Flotation 2
Flotation 2
,/"
1. Senior Metallurgist, Magotteaux Australia Pty Ltd, Suite 4,83 Havelock Street, West
Perth WA 6005. Email: [email protected]
2. MAusIMM, Principal Metallurgist. Magotteaux Australia Pty Ltd, Suite 4, 83
Havelock Street, West Perth WA 6005. Email: [email protected]
3. Senior Metallurgist, Kagara Zinc Ltd, Mount Garnet Operations, 4 Zinc Street, Mount
Gamet Qld 4872. Email: [email protected]
ABSTRACT
High chrome grinding media has been used in the mining industry for quite some time, as it
often proves economical on a wear basis. However it is unusual for an operation to change to
high chrome solely due to its metallurgical benefits. Kagara's Mount Garnet operation made
the change from forged steel to high chrome media after laboratory testwork indicated that
copper grades and recoveri~s would improve. The outcome was remarkable with positive
changes in pulp chemistry throughout the circuit accompanied by marked improvements in
copper, lead and zinc grades and recoveries. Reagent dosage rates were reduced, often by
more than half, and the grinding media wear rate dropped to less than a third of the wear rate
of the forged steel media.
The economic analysis showed that out of the total increase of revenue and savings from
metallurgical improvements and reagent and media consumption reductions the metallurgical
improvements made up approximately 73 per cent, the reduced reagents saved 25 per cent
and the savings through reduced media wear contributing by just over one per cent. This
indicates that the popular view of high chrome media only being trialled in the plant for wear
economics is outdated and the focus of operations should be shifted to using high chrome
media for the metallurgical benefits.
Theory
The use of forged steel (high carbon, low alloy steel) grinding media to liberate valuable
minerals prior to flotation is common practice due to the relatively low unit cost of this media
compared to other forms. However from a wear and chemistry perspective, forged steels may
not be the most cost effective proposition. It is known that total wear of grinding media is due
to a combination of corrosive, abrasive and impact wear. but it is difficult to separate the
contribution from each mechanism during wet grinding (Iwasaki et aI, 1985). Despite this,
investigations into wear mechanisms have been conducted (Natarajan and Iwasaki, 1984 and
Iwasaki el aI, 1985), and have reported that corrosive wear results in more than half of the
total media wear and others report it is accelerated when in the presence of reactive sulfide
minerals (Adam et ai, 1984; Iwasaki et ai, 1983; Isaacson, 1989).
Besides the obvious consumable cost, the impact of corrosive grinding media wear on
downstream processing is quite remarkable and has been studied for some time. Metallic
corrosion reactions in aqueous solutions are electrochemical in nature (Isaacson, 1989) which
means that galvanic interactions occur in every grinding media-sulfide mineral system. In
33
._--- ........
__
._--.-
_.,------_
...........",. '.
many reported cases, galvanic interactions have been found to adversely affect the flotation
response of sulfide minerals by affecting their surface properties (Adam et ai, 1984; Iwasaki
et ai, 1983; Natarajan and Iwasaki, 1984; Kocabag and Smith, 1985; Yelloji Rao and
Natarajan, 1989a and 1989b; Forssberg et ai, 1993; Cheng et ai, 1993; Yuan et ai, 1996;
Greet and Steinier, 2004, Greet et ai, 2005).
Electrochemical reactions are made up of the sum of two half reactions as one cathodic, or
reduction, and the other anodic, or oxidation. In a grinding media - sulfide mineral system,
due to the differences in rest potentials, complex galvanic cells are set up where the sulfide
mineral can act as an anode or cathode, depending on its contact with other sulfide minerals,
media and reagents (Cheng, Smith and Iwasaki, 1993). Generally, the forged steel balls act as
anodes relative to the sulfide mineral cathode:
(1)
-+
20R
(2)
Fe
-+
+ 2e'
The dissolution of iron ions from the grinding media bonds with the hydroxyl species and
fonns hydroxyl complexes of iron which are hydrophilic and have been seen as coatings on
the sulfide mineral surfaces (Adam et ai, 1984; Natarajan and Iwasaki, 1984; Yelloji Rao and
Natarajan, 1989a and b; Forssberg et ai, 1992). These iron hydroxide surface coatings affect
the floatability of the sulfide minerals, not only due to their hydrophilic nature but also by
hindering effective collector adsorption. Interestingly, it has been noted that without a
galvanic contact between the sulfide and the iron, the values of mixed potentials were in the
vicinity of the xanthate/dixanthogen redox potential. However, when in contact with iron, the
mixed potentialS shifted towards a more negative value, making them too reducing for a
xanthate collector to be adsorbed effectively (Adam et alI9M).
