CSRP08 Proceedings Lowres - Pdf#page 49

CSRP08
Delivering sustainable solutions to the minerals and metals industries

Produced by
CENTRE FOR SUSTAINABLE RESOURCE PROCESSING
26 Dick Perry Avenue, Kensington WA 6152 Australia
2nd annual conference
18 19 November 2008
Brisbane, Queensland, Australia
PROGRAM
TUESDAY 18 NOVEMBER 2008 (The Long Room)
8:30 Welcome, OHS, Introduction and Summary of CSRP Highlights
Stevan Green (CEO, CSRP) and Erica Smyth (Independent Director, CSRP)
8:45 Keynote Speaker
Dr John Cole (General Manager, Queensland Office of Clean Energy)
09:15 [1] ZERO WASTE & MINOR ELEMENTS
Chair: Warren Bruckard (CSIRO) (Q&A to follow each session 1 presentation)
Developing SUSOP a structured, methodical mechanism for incorporating sustainability principles
into the design and operation of minerals processing plants
Ben McLellan (Sustainable Minerals Institute, University of Queensland)
Overview of the Early Removal Process Small Scale Development Work
Warren Bruckard (CSIRO Minerals)
Thermodynamics of Selective Roasting of Arsenic from Tennantite Containing Copper Concentrates
Steven Wright (CSIRO Minerals)
Techno-Economic Assessment of Early Arsenic Removal from Copper Ores
Nawshad Haque (CSIRO Minerals)
10:30 morning tea
11:00 [2] CO2 BREAKTHROUGH IN METAL PRODUCTION
Chair: Sharif Jahanshahi (CSIRO)
Water Conservation in the Mining Industry case study of gold mining and refining
Goen Ho (Murdoch University)
Overview of the CO2 Breakthrough Program and Linkage to IISI
John Mathieson (BlueScope Steel Research)
Survey of Sustainable Biomass Resources for Iron and Steel Industry
Nawshad Haque (CSIRO Minerals)
Demonstration of Recarburisation of Liquid Steel Using Renewable Carbon at OneSteel
Michael Somerville (CSIRO Minerals)
Piloting the Integrated Dry Granulation and Heat Recovery Process at CSIRO
Dongsheng Xie (CSIRO Minerals)
Value of the Projects Industry Perspective
Phillip Ridgeway (OneSteel)
Panel Discussion (All session 2 speakers available for Q&A)
12:40 networking lunch
1:30 [3a] ENERGY EFFICIENT LIBERATION & COMMINUTION
Chair: Malcolm Powell (University of Queensland)
Minerals, Metals and Sustainability a text book for earth resources graduates of the future
John Rankin (CSIRO Minerals)
Demand-Supply Interaction on Future Mining Resource Production: The Coal Model
Geoffrey Evans (University of Newcastle)
Overview of Eco Efficient Liberation
Malcolm Powell (JKMRC, University of Queensland)
Energy Efficient Circuits and Progressive Liberation
Zeljka Pokrajcic (PhD Student, JKMRC, University of Queensland)
Fair Measures of Energy Usage
Fiesal Musa (PhD Student, JKMRC, University of Queensland)
Improving Grinding Efficiency with the IsaMill
Rob Morrison (JKMRC, University of Queensland)
2:45 afternoon tea
3:00 [3b] ENERGY EFFICIENT LIBERATION & COMMINUTION
Chair: Malcolm Powell (University of Queensland)
High Pressure Grinding Rolls Multiple Pass
Marko Hilden (JKMRC, University of Queensland)
Update on the JK Rotary Breakage Tester
Toni Kojovic (JKMRC, University of Queensland)
Improvement of Energy Efficiency of Rock Comminution through Reduction of Thermal Losses
Nenad Djordjevic (JKMRC, University of Queensland)
Towards a Virtual Comminution Machine
Rob Morrison (JKMRC, University of Queensland)
The New Energy Logging from the Discrete Element Method
Nirmal Weerasekara (JKMRC, University of Queensland)
Progress on the Unified Comminution Model
Malcolm Powell (JKMRC, University of Queensland)
4:30 end presentation sessions
4:35 [4] PARTICIPANTS FORUM (Lady Thiess Room)
Chair: Philip Bangerter (Hatch)
The purpose of the Participants Forum is to provide a forum for consultation among the CSRP Participants.
7:00 Pre-dinner drinks and canaps served on the Riverside Terrace, Customs House
7:30 Conference Dinner in The Long Room, Customs House
WEDNESDAY 19 NOVEMBER 2008 (The Long Room)
08:30 Welcome and Information on CSRP2
Stevan Green (CEO, CSRP)
09:00 [5] INDUSTRY PANEL SESSION
Chair: Erica Smyth (Director, CSRP)
This is an opportunity for our industry participants to tell us what we can do to better serve their needs and to
provide an opportunity to respond to day 1 of the conference.
10:00 morning tea
10:30 [6] GEOPOLYMERS
Chair: Arie van Riessen (Curtin University of Technology) (Q&A to follow each session 6 presentation)
Australian Life Cycle Initiative (AusLCI) and CSRP database: Australian data
Damien Giurco (Institute for Sustainable Futures, University of Technology Sydney)
Highly Sensitive Quantitative Phase Analysis of Fly Ashes for Use as Geopolymer Source Materials
Ross Williams (Centre for Materials Research, Curtin University)
Synthesis Paths and Performance Evaluation of Geopolymer Binder Systems Derived from Major
Mineral Processing and Mining Waste Feedstock Materials
Kwesi Sagoe-Crentsil (CSIRO Materials Science & Engineering)
Managing Coal-Fired Power Station Solid By-Products
Terry Gourley (Geopolymer Alliance)
Thermal Character of Geopolymers Synthesised from Class F Fly Ash Containing High
Concentrations of Iron and -Quartz
William Rickard (Centre for Materials Research, Curtin University)
Rational Utilisation of Fly Ash for Geopolymer Processing
Temuujin Jadambaa (Centre for Materials Research, Curtin University)
12:15 networking lunch
1:15 [7] EDUCATION
Dan Churach (CSRP) and Jim Avraamides (CSRP)
1:45 [8] BAUXITE RESIDUE
Chair: Evan Jamieson (Alcoa)
Development of a Sustainability Roadmap for the Kwinana Industrial Area (WA)
Karin Schianetz (Centre of Excellence in Cleaner Production, Curtin University)
Production and Application of Red Sand as ReSand
The Role of CSRP in Facilitation of a Sustainability Project
Evan Jamieson (Alcoa)
Understanding Bauxite Residue from the Communitys Perspective A review of a community
consultation in the Peel-Harvey Catchment
Catherine Pattenden (Sustainable Minerals Institute, University of Queensland)
Using LCA to Compare the Environmental Performance of Bauxite Residue, Lime and Bauxsol in
the Treatment of Acid Mine Drainage at Mt Morgan Mine
Daniel Tuazon (University of Queensland)
3:00 Conference Close
Stevan Green (CEO, CSRP)
CONTENTS
Developing SUSOP
A structured, methodical mechanism for incorporating sustainability principles into

the design and operation of minerals processing plants......................................................................................... 6
Thermodynamics of Selective Roasting of Arsenic from Tennantite Containing Copper Concentrates ........... 8
Techno-Economic Evaluation of Early Arsenic Removal from Copper Ores ................................................... 11
Water Conservation in the Mining Industry Case study of gold mining and refining.................................... 13
Overview of the CO2 Breakthrough Program and Linkage to IISI .................................................................... 17
Survey of Sustainable Biomass Resources for the Iron and Steel Industry........................................................ 21
Production of Charcoal for Recarburisation Plant Trials..................................................................................... 23
Piloting the Integrated Dry Granulation and Heat Recovery Process at CSIRO............................................... 26
Minerals, Metals and Sustainability A book for the future .............................................................................. 28
Demand-Supply Interaction on Future Mining Resource Production: The coal model .................................... 30
Applying DEM Outputs to the Unified Comminution Model ............................................................................ 32
Energy Efficient Comminution Circuits A modified grinding strategy and the selection of a target product
size........................................................................................................................................................................... 34
Improving Grinding Efficiency with the IsaMill............................................................................................. 38
Multiple-pass High Pressure Grinding Rolls Circuits ......................................................................................... 40
Update on the JKRBT (JKMRC Rotary Breakage Tester).................................................................................. 42
Improvement of Energy Efficiency of Rock Comminution through Reduction of Thermal Losses ................ 46
Towards a Virtual Comminution Machine........................................................................................................... 48
The New Energy Logging from the Discrete Element Method .......................................................................... 50
Australian Life Cycle Initiative (AusLCI) and CSRP Database: Australian Data............................................. 52
Quantitative Phase Analysis of Fly Ashes for Use as Geopolymer Source Materials....................................... 54
Synthesis Paths and Performance Evaluation of Geopolymer Binder Systems Derived from Major Mineral
Processing and Mining Waste Feedstock Materials............................................................................................. 56
Managing Coal-Fired Power Station Solid By-Products..................................................................................... 58
Thermal Character of Geopolymers Synthesised from Class F Fly Ash Containing High Concentrations of
Iron and u-Quartz ................................................................................................................................................... 60
Rational Utilisation of Fly Ash for Geopolymer Processing............................................................................... 62
Sustainability Roadmap for the Kwinana Industrial Area / Extended Abstract................................................. 65
Production and Application of Red Sand
TM
......................................................................................................... 67
Using Life Cycle Assessment to Compare the Environmental Performance of Bauxite Residue, Lime and
Bauxsol in the Treatment of Acid Mine Drainage at Mount Morgan Mine................................................... 70
Developing SUSOP
A structured, methodical mechanism for incorporating

sustainability principles into the design and operation of minerals processing plants
B. McLellan and G. Corder
University of Queensland, Sustainable Minerals Institute, Centre for Social Responsibility in Mining,
QLD
Introduction, Background and Objectives
The current key activity of the CSRPs
Sustainable Development (SD) Program is the
development of the SUSOP
(SUStainable
OPerations) mechanism for incorporating SD
principles into the design and operation of
industrial processing plants. SUSOP
will fill a
major gap in the current industry approach to SD
integration in design and operation, by providing a
comprehensive and rigorous mechanism to:
Generate feasible plant design and operating
options according to SD principles.
Evaluate the sustainability benefits and
impacts of these options.
Assess each option using an SD-based
decision-support process.
Support the engineering, project management
and subsequent operations of projects.
SUSOP aims to:
Guide the users to the tools they should apply
to evaluate the sustainability impact of their
operations.
Demonstrate the systematic links between
activities across the project or production
cycle.
Provide a framework of SD principles to
guide possible project and production options.
Generate an SD Balance Sheet that illustrates
the benefits and impacts of each option on a
chosen sustainability framework.
Use the SD Balance Sheet to determine the
impact on over-arching SD principles.
Incorporate a dynamic element to predict the
sustainability impacts of each option under
future scenarios.
Methodology
The development of SUSOP draws on three
strands of research:
Examination of literature on the fundamentals
and available SD tools to clarify the state of
play and identify gaps.
Case study development with industrial sites
and consulting companies.
Development of in-house databases and
technology assessment tools to support
SUSOP.
Key Results
The key result of the work so far has been the
development of the SUSOP concept, based on
the collaborative efforts of the CSRP SD program
working group. The acknowledgement that the
processing industries include a vast array of
activities with flow-on effects from pre-feasibility
and planning through to remediation and
restoration have led to the first key element of
SUSOP - entry points across the project and
production cycles. Each industry user can locate
their activity within one of the boxes in Figure 1.
From that starting point, users are guided through
the following process (Figure 2), and will be
directed to key tools and relevant information at
each step.
Figure 1: Multiple entry points on the project / production
cycle (after Allen et al., 1997).
The SD Opportunity Assessment Process for
identification and evaluation of sustainable
development (SD) opportunities is the most
intensive part of SUSOP and requires the most
input from the project proponent or operator.
Figure 2: SUSOP concept diagram.
The initial step, Familiarisation, will ensure the
users and relevant stakeholders are aware of the
meaning of sustainable development and
sustainability in the context of their operation.
Achieving this level of understanding may involve
workshops or training in SUSOP
and SD
concepts.
Following the Familiarisation step, SUSOP
will
guide users through Goal Scoping in an SD and
specific operational sense (i.e. What is the goal of
the project operationally, and what are the key SD
concerns?). Identifying the goals and context
enables the SUSOP
study to be developed
according to an agreed scope.
The Identification step takes the goals and
context, and examines the operation in an effort to
develop options for improving SD performance.
Screening tools and workshops are some of the
elements that can be applied in this step.
Evaluation then guides the user to a selection of
tools or approaches with which to quantify and
rank these opportunities. Examples include life
cycle assessment, ecological footprint, social
assessment, industry ecology, green engineering
etc. A range of approaches will be required for a
comprehensive analysis. An important part of the
research is to identify additional tools that dont
currently exist. Development of these tools will
form part of the research agenda for CSRP 2.
The outputs from this Identification and
Evaluation process then feed information back
into the selected Sustainable Development
Framework. Frameworks such as the 5 Capitals
Model, corporate frameworks or Triple Bottom
Line are all equally valid for use in SUSOP
.
These results will be used to generate an SD
Balance Sheet indicating the benefits and impacts
of the operation or selected alternatives.
The information from the SD Balance Sheet,
combined with Conventional Business Case
data, can then be used in the Decision Support
using tools such as Multi-Criteria Analysis, to
produce a Decision based on the incorporation of
agreed SD principles.
Monitoring and Evaluation is the ongoing
review of the selected options, once implemented.
This step ensures that the expected SD benefits
are achieved and any negative impacts are
minimised. SUSOP
will be able to be applied

periodically or at each stage of development in
order to identify emerging sustainability
opportunities.
Highlights / Benefits
Development of the SUSOP
concept that
incorporates a multiple-entry points approach
across the project and production cycles and
utilises a known or customised SD framework
which is then the basis for generating an SD-
based decision.
Conclusions and Future Direction
The SUSOP
concept is complete. Industry-based

case studies are and will be conducted leading to
the publication of SUSOP
Mk I, a formal
documentation of the concept and process, which
is due for completion in 2010.
Acknowledgements
This work was carried out under the auspices of
the Centre for Sustainable Resource Processing,
which is established and supported under the
Australian Governments Cooperative Research
Centres Program. The authors also acknowledge
the key contributions of the SD Program working
group.
References
1. B.C. McLellan and G.D. Corder SUSOP Concept
Document. CSRP Report. July 2008.
2. Allen, D., F. Consoli, et al. (1997). Public policy
applications of Life-Cycle Assessment. SETAC.
Thermodynamics of Selective Roasting of Arsenic from Tennantite Containing
Copper Concentrates
S. Wright, W. Bruckard and F. Jorgensen
CSIRO, Minerals Down Under National Research Flagship, VIC
Background
The CSRP project 2D8 (Early removal and safe
disposal of arsenic and other minor elements
during base metal processing) examined the
selective removal of arsenic from a copper
concentrate and its safe disposal through
alternative processes. While the concept was
developed several years ago, it was only exposed
and evaluated recently (Jahanshahi et al., 2006).
The proposed approach is applicable to ores
where the mineralogy allows separation by
flotation into high and low arsenic (As) streams.
A sample of high As ore from an NSW mine was
chosen as the raw material to demonstrate the
overall treatment at small scale. Differential
flotation produced two copper concentrates
(Bruckard and Davey, 2007). The high As
fraction required further treatment to produce a
material suitable for blending with the low As
fraction for sale. Roasting has potential to remove
the As from the high As stream and produce a
low-volume fume for disposal (Jorgensen, et al.,
2007). This is a major benefit as dispersion of As
into several streams during smelting is minimised.
Roasting can be performed at the
mine/concentrator, away from urban centres (see
Jahanshahi et al., 2006 for more details).
Encapsulation of the fume in geopolymers, (Brew
and Vance, 2007) or other materials such as slags
and concrete with high resistance to leaching
provide an outlet for the fume in the mine as an
engineering material as well a means of safe
disposal. This paper investigates the conditions
needed to selectively remove As from concentrate
through a roasting step. The approach taken
involves use of thermodynamic free energy
minimisation to simulate the high temperature
reactions taking place during the selective
roasting of high As copper concentrate.
Roasting Simulations
The simulations were performed with the CSIRO
Thermochemistry system (Turnbull and Wadsley,
1988). Composition of the simulated concentrate
is given in Table 1. The thermodynamic data of
all relevant As species was included in the system
definition, as well as the potential reaction
products of the concentrate. The thermodynamic
data for Tennatite was taken from the literature
(Seal et al., 1990).
Table 1: Input species used in simulating roasting in a
neutral atmosphere.
Species Mass (g) Species Mass (g)
N
2
gas 100.0
Cu
10
Fe
2
As
4
S
13
11.2 MgO 1.2
Cu
5
FeS
4
20.2 SiO
2
39.0
CuFeS
2
16.4 Al
2
O
3
8.9
FeS
2
0.27 CaO 2.4
The effects of various initial conditions were
evaluated at temperatures between 600 to 800C
with the focus on the As deportment and the
stability of the As containing phases. The results
of various scenarios follow.
Sulphur Dioxide and Temperature
The distribution of As to the gas phase is shown in
Figure 1 at temperatures between 600C and
800C for gas atmospheres containing up to 4%
sulphur dioxide. In a neutral atmosphere (SO
2
free), the removal of As approaches 100% at
600C. However as the SO
2
pressure increases,
complex behaviour with respect to the As
distribution is observed. This can be attributed to
the relative stability of solid calcium arsenate
(Ca
3
As
2
O
8
), calcium sulphide (CaS) and calcium
sulphate (CaSO
4
). Removal of As from the
concentrate approaches 100% in up to 2% SO
2
atmospheres at 650, 700 and 750C.
0
20
40
60
80
100
0 0.01 0.02 0.03 0.04
SO2 partial pressure (atm)
A
s

d
i
s
t
r
i
b
u
t
e
d

t
o

G
a
s

(
%
)
600
650
700
750
800
Figure 1: Effect of SO2 partial pressure on the selective
removal of As from Cu concentrate.
The stability of calcium phases at 800C is shown
in Figure 2. At low SO
2
partial pressures, CaS is
the dominant phase, but its proportion is reduced
as the SO
2
partial pressure increases, and the
proportion of Ca
3
As
2
O
8
increases. Consequently
less As distributes to the gas phase. Above 4%
SO
2
in the gas, CaSO
4
becomes stable relative to
CaS and Ca
3
As
2
O
8
and the As deporting to the gas
increases. This trend is also repeated at
temperatures below 800C, e.g. Figure 3 shows
the stability range at 650C. As the temperature
decreases, the pSO
2
where CaSO
4
becomes stable
decreases, and is less than 1% at 650C.
0
20
40
60
80
100
0 0.01 0.02 0.03 0.04 0.05
pSO
2
(atm)
C
a

p
h
a
s
e

d
i
s
t
r
i
b
u
t
i
o
n

(
%
)
Ca
3
As
2
O
5
CaS
800C
CaSO
4
Figure 2: Distribution of calcium phases at 800C as a
function of SO
2
partial pressure.
0
20
40
60
80
100
0 0.002 0.004 0.006 0.008
pSO
2
(atm)
C
a

p
h
a
s
e

d
i
s
t
r
i
b
u
t
i
o
n

(
%
)
Ca
3
As
2
O
5
CaS
650C CaSO
4
Figure 3: Distribution of calcium phases at 650C as a
function of the SO
2
partial pressure.
Partial Roasting with Air
Addition of oxygen (O
2
) to partially roast the
concentrate yielded different results than
equilibrating a N
2
-SO
2
gas with the concentrate.
In Figure 4 sufficient O
2
was added to N
2
to
produce SO
2
partial pressures matching those
shown in Figure 1. At the same partial pressures
of SO
2
the arsenic removal from the concentrate
was lower with the N
2
/O
2
input than in the N
2
/SO
2
input. The equilibrium sulphur pressure (pS
2
) is
lower in the case of addition of N
2
and O
2
, and
consequently the equilibrium vapour pressure of
As in the gas was lower. If too much air added so
that the pSO
2
is sufficiently high, then magnetite
(Fe
3
O
4
) and ferric arsentate (Fe
3
As
2
O
8
) can be
formed.
0
20
40
60
80
100
0 0.005 0.01 0.015 0.02
SO2 partial pressure (atm)
A
s

d
i
s
t
r
i
b
u
t
e
d

t
o

G
a
s

(
%
)
700C
N2 + SO2
N2 + O2
Figure 4: Effect of oxygen addition at 700C (plotted as a
function of pSO
2
) on the separation of As.
Addition of Pyrite
Addition of a sulphidant to increase pS
2
increased
As separation. Under conditions where the As
recovery would normally be low (700C,
pSO
2
=0.011 and 800C, pSO
2
= 0.0029) addition
of pyrite (FeS
2
) increased the removal of arsenic
(Fig. 5).
0
10
20
30
40
50
60
70
80
90
100
0 5 10 15
FeS
2
added (g/100g)
G
a
s
e
o
u
s

A
s

d
e
p
o
r
t
m
e
n
t

(
%
)
700C
800C
Figure 5: Effect of FeS
2
added to concentrate on the
recovery of As.
Increasing the sulphur partial pressure increased
the stability of CaS. At 700C maximum removal
of As was achieved by adding 11 g of FeS
2
per
100 g of dry concentrate. At 800C, addition of
25 g of FeS
2
increases the Arsenic deportment to
the gas phase to 100%.
Conclusions
The proportion of arsenic reporting to the gas
phase is related to the temperature and the
thermodynamic stability of Ca
3
As
2
O
8
, CaS and
CaSO
4
. Roasting conditions exist at temperatures
between 600 and 700C where arsenic can be
removed from an arsenic bearing concentrate by
using an a neutral (N
2
) atmosphere or an
atmosphere with a low concentration of SO
2
.
Partial oxidation of the concentrate is not advised,
as the simulations show that the sulphur potential
of the system is lower than the equivalent N
2
/SO
2
gas mixture and that magnetite (Fe
3
O
4
) can be
formed. If too much air is added, ferric arsenate
(Fe
3
As
2
O
8
) becomes stable and arsenic is retained
in the concentrate.
Addition of FeS
2
to the concentrate at
temperatures above 700C increases the sulphur
partial pressure and enhances the stability of CaS
compared to that of Ca
3
As
2
O
8
. This increases As
removal.
There are several potential thermodynamic
operating windows to remove As from an As-rich
copper concentrate. The most appropriate will
depend on the rate at which As can be volatilised
from the concentrate, and the conditions where the
structure of the concentrate and the fuel value of
the concentrate are not compromised.
Acknowledgements
This work was carried out under the auspice of the
Centre for Sustainable Resource Processing,
Centres Program. This project was financially
supported by CSIRO and CSRP.
References
1. Bruckard W.J. and Davey K.J., 2007. Arsenic
Separation in Copper circuits Using Flotation, First
CSRP Annual Conference, Melbourne, 43-44.
2. Brew D.R.M. and Vance E.R., 2007. Arsenic
Remediation Using Geopolymers, First CSRP Annual
Conference, Melbourne, 47-48.
3. Jahanshahi S., Bruckard W.J., Chen C. and Jorgensen
F.R.A., 2006. Management of Minor Elements in the
Production of Base Metals, Green Processing 2006,
Newcastle, 25-30, AusIMM, Parkville.
4. Jorgensen F.R.A., Hall T.P. and Sanetsis S., 2007.
Selective Removal of Arsenic from a Copper
Concentrate, First CSRP Annual Conference,
Melbourne, 45-46.
5. Seal, R.R., Essene, E.J., Kelley, W.C., 1990. Canadian
Mineralogist, Vol 28, pp. 725-738.
6. Turnbull, A,G. and Wadsley, M.W., 1988, The CSIRO
Thermochemistry System, CSIRO Division of Mineral
Products.
Techno-Economic Evaluation of Early Arsenic Removal from Copper Ores
N. Haque and T. Norgate
CSIRO Minerals, VIC
Introduction
There are significant amount of undeveloped
copper ore deposits containing arsenic minerals
(e.g. enargite and tennantite) which can produce
copper concentrates but these are high in arsenic
(0.7%) (Smith and Bruckard, 2007). Arsenic
contents above 0.5% are generally not acceptable
to smelters. Recently, a new process has been
proposed at CSIRO Minerals to separate high
arsenic copper minerals from other copper
minerals in flotation (Jahanshahi et al., 2006;
Senior et al., 2006) based on exploiting
differences in pulp potential and pH. This
technology involves a two-stage flotation process
accompanied by an additional roasting stage.
There are less environmental impacts with this
new process because of reduced arsenic, sulphur
and SO
2
emissions to the land and air. However,
the new proposed process involves additional
capital (roaster and additional flotation units) and
operating costs. The aim of this techno-economic
evaluation study is to provide first estimates of
any additional benefits from the new flowsheet
compared with conventional copper processing
technology and to determine whether these
additional benefits justify the additional costs of
the process.
Methodology / Technique
Process modelling of the conventional (base case)
flowsheet and the new proposed flowsheet was
undertaken using the METSIM process simulation
software. The mass and energy balances obtained
from the METSIM simulations of the respective
flowsheets were used to estimate the operating
costs of both processes. Several assumptions
were made in estimating these costs as outlined
below.
The roaster is assumed to operate at 700C in a
nitrogen atmosphere and the reaction time in the
roaster is 6 min (Padilla et al., 2001). The roaster
fuel energy requirement was estimated to be 0.27
GJ/t of concentrate, while the electrical energy
consumption was calculated to be 1.5 kW based
on the estimated size (5.2 m long, 1.7 m diameter)
of the rotary kiln for the required feed rate.
Key Results / Findings
The estimated operating cost of the proposed
process was approximately 17% higher ($76/t
conc. cf. $65/t conc.) than the conventional
process. The contributions to the operating cost
of both processes are compared in Figure 1. The
cost of labour was the major contributor followed
by electricity and reagents.
0
10
20
30
40
50
60
70
80
90
R
e
a
g
e
n
t
s
E
le
c
t
r
ic
it
y
W
a
t
e
r
L
a
b
o
u
r
C
o
n
s

