LABORATORY FLOTATION RATE

TEST PROCEDURE FOR PGM, BASE

METAL SULPHIDE AND OXIDE ORES

FOR A MF1 RATE TEST

Eurus Mineral Consultants

Laboratory Float Procedure. Update 4 January 2012.
Page 1 of 21
 LABORATORY FLOTATION RATE TEST PROCEDURE FOR PGM, SULPHIDE
AND OXIDE ORES

APPLICATION
This test procedure applies to a single stage MF1 (i.e. primary Mill-Float) rate test. See

DEFINITIONS
A rate test is where a number of concentrates are collected over various time periods in
order to generate recovery-time, grade-time and mass-time curves. The data is used to
estimate flotation kinetics.

A batch test is when a single concentrate is collected for the full duration of the test.

OBJECTIVE
The purpose of this procedure is to generate information to achieve the following;
 Characterise the ore
 Determine metallurgical variability of samples, ore types and ore mixes from the
deposit
 Ascertain what metallurgical changes result in an improvement of
recovery/grade such as changes in grind, reagent type and/or addition, pH, pulp
density
 Generate data to determine flotation kinetics as a means of optimising flotation
plant performance

CHARACTERISTICS OF PGM, SULPHIDE AND OXIDE ORES

Once reagents are added to a PGM and sulphide ore the ore responds to the flotation
process for the duration of the test to produce recovery/grade/mass/time relationships.
Staged addition of reagents can be tested as an option as opposed to being a necessity.
For oxide ores the addition of a sulphidising reagent is a necessity. However, the activity
of the sulphidiser tends to be short-lived and after dosing, flotation may only proceed for
a relatively short time. Staged addition of sulphidiser and collector may therefore be
required at intervals during the batch/rate test. In this case, the air is turned off, reagents
added and conditioned, air tuned on and the test continued until the next stage addition.

Eurus Mineral Consultants

Laboratory Float Procedure. Update 4 January 2012.
Page 2 of 21
 3

EQUIPMENT
There are a number of standard laboratory flotation machines and cell sizes. The Denver
D1 and D12 machines are the most commonly used (Figure 1) and come with a variety of
cell and rotor/stator sizes from 1.5 to 8.0 l. (Figure 2). Tests using the 8.0 l cell require a
larger rotor/stator. The LabTech Essa machine with 2.5 l cell is shown in Figure 3 and
Figure 4 (note that annotations on these pictures are equally applicable to the Denver
machine).

Figure 1

Eurus Mineral Consultants

Laboratory Float Procedure. Update 4 Jan 2012.
Page 3 of 21
 4

Figure 2

Eurus Mineral Consultants

Laboratory Float Procedure. Update 4 Jan 2012.
Page 4 of 21
 A

Figure 3

Level of pulp

10-15 mm

Figure 4
Eurus Mineral Consultants
Laboratory Float Procedure. Update 4 January 2012.
Page 5 of 21
 6

SET-UP
Agitation and Air
Usual set-up conditions for 1.5 to 8.0 l cells are shown in Table 1. Cell air flowrates and
cell rpms are plotted in Figure 5. These running conditions are guidelines and are not
“cast in stone”. Cell rpm is chosen, usually on a visual basis, to provide adequate
agitation and suspension of solids but should not be too intense to cause spillage. The
intensity of agitation reduces once air is introduced into the system.

Cell volume and dimensions in Table 1 are detailed in Figure 6. With agitation and
airflow, pulp level should be about 10-15mm below the froth overflow lip. As per the
illustration in Figure 6, this measurement is approximately the length of a thumbnail (the
proverbial “rule of thumb”!). The froth paddles are sized to extend 10mm below the
overflow lip. When air is turned on, pulp level rises and it should rise to 10-15mm below
the froth overflow lip as represented in Figure 4. The cell should be marked and
calibrated for pulp level with agitation and no air so that with airflow, pulp level is 10-
15mm below the froth overflow lip. Froth height should be level with the overflow lip.