One method of preventing these oxidation reactions occurring is to substitute inert grinding
media into the system. As ceramics would be unsuitable in this situation and the expense of
stainless steel media is prohibitive, high chrome white iron should be considered. High
chrome grinding media initially rapidly corrodes to form a very hard chrome oxide layer
which effectively then prevents the exchange of electrons and halts the corrosion process.
This frees the oxygen in the system to oxidise the sulfide minerals such as pyrite and
pyrrhotite and therefore inhibit the flotation of these minerals (Buckley and Woods, 1985;
Buckley and Woods, 1987). The presence of oxygen is vitally important to assist in collector
adsorption and mineral floatability (Gaudin, 1974).
There have been many laboratory tests conducted comparing the effect of varying media
types during grinding on flotation performance, and even pilot scale operations where the
effect of changing the media type was not so clear, but minimal work has been conducted in
fuJI plant scale operations. This paper will demonstrate how changing the grinding media
from forged steel to a high chromium alloy at Kagara Zinc's Mount Garnet operations
effectively resulted in a statistical increase in copper, lead and zinc flotation performance.
Problem definition
Kagara Zinc is the lowest cost zinc producer in Australia, with interests in MunganalRed
Dome, the Greenvale region and Admiral Bay. The company operates two concentrators
34
Metallurgical Planl Design wui Operating S/Tategiu (Me/Plant 2006)
18.19 S.p/ember 2006 Per/h. WA
on its lease.
'The 'Surveyor' ore proved difficult to treat, with the main problem being poor selectivity for
chalcopyrite against pyrite. Table 1 clearly showS that the copper, lead and zinc minerals
were misreporting into the wrong concentrate with approximately 26 per cent copper
reporting to the lead and zinc concentrates, 19 per cent lead in the copper and zinc
concentrates, and nine per cent zinc in the copper and lead concentrates.
Table 1: Copper, lead and zinc grades and recoveries for Mount Garnet,
Fehrwary and March 1004.
Recovery
Fe, %
-1
Grade
Zn,%
Pb,%
Cu,%
Fe, %
Zn,%
100
Pb,%
100
Cu,%
100
100
16.8
16.2
7.4
0.8
4.1
Feed
2.1
8.5
66.2
20.9
0.3
1.3
17.1
3.1
Cu Con
7.1
74.5
14.8
5.2
2.9
57.7
16.5
12.0
Pb Con
84.7
10.6
11.6
7.2
50.3
11.9
9.8
80.8
Zn Con
6.0
6.3
7.3
Tailings
\
I
Ininal site visit and laboratory testwork completed by OPTIMET Laboratories (Wong, 2003a,
2003b) demonstrated a strong relationship between grinding media type and pulp aeration
with increased copper flotation kinetics and grades. Further laboratory tests by Magotteaux
confirmed Wong's results and demonstrated the metallurgical advantages of high chrome
grinding media (Kinal, 2004a, 2004b).
A plant survey was undertaken, highlighting that the Mount Gamet grinding and copper
flotation circuits contained very low levels of dissolved oxygen and reasonably low Eh values
which is conducive to pyrite flotation. It appeared that the oxygen was consumed during the
grinding process, possibly through grinding media corrosion, and therefore not available to
oxidise the pyrite to prevent its flotation. The amount of iron sulfide reporting to the copper
concentrate was therefore quite high, averaging 10.60 per cent for the period measured. This,
along with the problem of the valuable minerals misreporting to the wrong concentrates
indicated that there was a massive problem with selectivity throughout each of the three
circuits.
Laboratory
testing
Laboratory grinding and flotation tests were conducted to compare the effects of grinding
with forged and 30 per cent chrome media on copper flotation response. Pulp chemistry and
EDT A extractable iron measurements were taken during the flotation tests.
Table 2 shows the mill discharge and flotation feed pulps became considerably more
oxidising with the shift from forged to 30 per cent chrome grinding media. The more
oxidising conditions observed when the 30 per cent chrome grinding media was employed
were accompanied with an increase in dissolved Ox.ygen levels in the mill discharge. There
also appeared to be a decrease in pH which suggests that there may be more rapid oxidation
35
Metallurgical Plant DeSIgn and OperatIng Strategies (MeiPlant2006)
18 _ 19 September 2006 Perth, WA
of pyrite when the 30 per ,cent chrome grinding media was used. If this is true then it is
expected that selectivity for chalcopyrite against pyrite should improve.