&

m
a
in
t
F
u
e
l
A
s

s
t
a
b
ilis
a
t
io
n
T
o
t
a
l
O
p
e
r
a
t
i
n
g

c
o
s
t

(
$
/
t

c
o
n
c
e
n
t
r
a
t
e
)
Base case
New process
Figure 1: Comparison of operating cost contributions for
base case and new process.
Estimated revenue for both processes is shown in
Table 1. The price received for the final copper
concentrates from both flowsheets was based on
the copper content (96.7% payable Mining
Journal, December 2006), and an average copper
price over 2006/2007 of US$7000/t (Mining
Journal, November 2007).
Table 1: Revenue from copper concentrate for both
flowsheets.
Parameter Base case New process
Ore (t/h) 1000 1000
Concentrate (t/h) 32.8 34.1
Concentrate grade
- % Cu
- % As
26.4
0.53
25.4
0.08
Cu revenue
- $/t conc
- $/t ore
2234
73.20
2149
73.33
As penalty
- $/t conc
- $/t ore
12.53
0.41
0.00
0.00
Net revenue ($/t ore) 72.79 73.33
Economic benefits of proposed process
Based on the economic results (and underlying
assumptions), the net revenue derived from the
new proposed process over the base case
conventional process was estimated to be $0.54/t
ore. The increased operating cost of the new
process over the conventional process was
estimated to be $0.46/t ore, giving an overall
increase in revenue of $0.08/t ore. However, this
analysis does not take into account the increased
capital cost (additional flotation cells, roaster and
associated equipment). For the assumed base
case nominal flow rate of 1000 t/h ore federate
(approximately 8.0 Mtpa at 92% plant
availability), the additional revenue received after
taking the increased operating costs into account
corresponds to about $0.6 M/y.
Effect of roaster arsenic removal efficiency
The As in the final concentrate product increased
by a factor of about three (2388 ppm cf. 777 ppm)
at 60% efficiency compared with 90% efficiency.
Thus a roaster efficiency of only 60% would see
an arsenic penalty imposed on the final
concentrate of $1.46/t concentrate (or $0.05/t ore),
which reduces the net revenue from $0.54/t ore to
$0.49/t ore, still slightly above the increase in
operating cost ($0.46/t ore).
Benefits of the proposed new flowsheet include
the following:
Potential utilisation of underdeveloped copper
ore deposits that contain otherwise
unacceptable levels of arsenic.
Processing of this ore using the proposed
process is likely to be economically viable.
The penalty for arsenic content (i.e. US$3 per
0.1% As content above 0.2%, averaged from
various sources) in the copper concentrate
product is significantly reduced due to the
reduction of arsenic using the proposed
process.
If the smelter cannot accept copper
concentrates with arsenic content above 5000
ppm, the economics of the new process
improves significantly since the proposed
process is the only option for utilisation of this
ore.
Removal of arsenic from the flotation circuit
during processing, stabilisation and safe
storage or utilisation would yield additional
benefits.
Reduction of sulphur and SO
2
emission from
the smelter.
It should also be noted that smelter limits for
As are projected to become more stringent in
the future which will further enhance the
benefits of this new approach.
This evaluation showed that the yield of
concentrate will be slightly higher with the
proposed process compared with the conventional
(base case) process. This resulted in an increase
in net revenue (copper revenue less arsenic
penalty) of $0.54/t ore with the new process.
However, it should be emphasised that this
evaluation was based on the premise that the
conventional process was capable of producing a
final concentrate with an arsenic content low
enough to make it acceptable to smelters. If no
acceptable concentrate can be produced from the
ore by conventional processing, meaning a high
arsenic ore body cannot be mined and processed,
then the economic benefits of the new proposed
flowsheet increase dramatically, with potential
environmental benefits as a bonus. These first
estimates of the economic benefits of the new
proposed process are sufficiently encouraging to
warrant further development of the flowsheet.
Acknowledgements
This work was carried out with financial support
from CSIRO and the Centre for Sustainable
Resource Processing, which is established and
supported under the Australian Governments
Cooperative Research Centres Program. The
authors gratefully acknowledge Warren Bruckard,
Kevin Davey and Frank Jorgensen from CSIRO
Minerals for their assistance during this study.
References
1. S. Jahanshahi, W.J. Bruckard, C. Chen and F.R.A.
Jorgensen Management of minor elements in the
production of base metals. In: Green Processing 2006
3rd International Conference on the Sustainable
Processing of Minerals, 5-6 June, Newcastle, Australia,
AusIMM & CSIRO. pp 25-30. 2006.
2. R. Padilla, Y. Fan and I. Wilkomirsky Decomposition
of enargite in nitrogen atmosphere. Canadian
Metallurgical Quarterly 2001, 40(3): 335-342.
3. G.D. Senior, P.J. Guy and W.J. Bruckard The selective
floatation of enargite from other copper minerals a
single mineral study in relation to beneficiation of the
Tampakan deposit in the Philippines. International
Journal of Mineral Processing 2006, 81:15-26.
4. L.K. Smith, and W.J. Bruckard The separation of
arsenic from copper in a Northparkes copper-gold ore
using controlled-potential flotation. International
Journal of Mineral Processing 2007, 84:15-24.
Water Conservation in the Mining Industry Case study of gold mining and refining
R. Cocks, G. Ho
1
, S. Dallas
1
, M. Anda
1
and Aidan Giblett
2
(industry supervisor)
1
Murdoch University, Environmental Technology Centre, WA
2
Newmont, WA
Introduction
Water is an essential input to mining and mineral
processing and demand is increasing due to
growing activities. Climate change has resulted in
reduced water availability and this is the case in
arid parts of Australia, where many mining and
mineral processing operations take place.
Governments in Australia now require major
water users to prepare a Water (Efficiency)
Management Plan to achieve reduction in water
consumption. Water auditing is now a well
established tool for conducting a systematic,
structured and rigorous water use performance
assessment that includes costed options for water
conservation. There are benefits resulting from an
industry conducting a water audit that is not only
due to the savings in water costs, but also the
heat/energy and materials dissolved or suspended
in the water. Reduction in heated process water
results in reduction of not only water volume, but
also the cost of heating the water. Reduction in
water volume may also mean reduction in the cost
of pumping the water. Water conservation
includes reduction in the amount of pollutants in
water streams or load of pollutants. Final disposal
of the water will impact less on the environment.
If the pollutants are nutrients (Nitrogen and
phosphorus) the water will be suitable for
landscaping. Minimising final disposal of
wastewater is an objective of water auditing and
water conservation. Zero discharge of stormwater
is a requirement in many cases at mining sites,
and zero emission of contaminated water may be
feasible and an objective to be considered as part
of achieving sustainable development.
Water Auditing and Water Conservation
The water audit process commences with a review
of the audit scope and objectives. Once the
domain of the audit has been agreed, water flows
into and out of the domain are determined and
quantity and quality are measured. Water auditing
is different to environmental auditing ISO 14000
in that quantifying and qualifying water use for
the purposes of auditing water is only covered in
part by ISO 14010 and 14011, and the emphasis
of environmental auditing is compliance with
objectives. The water auditing process will, on the
other hand, provide the objectives for
environmental auditing by recommending or
specifying objectives for achieving water
conservation, that is the water conservation
strategy (Sturman et al., 2004).
In the water audit process water and material
balances are fully carried out in a systematic
manner. Once the scope and objectives of the
water audit have been established a schedule is
drawn up and may include a selection list of the
people that would contribute to, and or be part of
a water audit team. Resources are also tabulated in
the context of availability, and resources that may
be required and would need to be introduced into
the audit process, and for example the need for a
clamp-on Acoustic Doppler Velocimeter for water
flow measurement and meter calibration. A
proposed water audit procedure includes many
facets and starts with a flow diagram of the water
audit domain that identifies the number of unit
operations. Once a site-wide water balance model
has been produced, recommended future changes
to operations and processes are weighted against
the financial capacity for implementing change
and the economic justification to do so within the
scope of water conservation. The collation of data
from an initial site visit underpins a
methodological approach towards conducting a
water audit and achieving a foreseeable outcome
in terms of audit closure. Closure refers to the
difference between input and output flows or
material fluxes, and usually a maximum 10 %
difference is an objective. This also integrates
with the resources on hand and the ability of the
mining company to provide additional support for
ongoing measurement and assessment of the water
balance account.
Application to Gold Mining and Refining
Newmont The Gold Company owns and
operates the Waihi Goldmine in New Zealands
North Island, the Tanami Granites Operation in
the Northern Territory and the Jundee-Nimary
Mine in the Little Sandy Desert north of
Kalgoorlie, WA. Joint venture mines include the
Fimiston-Gidji operation run by Kalgoorlie
Consolidated Gold Mine (KCGM) in a 50-50
share arrangement with Barrick Gold Australia
and the Boddington Gold Mine where 66.67 % is
owned by Newmont and 33.33 % by AngloGold
Ashanti. An approach to sustainable sourcing, use
and reuse of water was met through the use of the
water auditing tool.
Water balance models have been designed for
each case study and in turn set a framework for
water auditing exercises to be carried out in
ongoing research. Historically, Newmont has
audited water use in the refining operation at its
respective mine sites. However, a general audit of
a site-wide water account has yet to be
maintained. A possible exception is the Waihi
Gold Mines compliance with New Zealands
Resource Management Act 1991(RMA) that sets
out a series of amendments pertaining to mines
using land, beds of lakes and rivers, the use of
water and water quality on discharge or seepage
from mining operations. To comply with water
quality conditions set by The RMAs Resource
Consents and Water Rights Bill, comprehensive
routine monitoring of flow rates and water quality
is carried out. With regard to process water
monitoring, the Tanami Granites Operation is by
far the most efficient water-user over its gold mill
and personnel accommodation domain.
Furthermore, several conclusive water audit
exercises have been conducted in 2001 and 2003
at the Granites. The recorded results bear
testament to the validity of conserving water and
accounting for every kilolitre used in gold-ore
processing. Therefore, the question is raised as to
how much raw water will be conserved by
accounting for all water uses and establishing
closure in a site-wide water audit at any of
Newmonts gold mine sites. An overall objective
of carrying out a water audit is to arrive at a water
management strategy for each mine site.
Newmonts water data gathering is
comprehensive both with meteoric water flows
that are described as non-process water stocks and
process water flows used in all aspects of gold
extraction from pits and declines through to
crushing/milling of gold ore and final gold
recovery. At each mine site water data were
collated to form an overall picture of various
water sources, flow rates and end uses.
Information gathering included contacting staff to
retrieve data and opening archival IT systems
where historical data could be obtained. As a
general rule the metallurgical departments were
the greatest source of information followed by
process operations staff and bore/dewatering
personnel. Data acquisition workbooks displayed
an annual consumption rate of raw and recycled
water stocks and excel spreadsheets produced
daily and weekly water consumption rates. From
this information water balance flow diagrams
were produced to inform a methodological
approach and further research by examining site-
wide water flows via a formal water audit
exercise. Examples of both non-process and
process water schematics are illustrated in Figure
1 and Figure 2 respectively:
Figure 1: Pre-audit site-wide water schematic of process
water sources and uses.
Figure 2: Pre-audit site-wide water schematic of non-
process water sources and flows.
Within a water management strategy, continuous
improvement of maintenance procedures support
site-operating systems and a structured
methodology monitors key performance indicators
for ongoing full compliance of, for example;
Waihi Gold Mines Consents permit. Each mine
site was examined in terms of fit for purpose
water use in various operations and for example;
types of water used in dust suppression,
dewatering, site services and elution/gold
recovery. Furthermore, in non-disturbed areas of a
mine sites domain, non-process or environmental
water flows are examined. This methodology aims
to build in water quality to the pre-audit
conditions of achieving audit closure and the
inclusion of a hierarchical table of water quality
and uses (Figure 4).
Results
Figure 3 shows the amount of raw water used per
ounce of gold produced at each mine site. The
figures are a reflection of raw water used for
crushing and milling ore and the amount of gold
produced in any one of the gold recovery
operations. Factors affecting the amount of raw
water used include the amount of gold produced
per tonne of processed ore, where water use
increases with a lower ratio of recoverable gold to
ore tonnage. Other factors not tabled include the
geochemistry of ore relating to rock types and
processes required to recover gold from differing
geological matrices. An example of the difficulty
of gold recovery from a certain ore type is
experienced at KCGM where the high sulphide
content in the Fimiston Pit ore requires further
processing at the Gidji Roaster plant and using
more water to recover the gold from an extra
processing cycle. Therefore, the comparisons are
only an indication of water use efficiencies that
will assist in benchmarking and building an over
all framework of a mine site water account and
ultimate water audit closure. Even though KCGM
uses more water than the other mine sites, a total
water audit of the KCGM domain would lead to
further efficiencies in its gold operations and
water uses.
Figure 3: Comparison of water efficiencies.
The future direction with NZs RMA 91 is that
management of non-process water diversion,
collection and treatment is central to the effective
control of environmental conditions over Waihi
Gold Mines leased area. Typically, sites with an
effective water management strategy maintain a
water balance model as an effective tool for
informed risk based decision making and this
applies to all Newmont mine sites.
The National Water Initiative underpins the
implementation of water use strategies at state
levels. Western Australias State Water Allocation
Strategy advocates the use of water metering at all
stages and classifications of water used in the
operation of, for example; any mining venture.
Within the strategy, a new metering program is
proposed and may be rolled out on a priority basis
with a view that comprehensive metering will
inform government agencies on the draw of water
from the consumptive pool of a region.
To achieve reliable, cost-effective and efficient
performance in water extraction and use, mine
sites will establish a rigorous basis for water
resource management and trading in order to
provide the best available monitoring and
measurement information on water availability
and use. This follows on from the States Rights
in Water and Irrigation (Approved Meters) Order
of 2003 and is relevant to the standards,
technologies and operating practices under review
by the National Water Commission. User-
compliance
with water access entitlement will be monitored
where responsibility lies with user (mine sites)
water meter reporting. The water classification
table sets the basis for a generic application of a
hierarchical approach to a water management
strategy and water audit process.
References
1. Sturman, J., Ho, G., Mathew, K. 2004. Water Auditing
and Water Conservation. London: IWA Publishing.
Water source TDS Mine usage Approx. usage Classification
Scheme < 1,500 mg/L Accom/ops < 1% Potable
Bore potable < 1,500 mg/L Accom/ops
RO elution
< 5% Fresh
Bore brackish 1,500-15,000
mg/L
SAG/Ball mills < 20% Raw
Bore saline 15,000-35,000
mg/L
SAG/Ball mills < 20% Raw
Paleo-channel
Hyper-saline
> 35,000 mg/L Dust suppression Variable Raw
Tailings decant 15,000-35,000
mg/L
Sag/Ball mills 60% of raw water
inflow
Recycled
Treated wastewater 1,500-15,000
mg/L
Revegetation
Tailings dam
< 1% Recycled
Hydrocyclones Barren
elate
15,000-35,000
mg/L
Sag/Ball mills Up to 6 cycles of
re-use
Re-used
Meteoric < 1,500 mg/L Tailings dam
Water storage
Variable Raw
Figure 4: Water Classification Table.
0
500
1000
1500
2000
2500
C
o
p
p
e
r
N
ic
k
e
l
L
e
a
d
Z
in
c
A
lu
m
in
iu
m
S
t
e
e
l
G
l
o
b
a
l

