figure 9: Separating zone diameter as a function of the discharge pressure
If the discharge pressure is increased too sharply, a particularly pure organic phase is recovered but
there is an increased loss of the organic phase into the aqueous phase and possibly also the solids. In
this case, the operator can optimally adjust the parameters throughput capacity, yield of organic
phase and degree of purity of the organic phase to the specific operating situation.
5 Summary
By means of the DControl
[
-
]
0 2 4 6 8 10
0
2
4
6
8
C
o
n
c
.
o
f
f
r
e
e
c
e
l
l
s
,
X
L
1
0
-
1
4
[
c
e
l
l
s
/
m
3
]
Time, t [day]
0
0.2
0.4
0.6
0.8
1.0
C
o
p
p
e
r
l
e
a
c
h
e
d
,
[
-
]
0
0.2
0.4
0.6
0.8
1.0
C
o
p
p
e
r
l
e
a
c
h
e
d
,
[
-
]
0 2 4 6 8 10
0
2
4
6
8
C
o
n
c
.
o
f
f
r
e
e
c
e
l
l
s
,
X
L
1
0
-
1
4
[
c
e
l
l
s
/
m
3
]
Time, t [day]
0 2 4 6 8 10
0
2
4
6
8
C
o
n
c
.
o
f
f
r
e
e
c
e
l
l
s
,
X
L
1
0
-
1
4
[
c
e
l
l
s
/
m
3
]
Time, t [day]
Konishi, Saitoh, Shuto, Ogi, Kawakita, Kamiya
Proceedings of Copper 2010 2786
About 65 % of the copper in the crude chalcopyrite ore was leached in 10 days in a batch reactor.
The bioleaching of copper in the crude chalcopyrite ore resulted in a significant increase in the free
cell concentration X
L
. The rates of bacterial growth and chalcopyrite oxidation for the crude ore
(CuFeS
2
3.3 %, Fe
3
O
4
26.7 %, and SiO
2
46.3 %) were almost the same as those for the chalcopyrite
concentrate (Cu 22.5 %, Fe 27.8 %, S 31.4 %). We conclude that gangue minerals in the crude ore,
such as magnetite and silica, contribute little to chalcopyrite bioleaching with A. briereleyi.
3.2 Bioleaching of crude chalcopyrite ore by A. brierleyi
Figure 4 shows the bioleaching behavior of crude chalcopyrite ore with A. brierleyi at 65 C and at
an initial pH of 1.8. Rapid leaching of chalcopyrite with A. brierleyi was obtained for the crude ore
and 65 % of the copper was leached in 10 days. By comparison, the bioleaching of iron in the crude
chalcopyrite ore was very slow and only about 10 % of the iron was leached in 10 days in the pres-
ence or absence of A. brierleyi. In other words, the leaching of copper took place selectively when
A. brierleyi was used on chalcopyrite because magnetite leaching with A. brierleyi was negligible.
These results demonstrate that thermophilic bioleaching with A. brierleyi is attractive as an eco-
nomical and environmentally friendly process for good copper extraction from a crude chalcopyrite
ore.
Figure 4: Rate data on the batch bioleaching of crude chalcopyrite ore at an initial mineral-liquid
loading ratio of W
0
/V = 5 kg/m
3
, an initial pH of 1.8 and 65 C: () copper leaching
with A. brierleyi at an initial cell concentration of X
T0
= 1.210
14
cells/m
3
; () iron
leaching with A. brierleyi at an initial cell concentration of X
T0
= 1.210
14
cells/m
3
;
() copper leaching without A. brierleyi (sterile control); () iron leaching without
A. brierleyi (sterile control).
0 2 4 6 8 10
0
0.2
0.4
0.6
0.8
1.0
Time, t [day]
F
r
a
c
t
i
o
n
o
f
m
e
t
a
l
s
l
e
a
c
h
e
d
,
[
-
]
0 2 4 6 8 10
0
0.2
0.4
0.6
0.8
1.0
Time, t [day]
F
r
a
c
t
i
o
n
o
f
m
e
t
a
l
s
l
e
a
c
h
e
d
,
[
-
]
Bioleaching of Crude Chalcopyrite Ores by the Thermophilic Archaean Acidianus Brierleyi
Proceedings of Copper 2010 2787
Bioleaching experiments with A. brierleyi were done to examine the effect of initial ore-liquid load-
ing ratio W
0
/V. The results of these experiments are shown in Figure 5 where the concentration of
free cells in the leach solution, X
L
, and the fraction of chalcopyrite leached, , are plotted as a func-
tion of operating time. When the initial amount of the crude chalcopyrite ore, W
0
/V, was increased
from 5 to 20 kg/m
3
, the free cell concentration, X
L
, increased markedly but the leaching fraction of
copper, , changed slightly.
Figure 5: Effect of initial mineral-liquid loading ratio W
0
/V on the rate of chalcopyrite leaching
with A. brierleyi at an initial pH of 1.8 and an initial cell concentration of
X
T0
= (0.8-1.2)10
14
cells/m
3
: () W
0
/V = 5 kg/m
3
; () W
0
/V = 10 kg/m
3
;
() W
0
/V = 20 kg/m
3
; () W
0
/V = 50 kg/m
3
.
Konishi, Saitoh, Shuto, Ogi, Kawakita, Kamiya
Proceedings of Copper 2010 2788
Figure 6 shows the effect of initial pH on the bioleaching of the crude chalcopyrite ore with A. bri-
erleyi at an initial ore-liquid loading ratio of 5 kg/m
3
. In the range of the initial pH between 1.2 to
2.0 the effects of pH on the bacterial growth and chalcopyrite leaching were slightly.
Figure 6: Effect of initial pH on the rate of chalcopyrite leaching with A. brierleyi at an initial
mineral-liquid loading ratio of W
0
/V = 5 kg/m
3
and an initial cell concentration of
X
T0
= (1.2-1.8)10
14
cells/m
3
: () pH 1.2; () pH 1.5; () pH 1.8; () pH 2.0.
Bioleaching of Crude Chalcopyrite Ores by the Thermophilic Archaean Acidianus Brierleyi
Proceedings of Copper 2010 2789
3.3 Bioleaching mechanism of copper in crude ore with A. brierleyi
The mechanisms of microbial action in sulfide mineral oxidation are usually discussed in terms of a
direct microbial attack on the sulfide and an indirect attack via ferric iron. For the bioleaching of
chalcopyrite with A. brierleyi, both the oxidizable metal moiety (Fe(II)) and the sulfide moiety are
considered to be simultaneously attacked by separate enzymes [10]:
4CuFeS
2
+ 17O
2
+ 2H
2
SO
4
4CuSO
4
+ 2Fe
2
(SO
4
)
3
+ 2H
2
O (1)
When ferric ions are present in the leach solution, chalcopyrite is chemically oxidized with ferric
ion:
CuFeS
2
+ 2Fe
2
(SO
4
)
3
CuSO
4
+ 5FeSO
4
+ 2S
0
(2)
In the presence of A. brierleyi, ferrous ions and elemental sulfur are microbially oxidized:
4FeSO
4
+ O
2
+ 2H
2
SO
4
2Fe
2
(SO
4
)
3
+ 2H
2
O (3)
2S
0
+ 3O
2
+ 2H
2
O 2H
2
SO
4
(4)
Because ferric ions are automatically supplied by the solubilization of chalcopyrite, the bioleaching
of chalcopyrite may occur by both direct microbial action and indirect chemical action.
To examine the mechanism of chalcopyrite leaching with A. brierleyi, bioleaching experiments
were carried out at different initial ferric ion concentrations, [Fe(III)]
0
, at an initial ore-liquid load-
ing ratio W
0
/V of 5 kg/m
3
(Figure 7). Chalcopyrite leaching was stimulated in the presence of A.
brierleyi. Chalcopyrite bioleaching at [Fe(III)]
0
= 0 kg/m
3
resulted in an increase in the liquid-phase
ferric ion concentration to 0.04 kg/m
3
after 10 days but no accumulation of ferrous iron in the leach
solution was found. At an initial ferric iron concentration of 1.00 kg/m
3
, the ferric ion concentration
in the solution increased to 1.04 kg/m
3
after 10 days. There was a marked difference in the ferric ion
concentration during the bioleaching. Nevertheless, the initial addition of ferric ions to the A. brier-
leyi cultures only had a slight effect on the rate of chalcopyrite leaching. A previous study [11] re-
ported that the reaction kinetics of chemical leaching exhibits a half order dependence on the ferric
ion concentration. If chemical leaching with ferric ion contributes to chalcopyrite leaching, the ini-
tial concentration of ferric iron, [Fe(III)]
0
, would have a significant effect on the bioleaching rate.
Figure 7 shows that the bioleaching rates are independent of the ferric ion concentration. This dem-
onstrates that chemical leaching by ferric ion has a negligible contribution to bioleaching with A.
brierleyi. Thus, we conclude that chalcopyrite leaching by A. brierleyi is predominantly due to the
direct microbial attack by A. brierleyi cells.
Konishi, Saitoh, Shuto, Ogi, Kawakita, Kamiya
Proceedings of Copper 2010 2790
Figure 7: Effect of initial ferric iron concentration [Fe(III)]
0
on the rate of chalcopyrite leaching
with A. brierleyi at an initial mineral-liquid loading ratio of W
0
/V = 5 kg/m
3
, an initial
pH of 1.2 and an initial cell concentration of X
T0
= (1.2-1.5)10
14
cells/m
3
:
() [Fe(III)]
0
= 1.0 kg/m
3
; () [Fe(III)]
0
= 0.5 kg/m
3
; () [Fe(III)]
0
= 0.0 kg/m
3
;
() [Fe(III)]
0
= 1.0 kg/m
3
; () [Fe(III)]
0
= 0.0 kg/m
3
.
0 2 4 6 8 10
0
0.2
0.4
0.6
0.8
1.0
C
o
p
p
e
r
l
e
a
c
h
e
d
,
[
-
]
Time, t [day]
0 2 4 6 8 10
0
1
2
3
4
5
6
C
o
n
c
.
o
f
f
r
e
e
c
e
l
l
s
,
X
L
1
0
-
1
4
[
c
e
l
l
s
/
m
3
]
Time, t [day]
Bioleaching of Crude Chalcopyrite Ores by the Thermophilic Archaean Acidianus Brierleyi
Proceedings of Copper 2010 2791
4 Conclusion
We investigated the mechanism and kinetics of crude chalcopyrite ore leaching by the thermophilic
archaean Acidianus brierleyi at 65 C and at pH 1.8-2.0 in a batch reactor. About 65 % of the cop-
per in the crude chalcopyrite ore (size range < 25 m; Cu 1.2 %, Fe 20.4 %, S 2.4 %, SiO
2
46.3 %)
was leached in 10 days in the presence of Acidianus brierleyi. The copper leaching rate for the crude
ore was almost the same as that for a chalcopyrite concentrate (Cu 22.5 %, Fe 27.8 %, S 31.4 %),
suggesting that the effects of the gangue minerals such as magnetite and silica is little. The bioleach-
ing of iron was very slow and only about 10 % iron was leached in 10 days in the presence or ab-
sence of A. brierleyi. In other words, chalcopyrite leaching selectively took place and magnetite
leaching was negligible. These results lead us to conclude that thermophilic bioleaching with A.
brierleyi is attractive as an economical and environmentally friendly process for good copper extrac-
tion from low-grade primary sulfides.
References
[1] Rawlings, D. E., and Johson, D.B., 2007, Biomining, Springer, Heidelberg.
[2] Donnati, E. R., and Sand, W., 2007, Microbial Processing of Metal Sulfides, Springer, Hei-
delberg.
[3] Clark, M. E., Batty, J. D., van Buuren, C. B., Dew, D. W., and Eamon, M. A., 2006, Bio-
technology in Minerals Processing: Technological Breakthroughs Creating Value, Hydro-
metallurgy, 83, pp. 3-9.
[4] J. A. Brierley, Acidophilic thermophilic archaebacteria: potential application for metal recov-
ery, FEMS Microbiol. Rev., 75 (1990) 287-292.
[5] J. A. Brierley, C. L. Brierley, Microbial mining using thermophilic microorganisms, in T.D.
Brock (ed.), Thermophiles: General, Molecular, and Applied Microbiology, John Wiley &
Sons, New York, 1986, pp. 279-305.
[6] Y. Konishi, S. Yoshida, S. Asai, Bioleaching of pyrite by acidophilic thermophile Acidianus
brierleyi, Biotechnol. Bioeng., 48 (1995) 592-600.
[7] Y. Konishi, S. Asai, M. Tokushige, T. Suzuki, Kinetics of the bioleaching of chalcopyrite
concentrate by acidophilic thermophile Acidianus brierleyi, Biotechnol. Prog., 15 (1999) 681-
688.
[8] Y. Konishi, M. Tokushige, S. Asai and T. Suzuki, Copper recovery from chalcopyrite con-
centrate by Acidophilic Thermophile Acidianus brierleyi in batch and continuous-flow stirred
tank reactors, Hydrometallurgy, Vol. 59, 2001, 271-282.
Konishi, Saitoh, Shuto, Ogi, Kawakita, Kamiya
Proceedings of Copper 2010 2792
[9] M. P. Silvermann, D. G. Lundgren, Studies on chemoautotrophic iron bacterium Ferrobacil-
lus ferrooxidans. I. An improved medium and a harvesting procedure for securining high cell
yields, J. Bacteriol., 77 (1959) 642-647.
[10] H. L. Ehrlich, Geomicrobiology, 4
th
Ed., Marcell Dekker Inc., New York, NY, USA, 2002.
[11] P. B. Munoz, J. D. Miller, M. E. Wadsworth, Reaction mechanism for the acid ferric sulfate
leaching of chalcopyrite, Metall. Trans. B, 10B (1979) 149-158.
Proceedings of Copper 2010 2793
Hybrid Flotation Newly Developed Flotation
Technology for Increased Recovery
Especially in the Finest Particle Fractions
W. Krieglstein, L. Grossmann
Siemens AG, Industry Sector, Industry Solutions Division, Metals Technology, Mining
(I IS MT MI)
Schuhstrae 60
Erlangen, Germany
Keywords: Hybrid flotation, Los Pelambres
Abstract
This paper presents a study to estimate the effect of Siemens SIMINE
CIS
HybridFlot flotation tech-
nology on the processing of copper-molybdenum sulphide minerals. A pilot flotation cell was tested
in the Antofagasta Minerals S. A. (AMSA) operation Minera Los Pelambres (MLP), Chile, in 2007.
In Hybrid Flotation, the so-called pneumatic spray-in principle is combined with the column
method. Using SIMINE
CIS
HybridFlot flotation cells, an increase of molybdenum recovery in the
selective process is achieved particularly with regard to finest and coarsest particles.
1 Introduction
Froth flotation is an important method for the separation and enrichment of mineral raw materials.
This separation method is based on the interfacial phenomenon in the three-phase system liquid-
solid-gas [1]. Among others, flotation reagents play a decisive role in the separation process [2]. By
using these reagents, the flotation or non-flotation of particular minerals can be regulated. Ahead of
froth flotation, grinding is required for the liberation of the valuable minerals from the gangue to
achieve a good concentrate grade and recovery. Fine and ultra fine particle fractions generated by
the grinding process may hinder the downstream separation process especially in terms of recovery
and turn the flotation into a complicated process. Fine particles have a small mass and relatively
high surface area.
These properties can affect and result in the following phenomena: a higher dissolution rate in fluid,
a rigidity of froth, a high pulp viscosity, a heterocoagulation, a low probability of collision between
Krieglstein, Grossmann
Proceedings of Copper 2010 2794
particles and gas bubbles, a low particle momentum, and a gangue particle entrainment into the con-
centrate product [3].
This may also lead to a higher consumption of flotation reagents and to undesirable coating of gas
bubbles and valuable particles with ultrafine slimes [3]. It is also believed that overcoming the en-
ergy barrier between particle-particle and particle-bubble becomes more difficult for fine parti-
cles [3]. A continuously increasing appearance of ores with falling head grades and their benefici-
ation effect the flotation of fine particles to become more and more relevant in the future [4].
The Mining segment at the Siemens Metals Technologies Unit is currently developing a portfolio
for flotation cells, aiming to be able to provide complete flotation process solutions in future. In this
regard, Siemens Mining Technologies has developed and built a pilot flotation cell, commissioned
and tested in AMSAs Minera Los Pelambres (MLP), Chile. In this paper, the test results using the
Hybrid Flotation method in the molybdenum beneficiation process will be shown.
2 Flotation
One of the main challenges for flotation equipment providers is the development of flotation cells
which disperse a sufficient amount of fine air bubbles into the pulp for the recovery of fine grinded
valuable particles among coarser particle fractions.
The flotation process takes place under following conditions:
- A particle-gas bubble contact is given,
- Particles attach to bubbles and form aggregates,
- The particle-bubble-aggregates do not detach during floating from the pulp zone into the froth
zone of the machine and are transported from there into the concentrate collecting devices.
The probability P for a particle being recovered during the flotation process depends on the prob-
ability of the particle-bubble collision P
c
, the probability of attachment of particles to the bubbles P
a
and the probability of detachment of the particles from the bubble P
d
[3, 5].
( )
d a c
P P P P = 1 (1)
In literature, a generalized formula can be found for the probability of the particle collision with gas
bubbles P
c
depending on the particle size d
p
and the bubble d
b
size [5]:
n
b
p
c c
d
d
k P
= (2)
The Parameters k
c
and n are dependent on the Reynolds number Re i.e. the hydrodynamic milieu
during the flotation process.
Hybrid Flotation
Proceedings of Copper 2010 2795
The probability of the attachment depends, among others, on the hydrophobicity of the parti-
cles [3, 5]. The most important mechanism of the flotation is the attachment of particles to the gas
bubbles and it can be influenced by the addition of flotation reagents [3].
The probability of the detachment depends, among others, on the collision with other particles and
on the turbulence of the flotation cell [3] and can be defined as follows [5]:
5 . 1
max ,
=
p
p
d
d
d
P (3)
where d
p,max
is the particle size of the major floatable particle.
Furthermore, bubble and particle size distributions and their relation play a decisive role in the flota-
tion process [2, 3, 5].
As a consequence, research and development activities are focused on the hydrodynamic character-
istics of the flotation cells beside the areas of specific energy consumption, maintainability and
others.
3 Hybrid Flotation
In the design of Siemens SIMINE
CIS
HybridFlot flotation cell, a so-called pneumatic spray-in prin-
ciple is combined with the column method (Figure 1). The cell operates without an agitator, because
the ore slurry is sprayed into the cell by high-pressure nozzles. The gas is added to the ore slurry in
mixing chambers before the 3-phase mixture enters into the cell. The special design of the nozzles
ensures the generation of finest gas bubbles. This considerably improves the frequency of contact
between the gas bubbles and very fine particles (Equation 2), and also improves the ability of the
particles to stick to these bubbles. The resulting mixture is sprayed into the flotation cell in a tan-
gential arrangement.
Krieglstein, Grossmann
Proceedings of Copper 2010 2796
Figure 1: SIMINE
CIS
HybridFlot flotation cell [6]
Additional gas is added to an internal column in the second stage (Figure 1), which ensures that the
ore particles not captured in the first stage can collide with the gas bubbles of the second stage
and be transported to the surface. The froth concentrate produced during the process is taken to the
edge of the flotation tank and drained off. Short retention times of the pulp in the machine and new
patented additional drain gutters reduce the risk of loosing already captured particles. The
SIMINE
CIS
HybridFlot flotation cell as tested at Minera Los Pelambres can be operated with pulp
feed having up to 50 percent solid content and has a nominal capacity of up to 400 m/h.
4 Flotation Process in Minera Los Pelambres
In Minera Los Pelambres, the porphyry copper minerals like chalcopyrite, chalcocite, bornite, covel-
lite and molybdenite are processed with typical porphyry ores head grades for copper and molybde-
num. The recovery of molybdenum is carried out as a by-process of the flotation of copper sulphide
minerals. Concentration and separation takes place in two steps, carried out in a bulk flotation plant
and a molybdenum plant. In the bulk flotation process, rougher, scavenger and cleaner flotation are
Hybrid Flotation
Proceedings of Copper 2010 2797
implemented with the associated concentrate and regrind stages [2]. Bulk copper-molybdenum con-
centrate goes to a thickener prior to its treatment in the molybdenum plant (Figure 2). The
SIMINE
CIS
HybridFlot flotation cell was implemented in the existing flotation circuit as a pre-
rougher cell. The concentrate from the SIMINE
CIS
HybridFlot flotation cell was led to a pneumatic
cell which was operated as cleaner and produced final concentrate. The tailings from the SIMINE
CIS
HybridFlot flotation cell were fed to the rougher flotation bank of the selective process.
Figure 2: Flow sheet of the molybdenum beneficiation process in Mine Los Pelambres
5 Test Results
This paper presents the results of the in-plant equipment testing of the SIMINE
CIS
HybridFlot flota-
tion cell. During the in-plant testing of the SIMINE
CIS
HybridFlot flotation cell at Los Pelambres,
samples were collected from the feed, concentrate and tailings flows. Primarily, the SIMINE
CIS
Hy-
bridFlot flotation cell was integrated into the existing flotation circuit and was operated for 4 hours
and respectively 3 samples were taken from the flows. The concentrates from the regular process
and the concentrate of the SIMINE
CIS
Hybrid Flot flotation cell were united, see Figure 3. After-
wards, the cell was disconnected from the flotation process. The plant was operated then without the
SIMINE
CIS
HybridFlot flotation cell until a stabilisation of the process (after approx. 2 hours) was
achieved. The sampling for characterising the existing flotation circuit began after the stabilisation.
As a consequence from this test set up, the concentrate mix consisting of concentrate from the con-
Krieglstein, Grossmann
Proceedings of Copper 2010 2798
ventional mechanical rougher cells and concentrate from the SIMINE
CIS
HybridFlot flotation cell
was compared with concentrate coming only from the conventional mechanical rougher cells. The
tests were realized at different feed mass flow levels, ranging from 100 to 120 and 140 tons per hour
and the mass concentration was uniform. At the molybdenum concentrate sampling point (Figure 3),
the different concentrate samples were taken and analysed.
Figure 3: Test set up and Sampling points
The results of the analysis of the samples taken are illustrated in Figure 4. The metallurgical perform-
ance of the existing molybdenum flotation circuit could be enhanced by an increasement of molybde-
num recovery at per average 1.2 percentage points. This improvement of the metallurgical perform-
ance is associated with a higher selectivity of the flotation process. The recovery of copper and iron
into the concentrate is dropped by an average of 1.7 and 1 percent respectively. For the recovery of
molybdenum, the optimum operation range was found at the solid mass flow Q
s
= 120 t/h.
Hybrid Flotation
Proceedings of Copper 2010 2799
Figure 4: Recovery of molybdenum, copper and iron in the flotation process
Figure 5 demonstrates the selectivity of the existing molybdenum flotation circuit with and without
the SIMINE
CIS
Hybrid Flot flotation cell in operation. In Figure 5, the content of valuable mineral
(molybdenum) and the unwanted minerals in the concentrate depending on the solid mass flow are
illustrated. The flotation is seen to be improved by integration of the SIMINE
CIS
HybridFlot flota-
tion cell and the molybdenum grade is noted to be increased by 0.6 percent on average. It can be
concluded that the selectivity of the process is higher when the SIMINE
CIS
HybridFlot flotation cell
is integrated in the flotation circuit. The copper and iron grades in the concentrate drop by an aver-
age of 1.1 points when the SIMINE
CIS
HybridFlot flotation cell is integrated into the flotation cir-
cuit. The highest molybdenum grade in the concentrate was achieved at low solid mass flow
Q
s
= 100 t/h in the investigated range.
By integrating the SIMINE
CIS
HybridFlot flotation cell into the existing flotation circuit, an increase
of molybdenum recovery and grade in the concentrate can be achieved with simultaneously increas-
ing of the selectivity of the flotation process.
With the SIMINE
CIS
HybridFlot flotation cell, the valuable particles can be removed from the pulp
quickly. The residence time in the cell was between 3.8 and 5.3 minutes. In the existing flotation
bank at Los Pelambres, the pulp has a typical residence time of approx. 20 to 30 minutes, other se-
lective rougher processes do show even much higher residence times.
100 120 140
0
10
20
30
40
50
60
70
80
90
100
R
e
c
o
v
e
r
y
i
n
%
Solid Mass Flow in t/h
Molybdenum (with Hybrid Flotation Cell)
Molybdenum (without Hybrid Flotation Cell)
Copper (with Hybrid Flotation Cell)
Copper (without Hybrid Flotation Cell)
Iron (with Hybrid Flotation Cell)
Iron (without Hybrid Flotation Cell)
Krieglstein, Grossmann
Proceedings of Copper 2010 2800
Figure 5: Molybdenum, copper and iron grades in the concentrates
Recovery and molybdenum grade in the concentrate for the molybdenum rougher flotation process
are summarized in Table 1, classified by different particle size fractions. Analysis of particle sizes
was carried out by sieve tests.
Table 1: Molybdenum recoveries and grades in the concentrates of the rougher flotation process
Particle Size Hybrid Flotation Recovery Grade
d < 10 m ON 97.5 % 5.9 %
OFF 95.2% 6.6 %
10 m < d < 44 m ON 97 % 6.9 %
OFF 96 % 6.9 %
44 m < d < 74 m ON 97.8 % 38.1 %
OFF 96.3 % 29.9 %
74 m < d ON 96 % 37 %
OFF 90.2 % 29.7 %
100 120 140
0
5
10
15
20
25
30
35
C
o
n
t
e
n
t
i
n
%
Solid Mass Flow in t/h
Molybdenum (with Hybrid Flotation Cell)
Molybdenum (without Hybrid Flotation Cell)
Copper (with Hybrid Flotation Cell)
Copper (without Hybrid Flotation Cell)
Iron (with Hybrid Flotation Cell)
Iron (witout Hybrid Flotation Cell)
Hybrid Flotation
Proceedings of Copper 2010 2801
The highest increase in recovery is achieved for the particle fractions with a mean particle diameter
d > 74 m and d < 10 m. It can be assumed that the improvement of the metallurgical performance
of the molybdenum rougher flotation process is attributable to the previous recovery of fine particles
by the SIMINE
CIS
HybridFlot flotation cell. A plausible explanation of the data shown in table 1 can
be given by the assumption that two major effects are responsible for the overall metallurgical per-
formance improvement by implementing the SIMINE
CIS
HybridFlot flotation cell: Firstly, the addi-
tionally recovered fine particles fractions per se lead to a higher overall recovery without having a
negative effect on the grade. Secondly, the removal of these fine particle fractions might positively
influence the ability to recover coarser particle fractions in the down stream mechanical rougher
cells by preventing undesired effects like coating of valuable coarser mineral particles and gas bub-
bles by fine particles.
The findings and assumption as described above lead to the necessity of a more detailled analysis of
the influence of particle sizes. The cumulative undersize of the feed is shown in Figure 6. It can be
seen that the major part of the particles in the feed (d
50
= 18 m) is fine.
Figure 6: Cumulative undersize of the feed measured by a sieve test
The cumulative undersize of the concentrate of the SIMINE
CIS
Hybrid Flot flotation cell, measured
by Cyclosizer, is illustrated in Figure 7. It can be noted that the majority of the particles in the con-
centrate (approx. 58 %) has a particle size smaller than 10 m. This underlies the assumption that
Krieglstein, Grossmann
Proceedings of Copper 2010 2802
the SIMINE
CIS
HybridFlot flotation cell has a high recovery efficiency especially for fine and ultra
fine particle fractions.
Samples were drawn form the inner cylinder section (column method) and the peripheral section
(spray-in principle) of the SIMINE
CIS
HybridFlot flotation cell, see Figure 1. It was assumed that the
majority of the fine particle fractions are recovered in the first flotation stage in the upper section of
the SIMINE
CIS
HybridFlot flotation cell, i.e. by the spray-in principle. On the contrary, particles in
the in-cylinder are slightly finer than those in the peripheral part of the SIMINE
CIS
HybridFlot flota-
tion cell. This minimal effect can be explained by the sampling. The froth from in-cylinder is led to
the peripheral part of the cell (Figure 1). Therefore, both different froth products generated by the
two different principles in the SIMINE
CIS
HybridFlot flotation cell were mixed, before the entity of
both froth products was led out of the machine.
It is for this reason that a comparison of the different froth products was only possible by comparing
a froth mix against the product of the second stage, e.g. in-cylinder principle. A significant differ-
ence in terms of recoveries and grades of molybdenum of both froth products generated by the two
principles in the machine could therefore not be proven.
Figure 7: Cumulative undersize of the concentrate of SIMINE
CIS
HybridFlot flotation cell
All results presented so far show that an implementation of a single SIMINE
CIS
HybridFlot flotation
cell as an additional pre-rougher stage into an existing copper-molybdenum selective flotation pro-
cess can improve recovery and selectivity of the overall process, attributable to the improved re-
Hybrid Flotation
Proceedings of Copper 2010 2803
covery of fine particle fractions, the consequential positive impact on the downstream mechanical
rougher cells and the simple provision of additional process volume.
To also address the question to which extent a single Siemens cell may replace an entire bank of
mechanical rougher cells, the results of the metallurgical performance test of the single SIMINE
CIS
HybridFlot flotation cell are summarized in Table 2, classified by 2 different particle size fractions.
Table 2: Molybdenum recovery and grade with SIMINE
CIS
HybridFlot flotation cell
Particle Size Recovery Grade Enrichment factor approx.
d < 44 m 18.5 % 15.9 % -
d > 44 m 37.0 % 50.0 % -
Overall 22.1 % 24.0 % 8 10
The overall recovery of the SIMINE
CIS
HybridFlot flotation cell was 22.1 % with an overall 24 %
molybdenum grade in the concentrate, resulting in a range of 8 to 10 for the enrichment factor. It
can be concluded that, as expectable, a single SIMINE
CIS
HybridFlot flotation cell has a lower re-
covery than usually a full size mechanical rougher flotation bank has. On the other hand, the achiev-
able molybdenum grade for the concentrate respectively the ratio of concentration is very high with
the SIMINE
CIS
HybridFlot flotation cell.
6 Conclusions
The test with the first SIMINE
CIS
HybridFlot flotation pilot cell in the selective copper-molybdenum
process showed an increasement of molybdenum recovery by more than 1 percentage point per av-
erage of the overall rougher process. Depending of the feed load of the process, affecting the opti-
mum operation range of the SIMINE
CIS
HybridFlot flotation cell, overall recovery increasement by
more than 2 percent could be reached. The recovery increase is shown especially in the fine particle
fractions (d < 10 m), but also in the coarser particle size fractions (d > 74 m). It can be assumed
that this improved overall recovery of the coarser particles is due to a reduction of an undesired
coating of the valuable particles and gas bubbles with ultrafine slimes by recovering fine particle
fractions, operating the SIMINE
CIS
HybridFlot flotation cell as a pre-flotation stage of the process.
Furthermore, the ability to create fine gas bubble sizes, very short retention times and a very quick
removal of the froth products lead to an increased recovery of fine particle fractions. The molybde-
num concentration could be enriched from a typical concentration of less than 3 % in the feed pulp
up to approximately 24 % percent on average and up to nearly 40 % as maximum in the concentrate
product of the SIMINE
CIS
HybridFlot flotation cell using only one cell. The fraction of fine particles
and coarse particles which are lost in the existing flotation process can be recovered with good effi-
ciency by integration of the SIMINE
CIS
HybridFlot flotation cell into the process as pre-rougher
stage.