Table 2: Pulp chemistry data for laboratory mm discharge.and copper circuit feed for
plant reed samples ground with forged and 30 per cent chrome media.
Flotation feed
M ill discharge
Eh,mV
(SHE)
pH
DO, ppm
7.01
0.00
Forged standard
67
354
6.52
4,26
30% Cr standard
Media
Eh,mV
(SHE)
pH
DO, ppm
S8
6.79
0.00
204
6.36
0.00
The copper grade recovery curves (Figure 1) demonstrate that the forged standard test (no
aeration) is bent over, indicating that another mineral is preferentiaHy floating first, which is
similar behaviour to that noted in the Mt Gamet plant. The flotation rate of the chalcopyrite is
quite slow, and results in low concentrate grades. When the 30 per cent chrome grinding
media is substituted for, the forged media, the copper flotation performance markedly
improves, with a faster flotation rate and higher grade of copper in the fust concentrate. The
improvement in copper grade when grinding with 30 per cent chrome media is due to better
selectivity against galena, sphalerite, iron sulfides and non-sulfide gangue (Table 3).
.~
O'
15.0
\ e
10.0
\ :i"
0
GO
1:1
C>
. 5.0
l_
00
0.0
20.0
40.0
60.0
80.0
RlICovery . Cu(%)
100.0
._--------Figure }; Copper grade recovery curves for copper rougher tests completed 00
Surveyor ore comparing the effect of grinding with forged and 30 per cent chrome
media.
36
Metallurgical Plant Design and Operating Strategies (Me/Plant 2006)
18 _ /9 September 2006 Perth, WA
Table 3: Copper and silver grades llnd silver and diluent recoveries at 66.2 per cent
copper recovery for copper rougher tests completed on Surveyor ore comparing the
effect of grinding with forged and 30 per cent chrome media.
Recovery
Grade
NSG,%
Media
IS,%
Zn,%
Pb,%
Ag,%
Ag, ppm
Cu,%
1.78
2.60
2.93
4.62
15.54
604
14.35
Forged
0.91
1.87
1:62
2.65
14.90
750
18.96
30% Cr
Digital photographs were taken during these flotation tests and can be seen in Figures 2 and
3. The flotation rate of copper is clearly depicted in these photographs, with the 30 per cent
chrome test obviously generating far superior copper metallurgy than when grinding with
forged media despite the same flotation conditions.
Figure 3a-d: Photos oftbe first, second, third and forth concentrates ofa copper
roughing flotation test on Surveyor ore ground witb 30 per cent chrome media.
Plant tria)
From the positive results generated from the site laboratory testwork, Mount Gamet
personnel decided to undertake a plant trial, substituting 18 per cent chrome grinding media
into the SAG mill.
A series of pulp chemistry and EDT A surveys were conducted before and after changing the
media charge. Figure 4 shows the Eh throughout the primary grinding circuit was much more
oxidising than when measured previously while grinding with forged media. The pH (Figure
5) remained nominally the same as lime was added to a specific pH value. The dissolved
oxygen profile saw the most dramatic change, with the dissolved oxygen levels when
grinding with high chrome media up to over 3.5 parts per million in the cyclone overflow
(Figure 6). The higher levels of dissolved oxygen, along with more oxidising pulp potentials,
are strong evidence that the change in grinding media from forged to high chrome involves a
significant change in the corrosion mechanism. Further, as the grinding media was the only
parameter that was consciously altered, the observed changes can be attributed directly to the
high chrome grinding media.
Meto/lurgical Plant De.ign and Operating Strategies (MetPlant 2006)
18 _ 19 September 2006 Perth. WA
37
"
,.~!
:1
, '
The EDT A extractable iron profile (Figure 7) supports this, with the levels of oxidised iron
decreasing by almost a third in the cyclone overflow stream .
.. ._ .... ...~_.,.....~~.. ..~....
~,~"
r-
_,_._-----"-----------06-02-04
F orgad.l
...05-0lH)4 18% 9.
., ~
~i
oj;
Circuit Position
~
!
~!
:J
Figure 4: Eh profiles through the grinding and fiotation circuits oftbe Mount Garnet
concentrator, comparing forged grinding media to 18 per cent chrome grinding media.