G
H
G

e
m
i
s
s
i
o
n
s

(
M
t

C
O
2
e

/

y
)
0
10
20
30
40
50
A
u
s
t
r
a
l
i
a
n

G
H
G

e
m
i
s
s
i
o
n
s

(
M
t

C
O
2
e

/

y
)
Global Australia
Figure 1: Australian and global annual greenhouse gas
emissions for various primary metals (Norgate et al).
9
Towards Carbon Neutral Metal & Cement Production
Reduce GHG &
Limestone
System innovation builds on industrial ecology, regional synergi es,
and collaboration across normal business boundaries
VALUE
IMPACT
Wheat Power Iron & Steel Cement
Bio -
mass
Char
Farming Pyrolysis
Tree
Planting
Soil Quality
Zero Waste
-
Smelting
Planting
Soil Quality
(Salinity Control)
GHG Neutral Renewable
Energy
Cement
Making
Slag Dry
Granulation
Power
Reduce Emission
& Water Usage
Slag
Hot
- Reducing GHG Emissions, Fresh Water Usage and Salinity
Figure 2: Conceptual flowsheet for low GHG steel and
cement production using biomass derived charcoal and dry
granulation of slags.
Overview of the CO2 Breakthrough Program and Linkage to IISI
S. Jahanshahi
1
, J.G. Mathieson
2
and P. Ridgeway
3
1
CSIRO, Minerals Down Under National Research Flagship, VIC
2
BlueScope Steel Research, NSW
3
OneSteel, NSW
Introduction
Currently the steel industry is a moderate
contributor to global GHGs, but this ranking may
change given the initiatives on reducing GHGs by
the power generation and transport industries and
the rapid growth in demand for steel. Virgin
production typically produces around 2 t-CO
2
e/t-
steel it is thus imperative to identify major
opportunties for reducing the industrys energy
consumption and GHG emissions and develop
sustainable solutions/technologies in response to
the challenges ahead.
Major Opportunities in Reducing GHG
Emissions
CSIRO life cycle assessment studies have
indicated that extraction and refining of metals
from concentrates are the most energy intensive
steps in the mineral cycle
(i,ii)
, and are thus
responsible for the major portion of GHG
emissions. Taking into account the tonnage of
different metals being produced in Australia and
globally, it then becomes clear that major
opportunities for reducing GHG emissions in the
sector are in the production of steel from iron ore
and aluminium from bauxite (see Figure 1).
Through collaboration with the steel industry,
CSIRO and CSRP are involved in development
and demonstration of technologies for step-change
reductions in GHG emissions in metal extraction
and refining processes. These include:
use of renewable carbon as fuel and reductant
integrated heat recovery and dry granulation
of slags.
These technologies could be linked in the value
chain, as shown in Figure 2, to result in a
reduction of over 500 million tonnes per annum of
GHGs globally. It is interesting to note that the
potential reduction in GHGs resulting from
implementation of these technologies is similar to
the current level of total GHG emission by
Australia.
Renewable Fuels and Reductants
The steel industry is almost entirely dependent on
fossil carbon for fuel and reductant and
contributes significantly to GHG emissions in
Australia and globally. Processed biomass (e.g.
char) is a potential source of renewable energy
and carbon for metallurgical reactors, providing
an essentially GHG-neutral form of fuel/reductant.
By-products (e.g. bio-diesel or electricity) from
production of char from biomass can also
contribute to reduction of GHGs from industry.
The supply of biomass should (and can) be
managed in sustainable or even restorative ways
to avoid adding stress to agricultural and forestry
eco-systems. Our studies have shown that partial
replacement of fossil carbon by bio-char over the
next decade or two will create a demand for
biomass supply of up to 10 million tonnes per
year. Such volumes of biomass could
theoretically be supplied in a sustainable way
through use of large volumes of forestry and saw
mill residues currently being produced and/or
plantation of short rotation-cycle woody biomass,
such as the oil mallee that could meet the
metallurgical carbon requirements of Australias
steel industry. The innovative challenge is to find
viable solutions that capture the inherent
sustainability opportunities associated with the
use of biomass char and other forms of renewable,
recycled or waste carbon. Furthermore, by using
deep-rooted biomass that has been grown in
salinity-prone areas, we will also help to improve
soil quality in those areas. It is worth noting that a
plantation program of short rotation trees (oil
mallee) has already commenced in the WA wheat
belt to lower the water table and hence address the
salinity issue in this region. Full rehabilitation of
salinity-affected farmland will produce 25 million
tonnes per annum of biomass. The local and
overseas steel industries could potentially provide
an excellent market opportunity for such volumes
of renewable carbon.
In addition to the environmental benefits
associated with substitution of biomass-derived
charcoal for coal, there are also some economic
benefits. Studies by Mathieson
(i)
have shown the
value-in-use of pyrolysied woody biomass could
be up to 80 per cent higher than a reference coal
used for pulverised coal injection into a modern
blast furnace. This study shows that conditions
used for pyrolysis of biomass could result in
production of a range of chars with varying value-
in-use, thus it may be possible to design and
produce semi-charcoals that optimise blast
furnace heat and mass balance. This concept of
designer chars is being pursued through the
International Iron & Steel Institutes CO
2
Breakthrough Program, and is focusing on the
production of chars with tuned properties for
different applications in the iron and steel making
processes. These chars are aimed at replacing a
significant portion of the fossil fuel and reductant
used by the industry. In particular, we aim to
develop technologies to replace 100 per cent of
tuyere-injected fuels, i.e. about 30 per cent of
blast furnace fuel (excluding stoves), 50 per cent
of solid fuel used in sintering of iron ore, 2 to 10
per cent of coking coal blend and 100 per cent of
carbon used for recarburiser and slag foaming
agent in steel making over the next 5 to 15 years.
Integrated Heat Recovery and Dry
Granulation of Slags
In most modern smelters, a significant portion of
high-grade waste heat is recovered from processes
through the off-gas handling systems. However,
the thermal energy that resides in the slag is lost to
the atmosphere. Capture and use of such high-
grade waste heat is technically feasible and
economically attractive. Studies have shown
large potential benefits associated with the
application of suitable technology in integrated
metallurgical plants. The challenge has been in
development of a versatile technology platform
that maximises use of by-product energy and
material. This is being addressed through
development of an integrated process for heat
recovery and conversion of the slag into a glassy
by-product suitable for cement production.
An integrated steel plant producing one million
tonnes of steel a year, also produces 300 000
tonnes of slag. The Australian steel industry
produces more than two million tonnes of slag
each year. Blast furnace slags are generally either
water-granulated to produce glassy particles that
can be used in cement production or air-cooled in
large pits for land-fill or low-value applications
such as road base materials. Steelmaking slags are
generally air-cooled and land filled or used in
road-based applications.
Since 2002, CSIRO has been involved in
development of a new technology for waste heat
recovery and dry granulation of slags. The
CSIRO process involves feeding molten slag on to
a disc rotating at high speed. This breaks the slag
into small droplets that solidify rapidly, producing
glassy granules with similar properties to those
produced by water granulation. The granules can
be used as supplementary cementitious material in
the production of cement, which is a key
constituent of concrete. Globally, about 2.3 billion
tonnes of cement is produced every year.
Producing one tonne of Portland cement
consumes about 3000 mega-joules of electrical
and thermal energy and emits about 800
kilograms of carbon dioxide. Granulated slag can
substitute for up to 70 per cent of Portland
cement, leading to significant energy and GHG
reductions.
The conventional wet granulation method
involves significant capital costs, consumes large
amounts of water, generates acid mists and does
not recover any of the valuable heat. With lower
capital cost and benefits in heat recovery and
reduced pollution, dry granulation is an attractive
alternative to conventional wet granulation. A
recent economic assessment of CSIROs new
integrated process identified potential annual
savings in fuel costs alone of two to three dollars
per tonne of steel, which translates to about $20
million annually for the Australian industry and
over $3 billion for the global steel industry.
Successful development and commercialisation of
the technology in Australia will result in
reductions of GHG emissions by millions of
tonnes, reduction of water consumption by
thousands of millions of litres and conversion of
millions of tonnes of by-product slag into cement.
CSIROs pilot-plant work has provided insights
into process design and control, and how slag
atomisation can be optimised to produce fine
granulates. The results have demonstrated how
enhanced fast cooling can improve granulate
handling and have assisted in the design of a
compact unit to reduce cost and take advantage of
heat recovery. Work is now progressing on
optimising the scale-up of the integrated process
using a combination of computational fluid
dynamics modelling and pilot-plant measurement,
product evaluation and plant measurements (e.g.
current process slag temperature off-takes) at
OneSteel and BlueScope Steel sites.
Value of the Projects Industry
Perspective
Industry engagement is a critical success factor in
development and implementation of the emerging
technologies.
BlueScope Steel and OneSteel are the leading
firms in the Australian iron and steel industry,
employing over 20,000 people across several
hundred sites, contributing over $1.6 billion per
annum in exports, and servicing customers in the
manufacturing, infrastructure, agriculture and
building & construction sectors. Steel
manufacture accounts for approximately 3% of
Australia's annual greenhouse gas emissions. The
iron and steel manufacturing industry, which is
emissions-intensive and trade-exposed, is one of
the industry sectors that will be most affected by
the proposed Australian Carbon Pollution
Reduction Scheme (CPRS).
BlueScope Steel and OneSteel believe reducing
GHG emissions is a global problem that requires a
global response. Our companies recognise the role
of the Australian iron and steel industry, along
with other sectors of the Australian economy, and
the global industry, in transitioning to a low
emissions economy in a sustainable and
economically responsible way.
Whilst viable incremental energy and greenhouse
efficiency improvements will continue to be
sought, realistically any breakthrough iron and
steelmaking technology to radically reduce the
emission of CO
2
in the iron reduction process is
likely to be medium-term (5-10 years) to long-
term (greater than 10 years) in terms of
timeframes required in new technology
development, commercial proving, and take-up in
capital lead times within companies in an industry
where assets are long-lived between major
upgrades.
Both BlueScope Steel and OneSteel are actively
supporting the dry slag granulation and heat
recovery and biomass R&D projects given their
theoretical potential to deliver some significant
GHG savings in the medium term, with some
potential within the short-term. In supporting the
technical work of CSIRO Minerals, we as a team
contribute on strategic direction, technical
application advice, industry R&D and in-kind site
trial and sample support, apart from cash funding
contributions.
In regard to the biomass project, if 10% of the
carbon consumption of the world iron and steel
industry was able to be replaced with biomass
(charcoal), this would be approximately
equivalent to abating the GHG of Australia's
national GHG inventory per annum. The
CSIRO/BlueScope/ OneSteel biomass project
reported on at this conference is a valuable set of
work where the product chain from biomass
material sourcing, transport and logistics,
pyrolysis techniques, and application is being
researched. Just as 'oils ain't oil's, charcoals are
not all the same. Charcoals must be designed in
properties against particular applications with the
steel industry from steel recarburisation, to coke-
making additions, to sintering, to blast furnace
injection. This then required differing pyrolysis
condition, and hence a pyrolysis unit that can dial
up different charcoal properties (% moisture, %
volatiles etc). To make use of forestry and other
biomass source waste residues, this pyrolysis
unit then must consider feed variation
requirements. There are linkages, opportunities
and barriers that must be investigated between
these process chain steps in consort.
The work also seeks to bridge the 'chicken and
egg' syndrome of steelmakers waiting for biomass
availability, and potential biomass producers
waiting for steelmakers to generate a stable
market. The project seeks to fast-track R&D in
biomass uses that have potential for early
application, even if not involving large GHG
abatement, such as use as an alternative for steel
recarburiser. In this way both the steel industry
and external parties down the supply chain will be
shown practical demonstration of biomass use to
stimulate interest and support in other larger areas
of potential biomass application. It also provides
information key to other areas of work.
The first plant trial on use of bio-char in steel
making for recarburiser at OneSteels Sydney
Steel Mill is currently in preparation to be
undertaken with the next few months. Parallel
activities on sustainable supply of biomass and
techno-economics of the technologies and supply
chain have been commissioned through
collaborative projects involving a number of
stakeholders representing mineral and agri-
industries, local councils, state and local
governments, investors, other technology
providers and steelmaking companies.
The project on dry slag granulation and heat
recovery has also attracted interest from the
Australian steel industry, cement producers and
engineering firms. The project team at CSIRO is
currently evaluating the conceptual integrated
process through a pilot plant campaign at CSIRO
before plant trials at one of the iron making plants
in Australia. There is already very strong interest
expressed to CSIRO from parties around the
world in the CSIRO prototype work, and it
appears to have significant commercial value
if the research and development work reaches it
goals and becomes commercially viable and
proven. This project has several environmental
attributes as discussed in the conference report,
and in particular for OneSteel where traditional
wet slag granulation has not been viable in
utilising its generated blast furnace slag for
cement application, a breakthrough in this
technology is of interest. Particularly that it is dry
granulation rather than wet, is a very dry region of
Australia relying on the Murray River.
The International Iron & Steel Institutes CO
2
Breakthrough Program has recently acknowledged
the transformational nature of the biomass and dry
granulation projects and enrolled these R&D
projects as part of its worldwide portfolio of
projects. This allows engagement of a broader
range of collaborators and sponsors, and possibly
quicker technology uptake by overseas iron and
steel producers in the future.
Acknowledgements
This work was carried out under the auspices and
with the financial support of the Centre for
Sustainable Resource Processing, which is
established and supported under the Australian
Governments Cooperative Research Centres
Program. Financial support from CSIRO, through
the Minerals Down Under National Research
Flagship, OneSteel and BlueScope Steel is also
acknowledged.
References
1. T. Norgate, S. Jahanshahi and W.J. Rankin, Assessing
the environmental impact of metal production processes,
Journal of Cleaner Production, 15, pp.838-848, 2007
2. T. Norgate and S. Jahanshahi , Opportunities for
reducing energy consumption and greenhouse gas
emissions in mineral processing and metal production;
In Proceedings of the Chemeca 2007 conference,
Melbourne, 2007.
3. J. G. Mathieson, The Value-in-Use of Some Biomass-
Derived Blast Furnace Injectants, BlueScope Steel
unrestricted report, BSR/N/2007/071, December 2007.
Survey of Sustainable Biomass Resources for the Iron and Steel Industry
N. Haque
1
, M. Somerville
1
, S. Jahanshahi
1
, J.G. Mathieson
2
and P. Ridgeway
3
1
CSIRO Minerals, VIC
2
BlueScope Steel Research, NSW
3
OneSteel, NSW
A collaborative project between BlueScope Steel,
OneSteel and CSIRO through CSRP commenced
in late 2006 to identify, evaluate and demonstrate
specific opportunities where biomass or
sustainable and renewable carbon can be used in
iron making and steelmaking processes. As one
of the activities of this project, a survey of
currently available biomass resources from
several regions of Australia was carried out and is
reported here. Residues have been targeted
because of their current low value, waste disposal
issues and being readily available in some parts of
Australia. Potential waste biomass categories are:
woody biomass from forestry and wood
processing, residues from agriculture and
horticulture and biomass from woody weeds.
Figure 1: Map showing the regions for the survey of
biomass.
Methodology
The data collection was based on interviews and
meetings with personnel from the state forest
agencies and wood processing industries. Other
data were collected from the open literature,
including forest management plans and plantation
inventory data. The amount of non-forestry
resources was estimated from the production of
the agricultural and horticultural industries.
The green wood volume has been converted to an
oven-dry biomass equivalent, based on reasonable
conversion factors for harvest recovery in the
forest, greenmill recovery, drymill recovery in the
sawmill (Ximenes et al., 2005; Ximenes and
Gardner, 2006), wood density (Bootle, 2005;
Kininmonth and Whitehouse, 1991) and moisture
content (MC). Finally, the estimated amount of
residues has been reported as equivalent oven-dry
tonnes unless otherwise stated. Dry weight has
been taken as the basis because biomass achieves
this stable reference condition after drying at
1052C for 24 hours. Prices are currently paid
for the bone-dry weight of wood chips by the fibre
based mills and the well-established export
market.
Total dry residues from all regions were estimated
to be around 7.5 Mt per year. Forest residues (e.g.
wood chips, reject logs, out-of-specification logs,
bark, stump, branches, foliage (leaves and twigs)
and other biomass in the tree crown) are 45% of
total residues. Wood processing residues (e.g.
chips, sawdust, shavings, off-cuts) are 30% of
total residues. Non-forestry residues (e.g.
biomass from grain crops such as wheat and
maize, olive pomace, grape skin, almond waste,
cut flower waste and sugarcane based residues
(e.g. bagasse, infield cane crop residues), woody
weeds such as Camphor Laurel and waste from
macadamia nut processing) account for 25% of
the total biomass resources. Sample residues are
shown in Figure 2.
Figure 2: Photos of sample residues in the forest and during
processing.
The estimated weights of biomass residues are
shown in Table 1. Residues generated in the
forest during harvesting are largely uncommitted
Northern NSW
Central NSW
South East NSW
Southern NSW SE SA, SW Vic
Southern Vic
by the plantation owners. In NSW, the majority
of the plantations are owned by state forests,
whereas in Victoria and the GT region, plantations
are owned by both state forests and also private
companies. The electrical power generation
companies are the sector most interested in these
residues. Recently, there is some interest from the
bio-fuel companies for making bio-crude or
transport fuel.
Table 1: Consolidated annual production of biomass
residues from all regions studied.
Resource category Thousand dry tonnes
per annum
Forestry harvesting residues 3,385
Wood processing residues 2,255
Non-forestry residues 1,905
Total 7,546
Figure 3 shows the estimated requirements for
charcoal by the steel industry in Australia for
various categories of potential applications
(Langberg et al., 2007; Mathieson, 2008). In the
medium term (5-10 years), the theoretical
requirement will consume about 24% of the
charcoal making potential (2.3 Mt/yr) from the
sustainable biomass available from the regions
investigated. However, based on 50% availability
of total biomass (i.e. from 7.6 Mt/yr to 3.7 Mt/yr)
for making charcoal, 48% of the total estimated
charcoal can supply all the iron and steel
industrys requirements.
Figure 3: Estimated charcoal requirement for the Australian
steel industry (*assumes R&D success, and commercial
viability).
This assumes that the Australian iron and steel
industry requirements are represented by the two
main steel companies, BlueScope Steel and
OneSteel. This will also depend on the success of
R&D into the supply chain, pyrolysis, and
biomass application in terms of technical,
environmental and economic parameters.
Potential utilisation of, and value-adding by,
undervalued residues from the forestry,
agricultural and horticultural industries.
Likely large net reduction of CO
2
emission
from the iron and steel industry.
There appears to be sufficient biomass residues to
supply the estimated theoretical charcoal
requirements that may potentially be utilised in
the Australian iron and steel industry for
metallurgical processing in the short to medium
term.
It is envisaged that this study will be expanded
into a more detailed and focussed evaluation of
biomass sources concurrent with the other
research and development works on biomass
processing, transport, pyrolysis and steel industry
application.
The necessity for high quality biomass for
charcoal making (low contamination as evidenced
by ash content and low variability of products)
also requires future study on the issues related to
supply of high quality biomass from existing
sources.
Acknowledgements
This work was carried out by CSIRO under the
auspice of the Centre for Sustainable Resource
Processing, which is established and supported
under the Australian Governments Cooperative
Research Centres Program, with financial support
from BlueScope Steel, CSIRO, OneSteel and
CSRP.
References
1. K.R. Bootle Wood in Australia: Types, Properties and
Uses. 2nd Edition. McGraw-Hill Book Company,
Sydney. 2005.
2. J.A. Kininmonth and L.J. Whitehouse Properties and
Uses of New Zealand Radiata Pine. Volume 1: Wood
Properties, Forest Research Institute, Rotorua, New
Zealand. 1991.
3. D.E. Langberg, M.A. Somerville and T. Norgate The
use of charcoal in the iron and steel industry. DMR
3278, CSIRO Minerals. 2007.
4. J.G. Mathieson Private Communication, Clayton, Oct
2008.
5. F.A. Ximenes, W.D. Gardner and J.F. Marchant
Carbon flow following the harvest of blackbutt trees
and their conversion into sawn productions. Research
Report No 41, DPI NSW. 2005.
6. F.A. Ximenes and W.D. Gardner Biomass allocation
following radiata pine thinning operations in southern
NSW forests. CRC for Greenhouse Accounting Draft
Report by DPI NSW. 2006.
Production of Charcoal for Recarburisation Plant Trials
M. Somerville
1
, J. Mathieson
2
, P. Ridgeway
3
, and M. Davies
3
1
CSIRO Minerals, VIC
2
BlueScope Steel, NSW
3
OneSteel, NSW
The recarburisation of molten steel has been
identified as a suitable first step in the utilisation
of renewable and sustainable carbon in the iron
and steel industry
1
. Recarburisation is the
addition of extra carbon to molten refined steel to
produce particular steel grades. In general
recarburisers need to have a high carbon content
(>95 %), with low ash, volatiles and moisture.
Traditional recarburisers include materials such as
calcined anthracite or other high grade carbon
sources but could also include charcoal. Previous
kilogram-scale work conducted at CSIRO
2
showed that:
Charcoal readily dissolved into molten steel at
high rates of recovery
Carbon recovery to steel was directly related
to the volatile content of the recarburiser
Alkali components in the recarburiser ash can
be reduced and report to the gas stream.
In light of the success of this preliminary work,
the decision was made to proceed to plant-scale
recarburisation test work using charcoal at
OneSteels Sydney EAF Steel Mill (SSM) at
Rooty Hill during 2008-09.
The major part of the planning for these trials was
the preparation of a large amount of charcoal (3-4
tonnes). The key specification of the charcoal
was a volatile and moisture content of less than 1
wt %. Previous experience indicated that a
pyrolysis temperature of 1000 C would be
required to produce charcoal of this quality.
The Illawarra Coke Company (ICC) agreed to
support the project by allowing one of their coke
ovens at their Corrimal plant to be used in a
preliminary test and a further five ovens to be
used to produce the main batch of charcoal. In
normal operations 20-30 tonnes of fine locally
sourced coal is loaded into the ovens through a
port in the oven roof. Heat is generated through
the combustion of volatile components above the
coal bed. At the end of a production cycle (2-3
days), the coke is pushed from the ovens using a
ram. The coke is then quenched, crushed and
sized The coke ovens at Corrimal were suitable
for this charcoal making exercise because they
can provide a stable high temperature of about
1000 C and the ovens were available with
convenient access. Heat was available in the
empty coke ovens from hot walls and from
operating adjacent ovens.
This presentation will cover some experimental
results generated during the preliminary and main
charcoal making process at Corrimal.
Preliminary Trial
Methodology / Experimental Technique
The aim of the preliminary trial was to determine
the time required to produce low volatile charcoal
from supplied E. Rossie wood (or scribbly gum)
using the Corrimal coke ovens operating at about
1000 C.
Seasoned wood was cut into lengths of about 1
metre and had a diameter of between 3 and 10 cm.
The wood logs were loaded into a standard 200
litre fuel drum. Thermocouples fitted to the
drums measured the wood and furnace
temperatures during the trial.
Three drums of wood were added to a recently
emptied oven. One drum was removed after 2, 4
and 6 hours. Once removed, each drum was
quenched with water to prevent burning of the
charcoal. Representative samples of the charcoal
were collected and analysed at HRL
Technologies. Figure 1 shows a photograph of
the drums of wood inside the Corrimal coke
ovens.
Figure 1: Photograph of drums of wood inside the Corrimal
coke ovens.
Figure 2 shows the temperature within each drum
and the temperature of the coke oven during the
pyrolysis trial. These plots show the oven
temperature reached a maximum of about 1130
C, probably due to combustion of the evolved
volatiles, after about 1 hour, before decreasing to
about 960 C after 6 hours. The wood reached the
oven temperature, about 1000 C, after 2 hours,
before decreasing slightly over the next 4 hours to
about 980 C.
Table 1 shows the proximate analysis of the initial
wood, the three samples of charcoal (2, 4 and 6
hours) and a typical Recarburiser used at SSM.
The high moisture content of the charcoal
samples at about 70 % is due to the water
quenching of the hot charcoal. The 4 and 6 hour
charcoal samples showed increasing ash contents
and a corresponding decrease in fixed carbon
content. This is most likely due to extra burning
of the charcoal during the extended stay in the
oven.
0
200
400
600
800
1000
1200
7:00 8:00 9:00 10:00 11:00 12:00 13:00
Time
T
e
m
p
e
r
a
t
u
r
e

(
C
)
drum 2
drum 1
drum 3
oven
Figure 2: Wood and oven temperatures during pyrolysis
trial.
The composition of the 2 hour charcoal sample
compares very favourable with the recarburiser:
The ash content is lower, the volatile content is
about the same, while the fixed carbon content is
slightly higher. These results indicate that 2 hours
residence time of wood in a coke oven at about
1000 C should be adequate to achieve a low
volatile content of the charcoal while minimising
burning and hence the ash content.
Table 1: Proximate analysis of the original wood, charcoal
samples and SMM recarburiser.
Sample Moistur
e (%)
Ash
(db%)
Volatile
(db%)
FC
(db%)
Wood 15.7 0.2 78.8 21.0
Char 2 hr 67.6 1.5 0.7 97.8
Char 4 hr 69.7 3.1 0.2 96.7
Char 6 hr 69.3 6.9 0.4 92.7
SSM
recarb.
0.6 5.3 0.7 94.0
Charcoal Making for Plant Trials
Approximately 20 tonnes of wood (E. Rossie) was
supplied from a property near Colinton NSW.
The wood was delivered as debarked logs 1.8
metres in length and from about 60 mm to about
200 mm in diameter. Approximately 40 logs were
strapped together to form bundles using
conventional steel strapping. Each bundle
weighed about 1 tonne. Figure 3 shows a
photograph of the wood bundles.
During the charcoal making four bundles were fed
into each oven using a site fork lift. The charcoal
was removed from the oven using the coke
discharge ram which pushed the charcoal into the
waiting buckle of a larger front end loader (shown
in Figure 4). The charcoal was dumped into a
large steel bin where it was quenched with water
to minimise burning due to exposure of the hot
charcoal to air
Figure 3: Photograph of wood bundles used to produce
charcoal for recarburisation trials.
The aim was to allow the wood to reach 1000 C
before pushing and subsequent quenching of the
charcoal. The results of the previous drum test
work indicated that this should occur by about 2
hours.
The temperature of the fourth (last charged)
bundle of wood from each oven was monitored
during the charcoal making run. Grab samples of
the quenched charcoal from each oven were
collected and analysed by HRL Technologies.
Figure 4: Photograph of hot charcoal being removed from
the Corrimal coke ovens.
Figure 5 shows the temperatures recorded for each
oven during the charcoal making. The x-axis
shows the time from the final bundle addition for
each oven. The typical behaviour evident from
Figure 5 is a rapid increase in temperature to
about 400 C, followed by a plateau of about 3
hours and finally an increase in temperature to
approach the oven temperature of about 1000 C.
The temperature of one oven (33) remained
stubbornly low, but still displayed the temperature
plateau albeit at a lower temperature than the
other ovens. The reason for this lower
temperature is not known but may be due to the
position of the thermocouple in the wood bundle.
The temperature of wood/charcoal increased
much faster in the preliminary trial than in the
main charcoal making exercise. This is likely to
be due to a more open wood pile in the trial which
increased gas flows and hence convective heat
transfer during pyrolysis. The collapse of the
wood bundles into a bed of smouldering charcoal
would have resulted in relatively slow heat
transfer and hence a slower rate of pyrolysis.
0
200
400
600
800
1000
1200
0:00 1:00 2:00 3:00 4:00 5:00 6:00
Time from final charging (hr:min)
T
e
m
p
e
r
a
t
u
r
e

C
oven 39
oven 37
oven 35
oven 33
oven 31
Figure 5: Temperature of the wood in each oven during
charcoal making.
Proximate analysis results are shown in Table 2.
These results show that the target volatile content
of the charcoal (less than the 1 %) was reached.
The ash content was generally less than 1 % but
the charcoal from oven 37 contained 1.6 % ash.
The average ash content was 0.85 %. Table 1
shows that the ash content of the initial wood was
about 0.2 %. This indicated that the recovery of
material to charcoal is about 23 % and suggests
that the charcoal yield from the original 20 tonnes
of wood may be around 4.6 tonnes. However
some of the ash on the charcoal is likely to have
been washed away during water quenching.
Hence, the charcoal yield was probably less.
Table 2: Proximate analysis of charcoal samples collected
from each coke oven.
Sample Moisture
(%)
Volatiles
(db%)
Ash
(db%)
FC
(db%)
Oven 31 58.3 0.83 0.83 98.3
Oven 33 63.5 0.80 0.73 98.5
Oven 35 50.4 0.29 0.63 99.1
Oven 37 59.1 <0.01 1.59 98.4
Oven 39 51.6 <0.01 0.48 99.5
This project work was successful in producing a
significant quantity of charcoal of the required
volatile content for recarburising plant trials. The
next stage in processing the charcoal is drying
followed by crushing and sizing to a suitable
product for SSM plant operations. Issues with the
preparation of a final charcoal product of less than
1 % water need to be finalised before the
recarburising trials can be completed.
Acknowledgements
Centres Program. Financial support for this work
from CSIRO Minerals Down Under Flagship,
BlueScope Steel and One Steel are also
acknowledged. The cooperation of the
management and staff of ICC prior and during the
trials at Corrimal is gratefully recognised.
References
1. D.E. Langberg, M.A. Somerville and T. Norgate, The
Use of Charcoal in the Iron and Steel Industry, CSRP
Report DMR-3278, October 2007.
2. D. Langberg, M. Somerville and B. Washington,
Recarburisation of Molten Steel Using Charcoal.
CSRP Report DMR-3277, September 2007
Piloting the Integrated Dry Granulation and Heat Recovery Process at CSIRO
D. Xie, J. Donnelley, R. Flann, Y. Pan, S. Sanetsis and B. Washington
CSIRO Minerals, VIC
Introduction
The Australian steel industry produces more than
2 million tonnes of molten slags at around 1500C
each year, which contain a large amount of
sensible heat. Currently, these slags are either air
cooled (Fig.1) or water granulated with some
adverse environmental impact such as air
pollution and excessive water usage. The large
amount of waste heat is not recovered.
Figure 1: Slags were poured and air cooled in a slag pit at
OneSteel Whyalla Steelworks.
Dry slag granulation1 is emerging as an attractive
alternative to water granulation to provide an
environmental friendly solution. Some major
design difficulties however have been experienced
in previous studies. Recent work at CSIRO since
2002 has resulted in major progresses in the
process design to overcome some of these
difficulties.
To build on the success of CSIROs work, a four
year CSRP project has commenced in August
2006 to extend the dry granulation work to
incorporating heat recovery.2 The project is also
jointed sponsored by Australian steel industry
(BlueScope Steel and OneSteel). The key
objectives were:
To dry granulate molten blast furnace slags
and produce glassy slag products for cement
manufacture. Thus, reducing GHG emission
from cement industry.
To recover high grade waste heat from molten
slags, and utilise the heat in steel plants to
reduce energy cost and hence GHG emission.
To reduce water consumption and emission of
sulphur to atmosphere.
Dry slag granulation could involve pouring
molten slag onto a spinning disc or cup to produce
fine slag droplets using centrifugal forces as
shown in Figure 2. In comparison with alternative
approaches such as air blast and rotary drums, the
spinning disc dry granulation is more efficient in
producing fine slag droplets for fast quenching
and efficient heat recovery.
Figure 2: Dry granulation of molten slags at a spinning disc.
The dynamic process involves fast droplet
breakup, rapid cooling and solidification. There
are several major technical challenges: including
(1) optimal disc design and process control to
provide smooth and well-controlled granulation;
(2) fast quenching and solidification of droplets to
produce glassy granulates for cement
manufacture; (3) compact reactor design for
recovering high grade heat and reducing the
capital and operation cost. A conceptual process
based on a two-step dry granulation/heat recovery
operation has been developed. This is shown
schematically in Figure 3.
Hot air >6 00C
Granules <5 0C
Hot air >6 00C
Air 25C
Air 25C
Dry granulation
(Spinning disc)
Heat exchanger
(Packed bed)
Slag
1500C
Slag
1500C
Drying
Preheating
Steam
Power
Desalination
Cement
Solid ~ 900C
Figure 3: Conceptual flowsheet for the new integrated dry
granulation and heat recovery process.
In this process, the molten slag is first atomised in
a dry granulator and slag droplets generated are
fast quenched and solidified to produce glassy
granules. The hot granules are then removed to a
packed-bed heat exchanger to extract the
remaining heat contained in the slag. The waste
heat is recovered in the form of hot air, which
could be used on-site for drying, preheating and
other purposes.
A new pilot plant was constructed (Figure 4) at
CSIRO to prove the conceptual design. A series of
tests using industrial blast furnace slags were
carried out to establish key operating parameters
and further optimise the process design.
Figure 4: New pilot plant at CSIRO.
Some technical and operational challenges were
experienced, including efficient air delivery and
smooth handling of hot granulate collected. These
challenges have been resolved through design
modification and process optimisation. The pilot
plant could now be run smoothly to produce fine
granulates as shown in Figure 5.
Figure 5: Slag products from dry slag granulation.
The slag products collected were subjected to size
analysis. The results indicated that more than 90%
of the products were less than 1.5 mm in diameter
as shown in Figure 6. These slag samples are
being further assessed for their suitability for
cement applications.
0
10
20
30
40
50
60
70
80
90
100
0 1000 2000 3000 4000 5000
Aperture (_m)
C
u
m
u
l
a
t
i
v
e