Krieglstein, Grossmann
Proceedings of Copper 2010 2804
7 Outlook
Based on the achieved results, ongoing research and development works are currently done to de-
scribe the hydrodynamic characteristics of both the two basical principles in the SIMINE
CIS
Hybrid-
Flot flotation cell separately as well as an integrated, interacting system. To be able to address these
fields of research work, a laboratory device is set up and operated accordingly. Furthermore, a
downscaled mobile test unit of the SIMINE
CIS
HybridFlot flotation cell is available to be able to
conduct further test work in operational industrial sites.
Once detailled hydrodynamic characterizations and kinetic modelling are available in more details,
partial process solutions and in a further step fully integrated flotation processes can be elaborated
by combining and cascading SIMINE
CIS
HybridFlot flotation cells if applicable in different sizes.
Prospectively, Siemens is also striving to transfer its SIMINE
CIS
HybridFlot flotation technology
from copper-molybdenum selective processes into applications in copper bulk and other minerals
flotation processes.
Acknowledgments
Siemens AG would like to acknowledge the contributions of the personnel at the Minera Los Pe-
lambres operation of Antofagasta Minerals S.A., Chile. This project would not have been possible
without Mr. Dalibor Dragisevic, Mr. Nestor Villarroel, Mr. Jorge Cortinez and Mr. Andres Soto.
The authors wish to express their gratitude to them because of their invaluable openness, assistance
and support during the project. Their commitment in terms of time and openness to use new tech-
nology is gratefully acknowledged.
Hybrid Flotation
Proceedings of Copper 2010 2805
References
[1] WILLS, B. & NAPIER-MUNN, T. (2006): Wills Mineral Processing Technology: An Intro-
duction to the Practical Aspects of Ore Treatment and Mineral Recovery, Oxford (Butterworth-
Heinemann) ISBN 0-8247-9264-5
[2] BULATOVIC, S. M. (2007): Handbook of Flotation Reagents. Chemistry, Theory and Practice,
Amsterdam (Elsevier) ISBN 978-0-444-53029-5
[3] MATIS, K. A. (1995): Flotation science and engineering, New York (Marcel Dekker) ISBN
978-0-750-64450-1
[4] CROZIER, R. D. (1992): Flotation: theory, reagents and ore testing, Oxford (Pergamon)
ISBN 0-08-041864-3
[5] LI, R. (1996): Untersuchung ber Flotationsgrundlagen unter besonderer Bercksichtigung der
Korngre, Aachen (Shaker) ISBN 3-8265-1635-4
[6] VIDUYETSKY, M. G., GARIFULIN, I. G. & MALZEW, V. A. (2006): Patent WO 069995,
Pneumatic flotation column comprising a foam collecting container
Proceedings of Copper 2010 2806
Proceedings of Copper 2010 2807
Perspectives of Copper Mining Industry
Development in Poland
Ph.D. Eng. Jan Kudeko, Ph.D. Jacek Pyra, Ph.D. Eng. Jerzy Sobociski
KGHM CUPRUM Ltd Research and Development Centre
ul. Gen. Sikorskiego 2-8
53-659 Wroclaw, Poland
Keywords: Mining, copper, production, profit
Abstract
To determine mining profits possibility, copper ore and accompanying minerals were analyzed on
the basis of the copper ore resources. Future technical and economic developments of the copper
industry in Poland for the next ten years were evaluated considering actual state of copper ore stock
balance. Presented production-economic results obtained in the last 12 years by KGHM Polska
Miedz S. A. led to estimate company effectiveness's forecast in aspect of macroeconomic conditions
and estimated production values and costs.
1 Introduction
Development policy of domestic copper mining industry is a very important issue for the Polish
economy especially the regional one. Defining it at the very early phase of investment cycle, permits
not only for selection of adequate development strategy but also contribute to prepare the most op-
timum procedure of already chosen policy. Type of mineral, quality of deposit, presence of base and
accompanying minerals together with development of winning and processing technology, deter-
mine to the great extend, the development trends of KGHM Polska Miedz S. A. However, the
changes in word market situation, marked metals demand or even employment opportunities in the
region of possible deposit occurrence have a substantial impact on changes of previously deter-
mined development policy. Adapting to the market situation, basing on business analyzes of already
confirmed metals deposits at established assumption, it is possible to determine the project profit-
ability conditions and choose the most cost effective of deposit extraction as well and time of its
duration. Expanding the owned resource base by adding new deposit or widening the boundaries of
already mined deposit makes the possible mining period longer. Thus, very important are the explo-
ration works on perspective areas, where depending on economic conditions, further activities may
be taken.
Kudeko, Pyra, Sobociski
Proceedings of Copper 2010 2808
Using new technologies in mining of copper ore results in more cost effective production, what has
the direct relation with copper production profitability, even when mining the ore having worse
parameters. However utilization of these technologies gives the measurable results in longer time
perspective, due to the substantial costs of their winning and implementation.
The impact of world market metal prices fluctuation on the worthwhile mining of the poly-metallic
deposits is presented in the paper. This was the base to analyze the KGHM Polska Miedz S. A.
development trends in the future.
2 Reserves of copper deposits in Poland
Polish copper deposits are of sedimentary type and occur in the Zechstein (Permian) formation on
the Foresudetic Monocline, as well as in Northsudetic Basin. Presence of 14 copper deposits con-
firmed: 11 on the Foresudetic Monocline and 3 in the Northsudetic Basin.
At present 6 deposits on the Foresudetic Monocline: Lubin-Malomice Polkowice,
Sieroszowice, Radwanice Wschod, Rudna and Glogow Gleboki Przemyslowy (GG-P) are
mined. Other deposits such as Bytom Odrzanski, Gaworzyce, Glogow, Radwanice Zachod
and Retkow are now not developer are the future resource base.
On the Northsudetic Basin 3 deposits: Niecka Grodziecka", Wartowice" and Nowy Kosciol" are
located.
Deposits of copper ore mined by KGHM are located on Lower Silesia between Lubin and Glogow.
The deposit area confirmed by exploration extends 40 km along the longness and 20 km along the
dip on the depth of from 600 to 1380 meters. The orebody occurs within the Zechstein sedimentary
formation, which is inclined in the form of monocline towards north-east.
Economic reserve base of KGHM copper deposit as for 31.12.2005 are 922 million Mg of ore,
21.3 million Mg of copper and 58.5 thousand Mg of silver (Table 1).
The biggest amount of reserves bound in pillars is in Lubin-Malomice and Rudna deposit
about 150 million Mg each, mainly in protection pillar of Lubin town and protection pillar of Ze-
lazny Most tailings management facility.
Getting the reserves of Glogow Gleboki-Przemyslowy deposit will let to restore the base of re-
sources supposed economic from years 1998-1999 and base of economic reserve base from early
nineties of previous century (Table 1).
In further time perspective there will be a possibility to enlarge the resources of copper mining in-
dustry by the resource supposed economic of copper ore occurring on the following reserve areas
Gaworzyce (44.8 million Mg) and "Radwanice Zachod" (18.6 million Mg) located west from
Sieroszowice and Radwanice Wschod deposits as well as deposits located along the inclination,
towards north-east and north from Glogow Gleboki deposit Przemyslowy i.e. Bytom Odrzan-
Perspectives of Copper Mining Industry Development in Poland
Proceedings of Copper 2010 2809
ski (31.5 million Mg) and Retkow (135.8 million Mg). Due to the absence of present economic
criteria for copper deposits below 1250 m, reserve area Glogow does not have the confirmed sup-
posed economic resources of copper ore.
In so called Old Miedz Basin the deposit with potential mining capability is joint depost Niecka
Grodziecka and Wartowice (Table 2, 3). Total supposed economic resources are there 84.2 million
Mg of ore, and contain 1.3 million Mg of copper and 4.6 thousand Mg of silver.
Presence of perspective supposed economic resources of copper ore should be bound with both the
foresudetic area (Zechstein deposits), and sedimentary-magmatic rock of north periphery of Upper
Silesian Coal Basin (GZW). However extracting the foresudetic deposits is very impedimental due
to the unfavorable geological conditions, while north periphery of GZW in not enough explored yet.
Probable reserves in Zechstein formation on the Foresudetic monocline amount 29.7 billion Mg of
ore (165 million Mg Cu). Explored reserves of D
1
category (west from Wartowice deposit and Gaw-
rony Scinawa area) are estimated to be about 3.9 billion Mg of ore (about 17 million Mg of Cu).
These are areas with relatively high level of exploration, but mostly with not enough high metals
concentration within two meters interval. Probable reserves of copper ore explored in D
2
category
are estimated to be 25 billion Mg of ore (150 million Mg of Cu). Exploratory boreholes network is
very irregular since most of them were drilled by petroleum industry and their spacing is adapted to
gas and oil prospection. The situation is also made more complicated by the fact that they may be
located totally within the area of gas symptoms, making difficult the future mining (Table 4).
Kudeko, Pyra, Sobociski
Proceedings of Copper 2010 2810
Table 1: Reserves, supposed economic and not supposed economic resources of copper KGHM
Polska Miedz S. A. deposit as for 31.12.2005
Perspectives of Copper Mining Industry Development in Poland
Proceedings of Copper 2010 2811
Table 2: Niecka Grodziecka copper deposit-reserves specification
Type of re-
sources
Exploration
stage
Supposed economic resources Not supposed economic resources
Ore
[thousand Mg]
Cu
[thousand Mg]
Ore
[thousand Mg]
Cu
[thousand Mg]
Out of pillars A+B 513 7.5 - -
C
1
6856 108.3 2293 31.2
C
2
28,244 348 927 10.8
subtotal 35,613 463.8 3220 42.0
In protection
pillars
A+B 1,740 29.2 - -
C
1
8223 155.4 7 0.1
C
2
809 17.1 - -
Subtotal 10,772 201.8 7 0.1
Total reserves 46,385 665.6 3227 42.1
Table 3: Copper resources in Wartowice deposit to the depth of 1250 m
Type of resources
Exploration
stage
supposed economic resources
notsupposed economic re-
source
Ore
[thousand Mg]
Cu
[thousand Mg]
Ore
[thousand Mg]
Cu
[thousand Mg]
Out of pillars C
1
20,153 271 7965 88
In protection pillars C
1
17,650 327 1396 15
Total C
1
37,803 598 9361 103
Table 4: Level of probable resources of copper ore on the depth 1250-2000 m (cat. E)
(Rydzewski et al., 1996)
Item Calculative field
Interval of
depth
[m]
Field
area
[km
2
]
Thickness of
copper
bearing
inteval
[m]
Average
content of
Cu
e
[%]
Perspective resources
Ore
[million Mg]
Cu
[million Mg]
1
Foresudetic
Monocline area
III
(15 boreholes)
1250-1500
1500-2000
50
400
5.0 0.64 5600 35
2
Foresudetic
Monocline - area
IV
(Borzecin
Sulmierzyce)
(27 boreholes)
1250-1500
1500-2000
350
2030
3.08 0.55 170,000 86
Kudeko, Pyra, Sobociski
Proceedings of Copper 2010 2812
3 Economic premises for Wartowice Niecka Grodziecka
deposits management
Using the results of Analyze of grounds for Niecka Grodziecka deposits (Konrad, Wartowice)
management study, the analyze of cost-effectiveness of Wartowice Niecka and predicted macro
economic conditions.
Essential parameters decisive in determining the investment project profitability in poly-metallic
ores mining and processing industry are as follows:
deposit resources,
basic and accompanying elements content,
annual production figures,
time period of investment phase,
investment costs of deposit development and processing plant construction
costs of mining and processing,
prices and possibilities of produced metals marketing.
Among the above parameters, characterizing the project, the most difficult to estimate, especially in
longer time perspective, are the macroeconomic values such as copper price, silver price and ex-
change rates. Therefore predicting the values of those parameters the trends in last years were taken
into consideration (Figure 1 and 2).
Carrying the economic analyze in order to define the value of profit from extraction and processing
the ores from Wartowice Niecka Grodziecka deposits for the predicted schedule of production
based on reserves, the twenty years period of mine and concentrator operation was assumed.
The calculations were carried out using the method of drawn calculus for the technological circuit,
at the assumption, that the Company (mine, concentrator and metallurgical plant) final product is a
electrolytic copper and metallic silver.
Calculation of profit from selling the produced electrolytic copper and silver obtained as a result of
ore extraction and processing from Wartowice Niecka Grodziecka deposits consisted in defining
the following values:
costs of extraction,
costs of flotation processing,
costs of metallurgical processing Cu,
costs of metallurgical processing Ag,
investment cost for mine and concentrator construction,
depreciation of the above investments.
Perspectives of Copper Mining Industry Development in Poland
Proceedings of Copper 2010 2813
The sum of calculated costs gave the production costs.
The production income is composed of:
income from electrolytic Cu sale calculated as product of copper amount and price of 1 Mg of Cu
in USD and American dollar exchange rate,
income metallic Ag sale calculated as product of silver and price of 1 kg of Ag in USD and
American dollar exchange rate.
Gross profit was computed as a difference between the income and costs of production.
The following values describing the deposit and investment project were used in calculations [7]:
reserves of deposits 60,000,000 Mg,
content of Cu in the ore 1.23 %,
content of Ag in the ore 46.0 g/Mg,
production 3,000,000 Mg/year,
period of investment 11 years,
discounted investment costs 1100 million USD,
unit costs of extraction 33 USD/Mg,
unit costs of flotation processing 10 USD/Mg,
unit costs of metallurgical processing Cu 900 USD/Mg Cu,
unit costs of metallurgical processing Ag 13.7 USD/kg Ag.
Additionally the following values of yields obtained in flotation and metallurgical processing were
assumed:
flotation yield of Cu 90 %,
metallurgical yield of Cu 97 %,
flotation yield of Ag 86 %,
metallurgical yield of Ag 95 %.
The following values of macroeconomic parameters were used:
electrolytic copper price 6000 USD/Mg,
metallic silver price 386 USD/Mg (12 USD/troz).
Kudeko, Pyra, Sobociski
Proceedings of Copper 2010 2814
Figure 1: Average prices of electrolytic copper on LME in years 1996-2008
Figure 2: Average prices of metallic silver on LBM in years 1996-2008
Using those parameters the gross profit for example year of mine operation was calculated. More-
over the sensitivity analysis of annual gross profit value and cumulated gross profit for the whole
operation period of project on the copper price (Figure 3 and 4).
Perspectives of Copper Mining Industry Development in Poland
Proceedings of Copper 2010 2815
Figure 3: Sensitivity of annual gross profit on copper price change (Wartowice-Niecka
Grodziecka project)
Figure 4: Sensitivity of cumulated gross profit on price change of electrolytic copper (Wartowice-
Niecka Grodziecka project)
Analyzing the presented economic figures calculated for Wartowice Niecka Grodziecka deposit
management project it is evident that profit from flowing production will be maintained only if the
average copper price within the next three decades is not lower than 6000 USD per Mg.
Kudeko, Pyra, Sobociski
Proceedings of Copper 2010 2816
Thus the perspectives of Polish copper industry development at present conditions may be are prom-
ising only for KGHM Polska Miedz S. A.
4 Conditions for copper industry development in Poland in
KGHM Polska Miedz S. A.
KGHM Polska Miedz S. A. it is a company having almost 50 years long tradition, applying the most
modern methods of extraction and processing of copper ore, being the world leader of high quality
copper and silver as well as other raw materials. Copper ore is extracted from the Europe biggest
and one of the world biggest copper deposits, which is located between Glogow, Lubin and
Sieroszowice. The deposit has the area of about 550 km
2
. The company extracts the copper ore in
three mines: Lubin, Polkowice - Sieroszowice and Rudna. Annual output is nearly 30 million Mg of
ore while copper production exceeds 500,000 Mg of electrolytic copper and over 1200 Mg of metal-
lic silver due to processing the ore from own resources and purchased concentrates. Moreover small
amounts of nickel (nickel sulfate), crude lead, metallic selenium, and platinum-palladium sludge are
produced. In recent years increased also the metallic gold production what was the result of process-
ing the purchased concentrates with higher, than in domestic concentrates, Au content.
Granted in 2004 by Minister of Environment the concession for mining the copper ore from
Glogow Gleboki Przemyslowy (GG-P) deposit, what increased the reserves by 26 %. KGHM Pol-
ska Miedz S. A. currently extracts 6 deposits: Lubin-Malomice Polkowice, Sieroszowice",
Radwanice Wschod, Rudna and Glogow Gleboki Przemyslowy.
Production figures of the Company in years 1996-2007 are shown on drawings: 5 output extrac-
tion, 6 concentrate production, 7 electrolytic copper production, 8 production of metallic sil-
ver. Additionally changes of copper content in the ore and copper in concentrate changes are shown
in Figures 9 and 10.
Perspectives of Copper Mining Industry Development in Poland
Proceedings of Copper 2010 2817
Figure 5: Output in KGHM Polska Miedz S. A. mines
Figure 6: Concentrate production in KGHM Polska Miedz S. A.
Kudeko, Pyra, Sobociski
Proceedings of Copper 2010 2818
Figure 7: Production of electrolytic copper in KGHM Polska Miedz S. A.
Figure 8: Production of metallic silver in KGHM Polska Miedz S. A.
The presented results show that the production in KGHM mines since 2002 had been increasing
gradually up to the maximum level reached in 2006. Production of copper during last years in in-
creased first of all due to enlarged purchase of concentrates, while silver production since 2003 has
dropped as a result of lower metals (both copper and silver) content in own resources, what was also
noticed in copper content reduction in concentrate.
Perspectives of Copper Mining Industry Development in Poland
Proceedings of Copper 2010 2819
Figure 9: Copper content in output mined in KGHM Polska Miedz S. A. mines
Figure 10: Copper content in KGHM Polska Miedz S. A. concentrate
Carrying the economic analyze in order to define the value of profit from extraction and processing
the ores from GG-P deposit for the predicted schedule of production based on reserves, the period
from 2008 to 2040 of mine and concentrator operation was assumed. The calculations were carried
out using the method of drawn calculus for the technological circuit, at the assumption, that the
Company (mine, concentrator and metallurgical plant) final product is a electrolytic copper and me-
tallic silver.
In this analyze the following values describing the deposit and investment project were used [6]:
Kudeko, Pyra, Sobociski
Proceedings of Copper 2010 2820
deposits reserves 204,000,000 Mg,
content of Cu in the ore 2.08 %,
content of Ag in the ore 64.7 g Mg,
annual average mining production 6,000,000 Mg/year,
investment period (till getting the average mining level) 15 years,
investment cost of region development and preparing for extraction 660 million USD,
unit costs of extraction 50 USD/Mg,
unit costs of flotation processing 10 USD/Mg,
unit costs of metallurgical processing Cu 870 USD/Mg Cu,
unit costs of metallurgical processing Ag 13.7 USD/kg Ag.
Additionally the following values of yields obtained in flotation and metallurgical processing were
assumed:
flotation yield of Cu 90 %,
metallurgical yield Cu 97 %,
flotation yield of Ag 86 %,
metallurgical yield of Ag 95 %.
The following values of macroeconomic parameters were used:
electrolytic copper price 4500 USD/Mg,
metallic silver price 386 USD/Mg (12 USD/troz).
Using those parameters the gross profit for example year of mine operation from GG-P was calcu-
lated. Moreover the sensitivity analysis of annual gross profit value and cumulated gross profit for
the whole operation period of project on the copper price (Figures 11 and 12).
Perspectives of Copper Mining Industry Development in Poland
Proceedings of Copper 2010 2821
Figure 11: Sensitivity of annual gross profit on electrolytic copper price for production in GG-P
mine
Figure 12: Sensitivity of cumulated gross profit value on electrolytic copper price for production in
GG-P mine
5 Prognosis of production value and KGHM Polska
Miedz S. A. efficiency
Basing on production program elaborated for Appendixes to Copper Deposit Management Design
of Lubin-Malomice, Polkowice II, Sieroszowice I, Rudna I, Rudna II, Radwanice
Wschod and Glogow Gleboki-Przemyslowy mines, the prognosis of extraction level as well as
Kudeko, Pyra, Sobociski
Proceedings of Copper 2010 2822
copper and silver production from own reserves for years 2008-2040 was made. Studying the cur-
rent trends in extraction and processing cost changes and predicting the values of macroeconomic
parameters during next decades, the economic analyze was made with defining the predicted net
profit from selling the copper and silver from own reserves.
Forecast of production volume, of production costs and of net profit was made using in calculation
the following values:
unit costs of extraction 47.7 USD/Mg,
unit costs of flotation processing 10 USD/Mg,
unit costs of metallurgical processing Cu 870 USD/Mg Cu,
unit costs of metallurgical processing Ag 13.7 USD/kg Ag,
Cu flotation yield 90 %,
Cu metallurgical yield 97 %,
Ag flotation yield 86 %,
Ag metallurgical yield 95 %.
The following macroeconomic parameters were used:
electrolytic copper price:
2009 6500 USD/Mg,
2010 6000 USD/Mg,
2011-2040 4500 USD/Mg,
metallic silver price:
2009 14 USD/troz,
2010 13 USD/troz,
2011-2040 12 USD/troz.
Perspectives of Copper Mining Industry Development in Poland
Proceedings of Copper 2010 2823
Figure 13: Copper production costs vs. copper prices
In Figure 13 the comparison of copper price changes and production costs of electrolytic copper in
years 1996-2008 is presented. The charts show the substantial increase of production costs since
2004 when metal prices started to substantially increase.
Figure 14: Predicted mining production in KGHM mines
Kudeko, Pyra, Sobociski
Proceedings of Copper 2010 2824
Figure 15: Predicted production of electrolytic Cu from own KGHM reserves
In Figures 14, 15 and 16 the predicted volume of extraction in KGHM mines and predicted Cu and
Ag production from own reserves are presented. Extraction level should be stable until 2030 (28-
30 million Mg), only in last decade the drop of production to the level of 20-25 million Mg is ex-
pected. Prognosis of copper and silver production during the next three years is on the level of
400 thousand Mg of Cu and 1000 Mg of Ag due to carrying out the mining operations in areas with
lower metals content in the deposit. Starting the extraction in Glogow Gleboki region will let to
increase Cu production and especially Ag production due to better ore parameters in this area (aver-
age Cu content of 2.08 % and Ag content of 65 g/Mg). In Figure 17 the predicted net profit is pre-
sented, which at the copper price of 4500 USD/Mg Cu is on the level of 160-210 million USD for
the year.
In Figure 18 the predicted unit cost of electrolytic copper production taking into consideration the
revenue from silver sale, is shown.
Perspectives of Copper Mining Industry Development in Poland
Proceedings of Copper 2010 2825
Figure 16: Predicted production of metallic Ag from own KGHM resources
Figure 17: Predicted net profit from electrolytic Cu production from own KGHM resources
Kudeko, Pyra, Sobociski
Proceedings of Copper 2010 2826
Figure 18: Predicted unit costs of electrolytic copper production from own KGHM reserves
6 Conclusions
1. Among currently prospected resources of copper deposits only the deposits extracted KGHM
Polska Miedz S. A. have the economic importance and only the development of those deposits
extraction (Glogow Gleboki Przemyslowy) and if necessary development of Gaworzyce
and Radwanice Zachod deposits will be cost-effective.
2. Mine construction and mining of Wartowice Niecka Grodziecka deposit will be profitable if
copper price, during next three decades, is not lower than 6000 USD/Mg.
3. Decrease of mineralization in copper deposit extracted by KGHM Polska Miedz S. A. in current
five-year period does not have the substantial impact on the Company economic results.
4. Development and mining of Glogow Gleboki Przemyslowy (GG-P) deposit allow to main-
tain the production on the current level during next two decades.
Perspectives of Copper Mining Industry Development in Poland
Proceedings of Copper 2010 2827
References
[1] KGHM Polska Miedz S.A.: Annual Report 1998, Lubin.
[2] KGHM Polska Miedz S.A.: Annual Report 2005, Lubin.
[3] KGHM Polska Miedz S.A.: Annual Report 2006, Lubin.
[4] KGHM Polska Miedz S.A.: Annual Report 2007, Lubin.
[5] KGHM Polska Miedz S.A.: Annual Report 2008, Lubin.
[6] SADECKI Z. et al., (2006): Appendix no. 1 to Copper Deposits Management Design for
Glogow Gleboki-Przemyslowy. KGHM CUPRUM sp. z o.o. CBR, Wroclaw.
[7] SOBOCINSKI J. et al., (2007): Study on reasons of Grodziec Basin deposit (Konrad, War-
towice) management. KGHM CUPRUM sp. z o.o. CBR, Wroclaw.
Proceedings of Copper 2010 2828
Proceedings of Copper 2010 2829
Control of Bubble Size in a Laboratory Flotation
Column
Miguel Maldonado, Dr. Andr Desbiens, Dr. ric Poulin Alberto Riquelme
Dr. Ren del Villar
Universit Laval Universidad de Concepcin
Department of Electrical and Computer Engineering Department of Electrical Engineering
Department of Mining, Materials and Metallurgical Engineering
LOOP (Laboratoire dobservation et doptimisation des procds)
Qubec (QC), Canada Concepcin, Chile
Keywords: Flotation column control, frit-and-sleeve sparger, Wiener model, IMC control,
bubble size control, bubble size distribution
Abstract
Gas dispersion properties have proven to be key variables of the flotation process. Among these
properties, bubble surface area flux (BSAF or S
b
) has been reported to linearly correlate with the
flotation rate constant; therefore, it is a potential manipulated variable to achieve a desired metallur-
gical performance. BSAF can be represented as a combination of two other gas dispersion proper-
ties: the superficial gas velocity and the Sauter bubble mean diameter; thus, controlling S
b
implies
controlling bubble size and superficial gas velocity. This work focuses on the control of the Sauter
mean bubble diameter. To improve BSAF controllability, a so-called frit-and-sleeve sparger device
was implemented in a laboratory flotation column to regulate bubble size independently from super-
ficial gas velocity. For control purposes, the Sauter bubble mean diameter was indirectly calculated
from the bubble size distribution, estimated by using a Gaussian mixture model. An IMC controller
based on a Wiener model, was implemented in the laboratory flotation column set-up. Tracking
performance and rejection of gas velocity and unmeasured frother concentration variations were
then evaluated.
1 Introduction
Flotation is a commonly used method for separating valuable minerals (metal containing) from use-
less mineral (gangue). Its performance is determined by the valuable-mineral concentrate grade and
recovery. Whereas the first of these two variables can be measured on-line using an X-ray on-stream
analyzer (OSA), the latter must be estimated from steady-state material balance, which strongly lim-
Maldonado, Desbiens, Poulin, del Villar, Riquelme
Proceedings of Copper 2010 2830
its its use for regulatory control purposes. Moreover, the long sampling times of these OSA devices,
usually multiplexed, favour the use of a hierarchical control where secondary variables are con-
trolled to reject the frequent disturbances occurring in this type of process.
Recent studies [1] have shown that the performance of a flotation device basically depends on three
factors: the particle floatability, the froth recovery and the bubble surface area flux, a combination
of two other gas-dispersion properties, superficial gas velocity and bubble size [2]. This finding
suggests that for a well conditioned pulp, a given metallurgical performance can be achieved by
modifying froth recovery and bubble surface area flux. Consequently, two control approaches have
been proposed. The first focuses on controlling some froth characteristics, such as froth speed,
which have been widely applied [3-5], or froth appearance [6]. The second approach aims at con-
trolling some gas dispersion properties in the collection zone, such as gas hold-up and superficial
gas velocity [7-10]. This approach has been motivated by the recent availability of industrial gas-
dispersion sensors [11, 12]. This article extends previous works on control of some gas dispersion
properties, to the control of bubble size, represented by the Sauter mean bubble diameter, a key step
towards controlling the bubble surface area flux.
Bubble surface area flux can be mathematically expressed as follows:
32
6
g
b
J
S
d
= (1)
where J
g
is the superficial gas velocity and d
32
is the Sauter mean bubble diameter. Since there is no
uncertainty associated to Equation (1), controlling BSAF implies controlling superficial gas velocity
and the Sauter mean bubble diameter.
In flotation column operation, bubble size is affected by frother type and concentration, gas rate and
sparging system. Frother dosage regulation is usually implemented using a ratio feed-forward con-
trol based on the actual tonnage processed. Nevertheless, because of the limitations of feed-forward
control in the presence of modelling uncertainties and unknown disturbances, such as frother persis-
tency, evaporation rate and most importantly, the effect of reprocessed water (still containing some
residual frother), frother concentration in a given flotation machine is hard to assess. Consequently,
and despite the fact that the authors have already proposed a method to evaluate on-line the frother
concentration [13], in this work this variable is considered as an unmeasured disturbance. Since
superficial gas velocity also modifies bubble size [2, 14], it influences BSFA directly through the
numerator of Equation (1) and indirectly by affecting the bubble size (denominator). Whenever fea-
sible, the sparger system adds another control degree of freedom to modify bubble size in flotation
columns. To take advantage of this fact, in this work a frit-and-sleeve sparger [15] was imple-
mented, which allows the modification of the bubble size independently from gas velocity, thus
improving BSAF controllability.
Control of Bubble Size in a Laboratory Flotation Column
Proceedings of Copper 2010 2831
To obtain the benefits of feedback control, bubble size must be measured on-line. For diagnosis
purposes, Sauter mean bubble diameter is usually calculated after a set of collected images have
been processed by using the following equation:
3
1
32
2
1
N
i
i
N
i
i
d
d
d
=
=
=
(2)
where d
i
is the equivalent diameter of bubble i and N is the number of counted bubbles. This calcu-
lation assumes a steady state condition, i.e. the data points (bubble sizes in this case) are sampled
from a steady bubble size density function, and as such, is not suitable for control purposes. In this
work an on-line estimation of the Sauter mean bubble diameter based on the bubble size density
estimation was implemented.
2 Flotation column set-up
The flotation column used in this work is composed of three sections made of polycarbonate tubes
for a total height of 5 m; the internal diameter of the bottom, intermediate and upper sections are
respectively 15.24 cm, 10.16 cm, and 5.08 cm. A frit-and-sleeve sparger is mounted in the bottom
part of the column as shown in Figure 1. Gas flow rate is measured through a mass flow sen-
sor/controller (Aalborg model GFC17), which also provides an estimate of the volumetric flow
based on a reference condition (21.1
o
C and 101.3 kPa). Therefore, its readings must be converted to
the actual tests conditions of temperature and pressure, measured by the sensors shown in Figure 1
using the following equation:
2
1033.23 ( ) 273.15
1033.23 ( ) 294.16
o
ref
g g
a
T C
J J
P cmH O
( ( +
=
( (
+
(3)
where P
a
is the absolute pressure measured in cm H
2
O, T is the temperature in degrees Celsius and
J
g
ref
is the gas velocity calculated from the air mass flow meter readings at reference conditions. A
differential pressure transmitter DP (model ABB 264DS) was tapped between 250 cm and 320 cm
above the sparger to measure gas hold-up. For a two-phase system (air-water), this latter can be
measured using the following relationship:
(%) 100
g
P
L
= (4)
where P is the pressure differential in cm H
2
O and L is the distance between taps, here 70 cm.
Data acquisition was performed by a HMI/SCADA software (iFIX
operating system. An Opto 22 I/O system was used to centralize sensor and actuator signals. A
Maldonado, Desbiens, Poulin, del Villar, Riquelme
Proceedings of Copper 2010 2832
modified version of the McGills bubble viewer [11] was implemented to measure bubble size. The
image processing system was performed by a dedicated computer using MatLabs Image Acquisi-
tion toolbox
and Image J free license software which was used to process the collected images to
obtain bubble sizes. A communication link between MatLab applications running on the control
station and the image processing computer was implemented using MatLab's Instrumentation tool-
box
(6)
1
1, 0 1, 1, ,
M
j j
j
w w j M
=
= =
(7)
( ) 1, 1, ,
j
x dx j M
+
= =
(8)
where M is the model order, i.e. the number of components considered. The coefficients w
j
are
called mixing parameters and represent the prior probability of a data point having been generated
from component j of the mixture model.