Guidelines for air addition are given in Table 1. During a test-run an appropriate air set-
point was established at about 50% on the rotameter indicated by the line and arrow in
Figure 3. Pulp height was about 15mm below the overflow lip (see Figure 4).

Sample Mass
The Denver 2.5 l cell has a live volume of 2.4 l. This is sufficient for 1,0 kg of sample and
gives 32% solids in the cell which is a typical float plant feed density. In this case the
laboratory mill should be set-up for 1.0 kg of sample as it is practical and best to mill the
same mass of sample that will be used in the subsequent flotation test.

The LabTech Essa flotation cell has a live volume of 2.0 l. To float at between 30 and 35%
solids Table 2 below shows that 800 gr of solids meets this requirement and allows about
1.25 l of water to be used to wash the solids from mill to float cell.

From a visual analysis using the 2.5 l cell, the Essa flotation machine provided adequate
agitation at 1200rpm. This is the maximum that the machine can operate at. Usual rotor
speeds for the Denver cell are shown in Table 1.
Eurus Mineral Consultants
Laboratory Float Procedure. Update 4 Jan 2012.
Page 6 of 21
 Measurements for the Denver D12 Laboratory Machine and
Cells
1.5 litre Cell 2.5 litre cell 5.0 litre cell 8.0 litre cell
A mm 115 135 155 195
B mm 115 135 155 195
C mm 165 205 240 295
D mm 120 160 215 260

Nominal Volume (l) ) [A*B*D] 1.59 2.92 5.17 9.89

Active Volume to 15mm below

overflow lip less 10% for 1.25 2.38 4.32 8.38
Mechanism (l) ) [A*B*(D-15)]
Usual Airflow (l/min) 6.0 7.5 9.2 11.0
Airflow per Active Volume
4.80 3.15 2.13 1.31
(min^-1)
Airflow per Nominal Volume
3.78 2.57 1.78 1.11
(min^-1)
Usual Cell rpm 1200 1500 1650 1800

Table 1

FLOAT CELL
Ore SG (t/m³) 2.7 Choice Choice
Lab Cell
2.0 2.0 2.0 2.0 2.5 2.5 2.5
volume (l)
% solids 30.0 35.0 38.0 32.0 30.0 35.0 32.0
Pulp RD 1.233 1.283 1.315 1.252 1.233 1.283 1.252
Pulp (kg) 2.47 2.57 2.63 2.50 3.08 3.21 3.13
Solids (kg) 0.740 0.898 1.000 0.800 0.925 1.122 1.000
Water (l) 1.73 1.67 1.63 1.70 2.16 2.08 2.13
MILL
Mill % Solids 65.0
Pulp RD 1.693 Choice Choice
Pulp (kg) 1.138 1.381 1.538 1.231 1.423 1.727 1.538
Solids (kg) 0.740 0.898 1.000 0.800 0.925 1.122 1.000
Water (l) 0.40 0.48 0.54 0.43 0.50 0.60 0.54

WATER AVAILABLE TO WASH FROM MILL INTO

CELL
Water (l) 1.33 1.18 1.09 1.27 1.66 1.48 1.59

Table 2

Eurus Mineral Consultants

Laboratory Float Procedure. Update 4 January 2012.
Page 7 of 21
 8

Denver D12 Laboratory Flotation Machine

Cells Airflow per Active Volume Airflow per Nominal Volume RPM
13.0 2000

12.0
1800
11.0
1600
10.0
Airflow per Active Volume (min^-1)
Usual Air Flowrate (l/min) &

9.0 1400

8.0 1200

Cell rpm
7.0
1000
6.0

5.0 800

4.0 600

3.0
400
2.0
200
1.0

0.0 0
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5
Cell Volume (l)

Figure 5

15mm

Level of
pulp
D

A
B

Figure 6

Eurus Mineral Consultants

Laboratory Float Procedure. Update 4 Jan 2012.
Page 8 of 21
 9

Froth Paddles
Froth paddles should be sized and shaped as per the illustration in Figure 7. The paddles
should be half the width of the cell, extend 10mm below the froth overflow lip and have
handles. When collecting froth the paddle should sit on the cell wall so that froth
collection depth is constant. The paddle can then slide along the cell wall from back to
front ie “A” to “B” in Figure 11.