12.0
11.0:
10.0 ~I
. I
9.0
X
Q.
8.0 '.
__
7.0 )
08-02-04
Forged
__ 05-0&-'l418% Cr
6.0
1
5.0 c
4.0
., ~
o'E
TI
o~'"
Figure 5: pH profileS through tbe grinding and flotation c:ircuits of the Mount Gamet
concentrator, comparing forged grinding media to 18 per cent cbrome grinding media.
38
8.0
II
7.0
1 I....
06-02-04 Forged
18% Cr
-ir- 05-06-04
16 oJ
:
11i
II
'
!'
5 oj
I oS
4,0 [
Ii ' \
8
!
i
II
;
3.0 ':
i
2.0;
1.0
I.,
_JI'
0.0
'!
~~
I,
l;l
~I
S~
(/) '6
d'~
Circuit position
~)
<3
I~
l~
e~'
eo!
~
,.';j
!g
e~i I
~ -
L.._~-------------------------__
-_--I
Figure 6: Dissolved oxygen profiles through the grinding and flotation circuits of the
Mount Garnet concentrator, comparing forged grinding media to 18 per cent chrome
grinding media.
0,35
T-. --------,-------
I .... 06-o2-<l4
0,30 j
i~
-e- 05-06-04
Ii
Forged \.
18% Cr
0,25-
i{f
Ii
III
0.20\
\ ~ 0,15'
i fa
;~
0',10
I
\.
0,05
0.00 ..--------------------
I
I
-9/.
1Il~
g~
~-8
l.l
.--_.
....- ....-,.. .
lij
~n
: (/)~
~
~~
11
.!!
l.l>~
0
il'E
~
:>
Circuit Position
l.l
!g
!g
.Cl
c:
J!
Q.
J!
'"c:
~
'iii
c:
u:
Figure 7: EDTA extractable iron profiles tbrougb tbe grinding and flotation circuits of
the Mount Garnet concentrator, comparing forged grinding media to 18 per cent
chrome grinding media.
39
Metallurgical performancc
. Monitoring the shift data continued for five months after the high chrome media was
substituted into the SAG mill. fu this time the process water was diluted with a higher ratio of
raw water in order to continue along the path of cleaning the system to promote better copper,
lead and zinc metallurgy. Therefore the following improvement in shift performance can not
be attributed solely to the grinding media, rather a combination of the high chrome media and
cleaner water, as well as any' other operational factors.
Figure 8 plots the copper circuit performance over seven months. The 'forged' period has
been taken from after the copper circuit collector was changed and before the 18 per cent
chrome was added to the SAG mill. The 'high chrome' period was taken from after the 1g per
cent chrome was added to the SAG mill and has been cleaned to remove periods such as
shutdowns and out of the ordinary reagent usage. The lead and zinc circuit performances
were treated in a similar fashion, but the data is not shown.
100.0
[-I
<)
Raw Wilder
50.0
Replaced
RT013 with
45.0
OSP009
I
"
90.0
18% C,
in SAG
.80.0
40.0
70,0
35.0
60.0 ~
o
I
\
... . ."
30.0
;;"
I\ t5.,
~; "" -._,
v.v".. \'1").~~
~
~ 25.0
\:1
20.0
_\
1. - -.-
50.0
40.0 .~
_.'1/
30.0
15.0
20.0
10.0
10.0
~
CD
.:2
0.0
0.0
.i,
l._
.,1
ecember,\ January,
2003
\ 2004
\FebrUary.\
2004
G Cu grade
~ Cu recovery]
Figure 8: Copper circuit performance for the Mount Garnet concentrator for tbe period
from December 2003 to July iOO4, showing the effect of cbanging grinding media from
,
forged to high chrome.
Clean data was extracted from the shift performance and comparisons made between the
periods when grinding with forged and high chrome media. Table 4 shows that there was a
noticeable difference in copper, lead and zinc perfonnance when the grinding media changed,
however other factors such as head grade and feed tonnage were a~so different.
40
Metallurgical Plant Design and Operali/lg Strategies (MetPlant 200~)
18.19 September 2006 Perth. WA .
---------~_
...........
__.
To compare the test condition (high chrome grinding media), a multiple variable regression
model can be used. If there is a significant difference between the two cases, then the process
can be said to have been altered by the test condition (Napier-Muon, 2005). Using the
cleaned data, tonnage and feed grade along with dummy variables of zero for forged and one
for high chrome were tabulated against recovery and modelled using the multiple linear least
squares regression model from Excel.