The component densities could be any density function or even a combination of density functions
with different functional forms. Nevertheless, depending on their properties and structure, the pa-
rameter identification might become very difficult. An interesting case is obtained when the compo-
nents
j
(x) are chosen to be fixed Gaussian functions.
( )
2
2
1 1
( ) exp
2 2
j
j
x
x
=
`
)
(9)
where
j
, j =1,,M , while M and are fixed parameters.
Minimizing the log-likelihood of the samples and applying a stochastic gradient descent approach
the following recursive equation is obtained [17]:
( 1) ( ) ( / ( 1)) ( )
j j j
w k w k P j x k w k ( + = + +
(10)
When is selected to be constant, it introduces an exponentially decaying envelop (1- )
k-i
applied to the influence of the data point x(k-i) [18]. P(j|x(k)) is the posterior probability determined
using Bayes theorem as follows:
( ( ))
( | ( ))
( ( ))
j j
x k w
P j x k
f x k
= (11)
The Sauter mean bubble diameter can be estimated on-line from the Gaussian mixture parameters as
follows [19]:
( )
( )
3 2 2
1
32
2 2
2
1
, ( ) ( 3 )
( )
( )( )
,
M
j j j
j
M
j j
j
x f x k dx w k
d k
w k
x f x k dx
+
=
+
=
+
= =
+
(12)
Control of Bubble Size in a Laboratory Flotation Column
Proceedings of Copper 2010 2835
Figure 3 shows an example of bubble size density estimation using the previously described
algorithm.
Figure 3: Example of a bubble size density estimation using a mixture of thirty equally-spaced
fixed Gaussian distributions ( = 0.005).
5 The Wiener model
A Wiener model is used to represent the dynamic relationship between superficial water velocity
and Sauter mean diameter. Wiener models consist of a linear system in series with a memory-less
(static) nonlinear element, as shown in Figure 4. When this model is used for control purposes, the
nonlinear element must be selected in such a way that the nonlinearity function is invertible.
Figure 4: Wiener model structure.
The Wiener model was identified in two steps; first, the static nonlinear element was determined by
using steady-state information, then, the linear dynamic element was identified by using step re-
sponse information. The technique used in either case is described hereafter.
Maldonado, Desbiens, Poulin, del Villar, Riquelme
Proceedings of Copper 2010 2836
5.1 Static nonlinear element
To explore the static relationship between superficial water velocity in the sparger and Sauter mean
bubble diameter, several experimental tests were conducted in the laboratory flotation column. Fig-
ure 5 contains three subplots, each for a given superficial gas velocity, i.e. 0.5, 0.8 and 1.1 cm/s. For
a given superficial gas velocity, each subplot shows a monotone decreasing steady state relationship
between superficial water velocity in the sparger and the Sauter mean bubble diameter for three dif-
ferent frother concentrations. In general, it can be observed that the effect of frother concentration is
to shift this decreasing relationship without significantly modifying its shape.
Figure 5: Static relationship between Sauter mean diameter and the superficial water velocity in
the sparger for different frother concentration and superficial gas velocities.
Figure 6 shows the same pattern; in this case, it can be observed that superficial gas velocity acts by
shifting the nonlinear relationship between superficial water velocity and Sauter mean diameter.
Figure 6: Static relationship between Sauter mean diameter and the superficial water velocity
through the sparger for different frother concentration and superficial gas velocities.
Control of Bubble Size in a Laboratory Flotation Column
Proceedings of Copper 2010 2837
Figure 7 shows the static gain (d
32
/J
ls
) corresponding to the nonlinear relationships shown in
Figures 5 and 6 as a function of the superficial water velocity in the sparger. The selected nominal
gain shown in Figure 7 corresponds to the following nonlinear relationship between Sauter mean
diameter and superficial water velocity:
0.256
32
3.706 0.226
ls
d J
= (13)
where
32
d and
ls
J are respectively the steady-state Sauter mean diameter and superficial water
velocity.
Figure 7: Static gain of the system vs superficial water velocity in the sparger.
5.2 Linear dynamic element
A simple unitary gain, first-order lag system with time delay was used to represent the dynamic be-
haviour between superficial water velocity in the sparger and the Sauter mean diameter. Notice that
the static gain is captured by the nonlinear static block. The following nominal model was identified
using step response information:
90
( )
90 1
s
e
G s
s
=
+
(14)
Figure 8 shows the Wiener intermediate variable v(t) (see Figure 4), in this case, a filtered version of
the superficial water velocity obtained by inverting the nonlinearity function (Equation 13) and the
Maldonado, Desbiens, Poulin, del Villar, Riquelme
Proceedings of Copper 2010 2838
simulated response of the identified linear system (Equation 14). It can be seen that the Wiener
model is able to capture reasonably well the nonlinear dynamic behaviour of the process.
Figure 8: Wiener model validation.
6 Wiener model based control
Assuming there is no uncertainty in the non-linear block, the static non-linearity can then be com-
pletely removed by performing the inverse of the non-linearity (see Figure 9). Consequently, the
controller synthesis can be performed by the classical internal model control (IMC) framework, i.e,
by minimizing the H
2
norm of the error e for a given input v:
2
( ) ( ) 2
min min 1 ( ) ( ) ( )
Q s Q s
e G s Q s v s ( =
(15)
For a step input, i.e. v(s) = 1/s and a stable system factored into
( ) ( ) ( )
M A
G s G s G s =
where
( )
A
G s
is an
all-pass transfer function, containing right half plane zeros and delays, and
( )
M
G s
a minimum phase
transfer function, the optimal H
2
controller is:
1
( ) ( )
M
Q s G s
=
(16)
Then, the controller is augmented by a filter F(s):
( ) ( ) ( ) Q s Q s F s =
(17)
where the filter F(s) is usually selected to be:
( )
1
( )
1
r
F s
s
=
+
(18)
0 2000 4000 6000 8000 10000 12000 14000 16000 18000
0
50
100
150
200
250
300
350
time (s)
i
n
t
e
r
m
e
d
i
a
t
e
v
a
r
i
a
b
l
e
,
v
(
t
)
v(t)
estimated
v(t)
Control of Bubble Size in a Laboratory Flotation Column
Proceedings of Copper 2010 2839
and r is selected to obtain a realizable controller. The complementary sensitivity function is given
by:
( , ) ( )
( , )
1 ( , ) ( ) ( )
Q s G s
T s
Q s G s G s
=
( +
(19)
The maximum peak (Mp) tuning [20] is used to find the value of IMC filter for parametric uncer-
tainty in the process model. The objective is to find the smallest IMC filter time-constant that as-
sures that no closed-loop frequency response (complimentary sensitivity function) will have more
than specified Mp (i.e., the specified maximum peak).
Considering a 20 % uncertainty in time delay and time constant and a 5 % uncertainty in the static
gain of the nominal process model (Equation 14), the following IMC controller was obtained:
90 1
( )
82.6 1
s
Q s
s
+
=
+
(20)
For implementation purposes, this controller was discretized with a sampling time of 5 seconds.
Figure 9: IMC based Wiener model.
7 Experimental results
To evaluate performance of the control system, the laboratory flotation column was first filled with
a solution of 10 ppm frother concentration Dowfroth 250. Then, changes on superficial gas velocity
and frother concentration were implemented to simulate disturbances affecting the bubble size con-
trol loop.
Maldonado, Desbiens, Poulin, del Villar, Riquelme
Proceedings of Copper 2010 2840
Figure 10 shows the controlled variable, i.e, Sauter mean diameter, and the manipulated variable,
the superficial water velocity in the sparger. A good tracking performance can be observed when the
pump speed is not in a saturated condition. Approximately at 12,000 s, a gas velocity step change
from 0.5 to 0.8 cm/s was implemented. This originated an increase in bubble size, which was rap-
idly compensated by increasing superficial water velocity in the sparger. Later on, at about 13,000 s,
the bubble size set-point was decreased to 0.6 mm, but this new set-point was not achievable since
the pump speed saturated (at J
ls
= 340 cm/s). Finally, frother concentration was changed from
10 ppm to 20 ppm at around 16,000 s. It can be seen that the bubble size was consequently affected
(reduced), but the control system immediately reacted by decreasing the superficial water velocity,
which allowed the bubble size to return to its previous value (set-point).
Figure 11 shows the time evolution of both gas hold-up and bubble surface area flux for test condi-
tions shown in Figure 10. A significant correlation can be seen between gas hold-up and bubble
surface flux, initially suggesting that both variables carry similar information and consequently ei-
ther variable could be used for control purposes.
Figure 10: IMC based Wiener model: Tracking and regulation performance.
Control of Bubble Size in a Laboratory Flotation Column
Proceedings of Copper 2010 2841
Figure 11: IMC based Wiener model: relation between Sauter mean diameter, gas hold-up and
BSAF.
8 Discussion
It has been shown that for controlling BSAF, reference values for superficial gas velocity and Sauter
mean diameter must be provided. Different combinations of gas velocity and Sauter mean diameter
could generate the same BSAF, thus an optimal combination of these variables producing a desired
BSAF must be determined based on the desired metallurgical performance. Something similar oc-
curs with gas hold-up, since different conditions of gas velocity, frother concentration and bubble
size can generate the same gas hold-up value. As it is widely accepted that bubble size plays a key
role on flotation performance, one possible control approach is to regulate bubble size by modifying
the superficial water velocity in the sparger and to regulate bubble surface area flux (or gas hold-up)
by manipulating gas velocity.
One potential problem with controlling bubble size via Sauter mean diameter is that all the available
information regarding the shape of the bubble size distribution such as multi-modal and tailing
Maldonado, Desbiens, Poulin, del Villar, Riquelme
Proceedings of Copper 2010 2842
behaviour is completely lost in this averaging exercise. In fact, it is possible to generate very differ-
ent bubble size distributions having the same Sauter mean diameter. Therefore, using this latter for
control purposes will be appropriate only for unimodal narrow bubble size distributions.
In this study, frother concentration was considered to be an unknown disturbance. However, since it
has a strong effect on bubble size, its use as a manipulated variable deserves to be explored. More-
over, for mechanical cells where no spargers are used, it becomes a more relevant manipulated vari-
able to modify bubble size.
9 Conclusions
This work explored the control of Sauter mean diameter as a first step towards the control of bubble
surface area flux. A frit-and-sleeve sparger was implemented to allow the modification of the bub-
ble size by injecting water through a gap.
The Sauter mean diameter was calculated on-line from the measured (image processing) bubble size
distribution. A Gaussian mixture model was proposed and its parameters were determined by mini-
mizing the log-likelihood of the data points and using a gradient descent method.
A Wiener model-based IMC was proposed to control the Sauter mean diameter by modifying the
superficial water velocity through the frit-and-sleeve sparger. Good tracking performance (bubble
size set-point) and rejection of unmeasured changes in superficial gas velocity and frother concen-
tration were obtained in a laboratory flotation column.
Acknowledgements
This work was conducted under the project NSERC (National Science and Engineering Research
Council of Canada) with the following industrial partners: COREM, Xstrata-Ni Strathcona Mill,
Agnico-Eagle Mines Division Laronde.
Authors would like to acknowledge the McGill mineral processing group for providing a frit-and-
sleeve sparger prototype. M. Maldonado would also like to thank the Chilean Council for Science
and Technology (Conicyt) for partially financing his Ph.D. studies at Universit Laval.
Control of Bubble Size in a Laboratory Flotation Column
Proceedings of Copper 2010 2843
References
[1] GORAIN, B.K., FRANZIDIS, J-P., MANLAPIG, E.V. (1997): Studies on impeller type,
impeller speed and air flow rate in an industrial scale flotation cell Part4: Effect of bubble
surface area flux on flotation performance. Minerals Engineering, Vol.10, No.4, pp.367-379.
[2] FINCH, J.A. AND DOBBY, G.S. (1990): Column Flotation, Pergamon Press, Oxford (UK).
[3] BARRIA, A., VALDEBENITO, M. (2008): Implementation of rougher flotation control sys-
tem at Codelco Chile, Andina Division. Procemin 2008, 5
th
International Mineral Processing
Seminar, Santiago, Chile, pp.215220.
[4] CORTES, G., VERDUGO, M., FUENZALIDA, R., CERDA, J., CUBILLOS, E. (2008):
Rougher flotation multivariable predictive control; Concentrator A-1 division Codelco Norte.
Procemin 2008, 5
th
International Mineral Processing Seminar, Santiago, Chile, pp. 315325.
[5] MOILANEN, J., REMES, A. (2008): Control of the flotation process. Procemin 2008, 5
th
In-
ternational Mineral Processing Seminar, Santiago, Chile, pp. 305313.
[6] LIU, J.J., MACGREGOR, J.F. (2008): Froth-based modeling and control of flotation proc-
esses. Minerals Engineering, Vol.21, pp.642651.
[7] BERGH, L.G. and YIANATOS, J.B. (1993): Control alternatives for flotation columns. Min-
erals Engineering, Vol. 6, No. 6, pp. 631-642.
[8] CARVALHO, T. and DURO, F. (2002): Control of a flotation column using fuzzy logic
inference. Fuzzy Sets and Systems, Vol.125, pp.121-133.
[9] PERSECHINI, M.A.M, JOTA, F.G., PERES, A.E.C. and GONALVES, F (2004): Control
strategy for a column flotation process. Control Engineering Practice, Vol.12, pp.963-976.
[10] MALDONADO, M., DESBIENS, A. and DEL VILLAR, R. (2009): Potential use of model
predictive control in optimizing the column flotation process. International Journal of Mineral
Processing, Vol.93, pp.26-33.
[11] GOMEZ, C.O. and FINCH, J.A. (2007): Gas dispersion measurements in flotation cells. In-
ternational Journal of Mineral Processing, Vol.84, pp.51-58.
[12] OKEEFE, C., VIEGA, J., and FERNALD, M. (2007): Application of passive sonar technolo-
gy to mineral processing and oil sands applications. Proceedings of the 39
th
Annual Meeting
of the Canadian Mineral Processors, CIM, Ottawa (Canada), pp.429-457.
[13] MALDONADO, M., DESBIENS, A, DEL VILLAR, R. and AGUILERA, R. (2009): On-line
estimation of frother concentration in flotation processes. 48
th
Annual Conference of
Metallurgists, Sudbury, Canada, pp.35-146.
Maldonado, Desbiens, Poulin, del Villar, Riquelme
Proceedings of Copper 2010 2844
[14] NESSET, J.E., HERNANDEZ-AGUILAR, J., ACUA, C., GOMEZ, C.O. and FINCH, J.A.
(2006): Some gas dispersion characteristics of mechanical flotation machines. Minerals Engi-
neering, Vol.19, pp.807-815.
[15] KRACHT, W., GOMEZ, C.O and FINCH, J.A. (2008): Controlling bubble size using a frit-
and-sleeve sparger. Minerals Engineering, Vol.21, pp.660-663.
[16] BISHOP, C.M. (1996): Neural network for pattern recognition, Oxford University Press, 1st
edition.
[17] TRVN, H. (1991): A neural network approach to statistical pattern classification by semi-
parametric estimation of probability density functions. IEEE Transactions on Neural Net-
works, Vol.2, No.3.
[18] ZIVKOVIC, Z. and VAN DER HEIJDEN, F. (2004): Recursive unsupervised learning of fini-
te mixture models. IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol.26,
No.5, pp.651-656.
[19] MALDONADO, M., DESBIENS, A., DEL VILLAR, R., GIRGIN, E. and GOMEZ, C.
(2008): On-line estimation of bubble size distributions using Gaussian mixture models.
Procemin 2008, 5
th
International Mineral Processing Seminar, Santiago, Chile, pp.389-398.
[20] STRYCZEK, K, LAISECA, M, BROSILOW, C, LEITMAN, M. (2000): Tuning and design
of single input, single output control systems for parametric uncertainty. AIChE Journal, Vol.
46, No.8, pp.1616-1631.
Proceedings of Copper 2010 2845
Flow Process in the Aerator of the
Flotation Machine Preliminary Simulations
Adam Maka, Adam Fic, Andrzej Sachajdak, Ireneusz Szczygie
Silesian University of Technology
Institute of Non-ferrous Metals, Sowiskiego 5
Institute of Thermal Technology, Konarskiego 22
44-100 Gliwice, Poland
Keywords: Flotation, technology, computation, CFD modeling
Abstract
In this paper, the numerical aspect of the flow phenomena in the flotation machine is presented. The
efficiency of the flotation process strongly depends on the fluid flow in the main part of the ma-
chine: in the aerator. This element is responsible for mixing suspended solids and dispersing air.
The paper presents preliminary results of mathematical modeling. The process of flotation consists
of a number of phenomena which provide serious numerical difficulties. One can enumerate rota-
tion, two phase flow, foam formation etc. To the knowledge of authors there is no complete numeri-
cal model available for the flotation machine. The long term task of the project is to create a com-
plete model of the machine. Such a model would be very helpful in the process of constructing and
modernization of the flotation machine. As it was mentioned, due to the difficulties connected with
flotation phenomena modeling, only some of them were taken under consideration. First step, de-
scribed in this paper, shows how to handle one phase flow in the aerator. The commercial package
Ansys Fluent was utilized for the analysis. The results were compared with the measurements per-
formed on the real machine. Obtained results are satisfying and encouraging for further develop-
ment.
1 Introduction
Flotation is the industrial process of particle separation by dissolving air in the pulp under pressure
and then releasing the air at atmospheric pressure into a flotation tank. The released air forms tiny
bubbles which adhere to the suspended matter causing it to float to the surface of the water where it
may then be removed by a skimming device.
Maka, Fic, Sachajdak, Szczygie
Proceedings of Copper 2010 2846
The process is realized in the flotation machine. An example is shown in Figure 1. In the machine
complicated, multiphase (gas, liquid and solid) fluid flows take place. In the near past, modeling of
such a process was impossible due to the lack of sufficient computer capacity and the processor ef-
ficiency. So, construction of flotation machines based mainly on empirical relationships and on the
experience of the constructors. Nowadays, the numerical simulations of flotation phenomena be-
came possible due to powerful enough machines as well as due to the sophisticated CFD packages
like Fluent or CFX. Recently first attempts of modeling flotation processes using CFD packages can
be noticed [1, 3, 4]. There are even some thematic conferences devoted to numerical modeling of
flotation [4]. Unfortunately, these first attempts provide only simplified models of partial phenome-
na taking place during flotation. So, there is still a lot of work to do in this field.
The flotation machine consist of tank, rotor and stator (aerator), air feed mechanism as well as pulp
delivery and discharge mechanism. Air bubbles are provided to the pulp by the aerator which con-
sists of rotor and stator. The aerator is shown in the Figure 2. Details of the aerator construction
strongly influence the quality of the final products, as well as operational parameters of the ma-
chine: air and power consumptions. The Computational Fluid Dynamics (CFD) tool describing the
behaviour of the fluids in the flotation machine, especially in the aerator surroundings would be
very helpful in the optimization process of the flotation machine. The partial model of such machine
was presented in [1].
Figure 1: Flotation machine
Flow Process in the Aerator of the Flotation Machine Preliminary Simulations
Proceedings of Copper 2010 2847
The fluid flows in the flotation machine show complex physical dynamics which have a reflection
in complicated mathematical expressions. The features of the flow are:
three dimensional flows,
turbulent flows,
multiphase flows,
unsteady state.
Figure 2: Aerator
The phenomena in the flotation machine are described by:
continuity equation,
momentum equation,
energy equation,
turbulence model,
rotation model,
material transport,
foam formation model,
phase model.
Maka, Fic, Sachajdak, Szczygie
Proceedings of Copper 2010 2848
At this stage of computation, the energy balance, multiphase flow and foam formation are omitted
what makes the system of partial differential equations simpler, but what does not mean it is simple.
With the mentioned simplifications the governing equations can be presented as follows:
continuity equation:
(1)
momentum equation:
(2)
turbulence model.
For the solution of the system of equations the commercial package Fluent was employed. Fluent
utilizes the controlled volume method to convert the governing differential equations into algebraic
ones. Unfortunately, the set of governing equations is nonlinear, so an iterative procedure of solu-
tion is necessary. Additionally, the swirling elements of the aerator are the source of serious troubles
during the creation of a numerical model. There are several possibilities of rotation consideration:
Mixing Plane Model, Multiple Reference Frame Model and Sliding Mesh Model. All of them were
taken under consideration, but as the investigations showed, not all of them can be applied. Mixing
plane model can not be applied due to the multiphase flow. The distance between the rotor and sta-
bilizer (stator) blades is low (about 30 mm), which is the source of the transient character of the
phenomena taking place in the aerator. Due to that the only model of rotations which can be used is
Sliding Mesh. The rotating and stationary parts of the aerator should be divided by a so-called
interface. The mesh in the rotating region slides over the stationary cells along this interface. The
negative effect of such an attempt realizes in the long term of computations. The blade transfer be-
tween the appropriate stator blades should be stretched to the several dozens of time steps, which by
the practical rotational speeds of the rotor makes this time step extremely short.
The presented system of governing equations has to be accomplished with the well-matched pack of
initial and boundary conditions. Improper selection of boundary conditions can result in enlarging of
computational time, or, what is worse, in loss of convergence.
The following boundary conditions were set:
wall: no slip condition at the walls of the tank
rotor: rotation 30 rad/s
free surface at the top of the tank
in the first stage of simulation no air inlet were defined
As the initial condition a motionless rotor and motionless fluid was assumed.
( ) 0 = +
t
p
Dt
D
=
w
w)
2
(
r
= 80 % the following concentration results were obtained:
60 dm
3
/h =86.2 % =7.8 % =23.1 %
80 dm
3
/h =89.8 % =8.0 % =23.3 %
110 dm
3
/h =88.8 % =7.9 % =23.4 %
copper recovery in concentrate,
r
gangue recovery in tailings.
Similarly, the influence of the rotational impeller speed on flotation results was investigated. After
recalculating the results, using Fuerstenau`s curves for the same gangue recovery in tailings
r
= 80 %, the following concentration indexes were obtained:
590 rpm =86.1 % =7.7 % =23.3 %
670 rpm =89.8 % =8.0 % =23.3 %
790 rpm =88.8 % =7.9 % =23.2 %
copper recovery in concentrate,
r
gangue recovery in tailings.
Studies on the influence of aeration and rotational impeller speed showed that at given particle size
distributions of ore (90 % <0.025 mm) and in tested flow sheet [19], the most advantageous condi-
tions for flotation occurred at an air consumption amounting to 80 dm
3
/h (1.33 m
3
/h/m
3
of slurry), a
collector consumption of 100 g/Mg and at a rotational impeller speed of 670 rpm.
Except of grain size, hydrodynamic conditions and aeration, the flotation results depend also on the
type and amount of the flotation reagents added, especially frothers and collectors, which determine
Analysis of Fine Particles Behaviour in Flotation of Polish Copper Ores
Proceedings of Copper 2010 2867
Eh and pH levels in floated slurry. Their control during flotation helps to affect on flotation results,
since the copper sulfide minerals float efficiently only at specific values of Eh and pH [21]. The
influence of the potassium xanthate consumption, within the range from 70 to 150 g/Mg, was tested
only during this study. Tested ore was ground to 90 % <0.025 mm. Frother consumption was the
same in all experiments of this series and amounted to 100 g/Mg. The results are presented on
Figure 3 in the form of Fuerstenau`s curves.
Figure 3: Results of flotation of feed ground to 90 % <0.025 mm for different consumption of
potassium ethylo-xanthate: 70, 100 and 150 g/Mg. (curves in Fuerstenau pattern). Flota-
tion conditions: aeration 80 dm
3
/h, impeller rotation 670 rpm
The results were recalculated basing on Fuerstenau`s curves for the same value of gangue recovery
in tailings
r
= 80 % (see Figure 3). The following concentration results were obtained:
70 g/Mg =89.1 % =8.0 % =23.3 %
100 g/Mg =89.8 % =8.0 % =23.3 %
150 g/Mg =87.8 % =7.9 % =23.2 %
copper recovery in concentrate,
r
gangue recovery in tailings.
The best results were obtained for xanthate consumption using up to 100 g/Mg, which is a bit higher
than the standard consumption. Further increase of xanthate consumption, up to 150 g/Mg, gave
slightly worse results, which can often be observed in the case of xanthates.
In order to determine the flotation performance of individual particle classes of ore, the products of
standard ore flotation having 90 % particles below 0.025 mm were screened on 0.005, 0.010, 0.015
and 0.025 mm by micro-sieves and Cu content as well as types of minerals presented in the obtained
0
20
40
60
80
100
85 90 95 100
Cu recovery in concentrate, %
R
e
c
o
v
e
r
y
o
f
g
a
n
g
u
e
m
i
n
e
r
a
l
s
i
n
t
a
i
l
i
n
g
s
,
%
brak wzbogacania
70 g/Mg
100 g/ Mg
150 g/ Mg
no concentration
Potulska
Proceedings of Copper 2010 2868
fractions were determined. This was the basis to define the liberation of sulfide minerals (Figure 4)
as well as their floatability (see Figure 5) for all particle classes.
In the finest classes below 0.010 and 0.010-0.015 mm, liberation of sulfide minerals is practically
the same for all flotation products. Only in the most coarse classes (0.025 and above 0.040 mm) one
can see the evident difference between liberation level for concentrate and for tailings. Into concen-
trate, independently from particle class size, report almost solely liberated sulfide mineral grains.
The highest Cu recoveries are achieved for 0.010-0.015 and 0.015-0.025 mm size classes. While
drop of recovery for sizes above 0.025 mm can be explained by insufficient liberation of sulfide
minerals, for sizes below 0.010 mm, the recovery drop must be caused by specific features of the
fine particles flotation, because as show Figure 4, their liberation is adequate.
Figure 4: Liberation levels of sulfide minerals Figure 5: Copper distribution in all size
in flotation products for feed 90 % of classes of flotation products
particles <0.025 mm. (relative values) (concentrate + middlings)
Among possible reasons for worse flotation results of the finest size classes, the energetic condi-
tions of fine particles surface, improper hydrodynamic conditions during flotation, insufficient con-
sumption of collector, too short flotation time, too wide range of particle size classes, etc. must be
considered.
Based on the achieved results, the chart presenting the floatability of individual sulfide minerals vs.
their particle size distributions is shown in (Figure 6).
70
80
90
100
0 10 20 30 40 50
Particle size, mm
C
u
r
e
c
o
v
e
r
y
,
%
Analysis of Fine Particles Behaviour in Flotation of Polish Copper Ores
Proceedings of Copper 2010 2869
Figure 6: Mineral recoveries for the individual size classes of concentrate. Particle size distributi-
on of feed 90 % <0.025
mm, air consumption 80 dm
3
/h, impeller rotation 670 rpm
and collector consumption 70 g/Mg
Graphs of floatability and recoveries for the main sulfide minerals indicate, that the best floatability
for all size classes of ore being tested, have galena and chalcocite with digenite. Flotation recoveries
of those minerals practically do not depend (galena) or only slightly depend (chalcocite with dige-
nite) on class size. Visibly worse flotability have bornite and chalcopyrite, and definitely the worst
coveline and pyrite with markazite.
>0.025 0.015-0.025 0.010-0.015 <0.010
Particle size, mm
0
10
20
30
40
50
60
70
80
90
100
Bornite
Chalcocite with digenite
Chalcopirite
Pirite with markazite
Galena
Coveline
Potulska
Proceedings of Copper 2010 2870
5 Conclusions
Research results enable to formulate the following conclusions:
1. Grinding fine mineralized ore from Polkowice mine to 90 % <0.025 mm warrants better condi-
tions of flotation than in the case of grinding to 60 and 80 % <0.025 mm. At liberation of sulfide
minerals amounting 95 % (90 % <0.025 mm), maximum recovery of copper bearing minerals in
froth product equal to plants concentrate, amounting to 97-98 %, is observed for particles
within the range of 0.015-0.025 mm. Particles bigger than 0.040 mm are usually poorly libe-
rated.
2. Maximum copper recovery in flotation concentrate containing 25 % Cu, equal to plants concen-
trate, is observed for 0.010-0.025 mm size class. For <0.010 mm size class, under the experi-
mental conditions, a drop of recovery is visible, according to published data, concerning the
floatability of sulfide minerals.
3. The increase of rotational impeller speed in the flotation cell from 590 to 670 rpm, gives a cer-
tain improvement of floatability for all tested particle size distribution ranges. The increase of
slurry aeration also has an advantageous influence on flotation of all particle classes.
4. Concentration of the main sulfide minerals during flotation of Polkowice-Sieroszowice ore,
ground to 90 % < 0.025 mm, may be ranked in the following decreasing order: galena, chalco-
cite with digenite, chalcopyrite, bornite and coveline with pyrite and markasite. Floatability of
chalcocite and galena for all tested size classes, from above 0.025 to below 0.010 mm, changes
in a relatively low degree. However, coveline and pyrite with markasite show the worst floata-
bility of all tested ranges of particle size distributions, especially for the finest class below
0.010 mm.
Acknowledgments
During this study, as one of the analytical methods, the optical analysis was used. Hereby, I would
like to thank Ph.D. Antoni Muszer for carrying out the time-consuming mineralogical investigations
and for his help during the completion of this study.
This study as a research project was financed by the Polish Ministry of Science and Higher Edu-
cation, from its resources for science investigations for 2005 2007. It was also supported by funds
of First Scholarship Program of Integrated Operational Program for Regional Development, where
75 % of founds were covered by the European Social Fund and 25 % by the Polish State budget.
Analysis of Fine Particles Behaviour in Flotation of Polish Copper Ores
Proceedings of Copper 2010 2871
References
[1] TAGGART A. F., (1956): Handbook of Mineral Dressing, Ores and Industrial Minerals. John
Wiley and Sons, New York, Section 12: 92, 97
[2] PEASE J. D., YOUNG M. F., CURRY D., JOHNSON N. W., (2004): Improving fines recov-
ery by grinding finer. MetPlant, Centenary of Flotation Symposium, Brisbane, Australia, 117
[3] PEASE J. D., CURRY D., YOUNG M. F., (2006): Designing flotation circuits for high fines
recovery. Minerals Engeneering, 19: 831840
[4] HARBORT G., MURPHY A., VARGAS A., YOUNG M., (1999): The introduction of the
Isa-Mill for ultrafine grinding in the Mt Isa Lead/Zinc concentrator. For presentation at Ex-
temin 99, Arequipa, Peru, Sept., (Xstrata Technology), 18
[5] GAUDIN A. M., (1963): Flotacja. Tum. z II wydania oryginau: J. Olszewski, T. Piaseczny.
Wyd. lsk, Katowice
[6] TRAHAR W. J., (1976): The selective flotation of galena from sphalerite with special reffer-
ence to the effect of particle size. International Journal of Mineral Processing, 151166
[7] TRAHAR W. J., WARREN L. J., (1976): The flotability of very fine particles a review.
Inter-national Journal of Mineral Processing, 3: 103131
[8] LYNCH A. J., JOHNSON N. W., MANLAPIG E. V., THORNE C. G., (1981): Mineral and
coal flotation circuits. W: Fuerstenau, D.W. Ed., Dev. Miner. Process, Elsevier Science Pub-
lisher B. V., Vol. 3, Amsterdam
[9] POTULSKA A. (2006): Wpyw drobnego mielenia na flotacj rud cynkowo-oowiowych.