%

p
a
s
s
i
n
g
T15_R7
T15_R8
T16_R3
T16_R4
T17_R1
T18_R1
Figure 6: Accumulative size distribution of dry granulated
slags.
A new integrated dry slag granulation pilot plant
has been designed, constructed and commissioned
at CSIRO. The pilot plant is being used to prove
the concept of integrated dry slag granulation and
heat recovery. Test results with industrial blast
furnace slags have confirmed the feasibility of the
integrated process.
The new integrated slag treatment process will
allow for full value recovery (slag products and
heat recovery) from molten slags with a potential
reduction of 1.8 million tonnes GHG emission
each year for Australia.
Pilot scale tests have shown that the integrated
dry granulation and heat recovery is feasible.
Slag products are being assessed for
suitability for cement application.
CFD modelling and plant measurements are
being carried out for anticipated scale up and
plant trials in 2009.
Acknowledgements
Centres Program. The work is also jointly
sponsored by CSIRO, OneSteel and BlueScope
Steel.
References
1. Featherstone, W B: "Slag Treatment Improvement by
Dry Granulation," Iron and Steel Engineer, 1998, July,
42.
2. D Xie, Y Pan, R Flann, B Washington, S Sanetsis, J
Donnelley, T. Norgate, S Jahanshahi, Heat Recovery
from Slag through Dry Granulation, The 1st CSRP
Annual Conference, 21-22 November 2007, Melbourne,
Australia.
Minerals, Metals and Sustainability A book for the future
W.J. Rankin
CSIRO Minerals, VIC
A project has recently been established within the
Centre for Sustainable Resource Processing
(CSRP) with the objective of writing and
publishing a book which describes, within a
sustainability framework, the nature of mineral
resources, the role that minerals and their products
play in society, and the science and technology of
the minerals industry.
While there are many books on ores and
mineralogy, mining, and processing of metals
from elementary, descriptive books through to
advanced technical books these are usually
written from a relatively narrow discipline
perspective. The proposed book will have a
multi-disciplinary approach which integrates the
physical and earth sciences with the social
sciences, ecology and economics.
The intended readership of the book will be wide
and the book will be written in a manner to be
easily understood by readers with diverse
academic levels, backgrounds and interests.
However, it will be aimed particularly at:
students of engineering and applied science
(mining, metallurgical, civil, chemical,
mechanical, environmental) and geology, as
an integrated overview of the minerals
industry and/or as an introduction to specific
aspects of the industry;
practising engineers, geologists and scientists
who want a modern, and broader perspective
of the minerals industry;
students of economics, social sciences and
related disciplines who need an overview of
the minerals industry;
professionals in government service in areas
such as resources, environment and
sustainability; and
non-technical professionals working in the
minerals industry or in sectors servicing the
minerals industry.
The planned book will bring together recent
knowledge and understandings of the role of
minerals in modern society and the implications
of sustainable development on the use of minerals
and mineral products and of the minerals industry.
It will draw on a wide body of knowledge
including the work of CSRP.
It is hoped the book, when published, will have an
impact on the approach and content of relevant
undergraduate and graduate courses and that it
will result in an increase in the knowledge and
understanding by graduates and professionals in
industry and government service.
Methodology
The working title for the book is: Minerals,
Metals and Sustainability. The text will be
written around thirteen chapters arranged in a
logical sequence. Chapters 1 to 4 will introduce
readers to the concept of the earth as the source of
all materials, how materials are utilised in society
(with particular focus on inorganic materials), and
the issue of sustainability in relation to finite
resources. Chapters 5 to 8 will examine two
important classes of materials, metals and cement,
in detail. Chapters 9 to 13 will address the
sustainability challenges.
The proposed chapter titles are:
Chapter 1 Introduction
Chapter 2 Mineral resources
Chapter 3 Minerals and their role in society
Chapter 4 The Sustainability challenge and
opportunity
Chapter 5 Why are metals special?
Chapter 6 How metals are produced
Chapter 7 Cement Glue of industrial society
Chapter 8 Alternatives to metals and cement
Chapter 9 Minerals and the environment
Chapter 10 Natural limits What are they?
Chapter 11 Towards zero waste
Chapter 12 What does the future hold?
Chapter 13 A sustainability strategy for minerals
and metals
In general, no prior specialist knowledge will be
assumed other than an elementary knowledge of
chemistry and physics.
To make the book accessible, Chapters 1 to 4, 5 to
8 and 9 to 13 will be able to be read as
independent, free-standing blocks if desired and,
where necessary, cross-references will be
provided. For the same reason, more specialist
material, which could be omitted without loss of
continuity, will be placed in the later sections of
chapters. In this manner, the entire book, or
sections of it, will be capable of being studied at
various depths depending on the readers needs
and background.
The author, John Rankin, will draw on expertise
from a wide range of researchers and industry
personnel as needed through group and one-on-
one discussions, and will seek peer review of
chapter drafts as they are prepared by key
researchers and an Editorial Panel. Much of the
knowledge and understanding required for the
project is available within CSRP and its
Participants.
Progress
The project was formally established as a CSRP
project in mid-August 2008. In the 12-month
period leading up to this, the scope of the book
was developed and the proposal was discussed
with stakeholders. There was generally
widespread support for the project and this led to
commitment to support a two-year project within
CSRP. The project is fully resourced, with
funding being provided by CSRP, CSIRO
(through the Minerals Down Under Flagship) and
seven industry sponsors (BHP Billiton, Rio Tinto,
Alcoa, Xstrata, Hatch, BlueScope Steel,
OneSteel). CSIRO Publishing and the AusIMM
have indicated a strong interest in publishing the
book. Formal writing of the text commenced in
mid-August.
Demand-Supply Interaction on Future Mining Resource Production: The coal model
S. Mohr and G. Evans
University of Newcastle, Chemical Engineering, NSW
Introduction
Future supply of mineral and energy resources is
commonly predicted by applying Hubberts bell
curve analysis to historical production [1,2,3]. The
approach has a number of limitations. Firstly, it is
not theoretically immediate why production
should follow a bell curve distribution, and indeed
there are a number of cases where production is
asymmetric. Secondly, there is no mechanism for
accounting for changes in production rates by
external influences such as wars, economic
recessions, international crises etc. In this study an
alternative model is described, which includes
demand-supply interaction and production
schedule inputs for production output. The
Oceania coal industry will be used to demonstrate
this model.
Modelling Approach
The model is based on a market approach,
whereby the supply of a resource, such as oil or
coal, is influenced by: demand for it, production
capacity, and the amount of reserves available to
supply that market. The market is defined
depending on the nature of the resource. For
example, oil is shipped globally so it would be
appropriate to consider the global market.
Conversely, the global transport of natural gas is
relatively minor, so it would be more appropriate
to consider a regional or continental market.
The modelling approach for mineral production
from mining operations is described below:
1. Identify what the market is, regional, global,
etc, and define the intrinsic demand for the
resource. Here, intrinsic means the demand
that the supply is aiming for. Market
expectation (or demand) is a function of
supply availability; e.g. if supply cannot
satisfy demand then the market must adjust to
reduce demand.
2. Define the Ultimately Recoverable Resources
(URR) available to the market, and these are
obtained from literature values.
3. The URR are used to calculate the number of
mines with a given production capacity.
Demand-Supply interactions regulate the
scheduling of individual mine outputs,
including start-up and shut-down durations,
and steady-state production rates based on
historical data.
4. An iterative procedure is applied, whereby
intrinsic demand and supply try to equalise.
When sufficient capacity is available the
production is equal to the intrinsic demand,
brought about by a combination of production
from existing operating mines and by bringing
new mines on-stream. Increased production
from existing mines can also be implemented
in an effort to achieve the intrinsic demand. A
point is reached, however, where production
is not sufficient to meet existing intrinsic
demand, so the intrinsic demand must be
reduced.
Figure 1: Schematic of Mineral Model.
The model is illustrated in Figure 1. The approach
is implemented iteratively, so at a given point in
time, t, we know the following: the intrinsic
demand of the market M
D
[t], the total supply to
the market M
S
[t], the number of mines online M[t]
and the production, P
j
[t], and scheduling for each
mine.
From time, t, information, intrinsic demand and
supply for time t+1, are obtained as follows:
The intrinsic demand M
D
[t+1] is given by:
] [
] [ ] 1 [
t k
D D
D
e t M t M = + ,
(1)
where k
D
[t] is the rate variable, which is a function
of the equilibrium value, k
D0
, and the difference
between demand and supply at time, t, such that:
j
j
J
|
|
\
[
=
] [
] [ ] [
] [
1 0
t M
t M t M
k k t k
S
D S
D D
,
(2)
where k
1
is a proportionality constant.
The supply, M
S
[t+1], in year, t+1, is the sum of
the production from each mine, P
j
[t+1], i.e.:
[ 1]
1
[ 1] [ 1]
M t
j
S
j
M t P t
+
=
+ = +
.
(3)
In eq. (3), M[t+1] is the number of mines online at
time t+1, which is given by:
( )
[ ]
[ ]
[ 1] [ ]
S
S
M t
k t
URR
T T
M t M M M t e

+ =

,
(4)
where M
T
is the total number of mines, based on
the URR and the production scheduling applied to
all mines, and k
S
[t] is the supply rate variable and
is given by:
j
j
J
|
|
\
[
+ =
] [
] [ ] [
] [
2 0
t M
t M t M
k k t k
S
D S
S S
,
(5)
where k
S0
is the equilibrium supply rate variable
and k
2
is a constant.
Production for a given mine is assumed to take 4
years to linearly ramp up to the maximum
production, M
P
, and similarly 4 years to ramp
down to zero production at the end of its useful
life. There is scope for an individual mine to
upgrade production in response to demand to a
level of 2M
P
. The number of mines that upgrade
in the t
th
year, M
U
[t+1], is equal to:
|
|
|
|
j
j
|
|
= + ] [
] [
] [ ] [
] 1 [
3
t M k
t M
t M t M
k t M
U
S
D S
U
,
(6)
where k
3
is a proportionality constant, and k
U
is
the minimum gap needed between intrinsic
demand and supply before mines start upgrading.
Wars, financial slowdowns, etc, can reduce
intrinsic demand, which can be manually inputted
into the model, causing mines to be taken offline.
As described above, the model incorporates an
iterative approach, whereby production from
individual sources is optimised so that the demand
is met. When this can no longer be achieved by
the resources available, then the demand is revised
downward. The result is a demand-supply
interaction reflecting the reality of finite resources
and the market responding accordingly.
Example: Oceania Coal Demand-Supply
The model output is illustrated for the demand-
supply of coal in the Oceania market. Sources of
coal are Australian bituminous, sub-bituminous,
lignite, and other sources (e.g. New Zealand). The
corresponding URR values are given in Table 1.
The other inputs to the model are the constants
described in eqs. (1)-(6). These constants affect
the intrinsic demand and supply, and were
approximated by fitting the model to the European
coal market which has already peaked.
Table 1: Oceania Coal URR and peak year.
Type URR (Gt) Peak year
Aust. Bituminous 51 2058
Aust. Sub-bituminous 3 2033
Aust. Lignite 40 2088
Other 1 2056
Total 95 2066
The predicted supply and demand for coal types in
Oceania are shown in Figure 2. Also shown
(dotted line) is demand based on the current
growth rate.
Figure 2: Demand-Supply Model for Oceania.
It can be seen that future supply curves for each of
the different coal types is not a symmetric bell
curve, as assumed by Hubbert. This is due to the
intrinsic demand-supply interaction being based
on the historical data for the European market,
which itself is not symmetric. While a different
demand-supply interaction would change the
shapes of the curves, the peak years do not vary
greatly (less than 10 year) to those given in Table
2. Current projections show supply cannot meet
intrinsic demand after 2030, which is well below
the current business-as-usual scenario. The model
predicts a peak in total supply of coal occurring at
around 2066, and beyond that is a rapid decline in
production as the URR is quickly exhausted.
Acknowledgements
The Authors acknowledge the CSRP for their
financial support.
References
1. Zittel W., Schindler J., Energy Watch Group,
http://www.energywatchgroup.org/fileadmin/global/pdf/
EWG_Report_Coal_10-07-2007ms.pdf.
2. Laherrere J., Groningen Annual Energy Conv. 2006,
http://www.oilcrisis.com/Laherrere/groningen.pdf
3. Hubbert M. K., World Wildlife Funds Conference, The
Fragile Earth: Towards Strategies for Survival, San
Francisco, 1976.
Applying DEM Outputs to the Unified Comminution Model
M.S. Powell
1
, N. Weerasekara
1
, I. Govender
2
and M. Khanal
1
1
University of Queensland, JKMRC, QLD
2
University of Cape Town, Centre for Minerals Processing, South Africa
Introduction
The concept of a unified comminution model
(UCM) was formulated with the vision of
bringing all comminution models onto a common
base. The structure of the model relies on
knowledge of the mechanical environment in the
comminution devices - the Discrete Element
Method (DEM) has been used as the tool to
simulate this.
The model structure tackles the fundamental
causes of rock breakage, considering them to be
independent of the comminution equipment. The
equipment is considered to be the mechanism that
applies a given comminution environment to the
ore particles. Thus, once an understanding of the
processes of ore breakage is gained, the
mechanical environment from any type of
equipment can be overlaid to provide a model that
predicts the production of broken material.
Model Definition
The Unified Comminution Model (UCM) traces
the mechanical collision environment experienced
by particles in a comminution device and
calculates the resultant damage to the particles in
order to predict the progeny of the device.
The structure and key inputs to the model are
illustrated in Figure . The central block outlines
the model principle, and the peripheral blocks
outline the inputs that are required for the model.
A full description of the model is available in
Powell et al 2008.
DEM Outputs
The energy spectrum provided by DEM
simulations is divided into the tangential and
normal specific collision energies (Figure 2) per
rock size. This figure is rich with information,
showing the difference in tangential versus normal
energies; the increase in specific energy as the
rock size decreases; The reduction in the
maximum specific energy experienced by larger
rocks; and the dramatic increase in total number
of impacts for the smaller rocks (there are many
more of them compared to large rocks). These
individual energy spectra per size are used as the
input to the UCM breakage model.
Figure 1: UCM model structure.
The following observations can be drawn from the
plot of the energy spectra:
A tiny fraction of impacts have sufficient
energy to break rocks in one impact (Ecrit
0.28 kWh/t)
a few percent of collisions can break rocks
through multiple impacts (Eo 3.7 E-3 kWh/t).
the vast majority of the impact energy is
insufficient to cause bulk body breakage
the smaller particles receive considerably
more impacts
the smaller particles are subjected to
considerably higher specific energies of
impact.
This simulation output emphasises the importance
of the low energy surface impact abrasion
damage, and of the ball mill mode of breakage.
Therefore rounding and abrasion are likely to be
the dominant comminution mechanisms for
particles larger than 63 mm, and incremental
breakage important for particles in the size range
of 5.6 to 63 mm.
Applying these rates to the contents of a mill
allows the rate of breakage of each particle size to
be calculated. Adding the breakage size
distributions, that are a function of the impact
energy, to this information yields the overall size
distribution of the progeny. This is being applied
to the modelling of a pilot SAG mill in the first
instance.
Model principle
Breakage process
+
Mechanical environment
Breakage product
+
transport function
+
Time stepping
Product
DEM input
Mechanical environment
Contact detection
energy, angle, force
Mode of contact
Probability distribution
mode & partner
Distribution of mode
PB
Framework
Particles
+
Breakage
then
Convert to
mass
PB by mass
Tomography, MLA
Breakage testing
Mode of contact
Tests to reproduce these
Input - force, energy

Output size dist, liberation
Transport function
Mechanistic flow relationship
Later
SPH/DEM
Tangential energy distributions
1.E+00
1.E+01
1.E+02
1.E+03
1.E+04
1.E+05
1.E+06
1.E+07
1.E-10 1.E-09 1.E-08 1.E-07 1.E-06 1.E-05 1.E-04 1.E-03 1.E-02 1.E-01 1.E+00 1.E+01
Specific energy, kwh/t
c
u
m
u
l
a
t
i
v
e

n
o

o
f

i
m
p
a
c
t
s
180mm
125mm
90mm
63mm
45mm
31mm
22mm
16mm
11mm
8mm
5.6
Normal energy distributions
1.E+00
1.E+01
1.E+02
1.E+03
1.E+04
1.E+05
1.E+06
1.E+07
1.E-08 1.E-07 1.E-06 1.E-05 1.E-04 1.E-03 1.E-02 1.E-01 1.E+00 1.E+01
Specific energy, kwh/t
c
u
m
u
l
a
t
i
v
e

n
o

o
f

i
m
p
a
c
t
s
180mm
125mm
90mm
63mm
45mm
31mm
22mm
16mm
11mm
8mm
5.6
Figure 2: Specific energy distributions per particle size
class.
Figure 3: Fitted function to energy distributions.
The curves given in Figure 2 have been fitted to
generic functions, that allow interpolation
between sizes and a simple functional form to be
used in the UCM energy distribution model,
Figure 3.
We have also shifted from the traditional use of
the spring and dashpot energy to using the input
kinetic energy to analyse the breakage. This
energy corresponds to the energy that is measured
in the breakage tests, and is very different to the
dashpot energy, as illustrated in Figure 4.
Figure 4: Comparison of Dashpot and kinetic energies at
collision.
Conclusions
The principle of the application of the UCM has
been demonstrated, and the DME modelling
capability is considerably advanced. However, the
ore testing criteria have delayed proper model
validation, which is currently underway. This
clearly indicates the need for advanced pilot and
ore breakage characterisation work, which will
allow accurate validation and calibration of
advanced mill models, and models of other
comminution and classification equipment.
There are a number of areas to be addressed in the
capability of DEM:
Apportioning of energy between particles
Energy available for breakage during a
collision
Left-over energy- input power not accounted
for in particle damage
Sub-DEM size material
Solids and slurry transport
Solids discharge function.
References
1. Powell, M.S., Govender, I., and McBride, A.T., (2008).
Applying DEM outputs to the unified comminution
model the SAG mill. Minerals Engineering, Special
edition DEM07. DOI: 10.1016/j.mineng.2008.06.010
E
Ecrit
Energy Efficient Comminution Circuits A modified grinding strategy and the
selection of a target product size
Z. Pokrajcic
Introduction
This paper describes two techniques that can be
used in the design of comminution circuits for
improved efficiency and a reduction in total
energy consumption. They are:
A modified grinding strategy where efficient
grinding equipment is used and techniques
employed to decrease grinding media
consumption
A more reasoned and thorough approach to
the selection of target product size for a
comminution circuit involving a better
understanding and interpretation mineral
liberation data.
Background
There has been much discussion about the energy
intensive nature of comminution processes. It has
been quoted several times that comminution
processes account for 3%-4% of the global
electrical energy consumption. However, the
comminution process remains inherently
inefficient. It is understood that 85% of the energy
used comminution is dissipated as heat, 12% is
attributed to mechanical losses and only 1% of the
total energy input is used in size reduction of feed
material (Alvarado et al. 1998).
Another aspect of comminution which contributes
significantly to the energy intensity of the process
is the consumption of media and mill liners. The
quantity of energy necessary for the fabrication of
the media and/or liners from extraction, to
transport, to manufacture and assembly is
considered part of the total energy utilisation of
the comminution process. This is also quite an
energy intensive process and opportunities to
minimise steel consumption in comminution
circuits should be realised.
The third factor which contributes significantly to
the energy intensity of comminution circuit is the
chosen target product size, or grind size. As the
target product size decreases the energy required
to achieve the product size increases significantly.
As the particles become smaller their strength and
therefore their resistance to breakage increases.
For small particles as internal flaws are depleted,
the observed strength approaches the high
intrinsic strength of the solid.
This is the prospect facing all comminution design
and operating engineers. Hence, there is an
increasing need for more efficient grinding
systems and innovative comminution technologies
that not only increase the efficiency of the
comminution process but also act to minimise the
consumption of steel media and liners. This
together with a new approach to choosing a target
product size can significantly increase the energy
efficiency of a comminution circuit.
The two parts of this investigation were
undertaken according to the following:
A modified grinding strategy
All of the modifications made to the configuration
of an existing comminution circuit were
undertaken using the modelling and simulation
package JKSimMet. It is a valuable tool in
predicting the performance of a comminution
circuit, including the equipment power draw,
product size distributions, classification
efficiency, etc.
The aim of the eco-efficient modifications was to:
Reduce the direct energy usage. That is,
decrease the grinding power used in the size
reduction process.
Reduce the indirect energy usage. That is,
decrease the consumption of comminution
consumables such as steel media and mill
liners.
The strategy used to achieve these aims involved
better utilisation of the comminution energy. This
can be achieved by purposely targeting the
comminution energy at specific problematic size
fractions such as the critical size material in the
primary mill and the very fine fraction in the
secondary mill. Through the use of more efficient
comminution devices such as HPGR (high
pressure grinding rolls) a product size distribution
that is narrower in size range can be generated.
This size distribution contains less material in the
coarse fraction to aid overall mineral liberation
and less material in the fine fraction to minimise
the effects of slimes.
The strategy also exploited opportunities to
minimise the load in grinding mills and in some
cases employ autogenous grinding techniques to
decrease indirect energy consumption.
Selection of a target product size
This concept was illustrated using a sample from
of rod mill discharge from Mount Isa. Rod mill
discharge can be equated to SAG mill discharge in
terms of product size and product size distribution
in SABC circuit (SAG mill with recycle crushing
and a single stage ball mill).
The sample underwent staged grinding, using a
laboratory scale rod mill and ball mill to produce
samples of product sizes ranging from
P
80
=0.015m to P
80
=1.500mm.
Samples from each of the product sizes were
analysed using MLA techniques to assess the
extent of liberation of individual mineral
components.
In order to maximise the potential target grind size
and therefore minimise required energy input for
the circuit, the following concepts were applied:
minerals were grouped to form larger entities
and therefore deemed liberated at coarser
sizes
particles were deemed recoverable in a
flotation process, as opposed to liberated,
when the particle contains greater than 15%
mineral of interest
Together the above two points act not only to
increase the target product size but it also enable
the separation and rejection of material (typically
liberated gangue) from the circuit. The rejection
of gangue from the feed stream can limit the
amount of over-grinding but more importantly it
will reduce the mass of material proceeding to the
next stage of size reduction and/or separation.
This allows for more efficient separation and
significantly lower energy requirements for
downstream size reduction.
A modified grinding strategy
Eco-efficient modifications were made to an
existing SABC according to the following:
A pre-crush stage using a HPGR was included
to minimise the generation of critical size
material in the primary mill. The aim was to
reduce a portion of the feed material below
critical size and generate more fines to
decrease the load in the primary mill. Not all
of the feed material was pre-crushed to allow
for sufficient abrasion breakage in the primary
mill.
The SAG mill was converted to an AG mill
and the ball mill was converted to a pebble
mill to decrease liner and media consumption
Existing mill sizes were reduced
When the performance of this new circuit,
denoted ABC-HPGR-PM, was compared to the
original SABC circuit, the following (based on
JKSimMet and mass balancing predictions) was
observed.
The ABC-HPGR-PM circuit produced a both
a finer and steeper product size distribution
which promoted better flotation performance,
approximately a 2% increase gold recovery
Direct energy comparisons show that the
ABC-HPGR-PM circuit had an operating
work index of 15.9kWh/t compared to
24.3kWh/t for the base case SABC circuit.
This is an energy saving greater than 8kWh/t
for the whole circuit. Significant to say the
least.
Comparison of indirect energy consumption
for the two circuits is showed that again the
ABC-HPGR-PM out performed the standard
SABC circuit with a 79% saving in indirect
energy based on media and liner wear alone.
The SABC circuit consumed 3535kW while
the ABC-HPGR-PM circuit consumed
753kW. These figures are based on 6kWh/kg
required to manufacture the grinding
consumables.
Selection of a target product size
Analysis of the MLA data for the various product
sizes to include liberated and recoverable
particles, showed that:
70-80% of galena and sphalerite is liberated at
P
80
=76m
70-80% of galena and sphalerite is
recoverable at P
80
=150m
When galena and sphalerite are combined and
analysed as one mineral, 85% of this mineral
is recoverable at P
80
=150m
>95% of the sulphides (combined galena,
sphalerite, pyrite and pyrrhotite) are
recoverable at P
80
=1607m
45% of non-sulphide gangue is liberated at
P
80
=1607m
These points can be used in the design a
comminution circuit where product size is finer
than 1.5mm.
The aim is to exploit the recoverability of valuable
minerals and the rejection of liberated gangue in a
separation step at the coarsest product size.
The separation step could be a fotation stage or a
relative density separation stage. Ore properties will
determine the most suitable method.
The chosen circuit may resemble that shown in
Figure 1.
Figure 1: Design of separation circuit post primary
grinding.
The circuit Ieatures a coarse particle fotation step to
separate the liberated non sulphide gangue to tail Irom
the recoverable sulphides. The sulphides then proceed
to a Iurther stage oI grinding beIore being separated
in to valuable sulphides (galena and sphalerite) and
iron sulphide gangue which reports directly to tail.
The valuable sulphides are then separated in the last
grinding stage beIore proceeding to respective lead
and zinc fotation circuits Ior concentrate recovery.
The eIIect on energy consumption resulting Irom the
rejection oI liberated non-sulphide gangue prior to
the frst stage oI size reduction is shown in Table 1.
Table 1: Energy saved by rejecting the liberated non-
sulphide gangue at the head of the separation circuit.
Energy savings can be as high as 66 when removing
almost 50 oI the circuit Ieed material Irom Iurther
size reduction, the rejected material is primarily coarse
(P801.607mm) liberated non-sulphide gangue.
Opportunities exist to improve comminution
eIfciency and decrease comminution circuit energy
consumption by modiIying the standard SABC
circuit to include:
a HPGR pre-crush stage
autogenous grinding
The strategy is to target comminution energy at
problematic fractions such as critical size material
by decreasing re-circulating loads and employing
autogenous grinding techniques.
It is also possible through a better understanding oI
mineral liberation characteristics of coarse stream
(P80 ~1.5mm) in a comminution circuit to exploit
opportunities to reject liberated gangue Irom the
stream and therefore the comminution circuit thereby
reducing the size and grinding energy requirement
Ior downstream processes. These elements allow
provision Ior a coarser grind product size based on
the liberation and recoverability characteristics of
minerals.
The combination oI the modifed grinding strategy
together with a more reasoned approach to product
size selection enabling the rejection liberated gangue
Irom the circuit can decrease signifcantly total
energy consumption oI the circuit while improving
the operating eIfciency.
Acknowledgements
Dr. Toni Kojovic Ior helping to build and assess
the integrity oI the JKSimMet models.
Ridgeway site personnel Ior site survey and
analysis oI samples.
Mike Larson Ior collecting the rod mill discharge
sample at Mt. Isa.
Mitch Alexander Ior the staged grinding and
sizing oI sample.
Gerda Goy and the JK Center MLA crew Ior MLA
investigations.
ProI. Bill Johnson Ior imparting his knowledge
on liberation data analysis and coarse particle
fotation.
Dr. Rob Morrison Ior his continued guidance and
direction in the unravelling oI this project and its
purpose.
CSRP Ior the Iunding oI this research.
References
1. Alvarado, S, Alguerno, J, Auracher, H, and Casali, A,
1998. 'Energy-exergy optimization oI comminution.
Energy (OxIord), 23(2), pg 153-158.
2. Atasoy, Y., W. Valery, et al. (2001). Primary verses
secondary crushing at St. Ives (WMC) SAG mill circuit.
SAG 2001, Vancouver.
3. Daniel, M, 2007. Energy efficient liberation using
HPGR technology. PhD thesis. University of
Queensland, Brisbane. (unpublished)
4. Grano, S. (2006). Ian Wark Research Institute. Annual
Report 2006. U. o. S. Australia.
5. Huls, B. J. and G. S. Hill (2006). "The coarse particle
recovery process." CIM Bulletin 99(1093): 68-74.
6. Johnson, N. W. and P. D. Munro (2002). Overview of
flotation technology and plant practice for complex
sulphide ores. Mineral processing plant design, practice
and control, Vancouver, SME.
7. Laslett, G. M., D. N. Sutherland, P.Gottlieb, and N.R.
Allen, (1990). "Graphical assessment of a random
breakage model for mineral liberation." Powder
Technology 60(2): 83-97.
8. Manlapig, E. V., D. J. Drinkwater, P.D.Munro, N.W.
Johnson, and R.M.S Watsford, (1985). Optimisation of
grinding circuits at the lead/zinc concentrator, Mount Isa
Mines Ltd. Symposium on Automation for Mineral
Resource Development, Brisbane, Australia.
9. Munro, P. D. (1993). Lead-zinc-silver ore concentration
practice at the lead-zinc concentrator of Mount Isa
Mines Limited, Mount Isa, QLD. Australasian Mining
and Metallurgy - The Sir Maurice Mawby Memorial
Volume. J. T. Woodcook and J. K. Hamilton.
Melbourne, AusIMM. 1: 498-503.
10. Napier-Munn, T. J., Morrell, S., Morrison, R.D. and
Kojovic, T., (1996). Mineral Comminution Circuits:
Their Operation and Optimisation, pp. 49-92, (Julius
Mineral Kruttschnitt Mineral Research Centre:
Brisbane)
11. Pokrajcic, Z, and Morrison, R, 2008, A simulation
methodology for the design of eco-efficient
comminution circuits. IMPC 2008, Beijing. IN PRINT.
Improving Grinding Efficiency with the IsaMill
M. Larson
1
, R. Morrison
1
, F. Shi
1
and M.F. Young
2
1
2
Xstrata Technology, QLD
As mining companies have been called on to
process ever finer grained deposits they have had
to turn to new technology to economically treat
these ores. One example of this is the IsaMill
TM
.
The IsaMill
TM
was originally developed to grind
fine grained Mount Isa and McArthur River
lead/zinc ores to sizes finer than 7 micron.
The IsaMill
TM
is a horizontal stirred mill that takes
advantage of high rotational speeds and fine inert
media to efficiently grind to liberation sizes that
were previously uneconomical, at best, and
sometimes were simply not achievable.
A laboratory investigation was conducted to
determine the effects of the IsaMill
TM
operating
variables on power draw and power efficiency,
using a 4 litre M4 IsaMill
TM
at the JKMRC. The
results of the test work were used to construct a
preliminary IsaMill model.
The laboratory standard signature plot is used to
characterize IsaMill performance and grinding
energy requirements (Weller and Gao, 1999). This
has been shown to reliably translate laboratory
scale results to mine site operations. Laboratory
testing was undertaken with the M4 IsaMill to
determine which operating variables affect the
grinding energy efficiency and to what extent they
impact the grinding energy efficiency. A
campaign of test work was conducted using
copper concentrate from the Mount Isa Mine that
had a F80 =64 microns. Variables tested included
the mill speed, feed pump volumetric flow rate,
grinding media type, grinding media size, media
filling and feed pulp density. In addition viscosity
was investigated as it relates to the feed pulp
density and grinding energy efficiency
Mill Speed vs Power
y = 0.0017x - 1.866
R
2
= 0.986
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
1400 1600 1800 2000
Mill RPM
N
e
t

