AIR RECOVERY - Workspace - Imperial College London

Report
UNLOCKING OPTIMAL FLOTATION:
is the AIR RECOVERY the key?
Jan Cilliers
Royal School of Mines
Imperial College London
Outline
The Origins of Air Recovery
• Modelling Flotation Froths
• Useful froth equations
Air Recovery Application
• Measuring air recovery
• Air rate effect and flotation performance
• Bank air profiling using air recovery
What is the Air Recovery?
•Air leaves a flotation
cell by bursting on the
top of the froth or
overflowing into the
concentrate.
Air leaving froth by bursting
at top surface
Air overflowing
the weir as froth
•The AIR RECOVERY
is the fraction of the
air that that overflows
(and does not burst)
Froth
concentrate
Air into the cell
The Origins of Air Recovery
• Modelling Flotation Froths
• Useful froth equations
Froth Flotation and Froth Physics
The surface chemistry determines
whether the minerals can be separated
The froth physics determines how well
the separation happens
Requires a froth-phase model describing
the physics
A Flowing Froth Model - components
• Froth motion
• Liquid flow in the froth
• Solids motion
Froth Structure:
The Physics of the Froth
Films between bubbles
Plateau borders
Froth motion from pulp to concentrate
Laplace equation
gives velocity
Boundary conditions:
1. Shape of tank and
launders
2. Air entering the froth that
overflows:
AIR RECOVERY (%)
Froth Flow in Radial Equipment Designs
Liquid Flow in the Froth
Three balanced forces act on the liquid in
Plateau borders:
Gravity, capillary and viscous dissipation
2
 k 2  A  2


v

A

A
x
λ 
 k2 A 2 
A
vx 


 2 A  x 
x
x
x 

λ 
A

   k2 A
 v x A
x 
x

2
2



v



A
k

A

A

A
y
2
   k 2 A 2 
 λ  2k1 A

A
vy 

y 2 A  y 
y
y
y 



λ 
A
2
   k1 A  k 2 A
 v y A   0
y 
y

Liquid Motion and Content
Solids Motion
1. Attached Solids
Particles attached to bubbles
move with the froth
Most particles are detached due
to coalescence (>95%)
2. Unattached Solids:
Particles move in the Plateau
borders
Follow the liquid, settle and
disperse
Overflow into concentrate
Mineral and Waste Particles
Example of motion in Plateau borders
Valuable Mineral
Gangue Minerals
Mineral grade in froth
Froth Launder Design:
Effect of forcing froth to flow inwards or outwards
INTERNAL
CHANNEL
Internal Launder
CHANNEL 1
CHANNEL 2
Two Launders
Model validation
Tracking particles in
flotation using PEPT
Model validation
Tracking particles
in flotation using
PEPT
Simplified Equations for Flotation Modelling
Water flowrate to concentrate
Entrainment factor
Froth recovery
(α<0.5)
Water flowrate to concentrate
Ql  k
Acolumn J
d
2
bubble
2
g
 (1   )
ENTRAINMENT FACTOR
Ratio of gangue recovery to water recovery
1.5


v
h
settling froth


Ent  exp 
 D J g (1   ) 


Froth Recovery
R froth
 J g 


v

 settling 
f
2
 dbubble,in

d
 bubble,out




f
2
Froth Modelling Summary
• Froth physics determines the effectiveness
of the flotation separation
• Complex froth zone simulators are available
for operation and design
• Simplified models have been developed for
liquid recovery, froth recovery and
entrainment, based on the physics
All the froth models include
THE AIR RECOVERY
Air Recovery Application
• Measuring air recovery
• Air rate effect and flotation performance
• Bank air profiling using air recovery
Air recovery.. a reminder
Air leaves a flotation
cell by bursting on the
top of the froth or
overflowing into the
concentrate.
Air leaving froth by bursting
at top surface
Air overflowing
the weir as froth
The AIR RECOVERY is
the fraction of the air
that that overflows
(and does not burst)
Froth
concentrate
Air into the cell
Measuring the air recovery
Air leaving
through
bursting
Overflowing
froth height
Air flowing
over lip
Air Recovery =
Volumetric flowrate air overflowing
Air flowrate into cell
Air In
Volumetric flowrate air overflowing
= overflowing velocity x overflowing
froth height x lip length
Air Recovery shows a maximum (PAR)
at a specific air rate
Why is there a Peak in Air Recovery (PAR)?
Optimum balance
between froth stability
and motion
Air
Recovery
Bubbles heavily loaded
Stable, but move slowly
Bubbles under-loaded
Unstable, burst quickly
Air Velocity into Flotation Cell
Predicting air recovery – theory
0.25
P
*
Crit
0.2

1
v *g 1   
 N v *g

0.15
0.1
v*g 
vg
0.05
vg ,int
0
P
*
Crit
PCrit

PCrit ,int
0
2
4
6
8
10
vg*
0
-0.05
-0.2
12
14
16
Air Recovery and flotation performance
Air rate that gives highest air recovery also gives
highest mineral recovery
Froth appearance
• Air rate 8m3 min-1
• Air rate 12m3 min-1
• Air recovery 70%
• Air recovery 40%
Why does the Air Recovery affect flotation?
Optimum balance between froth
stability and motion
High recovery and grade
Metallurgical
Recovery

Air
Recovery
INCREASE AIR
REDUCE AIR
Reduce grade
Increase recovery
Increase grade
Increase recovery

Bubbles heavily loaded
Stable, but move slowly
Bubbles under-loaded
Unstable, burst quickly