Prace Naukowe Instytutu Gornictwa Politechniki Wrocawskiej, nr 116. Seria: Konferencje ;
nr 47, Wrocaw
[10] BORTEL R, (1967): Wpyw mineraw ilastych na flotowalno siarczkw metali niee-
laznych. Praca Doktorska, Politechnika lska, Gliwice
[11] WIENIEWSKI A., (1994): Problemy zwikszenia selektywnoci wzbogacania flotacyjnego
rud metali nieelaznych. Konferencja Komitetu Grnictwa PAN Aktualne zadania nauki w
grnictwie, Ustro
[12] GROTOWSKI A., (1997): Koncepcja rozwiza przerbki z zastosowaniem nowych tech-
nologii flotacji w aspekcie kompleksowego zagospodarowania grniczego wynikajcego z
etapu I, sprawozdanie CBPM Cuprum, Wrocaw
[13] USZCZKIEWICZ A., WIENIEWSKI A., (2006): Kierunki rozwoju technologii
wzbogacania rud w krajowym przemyle miedziowym. Grnictwo i Geoinynieria, Rok 30,
Zeszyt 3/1, AGH Uczelniane Wydawnictwa Naukowo-Dydaktyczne, ISSN: 1732-6702, 181
196
Potulska
Proceedings of Copper 2010 2872
[14] GROTOWSKI A., PAKULSKA B., MIZERA W., (2004): Wstpne okrelenie moliwoci
zastosowania mynw IsaMill do domielenia rudy miedzi w KGHM Polska Copper S.A.,
sprawozdanie CBR Cuprum sp. z o.o. sprawozdanie nr U-004/KU/04
[15] KIJEWSKI P., JAROSZ J., (2007): Odmiany litologiczne rudy. In: Monografia KGHM Pol-
ska Mied S.A. cz II, Praca zbiorowa pod redakcj Piestrzyski A., Wyd. KGHM Cuprum
CBR, Spka z o. o., Wrocaw, Lubin, 463472
[16] SPALISKA B., STEC R., SZTABA K., (2007): Miejsce i rola przerbki rudy w kompleksie
technologicznym KGHM Polska Mied S.A.. In: Monografia KGHM Polska Mied S.A..
cz II, Praca zbiorowa pod redakcj Piestrzyski A., Wyd. KGHM Cuprum CBR, Spka z
o. o., Wrocaw, Lubin, 463472
[17] TUMIDAJSKI T. et al., (2007-2008): Report of Investigation, Project No. KGHM-ZW-U-
0011
[18] GROTOWSKI A., DBKOWSKI R., B., HENZEL S., MIZERA A., MIZERA W.,
PAKULSKA B., SADECKI Z., SZAFRAN A., (2006): D2.2 Evaluation of the existing tech-
nologies used for mining and processing of black shale ore in KGHM Polish Copper. Propos-
als of new technologies for selective separation of shale material from Lubin and Talavivara
shale deposit, sprawozdanie KGHM Cuprum sp. z o.o.
[19] POTULSKA A. (2008): Wpyw drobnego mielenia na flotacj krajowych rud miedzi. Roz-
prawa Doktorska, Wrocaw
[20] DRZYMAA J., AHMED H. A. M., (2005): Mathematical equations for aproximation of
sepa-ration results using the Fuesrtenau upgraiding curves. International Journal of Mineral
Processing, Vol. 76, No. 1-2, April 4, 5565
[21] RICHARDSON P. E, WALKER G. W., (1985): The flotation of chalcocite, bornite, chal-
copy-rite, and pyrite in an electrochemical flotation cell. In: Proceed 5th Int. Min. Process.
Congr., Cannes, 198210
Proceedings of Copper 2010 2873
HPGR versus SAG Milling Technology in
Hard-Rock Mining Review and Analysis
Irshad Rana, Kris Chandrasekaran Ken Wood
Fluor Enterprises, Mining & Metals San Francisco Office Fluor Canada Limited
4140 Dublin Blvd., Suite 300 1075 West Georgia St., Suite 700
Dublin, California, USA Vancouver, Canada
Keywords: HPGR, SAG milling, high pressure grinding rolls, hardrock mining, cement, diamond,
kimberlite, copper
Abstract
High Pressure Grinding Roll (HPGR) technology has been developed based on the pioneering stu-
dies carried out by Professor Klaus Schoenert in Germany on single and packed bed particle fracture
under compression. This work demonstrated significantly lower specific energies for comminution
than those obtained from grinding ball mills. The machine consists of a pair of counter rotating rolls
with one fixed and the other floating. Gravity fed between the rolls, ore crushing takes place by the
mechanism of inter-particle breakage by compression exerted by the hydro-pneumatic spring on the
floating roll. The cement industry became the early adopter of HPGR technology in treating rela-
tively softer material (klinker). Other applications have since followed with kimberlite secondary
and tertiary crushing in the diamond industry, autogenous mill pebble crushing in iron ore
processing, limestone crushing, concentrate fine grinding and a number of others, such as phos-
phate, gypsum, titanium slag, gold ores and coal. Cyprus Sierrita operations in Arizona, USA, were
the first to attempt a large scale HPGR application in 1996 on harder and more abrasive copper ore.
Although the trial outcome did not result in the retrofit of HPGRs to the Sierrita fine crushing cir-
cuit, Phelps Dodge later purchased Cyprus Sierrita and, possibly capitalizing on this experiment,
became the first mining company to introduce HPGRs on a world class copper project at Cerro
Verde (Peru). Phelps Dodge was later purchased by Freeport McMoRan and the company now op-
erates four HPGR units (2.4 m x 1.7 m) instead of a traditional SAG mill circuit at Cerro Verde and
also operates two HPGRs (2.0 m x 1.8 m) at their Grasberg Property in Indonesia. Boddington and
Bendigo gold projects in Australia have also included HPGR units in their plants. Although the
technology is offering attractive possibilities when compared with SAG milling, site specific factors
must be carefully considered in decision making, such as ore wetness, stickiness, abrasion index and
work index etc. Based on various characteristics including plant layout, the capital costs for HPGR
installations are 10 to 25 % higher than the SAG milling process. This higher CAPEX is tempered
Rana, Chandrasekaran, Wood
Proceedings of Copper 2010 2874
by 15 to 25 % reduction in energy cost with potential overall operating cost savings in the range of
10-20 %. As inter-particle crushing promotes micro-cracking, an additional benefit of improved
recoveries from downstream processing may also be realized. This paper will also present an over-
view of the HPGR application in heap leaching technology where micro-cracking could have a posi-
tive influence on metal recoveries.
1 Introduction
Since the application of HPGRs at Cerro Verde (Peru) in 2006 instead of a traditional SAG milling
circuit, the technology has been receiving close scrutiny for projects in the hard-rock mining indus-
try (copper & gold ores). Although Professor Klaus Schoenert at Karlsruhe & Clausthal Universities
in Germany (early 1980s) demonstrated significantly lower specific energies for crushing in a
packed particle bed under compression than those obtained in the tumbling ball mills, the technolo-
gy was not adopted by the mining industry until in the late 1990s.
The cement industry was the first to accept this new technology in the late 1980s for the softer
clinker crushing and several hundred machines are now operating for this sector worldwide. Most of
the applications are in pre-treating feed for the ball mills and this has significantly improved the
downstream grinding capacity and the total energy consumption has been reduced as a result.
In the traditional mineral processing industry, diamond mining in South Africa began utilizing
HPGRs to promote controlled micro-cracking in ores to liberate diamonds without damaging the
product quality. This application has now been adopted in Australia and Canada as well and the
technology has become a standard in most new diamond plants. In iron ores, HPGRs are used large-
ly for pre-treatment of pellet feed to increase plant capacities and for pebble crushing. This has been
now widely accepted in Brazil, India, Sweden, Russia and the USA.
Freeport McMoRans Sierrita operations were the first in copper mining to perform plant tests of
the HPGR technology in 1996 in an effort to replace the existing tertiary crushers. Sierrita flowsheet
is three stage crushing followed by ball milling. The feed material was abrasive quartz diorite with a
nominal particle size of less than 50 mm. However, due to the excessive wear problems on the then
available roll surface protection materials, the pilot test was abandoned. The Sierrita pilot test work
and other efforts in the hard-rock mining led Freeport to review the HPGR technology for the Feasi-
bility Study of Cerro Verde Project in Peru instead of the traditional SAG milling flowsheet. The
study by Fluor Canada in Vancouver showed improved economics with HPGRs and the commis-
sioning of Cerro Verde HPGRs in 2006 marked the culmination of the efforts by many in the indus-
try over the past number of years to have the technology accepted as a legitimate alternative to the
conventional grinding approach for hard-rock mineral processing. Other projects that have recently
adopted HPGR technology include Freeport (Indonesia), Boddington (Australia) and Amplats Pot-
gietersus (South Africa).
HPGR versus SAG Milling Technology in Hard-Rock Mining Review and Analysis
Proceedings of Copper 2010 2875
Freeport Sierrita and Bagdad Operations in Arizona, USA, studied the use of HPGRs again in 2008.
The application was evaluated at the feasibility level (Fluor San Francisco Office located in Dublin,
California, USA) for quaternary crushing mode to increase the milling capacity at Sierrita. At Bag-
dad the expansion was studied to utilize HPGR instead of a conventional SAG milling flowsheet
with two stage crushing followed by HPGRs and a ball mill. After the completion of feasibility
studies, both projects were postponed due to the world economic crisis at that time.
2 HPGR Technology
The machine consists of two counter rotating rolls mounted in a robust frame with each roll sup-
ported by large bearing blocks. One of the bearing blocks is fixed while the other is allowed to float
on rails. Pressure is applied on the floating roll against the fixed by an oil hydraulic system. A stop
block prevents the rolls from contacting each other. Figure 1 provides a typical schematic cross sec-
tion of a HPGR.
Figure 1: Schematic Representation of HPGR (permission from Anguelov et al. [3])
The HPGR machinery is described in detail in numerous papers and vendor brochures along with
the definitions of terminology and the ways to carry out testing/sizing procedures. This article on the
other hand will present the main distinguishing features of the technology:
Rana, Chandrasekaran, Wood
Proceedings of Copper 2010 2876
Ore is choke fed, by gravity, between the rolls and the crushing takes place by a mechanism of
autogenous, compressive breakage in the particle bed. This is in contrast to the tumbling SAG or
ball mills where particles are broken in a random hit and miss manner.
Crushing is accomplished largely by the pressure exerted on the floating roll by the system hy-
draulics and this pressure could vary (50-250 MPa) depending upon the ore characteristics, roll
diameter and nature of the roll surface protection. The compressive forces involved render the
ore density between the rolls in the range of 70-85 % solid by volume. The hydraulic system al-
lows the rolls to open to pass hard metal tramp material, protecting the rolls.
Both rolls are identical and interchangeable and with variable speed drive the roll speeds are ad-
justable to obtain optimum grinding conditions. Pressure on the floating roll and the roll speed
are the most important operating parameters and can be controlled by an operator.
Depending upon the ore type the HPGR product could be compressed into a cake, flake or bri-
quette. The downstream processing may require de-agglomeration in a scrubber etc. Material
handling systems must account for the sheet/flake-like material characteristics.
HPGR product has a greater percentage of fines than can be found in a conventional crusher and
the coarser particles contain extensive micro-cracking which reduces the grinding work needed
in downstream processing to liberate minerals/metals from the ore [1].
The authors experience analysing testwork indicates that for the samples tested, HPGR product
contains a similar amount of fines to SAG product but the HPGR final product has less extreme
fines (slimes) and less material in the coarse fractions. It is felt that this has advantages for sub-
sequent flotation.
Micro-cracking is shown to be beneficial for heap leaching processes where the solution must
penetrate deep into the particles during leaching processes to dissolve metal and enhance kinetics
to generate the pregnant leach solution (PLS).
Diameters of rolls typically range from 0.5 m to 2.8 m and the roll widths vary from 0.2 m to
1.8 m. For comparison, Cerro Verde HPGRs are: 2.4 m diameter and 1.65 m wide processing
2500 tonnes per hour through each unit with a power rating of 5000 kW.
As the ore dwell time between the rolls is short, the machine settings can be easily altered by the
control system, permitting the ease of quick process control and compensation for the fluctuating
feed ore properties.
HPGR versus SAG Milling Technology in Hard-Rock Mining Review and Analysis
Proceedings of Copper 2010 2877
HPGR technology is characterized by:
Minerals are liberated generally along the grain boundaries,
High and steady throughput rates if the ore is dry and non-sticky,
Reduced noise levels,
Low and manageable dust loading,
The penetration of the HPGR technology in the hard-rock mining application was limited by
the wear resistance of the roll surfaces. However, recent advances in wear-resistant materials
and surfaces have improved roll service lives in the range of 2000 to 6000 hours for the abra-
sive hard rocks. The roll service lives can exceed 15,000 hours when treating softer materials.
The following three manufacturers are currently supplying HPGR machines:
Polysius (a Thyssen Krupp Company),
Kppern,
KHD Humbolt Wedag AG.
Except for minor differences, the three suppliers use largely similar technology.
Polysius employs a roll design with high aspect ratio (bigger diameter with smaller width) whereas
KHD and Kppern utilize a lower aspect ratio. Larger diameter rolls are expensive but offer a wider
operating gap and the roll surfaces are exposed to a smaller portion of the material processed. This
machine characteristic helps improve the roll wear life.
Tungsten carbide studs are being used to generate an autogenous wear layer on the roll surface un-
der a patent by KHD. Polysius and Kppern are also offering HPGRs with this innovation under a
license from KHD. Different roll surfaces are available, such as: pattern welded, studded, Hexadur
lined, and chevron welded. Equipment manufacturers provide detailed brochures with the available
roll surfaces based on the applicable ore hardness.
The HPGR has a lower unit capacity than SAG mills and as such is more readily scalable for
increasing plant capacity, although conveying systems may prove to be limitations to expansion.
3 Screening Parameters for HPGR vs. SAG Milling
3.1 Impact crushability work Index
The Bond Crushing Work Index of 13 and higher indicates that the ore to be crushed is a hard ore.
An impact work index of 20 and higher indicates that the ore is a very hard ore. The hard competent
rock is a candidate for the application of HPGR.
Rana, Chandrasekaran, Wood
Proceedings of Copper 2010 2878
The compressive strength of the material to be crushed determines the amount of energy that can be
absorbed by the material. This in turn determines the HPGR power requirement.
3.2 Abrasion Index
The abrasion index of the material to be crushed will determine the metal wear of the HPGR liner
material. Unless the feed ore has a very high abrasion index, HPGR is suitable for size reduction
such that single stage milling in a ball mill subsequent to the HPGR can be adopted for the grinding
circuit. With a high abrasion index, SAG mills will also experience high media consumption and
liner wear.
3.3 Ore Characteristics
Ore containing significant amounts of clay and talc minerals along with excessive moisture can
cause problems in maintaining a stable autogenous layer between the studs. The recommended feed
moisture is up to 4 %. Feed moisture significantly in excess of this will cause operational material
handling problems with the HPGR and associated bins, feeders, upstream crushers, conveyor trans-
fers, etc. Higher feed moisture also increases metal wear. These are conditions for which Autogen-
ous and SAG mills were originally applied.
3.4 Plant Layout
The layout of the HPGR facilities influences the economics due to the physical site conditions. The
use of HPGR in lieu of SAG mill needs the following additional facilities.
The primary crushed ore secondary crushing and screening facility. The secondary crusher oper-
ates in closed circuit with the screens.
HPGR feed bin and HPGRs.
Fine screens operating in closed circuit with HPGRs.
Conveyors for material transfer between facilities.
Dust control systems.
All the above facilities will occupy a relatively large area. In hilly terrain, especially where different
process facilities are required to be located at different elevations, the plant area comes at a pre-
mium. In addition, the terrain could involve a large quantity of earthwork for site preparation and
this quantity could be overwhelming. Even though the HPGR technology may be suitable to adopt,
the project economics may not support HPGR application.
HPGR versus SAG Milling Technology in Hard-Rock Mining Review and Analysis
Proceedings of Copper 2010 2879
4 Down-stream Processing Considerations
HPGR technology offers certain unique down-stream considerations in hard-rock mining applica-
tions due to the micro-cracking induced in the crushed particles, as described below.
4.1 Reduced Work Index
Laboratory tests have been carried out on various ores using the standard Bond grinding index pro-
cedures and have indicated reductions in Bond ball mill work index (Wi) on the HPGR product
compared with the original feed. This would, of course, vary from ore to ore and is dependent on the
grind size tested, but there have been reported reductions in the range of 0-25 %. Some recent cop-
per project tests have shown Wi reductions of 6-21 % [1]. HPGR suppliers have created grindability
test methodologies that start with a coarser feed than the Bond ball mill work index test and as a
result take into account the effect of micro-cracking on the coarser size fractions as well as the in-
creased production of fines from the HPGR compared to conventional crushers. These tests show
reductions in ball mill energy requirements more in the range of 10-40 % for the same samples as
the Wi values noted above, compared to conventionally crushed feed [1]. In comparison to SAG
mill product, only the Wi reduction would be expected to be evident since both SAG and HPGR
products have similar fine content. Such reduction in ball mill grinding energy would result in either
smaller ball mill grinding equipment, higher downstream capacity for grinding ore or a better grind
for a similar capacity. The mill operators therefore could opt for either higher capacity or better
metal recovery from the grinding fineness.
4.2 Flotation Recovery
Flotation losses tend to be greatest in the extreme fines and the coarse fractions. SAG product has
more material in both these fractions compared with HPGR product and can be assumed to be more
susceptible to recovery losses. In addition, HPGR product tends to break along grain boundaries,
leading to better liberation at coarser sizes. This may allow a coarser grind for the same recovery or
higher recovery at a given grind compared to SAG product.
4.3 Leach Recovery
Comparison of gold leach recoveries on a size by size basis on both South African and a refractory
Nevada ore showed improved recovery for ore that had been treated in an HPGR circuit compared
to conventional. Gold leach recoveries of 5-20 % higher were experienced with the HPGR
product [2].
Rana, Chandrasekaran, Wood
Proceedings of Copper 2010 2880
4.4 Ball Mill Optimization
The relatively tightly sized HPGR product after screening allows optimization of ball mill media
size for further improvement in ball mill efficiency and capacity in a close circuit configuration. The
reduced coarse material content in the slurry can also reduce wear on ancillary equipment such as
pumps, piping and cyclones compared to SAG circuits.
4.5 HPGR in Heap Leaching
Heap leaching of copper and gold ores is a major consideration for HPGR technology. The main
advantages for HPGR units versus the conventional crushing circuits are:
Micro-cracking of particles would permit better penetration of lixiviant,
Better mineral liberation and exposure to leaching,
Increased oxygen penetration.
Due to the factors above, the following observations have been reported in heap leaching regimes:
HPGR crushed ores exhibit better solution percolation compared with the conventional
crushing [1].
Approximately 60 % of the coarse ore particles (+100 mesh) are micro-fractured and hence the
diffusion paths to metal minerals are significantly better. This results in better metal recoveries.
The overall reaction kinetics are significantly improved for the ores with slow penetration rates.
Improved lixiviant accessibility through micro-cracks leads to reduction in leach reagent con-
sumption rate.
5 HPGR vs. SAG Milling Cost Considerations
In light of various trade off studies carried out recently, HPGR capital costs are 8-12 % higher than
the comparable SAG mill costs. This project capital gap appears to be narrowing as the HPGR tech-
nology is improving and the applications are better understood and analyzed. The increased capital
is mainly due to the requirement for a secondary crushing circuit and the associated conveyors and
screens to handle the processing. The material handling required with the HPGR mandates a larger
layout footprint and the associated civil earthworks.
HPGR energy utilization is more efficient as crushing takes place by a mechanism of autogenous
compressive breakage in a particle bed in contrast to the tumbling SAG mill where particles are
broken in a random hit and miss manner. Thus the comparatively higher HPGR capital cost is tem-
pered by 15-25 % reduction in energy cost with potential overall operating cost savings in the range
HPGR versus SAG Milling Technology in Hard-Rock Mining Review and Analysis
Proceedings of Copper 2010 2881
of 10-20 %. Next to the energy cost, absence of grinding media cost (ball consumption) is a major
factor in reducing the HPGR operating cost.
Anguelov et al. [3] presented the results of trade off studies, in a recent paper, between the HPGR
and the SAG milling technologies for various hard-rock mining projects. Without naming the
projects involved, given below are the results for capital and operating costs from the referenced
publication:
Table 1: Overall Mill Operating Cost
Table 2: Overall Capital Cost
Fluor San Francisco Office has carried out studies recently between HPGR and SAG milling circuits
and produced similar results (HPGR capital + 10-12 %).
Projects SAG Mill HPGR
Project A 5.30 $/t 4.56 $/t
Project B 2.24 $/t 1.63 $/t
Project C 4.98 $/t 3.62 $/t
Project D 2.66 $/t 2.03 $/t
Projects HPGR vs. SAG
Project A + 6.4 %
Project B - 14.3 %
Project C + 8.2 %
Project D + 9.6 %
Rana, Chandrasekaran, Wood
Proceedings of Copper 2010 2882
6 Conclusions
Based on the review and analysis of HPGR versus SAG Milling above, the following conclusions
are presented:
HPGR technology should be evaluated in light of the site specific factors for greenfield and
brownfield projects. This includes the available layout space, topography of the area, energy cost,
labor cost and the remoteness of the project site. In a brownfield project, the evaluation must
consider the fitting in of the HPGR technology within the current mix of the process flowsheet
equipment.
Wet, sticky feed and ores containing excessive clays should be evaluated critically for the materi-
al flow bottlenecks in HPGR type flowsheets. These are probably better suited to SAG milling.
Extensive laboratory and pilot plant testwork should be carried out before deciding between the
HPGR and SAG milling circuit. This would shed light on the energy consumption, ore hardness
and ore abrasiveness indices.
HPGR capital costs are expected to be 8-12 % higher than the SAG mill circuit, under the current
state of technology.
Overall operating costs are expected to be 10-20 % lower than the SAG mill circuit due to the
lower energy cost with the HPGR circuit and the absence of grinding media cost (roll liners con-
sidered).
Energy costs are expected to be 15-25 % lower for the HPGR circuit compared with the SAG
mill.
Acknowledgments
The authors wish to thank Fluor Enterprises and the management of Fluor San Francisco Office
(Mining & Metals) for granting permission to publish and allowing resources to develop the paper.
References
[1] Unpublished test data on copper/gold ores from North and South America
[2] PATZELT, N., KNECHT, H., BAUM, W. (1995): Case made for High Pressure Roll Grinding
in Gold Plants Mining Engineering; June 1995, 524-529
[3] ANGUELOV, R., GHAFFARI, H., ALEXANDER, J.: High Pressure Grinding Rolls (HPGR),
an Alternative Technology versus SAG Milling, Conference: Communition of Ores, Common
Wealth of Cornwall, UK. 2008 & China Mining Conference, 2008
Proceedings of Copper 2010 2883
Selective Leaching of Arsenic from Copper Ores
and Concentrates Containing Enargite in
NaHS Media
William Tongamp, Yasushi Takasaki, Atsushi Shibayama
Akita University
Faculty of Engineering and Resource Science
11 TegataGakuen cho
Akita City 0108502, Japan
Keywords: Selective leaching, arsenic, copper resources, enargite, sodium hydrosulfide
Abstract
Conventional smelting processes cannot treat Cuores/concentrates containing significant amount
of toxic arsenic, and development of pretreatment methods to remove arsenic (As) is needed. This
work reports selective leaching of As from enargite (Cu
3
AsS
4
) contained in Cuores/concentrates
using NaHS media. Samples were obtained from various sources with As content ranging from 0.9
to 14 wt. % and Cu content from 20 to 40 wt. %. Leaching experiments were conducted at 30-95 C
under atmospheric pressure with pulp densities from 100 to 1000 g/L. Concentration of NaHS was
varied from 50 to 200 g/L, and 50 to 100 g/L of NaOH was added. The kinetic analysis of the data
found that the reaction is controlled by a product layer diffusion process with an activation energy of
70.3 4.7 kJ/mol. Chemical analysis and X-ray diffraction analysis of the leached samples showed
that As content in the samples were reduced to less than 0.5 wt. % within 3 to 6 h even at high pulp
density of 1000 g/L, and that Cu
3
AsS
4
was transformed to Cu
2
S.
1 Introduction
Enargite (Cu
3
AsS
4
) and tennantite (Cu
12
As
4
S
13
) are the main minerals with high As content that
associate with copper sulfides such as covellite (CuS), chalcocite (Cu
2
S), chalcopyrite (CuFeS
2
),
Bornite (Cu
5
FeS
4
), and are normally recovered by flotation and offered as Cu-resource to smelters
for Cu recovery. The high content of As effectively reduces value of the Cu-resource due to envi-
ronmental implications of toxic As that require additional processing steps for containment of As.
As demand for Cu continues to grow, the need to source Cu from all resources including those that
contain high As is necessary and as such more economical and environmentally friendly processing
Tongamp, Takasaki, Shibayama
Proceedings of Copper 2010 2884
options will be sought. Reviews on Curesources containing enargite are covered widely in litera-
ture by many researchers [1-8].
A method to remove As from the Cuores/concentrates containing enargite is flotation [2, 5, 9-13];
by employing slurry potential and pH control with depressants and activators, a clean concentrate
with As content < 0.5 wt. % are recovered. Conversion of enargite to CuO or Cu
2
S by selective dis-
solution of As into solution with sodium hypochlorite (NaClO) or sodium sulfide (Na
2
S) in alkaline
solutions with NaOH is also effective to remove As from the Cu-ores/concentrates [1-5, 6, 14-17].
Other hydrometallurgical treatment options through chloride and sulfates leaching, or in ammonical
solutions and at elevated temperatures and pressures is reported in many literature listed above and
elsewhere [18, 19].
In a recent study, the same authors have introduced sodium hydrosulfide (NaHS) as a leaching me-
dia for the conversion of enargite (Cu
3
AsS
4
) to Cu
2
S in NaHSNaOH solution [20], and have
shown that near complete removal of As (> 99 %) in enargite can be achieved.
The behavior of enargite conversion in NaHS media is very similar to that in sodium sulfide (Na
2
S)
media [2, 4, 6, 14, 17], however, selection of NaHS was based on its application in copper flotation
and also its higher sulfur content (~43 %) per unit weight NaHS as compared to Na
2
S (~13 %) since
leaching of As from enargite is a function of sulfur [S
2-
] concentration.
In the present study, the effects of parameters such as particle size, slurry stirring rate, NaHS con-
centration and temperature were investigated to study the reaction kinetics of enargite in NaHS
NaOH media.
2 Experimental
2.1 Sample preparation
Enargite ores and Cu-concentrates samples used in this work have been obtained from various
sources with differing amounts of As ranging from 0.8 to 14 wt. % and Cu from 10 to 40 wt. %. Ore
samples were first prepared by crushing and grinding followed by sizing of both ores and concen-
trates to obtain a size fraction of -75+35 m for all leaching experiments. Samples with As content
< 1.0 wt. % were also blended with enargite ore with As content >8.0 wt. % to obtain a composite
with As content at 3.0 wt. %. Sample information and chemical analysis of the original samples is
given in Table 1.
Selective Leaching of Arsenic from Copper Ores and Concentrates
Proceedings of Copper 2010 2885
Table 1: Main chemical and phase compositions in Cuores/concentrate samples with As impu-
rity from various sources.
Sample identity As (%) Cu (%) Fe (%) main composition
(1) Cu conc. 0.81 26.62 25.40 CuFeS
2
, FeS
2
(2) Cu conc. 0.95 22.60 18.17 CuFeS
2
, FeS
2
(3) Cu conc. 1.67 11.49 9.71 CuFeS
2
, FeS
(4) Cu conc. 3.11 33.46 14.34 FeS
2
, Cu
2
FeSiS
4
(5) Enargite ore 8.32 33.71 16.34 Cu
3
AsS
4
, CuFeS
2
(6) mixture (1) + (5) 3.00 25.84 21.04 -
(7) mixture (2) + (5) 3.00 25.69 21.08 -
(8) Enargite ore 3.65 12.32 23.37 Cu
3
AsS
4
, FeS
2
2.2 Leaching
All leaching experiments were conducted in a 200 ml Teflon beaker immersed in a water bath and
total solution volume was kept constant at 100 ml, fully enclosed with a thermometer fixed for
monitoring temperature and cooling system to prevent solution evaporation. NaHS as leaching me-
dia and NaOH to maintain high pH and prevent sulfur hydrolysis were obtained as pure chemical
reagents from Wako Chemicals Japan and added at (50200) g/L.
After adjusting temperature to set point, samples at varying pulp densities ranging between 100 and
1000 g/L were weighed and introduced into the solution containing NaHS. Slurry temperature dur-
ing leaching was varied between 30 to 95 C to evaluate the effect of temperature; most tests were
conducted at 80 C in this work and leaching time varied between 5 min to 10 h. At the end of each
experimental run, the solids were separated from solution by filtration to obtain filtrate and residue
for analyses.
Tongamp, Takasaki, Shibayama
Proceedings of Copper 2010 2886
Figure 1: Schematic illustration of experimental flowsheet for enargite leaching and main condi-
tions.
2.3 Characterization and analyses
X-ray diffraction (XRD) powder analysis was carried out using Rigaku, RINT2200 / PC system with
a CuK irradiation source ( = 1.5405 ) for starting samples and residues after leaching to observe
changes in phase formations. Solution potential and pH both at the beginning and end of each leaching
experiment was measured using laboratory Eh/pH meters to observe changes and relate to relationship
of Eh and pH conditions in enargite transformation. Analyses of filtrates were carried out using an
Inductively Coupled Plasma (ICPAES/OES) atomic/optical emission spectrometer, SPS3000 (Seiko
Instruments Inc.) to monitor dissolution of As, Cu and Fe during leaching.
3 Experimental results
3.1 Effect of temperature
The influence of temperature on As dissolution was investigated by conducting experiments from
30 to 95 C using the -75+32 m fraction of enargite ore (8) while keeping reagent concentrations
fixed at 100 g/L (1.35 M) NaHS + 50 g/L (1.25 M) NaOH and the results obtained are shown in
Figure 2.
(crush/grind/sizing/blending)
NaSH + NaOH +
H
2
O
Sample preparation
Na
3
AsS
3
solution
Selective leaching
Solid/Liquid separation
Arsenic treatment Clean Cu conc
Experimental conditions
Pulp density: 100 1000 g/L
NaHS addition: 50 200 g/L
Leaching time : 10 min 12 h
Temperature: 30 95 C
Stirrer speed: 150 750 rpm
NaOH addition: 50 100 g/L
Enargite
Selective Leaching of Arsenic from Copper Ores and Concentrates
Proceedings of Copper 2010 2887
Figure 2: Arsenic dissolution behavior as a function of leaching time for a 1.0 g enargite ore (8)
Leaching of -75+32 m at different temperatures in a 100 g/L (1.35 M) NaHS and
50 g/L (1.25 M) NaOH solution. Stirring rate was fixed at 550 rpm.
These data shows that temperature exerts significant effect on As leaching rate. At 30 C, As disso-
lution reached only 13 % in 30 min and shows no improvement after that. Increasing temperature to
40 C, As dissolution reaches 20 % in the first 30 min and reaches 36 % in 120 min. Further in-
crease in temperature, As dissolution significantly increases reaching over 99 % in 120 min for sam-
ple leached at 80C and over 99 % in 90 min for sample leached at 95 C.
In a second case, another sample; Cu-concentrate (3) containing 1.67 wt. % As was leached for 3 h
at temperatures ranging between 30 and 95 C but at a much higher pulp density of 500 g/L is
shown in Figure 3. NaHS and NaOH additions were fixed at 100 g/L each and these corresponds to
S
2-
/As ratio of 16.20 (mol/mol ratio). At 30 C As dissolution reached only 10 %, but with increase
in temperature, As dissolution reached over 47 % at 60 C, 87 % at 80 C, and over 95 % at 95 C.