Mechanism
Froth Paddle

Froth and froth height

Overflow Lip
A B
Pulp Level

A: 10mm below
overflow lip

B: 15-25mm
below overflow lip

Figure 7

Determining Froth Collection Times

There are four important issues for both assayed elements and the mass/gangue
component. The number and timing of concentrates collection should be arranged so
that;
1. The shape of the recovery-time curve can be accurately described mathematically
2. The fast floating fraction collected at the start of the test is adequately represented
and measured (see Figure 9)
3. The slow floating fraction collected towards the end of the test is adequately
represented and measured (see Figure 9)
4. Sufficient sample is generated so that all analytes can be assayed. If this is not
possible then duplicate/triplicate tests should be done.

Eurus Mineral Consultants

Laboratory Float Procedure. Update 4 Jan 2012.
Page 9 of 21
 10

In points 3 and 4 above, flotation concentrates should be timed so that the shape and
position of the fast and slow floating sections of the curve can be well delineated and the
slow floating section should be approaching or have approached a plateau, ie the
recovery-time curve should have two distinct components with a discernable inflection
point as per the bottom curve in Figure 9. If the recovery-time curve trend is described by
a straight line as shown by the mass curve (red line, triangle markers) in Figure 10 then
the test should be redone for a longer time to generate a curve with two distinct fast and
slow sections. An example of this is shown in Figure 10 where the original test had to be
redone for an extended flotation time from 23 to 40 minutes.

From a cost and sample processing point of view the minimum number of concentrates to
adequately describe the recovery-time curve and the ore’s flotation characteristic is four
timed at 1, 3, 10 and 20 minutes. The first concentrate has the smallest mass and if this is a
problem for assay, then the first two concentrate collection times can be moved to 2 and 4
minutes without adversely affecting the accuracy of estimating fast floating kinetics.

The table below recommends concentrate collection times for three categories of ore
according to how fast mineral and gangue float and how quickly the recovery-time curve
forms a plateau. The choice of collection times should suit the slowest floating
component which is usually mass. For example, a 20 minute float would suit the ores in
Figure 9 and the top curve in Figure 10, but only a 40 minute float would be suitable for
an ore with a mass-time curve in Figure 8 and the bottom curve of Figure 10.
A B C
Slow Floating Ores
Average to Fast Average to Slow
and/or linear mass-
Floating Ores Floating Ores
time ores
Time Time Time
Conc Conc Conc
(min) (min) (min)
1 1 1 1 2 1
3 2 3 2 5 2
7 3 15 3 20 3
20 4 30 4 40 4
Alternately: Alternately:
Time Time
Conc Conc
(min) (min)
1 1 2 1
3 2 5 2
7 3 10 3
20 4 25 4
30 5 40 5

Eurus Mineral Consultants

Laboratory Float Procedure. Update 4 Jan 2012.
Page 10 of 21
 11

The following excerpt from “Using Simulation to Understand Metal and Mineral
Flotation Performance at one of Falconbridge’s Base Metal Operations”; M. P. Hay et al,
Nickel 05 Conference, Cape Town serves to outline the importance of not generating a
linear recovery-time curve.

Determination of Flotation Kinetics

Fitting Kelsall’s parameters to a normal recovery and grade profile such as for Copper was
achieved with a high level of accuracy. In contrast, the cumulative mass-time profile in Figure
8 posed quite a problem because it was essentially linear with a correlation coefficient, r2, of
0.976. It is not possible to fit a two component equation to a linear or near-linear relationship.
In cases such as this a solution is often generated where both fast and slow floating rates have
equal values. This is mathematically correct but does not occur in practice. When both rates
are equal the fast floating fraction can be assigned any value between 0.0 and 1.0 without
making any difference to the resulting mass recovery-time profile and these kinetic data are
useless for simulation. For example the two laboratory test profiles in Figure 8 can be
described by the two-component Kelsall equation;

Cumulative mass recovery = (100 - θ) [1 – exp(-kGF* t)] + θ[1 - exp(-kGS* t)]

where kGF=kGS=0.0160 with θ being any value – as described by the dashed line.