Air Velocity into Flotation Cell
Air Recovery Application
• Measuring air recovery
• Air rate effect and flotation performance
• Bank air profiling using air recovery
Air rate profiling
The air rate profile in a flotation bank affects the
performance
• Two strategies:
1. Determine the best air rate profile
– Vary distribution of a set total air addition
1. Determine the optimal total air addition
– Vary the total air addition with a set air profile
12
10
10
-1
12
Inlet air rate / m min
8
3
Inlet air rate / m3 min-1
Air rate profiling approaches
6
4
8
6
4
2
2
0
0
Balanced
Increasing Decreasing
Humped
1. Different air profiles with
same total addition
(e.g. Cooper et al., 2004)
Low
Intermediate
High
2. Different air addition
with the same profile
(Hadler et al., 2006)
Air Profiling Strategies
1. Determine the best air rate profile
– Vary distribution of the total air addition
– Increasing profile typically improves performance
e.g. Cooper et al., 2004; Gorain, 2005; Hernandez-Aguilar and
Reddick, 2007; Smith et al., 2008
1. Determine the optimal total air addition
Determining the air rate profile
• Increasing profile typically yields better performance
Higher cumulative grade for same cumulative recovery
(e.g. Cooper et al., 2004)
Introduction: Previous work
1. Determine the best air rate profile
1. Determine the optimal total air addition
– Best performance at air rate giving
Peak Air Recovery (PAR)
e.g. Hadler et al., 2006; Hadler and Cilliers, 2009
Cu Rougher Performance:
25%
35%
34%
33%
32%
31%
30%
29%
28%
27%
26%
25%
76.3%
Cumulative
recoveries:
75.6%
0%
20%
40%
60%
80%
Cumulative Recovery (% Cu)
As Found
Peak Air Recovery
100%
Cumulative air recovery (%)
Cumulative Grade (% Cu)
Grade-Recovery and Air Recovery
20%
15%
10%
5%
0%
As Found
Peak Air
Recovery
Study performed in two stages
1. Air rate profiling tests
2. Air recovery optimisation (PAR) tests
First direct comparison of the two approaches
Stage 1: Air rate profiles
1. Air rate profiling tests
– Three profiles tested, the ‘Standard’ and
two others, all adding same total air
2. Air recovery optimisation
Air rate profiling: Air rate profiles
Air flowrate / m 3 min -1
10
Profile
8
6
4
2
0
Standard
Stepped
Sawtooth
Total air
addition
(m3 min-1)
Standard
30.3
Stepped
28
Sawtooth
29.5
Air rate profiling: Performance
60
Standard
Stepped
Sawtooth
Cumulative Air Recoverey / %
Cumulative upgrade ratio
160
120
80
40
40
20
0
20
30
Cumulative recovery / %
40
0
D
Stepped
Sawtooth
Standard
Air rate profiling: Findings
• Order of cumulative Cu recovery is same as cumulative air
recovery
– Sawtooth > Stepped > Standard
Mineral recovery and air recovery qualitatively linked
Stage 2: Peak Air Recovery test
1. Air rate profiling test
2. Air recovery optimisation
– Preliminary tests to find PAR air rates
– Test conducted at PAR air rates
– Total air added same as ‘Standard’ profile
Air recovery optimisation:
Preliminary tests
Air recovery / %
80%
60%
40%
20%
Cell A
Cell B
0%
5
7
9
Air flowrate / m3 min-1
11
Air recovery optimisation:
Air rate profiles
Air flowrate / m 3 min -1
10
Profile
8
6
4
2
0
Standard
Stepped
Sawtooth
PAR
Total air
addition
(m3 min-1)
Standard
30.3
Stepped
28
Sawtooth
29.5
Peak Air
Recovery
28
Air recovery optimisation: Air recovery
Cumulative air recoverey / %
80
Standard
Stepped
Sawtooth
PAR
60
40
20
0
A
B
Cell
C
D
Air recovery optimisation:
Performance
Cumulative upgrade ratio
160
Standard
Stepped
Sawtooth
PAR
120
80
40
0
20
30
40
Cumulative recovery / %
50
Air recovery optimisation:
Performance of first cell
• Effect of air rate:
– Recovery
maximum at PAR
air rate
– Upgrade ratio
decreases with
increasing air rate
Air profiling using air recovery: Summary
• Air profiling can significantly improve flotation
performance
• The performance improvement is a froth
effect; rate kinetics alone cannot explain it
• The air rate giving the highest air recovery
(PAR) also gives the best flotation
• The PAR method simultaneously determines
the optimal bank air rate and distribution
Summary and Conclusions
• Froth physics determines the effectiveness of flotation
• Froth models indicate important variables – this is the
origin of AIR RECOVERY
• Air recovery is affected by air rate; there is an air rate at
which the air recovery is a maximum (PAR)
• The Peak Air Recovery (PAR) methodology
simultaneously establishes the correct air addition rate
and the best air rate profile for a flotation bank
• Significant improvements observed; plant control strategy
Acknowledgements
• Rio Tinto Centre for Advanced Mineral Recovery at
Imperial College London
• Froth and Foam Research team
• Intellectual Property Rights
• The peak air recovery-based froth flotation optimisation
methodology is protected by a PCT-stage patent
application, covering most of the countries of the world,
with additional protection in Chile and Peru
Questions?
UNLOCKING OPTIMAL FLOTATION:
is the AIR RECOVERY the key?
Jan Cilliers
Royal School of Mines
Imperial College London

similar documents