To reduce As content from 1.67 wt. % to <0.50 wt. %, up to 70 % dissolution of As in the sample is
required and this result indicates that within 3 h and at 80 C As content in original sample was re-
duced to about 0.22 wt. %. Based on this result, all proceeding experiments were conducted at
80 C and leaching times varied from 1 to 6 h.
0
20
40
60
80
100
0 20 40 60 80 100 120
30 C
Leaching time, t / min
A
r
s
e
n
i
c
r
e
m
o
v
a
l
(
%
)
50 C
60 C
70 C
80 C
95 C
Tongamp, Takasaki, Shibayama
Proceedings of Copper 2010 2888
Figure 3: Dissolution behavior of As from Cu-concentrate (3) at different temperatures. Leaching
time was fixed at 3 h, pulp density at 500 g/L with NaHS and NaOH additions
at 100 g/L.
3.2 Effect of NaHS and NaOH concentration
For subsequent leaching experiments to investigate effect of NaHS and NaOH concentrations on As
leaching, temperature was kept constant at 80 C and pulp density fixed at 500 g/L. Cu-concentrate
(3) with As content at 1.67 wt. % was leached for 1 to 4 h and varying NaHS/NaOH concentrations
and results obtained is shown in Figure 4. At 100 g/L NaHS and NaOH additions respectively
(S
2-
/As = 16.20), As dissolution rapidly increases to 70 % in the first 1 h and gradually increases to
above 80 % within 3 h and almost complete dissolution in 4 h. This shows As content in this sample
could be reduced to below 0.5 wt. % within 1 to 4 h. However, when NaHS addition is reduced to
50 g/L (S
2-
/As = 8.10), overall dissolution does not exceed 40 %. When addition of NaOH is re-
duced to 50 g/L and maintaining NaHS at 100 g/L, dissolution rate of As is slightly decreased and
only reaches 70 % dissolution after 3 h for the same sample. This result indicates that at 500 g/L
pulp density and leaching temperature of 80 C, concentration of both NaHS and NaOH of at least
100 g/L (S
2-
/As = 16.20) is required to reduce As content in the sample to <0.50 wt. %
within 2 to 4 h.
0
20
40
60
80
100
9
5
.
4
0
%
8
7
.
9
1
%
4
7
.
1
7
%
1
0
.
0
5
%
95
o
C 80
o
C 60
o
C 30
o
C
A
r
s
e
n
i
c
d
i
s
s
o
l
u
t
i
o
n
(
%
)
Leaching temperature, T /
o
C
Selective Leaching of Arsenic from Copper Ores and Concentrates
Proceedings of Copper 2010 2889
Figure 4: Effect of NaHS and NaOH concentrations on As dissolution as a function of leaching
time at 80 C and pulp density of 500 g/L. (a) Both NaHS/NaOH addition at 100 g/L;
(b) NaOH addition at 50 g/L; (c) NaHS addition at 50 g/L.
3.3 Effect of pulp density on leaching
Varying pulp densities at 100 to 600 g/L was investigated to determine optimum pulp density at
which As content in the samples could be reduced to <0.50 wt. % within 6h and at low NaHS/NaOH
concentration and leaching temperature fixed at 80 C. As discussed in previous sections above,
results at 500 and 600 g/L show that such an objective could be established as shown in Figure 5. At
600 g/L (S
2-
/As = 13.50), leaching rate is slightly lower than 500 g/L and As dissolution reaches
70 % after 2 h, however it is also clear that As content can be reduced to <0.50 wt. % within 3 to
6 h. For a pulp density of 100 g/L, complete removal of As was achieved within 3 h under similar
conditions.
Varying addition of NaHS and NaOH to observe As dissolution of sample (7) with a pulp density of
1000 g/L was performed and the results are shown in Figure 6. At 200 g/L NaHS addition
(S/As = 6.7) over 80 % of As is dissolved in 1 h and reaches over 95 % in 6 h and total dissolution
achieved in 10 h. Decreasing NaHS addition to 150 g/L (S/As = 5.1) and 100 g/L (S/As = 3.4), the
As dissolution is reduced accordingly and reaches 70 to 80 % in 6 h, after which it remains consis-
tent at prolonged leaching up to 10 h. This result suggests that a S/As molar ratio of at least 6 is re-
quired from a stoichiometric requirement of S/As = 1.5 to reduce As content from 3.0 wt. % to be-
low 0.5 wt. % for the Cuconcentrate samples mixed with enargite ore within 6 h.
0 1 2 3 4
0
20
40
60
80
100
(c): S/As = 6.1
(b): S/As = 12.3
(a): S/As = 12.3
A
r
s
e
n
i
c
d
i
s
s
o
l
u
t
i
o
n
(
%
)
Leaching time, t / hours
Tongamp, Takasaki, Shibayama
Proceedings of Copper 2010 2890
Figure 5: Dissolution behavior of As from Cu-concentrate (3) at different pulp densities as a func-
tion of leaching times at 80 C. NaHS/NaOH concentrations fixed at 100 g/L.
Figure 6: Arsenic leaching behavior of blended sample (7) with arsenic content at 3.0 wt. % at a
very high pulp density of 1000 g/L. Effect of NaHS and NaOH addition on arsenic dis-
solution as a function of leaching time at 80 C was investigated and aggressive stirring
at 650 to 750 rpm was performed to keep sample evenly dispersed in slurry.
X-ray diffraction analysis of starting enargite ore (5) with As content of 8.32 wt. % and compared to
leach residues obtained after leaching for different times; 1, 3 and 6 h is shown in Figure 7. The
result shows near complete disappearance of characteristic peaks of Cu
3
AsS
4
from leach residues
obtained after 1 h. On the other hand, characteristic peaks of Cu
2
S clearly appear from residues ob-
tained after 1 h leaching and their intensity increases significantly as leaching progresses.
0
20
40
60
80
100
0 1 2 3 4 5 6
600 g/L (S/As = 10.2)
500 g/L (S/As = 12.3)
A
r
s
e
n
i
c
d
i
s
s
o
l
u
t
i
o
n
(
%
)
Leaching time, t / hours
100 g/L (S/As = 61)
0
20
40
60
80
100
0 1 2 3 4 5 6 7 8 9 10
A
r
s
e
n
i
c
d
i
s
s
o
l
u
t
i
o
n
(
%
)
200 g/L NaHS (S/As = 6.7)
Leaching time, t / hours
100 g/L NaHS (S/As = 3.3)
150 g/L NaHS (S/As = 5.1)
Selective Leaching of Arsenic from Copper Ores and Concentrates
Proceedings of Copper 2010 2891
Figure 7: X-ray diffraction pattern of original enargite ore (5) and the leach residues obtained after
leaching for different times at 80 C. Pulp density, NaHS and NaOH additions were kept
constant at 500 g/L and 100 g/L respectively.
3.4 The leaching mechanism
The transformation or leaching mechanism of enargite (Cu
3
AsS
4
) to Cu
2
S in NaHS and NaOH me-
dia can be represented by Reaction (1) below. The dissociation of NaHS added as a leaching me-
dium can be represented by Reaction (2), which further dissociates to provide S
2-
as shown by Reac-
tion (3) in the presence of OH
-
(NaOH). Therefore, maintaining high pH in high alkaline region
(> 12.5) prevents the hydrolysis of the S
2-
ions to HS
-
, and maintains a high concentration of S
2-
ions
to dissolve As from Cu
3
AsS
4
.
2Cu
3
AsS
4
+ 3NaHS + 3NaOH = 2Na
3
AsS
4
+ 3Cu
2
S + 3H
2
O (1)
NaHS Na
+
+ HS
-
(2)
HS
-
+ OH
-
S
2-
+ H
2
O (3)
The stoichiometric requirements according to Reaction (1), shows that one mole of As would re-
quire 1.5 moles of sulfur (S/As = 1.5) for dissolution of As from enargite. The transformation of
Cu
3
AsS
4
to Cu
2
S as represented by Reaction (1) in NaHS media is similar that for Na
2
S. NaHS was
chosen as reagent for this work due to its application in Cu flotation and higher sulfur content (2 to
3 times higher).
10 20 30 40 50 60 70
2 /degree
0h
1h
3h
Cu
3
AsS
4
Cu
2
S
FeS
2
SiO
2 6h
3
(
1
)
2
/
3
+
2
(
1
)
Leaching time, t / min
0.00270 0.00285 0.00300 0.00315 0.00330 0.00345
-14
-13
-12
-11
-10
-9
-8
R = -0.99103
RootMSE (SD) = 0.29009
l
n
k
1/T
E
a
= 70.26 4.74 kJ/mol
Tongamp, Takasaki, Shibayama
Proceedings of Copper 2010 2894
To investigate effect of NaHS concentration, the (-75+32 m) fraction was leached in 50 (0.68 M),
75 (1.01 M), 100 g/L (1.35 M) + 50 g/L (1.25 M) NaOH for 120 min. Slurry temperature and stir-
ring rate were fixed at 80 C and 550 rpm respectively. At 100 g/L (1.35 M) NaHS addition, over
99 % As is removed within 2 h, however, the leaching rate decreases when NaHS concentration is
reduced to 75 (1.01 M) and 50 g/L (0.68 M), reaching up to 85 % and 55 % for the two NaHS con-
centrations respectively in 2 h. Low As removal at 50 and 75 g/L NaHS addition could be attributed
to by hydrolysis of S
2-
ions to HS
-
ions as in both cases pH values decreased from 13.7 for 100 g/L
NaHS addition to below pH 12.7 and pH 12.3 for 75 and 50 g/L NaHS respectively. For tests at
100 g/L NaHS addition, the slurry pH and potential (Eh vs. SHE) remained consistently over 13.5
and Eh values ranging between -500 to -550 mV.
To investigate effect of stirring rate, the (-75+32 m) fraction was leached in
100 g/L (1.35 M) NaHS + 50 g/L (1.25 M) NaOH at 80 C under three different stirring rates. At
550 rpm, over 99 % of arsenic removal was achieved within 2 h; however, on decrease in stirring
speed arsenic removal proceeds slower and reached 89 and 82 % in 2 h for 360 and 200 rpm,
respectively. These results suggest that the leaching of enargite in NaHSNaOH media is of diffu-
sion control, since a chemical reaction controlled process would show not much effect of stirring
speed on the leaching rate.
4 Conclusion
Selective leaching of As impurity in Cu-ores/concentrates containing enargite (Cu
3
AsS
4
) by leaching
in NaHS and NaOH media was investigated in this work. The results obtained and discussed can be
summarized as follows:
1. As contents in Cuores/concentrates can be reduced to <0.50 wt. % within 1 to 3 h for samples
with As content <4.0 wt. % and within 3 to 6 h for samples with As content <10.0 wt. % at tem-
peratures between 80 to 90 C. Such result was also obtained for pulp density at 1000 g/L.
2. Enargite (Cu
3
AsS
4
) was transformed to Cu
2
S as a result of reduction of As
5+
in enargite being
reduced to As
3+
(Na
3
AsS
4
) by S
2-
offered by NaHS. Xray diffraction analysis of leach residues
and solution pH (> 12.50) and potential (< -500 mV (vs. SHE)) measurements correspond well
with this result.
3. Selective leaching of As from enargite (Cu
3
AsS
4
) and subsequent formation of Cu
2
S in
NaHSNaOH system occurs according to the reaction:
2Cu
3
AsS
4
+ 3NaHS + 3NaOH = 2Na
3
AsS
4
+ 3Cu
2
S + 3H
2
O.
4. The kinetics of As leaching from enargite in NaHSNaOH solution was well represented by the
diffusion controlled model; kt = 1-3 (1-)
2/3
+2 (1-) and the activation energy was calculated to
be 70.26 4.74 kJ/mol.
Selective Leaching of Arsenic from Copper Ores and Concentrates
Proceedings of Copper 2010 2895
As impurity in Cu-ores/concentrates can be successfully removed be leaching in NaHS media as
shown by the results in this work. Treatment of solution from the leaching operation for removal of
As is a next step for complete process development.
References
[1] LATTANZI P., DA PELO S., MUSU E., ATZEI D., ELSENER B., FANTAUZZI M., ROSSI
A.: Enargite oxidation: A Review, EarthScience Reviews, 2008, 86, pp. 6288.
[2] FILIPPOU D., STGERMAIN P., GRAMMATIKOPOULOS.: Recovery of metal values
from CopperArsenic minerals and other related resources, Mineral Processing & Extractive
Metall. Rev., 2007, 28, pp. 247298.
[3] MIHAJLOVIC I., STRBAC N., ZIVKOVIC Z., KOVACEVIC R., STEHERNIK M.: A po-
tential method for arsenic removal from copper concentrates, Miner. Engin., 2007, 20, pp. 26
33.
[4] BAL P. and ACHIMOVIOV M.: Selective leaching of antimony and arsenic from me-
chanically activated tetrahedrite, jamesonite and enargite, Int. J. Miner. Process, 2006, 81, pp.
4450.
[5] GUO H. and YEN W.T.: Selective flotation of enargite from chalcopyrite by electrochemical
control. Miner. Engin., 2005, 18, pp. 605612.
[6] VIALS J., ROCA A., HERNNDEZ M.C., BENAVENTE O.: Topochemical transforma-
tion of enargite into copper oxide by hypochlorite leaching, Hydrometallurgy, 2003, 68, pp.
183193.
[7] WELHAM N.J.: Mechanochemical processing of enargite (Cu
3
AsS
4
), Hydrometallurgy, 2001,
62, pp. 165173.
[8] PADILA R., FAN Y., SANCHEZ M., WILKOMIRSKY I.: Processing high arsenic copper
concentrates. In: Sanchez, M.A., Vergara, F., Castro, S.H., (Eds.), 1998, Environment, Inova-
tion in Mining and Mineral Technology vol. 2. University of Conception, Chile, pp. 603612.
[9] SENIOR G.D., GUY P.J., BRUCKARD W.J.: The selective flotation of enargite from other
copper minerals a single mineral study into relation to beneficiation of the Tampakan de-
posit in the Philippines, Int. J. Miner. Process, 2006, 81, pp. 1526.
[10] KANTAR C.: Solution and flotation chemistry of enargite, Colloids and Surfaces A: Physical
and Engineering Aspects, 2002, 210, pp. 2331.
[11] FORNASIERO D., FULLSTON D., LI C., RALSTON J.: Separation of enargite and tennan-
tite from non-arsenic copper sulfide minerals by selective oxidation or dissolution, Int. J.
Miner. Process, 2001, 61, pp. 109119.
Tongamp, Takasaki, Shibayama
Proceedings of Copper 2010 2896
[12] PAUPORT TH., SCHUHMANN D.: An electrochemical study of natural enargite under
conditions relating to those used in flotation of sulfide minerals, Colloids and Surfaces A:
Physicochemical and Engineering Aspects, 1996, 111, pp. 119.
[13] MENACHO J.M., ALIAGA, W., VALENUELA R., RAMOS V., OLIVARES L.: Selective
flotation of enargite and chalcopyrite, Minerals, 1993, 48 (203), pp. 3339.
[14] BAL P. and ACHIMOVIOV M.: Mechanochemical leaching in hydrometallurgy of
complex sulphides, Hydrometallurgy, 2006, 84, pp. 6068.
[15] CURRELI L., GHIANI M., SURRACCO M., ORR G.: Beneficiation of a gold bearing
enargite ore by flotation and As leaching with Nahypochlorite. Minerals Engineering, 2005,
18, pp. 849854.
[16] DELFINI M., FERRINI M., MANNI A., MASSACCI P., PIGA L.: Arsenic leaching by Na
2
S
to decontaminate tailings coming from colemanite processing, Miner. Engin., 2003, 16, pp.
4550.
[17] BAL P., ACHIMOVIOV M., BASTL Z., OHTANI T., SNCHEZ M.: Influence of
mechanical activation on the alkaline leaching of enargite concentrate, Hydrometallurgy,
2000, 54, pp. 205216.
[18] HERREROS R.O., QUIROZ HERNNDEZ M.C., VIALS J.: Dissolution kinetics of enar-
gite in dilute Cl
2
/Cl media, Hydrometallurgy, 2002, 64, pp. 153160.
[19] DUTRIZAC J.E. and MACDONALD R.J.C.: The kinetics of dissolution of enargite in acidi-
fied ferric sulfate solution, Can. Metall. Q., 1971, 11(3), pp. 469476.
[20] TONGAMP W., TAKASAKI Y., SHIBAYAMA A.: Removal of arsenic from copper
ores/concentrates by alkaline leaching in NaHS media, Hydrometallurgy, 2009, 98, pp. 213
218.
Proceedings of Copper 2010 2897
Copper Leaching from Molybdenite in Acidic
FeCl
3
Solutions with FeCl
2
Yan Zhang, Narangarav Tumen-Ulzii, Zhibao Li
Chinese Academy of Sciences
Institute of Process Engineering
National Engineering Laboratory for Hydrometallurgical Cleaner Production Technology
Beijing 100190, China
Keywords: Molybdenite, copper, leaching
Abstract
This work aims to remove copper from molybdenite ore by leaching. The leaching process of
covellite and chalcopyrite was investigated in the media of FeCl
3
-FeCl
2
-HCl. The results showed
that leaching temperature, solid to liquid ratio and the concentration of FeCl
3
and FeCl
2
(or redox
potential) affected on the leaching significantly. Numerical simulation showed that the hydrogen
ion activity and the boiling point of the solution increases with the addition of FeCl
2
. Data analysis
using a shrinking core model suggested that the leaching process is controlled by chemical reaction
with activation energy of 79 kJ/mol.
1 Introduction
Molybdenum disulfide (MoS
2
) with the purity of more than 98 % is a good solid lubricant, and it is
used in automobile, aerospace and military field. Molybdenite concentrate, as usually marketed,
contains the mineral MoS
2
, along with silicious material, iron, copper, and other impurities [1].
Copper is an undesired impurity in molybdenite concentrate. Selective leaching of copper is the
premise to get high purity MoS
2
.
Copper occurs mainly as covellite (CuS), chalcopyrite (CuFeS
2
), chalcocite (Cu
2
S) and cuprite
(CuO) in molybdenite ore. Many approaches have been reported for the leaching of copper from
molybdenite concentrate, including leaching with dichromate-sulphuric acid mixture [2], HF & HCl
[3], FeCl
3
in acidic solutions [4] and other chemicals
.
Usually, both oxidizing agent and acid are
included in these methods. The oxidizing agents commonly used for the leaching of copper include
Zhang, Tumen-Ulzii, Li
Proceedings of Copper 2010 2898
Cu
2+
, Fe
3+
, O
2
, Cl
2
and so on [5]. Among them, Fe
3+
is more efficient and can be recovered more
easily than others. When it comes to acid, HCl and H
2
SO
4
are commonly used in copper leaching and
HCl made the leaching more easily than H
2
SO
4
[6, 7].
FeCl
3
and HCl were selected as the oxidizing agents and acid respectively in the leaching solutions
for this work. Moreover, FeCl
2
was also added to the solution. On one side, FeCl
2
can increase the
activity of H
+
, on the other, it can increase the boiling point of the solution [8, 9]. Furthermore, the
redox potential, determined by [Fe
3+
] / [Fe
2+
] ratio, decreases with FeCl
2
addition.
Mineralogical analysis showed that CuS and CuFeS
2
were the main impurities in the molybdenite
concentrate used in this work. The objective of the present work is to remove these copper
sulphides from molybdenite concentrate. Experiments were carried out to investigate the effects of
temperature, solid to liquid (S/L) ratio, and concentration of Fe
3+
and Fe
2+
. The kinetic of the copper
extraction were also studied.
2 Experimental
2.1 Materials
The molybdenite concentrate used in this work was obtained from the Erdenet Mine of Mongolia.
The particle size of the concentrate was from 200 to 500 mesh. The chemical composition
determined by X-Ray Fluorescence (XRF) is shown in Table 1. The phase identification and image
analysis were achieved by polarizing microscope and scanning electron microscope-energy
dispersive spectrometry (SEM-EDS). Polarizing microscope micrograph in Figure 1 combined with
SEM-EDS analysis in Table 2 showed that the main phases of the concentrate were MoS
2
, and the
impurities were CuS, CuFeS
2
and SiO
2
.
Copper Leaching from Molydenite in Acidic FeCl
3
Solutions with FeCl
2
Proceedings of Copper 2010 2899
Table 1: The elements content of the molybdenite concentrate by XRF
Chemical species Content (% w/w) Chemical species Content (% w/w)
Mo 50.09 Ca 1.78
S 37.01 Mn 0.06
Fe 4.30 Al 0.09
Cu 2.71 K 0.34
Si 1.96 Mg 0.34
Pb 0.07 Cl 0.41
Figure 1: The phases of molybdenite concentrate investigated by polarizing microscope
(1-CuS, 2-SiO
2
, 3-MoS
2
, 4-CuFeS
2
, 5-MoS
2
)
2.2 Methods
Leaching experiments were carried out in a round bottom glass reactor of 500 ml. The leaching
experiments were done under the following conditions: leaching time, 04 h; temperature,
70100 C; solid / liquid ratio, 0.1-0.35 kg/L; the concentration of FeCl
3
, 100-200 g/L; the
concentration of FeCl
2
, 200-300 g/L; stirring speed, 300 rpm; and pH=0. The leaching solution was
filtrated and analyzed by atomic absorption spectrophotometer (AAS).
Zhang, Tumen-Ulzii, Li
Proceedings of Copper 2010 2900
Table 2: The SEM-EDS analysis of the position 1-5 in Figure 1
Elements
%
Position
O Si S Fe Cu Al Mo
1 1.14 0.36 33.05 0.36 65.13 - -
2 57.87 39.55 2.22 0.35 - - -
3 - - 46.73 - - 0.57 52.7
4 2.81 0.37 36.79 28.73 31.04 0.26 -
5 5.36 - 45.8 0.56 - 0.44 47.84
3 Results and discussion
3.1 Eh-pH equilibrium diagrams and electrochemical mechanism
Figure 3 and Figure 4 present Eh-pH equilibrium diagrams generated by chemical reaction and
equilibrium software HSC chemistry at 25 C for Cu-Fe-S-H
2
O and Fe-Cl-H
2
O systems. From
Figure 3 it can be seen that both CuS and CuFeS
2
dissolve at pH lower than 4. A comparison of
Figure 4 and Figure 3 shows the redox potential of Fe
3+
is much higher than that of CuFeS
2
and
CuS, and can dissolve them at pH lower than 0. So FeCl
3
solutions with pH=0 were prepared as the
leaching solution.
Equation (1) is reported to represent the dissolution of chalcopyrite in ferric chloride [10]. Covellite
dissolution in acidic FeCl
3
solutions can be represented by Equation (2) [11]. Consequently, the
oxidation proceeds according to an electrochemical mechanism (Equations (3), (4) and (5)) occurs
at the anode parts of the mineral, and iron reduction occurs at the cathode parts.
S FeCl CuCl FeCl CuFeS 2 5 4
2 2 3 2
+ + = + (1)
S FeCl CuCl FeCl CuS + + = +
2 2 3
2 2 (2)
Copper Leaching from Molydenite in Acidic FeCl
3
Solutions with FeCl
2
Proceedings of Copper 2010 2901
Anode:
0 2
2 S Cu e CuS + =
+
(3)
0 2 2
2
4 S Fe Cu e CuFeS + + =
+ +
(4)
Cathode:
+ +
= +
2 3
Fe e Fe (5)
Figure 3: Eh-pH diagram of the Cu-Fe-S-H
2
O system at 25 C
14 12 10 8 6 4 2 0
2.0
1.5
1.0
0.5
0.0
-0.5
-1.0
-1.5
-2.0
Cu - Fe - S - H2O - System at 25.00 C
C:\HSC5\EpH\CuFeS25.iep pH
Eh (Volts)
H 2 O L i mi t s
CuFeS2
CuS
Cu(OH)2
Cu(+a)
Cu(+2a)
Zhang, Tumen-Ulzii, Li
Proceedings of Copper 2010 2902
Figure 4: Eh-pH diagram of the Fe-Cl-H
2
O system at 25 C
3.2 Effect of leaching temperature
The effect of temperature on the copper extraction was studied over the range between 70-100 C,
at pH 0 and S: L = 1: 10. The concentrations of FeCl
3
and FeCl
2
in the leaching solutions are 100
and 300 g/L respectively. The leaching curves in Figure 5 indicate a marked effect of temperature
on the copper extraction. It can be observed that copper extraction increased from 31.9 % to 63.1 %
as the temperature increased from 70 C to 100 C after leaching for 4 hours.
14 12 10 8 6 4 2 0 -2
2.0
1.5
1.0
0.5
0.0
-0.5
-1.0
-1.5
-2.0
Fe - Cl - H2O - System at 25.00 C
C:\HSC5\EpH\FeCl25.iep pH
Eh (Volts)
H 2 O L imit s
Fe
Fe2O3
Fe3O4
Fe(OH)2
Fe(+3a)
Fe(+2a)
Copper Leaching from Molydenite in Acidic FeCl
3
Solutions with FeCl
2
Proceedings of Copper 2010 2903
Figure 5: Effect of temperature on the copper extraction in the ferric chloride leaching of
molybdenite concentrate (pH = 0, S: L = 1:10, FeCl
3
concentration: 100 g/L,
FeCl
2
concentration: 300 g/L)
3.3 Effect of solid: liquid ratio
The effect of solid: liquid (mass/ volume) ratio from 1:2.5 to 1:10 on copper extraction was studied
using an initial FeCl
3
concentration 200 g/L, FeCl
2
concentration 300 g/L, and temperature 100 C.
It can be seen from Figure 6 that the leaching rate gradually increases in copper leaching process
with the decrease of solid: liquid ratio, and the copper extraction of leaching with S:L = 1:10 is
almost 20 % higher than leaching with S:L = 1:2.5 after leaching of 4 hours. This may be because
of that the decrease in solid: liquid ratio enhances medium diffusion and makes the reaction more
sufficient [12].
0 50 100 150 200 250
0
10
20
30
40
50
60
70
C
o
p
p
e
r
E
x
t
r
a
c
t
i
o
n
(
%
)
Leaching Time (min)
70
0
C
80
0
C
90
0
C
100
0
C
Zhang, Tumen-Ulzii, Li
Proceedings of Copper 2010 2904
Figure 6: Effect of S: L on the copper extraction in the ferric chloride leaching of molybdenite
concentrate (pH=0, 100 C, FeCl
3
concentration: 200g/L, FeCl
2
concentration: 300g/L)
3.4 Effect of concentration of ferric chloride and ferrous chloride
The leaching of copper is an oxidation process for both chalcopyrite and covellite. Several
researchers have reported that the chalcopyrite dissolution rate depends on the redox potential of the
solution, and that the best result is achieved under moderately oxidizing conditions [13]. The
leaching of covelite would be also affected by the redox potential. In FeCl
3
FeCl
2
leaching
system, the initial redox potential of the leaching solution can be controlled by FeCl
3
/ FeCl
2
ratio,
according to the Nernst equation (Equation 6),
) (
) (
log 059 . 0 67 . 0
2
3
+
+
+ =
Fe a
Fe a
E
(6)
As shown in Figure 7, at a constant concentration of FeCl
2
(300 g/L), copper extraction increased
with increasing FeCl
3
concentration. And it can be seen in Figure 8 that copper extraction decreased
with increasing FeCl
2
concentration at a constant FeCl
3
concentration (100 g/L).
0 50 100 150 200 250
0
10
20
30
40
50
60
70
80
C
o
p
p
e
r
E
x
t
r
a
c
t
i
o
n
(
%
)
Leaching Time (min)
S:L=1:10
S:L=1:5
S:L=1:2.5
Copper Leaching from Molydenite in Acidic FeCl
3
Solutions with FeCl
2
Proceedings of Copper 2010 2905
Figure 7: Influence of FeCl
3
concentration on the copper extraction in the ferric chloride leaching
of molybdenite concentrate (pH=0, S:L=1:10, 100 C, FeCl
2
concentration: 300 g/L)
Figure 8: Influence of FeCl
2
concentration on the copper extraction in the ferric chloride leaching
of molybdenite concentrate (pH=0, S:L=1:10, 100 C, FeCl
3
concentration: 100 g/L)
0 50 100 150 200 250
0
10
20
30
40
50
60
70
80
200 g/L FeCl
3
- 300 g/L FeCl
2
100 g/L FeCl
3
- 300 g/L FeCl
2
C
o
p
p
e
r
E
x
t
r
a
c
t
i
o
n
(
%
)
Leaching Time (min)
0 50 100 150 200 250
0
10
20
30
40
50
60
70
80
C
o
p
p
e
r
E
x
t
r
a
c
t
i
o
n
(
%
)
Leaching Time (min)
100g/L FeCl
3
- 300g/L FeCl
2
100g/L FeCl
3
- 200g/L FeCl
2
Zhang, Tumen-Ulzii, Li
Proceedings of Copper 2010 2906
3.5 Synergetic effect of FeCl
2
The synergetic effect of FeCl
2
was studied by numerical simulation using OLI Systems Stream
Analyzer software package [9]. As can be seen in Figure 9, the hydrogen ion activity (a
H
+
)
increases rapidly with the increase of the concentration of FeCl
2
. The value of a
H
+
enhanced more
than two times duo to the addition of 3 mol/ kg FeCl
2
. Then the efficiency of the leaching can be
enhanced when FeCl
2
is added into the leaching solution. Moreover, the addition of FeCl
2
can
increase the boiling point as shown in Figure 10. The curves in Figure 5 indicate that higher copper
extraction occurs at higher temperature. So the optimum leaching temperature is near the boiling
point of the solution, and increase in the boiling point of the solution can improve the leaching
efficiency of copper. As can be seen in Figure 10, the boiling point of the leaching solution is lower
than 102 C without FeCl
2
, but it increases to more than 107 C when 3 mol/ kg of FeCl
2
has been
added.
Figure 9: Influence of concentration of FeCl
2
on a
H
+
in a solution of HCl (0.1 mol/ kg) +
FeCl
3
(1 mol/ kg) + FeCl
2
(0-3 mol/ kg) at 25 C with the aid of OLI Systems Stream
Analyzer software
0.0 0.5 1.0 1.5 2.0 2.5 3.0
0.08
0.10
0.12
0.14
0.16
0.18
a
H
+
(
m
o
l
/
k
g
)
Concentration of FeCl
2
(mol/ kg)
Copper Leaching from Molydenite in Acidic FeCl
3
Solutions with FeCl
2
Proceedings of Copper 2010 2907
Figure 10: Influence of concentration of FeCl
2
on the boiling point of the solution of HCl
(0.1 mol/kg) + FeCl
3
(1 mol/kg) + FeCl
2
(0-3 mol/kg) with the aid of OLI Systems
Stream Analyzer software
3.6 Kinetic studies
The leaching kinetic was studied based on a shrinking core model using the experimental data
shown in Figure 5. As shown in Figure 11, the dissolution of copper was found to follow Equation
7 [10, 14],
t k X
s
=
3 / 1
) 1 ( 1
(7)
RT Ea
Ae k
/
= (8)
RT E A k
a
/ ln ln = (9)
where X is the fraction of leached copper amount at time t (min), and k
s
is the apparent rate constant
(1/min). This result implies that the leaching process is controlled by surface chemical reaction.
Based on Arrhenius equations (Equation 8 and Equation 9), an activation energy was evaluated as
79 kJ/mol from the Arrhenius plot (Figure 12).