P
o
w
e
r

D
r
a
w

(
k
W
)
15 micron
power draw
Figure 1: Mill speed vs. net mill power draw (at constant
media load).
The feed pump volumetric flow rate did not have
an effect on grinding energy efficiency.
Regardless of the pump speed the same signature
plot will still be created. The mill speed did not
have an effect on grinding energy efficiency and it
was found to be a reliable indicator of net mill
power draw between 1400 and 2000 RPM, as
shown in Figure 1. Less residence time was
required to grind to a given target size as the mill
speed was increased.
Similarly, a linear relationship was a good fit for
the net power draw when the media load was
varied (Figure 2). The two relationships should
provide a means in the future to better predict the
outcomes of varying operating conditions while
performing signature plot tests
Media Loading vs Net Power Draw
y = 0.0161x + 0.0156
R
2
= 0.9679
0.4
0.6
0.8
1
1.2
1.4
1.6
20 40 60 80 100
Media Load (%Full)
N
e
t

P
o
w
e
r

D
r
a
w

(
k
W
)
15 micron
power draw
Figure 2: Media loading vs. net mill power draw (at
constant mill speed).
The overall grinding efficiency tends to increase
as the media volume increases in the mill, with the
most dramatic effect occurring from the low end
of filling to the beginning of the generally
accepted operating range of 60-80% full, when
grinding to a target of P80=10 microns, as shown
in Figure 4. The media volume is the bulk volume
of the media at rest divided by the net volume of
the mill (volume of shaft and grinding discs
subtracted from total mill volume). The
inefficiency was most clearly seen when the mill
had very low media filling. When the mill had low
media filling the initial discharge density was
lower than the feed density indicating there was a
build up of solids inside the mill. The solids that
build up in the volume that would normally be
filled with grinding media could cause increased
power draw and be reducing the power efficiency.
M4 IsaMill Model
Two simple equations were developed relating the
overall efficiency of varying media loads and feed
pulp densities, based on the previous grinding
energy requirements graphs. The 100% efficiency
was set at a media filling of 80% and solids feed
pulp density of 45%, then the equations were
developed to relate the energy efficiency loss to
the changing variables. These equations were
applied to a set of six model validation tests.
Table 1: Model Validation Results using Mount Isa Cu
Concentrate.
Test Media
Volume
(L)
Density
(%weight)
kWh/t
model
kWh/t
actual
%
error
1 2.8 39 68.5 69.5 1.44
2 2.3 54.4 86 84.8 1.42
3 1.9 35 90.5 83.7 8.12
4 2.6 52 76.6 75.6 1.32
5 2.5 45 75.5 77 1.95
6 2.7 57.1 80.1 82.2 2.55
Fines Production Model
It was decided to investigate the form of the
relationship between new surface area generated
and energy input, based on work by McIvor
(2007) and later Musa and Morrison (2008).
McIvor and Musa both demonstrate that new
production of particles finer than a certain target
size is approximately linear with energy for rod
and ball mills. A similar simple relation was
attempted to allow grinding energy predictions
with changing the feed size.
The selected micron size for the modelling was
chosen as the 50% passing size of the final pass
from the Isamill test, and then the percent passing
the selected micron size for each pass in the
Isamill test was plotted against cumulative energy
for the Isamill test to that point.
This did not fit a linear relationship or any
variation of a power plot. It was determined
however that by squaring the percent passing
value and plotting it versus cumulative energy on
a normal plot a straight line connecting each
IsaMill pass with the feed at the 0 kWh/t could be
formed. This would have the advantage over the
standard signature plot that the feed material
could be included in the line created.
The main disadvantage of the signature plot is that
each percent passing line cannot be extrapolated
back to the feed size. This imperfection has to be
taken into account when estimating the effect of
changing feed sizes. With the new method, using
the percent passing a particular size, the line
should be able to be shifted left or right depending
on the change in feed size and using the same
slope to estimate the new power requirement.
This should be valid for any case where the same
media is capable of grinding the new feed size.
The work done thus far shows that through an
improved understanding of basic mill operating
variables, showing grinding energy efficiency of
the Isamill can be predicted and improved. The
ability to predict the mill performance with
coarser feed sizes using the fines squared function
will allow the design of grinding circuits
incorporating the IsaMill. This along with the
optimization of media size will assist to fully
optimize the entire circuit grinding efficiency
Acknowledgements
The authors wish to thank Xstrata Technology for
funding this project. This work was carried out
under the auspice of the Centre for Sustainable
Resource Processing, which is established and
supported under the Australian Governments
Cooperative Research Centres Program.
References
1. Weller K. and Gao M., 1999, Ultra-fine Grinding,
AJM Crushing and Grinding Conference, Kalgoorlie,
27-28th April 1999
2. McIvor R. and Finch J., The Finch-McIvor Functional
Performance Based Grinding Circuit Modelling System
3. Musa F. and Morrison R., 2008. A More Sustainable
Approach to Assessing Comminution Efficiency,
Comminution 08 Falmouth
Multiple-pass High Pressure Grinding Rolls Circuits
M. Hilden and M. Powell
High Pressure Grinding Rolls (HPGR) are widely
considered to be more energy efficient devices for
comminuting rock than tumbling mills such as
SAG mills. Recently, HPGR have been installed
and commissioned at Cerro Verde (Peru),
Mogolokwane (South Africa), and Freeport
(Indonesia) with reported energy savings of
around 19% over more conventional grinding
routes [1].
To date, HPGR have been installed as single unit,
either in open circuit or closed circuit with a
screen. A single HPGR unit is typically limited to
around 3 kWh/t, compared with SAG mills which
may have energy inputs exceeding 10 kWh/t, and
therefore despite being less efficient can generate
a finer product.
This project aims to determine whether a series of
HPGR units could be used to perform most of the
grinding duty in a process plant, increasing the
amount of breakage performed in these highly
efficient devices. Such a circuit could
substantially reduce the energy and grinding
media used in milling circuits and therefore
reduce the carbon footprint of new processing
plants.
Three ore samples have been tested to date. The
samples were a hard copper-gold ore, a softer
lead-zinc sulphide ore, and the third was a
porphyry copper ore. The samples were crushed to
-32 mm top-size and processed in a three-pass
HPGR circuit (Figure 1).
HPGR
HPGR
HPGR
Figure 1: Three HPGR in series.
Test work was carried out on two large pilot-scale
Kppern HPGRs at AMMTEC, Perth. The first
two samples were tested using 0.22 m (W) x 0.75
m (D) rolls in July 2007. The third sample was
tested using a newer 0.25 m (W) x 1.0 m (D)
HPGR (Figure 2) in March 2008.
Figure 2: Kpperns Andrew Gardula and JKMRCs
Malcolm Powell discussing the pilot-scale HPGR at
AMMTEC Perth.
The third sample was also tested in a wide range
of circuit configurations that included screening
between passes. Test work was also carried out
using one or two passes with the Kppern HPRG
with subsequent passes carried out at CSIRO
Pinjarra Hills using a laboratory-scale HPGR 0.10
m (w) x 0.25 m (D). In each case, up to three
passes of the HPGR were used.
The energy consumption and product size was
measured for each stage in the flow sheets to
allow energy efficiency of each circuit to be
determined.
The feed and product sizes from the samples
tested to data show that the three-HPGR circuit
produces a broad product size distribution, but
also that the magnitude of size reduction reduces
with subsequent passes. Two passes were found to
be effective; however the third pass continues to
generate fines without substantially reducing the
top size. The result is a flatter or broader product
size distribution as shown in Figure 3.
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
0.01 0.1 1 10 100
Size, mm
%

P
a
s
s
i
n
g
Feed
Pass 1
Pass 2
Pass 3
Figure 3: Product sizing for the three-stage circuit.
A number of three-stage circuit HPGR
configurations were compared with the above
three-pass HPGR circuit. Two approaches were
found to be effective in generating a steeper
product size distribution:
1. Removing excess fines from the feed to the
HPGR (Figure 4a), and
2. Using a smaller diameter rolls in the second
and/or third pass (Figure 4b).
A B
HPGR
HPGR
HPGR
2.0 mm Screen
48%
52%
1.0 m HPGR
1.0 m HPGR
0.25 m HPGR
HPGR
8 mm Screen
HPGR
HPGR
1.0 m HPGR
22%
78%
t
0.25 m HPGR
0.25 m HPGR
Figure 4: Two of the alternative flow sheets tested.
As shown in Table 1, these circuits produce a
similar-sized product using less energy compared
with three passes in the 1.0 m rolls. In the case of
flow sheet 4b, just two HPGR passes are needed
to grind to an equivalent product fineness
(measured using the 80% passing size, P80, and
the quantity of -150 microns fines generated), and
uses 33% less energy.
Table 1: Comparison of various three-pass HPGR circuits.
Pass 1 Pass 2 Pass 3
P80, um 9391 3788 2414
% -150um 24% 38% 47%
Energy, kWh/t 2.77 5.23 7.53
P80, um 8804 2824 1425
% -150um 22% 35% 45%
Energy, kWh/t 2.83 4.13 5.48
P80, um 4710 1764 1195
% -150um 27% 46% 53%
Energy, kWh/t 3.23 5.03 5.83
Triple-pass circuit Fig 1.
Modified Circuit Fig 4a.
Modified Circuit Fig 4b.
Test work confirms that the HPGR can be
used in series to efficiently generate a very
fine product. After three passes, around half
of the product can be made into final flotation
feed size.
Two flow sheet modifications further improve
the performance of the three-stage HPGR
circuit and use less energy, albeit using a
potentially more complex flow sheet.
This work has shown that a multi-pass HPGR
circuit can generate a high proportion of fines, but
only two passes appear to be beneficial without
either a classification stage or a reduction in roll
diameter. One or two additional samples will be
tested in coming months.
Acknowledgements
Centres Program. The large-scale test work was
carried out with the assistance of Kppern
Machinery Australia and the small-scale HPGR
test work was carried out by Steve Suthers of
CSIRO, Pinjarra Hills.
References
1. C. M. Rule, I. Smit, A. J. Cope, and G. A. Humphries,
"Commissioning of the Polycom 2.2/1.6 5.6MW HPGR
at Anglo Platinum's new Mogalakwena North
Concentrator," presented at Comminution 08, Falmouth,
2008.
Update on the JKRBT (JKMRC Rotary Breakage Tester)
T. Kojovic and F. Shi
The long history of comminution studies at the
JKMRC has included the development of
laboratory testing procedures, to provide material
characterisation data that are used in conjunction
with machine specific data in modelling and
simulation (Napier-Munn et al, 1996). The key
developments include:
Twin Pendulum
Drop Weight tester (DWT)
Automated Linear Impact Comminution
Evaluator (ALICE)
Short Impact Load Cell (SILC)
The DWT is an established and globally accepted
impact characterisation device, and has become
one of the standard impact tests for AG/SAG
milling ore characterisation. The DWT apparatus
and its associated data reduction technique were
developed so that the relationship between
specific energy input and resultant product size
could be determined (see Equation 1):
( )
Ecs b
e A t
.
10
1

= (1)
where t10 is a size distribution fineness index
defined as the progeny percent passing one tenth
of the initial mean particle size, Ecs is the specific
comminution energy (kWh/t), and A and b are the
ore impact breakage parameters determined from
DWT results. The index A*b has become well
known in the mining industry as a reliable
indicator of impact ore hardness, and essentially
describes the rate at which fines are produced
(t10) for a set amount of specific energy (Ecs).
The t10-Ecs relationship and A*b parameters are
used in size-reduction-modelling for crushers and
mills in the JKSimMet mineral processing
simulator (Wiseman and Richardson, 1991).
Since all tests using the DWT, SILC (Bourgeois
and Banini, 2002), or the Hopkinson bar are
conducted on single rock specimens positioned
manually on the anvil or flat surface, they are both
time consuming and expensive. For the test to
remain practical the rock samples being tested are
limited to 10 30 pieces for each size fraction,
which inevitably throws into question the
statistical validity of the derived ore
characteristics. Recent developments in DEM of
milling have revealed that small energy impacts
occur much more frequently than high energy
impacts (Djordjevic et al, 2004). As such it
appears a new requirement in breakage testing
will be the characterization of incremental
breakage at small impact energies. However,
testing of repetitive impacts at small energies
using the DW tester is very time-consuming and
hence impractical. Research into finding a rapid
breakage characterization device was clearly
warranted in an effort to overcome these
limitations.
The concept of using kinetic energy to crush rocks
seemed like a viable alternative for rapid breakage
characterization, since it no longer requires the
positioning of rock specimens manually on the
anvil. Industrial applications of this concept are
found in Vertical Shaft Impact (VSI) crushers and
laboratory pulverizers, which employ a rotor-
stator impacting system. However, both of these
devices are employed merely for particle size
reduction, not for breakage characterization, as the
exact amount of energy applied in the process is
not well controlled or measured. However, the
JKMRC found that VSI crushers can be modelled
from first principles, allowing useful simulations
(e.g. effect of rotor speed) to be carried out with
only the basic design data on the machine and
rock characteristics required (Kojovic, 1996;
Djordjevic et al, 2003). This suggests that the rock
characteristics may be inferred from the product
of the VSI, if the amount of energy applied in the
process can be precisely controlled and measured.
In response to the body of evidence supporting the
use of kinetic energy, in 2005 the JKMRC
comminution research team led by Shi and
Kojovic decided to thoroughly investigate the
feasibility of using the kinetic energy concept for
particle impact breakage characterization. A
prototype Rotary Breakage Tester (JKRBT), with
a rotor diameter of 360 mm, was designed and
manufactured by the JKMRC pilot plant
workshop team (Figure 1). The operating system
consists of a vibrating feeder, a rotor-stator
impacting device with its drive system, and an
operation control unit. Like the DWT, the JKRBT
also requires the ore particles be pre-sized into
narrow fractions. Particles of the selected size are
fed into the rotor-stator impacting system via a
vibrating feeder.
Figure 1: Photograph of the prototype JKRBT device, with
rotor-stator showing through the inspection window.
The JKRBT uses a rotor-stator impacting system,
in which particles gain kinetic energy while they
are spun in the rotor. They are then ejected and
impacted against the stator, causing particle
breakage. The specific energy of each impact in
the JKRBT, Ecs, is defined as the kinetic energy
Ek per particle mass m:
2
2
5 . 0
5 . 0
i
i k
V
m
V m
m
E
Ecs =

= = (2)
Since the particle mass does not affect the specific
energy in this type of impact breakage device, the
Ecs becomes solely dependent on the impact
velocity Vi. The research team has shown that the
specific energy of impact can be accurately and
precisely controlled in the JKRBT. They also
confirmed that the actual impact velocity in
practice is less than theoretical, requiring the
JKRBT to be calibrated first to achieve precise
breakage energy levels. The calibration is
machine specific and can be programmed into the
JKRBT control system to allow the user to enter
the required specific energy directly, or it can be
tabulated in the JKRBT Operating Manual as pre-
set speeds which can be entered manually to
achieve a desired specific energy level.
The JKRBT overcomes some of the limitations of
existing impact tests such as a lack of precision of
the energy input, the amount of time required to
run individual tests and the difficulties in testing
the smallest particles. To date, comparative
breakage tests using the prototype JKRBT device
and the traditional JKMRC Drop Weight Tester
showed that the two devices generate the identical
breakageenergy relationship for the same ore of
the same size, as shown in Figure 2.
0
5
10
15
20
25
30
35
40
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5
Ecs (kWh/t)
t
1
0

(
%
)
DWT
RBT
RBT repeat
Model Fit
Figure 2: Comparison of t10 vs. Ecs relation determined by
the JKRBT and DWT.
Statistical analysis indicates that the two testing
methods can generate statistically similar
breakage parameter A*b values, as illustrated in
Figure 3 which compares the A*b values of 16 ore
types determined by industrial JKRBT and DWT
respectively. This analysis accounts for the
difference in contact points and strain rates
between the DWT and JKRBT.
0
20
40
60
80
100
120
140
160
180
200
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
Data set
A
x
b