The confidence values show that in the copper, lead and zinc circuits, the grinding media type
is highly significant with respect to change in metallurgical performance. The standard errors
showed that there is a fair amount of scatter in the forged media data and it is obvious from
the time series chart in Figure 9 that the plant was noticeably more stable when high chrome
media was implemented.
The coefficients of the on/off variable give the difference in recovery between the on (high
chrome) and off (forged) cases. Table 5 lists these differences and the statistical confidence
that accompanies these figures.
As grinding media plant'trials occur over extended periods oftime to generate wear data, it is
very difficult to design the trial such that all other factors which potentially influence
recovery (eg tonnage, feed grade, mineralogy and operational changes) are removed.
Therefore although there has been a low level of statistical analysis completed on this data,
the calculated change in recoveries should be viewed with caution. Further, as the economic
analysis is based on the calculated recovery improvements, it follows that caution should be
maintained when studying the revenue increases.
41
Table 5: Difference between copper, lead and zinc circuit performance parameters
comparing forged to 18 per cent chrome media.
Confidence values
Difference (18% Cr
(%)
-Forged) ..
Parameter
>99.99%
+5.65 %
Recovery (%)
>99.99%
+4.07 %
Copper concentrate
Grade (%)
>99.99%
+5.37 %
Recovery (%)
99.89%
Lead concentrate
+2.16 %
Grade (%)
>99.99010
+3.60 %
Recovery (%)
99.84%
+0.97 %
Zinc concentrate
Grade (%)
Economic analysis
Increasing recovery at the same or better grade effectively means an increase in revenue
earned. Using an average throughput rate of 800 toones per day with an availability of 90 per
cent, the increase in concentrate production from the change to high chrome (including other
factors) can be calculated. Then, using the copper, lead and zinc prices from August 2004, a
dollar value equating to revenue increase can be demonstrated. Table 6 demonstrates that a
total of $A3.91 M in revenue has potentially been earned direc.tly from the improvements in
recovery. This does not take into account benefits in improved concentrate grade as the terms
of concentrate sale are confidential. However, it is known that a higher grade concentrate will
attract fewer penalties.
Table 6: Increase in revenue due to increased
Metal Price (LME
Metal
Copper
Lead
Zinc
27/0812004)
AUD $ 3990
AUD $ 1312
AUD $ 1377
Total
Apart from the metallurgical benefits documented in this case study, it should be pointed out
that there has been considerable saving through a reduction in the consumption of
consumables, i.e. grinding media and reagents (Table 7).
It is widely known that high chrome grinding media has superior wear rates to that of forged
media, which in some instances is sufficient to justify a change in media. The consumption of
grinding media at Mount Garnet dropped by approximately 68 per cent due to the change
from forged to 18 per cent chrome media in the SAG mill. This change was enough to offset
the higher cost of high chrome grinding media as well as increasing revenue (approx A$56
576 pa).
42
MetallurgIcal Plant DeSIgn and Operating Strategies (MetPlant 2()(J6)
18 _ 19 September 2006 Perth, WA
Table 7: Change in reagent usage for the Mount Garnet concentrator, comparing
forged to 18 per cent ch rome media, removing the errect of throughput variations.
Reagent
Change (%)
-68.17
Grinding media
Antiscalent
DSPOO9
Copper sulfate
Cyanide
Frother
Lime
5MBS
Xanthate
Zinc sulfate
..
-66.86
-66.48
.46.61
-48.65
+101.06
~68.80
-66.73
-59.38
-62.35
The usage rates of most of the .reagents also dropped by a remarkable amount, ranging from
46 to 68 per cent reductions. Only the frother saw an increase in use, which is likely due to
the addition of raw water to the circuit. The reduction in the collectors were expected, as the
collector-mineral reactions are electrochemical in nature and are therefore influenced by the
oxidation state of mmerals, galvanic reactions and dissolved ions in the pulp (Ekmekyi, et aI,
2005). Surface coatings of iron hydroxides as well as low dissolved oxygen levels as a result
of grinding with forged media creates a barrier to reagent attachment, so to reduce reagent
usage, cleaner surfaces and higher dissolved oxygen concentrations is necessary. This is
(;:
43
,---------~
..- ---------- --------------
1-'
Parameter
Savinw'revenue ($A
per annum)
$161 500
$ 56 576
1.05 %
Media
$2200000
$ 1 369116
25.37 %
Reagents
$74400000
$ 3971203
73.58 %
Metallurgy
Total
$76660000
$ 5 396 895
100%
CONCLUSIONS
The Mount Gamet copper-lead-zinc plant demonstrated poor metallurgical performance with
misreporting of the valuable minerals and poor selectivity for chalcopyrite against pyrite. The
pulp chemistry showed low pulp potentials and dissolved oxygen levels in most streams,
along with high EDT A extractable iron levels.