0.0 0.5 1.0 1.5 2.0 2.5 3.0
101
102
103
104
105
106
107
108
B
o
i
l
i
n
g
p
o
i
n
t
(
C
)
Concentration of FeCl
2
(mol/ kg)
Zhang, Tumen-Ulzii, Li
Proceedings of Copper 2010 2908
Figure 11: Variation of 1-(1-X)
1/3
over time at various temperatures
Figure 12: Arrhenius plot for the dissolution of copper from molybdenite concentrate based on the
apparent rate constants calculated from Figure 5
40 60 80 100 120 140 160 180 200 220 240 260
0.104
0.112
0.120
0.128
0.136
0.144
0.152
0.160
0.168
0.176
0.184
70 C
80 C
90 C
1
-
(
1
-
x
)
1
/
3
Time (min)
2.74 2.76 2.78 2.80 2.82 2.84 2.86 2.88 2.90 2.92
8.0
8.2
8.4
8.6
8.8
9.0
9.2
9.4
9.6
9.8
-
L
n
k
(
s
-
1
)
T
-1
/10
-3
(K
-1
)
Copper Leaching from Molydenite in Acidic FeCl
3
Solutions with FeCl
2
Proceedings of Copper 2010 2909
4 Conclusions
In this work, the dissolution process of covellite and chalcopyrite in molybdenite has been
investigated. It was found that dissolution rate of them increased with an increase in temperature,
and the optimum leaching temperature was near the boiling point of the solution. The S: L ratio
affect on the leaching since relatively less solid made the diffusion more easily. The initial redox
potential of the leaching solution was another important factor for the leaching, which can be
controlled by FeCl
3
/FeCl
2
ratio. The addition of FeCl
2
increased the hydrogen ion activity of the
solution and the boiling point of the solution. The kinetic study using a shrinking core model
suggested that the copper extraction rate is controlled by the surface reaction with activation energy
of 79 kJ/ mol.
Acknowledgements
The support of the National Nature Science Foundation of China (Grant No.20876161) and the
National Basic Research Program of China (973 Program, 2007CB613501) are gratefully
acknowledged.
References
[1] Byung-Su K., Manis K.J., Jinki J., Jae-chun L.: Leaching of impurities for the up-gradation of
molybdenum oxide and cementation of copper by scrap iron, Int. J. Miner. Process., 2008, 88,
pp. 7-12.
[2] Ruiz, M.C., Padilla, R.: Copper removal from molybdenite concentrate by sodium dichromate
leaching, Hydrometallurgy, 1998, 48 (3), pp. 313-325.
[3] Mankhand, T.R., Prasad P.M.: Lime enhanced hydrogen reduction of molybdenite.
Metallurgical Transactions, 1982, 13, pp. 275-282.
[4] Ikumapayi. F. K.: Purification of molybdenite concentrates, Master Thesis of Lulea
University of Technology, 2008.
[5] Herreros O., Vinals J.: Leaching of sulfide copper ore in a NaCl - H
2
SO
4
-O
2
media with acid
pre-treatment, Hydrometallurgy, 2007, 89, pp. 260-68.
[6] Lu Z.Y., Jeffrey M.I., Lawson F., An electrochemical study of the effect of chloride ions on
the dissolution of chalcopyrite in acidic solutions, Hydrometallurgy, 2000, 56, pp. 145-155.
Zhang, Tumen-Ulzii, Li
Proceedings of Copper 2010 2910
[7] Lu Z.Y., Jeffrey M.I., Lawson F., The effect of chloride ions on the dissolution of
chalcopyrite in acidic solutions, Hydrometallurgy, 2000, 56, pp. 18920.
[8] Demopoulos G. P., Zhibao L., Becze L., Moldoveanu G., Cheng T., Harris G. B.: New
technologies for HCl regeneration in chloride hydrometallurgy. ERZMETALL, 2008, 61, pp.
89-98.
[9] Demopoulos G. P., Zhibao L., Becze L., Moldoveanu G., Cheng T., Harris G.B.: New
technologies for HCl regeneration in chloride hydrometallurgy european metallurgical
conferenceEMC 2007, June 11-14, 2007, Dusseldorf, Germany.
[10] Mohammad A. H., Sam K., Adnan A. H.: Ferric chloride leaching of chalcopyrite: synergetic
effect of CuCl
2
, Hydrometallurgy, 2008, 91, pp. 8997.
[11] Antonijevic M.M., Dimitrijevic M.D., Stevanovic Z.O., Serbula S.M., Bogdanovic G.D.:
Investigation of the possibility of copper recovery from the flotation tailings by acid leaching,
Journal of Hazardous Materials, 2008, 158, pp. 23-3.
[12] Zeng L., Yongcheng C.: A literature review of the recovery of molybdenum and vanadium
from spent hydrodesulphurization catalysts Part II: Separation and purification,
Hydrometallurgy, 2009, 98, pp. 10-20
[13] Cordob E.M., Munoz J.A., Blazque M.L., Gonzale F., Ballester A.: Leaching of chalcopyrite
with ferric ion. Part II: Effect of redox potential, Hydrometallurgy, 2008, 93, pp. 88-96.
[14] Cordob E.M., Munoz J.A., Blazque M.L., Gonzale F., Ballester A.: Leaching of chalcopyrite
with ferric ion. Part III: Effect of redox potential on the silver-catalyzed process,
Hydrometallurgy, 2008, 93, pp. 97-105.
Proceedings of Copper 2010 2911
Recycling
Proceedings of Copper 2010 2912
Proceedings of Copper 2010 2913
Shifting Core Business Vision: From Copper to
Polymetallics A Recycling Point of View
Juan Ignacio Barturen Zabala
Codelco Chile
Once Norte 1291. Edificio Corporativo Codelco Norte
Calama, Chile
Keywords: Recycling, polymetallics
Abstract
It is everybody knowledge that mining business is a time and site restricted business due to ore
grade and technology availability to extend to a maximum its profits. This ore grade constant de-
crease is somehow balanced by both technologies breakthrough that allow recovering targeted ele-
ments (i.e. copper) from lower grade ores, and the possibility to recover others minerals that may
slow down cash costs growing through subproducts commercializing.
Chuquicamata mine is not an exception, and given its near 100 years of production life, lower
grades are no minor subject. Although almost every copper related recovery technology is being
used, or had being used, and several arsenic confining technologies had being developed to keep
producing copper in spite of this impurity; higher costs, human resource problems, and lower pro-
duction rates are unavoidable in the short term, and will keep increasing its effects. Something must
be made to keep Chuquicamata a profitable business.
In the last 10 years many social transformation had occur and still surprise us with new forms of
communication, science develops, etc., but there is one relevant change (to mine business) that
should shift how do we think the whole mining business: Recycling.
But recycling not just as a secondary business for better respect the environment and avoid danger-
ous waste, but a main crucial shift in how we evaluate our ores and products and tails.
How about to achieve commercialize every last atom of every element founded, whichever these
are, wherever flow they concentrated in the cooper process? How about to avoid waste?
Some by-products recovery had point in this direction, Molybdenum mainly, and also Rhenium,
Silver and Gold from anodic slug, Sulphuric acid and so on, but because those products are in such
high concentrations that no to do so was unthinkable, a very bad business.
Barturen
Proceedings of Copper 2010 2914
But, what about the rest of the elements contained in tails, slugs, intermediate process flows and the
like? What if there is more to recover (and profit)? To answer this is both a commercial and tech-
nological matter in the first place, is there a market for those elements? Is possible to recover those
elements without interfering the core business?
Nevertheless those questions are crucial for the success of such a challenge, first we have to face
one main, primary, foremost question: is this endeavor relevant for business? Is there any interest in
undertake this strategy? Some guidelines for these questions should be revealed.
1 Introduction
In this first XXI century decade many changes had happened, as to the start of this new millennium,
the complete deploy of the new economy of internet [1, 2], the arise of Chine and India as a
worldwide player [3], the loss of confidence in the economy thesis that build the subprime crisis [4],
the ever growing agreement in the Climate warming and its consequences, the ecology concept get
into our lives, among others changes had had a big impact in how we see and understand our world.
This changes had occur in every aspect of our lives, how we communicate between each other, how
companies treat their employees, buy theirs supplies, distribute their products. The supply chain
management have developed to a proper business by its own [5]; marketing had evolved from sell-
ing to bonding firm to the customer, and so on. This implies that companies had included many as-
pects into their strategy not evaluated before, bringing the theory of the stakeholders to front be-
cause profit and costs aren't the only aspects to analyse firms and the conclusions of such analysis
aren't enough to understand and withstand environmental changes and challenges.
But is the ecologic view of the world an aspect that is crucial now-a-days to the mining business,
and is going to increase its importance. From Lean production [6], to the Factor 4 [7] and other en-
vironmental thesis there is one point of agreement: recycling is not an option. Is a must in a world
deep concerned in maximize the output of the production factors: land, capital, labour, technology;
without any environmental cost or ambient damage.
Then, to better profit our resources is not just an enterprise policy, but a main goal for our society
due to the scarcity of them (especially of the minerals that allow built the world as we know it) and
to avoid the impact on the environment that waste produce.
2 Environment factors
In 2009, the high prices cycle that started on 2004 come to an end, and in some cases, a very painful
end. Many conclusions have been written these months regarding this, but also so many agree that if
Chine (and India) keeps its growing momentum, the world mining industry would be safe, or less
Shifting Core Business Vision: From Copper to Polymetallics
Proceedings of Copper 2010 2915
damaged. And this have been validated with facts, as most metal industry had not suffer any deep
damage, and most had recover their dynamism from 2
nd
Q of 2009.
But this might disguise the real problem or lessen the effects over an industry that had got into
structural problems everywhere. Mining operation faces all around the world supply problems,
mainly supply of water, energy and qualified workers. Although water & energy problems aren't
easy to resolve and may block the mine development and extraction, is the labour factor once wa-
ter and energy are solved the biggest challenge to keep mining business running. One main reason
for this is the location for these operations as they are remote located, far away from industry end
users, big cities and several of them are in deserts, jungles or the like. So, there are two ways to ad-
dress this problem: bring enough well prepared workers to mine sites, or prepare from the locals the
needed working force. Both alternatives faces the same problem in dealing with people so mining
facilities more and more depends on technology to diminish worker's lack of preparation.
Moreover, this highly specialized work force, as they became aware of their importance, put another
problem to the struggling mining industry: the strike menace. This fact, plus the ever lowering ore's
grade put forward one simple question, how can industry maximize the output of their labour fac-
tor?
On one hand, there is the possibility of, through education, increase the productivity of workers,
giving them tools for innovations in their work place that could led to patenting and knowledge, and
preparing them well enough in order to avoid work accidents.
On the other hand, technology allow to overcame this by developing process control, robots, intelli-
gent vehicles, energy efficient equipment, etc, that allow companies to lower demands of workers
unions and to keep controlled the payroll. In this sense, the core-business philosophy had an impor-
tant role in disaggregate large companies into single-focus, less integrate firms, and consequently,
less complex to manage the organization.
The industry struggles to keep running its operations despite these problems and restrictions, and
have found some relief focusing on their main business, but as environmental factor keep getting
stronger, and prices for every metal stays pushing, there must be a solution for these supply prob-
lems and a turnaround over the core-business vision.
3 The core-business/integration dilemma
Delocalization is one of the main effects over world economy of following core-business strategy as
the idea of focusing only in what make us different, by allowing other industries to emerge (ser-
vices mainly) in any country, as long as this choice bring cost and quality advantages.
To mining industry this also have been true, as companies define their business and got rid of any
operation not directly related to their core-business, and in some others respects it didn't, for one
reason mainly: you can not delocate mines and processing facilities. But mining and mineral proc-
Barturen
Proceedings of Copper 2010 2916
essing firms had deploy the core-business philosophy (with some exceptions) focusing in their main
products, which results in copper companies only producing copper, steel companies only concern-
ing in iron and so on. This had act as a blindfold to strategic vision.
From early 90's, the core business concept drove all industries to a narrow view of their relation to
the customer and the rest of the world. This concept have had a big impact in how mining compa-
nies think themselves, as many start to sell their secondary business to better focus on their core-
business. For sub product's development these have been disastrous as many plants had to close, and
no more integration projects have been developed. For porphyry copper mines, which characterized
for having many metals contained among the ore, these means that tails gets lost with no value add-
ed but water recovering (if mine is in a dry zone). Tailing dams are now even an environment prob-
lem due to the foreclose actions needed and soil remediation.
Many questions may arise from this fact: is not possible to lessen this problem? Is there any alterna-
tive? What else can be done from these materials?
The last question is very significant because the answer may lead to the conclusion that something
else can be done in terms of better profit from these wasted resources, which for certain have some
value due to its metal contents. It may be argued that the Fe, Si and other elements contained in tails
and the like have not the value that copper has, or that there is no technology available for recover
these elements economically, but it's also true that need brought always new forms to adjust industry
to new paradigms.
Figure 1 shows a typical copper processing operation, on the left hand, the sulphur processing line,
and on the right side, the oxide processing line. Is possible to state that -for a given amount of re-
sources located on mine site- there is a limited production capacity for the targeted element, as well
as limited technology availability for this element to be recovered. So, for the question of what else
can be done, it's imperative to look down to the waste, garbage flows produced by recovering
copper in this case. Several by-products are being recovered in some facilities from anodic Slug,
and by concentrating Sulphuric acid, for example, but once again arise the question of what else can
be done.
Shifting Core Business Vision: From Copper to Polymetallics
Proceedings of Copper 2010 2917
Figure 1: Copper processing description
There are two kinds of waste generated, one related to different flows and materials discarded by
mineral processing operations; and stocks, related to large amount of materials dispose as garbage.
So different strategies could be follow in recovering valuable elements from these wastes. The for-
mer waste, named discarded flow, could lead to discrete modules that will extract other sub-
products depending on which element is available without interfering core normal operation.
The latter, named stock consists in large deposits of tails and graves, which historically are being
dumped with no further use, but nevertheless contain valuable elements waiting to be extracted.
Then, its possible to think that new breakthroughs could follow the breakdown of the core-business
paradigm regarding new visions of what mining industry its all about, and where to find new value
in mining disregarded wastes.
The steel industry have seen these lasts years how big companies as ArcelorMittal start to integrate
new operations to protect themselves from price volatility, and supply shortcuts for their energy
(mainly coal) and special metals needs for steel alloys. This is the opposite direction followed by the
same company (both Arcelor and Mittal, previous the merge) that in previous years had sell some
coal mines, iron-rich sands, and minor business (as related to Steel production) following the core-
business drive, but facing the same problem of loosing operation flexibility, business stability, and
mainly, loosing strategic business opportunities related to supply or commercial ventures that could
allow to bypass as this year made clear- or lessen the effects of a worldwide economic crisis.
Furthermore, imagine the possibilities for mining industry in developing new technologies to re-
cover more minerals form the same ore, the impact of such an endeavour may lead to a complete
revolution not only in the know-how and knowledge management assets of the firm, but in getting
more from the labour factor, and then, increasing many times the financial output (i.e., stock value)
by getting into business accountability resources now labelled as waste.
Barturen
Proceedings of Copper 2010 2918
4 The no-waste business policy
As had been stated, is no longer possible to keep wasting so many resources both from an economic
and environmental point of view. So a major shift in the business vision is required. The proposition
is to a no waste policy that will (almost automatically) lead to a polymetallic production to better
profit the resources available in situ on the ore and mines worldwide.
The propose policy tends to unite both environmental and economic objectives into one main enter-
prise goal of increase value for the company and avoid environmental damages to earth.
For certain there are several restrictions in what can be done to follow this policy, among all is the
technology availability what came in first place, as is imperative this technology to achieve the ob-
jective of a polymetallic production evolution from a single-metal facility. Yes, economic reasons
had determined why some elements are recovered and others don't, but these economic reasons are
mainly defined by the technology that makes it possible and its operating costs.
Some guidelines for doing this may be followed:
Broad scope for metal recovery.
Analyze and quantified metal content in ore, tails, tails dam, graves and any waste deposit.
Screen metal occurrence for better focus on subproduct development.
Search for technology availability.
Design a proper business model for processing these products.
Adjust financial aspects for capital investing and technology development.
Encourage technology innovations.
Figure 2 describes the proposed no-waste policy process, the organizational responsible and the ex-
pected outcome for each step.
Shifting Core Business Vision: From Copper to Polymetallics
Proceedings of Copper 2010 2919
Figure 2: The No-Waste Policy
The target for a less volume of waste to dispose is achieve by a selective process intended to deter-
mine the economic potential to be recover by processing wasted materials, and the technical feasi-
bility to make this possible. Two main products could be defined, a less volume of waste to dispose,
which means a less costly disposal operation; and the innovation acquired by this search for new
metallurgic and chemical processes.
These guidelines are not exhaustive, and are yet to go through management approval, but the prima-
ry idea is to design a business strategy that, in the long term, allow the industry shift to a polymetal-
lic production intended to avoid any waste, given that every atom of element that goes through min-
eral processing had consume energy, water, chemicals and other resources, which in the long term
would be unacceptable to loss.
Maybe is 'common sense' that mineral processing only focused on what is profitable, disregarding
the rest, but is also true that 'common sense' isn't static, but is gradually increased by science and
technology [8]. Then, although now it seems that is not an option to recovers every element from
Barturen
Proceedings of Copper 2010 2920
the ore, this might easily change if some technology breakthrough happened, or the economic value
of this wasted elements increase to a certain level enough for allowing their recovery without eco-
nomic penalty.
This policy is meant to prepare the organization for seek and develop actions that would lead to a
polymetallic mineral processing, but is not a recipe to undertake this change, because this shall be
restricted by technology advances yet to come.
Maybe to declare that polymetallic is the future of the mining industry is a bold statement, but as
long as prices keep rocketing, ore's grades keep lowering and metal's world demand stay as strong as
consumer technology envisions new uses and aids to our lives, to dump any metal once crushed,
grind, floated, lixiviated or whatever process undertake will be a luxury that none would be able to
afford.
5 Conclusion
Our society will face major challenges in the near future regarding environmental problems, eco-
nomic paradigms changes, social stress and technology development. Most changes point in one
direction: to better use the resources available.
This means that not only climate global warming affect us as a distant not-so-clear menace, not only
the economy is changing its principles to avoid structural weakness, not only workers will become
more aware of their rights and not only technology will keep its trend of awesome advances; but
reassures that the scarcity concept contained in any extracting industry as mining and mineral
processing will be more and more relevant.
Organizations must be prepared for this change in how we value our resources, and mostly, how we
value waste; in order to keep its operations running in the long term. To do so, imply a major shift
in mining industry, that should drove them to maximize their productivity and resource recovery,
transforming -wherever possible- actual facilities into polymetallic processing operations that will
lead not only to a business economic burst, but also to lessen environmental impact.
Shifting Core Business Vision: From Copper to Polymetallics
Proceedings of Copper 2010 2921
References
[1] KELLY, K. (1999): New Rules for the New Economy, 16-18, New York.
[2] BLINDER, A.S. (2000): The Internet and the New Economy, Brookings Policy Brief, June,
N60, Washington.
[3] ENGARDIO, P. (2007): Chindia: How Chine and India are revolutionizing global business,
McGraw-Hill, New York.
[4] SOROS, G. (2008): The New Paradigm for Financial Markets: The Credit Crisis of 2008 and
what it means, Public Affairs, New York.
[5] MENTZER, J.T.; DeWitt, W; Keebler, J.S.; Min, S.; Nix, N; Smith, C.D.; Zacharia, Z.,(2001):
Defining Supply Chain Management, Journal of Business Logistics, Vol 22, N2.
[6] WOMACK, J.P.; JONES, D.T. (1996): Lean Thinking, Simon&Schuster, New York.
[7] Von WEIZSACKER, E.; LOVINS, A.B.; LOVINS, L.H., (1997): Factor Four: Doubling
Wealth Halving Resources, Allen&Unwin.
[8] BUNGE, M. (1996): Intuicin y Razn, p 152; Buenos Aires.
Proceedings of Copper 2010 2922
Proceedings of Copper 2010 2923
Solubility of Scorodite Synthesized by Oxidation
of Ferrous Ions
Tetsuo Fujita Etsuro Shibata, Takashi Nakamura
Dowa Metals & Mining Co., Ltd. Tohoku University, Institute of Multidisciplinary
Research for Advanced Materials
217-9 Shimo-kawabata Furumichi Iijima 2-1-1 Aoba-ku Katahira
Akita, Japan Sendai, Japan
Keywords: Scorodite, solubility, NaOH, NaCl, CaO
Abstract
An atmospheric scorodite synthesis process was developed, in which ferrous ions were oxidized by
oxygen gas in the presence of pentavalent arsenic ions. The synthesized scorodite was well-
crystallized in a short retention time of 1 to 7 h. The resulting scorodite particles were as large as
15 m in diameter, had a moisture content of less than 10 % were readily washable, and featured
excellent packing properties. This scorodite synthesis can be incorporated into a hydrometallurgy
process.
The solubility of scorodite synthesized via this novel atmospheric process was investigated. Envi-
ronmental batch leaching tests were performed according to the United States Environmental Pro-
tection Agencys Toxicity Characteristic Leaching Procedure (TCLP) test (method 1311), and the
Japanese Ministry of the Environment Notification No. 13 method, with minor modifications. In the
TCLP test, scorodite particles released a very low, almost negligible, concentration of arsenic in the
pH range 3 to 5, suggesting long-term stability. In leach tests with various pH solutions, it was
shown that the scorodite solubility was significantly affected by outside environmental factors such
as pH, rather than by long-term stability. In addition, the results showed that scorodite released high
concentrations of arsenic under specific leaching conditions. In particular, the presence of a combi-
nation of CaO and NaCl in alkaline leaching solutions had a significant effect on arsenic release.
Development of scorodite storage methods based on accurate evaluation of the conditions that cause
arsenic leaching, which will minimize the risk of environmental arsenic contamination, is crucial.
Fujita, Shibata, Nakamura
Proceedings of Copper 2010 2924
1 Introduction
Arsenic contamination is one of the worlds major environmental issues. Many methods have been
reported for the removal of arsenic from water environments [114]. Most reported methods involve
adsorption of arsenic on iron compounds and this phenomenon has long been a subject of interest
[1517].
Arsenic is present in ore impurities and occurs in the by-products of metallurgical processes. In the
past, arsenic compounds were used in pesticides, herbicides, fungicides, and wood preservers. How-
ever, because of the environmental impact of As
2
O
3
toxicity, many countries have banned the use of
arsenic-containing agricultural products and arsenic consumption has therefore greatly declined.
Thus, the production volume of arsenic far exceeds industrial demand, and the development of ef-
fective technologies for elimination, fixation, and storage of arsenic are highly sought after.
A novel atmospheric process for the synthesis of scorodite (FeAsO
4
2H
2
O) has been developed and
reported [1821]. The essential feature of this process involves precipitation of scorodite by oxidiz-
ing ferrous ions contained in a solution containing a high concentration arsenic (V). Scorodite crys-
tals release only very low or negligible levels of arsenic when tested according to the Japanese Min-
istry of the Environment Notification No. 13 method. However, this test result should not be
construed as the sole basis for judgment of scorodite stability. A wide range of experimental condi-
tions must be used for comprehensive analysis of scorodite leaching behavior.
Investigations into the leaching of arsenic have been widely reported in the literature. Nishimura and
Tozawa [23, 24] calculated the solubility products for FeAsO
4
and found that heat treatment en-
hanced its immobilization. Robins [25-29] conducted a series of studies on the solubility and stabili-
ty of scorodite, comparing the thermodynamic properties of FeAsO
4
and scorodite. Krause and Ettel
[30-32] examined the solubilities of ferric arsenate at various Fe/As molar ratios. Harris and Mo-
nette [33, 34] reported the solubilities of smelter/refinery-associated arsenic compounds, including
ferric arsenate. Emett and Khoe [35, 36] explored the effect of co-impurities such as cadmium and
lead on the solubility of arsenic. The recent work by Langmuir et al. [37, 38] provides new insights
into the thermodynamic behavior of scorodite. Bluteau and Demopoulos [39, 40] presented an ex-
tensive discussion of the mechanism of arsenic leaching from scorodite.
Although a large number of studies have been reported to date, they seem to fall short of establish-
ing the safest and most convincing methods for scorodite storage. This is primarily because of a lack
of evidence for long-term scorodite stability. The solubility is affected by time-dependent changes in
the stored scorodite environment, therefore research aimed at eliminating the risk of arsenic dissolu-
tion will be unlikely to yield robust conclusions regarding storage or disposal.
In this study, the solubility of crystalline scorodite synthesized by our novel atmospheric process
was investigated. Environmental leaching tests were performed according to batch leaching proce-
dures. The objectives of the present study were (1) to measure and compare the levels of arsenic
Solubility of Scorodite Synthesized by Oxidation of Ferrous Ions
Proceedings of Copper 2010 2925
leached from synthesized scorodite and (2) to examine the concentrations of arsenic leached from
scorodite into various aqueous solutions at various pH values to understand the scorodite solubili-
ties in various external environments.
2 Experimental
2.1 Materials
We used an arsenic (V) acid solution (labeled H
3
AsO
4
; content: 58-62 %; mean: 60 %). Analytical-
grade iron (II) sulfate 7-hydrate (FeSO
4
7H
2
O) was used as the iron source. All chemicals used in
the tests were purchased from Wako Pure Chemical Industries, Ltd. (Osaka, Japan). Compressed
oxygen gas (purity: 99.9 %, Akisan Kogyo K. K., Akita, Japan) was used as the oxidizing agent.
2.2 Synthesis of scorodite
The experimental procedure for the production of scorodite is shown in Figure 1. 4 L of arsenic test
solution (50 g/L) was measured and poured into a beaker, and a sufficient amount of FeSO
4
7H
2
O
was weighed and added to the solution to give an Fe/As molar ratio of 1.5. The total volume of so-
lution was 4 L. The beaker was placed on a heater with a titanium stirrer and the reaction mixture
was agitated at 200 rpm until it reached a pre-determined temperature. When the required tempera-
ture was attained, oxygen gas (4 L/min) was introduced into the mixture via a glass tube to begin
reaction, while stirring at 800 rpm. After the reaction had progressed for a pre-defined period of
time, part of the slurry was collected, cooled to 60 C, and its pH and oxidation reduction potentials
(ORP) were measured. The pH value was determined using manual temperature compensation
(MTC), set at 30 C. ORP was measured with an Ag/AgCl reference electrode, with a 3.3 mol/L
KCl filling solution. The remaining reaction slurry was pressure-filtered through a 1 mm pore-size
polytetrafluoroethylene (PTFE) membrane filter at 0.4 MPa to separate the solid from the liquid.
The filtrate was subjected to inductively coupled plasma (ICP) measurements.
The solid filter cake was re-suspended and washed three to five times with distilled water until the
pH of the wash water exceeded 3.9. The weight ratio of solid to distilled water was 1 to 10. After
each washing, the solid was separated from the liquid by pressure filtration. Finally, the solid was
dried at 60 C for 18 h and carefully pulverized in a mortar. The obtained solids were subjected to
chemical composition and powder characteristics analyses.
Fujita, Shibata, Nakamura
Proceedings of Copper 2010 2926
Figure 1: Procedure for scorodite synthesis
2.3 Analytical methods
ICP analyses were performed on the solid and on the post-reaction filtrate using inductively coupled
plasma atomic emission spectroscopy (ICPAES) (ICAP577, Nippon Jarrell-Ash Co. Ltd., Kyoto,
Japan) to determine their chemical compositions. Scanning electron microscopy (SEM) was con-
ducted on solids using a Hitachi FE-SEM S-4500 (Hitachi Ltd., Tokyo, Japan).
2.4 Leach tests
2.4.1 Long-term leach tests
Long-term leach tests were performed according to the United States Environmental Protection
Agencys Toxicity Characteristic Leaching Procedure (TCLP) test (method 1311), with minor mod-
ifications. The TCLP leach test is usually carried out as follows. A 100 g of sample and 2 L of acetic
acid buffer solution (pH 2:88 or 4.95) are mixed in a polyethylene bottle. Extraction is performed by
rotary shaking for 18 h and solid-liquid separation is carried out using a 0.6-0.8 m pore-size glass
fiber filter (GFF). In this study, a 5 g (dry weight) sample was added to 100 g of extraction fluid in a
high-density polyethylene sample bottle; the bottle was placed horizontally, and the mixture was
shaken for a period of from 6 h to 35 days. The maximum duration of 35 days was based on data
reported by Bluteau and Demopoulos [39, 40], which showed the attainment of equilibrium in 5
Prepared solution 4 Liter
As content: 50 g/L
Fe(II)SO
4
.7H
2
O
Fe/As molar ratio: 1.5
Temp.: 95
o
C
Stirring rate: 800 rpm
O
2
gas flow: 4 L/min
Reaction time: 7 h
Reaction
Sampling
Pressure filtration
Precipitate
Washing
Scorodite powder Solution
Solubility of Scorodite Synthesized by Oxidation of Ferrous Ions
Proceedings of Copper 2010 2927
weeks. After a pre-determined time, the sample bottle was removed from the shaker, the mixture
was filtered through a 0.2 m PTFE membrane filter, and the filtrate was analyzed by inductively
coupled plasma mass spectrometry (ICPMS) (SPQ9000, Seiko Instruments Inc., Chiba, Japan).
In these tests, the following buffers were used for the extraction fluid: 0.1 mol/L aqueous acetic acid
(pH 2.88), a mixture of 0.1 mol/L acetic acid solution and 0.1 mol/L sodium hydroxide solution
(pH 4.93), a mixture of 0.025 mol/L potassium dihydrogen phosphate and 0.025 mol/L sodium di-
hydrogen phosphate (pH 6.86), and 0.01 mol/L sodium tetraborate (pH 9.18). The samples were
sealed in the bottle under a normal air atmosphere.
2.4.2 Solvent-specific leach tests
Leach tests using different extraction fluids were carried out according to the method stipulated by
the Japanese Ministry of the Environment, with several modifications. After a 10 g sample and
100 g of extraction fluid were weighed and mixed in a high-density polyethylene bottle, the bottle
was sealed, placed horizontally and shaken for 6 h. Upon completion, the mixture was filtered
through a 0.20 m PTFE membrane filter, and the filtrate was analyzed by ICP-MS. These experi-
ments employed the extraction fluids listed in Table 1.
Table 1: Solvent conditions used in the 6 h leach test
Condition of solvent pH
Distilled water 5.74
TCLP 2.88 (0.1mol/L CH
3
COOH solution) 3.00
TCLP 4.93 (0.1mol/L CH
3
COOH and NaOH solution) 4.88
Alkaline water conditioned by CaO 12.52, 12.30, 10.75, 10.14, 9.00
Alkaline water conditioned by Mg(OH)
2
10.48, 10.41, 10.64, 9.00
Alkaline water conditioned by NaOH 12.50, 11.06, 9.70
Acid water conditioned by H
2
SO
4
3.96, 2.97, 1.97, 1.22, 0.35
Acid water conditioned by HCl 3.99, 2.99, 1.98, 1.11, 0.13
Acid water conditioned by HNO
3
3.98, 2.99, 1.99, 1.12, 0.12
8.2 g/L NaCl solution conditioned by CaO 12.50, 12.00, 11.06
8.2 g/L NaCl solution conditioned by H
2
SO
4
5.80, 2.99
16.5 g/L NaCl solution conditioned by CaO 12.51, 12.00, 11.01
16.5 g/L NaCl solution conditioned by H
2
SO
4
5.55, 2.99
33 g/L NaCl solution conditioned by CaO 12.51, 12.00, 11.05
33 g/L NaCl solution conditioned by H
2
SO
4
5.47, 2.96
Fujita, Shibata, Nakamura
Proceedings of Copper 2010 2928
3 Results and discussion
3.1 Synthesis and characteristics of scorodite particles
Figure 2 presents SEM images of the scorodite precipitate. Large scorodite crystals with low mois-
ture content (11.2 %) were formed, yielding a high arsenic precipitation percentage. The final pH
and ORP of the synthesis solution were -0.22 and 474 mV (vs Ag/AgCl), respectively.
Tables 2 and 3 summarize the results of scorodite leach tests with the different extraction buffers.
The pH values given in the tables were taken immediately after shaking was complete (before solid/
liquid filtration and separation). These results show that the scorodite crystals were quite stable,
with arsenic leaching less than 0.5 mg/L in the pH range 3 to 5, though arsenic leached at a concen-
tration of 3 mg/L at pH 6.9, and at a concentration of 300 mg/L at pH 9. The value of 300 mg/L in-
dicates that about 2 % of the arsenic was released from the scorodite.