v
a
l
u
e
s
Figure 3: Comparison of breakage parameters A*b
determined by the JKRBT and DWT (10% error bars on
DWT result).
The very positive feedback received from mining
companies to the JKRBT has prompted the
JKMRC to further validate the device through a
formalized experimental test program on
professionally designed and manufactured JKRBT
machines. The validation project started in March
2007 with the first industrialised unit installed and
commissioned at Anglo Research in
Johannesburg, South Africa. Barrick, BHP
Billiton, Rio Tinto and Teck Cominco have also
signed up to participate in the project. The current
status and site deployment is as follows:
Anglo Research (Crown Mines, South Africa)
x 2 machines
Barrick Gold (JKtech, Brisbane) - 12 months
agreement
BHP-Billiton (Newcastle Technology Centre,
Australia)
Rio Tinto (Kennecott, USA)
Teck Cominco (Trail, Canada)
Figure 4 shows a diagram of the industrialised
JKBRT, designed and manufactured by Russell
Mineral Equipment (RME).
Figure 4: The first industrialised JKRBT installed at Anglo
Research in South Africa.
Unlike the prototype device, the industrialised
JKRBT can be opened for easy access to clean the
breakage chamber. The lid is operated by an
electronically driven linear actuator. The counter-
weight at the back of the lid is designed to prevent
the lid from falling down by gravity should the
electronic system fail while the lid is opening or
closing. The industrialised JKRBT employs a
rotor 450 mm in diameter, and can treat particles
from 1 to 45 mm, at specific energy levels from
0.001 to 3.8 kWh/t. The use a rotary feeder has
provided a simple feeding system that also offers
an effective noise suppression mechanism,
reducing the noise to below 85dB at the highest
operating speed.
It is worth noting that the new breakage model
developed by the JKMRC (Shi and Kojovic,
2007) enables the user to accurately match the
DWT derived breakage parameters, A and b, from
JKRBT data on four (13.2 to 45 mm) instead of
the standard five DWT size fractions (13.2 to 63
mm). The model and application of JKRBT
concept to ore testing has been patented by The
University of Queensland.
Feedback from Anglo Research has indicated
higher productivity and better repeatability by the
first commercial JKRBT when compared with the
Drop Weight Tester. The results to date look very
promising and suggest the JKRBT will provide a
powerful and easy to use tool for rapid ore
breakage characterisation to the mining industry.
Tests have confirmed the device offers a rapid
method for determining the hardness of drill core
samples within the context of the AMIRA P843
geometallurgical project. For comparative testing,
Walters and Kojovic (2006) found the JKRBT
appears to be the best choice, followed by SMC
and PLT.
Some of the key features of the JKRBT are:
Treats many more particles hence can
generate statistically more valid results
Wide particle size range (1.4 45 mm), wide
energy range (0.001 3.9 kWh/t)
Rapid characterisation (1/8 - 1/10 of DWT
time confirmed by Anglo Research)
Accurately and precisely control energy
Excellent reproducibility
A new rapid breakage characterization testing
device, the JKRBT, has been developed by the
JKMRC. A detailed study using a prototype
machine confirmed the concept of using
controlled kinetic energy to characterize ore
particle breakage, prompting a comprehensive
validation using an industrialized JKRBT,
professionally designed and fabricated by Russell
Mineral Equipment (RME). The validation project
will be completed in 2008, with commercial
release expected in early 2009.
Application of the prototype JKRBT in
geometallurgical testing with the AMIRA P843
GeM project has confirmed the JKRBT device
offers a rapid and consistent method for
determining the hardness of drill core samples.
The JKRBT has the potential to revolutionize ore
testing, with applications in laboratory impact
breakage characterisation of rock samples, drill
cores, coal and other materials. Automation with
lab image sizing technology like the Camsizer
would further enhance its productivity. It also
stands poised to significantly advance the process
control of comminution circuits since it is readily
adaptable to on-line ore hardness measurement.
This concept will be explored in the current P9O
project.
Acknowledgements
The research and development of the JKRBT has
been funded by The University of Queensland,
JKMRC, JKTech, AMIRA/P9N Project, and
Centre for Sustainable Resource Processing
(CSRP). In-kind support from Anglo Research
staff during the commissioning of the first
industrial JKRBT unit is greatly appreciated.
References
1. Bourgeois, FS and Banini, GA, 2002. A Portable load
cell for in-situ ore impact breakage testing. Int. Journal
of Mineral Processing. 65, 3154.
2. Djordjevic, N, Shi, F and Morrison, RD, 2003. Applying
discrete element modelling to vertical and horizontal
shaft impact crushers. Minerals Engineering 16: 983-
991.
3. Djordjevic, N., Shi, F. and Morrison, R., 2004.
Determination of lifter design, speed and filling effects
in AG mills by 3D DEM. Minerals Engineering, 17,
1135-1142.
4. Kojovic, T., 1996. Vertical shaft impactors: predicting
performance, Quarry Aust Jrnl, Vol 4, No 6, pp35-39
5. Napier-Munn, T.J., Morrell, S., Morrison, R.D., and
Kojovic, T., 1996. Mineral comminution circuits: their
operation and optimisation. ISBN 0 646 28861. Julius
Kruttschnitt Mineral Research Centre.
6. Shi, F. and Kojovic, T. 2007. Validation of a model for
impact breakage incorporating particle size effect. Int.
Journal of Mineral Processing, 82, 156-163.
7. Van Latum, LA, 1985. The evaluation,
conceptualization and development of single particle
breakage testing apparatus. Masters Thesis
(unpublished), Technical University of Delft,
Netherland.
8. Walters, S., and Kojovic, T., 2006. Geometallurgical
Mapping and Mine Modelling (GEMIII) the way of
the future. Proc SAG 2006, Vancouver, Vol IV, pp411-
425.
9. Wiseman D.M. and Richardson J.M., 1991. JKSimMet
- the mineral processing simulator. Proceedings 2nd
Can Conf on Comp Applications in the Min Ind, (Eds
Paulin, Pakalnis and Mular), Univ B.C. and CIM, Vol II,
pp427-438.
Improvement of Energy Efficiency of Rock Comminution through Reduction of
Thermal Losses
N. Djordjevic
Introduction
Theoretical calculations show that only some 0.1
to 2 % of the energy supplied to rock during
comminution is effectively utilised for fracturing.
This project investigates the accumulation in
thermal energy of rock, using an advanced
thermal imaging camera, to quantify how much
energy is lost into increased thermal energy of
rock, and to explore potential relationships
between this heat accumulation, the physical
and/or mineralogical characteristics of the rock,
the produced fragment size distribution and
operational parameters of the comminution
process. Ultimately it is hoped that such better
understanding of heat losses, will generate options
for modification of operational and equipment
parameters of comminution equipment to reduce
heat losses and improve efficiency of rock
crushing.
Obtained results demonstrated that substantial
fraction of net energy delivered to rock ends up as
the heat energy radiating from the rock after
crushing. Obviously this indicates more energy is
delivered to the rock that what is required to crush
the rock. From the practical point of view,
question is what can be done with existing
equipment to reduce such thermal losses. We were
focused on the potential improvement of the
efficiency of crushing that can be archived
through modification of the feed size distribution.
Testing was performed during HPGR crushing of
Rio Tintos Bingham Canyon copper ore.
Thermal Radiation
Any object will emit energy due to its
temperature, as long as temperature of the object
is above absolute zero. From the measured
radiation energy and emissivity constant of the
surface, it is possible to calculate surface
temperature of the materials. Assuming that
surface temperature represents average
temperature of the materials, it is then possible to
calculate thermal energy stored within materials,
from equation.
Experimental Results
We performed a range of rock crushing
experiments which was monitored using sensitive
infrared (IR) camera. Measurements of thermal
losses that occurred during High Pressure
Grinding Roll (HPGR) crushing of local basalt
were performed. HPGR crushing was performed
on the six samples of equal mass (18.86kg). Initial
pressure of the HPGR were set to 20, 30 (2
samples), 40, 50 and 60bar. Infrared imaging was
performed during crushing, Figure 1. Crushing
products were sized and compared against feed
size distribution and introduced net comminution
energy.
Figure 1: Setting used during HPGR testing.
Increase in the specific comminution energy
(applied pressure), results in finer fragmentation.
However, there is tendency that after certain
amount of energy is introduced, further increase in
consumed specific energy, produce only minor or
insignificant improvements in fragmentation. This
is observed in the diagrams of fragment size
distributions which show asymptotic behaviour of
the fragment size distribution as applied pressure
increase, or amount of net energy consumed
increases, Figure 2.
HPGR Crushing of Mt. Morrow Basalt
0.0
10.0
20.0
30.0
40.0
50.0
60.0
70.0
80.0
90.0
100.0
0.1 1 10
size (mm)
c
u
m
u
l
a
t
i
v
e

(
%
)

p
a
s
s
i
n
g
Feed
20bar
30bar
40bar
50bar
60bar
Figure 2: Feed size distribution and fragment size
distribution after HPGR crushing under different pressure.
Results of IR imaging of the crushed rock, shows
that large fraction of net energy supplied to rock is
transformed in to heat, rising temperature of the
rock, Figures 3.
Figure 3: Infrared image for set pressure of 30bar, copper
ore (temperature scale min. 30degC, max. 60degC).
From the temperature and measured net energy
supplied to rock, efficiency of energy utilization
was calculated as function of the HPGR pressure
setting, Figure 4.
Efficiency of Utilization of Net Energy Consumed
0
10
20
30
40
50
60
20 30 40 50 60
Pressure (bar)
%
Figure 4: Efficiency of energy utilization as function of
HPGR pressure.
Conclusions
Presented results and interpretation are restricted
on comminution aspects of energy transfer and
thermal losses. We were able to perform
reproducible measurements of temperature
increase that occur during highly transient events
such as dynamic rock breakage. Obtained results
show that with increase of energy introduced,
there is increase in the temperature along the
fractured surface as well as increases in the
overall amount of thermal energy.
Results obtained during HPGR testing clearly
indicate that there is optimum intensity of
pressure to which given rock needs to be exposed.
Any further increase in pressure results in only
marginal increase in fragmentation and significant
increase in unproductive heating of rock.
Feed characterised with relatively narrow size
distribution (i.e., without large amount of very
fine material) is prone to more efficient crushing
at significantly lower pressure than feed with
same top size, but including relatively large
amount of fines. Experimental results indicate
that up to 40% of energy can be saved through
optimization of the applied pressure and
modification of feed size distribution.
Acknowledgements
Centres Program.
Towards a Virtual Comminution Machine
R.D. Morrison
1
and P.W. Cleary
2
1
2
CSIRO, Mathematical and Information Sciences, VIC
Towards the end of the 1990s readily available
personal computers became sufficiently powerful
when combined with an efficient numerical
code to use Discrete Element Modelling (DEM)
in two dimensions for models involving a few
hundred to a few thousand particles in
commercially available packages. Some
proprietary codes reported up to 200,000 particles
(Herbst and Nordell, 2001 and Cleary, 2001)
In early 2000, JKMRC and CSIRO-MIS agreed to
an informal collaboration with the objective of
testing various DEM approaches against detailed
process measurements. The initial collaboration
demonstrated that 3D-DEM using spheres was
sufficiently realistic for flow patterns and power
estimation within tumbling mills. The results were
reported in papers which were presented at SAG
2001 and in the technical literature. (Morrison et
al, 2001; Cleary et al, 2003; Cleary et al, 2008)
The VCM will allow a comminution machine
design which exists as a suitably detailed design
in a 3D Computer aided design file (CAD) to
simulate processing an ore (which has been
characterised by suitable test work) to predict
progeny, power consumption, wear and even
machine component loadings.
Methodology
A new comminution device typically requires
15-25 years for development and to gain a
sufficient degree of industrial acceptance to have
any chance of commercial survival.
Unsurprisingly, many promising devices suffer
from interruptions to their development schedule
or fail altogether for lack of development finance
as much as any technical shortcomings.
There are also broader reasons for seeking more
efficient ways to carry out comminution. It has
been estimated that comminution processes world
wide consume about 3% of all electricity
generated (La Nauze et al, 2002). This compares
with 1.3% by ore milling in North America (DOE
report, 1981). Consumption of grinding media
and wear resistant liners consumes about the same
order of energy in terms of green house gas
production (Musa and Morrison, 2007). Hence
the total energy equivalent is more like 6%. In
Australian terms, 10 and 20% are reasonable
estimates. The current substantial pressure to
reduce greenhouse gas production therefore
provides a strong incentive to devise more energy
efficient comminution devices, and to reduce the
calendar time required for their development.
The most serious obstacle in the traditional
approach to developing a new comminution
device is the time and cost of building successive
prototypes. Each prototype represents a
substantial investment in development time and
capital cost. Mechanical shortcomings such as
component rigidity may cause rejection of a
design with potentially good process performance.
If at least some of the required prototypes can be
constructed and assessed by simulation, then a
large reduction in development time and cost
should be possible. In many ways worse still are
machines which work well at lab and pilot scale
but cannot be scaled up to commercially viable
sizes.
Each prototype cycle can easily cost 6-12 months
of development time and tens to hundreds of
thousands of development dollars. Hence, a
system which can allow a good deal of the
development work to reside in a simulation model
has much to recommend it.
Experimental Technique
To be credible as a VCM, both the simulation
program and the ore characterisation process must
be thoroughly verified. Hence the VCM should
also be useful for modelling and further
development of existing comminution devices.
A VCM has the further advantage that variations
on a theme can more easily be considered
allowing estimation of the total energy footprint
in terms of electrical energy and wear whilst
treating identical simulated feed materials.
The Virtual Comminution Machine (VCM) must
be able to relate each collision type and energy
within a DEM simulation to the probable degree
of breakage of that particle based on
characterisation tests of a particular ore. As
several modes of comminution occur within
tumbling mills, relating collision spectra to
comminution represents a particular challenge.
The greater part of comminution energy is
expended in a slurry environment in wet
autogenous and semi-autogenous (AG and SAG)
mills and ball mills. The proportion of fine (minus
37 micron) material in the slurry has a strong
effect on its viscous properties. A reasonably
viscous slurry which can retain particles on the
surface of the grinding media helps to ensure
selection of those particles for breakage. Hence
the fluid and particle interactions are also
important for particle breakage and transport.
Models based on computational fluid dynamics
(CFD) are not well suited to the high shear, high
viscosity environment with complex free surface
behaviour found in wet tumbling mills. DEM can
be adapted to fluid modelling using Smoothed
Particle Hydrodynamics (SPH), see Cleary (1998)
and Cleary et al. (2007). The shear dependant
viscosity model developed by Shi and Napier-
Munn (2002) has been implemented in the CMIS
code using SPH as a way to model mineral
slurries.
In a DEM simulation, the forces exerted by each
particle on the equipment surfaces are estimated.
Therefore the relative wear rates can also be
estimated using an appropriate wear model.
Similarly, the total energy requirements for the
machine can be estimated by summing the
energies consumed by particle movement.
In short, the VCM objective is to model each
process within any comminution machine to an
acceptable level of accuracy so that it can be used
as virtual development environment for existing
and new comminution machines.
The VCM will always require powerful
computational capabilities. A parallel
devolvement is the UCM (Universal
Comminution Model) which will embed VCM
type relationships into a system suitable for
desktop PCs (Powell et al, 2008)
Acknowledgements
Centres Program.
References
1. Cleary, P. W., (1998), Modelling confined multi-
material heat and mass flows using SPH, Applied
Mathematical Modelling, Vol. 22, pp. 981-993.
2. Cleary, P. W., (2001), Charge behaviour and power
consumption in ball mills: Sensitivity to mill operating
conditions, liner geometry and charge composition, Int.
J. Min. Processing, Vol. 63, pp. 79-114.
3. Cleary, P. W., Powell M.S and Morrison R.D.(2008),
Applying Computationally Intensive Techniques to
Modelling of Comminution Devices, IMPC Beijing
4. Cleary, P. W., Morrison, R., and Morrell, S., (2003).
Comparison of DEM and experiment for a scale model
SAG mill, Int. J. Min. Processing, 68, pp. 129-165.
5. Cleary, P. W., Prakash, M., Ha, J., Stokes, N., and Scott,
C., (2007), Smooth Particle Hydrodynamics; Status and
future potential, Progress in Computational Fluid
Dynamics, 7, pp. 70-90.
6. DOE, 1981. Committee on Comminution and Energy
Consumption, NMAB-364
7. Herbst, J. A., and Nordell, L., (2001), Optimization of
the design of sag mill internals using high fidelity
simulation, Proceedings of the SAG conference, Eds.
Barratt, Allan and Mular,, (University of British
Columbia), Vol. IV, pp. 150 -164.
8. La Nauze, R.D and Temos, J., (2002), Technologies for
sustainable operation. CMMI Congress 2002, Cairns
Queensland, May AusIMM, pp. 27-34.
9. Morrison, R.D., Cleary, P.W. and Valery, W., (2001).
Comparing power and performance trends from DEM
and JK modelling. SAG 2001, Univ of British
Columbia, Vancouver, pp. 284-300.
10. Musa F, and Morrison, R. D., (2007). Assessing
comminution efficiency.XXII ENTMME / VII MSHMT
Ouro Preto-MG, Brazil
11. Powell M.S., Govender I. and McBride A.T. (2008)
Challenges in Applying the Unified Comminution
Model, IMPC Beijing
12. Shi, F. N. and Napier-Munn, T. J., (2002). "Effects of
slurry rheology on industrial grinding performance",
IntJMP, 65(3-4), pp. 125-40.
The New Energy Logging from the Discrete Element Method
N. Weerasekara and M. Powell
A discrete element modelling (DEM) simulation
for a typical mill section with several thousands of
particles generates massive amounts of output
data, running to several gigabits. Even with good
sampling techniques
1
this generates files of close
to a gigabyte.
Although some DEM software provides a GUI
based data analysis environment, their techniques
either require a considerable amount of processing
power or are not capable of delivering the
required inputs for the unified comminution
model (UCM) kind of modelling
2
.
Therefore a new data logging system was
developed to fulfill the following objective:
Provide flexibility in handling huge data sets
generated by typical DEM code.
Extract particle collision information from
DEM output database.
Summarise data to provide more useful
information for further comminution
modelling process.
Methodology
A C/C++ based framework was developed in
extracting and analysing the massive data sets
generated though the DEM simulation. The
following structure shown in Figure 1 and 2 were
adapted in implementation.
Figure 1: Phase I data extraction.
As show in Figure 1, Phase I of this system
interacts directly with the DEM output and
generates data sets which can be easily managed
and handled by the Phase II. The method adapted
in the Phase I will vary slightly depending on the
DEM output data structure of the DEM code used.
But the basic framework is similar to the
flowchart in Figure 1.
The basic framework for the Phase II is shown in
Figure 2. In this phase the extracted and grouped
data is further summarised depending on the
requirements for further modelling work.
Figure 2: Phase II data summarisation.
A series of DEM simulations were carried out
using EDEM software (Figure 3) and CSIRO
DEM code, for a 1.6m diameter pilot SAG mill
with roughly 30,000 particles. This new data
logging framework was tested and used in
analysing the outputs generated though both these
codes. These results are being used as inputs to
the UCM modelling.
Figure 3: EDEM output graphical view.
Group collision energy data based on
particle size classes and mill liner
Particle Class data
Particle IDs, No of collisions
at each time step
Particle IDs
Collision energy at each
collision
Analysis and summaries collision energy
data for required particle size classes and
mill liner
Summarized output
(e.g. Cumulative collision
energy spectrum)
e.g. Modules
AbsCollision
AbsNormalEnergy
AbsTangEnergy
RelCollision
BinEnergyLamda
DEM Output
(Raw Data Base
collision info, energy etc.)
Extract particle and surface specific
information
Extract collision information
(e.g. normal/tangential energy)
Particle Class data
Particle IDs, No of collision
at each time step
Group data base on particle classes
and liner surface
Log energy data at each collision
Particle IDs
Collision energy at each
collision
Key Results
From this energy logging system, normal and
tangential collision energy distribution for
different particle classes were observed (Figure 4)
across the full energy spectrum. The effect of
particle size and the number of particles present in
the mill on the cumulative collision energy per
particle is clearly shown. This provides an
excellent input for estimating the modes and
degree of breakage.
Figure 4: Cumulative collision energy spectrum for 12
particle classes (labelled C1 to C12), using CSIRO DEM
simulation.
It was also interesting to observe the number of
collisions occurring within a given particle size
class (Figure 5).
Figure 5: Cumulative number of collision across the normal
energy spectrum for 12 particle classes (labelled C1 to C12),
using EDEM simulation.
Number of collisions per particle (
DEM
) is another
important parameter for UCM modelling
2
, the
developed framework generated that quite well,
for example Figure 6 shows
DEM
variation across
the normal energy spectrum. The data logging
system also generates
DEM
variation for all the
particle classes simulated in the DEM code.
Figure 6: Number of collisions per particle (
DEM
) across
the normal energy spectrum, from 5 different DEM test cases
using EDEM.
Highlights and Benefits
Robust, efficient and faster data extraction,
analysis and summarisation process.
Demands less computational power and
memory.
Has the flexibility and potential to extract and
summarise other relevant information needed
out of the DEM simulation.
The C/C++ based framework was tested and used
in logging and summarising energy from DEM
outputs from two different DEM codes. This was
tested and used in Windows and Linux platforms.
It also proved to be more efficient and fast.
The future direction is to extend this framework to
extract and analysis multiple sets of DEM outputs
in parallel.
Acknowledgements
This work was carried out under the auspice and
funding of the CSRP, and with additional funding
from the JKMRC
References
1. P. W. Cleary, Recent advances in DEM modelling of
tumbling mills, Minerals Engineering, Volume 14, Issue
10, , October 2001, Pages 1295-1319.
2. M.S. Powell, I. Govender, A.T. McBride, Applying
DEM outputs to the unified comminution model,
Minerals Engineering, Volume 21, Issue 11, Pages 744-
750.
3. EDEM user manual, Copyright 2002-2008 DEM
Solutions.
Australian Life Cycle Initiative (AusLCI) and CSRP Database: Australian Data
D. Giurco
1
, P. Schmidt
2
and B. McClellan
2
1
University of Technology Sydney, Institute for Sustainable Futures, NSW
2
University of Queensland, Sustainable Minerals Institute, Centre for Social Responsibility in
Mining, QLD
The Australian Life Cycle Initiative (AusLCI) is
being developed by CSIRO and the Australian Life
Cycle Assessment Society (ALCAS). Its aim is to:
"provide a national, publicly-accessible database
with easy access to authoritative, comprehensive
and transparent environmental information on a
wide range of Australian products and services over
their entire life cycle"[1].
In parallel, CSRP researchers are working to
develop a database of readily accessible information
for exchange between CSRP participants to assist
with decision making for sustainability.
The aim of this extended abstract is to:
1. provide an overview of activities being
undertaken by AusLCI and, in particular, the
"metals" working group within AusLCI
2. describe the progress of a CSRP database and
how this relates to AusLCI
3. outline benefits of the CSRP database and
AusLCI initiative and future research needs.
AusLCI Initiative Overview and Approach
Data on which Life Cycle Assessment studies are
based is largely drawn from outside Australia (e.g.
Europe) and can be inappropriate to our local
context. AusLCI is the hub for gathering accurate
and reliable Australian data for use in LCA studies.
The initiative has several technical committees
considering issues of allocation and data quality and
guidelines covering data format, quality, and critical
review aimed at ensuring transparent data is used.
Several sector-based groups have formed to set the
research agenda for data collection within sectors
and engage industry participation (e.g. agriculture,
plastics, water, construction, metals). See [2] for
further information.
The metals group has identified:
need to further engage industry (e.g. through
roundtable workshop)
need to address data confidentiality when
seeking involvement
limited Australian data is publicly available.
AusLCI Future Directions
The AusLCI initiative will continue gathering data
across sectors for incorporation into its database
over the coming years. The priorities for the metals
sector include consulting with industry to identify
data gaps and to find funding to commission new
research for accurate data collection.
CSRP Database Overview
Independently of the AusLCI process the CSRP SD
program involves the task of producing a database
to support CSRP researchers. This commenced
independently of the AusLCI project due to the
longer time-frames required to deliver AusLCI data.
The aim of this is to provide a database of
predominantly quantitative relevant information that
would assist CSRP researchers in evaluating the
overall benefits or impacts of project outcomes.
This database is also intended to feed into the
developing SUSOP
mechanism options which will

incorporate Sustainable Development principles
into the design and operation of industrial
processing plants.
Methodology: CSRP Database Development
Development of the CSRP database over the recent
months builds on initial work during the
development of SD CAT (the SD Contributions
Assessment Tool). The database has been initially
structured to fit with the perceived needs of SD
assessment in the CSRP, as per Figure (overleaf).
To date, the database covers data categories of
electricity, transport, grinding media and reagents.
The data has been collated from public domain
sources such as ABARE, the Department of Climate
Change and industry reports. Minerals processing
data is contributed through other projects and
studies (for example see [3]).
CSRP Database: Interim Findings
Table 1 gives a brief overview of the current
coverage of data found from public domain
literature to date. Whilst transport and electricity
data are increasingly available ahead of emissions
trading, proprietary products such as reagents and
grinding media will require consultation with
producers, or estimated ranges of values.
Table 1: Data overview.
Data
category
Example Data Data
source
Coverage
Transport
(per net
tonne
kilometre)
Articulated
trucks
0.0705 kg CO2
0.0784 kg CO2
(FFC)*
AGO,
NPI,
ARA
Australian
average figures
for key modes of
freight transport;
Electricity
(per MWh)
Qld Grid
903 kg CO2
and
1046 kg CO2 /
MWh (FFC)
AGO,
NPI,
ESAA,
ABARE
ABS
Australian, fuel-
specific and state-
by-state data on
production and
emissions;
Grinding
Media
~ ~ Data collection
continuing
Reagents ~ ~ Data collection
continuing
*Full fuel cycle
The rapid expansion of data quantity and quality
for transport and electricity has been supported by
increasing requirements of carbon reporting. Data
availability for grinding media and reagents has
been less supported by these processes as shown
by limited public domain material.
A ready reference of useful data for CSRP
projects and from CSRP projects for sharing with
other CSRP participants will facilitate assessing
the sustainability benefits of new projects
CSRP Database Benefits
A comprehensive database of current data for
Australia will allow quicker, reliable
assessment of sustainable development
benefits of CSRP research outcomes and other
projects.
Provide a ready reference for assessing the
carbon intensity of ongoing operations and
areas for reduced impacts.
The next phase for database development will
involve direct application of energy data to more
specific minerals processing data, especially
around grinding media and reagents. Further
verification of data from multiple sources will
clarify data ranges and potential error margins.
Acknowledgements
the Centre for Sustainable Resource Processing,
Australian Governments CRC Program.
References
1. AusLCI website. http://www.auslci.com.au
2. Aus LCI Wiki site
http://alcas.asn.au/auslci/pmwiki/pmwiki.php/AusLCI/AusLCI
3. Norgate, T. E. and S. Jahanshahi (2006). Project 1A2:
Energy Issues Paper - Energy Use in Metallurgical
Processes and Related Greenhouse Gas Emissions.
Perth, Co-operative Research Centre for Sustainable
Resource Processing.
Figure 1: Initial structure of database (dark orange blocks in the figure have been completed to date, lightly shaded blocks are
partially complete, white blocks need data).
Quantitative Phase Analysis of Fly Ashes for Use as Geopolymer Source Materials
R.P. Williams, N. Chen-Tan and A. van Riessen
Curtin University of Technology, Centre for Materials Research, WA
How would it be if we could use cheap industrial
waste products to make concrete with superior
properties that will last for thousands of years,
similar to the concretes used in Ancient Rome?
That is exactly what is achievable by making
geopolymers with clays, fly ash or almost any
aluminosilicate material. The possible
applications of geopolymers are only limited by
scientists imaginations. Geopolymers are
currently in use as a concrete (cement
replacement) and have also shown promise in
specialised applications such as fire proof barriers,
ceramic precursors and refractory adhesives to
name just a few. Geopolymers have only short-
range ordering, hence are considered to be x-ray
amorphous.
Fly ash is the fine particle residue from burning
coal in thermal power stations. Fly ashes have
complex variable composition due to the
composition of the coal from which they are
derived and the operating conditions of the power
station. Fly ash consists of a mineral fraction,
which is inert, and an amorphous fraction, which
takes part in the geopolymerisation reaction. This
means the concentration and composition of the
amorphous fraction needs to be determined, in
order to manufacture geopolymers from different
fly ashes with reproducible properties.
The bulk elemental composition of the fly ash can
be obtained by XRF and the mineralogical (phase)
composition can be obtained using a quantitative
XRD technique such as Rietveld refinement,
allowing the amorphous composition to be
determined by difference [1, 2]. Further
understanding of the fly ash internal structure and
of significant factors in the chemistry of the
geopolymer formation, such as the role of
crystalline and amorphous phases and particle
size, can be obtained by from x-ray absorption
spectroscopy (XAS), scanning electron
microscopy (SEM) and transmission electron
microscopy (TEM). Compositional analysis of the
amorphous component of the precursors provides
a better understanding of the geopolymerisation
reaction.
For powder diffraction, each fly ash was
micronized for 5 minutes with a known mass of
internal standard (fluorite - CaF
2
), the resulting
slurry was dried at 105C for 24 hrs, then loaded
into 0.5 mm borosilicate capillaries (GLAS,
Germany). The remaining powder was stored in a
sealed vial for laboratory powder diffraction.
Powder diffraction (PD) data was collected at the
Australian Synchrotron (Melbourne, Australia). A
wavelength of 1.000625 angstrom was utilised,
the diffraction data was collected on the
MYTHEN position sensitive detector system. This
detector is comprised of 16 MYTHEN modules
each covering ~4.8 2, such that the system
covers ~80 2 with 0.2 gaps every 4.8 2. Two
dataset were collected per sample, each for 300
seconds, the repeat collected after moving the
detector bank 0.5, the two datasets were
subsequently joined using a conversion tool
described elsewhere [3].
Figure 1: Photo of the Powder Diffractometer at the
Australian Synchrotron.
The crystalline phases present in each sample
were determined using search match routines in
both Bruker-AXS EVA 10.0 and MDI Jade 6.0
software packages to search the powder
diffraction file (PDF) database.
Rietveld Quantitative Phase analysis was
performed using the Bruker-AXS TOPAS 4.0
software package. The crystal structures used
were selected from the Inorganic Crystal Structure
Database (ICSD) (version 2008/1). The line
profile function was refined using relevant
standards utilising the Thompson-Cox-Hastings
model. The background was modelled using an
order 2 Chebychev polynomial and two broad
split pseudo-Voigt peaks. The following
parameters were refined for each crystalline
phase: scale, lattice parameters, Lorentzian
crystallite size (i.e. sample contribution is a
Lorentzian function convolved such that the
FWHM varies as a function of cos), atomic
displacement parameter for phases > 5 wt%. The
data resulting from this process is the absolute
crystalline phase quantification and the
amorphous quantification.
Preliminary results from Rietveld quantitative
phase analysis of laboratory based x-ray
diffraction data showed that the method is viable,
however many of the minor phases are near the
detection limit providing poor accuracy and
precision in the quantification. Powder diffraction
data has been collected at the Australian
Synchrotron, the improved signal to noise and 2-
theta resolution lowers the detection limit of these
minor phases, improving accuracy and precision
of the quantification, these results will be
presented.
Table 1: A comparison with other studies of the
Quantitative Phase analysis of NIST SRM 1633b Fly
ash.
T
h
i
s