After laboratory tests indicated that metallurgical improvements could be found by changing
to a more inert grinding media, 18 per cent chrome media was loaded into the SAG mill for a
plant trial. The resulting changes included improved pulp chemistry with more oxidising Eh
and dissolved oxygen values, along with lower EDT A extractable iron percentages. The
metallurgical change!l were impressive, with increased grades and recoveries of copper, lead
and zinc with a corresponding reduction in grinding media consumption as well as reagent
dosage rates.
It is estimated that the change from forged to high chrome griJ)ding media contributed to an
increase in revenue of almost ASSA million per annum. Of this, metallurgical improvements
accounted for around 73 per cent of the increase. It must be noted that the improvements in
metallurgy were due to a variety of factors, including grinding media change, cleaner water,
and other operational factors.
This case study challenges conventional thinking which dictates that a change in grinding
media must be justified by a decrease in media consumption alone. The grinding circuit (and
the media used) is the primary conditioner and does have a significant impact on downstream
processing. So, it is imperative that these aspects be taken into account in any plant trial.
Further to this, studies investigating the most appropriate grinding media should be included
in the design and feasibility stage of all new ore body developments. The Mt Gamet ore
feasibility testwork did include grinding media effects and these did show a marked copper
circuit performance.
ACKNOWLEDGEMENTS
The authors would like to acknowledge
44
Metallurgical Plant Design and Operating Strategies (MetPlanJ 2006)
18. 19 September 2006I'erth. WA
(
REFERENCES
)'
"
Adam K., Natarajan K A and Iwasaki I (1984), Grinding media wear and its effect on the flotation of
, sulphide minerals, lnternational
spectroscopy
of oxidised pyrrhotite
surfaces,
Buckley A N and Woods R (1987), The surface oxidation of pyrite, Applied Surface Science, 27, pp
437 to 452,
Cheng X, Smith K A and Iwasaki I (1993), Electrochemistry
of chalcopyrite-pyrrhotite-mild
steel
interactions and its relevance to the flotation of complex sulphide ores, in the proceedings
Paul E Queneau International
Symposium:
of the
Volume I: Fundamental
Buswell
electrochemical
M A, Bradshaw
D J, Harris
P J (2005),
The value
and limitations
825 to 831,
Forssberg K S E, Subrahmanyam
of grinding
of
18, pp
method on
Plant
Design
- the primary
and Operating
conditioner,
Strategies
in the Proceedings
Conference,
pp
Greet C J, Kina1 J arid Steinier P (2005), Grinding media - Its effect on pulp chemistry
tictiori?, in the Proceedings of the Centenary of Flotation 2005 Symposium,
Australasian
of the
- Fact or
Isaacson A E (1989), Effect of sulphide minerals on ferrous alloy grinding media corrosion,
United
media on
45
.,.,"
Kocabag D and Smith M R (1985), The effect of grinding media and galvanic .interaetions upon the
flotation of sulphide minerals, in the Proceedings of Complex Sulphides - Processing
Concentrates
and By-Products
T J, (2005) An Introduction
Minerals Engineers
of Ores,
Statistics
and Experimental
Mineral Research
Design. for
Centre, University of
nd
Interactions
I3,PP 53 to 71.
with Metallurgical
team
95 to 109.
Yelloji Rao M K and Natarajan
Sulphide Minerals
K A, (1989(b))
1~112/2003.
Interactiqn
Grinding
Effect of Electrochemical
between
Flotation,
Interactions
Among
Journal
46
Metallurgical Plant Design and Operating Strategies (MetPlant 200~)
18 - J 9 September 2006 Perth, WA
Proceedings
Perth,. Wes~~rn.AtI$t.r~'l~
,',",i"
\.--
18 - 1'9,Seplember2G.G_~::~;"t
",'.
principal Sponsor
t8JlI.~
,A'4SENcto
"..- '~~~~t.m
Technology
Aj.\SCENTIS
.......,,~.nllsq
'1!l\()nl.com