Figure 2: SEM images of precipitated samples
30m 6m
Solubility of Scorodite Synthesized by Oxidation of Ferrous Ions
Proceedings of Copper 2010 2929
Table 2: Long-term scorodite leach tests under a normal air atmosphere using buffer solutions
with pH values of 2.88 and 4.93
Stirring
days
0.1 mol/L CH
3
COOH 0.1 mol/L CH
3
COOH and NaOH
pH
As Fe S
pH
As Fe S
mg/L mg/L mg/L mg/L mg/L mg/L
0.25 2.88 0.19 0.70 <10 4.95 0.16 0.06 <10
1 2.90 0.24 0.90 <10 4.95 0.16 0.06 <10
2 2.90 0.28 0.93 <10 4.96 0.25 0.07 <10
3 2.90 0.38 1.09 <10 4.95 0.26 0.06 <10
5 2.91 0.32 1.04 <10 4.95 0.20 0.07 <10
7 2.88 0.31 1.03 <10 4.94 0.23 0.07 <10
14 2.90 0.34 1.14 <10 4.96 0.19 0.09 <10
21 2.91 0.50 1.33 <10 4.96 0.18 0.07 <10
28 2.90 0.43 1.39 <10 4.95 0.20 0.12 <10
35 2.89 0.40 1.23 <10 4.95 0.18 0.09 <10
Table 3: Long-term scorodite leach tests under a normal air atmosphere using buffer solutions
with pH values of 6.86 and 9.18
Stirring
days
0.025 mol/L KH
2
PO
4
and NaH
2
PO
4
0.01 mol/L Na
2
B
4
O
7
pH
As Fe S
pH
As Fe S
mg/L mg/L mg/L mg/L mg/L mg/L
0.25 6.92 1.05 <0.04 <10 9.26 4.18 0.60 <10
1 6.92 1.50 0.06 <10 9.20 32.3 9.11 <10
2 6.91 1.53 0.06 <10 9.19 40.3 13.3 <10
3 6.91 1.88 0.07 <10 9.15 71.3 28.1 <10
5 6.91 2.13 0.12 <10 9.04 137.0 54.9 <10
7 6.89 2.07 0.09 <10 9.08 117.0 42.5 <10
14 6.91 2.49 0.13 <10 8.84 253.0 15.6 10
21 6.91 3.00 0.11 <10 8.75 321.7 24.4 <10
28 6.90 2.77 0.17 <10 8.69 312.0 26.6 10
35 6.89 3.14 0.15 <10 8.66 351.7 26.3 <10
Tables 4 and 5 show the results of 6-hour scorodite leach tests conducted using various solvents.
The leach tests using buffer solutions with pH values of 2.88 and 4.93 were performed again in this
series of experiments and the results differ from the corresponding data reported above. Table 5
gives the results of leach tests that involved the presence of a given amount of NaCl. In this experi-
ment, CaO or H
2
SO
4
was used to adjust the pH.
Fujita, Shibata, Nakamura
Proceedings of Copper 2010 2930
When Mg(OH)
2
or CaO was used, the leached arsenic concentration was as low as approximately
3 mg/L under strongly basic conditions (pH 10 or higher), whereas arsenic leached at the higher
concentration of 10 mg/L in the pH range 8-10. The arsenic leaching concentration under alkaline
conditions using NaOH was as high as 99.2 mg/L. The liquid-solid ratio was 10:1 in this leach test
and it was calculated that about 0.3 % of the arsenic was released from the scorodite.
When HCl, H
2
SO
4
or HNO
3
was employed, the arsenic leaching concentration was less than 1 mg/L
at pH 3, and 2 mg/L at pH 2. However, when the pH was reduced from 1 to 0, the concentration of
leached arsenic increased considerably. The sequence of acids arranged in decreasing order of arsen-
ic leaching was HNO
3
, H
2
SO
4
and HCl.
For leach tests conducted in the presence of NaCl, the arsenic leaching concentration was 0.3 mg/L
or less under acidic to neutral pH conditions, though it exceeded 1000 mg/L (dissolved As ratio:
approximately 3 %) under basic conditions.
Table 4: 6-hour scorodite leach tests with various solvents
Solvent
Initial
pH
Final
pH
Leachate
Solvent
Initial
pH
Final
pH
Leachate
As Fe S As Fe S
mg/L mg/L mg/L mg/L mg/L mg/L
Water 5.74 4.61 0.01 0.56 <10 H
2
SO
4
3.96 3.99 0.05 0.91 <10
TCLP 2.88 3.00 3.02 0.11 1.48 <10 H
2
SO
4
2.97 3.03 0.21 1.37 20
TCLP 4.93 4.88 4.89 0.16 0.21 <10 H
2
SO
4
1.97 1.99 1.83 2.87 210
CaO 12.52 12.17 2.55 0.09 <10 H
2
SO
4
1.22 1.28 11.00 9.84 1630
CaO 12.30 12.10 2.20 0.11 <10 H
2
SO
4
0.35 0.42 94.74 78.04 15130
CaO 10.75 8.36 3.91 0.54 <10 HCl 3.99 3.98 0.04 0.97 <10
CaO 10.14 6.83 0.51 0.27 <10 HCl 2.99 3.04 0.13 1.68 <10
CaO 9.00 5.14 <0.01 0.24 <10 HCl 1.98 2.03 1.06 2.94 <10
Mg(OH)
2
10.48 10.03 1.66 0.05 <10 HCl 1.11 1.12 6.70 7.73 <10
Mg(OH)
2
10.41 9.88 8.59 <0.05 <10 HCl 0.13 0.15 202 166 <10
Mg(OH)
2
10.64 9.26 6.66 0.12 <10 HNO
3
3.98 3.99 0.05 0.98 <10
Mg(OH)
2
9.00 7.58 1.27 0.60 <10 HNO
3
2.99 3.01 0.08 1.72 <10
NaOH 12.50 9.46 99.20 60.50 30 HNO
3
1.99 2.04 0.78 2.90 <10
NaOH 11.06 8.28 21.20 5.43 <10 HNO
3
1.12 1.12 3.89 5.93 <10
NaOH 9.70 5.83 0.09 0.05 <10 HNO
3
0.12 0.16 29.34 26.11 <10
Solubility of Scorodite Synthesized by Oxidation of Ferrous Ions
Proceedings of Copper 2010 2931
Table 5: 6-hour scorodite leach tests with various solvents in the presence of NaCl
Solvent Initial pH Final pH
Leachate
As Fe S
mg/L mg/L mg/L
NaCl=8.2g/L 12.51 10.47 927 29.9 30
NaCl=8.2g/L 12.00 9.55 258 26 <10
NaCl=8.2g/L 11.06 8.46 25.80 0.07 <10
NaCl=8.2g/L * 5.80 4.61 0.04 0.58 <10
NaCl=8.2g/L * 2.99 3.05 0.18 1.54 20
NaCl=16.5g/L 12.51 10.52 1041 26.7 40
NaCl=16.5g/L 12.00 9.71 278 2.76 10
NaCl=16.5g/L 11.01 8.32 22.05 <0.05 <10
NaCl=16.5g/L * 5.55 4.49 <0.01 0.66 <10
NaCl=16.5g/L * 2.99 3.05 0.10 1.55 20
NaCl=33g/L 12.51 10.54 1164 21.9 50
NaCl=33g/L 12.00 9.85 309 2.54 10
NaCl=33g/L 11.05 8.36 26.63 <0.05 <10
NaCl=33g/L * 5.47 4.42 <0.01 0.68 <10
NaCl=33g/L * 2.96 3.03 0.30 1.61 20
: CaO was used for adjustment of pH.
*: H
2
SO
4
was used for adjustment of pH.
The results of leach tests in Tables 2-5 indicate that the solubility of scorodite depends strongly on
pH. The experimental results indicate that high pH values increase the concentration of arsenic ac-
cording to the following reaction;
FeAsO
4
2H
2
O + H
2
O = Fe(OH)
3
+ HAsO
4
2-
+ 2H
+
(1)
The above results (Tables 4 and 5) indicate that scorodite leached at only very low concentrations in
the pH range 3-6. The data also show that scorodite leached at higher concentrations at pH values
below 3, and that it leached at high concentrations when the solvent was made alkaline by adding
NaOH. Considering that under neutral to acidic pH conditions only low concentrations of arsenic
leached into solvents with high NaCl concentrations, we may conclude that alkaline pH conditions,
not Na
+
ions, contributed to the enhanced arsenic leaching.
Low concentrations of scorodite leached into solvents containing calcium and magnesium ions. This
finding is consistent with the solubility products reported by Nishimura and Tozawa [28] for cal-
cium arsenate and magnesium arsenate. One point worth noting is that the above study by Nishimu-
ra and Tozawa found that no calcium arsenate leached at pH 8-10, resulting in an apparent discre-
pancy with the results reported here. The discrepancy arises because the leach test in their study
measured the leaching of scorodite into a calcium-containing aqueous solution. It is presumed that
Fujita, Shibata, Nakamura
Proceedings of Copper 2010 2932
in the test for leaching into the calcium-containing alkaline solution reported in this article, scoro-
dite leached a low level of arsenate, which then reacted with calcium ions, present in abundance to
form calcium arsenate, resulting in a very low arsenic concentration. Similar reaction schemes prob-
ably dominated in the case of magnesium.
It was observed that addition of NaCl to the Ca-containing alkaline solution promoted leaching of
arsenic from scorodite at a level similar to or higher than that of the NaOH-containing solution. This
observation indicates that the effect of lime on the suppression of arsenic leaching is cancelled out
by the presence of NaCl. Sodium ions may react with arsenic ions to form sodium biarsenate, how-
ever the sodium biarsenate will dissolve easily in an alkaline solution. The sodium ions probably
enhanced the dissolution of arsenic from scorodite, with the chlorine ions present also being able to
react with the scorodite ferric ions.
The combination of NaCl and CaO may constitute the composition of fly ash from blast furnaces
recently developed at industrial waste treatment sites. The results of leach testing in this research
demonstrate the absolute necessity of separating the storage locations of fly ashes from blast fur-
naces for scorodite precipitates because these materials, when dumped together, constitute a high
risk of arsenic leaching.
4 Conclusions
Stable scorodite (FeAsO
4
2H
2
O) particles were produced by introducing oxidizing gas into a reac-
tion mixture containing ferrous sulfate and a high concentration of arsenic (V) to convert ferrous
ions to ferric ions. The solubility of scorodite synthesized by this novel atmospheric process was
investigated. The long-term solubility of the new scorodite product was the same as previously re-
ported results. In addition, it was shown that scorodite solubility was significantly influenced by
surrounding environmental factors such as pH rather than by long-term stability.
The results can be summarized as follows:
(1) When Mg(OH)
2
or CaO was used, the leached arsenic concentration was as low as approxi-
mately 3 mg/L under strongly basic conditions (pH 10 or higher), whereas arsenic leached at the
higher concentration of 10 mg/L in the pH range 8-10.
(2) The arsenic leaching concentration under alkaline conditions using NaOH was as high as
99.2 mg/L.
(3) When HCl, H
2
SO
4
or HNO
3
was employed, the arsenic leaching concentration was less than
1 mg/L at pH 3, and less than 2 mg/L at pH 2. However, when the pH was reduced from 1 to 0,
the concentration of leached arsenic increased considerably. The sequence of acids, arranged in
order of decreasing effect on arsenic leaching, was HNO
3
, H
2
SO
4
, and HCl.
(4) For leach tests conducted in the presence of NaCl, the arsenic leaching concentration was
0.3 mg/L or less under acidic to neutral pH conditions, though it exceeded 1000 mg/L under
basic conditions.
Solubility of Scorodite Synthesized by Oxidation of Ferrous Ions
Proceedings of Copper 2010 2933
References
[1] WILKIE J.A., HERING J.G. (1996): Colloids Surf., A 107: 97110
[2] MANNING B.A., GOLDBERG S. (1996): Soil Sci. Soc. Am. J., 60: 121131
[3] JAIN A., RAVEN K.P., LOEPPERT R.H. (1999): Environ. Sci. Technol., 33: 11791184
[4] MENG X., BANG S., KORFIATIS G.P. (2000): Water Res., 34: 12551261
[5] GAO Y., MUCCI A. (2001): Geochimica et Cosmochimica Acta, 65: 23612378
[6] DIXIT S., HERING J.G. (2003): Environ. Sci. Technol., 37: 41824189
[7] KATSOYIANNIS I., ZOUBOULIS A., ALTHOFF H., BARTEL H. (2002): Chemosphere, 47: 325332
[8] AHN J.S., CHON C.M., MOON H.S., KIM K.W. (2003): Water Res., 37:24782488
[9] FUKUSHI K., SATO T., YANASE N. (2003): Environ. Sci. Technol., 37: 35813586
[10] MUNOZ J.A., GONZALO A., VALIENTE M. (2002): Environ. Sci. Technol., 36: 34053411
[11] GIASUDDIN A.B.M., KANEL S.R., CHOI H. (2007): Environ. Sci. Technol., 41: 20222027
[12] JANG J.H., DEMPSEY B.A. (2008): Environ. Sci. Technol., 42: 28932898
[13] MERCER K.L., TOBIASON J.E. (2008): Environ. Sci. Technol., 42: 37973802
[14] SARKAR A., BLANEY L.M., GUPTA A., GHOSH D., SENGUPTA A.K. (2008): Environ. Sci.
Technol., 424: 2684273
[15] HINGSTON F.J., ATKINSON R.J., POSNER A.M., QUIRK J.P. (1968): 9th Int. Congr. Soil
Science Transactions (Adelaide, Australia), Vol. I: 669678
[16] HINGSTON F.J., POSNER A.M., QUIRK J.P. (1971): Discussions of the Faraday Society,
No.52: 334342
[17] PIERCE M.L., MOORE C.B. (1982): Water Res., 16: 12471253
[18] FUJITA T., TAGUCHI R., ABUMIYA M., MATSUMOTO M., SHIBATA E., NAKAMURA T.
(2008): Hydrometallurgy, 90: 92102
[19] FUJITA T., TAGUCHI R., ABUMIYA M., MATSUMOTO M., SHIBATA E., NAKAMURA T.
(2008): Hydrometallurgy, 90: 8591
[20] FUJITA T., TAGUCHI R., ABUMIYA M., MATSUMOTO M., SHIBATA E., NAKAMURA T.
(2008): Hydrometallurgy, 93: 3038
[21] FUJITA T., TAGUCHI R., ABUMIYA M., MATSUMOTO M., SHIBATA E., NAKAMURA T.
(2009): Hydrometallurgy, 96: 189-198
[22] FUJITA T., TAGUCHI R., MATSUMOTO M., SHIBATA E., NAKAMURA T. (2009): Hydrometal-
lurgy, 96: 300-312
[23] NISHIMURA T., TOZAWA K. (1978): Reprint from the Bull. The Research Institute of Mineral
Dressing and Metallurgy, Tohoku University, 34: 1926
Fujita, Shibata, Nakamura
Proceedings of Copper 2010 2934
[24] NISHIMURA T., TOZAWA K. (1984): J. Min. Metall. Inst. Japan, 100: 11381144
[25] ROBINS R.G. (1980): 4
th
Joint Meeting of MMIJ-AIME, Technical Session D-1, (Tokyo, Ja-
pan), 2544
[26] ROBINS R.G. (1981): Metall. Trans. B 12B: 103109
[27] ROBINS R.G. (1983): 3
rd
Int. Symp. on Hydrometallurgy, Hydrometallurgy Research, Devel-
opment and Plant Practice, Eds. K. Osseo-Asare and J. D. Miller, (The Metallurgical Society
of AIME, Georgia, Atlanta), 291310
[28] ROBINS R.G. (1987): Am. Mineralogist, 72: 842844
[29] ROBINS R.G. (1990): EPD Congress 90, Ed. D. R. Gaskell, (The Minerals, Metals & Mate-
rials Society, Anaheim, California), 93104
[30] KRAUSE E., ETTEL V.A. (1985): Proc. 15
th
Annual CIM Hydrometallurgy Meeting, (CIM,
Vancouver, Canada), 5/15/20
[31] KRAUSE E., ETTEL V.A. (1988): Am. Mineralogist, 73: 850854
[32] KRAUSE E., ETTEL V.A. (1989): Hydrometallurgy, 22: 311337
[33] HARRIS G.B., MONETTE S. (1989): Arsenic Metallurgy Fundamentals and Applications, Eds.
R. G. Reddy, J. L. Hendrix and P. B. Queneau, (The Metallurgical Society of AIME, Phoenix,
Arizona), 469488
[34] HARRIS G.B., MONETTE S. (1989): Productivity and Technology in the Metallurgical Indus-
tries, Eds. M. Koch and J. C. Taylor, (The Minerals, Metals & Materials Society, Cologne,
Germany), 545559
[35] EMETT M.T., KHOE G.H. (1994): EPD Congress 1994, Ed. G. W. Warren, (The Minerals,
Metals & Materials Society, San Francisco, California), 153166
[36] KHOE G., CARTER M., EMETT M., VANCE E.R., ZAW M. (1994): 6th AusIMM Extractive Metal-
lurgy Conference, Ed. G. W. Warren, (AusIMM publications, Brisbane, Queensland), 281286
[37] LANGMUIR D., MAHONEY J., MACDONALD A., ROWSON J. (1999): Geochimica et Cosmo-
chimica Acta, 63: 33793394
[38] LANGMUIR D., MAHONEY J., ROWSON J. (2006): Geochimica et Cosmochimica Acta, 70:
29422956
[39] BLUTEAU M.C., DEMOPOULOS G.P. (2004): Waste Processing and Recycling in Mineral and
Metallurgical Industries V Fifth Int. Symp., Hamilton, Ontario, Eds. S. R. Rao, F. W. Harri-
son, J. A. Kozinski, L. M. Amaratunga, T. C. Cheng and G. G. Richards, (CIM, Montreal,
Canada), 439451
[40] BLUTEAU M.C., DEMOPOULOS G.P. (2007): Hydrometallurgy, 87: 163177
Proceedings of Copper 2010 2935
Biosolubilization of Copper from
Waste Electric Cables
Asst. Prof. S. Gaydardzhiev, D. Bastin, Eng. F. Goffinet Dr. P. F. Bareel
University of Liege Comet Traitements S. A.
Mineral Processing and Recycling GeMME
B 52, Sart Tilman Rivage de Boubier 25
4000 Liege, Belgium 6200 Chtelet, Belgium
Keywords: Copper, recycling, technogenic products, bio leaching
Abstract
The paper reports on results from a laboratory tests for bacterial leaching of copper from scrap ca-
bles. The studied material is a reject fraction obtained after dismantling and separation of electric
cables during recycling of end-of-life vehicles (ELV). The copper has been met predominately in
pure metallic form as tiny irregular shaped wires often coated with tin and well liberated from the
plastic isolations. For bringing copper in solution, a bacterially assisted agitative leaching with
mixed consortium of mesophylic microorganisms has been chosen. Continuous bacterial adaptation
of the cultures to the substrate has been envisaged in order to provide an efficient way for ferrous
iron regeneration during the leaching. It has been established that under optimal conditions of pH,
density and temperature it is possible to recover nearly the total copper within short leach duration.
The obtained pregnant leach solution could be subjected to subsequent copper recovery via solvent
extraction, while the solid leached residue could be considered as non-metallic material containing
plastics suitable for recycling.
1 Introduction
Technogenic wastes derived from recycling of End-off-Life Vehicles (ELV) are viewed as growing
stream for use as a secondary resource base. Recent statistic estimates the world automotive park as
nearly 900 million cars, a figure which grows annually at constant rate of 2.25 % [1]. Moreover, due
to the forthcoming implementation of the new EU regulation in 2015 stipulating automotive indus-
try to recycle cars up to 95 % by weight, there is considerable interest in developing post-shredder
technologies capable to increase the recycling rate for both nonmetallic and metallic vehicle com-
ponents. During ELV recycling, cars are dismantled and depolluted and delivered to shredding and
further to post-shredder treatment where various material streams are formed. The separation flow
Gaydardzhiev, Bastin, Bareel, Goffinet
Proceedings of Copper 2010 2936
sheets used often involve combination of wind sichters, screens, magnetic and eddy-current separa-
tors. The dismantled electric and electronic equipment like board computers, CD players, navigation
units, electric cables etc., are processed independently in order to concentrate targeted components
like plastics, non-ferrous metals, and glass. The material stream in which electric cable scraps are
predominately concentrated as a rule is not suitable for land-filling or as an additive for the civil
engineering sector due to the elevated content of metals. Therefore, quite often a dedicated mechan-
ical post-processing involving grinding by knife mills and gravity separation (shaking tables, zigzag
sichters, pneumatic separators, etc.) is envisaged. The purpose is to obtain enriched in copper frac-
tions acceptable by the refineries as by-product to usual feeds. However, the heterogeneity of the
material and the irregular shape of wires and plastics render the gravimetric separation difficult
which reflects in non-negligible losses of metal values with the tailing fractions. This was the case
of the present study dealing with material presenting a mixture between the light fraction from spiral
separation and reject fraction after separation in pneumatic table of ground scrap cables. Hydrome-
tallurgical treatment could recover the remaining copper from this material, provided the metal sur-
face is accessible to aqueous leach solutions. Although not completely separated from their plastic
isolation, the cables are relatively finely ground and the isolation loosened during fragmentation,
therefore the wire surfaces could be deemed accessible to lixiviants.
The use of microorganisms to bring metals from wastes into solution could offer a low-cost alterna-
tive to the classical hydrometallurgical processes. Bio assisted solubilization can be utilized to oxi-
dize zero valent copper through leaching by Fe(III) and sulphuric acid, where Fe(II) oxidation is
carried out by the microorganisms. Several studies have been performed recently mainly for recov-
ery metals from PCBs (printed circuit boards) with various technological aspects being discussed
by Brandl et al. [2], Ilyas et al. [3], Yang et al. [4], Pham et al. [5]. Bioleaching has been also inves-
tigated by Misra et al. [6] for recovery of lithium from spent secondary batteries. Biotechnological
recovery of non-ferrous metals from scrap cables has not been yet reported. Therefore, the present
work has an objective to evaluate the feasibility of this option and to investigate certain process re-
levant parameters such as pulp density, pH and effect of culture adaptation on the material upon the
degree of metals leaching. The bacterial adaptation to the substrate as suggested by Elzeky & Atia
[7] is an important parameter during bioleaching of mineral ores and it is expected this factor to play
an essential role due to the quite specific nature of the cable scraps.
2 Materials and Methods
2.1 Culture
A mixed consortium of acidophilic bacteria from the genus Acidithiobacillus ferrooxidans, Acidi-
thiobacillus thiooxidans, and Leptospirillum ferrooxydans supplied by the Department of
Geoecololgy, University of Mining and Geology in Sofia, Bulgaria has been used. Bacteria were
incubated under non-sterile conditions at 35 C in 250 mL Erlenmayer flasks on orbital shaker using
Biosolubilization of Copper from Waste Electric Cables
Proceedings of Copper 2010 2937
an iron containing 9K medium prepared as described in [8]. Standard shake flask serial transfer was
used for adapting the cultures to the substrate. After three serial transfers, inoculation of a 2-liter
fermentor has taken place by transferring the cultures having reached their logarithmic growth
phase.
2.2 Analysis
Copper, lead, zinc and iron in the leach solutions have been analyzed by atomic adsorption spec-
trometer (AAS Perkin Elmer) following filtration of the samples. It has been tried to determine fer-
rous iron by titration with K
2
Cr
2
O
7
, but the strong interference effect from the bivalent copper pre-
sent in the solution has rendered this method non applicable. Sulphuric acid and sodium hydroxide
of analytical grade (Merck AG, Darmstadt) have been used to maintain the desired pH during leach-
ing. Particle size analysis of the sample has been realised by two alternative methods: by using a
Zephyr image analysis system (Ochio, Belgium) and through dry sieve analysis.
2.3 Scrap cables
The as received from the car recycling company sample of scrap cables was subjected to characteri-
sation and leaching without pre-treatment. Its particle size distribution obtained through both image
analysis and dry sieving is presented in Figure 1. A quite good coincidence of particle size distribu-
tion curves obtained by the two methods could be noted. The image analysis system determines the
particle size while capturing objects in motion and the implemented algorithm yields the weighted
sieve diameter. The mean diameter of the material (d
50
) as well as d
10
and d
90
as given by the image
analysis system are derived as 1.27 mm, 0.43 and 2.03 mm respectively.
Gaydardzhiev, Bastin, Bareel, Goffinet
Proceedings of Copper 2010 2938
Figure 1: Particle size distribution of the waste cables as given by sieve and image analysis
The content of the principal metals of interest in the scrap cables and their repartition in the differ-
ent granulometric classes is given in Table 1. It should be noted that the principal metal of interest is
copper which is concentrated in majority in the mesh size < 500 m. However, the mass yield of
this class has been found less than 10 %. For quantitative determination of other elements which
could not be detected by the AAS, the sample has been subjected to XRF analysis. The analysis
gave the following elements in respective concentrations [%]: Ca 6.58; Sn 0.03; Al 1.09;
Cl 2.47; F 1.09; Mg 0.21; O 8.07; P 0.02; Ni 0.04.
Figure 2 presents a general view of the material under study. The immediate impression is that the
predominant part of the wires is well liberated and metallic copper could be well distinguished es-
pecially on the background of the black foamy-look composites. Visible are several pieces of wires
coated with tin and pieces of brass as well. It should be noted that PVC and polyethylene are the
plastics most often met in the automotive cables. Other type of plastics such as EFTE, PET and PU
are less common. The issue of eventual dissolution of plastics during the non-ferrous metals leach-
ing should be addressed in view assessing the level of bacterial tolerance towards chlorinated
products.
Biosolubilization of Copper from Waste Electric Cables
Proceedings of Copper 2010 2939
Table 1: Concentration of main metals of interest and their granulometric repartition
Screen opening [mm] Cu [%] Fe [%] Zn [%] Pb [%]
+ 2
- 2 + 1
- 1 + 0.5
- 0.5 + 0.3
- 0.3 + 0.15
- 0.15
0.87
2.15
9.96
43.18
60.10
37.10
0.11
0.11
0.19
0.36
0.43
0.97
0.03
0.08
0.25
0.11
0.09
0.24
0.20
0.14
0.15
0.18
0.17
0.15
Input material 9.83 0.17 0.12 0.15
Figure 2: View of the as received scrap cables
Gaydardzhiev, Bastin, Bareel, Goffinet
Proceedings of Copper 2010 2940
2.4 Leaching procedure
Two different modes of agitative leaching have been tested. The first one, aimed for evaluating the
material aptitude to leaching and for adapting the bacterial culture to the substrate, has taken place
inside 250 mL Erlenmayer flasks on an orbital shaker. The second mode leaching performed at in-
creased pulp density, has been realised in a 1-liter double wall reactor equipped with a glass stirrer
and agitator run at 800 rpm (Janke & Kunkel RW20). In both cases temperature has been kept
around 35 C, either by placing the shaker inside hot room or by heating the reactor with thermo-
static bath. The leach solution has been produced by fermentor operating in batch mode and inocu-
lated with the culture under consideration. The required volume of leach solution has been periodi-
cally drained and the fermentor refilled with a fresh 9K nutrient medium. At the beginning of the
leaching experiment, 1 litre solution produced by the fermentor has been placed inside the leaching
reactor and predetermined amount of cables added. The solution containing the biomass was charac-
terized by redox potential of minimum 690 mV and concentration of ferric iron about 8 g/L, sug-
gesting that bacteria have reached their logarithmic phase of growth and that almost complete oxi-
dation of the ferrous to ferric iron by the bacteria has taken place. The pH set value for the leaching
was 1.9, chosen in view optimal bacterial growth and for prevention of excessive built up of ferrous
hydroxides which could lead to copper losses. The adjustment of the pH to the set value has been
realised either through periodical check and correction or via continuous pH control. At the former
case, pH has been measured in a given interval and when values above 1.9 registered, sulphuric acid
(10N) added to adjust pH back to 1.9. At the latter case, a pH electrode placed in the agitation vessel
and micro dosing pump have been connected to a control system (Consort R305) operating in pH-
stat mode. The Eh value (vs. SHE) has been measured as well and registered on-line by a PC to-
gether with the pH. Bacterial leaching has been monitored through periodical sampling of the pulp
and every time sample was taken, an equivalent amount of distilled water has been added. After
bioleaching, the pulp has been subjected to solid liquid separation using a filter paper and the re-
maining solid analysed for mass balance calculation.
3 Results and discussions
3.1 Leaching in Erlenmayer flasks at low pulp density with culture
non-adapted to the scrap cables
The initial bioleaching tests have been done with the aim to adapt the bacterial cultures to the sub-
strate and to evaluate the acid consumption required for keeping pH below 1.9. 10 grams of cable
scrap material have been contacted with 150 ml bacterial solution in Erlenmayer flasks. It should be
noted that for these tests, the bacterial culture has been already adapted during previous studies to
copper concentration of 4 g/L coming from metal bearing substrate different than the cables. The
results regarding pH and Eh evolution as well as the acid consumption for keeping pH below 1.9 are
Biosolubilization of Copper from Waste Electric Cables
Proceedings of Copper 2010 2941
presented at Table 2, while Figure 3 shows the leaching kinetics for copper, zinc and lead. The
leaching has been terminated after 48 hours based on visual observation of pulp coloration and on
disappearance of copper wires. Iron concentration in the solution has not been measured since it has
been assumed that due to pH variation part of the iron has precipitated.
Table 2: Variation in pH and Eh during leaching and amount of sulphuric acid (10N) totally
consumed
Time [h] Eh [mV] pH measured pH adjusted H
2
SO
4
added cumulatively [mL]
0 690 2.04 - -
1 512 2.30 1.84 0.47
3 414 2.40 1.88 0.87
6 386 2.68 1.80 1.27
27 430 2.36 1.78 1.67
28 440 2.35 1.77 2.07
30 458 2.09 1.84 2.30
48 558 2.18 1.78 2.50
It could be seeen that after approximately 40 hours, following an innitial drop, the redox potential
potential begins to rise which well corraborates with the end of copper leaching and indicates the
emerging of regenerated from the bateria ferric iron. As shown in Table 2, the pH has to be adjusted
back all along during the leaching.This indicates that the material is alkaline in nature and consumes
acid.
Figure 3: Leaching rate of copper, zinc and lead. Conditions: 6 % (w/v) pulp density, Erlenmayer
flasks, 35 C, Cu in g/L, Pb and Zn in mg/L
0
1
2
3
4
5
6
7
0 10 20 30 40 50 60
Leaching time (h)
C
o
n
c
e
n
t
r
a
t
i
o
n
o
f
C
u
(
g
/
L
)
0
2
4
6
8
10
12
14
16
18
20
C
o
n
c
e
n
t
r
a
t
i
o
n
o
f
Z
n
a
n
d
P
b
(
m
g
/
L
)
Cu
Zn
Pb
Gaydardzhiev, Bastin, Bareel, Goffinet
Proceedings of Copper 2010 2942
The results shown in Figure 3 suggest that it took less than 48 hours to bring copper in solution with
recovery of about 99 %. Zinc is mobilized faster than copper, while lead is not leached at all and
remains in the solid phase. The concentration of zinc in the pregnant leach solution is very low
which is logicall owing to its low innitial content. On comparison of the results from Figure 3 and
Table 2 it could be noticed that copper leaching does not follow the redox potential trend. Therefore
the sulphuric acid added for keeping the pH below 1.9 (totally 2.5 mL) eventually contributes also
to leach copper. The rise in the redox potential at the end of the test shows that the bacterial culture
even not adapted to the same substrat has been able to regenarate ferrous to ferric iron. The plastic
materials ussually contain calcium carbonate as filler (in our case about 16 % as given by the XRF
analysis), which explains the alkaline nature of the cables. Consequently it has been important to
delineate the amount of acid consumed by the plastic and the amount of acid responsible for
mobilizing copper. To evaluate the degree of acid consumption, the material has been brought in
contact with sulfuric acid under agitative mode with pH being kept between 1.8 and 2. The test was
terminated after 50 hours when pH has stabilized and sulphuric acid total consumption has been
calculated as 0.12 g/g material. During this leaching test, characterized by absence of
microorganisms and ferric iron, only 2 % of copper has been brought in solution. This fact idicates
that the presence of bacteria and ferric iron do enable and do accelerate mobilization of copper. The
XRD analysis of the leached residue is presented in Figure 4. The characteristics pics are
suggesting that the predominate compound of the precipitate is synthetic gypsum, resulting from the
reaction between the sulphuric acid and the calcium carbonate.