s
t
u
d
y
W
i
n
b
u
r
n

e
t

a
l

2
0
0
0

[
1
]
W
a
r
d

a
n
d

F
r
e
n
c
h

2
0
0
6

[
2
]
Phase Name wt% (2 esd) wt% (2 sd) wt%
Mullite 20.9 0.2 21.8 1 22.2
Quartz 6.53 0.08 6.1 0.4 7.1
Maghemite 3.3 0.2 2
Magnetite 2.6 0.1 3.9 0.4 2.2
Hematite 1.5 0.1 2.2 0.4 1.1
Gypsum 1 0.2
Calcium
aluminate 1.1
Amorphous 65.1 0.3 65 2.2 64.3
To gauge the accuracy of the method employed in
this study a NIST standard reference material fly
ash was also characterised, the results are
compared to other studies in Table 1. In this study
the concentration of the iron oxides phases
(maghemite, magnetite and hematite) were
determined to have slightly higher total
concentrations than in the other studies (7.4, 6.1
and 5.3 wt% for this study, Winburn et al. [2] and
Ward and French [1] respectively), this is to be
expected, as the other studies employed Cu K
alpha
radiation for which there is significant
microabsorption effects with iron, resulting in a
negative bias on phase quantification of iron
containing phases [4]. The other phases are
comparable, however no gypsum nor calcium
aluminate were detected in the sample.
Rietveld quantitative phase analysis (RQPA) was
undertaken on several local fly ashes, the
amorphous concentration has been determined.
Future work involves using this data to calculate
the amorphous composition of each fly ash and
produce viable geopolymers with each.
Acknowledgements
Centres Program.
This research was undertaken on the Powder
Diffraction beamline at the Australian
Synchrotron, Victoria, Australia. The views
expressed herein are those of the authors and are
not necessarily those of the owner or operator of
the Australian Synchrotron
Thank you to Hans Fairhurst of Cockburn Cement
and Craig Heidrich of Ash Development
association for providing the fly ashes.
References
1. Ward, C.R. and D. French, Determination of glass
content and estimation of glass composition in fly ash
using quantitative X-ray diffractometry. Fuel, 2006.
85(16): p. 2268-2277.
2. Winburn, R.S., et al., Rietveld quantitative X-ray
diffraction analysis of NIST fly ash standard reference
materials. Powder Diffraction, 2000. 15(3): p. 163-172
3. Williams, RP and van Riessen A, Data processing of
diffraction data from the MYTHEN detector system on
the Powder Diffraction beamline at the Australian
Synchrotron. Centre for Materials Research (CMR)
Technical Note, 2008, Curtin University of Technology.
Note# 001
4. Madsen I.C., Scarlett N.V.Y., Cranswick L.M.D. and
Lwin T. Outcomes of the IUCR CPD Round Robin on
Quantitative Phase Analysis: Samples 1a to 1h, J.Appl.
Cryst, 2001 34, 409-426.
Synthesis Paths and Performance Evaluation of Geopolymer Binder Systems
Derived from Major Mineral Processing and Mining Waste Feedstock Materials
K. Sagoe-Crentsil and T. Brown
CSIRO, Materials Science and Engineering, VIC
Introduction
The base chemical and mineralogical composition
of several large volume industrial, mining and
mineral processing byproducts fall within the
compositional limits of feedstock materials
deemed suitable for Geopolymer binder synthesis.
Such materials include, fly ash from coal-fired
electric power plants, red mud residues from
alumina processing as well as blast furnace slag
from iron ore processing. Geopolymer research
has thus attracted considerable research interest
[1-2] in recent years, partly due to its potential to
utilize diverse industrial wastes in large scale
mining, geotechnical and civil construction
applications. In essence, the underpinning
chemical reactions governing aluminosilicate
solids and alkali silicate solutions [3-4] to form
Geopolymer binders can, in principle, be
amenable to a range of industrial waste forms.
This study therefore attempts to examine the
relationships between chemical formulation,
micro structure and mechanical properties of
selected fly-ash based Geopolymer systems to
explore compositional limits relevant to mining
waste streams eg red-mud mineralogy.
Specifically, this paper draws on experimental
studies to examine the relationships between
chemical formulation, microstructure and
mechanical properties of high silica Geopolymer
systems with SiO
2
/Al
2
O
3
> 15 as an alternative
binder to conventional geopolymers having
SiO
2
/Al
2
O
3
= 3-4.
Methodology
Several Geopolymer formulations with
conventional Si contents, termed low-Si
Geopolymer, were prepared with different
proportions of ingredients selected to allow the
effect of silica content to be assessed. Samples
with compositional ratios i.e. SiO
2
/Al
2
O
3
= 2.7 to
3.9 differing in the proportion of silica added in
solution compared to that obtained from the solid
fly ash component, as detailed in Table 1. The
high-silica samples required additional solid silica
to achieve required Si dosages. The formulations
were cured at 85C for 2 hours. Compressive
strengths were measured on 25.4 mm cubes and
SEM was performed on fracture faces.
The investigation focussed on the effects of molar
concentrations of SiO
2
through a combination of
electron microscopy and strength testing.
Table 1: Nominal chemical composition of Geopolymer
systems.
Nominal mix composition
Silicate series
SiO
2
/Al
2
O
3
Ratio
Na
2
O2.7SiO
2
Al
2
O
3
10H
2
O 2.7
Na
2
O3.0SiO
2
Al
2
O
3
10H
2
O 3.0
Na
2
O3.5SiO
2
Al
2
O
3
10H
2
O 3.5
Na
2
O3.9SiO
2
Al
2
O
3
10H
2
O 3.9
Key Results
Low-silica content Geopolymer mixtures:
The microstructures of Geopolymer samples
appeared similar to images previously published
for these materials [5, 6]. The condensed gel
phase in the samples consisted of rounded
growths with bridging between them (Fig. 1).
Variations exist for microstructures of
geopolymer systems with SiO
2
/Al
2
O
3
= 2.7 and
3.9. The fineness of the texture and the density of
the geopolymer increases greatly from SiO
2
/Al
2
O
3
= 2.7 to 3.9. The images reveal higher silica
samples contain more unreacted particles, and
also that crystals have grown on a number of the
unreacted fly ash particles, particularly in the
higher silica mixture (Si/Al 35) as shown in the
microstructure of Figure 1.
The compressive strengths mirrored observed
trends in the microstructure in the expected way,
in that the densest, finest grained samples were
also the strongest. In particular, the change in
strength with silica content was quite dramatic,
given that a molar ratio increase of 11% Si from
SiO
2
/Al
2
O
3
= 3.5 to 3.9 yielded a corresponding
62% strength increase.
Figure 1: SEM image of a low-silica Geopolymer
SiO
2
/Al
2
O
3
=3.9. Scale bar represents 1 m.
High-silica Geopolymer mixtures:
The high silica geopolymer systems exhibited
extremely dense microstructures; much denser
than conventional geopolymers. The dense
microstructures suggest that high silica
geopolymers could possess good mechanical
properties and bonding characteristics. Figure 2
shows the strength development characteristics of
Si-rich geopolymer systems with varying
SiO
2
/Al
2
O
3
content ranging from SiO
2
/Al
2
O
3
=
15-56. The compressive strength was a maximum
at around SiO
2
/Al
2
O
3
= 32. It must also be noted
that the pH dropped with increasing silica content
to about SiO
2
/Al
2
O
3
= 38 and remained constant
beyond that. The high-silica content formulations
were also characterised by excess water demand.
0
10
20
30
40
10 30 50 70
SiO
2
/Al
2
O
3
C
o
m
p
r
e
s
s
i
v
e

S
t
r
e
n
g
t
h

(
M
P
a
)
Figure 2: Effect of varying silica content on compressive
strength of high- Si Geopolymer systems.
Discussion
It was observed that the investigated fly ash-based
formulations had characteristic complex
microstructures and structural variety. In
particular, samples with the highest strength and
the densest microstructure tended also to contain
crystalline components.
Compressive strengths in general can be related to
both composition and microstructure in a logical
way. These general observations and their
implications on feedstock material selection may
be best understood by examining the fundamental
dissolution and condensation reactions occurring
during synthesis. Given that Si solubility, is
generally much less than aluminium under
alkaline conditions, the condensation reaction in
geopolymers will likely occur between
monomeric [Al(OH)
4
]
and a variety of silicate

species, including monomers and oligomers. The
silicate solutions used in the production of
geopolymers, typically have high SiO
2
concentrations above 5M, and M
2
O/SiO
2
ratios of
0.66~0.83 by the addition of NaOH which is
further likely to generate other polymeric silicate
species, as demonstrated by Barbosa et al[7]. In
relation to the potential use of industrial waste
streams e.g. red mud residues from the Bayer
process, the key reaction determinant will likely
occur during the condensation reaction between
monomeric [Al(OH)
4
]
and a variety of silicate

species, including monomers and oligomers.
Conclusions
Observations described in this paper broadly
follow expected theoretical correlations between
geopolymer mix composition, microstructure and
strength. It is observed that the high silica
geopolymer formulations investigated achieve
near optimal high strengths at SiO
2
/Al
2
O
3
35
and show characteristic low porosity and dense,
fine grained microstructures. The observed
synthesis parameters further suggest potential
beneficial and novel applications of red-mud as
raw feedstock material in a variety of building
product and civil construction applications.
References
1. Davidovits, J, Davidovits, M, Davidovits, N, US Patent
5342595, 1994.
2. Fernndez-Jimnez, A.; Palomo, A.; Criado, M.,
Cem. Concr. Res.,35, Issue: 6, June, 2005, pp. 1204-
1209.
3. Van Jaarsveld, J.G.S, van Deventer, J.S.J, and
Lorenzen, L, Met. Mater. Trans. B 29B (1998) 283
291.
4. Steveson, M, and SagoeCrentsil, K, J. Mater. Sci.,40
(2005) 4247-4259.
5. Phair, J.W, and Van Deventer, J.S.J, Miner. Eng. 14
(2001) 289304.
6. Van Jaarsveld, J.G.S, Van Deventer, J.S.J, and
Schwartzman, A, Miner. Eng. 12(1) (1999) 7591
7. Barbosa, V, MacKenzie, K, Thaumaturgo, C, Int. J.
Inorg. Mater. 2 (2000) 309.
Managing Coal-Fired Power Station Solid By-Products
J.T. Gourley
Geopolymer Alliance, WA
Coal-fired power stations are focused on producing
Energy as their primary product. As a general rule
they are forced to treat their secondary products
(flyash, bottom ash and heat) as waste products
since most see the effort to process and market
these bi-products as a distraction from the core
business. From the view of sustainable resource
processing this is unfortunate. With minimal effort
these materials can be converted to useful products
which can bring with them considerable greenhouse
gas credits when used to supplement or replace
existing products.
If addressed as a separate business all solid wastes
can be converted to useful products. Future market
demand will be such that all currently produced
wastes will be saleable as products and previous
wastes that are currently stored in tailings dams or
dry storage sites will be required to be recovered
and processed.
It is theoretically possible to convert all solid wastes
(flyash and bottom ash) from coal fired power
stations, into useful products for the construction
industry. Potential by-products include:
Activated carbon or char, as a filtering medium
or in the steel industry.
Coarse flyash as an extender for Ordinary
Portland Cement (OPC).
Cenospheres.
Conventional cement-sized flyash (Classified
flyash) as a supplementary cementitious
(SCMs) material for inclusion in OPC blends.
Coarsely milled Run-Of-Station (ROS) flyash
as a supplementary cementitious material for
inclusion in OPC blends.
Finely milled ROS ash as an OPC substitute.
Very finely milled classified flyash as a
substitute for silica fume.
Bottom ash as sand for use in geopolymer
concrete or in high volume flyash OPC
concretes.
If a low temperature drying stage is included, flyash
currently stockpiled in tailings dams can be brought
on stream to produce milled ash.
Why does the construction industry need such
products?
Figure 1 illustrates the predictions of the cement
industry for cement usage out to 2020, and the
consequent production rates of carbon dioxide. If
the industry is to attempt to meet community
expectations of significant reductions in greenhouse
gas emissions over this period then large volumes of
supplementary cementitious materials (SCMs) are
going to needed very quickly. The environmental
challenge facing the cement industry over the next
40 years means that ordinary Portland cement will
have to be largely replaced SCMs or fully, by the
use of geopolymer cements.
0
5 0 0
1 , 0 0 0
1 , 5 0 0
2 , 0 0 0
2 , 5 0 0
3 , 0 0 0
3 , 5 0 0
4 , 0 0 0
1 9 9 0 2 0 0 5 2020 (1) 2020 (2) 2020 (3)
G
l
o
b
a
l

O
u
t
p
u
t

(
m
i
l
l
i
o
n

t
o
n
n
e
s

/

y
e
a
r
)
SCM
Ce me n
CO2
Figure 1: Predicted cement production.
(Ref:http://www.concretesdc.org/Sustainability/Sustainability_h
ome.htm).
1990 and 2005 values are the actual global
quantities used (cement and SCMs) and the CO
2
masses generated from the production of these
binder masses.
2020 (1) is the prediction for business-as-usual
but with some attempt to increase SCM proportions
to 20% by mass of binder (the current hopes of the
cement and concrete industry; Ref
http://www.concretesdc.org/Sustainability/Sustaina
bility_home.htm).
2020 (2) is the prediction if the CO
2
levels are to be
kept at 2005 levels.
2020 (3) is the prediction if they are to be reduced
to 1990 levels.
The CO
2
contributions are based on the following
assumptions:
the cement clinker contribution reduces from
1.0 t/t in 1990 to 0.8t/t in 2020 as the fuels used
to fire the clinker kilns change;
the SCM contribution increases to 0.15 t/t in
2020 as milling of coarse flyash becomes
necessary to provide sufficient feedstock at
these high replacement levels.
Scenario (1), business-as-usual, is not an option if
the construction industry is take up its share of the
burden of addressing climate change.
Scenario (2) is just possible with OPC concrete as
usually OPC (the primary binder component) must
be in excess of flyash content for usable concrete
strengths.
Scenario (3) is not possible with OPC concrete
and normal classified flyash and is only
achievable with OPC-Engineered flyash/slag
blends, or with geopolymer concrete where the
SCM (flyash, slag, etc) becomes the (dominant)
binder fraction.
This latter scenario requires nearly 3 billion
tonnes of SCM per annum, worldwide.
The most likely SCMs to be used will be processed
fly ashs (EFAs) and/or ground granulated blast
furnace slags (GGBFSs).
The cost to grind flyash is significantly lower than
that of slags as the vast majority of medium to
coarse flyash particles are hollow spheres that crush
like eggshells. In addition such a process tends to
produce a single-mode particle size distribution
whereas grinding hard, solid slag particle tends to
produce bi-modal distributions; a population of
small chips and a population of large chipped parent
particles.
Just as particle size distribution (PSD) is important
in designing a low cost concrete aggregate skeleton,
so it is in binder design. Optimum particle packing
in binders allows the use of minimum quantities of
fluids to achieve the required workability or flow-
ability of the concrete. As OPC concrete strength
and durability characteristics are inversely
proportional to fluids content (air and water in the
case of OPC concrete) a perfectly packed binder
fraction allows the production of a superior
concrete. As Australian cements do not in general
get close to perfect packing with their PSDs, the
incorporation of EFAs and GGBFSs will produce
superior concretes and at the same time convert
wastes to valuable by-products and reduce
greenhouse gas emissions.
Geopolymer binders just use SCMs as the binder
and as a result of the kinetics of the
geopolymerisation process, allow the use of sintered
or pelletised aggregate particle made from flyash or
slag. Thus geopolymer concretes are well suited to
the use of aggregates which would otherwise be
shunned in conventional concrete mix designs.
In fact, it is possible to make a very commercial
geopolymer concrete entirely from ash and slag
wastes. Flyash can be pelletised using a
geopolymer process to form coarse aggregate.
Larger particles of granulated slag can be used as
coarse aggregate. Bottom ash or fine granulated
slag can be used as the fine aggregate (sand).
Coarse flyash can be used as the binder coarse
fraction. Classified flyash or lightly crushed Run-
Of-Station flyash can be used as the cement-sized
binder faction and fully crushed flyash (EFA) can
be used as the fine binder fraction.
Design your concrete mix for optimum particle
packing, add a small quantity of concentrated
alkaline solution and you have a low-cost, high
performance concrete which enhances the
sustainability of both the power generation and
construction industries.
The Geopolymer Alliance is working to encourage
the power industry to review these sustainability
opportunities and join with research providers
within CSRP to investigate possible process and
product options.
Thermal Character of Geopolymers Synthesised from Class F Fly Ash Containing
High Concentrations of Iron and -Quartz
W.D.A. Rickard and A. van Riessen
Background
Geopolymers are a synthetic inorganic polymer
produced by the alkali activation of
aluminosilicate raw materials. In recent years fly
ash has been proposed as a suitable
aluminosilicate source for use in geopolymeric
materials. The use of fly ash allows for a less
expensive alternative to pure sources as well as
utilising an industrial waste. Due to their
inorganic framework, geopolymers have excellent
thermal stability, far in excess of traditional
cements [1]. These characteristics allow
geopolymers and geopolymer composites to be
considered for use in high temperature
applications such as furnace linings, thermal
insulation and wall panels.
Previous investigations into to the thermal
properties of geopolymers have been conducted
using pure aluminosilicate sources or have not
considered the effect of the impurities introduced
through the fly ash on the bulk thermal properties.
This paper reports the thermal characteristics of
geopolymers prepared with a class F fly ash
containing 15 wt% iron oxide and 20 wt% u-
Quartz, upon heating to 900 C. Characterisation
techniques to be presented include thermal
expansion, TGA, DTA, XRD and SEM.
Geopolymer samples in this study were
synthesised using a Western Australian fly ash as
the aluminosilicate source material. Samples were
prepared to achieve compositional ratios of Si:Al
= 2.3, Na:Al = 0.85 and H
2
O:SiO
2
= 2.0 by
calculating concentrations based on the
amorphous components of the fly ash only. A DI-
24 Adamel Lhomargy dilatometer was used for
the thermal expansion measurements (Figure 1).
All measurements were conducted in accordance
with ASTM E831. Moulded dilatometry samples
were cast into a 1 mL syringe and cut to length to
achieve a cylinder of 15 mm length and 5 mm
diameter. Other dilatometry samples were cut
from larger samples using a diamond saw to
achieve an 8 x 8 x 15 mm rectangular specimen.
Larger, bulk, samples were moulded into 15 mm
diameter cylinders and cut to a 20 mm length.
Thermo-Gravimetric Analysis (TGA)/Differential
Temperature Analysis (DTA) measurements were
performed using a SDT Q600 V3.4 Texas
Instrument in both air and nitrogen atmospheres.
Figure 1: DI-24 Dilatometer.
Figure 2 shows the dilatometer curves for fly ash
geopolymers of various sample dimensions.
Included on the graph is a regional segregation of
the various expansion / contraction stages
observed during heating to 900 C. The regional
model expands on what was initially proposed by
Duxson et al. [2]. The geopolymers in this study
exhibited fundamentally different thermal
expansion at high temperatures to what has been
observed previously in the literature. It is believed
by the authors that this is due to the presence of
impurities introduced via the fly ash precursor and
the effect of sample dimensions.
Figure 3 compares the simultaneous TGA / DTA
thermograms of the geopolymers used in this
study exposed in both air and nitrogen. An
endothermic peak associated with the dehydration
of adsorbed water was observed at approximately
80 C for both atmospheres. A broad exothermic
peak at 400 C was observed for the sample in air
which was not observed for the sample in
nitrogen, suggesting the presence of an oxygen
dependant reaction.
Figure 2: Thermal expansion of fly ash geopolymers. Si :Al
= 2.3, H
2
O:SiO
2
= 2.0.
Three major mass change events were observed to
occur in the geopolymer samples upon heating to
1000 C in air. Dehydration of free water
accounted for 80 % of the total weight loss,
whereas dehydroxylation was approximately 10
%. Interestingly the sample gained weight above
600 C as the iron species oxidised (figure 3, air
TGA).
Figure 3: TGA / DTA results for a Collie fly ash
geopolymer.
The authors of this study suggest the following
two stage phase transition of ferrihydrite to
hematite describes the oxidation reaction observed
in figure 3;
OH FeO FeOOH +
Eq. 1
3 2 2
2 O Fe O FeO +
Eq. 2
Optical investigations revealed the exposure of
geopolymers to 900 C in air changed the colour
of the specimens from grey (similar to that of the
original fly ash) to dark red (similar to that of
hematite). The colour change was not uniform in
the interior of the larger samples. Large radial
cracking was observed in exposed samples.
Observed differences in the microstructure after
high temperature exposure included an increase in
porosity and a reduction in the amount of intact
fly ash spheres, as most had melted during
exposure.
Analysis of XRD patterns indicated an increased
concentration of hematite after heating to 900 C
(Figure 4). Other crystalline phases did not vary
significantly.
Figure 4: XRD patterns of the major hematite peaks.
Continuous line: Unexposed geopolymer, Dashed line:
Thermally exposed geopolymer.
Geopolymers synthesised from fly ash containing
high concentrations of iron oxides and crystalline
silica exhibited variable thermal properties at
temperatures greater than 500 C. This was due to
interactions of the secondary phases, most notably
the iron oxides, with the atmosphere and sample
degradation through cracking. The results of this
study suggest that quartz rich fly ashes do not
significantly reduce the thermal performance of
the resulting geopolymer as key characteristics
such as thermal expansion and phase stability
were largely unaffected during thermal exposure.
Geopolymers synthesised from iron rich fly ashes
were observed to reduce the thermal performance
of the sample due to phase transitions and
oxidisation reactions inducing variable thermal
expansion at high temperatures.
Acknowledgements
Centres Program.
References
1. Duxson, P., Provis, J.L., Lukey, G.C., Mallicoat, S.W.,
Kriven, W.M., van Deventer, J.S.J., Understanding the
relationship between geopolymer composition,
microstructure and mechanical properties. Colloids and
Surfaces A: Physicochemical and Engineering Aspects
2005. 269(1-3): p. 47-58.
2. Duxson, P., G.C. Lukey, and J.S.J. van Deventer,
Physical evolution of Na-geopolymer derived from
metakaolin up to 1000 C. Journal of Materials Science,
2007. 42(6): p. 3044-3054.
Rational Utilisation of Fly Ash for Geopolymer Processing
J. Temuujin, R. Williams and A. van Riessen
As demand for concrete as a construction material
increases, the production of Portland cement will
also increase. However, production of Portland
cement liberates a considerable amount of
greenhouse gas as a result of de-carbonation of
limestone in the kiln during manufacturing of
cement and the combustion of fossil fuels.
Furthermore, Portland cement is also among the
most energy-intensive construction materials,
after aluminium and steel [1]. Therefore, the
search for an alternative binder to replace Portland
cement binder is important endeavour.
Importantly, such so-called geopolymer
materials not only have comparable or superior
properties to traditional cementitious binders, but
also have low greenhouse emissions. The source
materials for the production of geopolymer are
usually fly ash, metakaolin and slag. Fly ash is
thought to be good candidate for source material
because it is the residue from coal burnt in a
thermal power plant and regarded as a waste
product. However, fly ash also contains some
unburnt carbon with the following effects [2]:
Increases the electrical conductivity of
concrete;
Changes colour of mortar and concrete (may
appear black);
Increases demand for water and additives.
Therefore, carbon free fly ash based geopolymers
are expected to have better mechanical properties
than carbon containing fly ash based
geopolymers.
For production of geopolymers low calcium class
F fly ash is preferred as the source material rather
than high-calcium class C fly ash as the presence
of high amounts of calcium may interfere with the
polymerisation process and alter the
microstructure. However, addition of a small
amount CaO to fly ash/cement system accelerates
the hardening process and increases strength [3].
In fact, one of the most important aspects of
adding calcium to a geopolymer system is its
ability to harden at ambient temperature.
In an attempt to improve the performance of fly
ash-based geopolymers and facilitate curing at
room temperature, fly ash was calcined and
calcium compounds were included in the
formulation, respectively.
Collie fly ash was used as the source material for
production of the geopolymer. For de-
carbonation, the fly ash was calcined at 500 and
800
o
C. Sodium silicate D-51 (PQ Australia Pty
Ltd.) and sodium hydroxide solution with 14 M
concentration were used as the activating
solutions.
Analytical grade calcium oxide (CaO) and
calcium hydroxide Ca(OH)
2
were used as calcium
sources. One, 2 and 3 wt. % calcium oxide was
substituted for fly ash in the geopolymer mixture.
The geopolymer samples were made with a Si:Al
ratio of 2.3 and a Na:Al ratio of 0.88. Mechanical
properties were evaluated by measuring
compressive strength of cylindrical samples.
Calcination of fly ash
Preliminary calcination of fly ash caused
approximately a 3% decrease of the amorphous
component of the fly ash while at same time
increasing the crystalline phases; mullite and
hematite. Therefore, 500 and 800
o
C calcined fly
ash can be considered as carbon free, but with
slightly different amorphous and crystalline
compositions. However, carbon free fly ashes
show weaker compressive strength than raw fly
ash. Figure 1 shows influence of fly ash
calcination temperature on compressive strength
of the geopolymers.
Figure 1: Influence of fly ash calcination temperature on
compressive strength of the geopolymers.
0
20
40
60
80
0 500 1000
Temperature,
o
C
C
o
m
p
r
e
s
s
i
v
e