Figure 4: XRD spectra of the residue after acid leaching
Biosolubilization of Copper from Waste Electric Cables
Proceedings of Copper 2010 2943
3.2 Leaching at controlled pH with adapted bacterial culture
The test realised in agitative vessel at density of 12 % (w/v) has been performed to study the effect
of the continuous pH control on the degree of copper mobilization. At this instance, the leach sus-
pension has contained an adapted to the cables bacterial culture. Figure 5 portrays the leaching ki-
netics for copper, zinc and lead and the total iron concentration, while Figure 6 gives the variations
in pH and Eh during the leaching.
Figure 5: Leaching rate of copper, zinc and lead. Conditions: 12 % (w/v) pulp density, agitative
reactor, 35 C, Cu in g/L, Pb and Zn in mg/L
From the results it could be depicted that likewise in the Erlenmayer tests, zinc is leached rapidly at
the very beginning. The total iron concentration is stable at the first 20 hours after which begins to
decrease. The reduction of iron in the process solution from 7.3 to 6.2 g/L after the first 20 hours up
to the end of copper leaching reaction (hour 80) accounted for around 75 % copper dissolution. Fe
concentration decreased slowly and steadily as the leaching proceeded and iron concentration
reached about 5 g/L at the end. Possibly, regardless the continuous pH control, some part of the iron
is possibly precipitated as jarosite. The Eh curve in Figure 6 indicates a rapid sink of the redox po-
tential at the beginning following from a prolonged period of stabilization at around 400 mV. Al-
though copper solubilization comes to the end after 80 hours, the potential remains stable at about
400 mV until the hour 100, after which it rises rapidly. The dissolution of copper has been increas-
ing steadily indicating that bacteria are well adapted to the light jarosite media and continue to rege-
nerate the iron regardless its slightly diminishing concentration. The amount of consumed acid per
gram material has been the same as the one estimated in the Erlenmayer leaching tests. The slight
Gaydardzhiev, Bastin, Bareel, Goffinet
Proceedings of Copper 2010 2944
decrease in zinc and lead concentration at the end of the leaching cycle could be due to sorption
effects caused by the formed iron precipitates.
Figure 6: Evolution of Eh and pH during leaching. Conditions: 12 % (w/v) pulp density, agitative
reactor, 35 C
4 Conclusions
The study although preliminary, allows drawing certain conclusions regarding the approach feasibil-
ity. The experiments of the leaching of scrap cables have indicated that metallic copper could be
recovered effectively by bio-hydrometallurgical process.
It has taken about 40 hours to leach the total copper from a suspension with 6 % density (w/v) using
a bacterial culture adapted to other substrate. Copper concentration has reached 6.3 g/L. The
adapted to the material bacterial culture has recovered about 100 % of the copper in 80 hours
(3.5 days) from a suspension at 12 w/v density and have brought copper concentration of 12.8 g/L in
the PLS. Hence increasing twice the pulp density has doubled the leaching time required to reach
same recovery, with copper concentration in PLS increasing also twice. This correlation shows that
there are no inhibition effects from the cable material on the bacterial culture, which is already resis-
tive following its initial adaptation to other metal bearing substrate. Slight variation in pulp pH in
certain margins does not influence leaching efficiency.
Neither the plastics nor the presence of tin have affected the bacterial activity and their ability to
regenerate iron. Ferric iron seems the main factor for copper mobilization, but iron speciation
method should be found in this particular case.
The final copper concentration makes the PLS suitable for further treatment via solvent extraction
for copper extraction with the additional advantage that the solution does not contain zinc.
Biosolubilization of Copper from Waste Electric Cables
Proceedings of Copper 2010 2945
References
[1] WINFIELD, P., HUTCHINSON, A. & PEMBERTON, M., (2007): Whole life vehicle waste
streams, Oxford Brookes University.
[2] ILYAS, S. et al., (2007): Bioleaching of Metals from Electronic Scrap by Moderately Thermo-
philic Acidophilic Bacteria, Hydrometallugy, 88, 180-188
[3] BRANDL, H., BOSHARD, R. & WEGMANN, M. (2001): Computer-munching Microbes:
Metal Leaching from Electronic Scrap by Bacteria and Fungi, Hydrometallugy, 59, 319-326
[4] PHAM, V. & TING, Y. (2009): Gold Bioleaching of Electronic Waste by Cyanogenic Bacteria
and its Enhancement with Bio-oxidation, Advanced material research, 71-73, 661-664
[5] YANG, T. et al., (2009): Factors Influencing Bioleaching Copper from Waste Printed Circuit
Boards by Acidithiobacillus ferrooxidans, Hydrometallugy, 97, 29-32
[6] MISHRA, D. et al., (2008): Bioleaching of Metals from Spent Lithium Ion secondary Batteries
using Acidithiobacillus ferrooxidans, Waste Management, 28, 333-338
[7] ELZEKY, M. & ATTIA, Y., (1995): Effect of Bacterial Adaptation on Kinetics and Mechanism
of Bioleaching Ferrous Sulfides, Chemical Engineering Journal, 56, B115-B124
[8] SILVERMAN, M. & LUNDGREN, D. (1959): Studies on the chemoautotrophic iron bacterium
Ferrobacillus ferrooxidans. I. An improved Medium and a Harvesting Procedure for securing
High Cell Yields, Journal of Bacteriololgy, 77, 642647
Proceedings of Copper 2010 2946
Proceedings of Copper 2010 2947
Dowa Mining Scorodite Process
Application
to Copper Hydrometallurgy
H. Kubo, M. Abumiya, M. Matsumoto
Dowa Metals & Mining Co. Ltd.
217-9 Shimo-kawabata Furumichi Iijima
Akita, Japan
Keywords: Scorodite, arsenic, ferrous oxidation, ferric arsenate, crystalline
Abstract
Dowa Metals & Mining Co. Ltd. has developed a new method for arsenic immobilization. Specifi-
cally, arsenic fixation is achieved by synthesizing crystalline scorodite (FeAsO
4
2H
2
O) under at-
mospheric conditions.
Since June 2008, a plant handling 30 MT/month of arsenic has been operating at the Kosaka
Smelter to produce crystalline scorodite from non-ferrous intermediate materials. The results from
this plants operation are reported here.
The process utilized by this plant consists primarily of three steps: 1) the process of leaching arsenic
from metallurgical intermediate materials containing arsenic; 2) the oxidation of As(III) to As(V) in
the leaching solution; 3) the crystallization of scorodite from the solution at 95 C under atmos-
pheric conditions.
This plant can use two types of intermediate products as the starting material. The first is arsenic
sulfide precipitated from the wastewater from sulfuric acid production. The other is liberation slime
from copper electrolysis.
The scorodite formed in this plant has good crystallinity, large particle size (>20 m on average),
high X-ray diffraction intensities, and filtering and handling attributes similar to those of laboratory-
synthesized samples. The concentration of arsenic with respect to the solubility of scorodite is low
enough to satisfy the environmental regulations of most countries.
This achievement may contribute to the arsenic immobilization methods used in the non-ferrous
smelting industry.
Kubo, Abumiya, Matsumoto
Proceedings of Copper 2010 2948
1 Introduction
Today, non-ferrous smelters are facing serious problems in handling low-grade ores and concen-
trates, which contain large amounts of various impurities. This is particularly evident in copper
smelting due to the continuously increasing trends seen in the amounts of arsenic present in concen-
trates. Hence, copper smelters are especially prone to operational difficulties. Therefore, establish-
ment of a reliable method for arsenic management is considered to be urgent.
Two types of widely known intermediate products in the copper smelting process contain large
amounts of arsenic. One is arsenic sulfide, which is precipitated from the wastewater solutions in
sulfuric acid plants, and the other is liberation slime created from copper electrolysis. These prod-
ucts, which also contain large amounts of copper, are usually recycled into the smelting process for
copper recovery. However, increasing arsenic charge from raw materials also increases the volume
of reiterative arsenic remaining in the process, and makes the operation of smelters gradually diffi-
cult. Then, the recycling of these products cannot be continued, and this finally results in an unde-
sired temporary stockpile of these intermediate products in an unstable form within the smelters
facilities. An appropriate and reliable method of treatment for arsenic extraction and fixation from
such intermediate products is urgently needed to cope with arsenics rising risks and problems such
as its outflow, exposure, and space for the safe storage of such arsenic-bearing products.
Studies on the treatment of arsenic-containing non-ferrous intermediate products have been per-
formed and reviewed [1-3]. However, a process which resolves this issue satisfactorily has yet to be
developed, and many unresolved problems remain.
Dowa Metals & Mining Co. Ltd. has developed a process for treating arsenic sulfide and liberation
slime for them to function as starting materials. This process involves simultaneous leaching of ar-
senic at a high extraction ratio and copper residue recovery. The copper residue is used as a resource
for copper smelting, and the dissolved arsenic in the solution is transformed into crystalline scoro-
dite (FeAsO
4
2H
2
O) by reaction with iron(II) ion and O
2
gas in the solution [4-6]. The scorodite is
considered to be the most stable arsenic compound ever known.
Based on the achievement process described above, DOWA began experiments to verify the indus-
trial capability of a plant with a capacity of 30 MT/month of arsenic since June 2008, and has suc-
ceeded in proving that it is possible to produce crystalline scorodite industrially on a commercial
scale.
This process was named and trademarked as DMSP
Plant
2.1 Plant design concept and features
Figure 1 shows the appearance of the DMSP
consists of three processes. The first process is arsenic leaching and recovery of copper
residues. The second process is oxidation of As(III) in the leached solution to As(V). The third
process is crystallization, immobilizing the arsenic dissolved in the oxidized solution obtained from
the oxidation process, by forming scorodite in a solution of sulfuric acid under atmospheric condi-
tions. Figure 2 shows a flow diagram of this process.
Kubo, Abumiya, Matsumoto
Proceedings of Copper 2010 2950
Figure 2: DMSP
flow diagram
2.3 Leaching process
In this process, arsenic contained in the materials is leached into solution, while copper is not
leached and remains as a residue. Two kinds of arsenic-containing products are treated as target
materials. One is arsenic sulfide obtained from sulfide treatment of the wastewater solutions from
sulfuric acid production, and the other is liberation slime created in copper electrolysis.
This process is completed by the application and development of the Kowa-process [7].
The essential reaction is represented by Equation (1).
As
2
S
3
+ 3CuSO
4
+ 4H
2
O 3CuS + 2HAsO
2
+ 3H
2
SO
4
(1)
The sulfuric acid generated by this reaction is consumed to leach copper from liberation slime.
Solid-liquid separation is carried out by pressure filtration. The obtained solution moves on to the
next oxidation process. The residue is supplied to the copper smelting process as a material.
High Arsenic Bearing Materials
(Arsenic sulfide, Liberation slime)
O
Leaching
Solution Residue
Oxidation Copper Smelter
Oxidized Solution
Crystallization
SCORODITE Effluent
Stockpile Solution Treatment
Oxidizing agent
FeSO7HO
O-gas
Dowa Mining Scorodite Process
plant
0
5
10
15
20
25
30
35
40
6 7 8 9 10 11 12 1 2 3 4 5
2008 2009
Period (year - month)
A
r
s
e
n
i
c
M
T
/
M
o
n
t
h
0%
20%
40%
60%
80%
100%
9 10 11 12 1 2 3 4 5
2008 2009
Period (year - month)
E
x
t
r
a
c
t
i
o
n
(
%
)
Dowa Mining Scorodite Process
plant
(MOE Notification No. 13)
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
0.45
0.50
0 10 20 30 40 50 60
Sample No.
L
e
a
c
h
a
t
e
A
s
(
m
g
/
L
)
Regulatory threshold: 0.3mg/L
Average: 0.07mg/L
Kubo, Abumiya, Matsumoto
Proceedings of Copper 2010 2954
Table 1: Arsenic solubility of scorodite produced at the DMSP
plant (other testing methods)
Sample a Sample b Sample c
Notification No. 13 (Japan) 0.06 0.07 0.03
TCLP (U.S. EPA Method 1311) 0.053 0.038 0.034
Availability Test (NEN 7341) 0.070 0.088 0.031
EP (U.S. EPA Method 1310B) 0.045 0.040 0.028
MEP (1) (U.S. EPA Method 1320) 0.31 0.24 0.24
MEP (2) 0.26 0.32 0.37
MEP (3) 0.46 0.48 0.54
MEP (4) 0.31 0.31 0.36
MEP (5) 0.37 0.33 0.37
MEP (6) 0.23 0.29 0.32
MEP (7) 0.23 0.29 0.31
MEP (8) 0.25 0.26 0.32
MEP (9) 0.28 0.51 0.33
Method
Leachate As (mg/L)
Table 2: Result of leaching test for hazardous heavy metals in scorodite produced at the DMSP
plant
Hg Pb Cd Se Cr(VI)
0.005 0.3 0.3 0.3 1.5
<0.0005 <0.01 <0.03 <0.01 <0.02
Threshhold limit (mg/L)
Test result (mg/L)
3.4 Characterization of Scorodite
The precipitates produced at the DMSP
Scorodite
ICDD-PDF: 00-37-0468
Lab. test sample
Plant made sample
Kubo, Abumiya, Matsumoto
Proceedings of Copper 2010 2956
Figure 7: Comparison of SEM images of scorodite
Figure 8: Particle size distributions of scorodite
Lab. test sample Plant made sample
20m 20m
0%
3%
6%
9%
12%
15%
1 10 100 1000
Diameter (m)
F
r
e
q
u
e
n
c
y
(
%
)
Plant made sample
Lab. test sample
Dowa Mining Scorodite Process
has been developed for arsenic immobilization and copper recovery in copper smelters.
Practical experiments have been subsequently carried out at a new plant with satisfactory results,
verifying that it is capable of handling 30 As-MT/month of arsenic. The plant start-up was achieved
without difficulty, which showed the great potential of the DMSP
................................................. 137
convergent divergent nozzle ................. 931
converter slag ........................................... 863
converters ................................................. 839
converting ....................................... 649, 2543
conveying ............................................... 2517
cooling ...................................................... 907
cooling media ......................................... 1247
cooling system .......................................... 699
cooling technology ................................. 1247
cooling water ............................................ 271
copper ........ 31, 271, 285, 301, 365, 379, 425,
.......................... 439, 517, 533, 561, 615, 631,
.......................... 649, 721, 749, 761, 793, 863,
........................ 1063, 1259, 1281, 1327, 1345,
........................ 1423, 1483, 1531, 1545, 1559,
........................ 1569, 1599, 1649, 1675, 1713,
........................ 1783, 1845, 1871, 1899, 1925,
........................ 1983, 2039, 2071, 2433, 2543,
.................................. 2589, 2673, 2693, 2737,
................................... 2807, 2873, 2897, 2935
copper alloys .............. 81, 185, 191, 207, 217
copper business ...................................... 2563
copper cathode ....................................... 1585
copper concentrate ............ 1095, 1183, 1725,
....................................................... 2575, 2747
copper concentrates .............. 893, 1013, 1273
copper concentrator ................................ 2235
copper converting ........................... 811, 1297
copper cooler ............................................ 971
copper demand ....................................... 2215
copper electrorefining ....... 1307, 1367, 1437,
...................................................... 1663, 1687
copper electrowinning .................. 1379, 1585
copper extraction technology ................. 1167
copper fabrication .................................. 2563
copper global production ....................... 2215
copper industry ....................................... 1167
copper infiltration .................................... 247
copper losses .......................................... 2637
copper losses to slag .............................. 1115
copper market ........................................ 2215
copper material properties ....................... 231
copper matte ................................... 931, 1287
copper matte smelting ............................ 1115
copper mine projects .............................. 2215
copper nanoparticles .................................. 21
copper ores ......................... 2453, 2655, 2859
copper production .................................. 1511
copper recovery .......... 399, 1095, 2029, 2781
copper refining ................... 1521, 1941, 2397
copper refining electrolysis .................... 1495
copper resources .................................... 2883
copper rod .................................................. 99
copper rod plasticity examination by AR test .. 13
Keywords Index
Proceedings of Copper 2010 3014
copper sector .......................................... 2249
copper selenide ....................................... 1815
copper slag ............................................. 2283
copper smelter ........................................ 2415
copper smelting ........................................ 731
copper solvent extraction ....................... 2059
copper telluride ...................................... 2473
copper trade ............................................ 2215
copper white metal ................................. 2283
copper-cobalt concentrate ........................ 587
corrosion ............................... 699, 1593, 1857
cost effective .......................................... 1143
cost escalation ........................................ 2101
cost function ........................................... 2171
cost reduction ............................... 1051, 2369
cost saving .............................................. 1585
costs ........................................................ 2101
crane operation ....................................... 1569
critical blast flow rate ............................... 839
CRT ........................................................ 2959
crucible melting furnace ........................... 173
crud treatment ........................................ 2737
crystalline ............................................... 2947
Cu-40Zn brass ............................................ 89
CuAg0.1 ..................................................... 99
CuAg0.10 ................................................. 137
Cu-ETP1 .................................................... 99
CuFe
2
P ....................................................... 81
CuMg0.2 ..................................................... 99
CuMn20Ni20 ............................................... 3
Cu-OFE ...................................................... 99
cupric ion leaching ................................. 1815
current ...... 1281, 1355, 1449, 1521, 1687, 1713
current efficiency ........ 1437, 1649, 1687, 1713
CuSn10 ...................................................... 81
cyanide ................................................... 2087
cyclone reactor ......................................... 879
data interoperability ............................... 2127
decopperized anode slime smelting ....... 1293
development ................. 387, 721, 1233, 1247
diamond ................................................. 2873
direct current .......................................... 1437
direct emissions ..................................... 2117
direct to blister furnace .......................... 1063
discourse communities ............................. 351
distribution ............................................. 1281
distribution coefficient ........................... 1495
distribution ratio of minor elements ...... 2415
dor matte refining ................................. 1293
dor metal refining ................................. 1293
dryer ....................................................... 2517
dust ........................................................... 893
dust emissions .......................................... 517
dusting rate ............................................. 1025
dynamic modeling .................................. 2235
ecobronze ................................................... 81
economic ................................................ 1999
Ecuprex .................................................. 1973
Keywords Index
Proceedings of Copper 2010 3015
EE-core ..................................................... 173
efficiency .................................................. 333
EIS .......................................................... 1909
electric furnace ......................................... 961
electric home appliances ........................ 2959
electro-chemical ....................................... 425
electrochemical dissolution .................... 1909
electrochemical potential ....................... 2747
electrochemistry ..................................... 1845
electrolysis ................................................ 399
electrolyte ............................................... 1307
electrolyte composition .......................... 1463
electrolyte temperature ........................... 1675
electrolyte treatment ............................... 1663
electrolytic .............................................. 1599
electrolytic copper ...................................... 13
electrolytic copper powder ....................... 399
electrorefining .......... 1355, 1531, 1675, 1713
electrorefining process ........................... 1327
electrowinning ......... 1237, 1253, 1281, 1355,
.................................. 1449, 1483, 1545, 1559,
.................................. 1617, 1635, 1649, 1699,
................................... 1713, 1973, 2137, 2189
emission ................................................... 713
emissions .......................................... 477, 649
enargite ..................... 1035, 1783, 1983, 2883
energy ..... 301, 561, 649, 713, 1617, 1635, 2543
energy efficiency .................... 185, 271, 2117
energy optimization .................................. 497
energy saving ...... 779, 831, 1449, 1511, 2369
energy saving ........................................... 831
engineering design ................................... 547
environment ... 285, 533, 561, 577, 721, 2543
environment protection .......................... 1167
environmental engineering ....................... 547
environmental impact .............................. 547
environmental improvement .................... 413
environmental preservation .................... 2563
environmental protection ....................... 2369
EPMA ............................................ 811, 1297
ethyl xanthate ......................................... 2765
EU regulations ......................................... 365
evaporation ............................................. 1663
excellence ................................................. 309
exhaust gas system ................................... 413
expansion ............................................... 1143
experimental .......................................... 1271
experimental estimation ........................... 919
extraction ............................. 893, 1259, 2721
ferric arsenate ......................................... 2947
ferric sulphate leaching .......................... 1815
ferrous oxidation .................................... 2947
Finite Element Analysis (FEA) ................ 587
fire refining .............................................. 649
flameless combustion ............................. 1051
flash dryer ................................................ 779
flash furnace ........................................... 1079
flash smelting ..... 669, 699, 1313, 2381, 2415
flash smelting furnace ............ 823, 879, 1155
Keywords Index
Proceedings of Copper 2010 3016
flash smelting process ............................ 2499
flotation ................... 2189, 2433, 2589, 2655,
.............................................. 2765, 2845, 2859
flotation column control ......................... 2829
flotation reagents optimization ............... 2453
flow conditions ....................................... 1675
fluid bed ................................................... 587
fluid flow .................................................. 823
fluidization ............................................. 1025
flux ........................................................... 685
fluxes ........................................................ 893
forging ...................................................... 155
freezing line ............................................ 2307
frit-and-sleeve sparger ............................ 2829
froth flotation ......................................... 2453
fugitive emission ...................................... 685
fugitive emissions .......................... 517, 2971
furnace campaign life ............................... 971
furnace inspection .................................. 1025
furnace integrity ............................... 971, 987
gas cleaning .............................................. 497
gas flow .......................................... 879, 1003
gas sparging ............................................ 1545
gas-gas cooler ......................................... 1173
gas-solid ................................................. 2483
glue ......................................................... 1687
gold ........................... 1259, 1393, 1403, 1783
granulation ............................................. 2283
graphite particles ........................................ 89
grinding ........................................ 2189, 2621
gypsum ..................................................... 465
hardrock mining ..................................... 2873
harvesting plan ....................................... 1569
Hatch furnace technology ........................ 971
head grade .............................................. 2171
health ........................................................ 561
heap leach .............................................. 2711
heap leach modeling .............................. 2703
heap leaching ......................................... 2693
hearth accretion ...................................... 2327
heat and mass balance ............................ 1063
heat loss ................................................. 2307
heat recovery .................................. 497, 1003
heat transfer ............................................ 2307
heat treatment ........................................... 155
heavy metals ............................................. 517
heavy oil consumption ............................. 779
high current density ...................... 1545, 1675
high pressure grinding rolls ................... 2873
high speed flow ...................................... 1079
high speed pressure measurement .......... 1079
high-arsenic ............................................ 1273
historic R+D ............................................. 945
history .................................................... 2721
hospital superbugs .................................... 217
hospital-acquired infection .............. 207, 217
HPGR ........................................... 2621, 2873
Hybrid flotation ...................................... 2793
Keywords Index
Proceedings of Copper 2010 3017
HydroCopper
........................................ 1959
hydrogen reduction ..................................... 21
hydrogen sulfide ..................................... 1973
hydrometallurgical ....................... 1403, 1871
hydrometallurgy ...... 1393, 1783, 1845, 2017,
................................... 2039, 2189, 2693, 2737
image analysis ........................................ 1237
IMC control ............................................ 2829
impact ....................................................... 387
implementation ...................................... 2499
improvement ........ 387, 699, 713, 1213, 1413
improvements ......................................... 1379
impure anodes ........................................ 1495
impurities .......................... 1327, 1463, 1663,
impurities in cathode copper ...................... 13
impurity management ............................. 2159
indirect emissions ................................... 2117
indoor air quality ...................................... 185
induction casting ........................................ 31
induction melting ....................................... 31
inductive heating ...................................... 173
industrial test .......................................... 1095
inertization ............................................. 2071
infection control ....................................... 207
information system ................................. 1423
infrared ................................................... 1123
injection .................................................. 2483
innovation ................................................ 577
intensified smelting process ................... 1313
investment .............................................. 1413
ion exchange .......................................... 1803
ionic liquids ........................................... 1247
iron ......................................................... 1379
iron-silicate slag ..................................... 1287
ISA 2000 ................................................ 1307
ISA process ............................................ 1521
ISASMELT
TM
................ 615, 749, 907, 1327
ITER ............................................................. 3
jacket ...................................................... 2307
Kayser Recycling System (KRS) ........... 2971
kimberlite ............................................... 2873
Konkola Copper Mines .......................... 1413
k- turbulence model .................... 2397, 2499
La Coipa mine ........................................ 1035
laboratory practices testing ...................... 191
labour productivity ................................... 439
lance ....................................................... 2483
leaching ..... 893, 1585, 1725, 1737, 1753, 1771,
............................... 1783, 1973, 1983, 1999, 2029,
................................. 2189, 2655, 2721, 2753, 2897
lead ......................................... 631, 863, 1699
lead anodes ............................................. 1593
lead free brass .......................................... 255
lead removal ............................................. 669
lead-free ..................................................... 89
liberation ...................................... 2703, 2859
lime ................................................ 465, 2765
limestone .................................................. 685
liquidus .................................................... 731
Keywords Index
Proceedings of Copper 2010 3018
liquidus temperature ............... 761, 811, 1297
LIX 7950 ................................................ 2087
LOI measurement ................................... 2433
Los Pelambres ........................................ 2793
low-grade ores ........................................ 2781
machinability .............................................. 89
macrosegregation ................................. 49, 65
maintenance optimization ...................... 1173
management ............................ 309, 319, 387,
management decisions ........................... 2127
management system ............................... 1423
manganese .................................... 1593, 1699
manganese dioxide ................................. 1483
markets ................................................... 2543
material cycle ............................................. 81
material flow management ..................... 2341
material storage and pre-treatment ......... 2971
matte converting ....................................... 945
matte making .......................................... 2527
measurement .................................. 517, 1079
mechanical properties .............................. 255
metal ....................................................... 2087
metal ions ............................................... 1857
metal recovery ................................ 863, 1213
metallurgy ................................................ 319
metals ....................................................... 425
metastable Cu1+xSe phase .................... 1815
METTOP-BRX-Technology .................. 1713
micro alloys ................................................ 99
microstructure .......................................... 255
mineralogy from elements ..................... 2381
minerals applications ............................. 2621
minerals references ................................ 2621
miniaturization ........................................... 81
mining ............................ 285, 319, 333, 2807
mining industry ................................ 533, 577
mining projects ........................................ 547
minor element deportment ..................... 2415
mist suppression agents ......................... 1271
Mitsubishi process ..... 685, 1143, 1199, 2563
Mitsubishi smelting process .................. 2483
model ..................................................... 1649
model application ................................... 2381
modeling .................... 919, 1003, 1063, 2711
moderate thermophile ............................ 2575
modernization .......................................... 961
modified phosphinic acid formulation ... 2059
Molecular Recognition Technology (MRT)
................................................................ 1941
molybdenum ................................ 2059, 2589
molybdenum recovery ............................ 2029
MRSA ...................................................... 217
multicomponent ........................................... 3
multi-frequency current ......................... 1355
NaCl ....................................................... 2923
NaOH ..................................................... 2923
neutralizing .............................................. 465
new technology ...................................... 1871
Keywords Index
Proceedings of Copper 2010 3019
nickel ...................................... 793, 863, 1663
nitrate solution ....................................... 1687
nodulation .............................................. 1599
Non-Destructive Testing (NDT) ............ 1123
NSC ........................................................ 2589
numerical simulation ................................ 823
off-gas ........................ 601, 1003, 1173, 2357
off-gas cleaning ........................................ 601
off-gas duct .................................. 1173, 1233
off-gas flow ............................................ 2499
off-gas line ............................................. 1003
off-gas recirculation ............................... 1051
on-line particle size analysis .................. 2267
operating ................................................. 2101
operation ....................................... 1143, 1345
operational cost ........................................ 497
operational stability .................................. 477
operations ................................................. 533
optimization ....... 721, 879, 1003, 1569, 2575
optimizing operations ............................. 1379
OPUS ..................................................... 2267
ORP ........................................................ 1393
oscillation ............................................... 1367
O-SR process .......................................... 1199
oxidation .............................. 793, 2443, 2753
oxyfuel ................................................... 1051
oxygen .......................................... 1495, 2543
oxygen burner ......................................... 1051
oxygen evolution anode ......................... 1511
oxygen free copper ................................... 119
oxygen free silver copper ......................... 137
oxygen-enriched bottom blowing .......... 2527
oxygenfree copper ...................................... 31
panel data ................................................. 439
parallel electrolyte flow ......................... 1713
passivation ............................................. 1559
passivation mechanism .......................... 1463
PCR ........................................................ 1675
penetration depth .................................... 2483
permanent cathode ................................. 1521
permanent cathodes ................................ 2295
phase diagram .......................................... 731
phase diagrams ......................................... 761
pilot molten layer reactor ....................... 1273
pilot tests ................................................ 2039
pilot trials ................................................. 945
planning ................................................. 2295
plant continuity ...................................... 2127
plant structuring ..................................... 2137
pneumatic ............................................... 2517
polarization ............................................ 1531
polymer additive .................................... 1531
polymetallics .......................................... 2913
potential ............................. 1737, 1753, 1771
practice production ................................ 2527
precious metal ........................................ 1287
precious metals ........................................ 631
precipitation ................................. 1599, 1973
Keywords Index
Proceedings of Copper 2010 3020
pressure leaching .......................... 2017, 2673
pressure oxidation .................................. 2589
prices ...................................................... 2101
primary materials ................................... 1899
process advisor ....................................... 2381
process advisor software ........................ 1155
process control ............................. 1063, 1155
process heat balance ................................. 631
process modernization .............................. 631
process optimization .................... 2369, 2397
production ...................................... 477, 2807
production management system ............. 2341
productivity .......................... 713, 1449, 2543
profit ....................................................... 2807
project ..................................................... 1871
protection ............................................... 1559
P-S Converter ......................... 919, 931, 1183
public health registration .......................... 191
pyrite ................................. 1845, 2721, 2765,
pyritic ores ................................................ 945
pyrometallurgy .............. 379, 615, 749, 1115,
............................................. 1183, 2443, 2637
Radiant cooling chamber (RCC) ............ 2357
rate-controlling step ............................... 1959
reaction shaft .......................................... 1313
reagents .................................................. 1925
real time transparency ............................ 2341
recovery .......................................... 301, 1259
recyclabilityy .............................................. 81
recycle ...................................................... 425
recycling ..... 379, 2117, 2159, 2913, 2935, 2959
redox potential ............................. 1725, 2753
reduction ...................... 477, 793, 2397, 2753
refinery ....................... 721, 1345, 2295, 1253
refractory ................................................ 2307
refractory engineering .................. 1233, 1247
refractory wear ....................................... 1123
residence time distribution ....................... 919
residue .................................................... 1259
residue disposal ...................................... 1035
resistance .................................................. 699
resource utilization ................................. 1167
resources .................................................. 333
rhenium .................................................. 1803
rhenium recovery ..................................... 893
risk and uncertainty ................................ 2249
risk assessment ......................................... 365
roasting ........... 1035, 1259, 1783, 2029, 2721
robotic .................................................... 1253
rod mill ..................................................... 271
rolled lead calcium tin alloys ................. 1593
rotary furnace ........................................... 721
rotating cylinder electrode ....................... 399
RotoFeed