s
t
r
e
n
g
t
h
,

M
P
a
Calcination of the fly ash resulted in a decrease of
the compressive strength of the samples from
55.7(9.7) MPa in non treated fly ash based
geopolymer to 44.3(5.4) MPa for 800
o
C calcined
fly ash based samples. Geopolymer prepared
using raw fly ash shows 0.54 (0.1) % volume
shrinkage while for the 500 and 800
o
C calcined
fly ash samples the shrinkage was 0.83 (0.25) and
5.45 (0.40)%, respectively. Also efflorescence
was observed on the surface of the calcined fly
ash based geopolymers. The results indicate that
the geopolymerisation reaction in the calcined fly
ash sample was less intensive and consequently
depends strongly on the surface conditions of the
fly ash. The difference between the raw and
800
o
C calcined fly ash in terms of the amorphous
and crystalline components is only 3 wt%. But
there are very large differences in terms of
geopolymerisation. The heat treatment or
calcination causes crystallisation of the
aluminosilicates and hematite on the surface of
the spherical particles. Such crystallised
aluminosilicate (mullite) and hematite prevent
dissolution of the aluminate and silicate species
by the activating alkali, consequently there is
excess alkali which is able to migrate to the
surface of the geopolymer during curing which
then crystallises as sodium hydrate carbonate.
Addition of Ca
Figure 2 shows compressive strength changes of
the calcium added geopolymer samples cured at
ambient and 70
o
C.
0
10
20
30
40
50
60
70
80
90
0 1 2 3 4
Added Calcium Compound, %
C
o
m
p
r
e
s
s
i
v
e

s
t
r
e
n
g
t
h
,

M
P
a
CaO added and
20Ccured
Ca(OH)2 added
and 70Ccured
Ca(OH)2 added
20Ccured
CaO added
70Ccured
Figure 2: Compressive strength versus wt% calcium
addition.
Adding calcium compounds improved the
mechanical strength of samples cured at room
temperature while reducing the strength of
samples cured at 70
o
C suggesting the possibility
of different mechanisms of hardening. The data
also indicates that the addition of Ca(OH)
2
is more
beneficial than CaO in terms of compressive
strength. Calcium compound addition is likely to
result in precipitation of calcium silicate hydrate
or calcium silicate aluminate hydrate phases and
at the same time improve the dissolution of the fly
ash in the alkaline medium and subsequently the
geopolymerisation reaction.
Based on the fly ash calcination experiments
presented here, it is suggested that removal of
unburnt carbon from fly ash by calcination is
counter productive. The calcination may
remove the carbon but the concomitant
changes to the surface of the fly ash limit the
degree of polymerisation.
Addition of the calcium compounds CaO and
Ca(OH)
2
improves mechanical properties of
the fly ash based geopolymers cured at room
temperature. Calcium compound addition
reduces mechanical properties of geopolymer
cured at elevated temperatures. Ambient
temperature cured samples exhibit
efflorescence because of the presence of the
excess alkaline solution caused by incomplete
dissolution of the fly ash spheres and low
dissolution rate. Efflorescence formed on the
Collie fly ash based geopolymer is sodium
phosphate hydrate.
As a means of removing unburnt, carbon
calcination of the fly ash is not considered to be
beneficial for both economic reasons and the
reduced physico mechanical properties of the
geopolymers. Addition of the calcium
compounds into geopolymer mixtures cured at
ambient temperature improves mechanical
properties. The reason for the improved
mechanical properties is believed to be due to an
increase in the geopolymerisation reaction and
possible formation of Ca-containing components.
Acknowledgements
Centres Program.
References
1. D.Hardjito, S.E.Wallah, D.M.J.Sumajouw, and
V.Rangan, Fly ash-based geopolymer concrete, Aust.
J. Struct. Eng. 2005, 6: 1-9
2. T.H.Ha, S.Muralidharan, J.H.Bae, Y.C.Ha, H.G.Lee,
K.W.Park, D.K.Kim, Effect of unburnt carbon on the
corrosion performance of fly ash cement mortar.
Construction and Build.Mater. 2005, 19: 509-515
3. S.Antiohos and S.Tsimas, Activation of fly ash
cementitious systems in the presence of quicklime Part
1. Compressive strength and pozzolanic reaction rate,
Cement and Concrete Res. 2004, 34: 769-77
Sustainability Roadmap for the Kwinana Industrial Area / Extended Abstract
K. Schianetz
1
, M. John
1
, D. van Beers
2
, C. Oughton
3
and D. Cooling
3
1
Curtin University of Technology, Centre of Excellence in Cleaner Production, WA
2
GHD, WA
3
Kwinana Industries Council, WA
The Kwinana Industrial Area (KIA) is by far the
largest and most diverse industrial processing
region (with supporting industries) in Western
Australia. As with elsewhere in Australia (and
globally), the KIA is facing sustainability
challenges on various fronts, such as increasing
water and energy scarcity, climate change, an
aging workforce, and growing community
expectations.
Through the Kwinana Industries Council (KIC)
the industries within Kwinana are working closely
on common issues in the KIA, including the
sustainable development of the region. Over the
past decade, significant progress has been made
by the industries towards the improvement of the
sustainability performance at the company and
regional level. Given the ever increasing
sustainability pressures, there is strong desire to
address sustainability challenges in the KIA,
through a pro-active and collaborative industry
approach. Since April 2004 the CSRP research
project on Capturing Regional Synergies in the
KIA is assisting the KIC and its members on
their journey towards sustainability. The overall
objective of the project is to provide practical
support to the companies in the KIA to develop,
evaluate and implement regional synergy
opportunities, thus creating greater eco-
efficiency
1
.
Within the framework of this project a
sustainability roadmap is now being developed to
assist the KIC and its industry members with
strategic decision making on the long-term
sustainability of the KIA. The broad aims of the
Sustainability Roadmap are to:
Guide KIC strategic decision-making on the
long-term sustainability of the KIA;
Provide the basis for stakeholder discussion
on sustainable development;
Scope sustainable development initiatives;
Document progress towards sustainability
goals and targets; and
Benchmark the sustainability performance of
the KIA with other industrial regions and
industry sectors in Australia and
internationally.
The paper will report on the development process
and the preliminary results of the sustainability
roadmap for the KIA, including a historical
perspective on milestones achieved so far and
future aspirations.
Development of a Framework of the KIA
Sustainability Roadmap
The Sustainability Roadmap for the KIA is work
in progress. In a first step a draft framework of the
KIA Sustainability Roadmap was developed
(Figure 1), consisting of a KIA Case-Studies
Matrix and KIC Sustainability Management
Matrix.
Past Present Future
Documentation of
KIA case-studies
KIA SUSTAINABILITY ROADMAP
Where are we now, where do we want to be, and how do we get there?
IN
D
IC
A
T
O
R
S
Current
performance
Aspirational
performance
targets
Time
K
I
A

s
u
s
t
a
i
n
a
b
i
l
i
t
y

p
e
r
f
o
r
m
a
n
c
e
Low
High
KIC action plans for
selected priority areas
Past
performance
KIC SUSTAINABILITY
MANAGEMENT MATRIX
IN
D
IC
A
T
O
R
S
KIA CASE-STUDY MATRIX
Figure 1: KIA Sustainability Roadmap Framework
1
The figure illustrates how sustainability indicators
are used to measure past and current performance
of the KIA and to set aspirational targets. The
KIA Case-Study Matrix could be used to
document significant past events and initiatives
that have led to the KIA sustainability
performance of the present. The KIC
Sustainability Management Matrix will assist the
KIC with formulating action plans for selected
priority areas to achieve the aspirational targets.
The Choice of Sustainability Indicators
The roadmap is based on sustainability indicators,
which relate economic, social and environmental
factors, and therefore provide integrated
information as to how the KIA is progressing
toward Triple Bottom Line sustainability.
The choice and use of indicators for sustainable
development is a critical determinant to sound
decision-making processes
2
. In particular, it is
argued that tailor made indicator sets, addressing
specific stakeholder concerns and supporting the
organisations strategies, are more likely to assist
stakeholders in the achievement of their goals
than approaches which prioritise reporting against
generic off the shelf indicators
3
.
The search for sustainability indicators is
evolutionary
2
. The necessary process is one of
collective learning. This learning process at the
KIC started out with a literature review of
regional sustainability indicator systems being
used internationally, and indicators used by
companies in the KIA. The Sustainability
Reporting Framework of the Global Reporting
Initiative (GRI) was used as a basis for the
development of the initial set of KIA
sustainability indicators
4
. The GRI framework is
the worlds most widely used framework for
reporting on an organisations economic,
environmental, and social performance.
The initial list of possible indicators has been
discussed with different stakeholder groups and
generated the framework of the KIC Sustainability
Management Matrix. At the heart of the matrix is
a set of key performance indicators for each of
seven priority themes, namely greenhouse gas
emissions and energy, water, process residues,
economic performance, air quality and noise,
community, and workforce. Through the KIC
Eco-Efficiency Committee, the matrix is being
populated with relevant primary and secondary
performance indicators, aspirational targets, and
subsequent strategies and action plans to achieve
the targets.
The selection of relevant sustainability indicators
is subject to the following criteria:
Measurable (relatively easily obtainable and
build upon existing data);
Understandable for internal and external
communications;
Influenceable by KIC and its industry
members;
Relatively small number of key indicators;
Provide holistic overview of the sustainability
performance of the KIA; and
Provide the basis for benchmarking KIA with
other (inter)national industrial areas.
As part of the extension of the CSRP Kwinana
Synergies Project, the KIC Sustainability
Management Matrix will be further developed
with KIC internal and external stakeholders. This
will include collection of historical industry data
to quantify selected sustainability indicators
(historical trend analysis), documentation of
Kwinana case-studies, and engagement with KIC
industry members, government and community.
The KIA Sustainability Roadmap is work in
progress. The work so far focused on the
development of the roadmap framework, the
review international sustainability indicators, and
the development of the KIC Sustainability
Management Matrix, including the selection of
indicators relevant to the KIC. The next steps
include setting of five year aspirational targets and
formulating subsequent strategies and action plans
to meet these targets. The focus will be on
activities where the KIC has an influence and can
make a positive impact. Engagement and
discussions with KIA stakeholders will be an
important part of this process. In its final form, the
KIA Sustainability Roadmap will enable ongoing
constructive stakeholder discussions (e.g.
government and community) on emerging local
issues and possible future directions for the KIA.
Acknowledgements
Centres Program. The authors wish to
acknowledge the commitment and support from
the Kwinana Industries Council and its members
to the research presented here.
References
1. Van Beers, D. 2008 Capturing Regional Synergies in
the Kwinana Industrial Area. 2008 Status Report,
CSRP, Perth, Western Australia.
2. Meadows, D. 1998 Indicators and Information Systems
for Sustainable Development. A Report to the Balaton
Group. The Sustainability Institute, Hartland Four
Corners, VT.
3. Warhurst, A. 2002 Sustainability Indicators and
Sustainability Performance Management. Report
commissioned by the Mining, Minerals and Sustainable
Development (MMSD) project of International Institute
for Environment and Development (IIED).
4. Global Reporting Initiative (GRI) 2000 Sustainability
Reporting Guidelines. Global Reporting Initiative,
Amsterdam.
Production and Application of Red Sand
TM
E. Jamieson, A. Jones, L. Guilfoyle, D. Cooling and S. Attiwell
Alcoa World Alumina, WA
The extraction of alumina from bauxite produces
a high volume by-product which is stored in
secure impoundments close to the refining
operations.
In Western Australia, typically half of this by-
product has a particle size in excess of 90 m due
to the high quartz content of bauxite from the
Darling Range. This coarse fraction can be further
processed to produce a general purpose sand
termed Red Sand.
Alcoa World Alumina Australia (Alcoa) operates
three alumina refineries in Western Australia,
producing a potential resource of up to 20,000
tonnes of Red Sand per day.
A variety of uses for bauxite residues has
previously been outlined by Jamieson et al (2005).
Primarily Red Sand
TM
has been deemed to have
physical properties suitable for use as a general
fill material within the construction industry and
road construction.
Attiwell (2007) has shown that Red Sand can
be produced to a standard suitable for branding
with the Centre for Sustainable Resource
Processing (CSRP) ReSand trade mark. The
CSRP brand ReSand represents by-product sand
from the mining industry independent of source
having met application and sustainability criteria.
Alcoa has now embarked upon production of
suitable quantities of Red Sand to provide to the
CSRP for demonstration projects as a part of the
CSRPs broader evaluation of ReSand.
To demonstrate the use of Red Sand as
ReSand a pilot plant has been constructed to
provide approximately 5,000 tonnes of material
for use in these trials.
The pilot plant was commissioned on 13/06/2008.
Production capacity has reached design
specification of 10 t/hr, with the plant being run
typically over an 8 hour period.
Figure 1: Delivery of the flat bottom classifier to
construction site.
Figure 2: Installation of dewater sand screen.
Key operational goals include consistent sand
washing and neutralisation via carbon dioxide
dosing within the flat bottom classifier.
Standard operation has been able to produce clean
sand washed free of fine residue and neutralised to
a pH of less than 10.5. This value can be lowered
depending upon product use requirements.
Figure 3: Sand stockpile in production.
Demonstration 1
The First application being demonstrated is the
use of sand for top dressing of turf.
Assessment of the Red Sand
TM
has ensured all
factors have been assessed as acceptable
including, leaching, composition, particle size,
dusting potential, phosphate retention and pH.
Preliminary trials have been successfully
completed on a cored golf course fairway,
resulting in improved drainage and greater
moisture penetration.
A larger demonstration is to use 1000 cubic
meters of sand at Fairbridge Village for top
dressing of the sports oval. Trials of Red Sand
TM
application to the oval (1 m
2
) have shown typical
top dressing performance as expected.
Figure 4: Grass growth through the sand top dressing,
before and after 25 days.
Alcoa has provided the sand including cartage to
site as part of its CSRP collaborative research
project.
Demonstration 2
The second major trial will be the use of Red
Sand in road construction. Assessment of
particle size, composition, leach analysis and
strength parameters have indicated trial materials
are well suited for this application.
Sand for this trial is currently being produced and
samples are being assessed to confirm results. A
freeway onramp has been earmarked by Main
Roads for construction using some 2000 m
3
of this
material in the first quarter of 2009.
Pilot plant operation has demonstrated
technical and operational capability.
Planning for demonstration projects utilising
the Red Sand
TM
are well underway
Collaboration under the guidance of the CSRP
leading to demonstrated viability of by-
product utilisation.
With the successful demonstrated uses of Red
Sand
TM
, Alcoa will be looking toward larger scale
production of Red Sand and commercialised
uses for the product.
Acknowledgements
the CSRP, which is established and supported
under the Australian Governments Cooperative
Research Centres Program. Special thanks to the
Wagerup Refinery Residue Area Group of Alcoa
World Alumina, Australia, for continued
assistance to pilot plant operations. Thanks to
Linatex Pty Ltd for delivery of the pilot plant and
associated assistance. Thanks are extended to the
Main Roads Department WA and also Chemistry
Centre WA for assistance with demonstration
trials.
Finally thanks to Iggy Castle of Fairbridge Village
for her enthusiastic support of the trial.
References
1. Jamieson, E.J., Cooling, D.J. & Fu, J. (2005) High
volume resource from bauxite residue. Proceedings of
the 7th International Alumina Quality Workshop. Perth
October 2005. pp 210- 213.
2. Attiwell, S. (2007). Classification of ReSand and
Leach Testing of Alkaloam and Red Lime. Centre
for Sustainable Resource Processing (CSRP) Inaugural
Conference. Melbourne 21st November, pp 73-74.
Using Life Cycle Assessment to Compare the Environmental Performance of
Bauxite Residue, Lime and Bauxsol in the Treatment of Acid Mine Drainage at
Mount Morgan Mine
D. Tuazon
1
, G.D. Corder
1
, L.Q. Harris
2
1
University of Queensland, Sustainable Minerals Institute, Centre for Social Responsibility in
Mining, QLD
2
Hatch Engineering, QLD
Introduction to Seawater Neutralised Red
Mud and Bauxsol LCA
In December 2005, a supplementary project to the
CSRPs Project 3C1 (Developing Local Synergies
in the Gladstone Industrial Area) was initiated to
investigate the holistic environmental merits of
utilising bauxite residue (or red mud) for acid
mine drainage (AMD) remediation at Mount
Morgan, a historical gold and copper mine in
regional Queensland approximately 40 km south-
west of Rockhampton.
The findings of the initial study [1] found that red
mud was better than lime from an environmental
perspective, with the former having a much lower
carbon dioxide inventory. The study and the
results gave further impetus to pursue an
extension LCA by considering the life cycle of
another neutralant Bauxsol. Bauxsol is a
proprietary, red mud derived product from Virotec
International which is designed to treat AMD.
Like red mud, Bauxsol was considered as a
candidate for remediation and the Department of
Mines and Energy commissioned a study to
investigate the feasibility of using different
neutralants at Mount Morgan. The study by EWL
Sciences Pty Ltd [2] concluded that lime was the
most favourable option based on economics,
practicality and technological considerations. The
result of this study was the planning and
subsequent installation of a lime dosing plant at
Mount Morgan [3].
Note that Bauxsol actually refers to many
different blends of related products which are
tailored to suit different remediation scenarios. In
the case of Mount Morgan, the blend Acid B
Extra was suggested. Nevertheless, in this
paper, the product analysed will be referred to as
Bauxsol.
Goal and Scope Definition
The life cycles of each of lime, seawater
neutralised red mud and Bauxsol involved
tracking the inventories of fuel usage, electrical
consumption and carbon dioxide emissions. The
carbon dioxide emissions inventory was decided
as the primary yardstick of comparison between
all neutralants. The functional unit and basis was
1000 cubic metres of untreated, acid mine
drainage water.
Life Cycle Determination
The life cycle of each neutralant involved the
extraction of raw materials and manufacture at
Gladstone (Queensland) before being transported
to Mount Morgan for remediation application and
disposal. In the cases of the lime and red mud
scenarios, the remediation application was using a
dosing plant which used grid electricity, whereas
with the Bauxsol scenario, a portable, on-site
processing facility was used for the remediation
process. The Bauxsol processing facility is
powered by fuel-powered generators. In all cases,
the grave of each life cycle involves disposing the
spent neutralant into the Mount Morgan pit.
Bauxsol presented another interesting life cycle
aspect, which was the possibility of additive lime
in the Bauxsol.
The cradles for each of the life cycles were the
locations where the raw materials for each of the
neutralants were extracted; these were:
Lime Scenario: East End Mine (Gladstone).
Seawater Neutralised Red Mud Scenario: Red
Mud Dam (Gladstone)
Bauxsol Scenario: Red Mud Dam
(Gladstone)
Life Cycle Inventory and Impact Analysis
Table 1 summarises the inventory data for the
base case. In the base case, it is seen that using red
mud produces 81% less carbon dioxide than using
lime and Bauxsol up to 92% less. However the
incremental benefit achieved by using Bauxsol
instead of red mud is lessened or even reversed if
lime is required to be added to the Bauxsol as
could be the case in at least some applications.
Table 1: Key Life Cycle Assessment Results (Basis: 1000 m
3
of AMD water).
Scenario
Fuel
(L)
Electricity
(kWh)
CO
2
(kg)
Lime 16 377 4378
Seawater
neutralised red
mud
190 164 853
Bauxsol 70 0 335
In addition to benefits in terms of carbon dioxide
inventory, the use of Bauxsol requires less
electricity than using either lime or red mud
(although it should be noted that, in this study, the
Bauxsol production and application facilities
are powered by fuel-powered generators) and less
fuel than using red mud. The reason for less fuel
is that Bauxsol is more potent as a neutralant
than red mud per unit mass, thereby reducing the
amount of fuel required to transport neutralant
materials to Mount Morgan (this reduction in fuel
usage is reflected in a drop in the carbon dioxide
inventory).
Sensitivity analyses revealed that the carbon
dioxide inventory of the Bauxsol scenario is a
strong function of the amount of lime that is
added into the treatment media. The reason for
this is that manufacturing lime produces a
significant amount of carbon dioxide and the
environmental burdens associated with adding
lime to Bauxsol quickly compromise the carbon
dioxide benefits of the neutralant compared to red
mud. Analysis showed that even with the worst
base neutralisation power of Bauxsol of 5 mol
H
+
/kg, the carbon dioxide inventory of Bauxsol
was still lower than that of both the red mud and
lime scenarios. When additive lime was
introduced into the Bauxsol treatment media,
the carbon dioxide inventory of Bauxsol was
still much lower than using lime but could easily
be higher than that of red mud depending on the
base neutralisation power of Bauxsol. The
combined effects of neutralisation power of
Bauxsol and additive lime on the Bauxsol
scenario carbon dioxide inventory can be
summarised graphically in Figure 1.
Conclusion
The study illustrates that Bauxsol, over its life
cycle, generates less carbon dioxide and uses less
fuel than both red mud and lime, although the
benefits of Bauxsol compared to red mud are
less than those of red mud compared to lime. The
results also show that the improved environmental
benefits of Bauxsol compared to red mud may
be marginal (or even negative) depending on the
amount of lime added.
Acknowledgements
This project is carried out under the auspice and
with the financial support of the Centre for
Sustainable Resource Processing, which is
established and supported under the Australian
Governments Cooperative Research Centres
Program.
The authors would further like to acknowledge the
contributions and support from Mansour Edraki
(The University of Queensland, Sustainable
Minerals Institute, Centre for Mined Land
Rehabilitation), Philip Bangerter (Hatch
Engineering) and Daniel Blair (Virotec
International).
Bauxsol and Acid B Extra are trademarks of
Virotec International.
Disclaimer
This study was intended to provide an objective
desktop investigation of the environmental aspects
of using red mud, lime or Bauxsol for
remediating acid mine drainage. It was not
intended to endorse (or otherwise) the use of any
particular neutralant, nor attest to their efficacies
as AMD neutralants.
References
1. Tuazon, D. and Corder, G.D. (2006) Life Cycle
Assessment of Seawater Neutralised Red Mud from
Gladstone for Treatment of Acid Mine Drainage at
Mount Morgan Mine. Centre for Social Responsibility
in Mining: Brisbane.
2. Jones, D. (2002) Technical Input to Contingency
Planning Process and Review of Water Treatment
Options for Water in the Open Cut Pit at Mount
Morgan. EWL Sciences: Darwin.
3. Unger, C., et al. (2003) Rehabilitation Plan for the
Mount Morgan Minesite Central Queensland.
Department of Natural Resources and Mines,
Queensland Government: Brisbane.
Figure 1: Bauxsol CO
2
LCI vs. Neutralisation Power of Bauxsol and Additive Lime Content.

CSRP08 Proceedings Lowres - Pdf#page 49

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

CSRP08 Proceedings Lowres - Pdf#page 49

Uploaded by

Copyright:

Available Formats

CSRP08

Delivering sustainable solutions to the minerals and metals industries

A structured, methodical mechanism for incorporating sustainability principles into

A structured, methodical mechanism for incorporating

will be able to be applied

concept is complete. Industry-based

Tests to reproduce these

Input - force, energy

mechanism options which will

and a variety of silicate

and a variety of silicate

You might also like