FALCONBRIDGE BMS ORE, LABORATORY ROUGHER RATE TESTS: MASS

Lab Test 1 Lab Test PT6 kGF=kGS=0.0160

35
% ROUGHER CONC MASS RECOVERY

30

25

20

15

10

0
0 2 4 6 8 10 12 14 16 18 20 22 24 26
TIME (min)

Figure 8 Laboratory Test Data: Mass

Eurus Mineral Consultants

Laboratory Float Procedure. Update 4 Jan 2012.
Page 11 of 21
 12

100

95
Slow Floating
90

85
LABORATORY RECOVERY

80

75
Fast Floating
Slow
70

65

60

55

50
Inflection point
45
Fast
40

35

30
0 2 4 6 8 10 12 14 16 18 20
FLOTATION ROUGHER RATE TIME (min)

Figure 9

FB BMS ORE, ROUGHER RATE TEST: COPPER AND MASS

Av of Lab Ro Rate +PT6 tests: Cu ReTest of Lab Ro Rate test: Cu

Av of Lab Ro Rate + PT6 tests: Mass ReTest of Lab Ro Rate test: Mass
100
95
COPPER
90
85
80
75
% ROUGHER CONC RECOVERY

70
65
60
55
50
45
40
35 MASS
30
25
20
15
10
5
0
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42
TIME (min)

Figure 10

Eurus Mineral Consultants

Laboratory Float Procedure. Update 4 Jan 2012.
Page 12 of 21
DETAILS OF LABORATORY FLOTATION PROCEDURE
Sample
There are two types of test designated by sample source;
1. Sample that has been milled in a laboratory mill, referred to as a “lab” test or
“lab” float and
2. Sample that has been taken from a flotation circuit, either pilot or plant scale,
referred to as a “hot” float.

Mixing and Reagent Addition

 Air should be off
 Set cell rpm
 Pulp level should be to the calibration mark
 If pH must be modified or reagents added, do so in the desired order and
addition rate.

Air Rate
LabTech Essa
The test is begun by opening the air. The aim throughout the test is to maintain the top of
the froth bed level with the overflow lip of the cell. The set point for air flow has been
marked at about 50% on the rotameter (indicated by the dark line and white arrow in
Figure 3). This setting is suitable for a 2.0/2.5 l cell. If a smaller or larger cell is used then
air rate will need to be adjusted as per Table 1. Air flow should not be increased greater
than the set point because an unstable situation can be created where air begins to blast
through the froth.

Air rate is adjusted by the knob marked “A” and flow is turned on/off by the switch
marked “B”. The ideal is to have previously set knob “A” and on starting the test, switch
“B” is then turned to “on” so that the test begins with the set point flow of air. This may
not always be possible depending on the floatability of the ore. If this is high, concentrate
production will be voluminous and a runaway froth will occur which will be
uncontrollable (ie concentrate overflow without having to use the paddles). For very
floatable ores it is best to begin the test by slowly opening the air and gradually building a

Eurus Mineral Consultants

Laboratory Float Procedure. Update 4 January 2012.
Page 13 of 21
 14

froth bed until it becomes level with the overflow lip. As soon as this is done timed
concentrate collection can begin.

What is required is that the froth is controlled at all times and is kept level with the
overflow lip of the cell. Thus at the start of the test, air valve “A” will need to be adjusted
and then gradually opened to the set point. After this, the level of pulp is topped up with
water to maintain froth level. All the time, froth is collected by paddle and the quantity of
froth removed is always controlled by the paddle design and the frequency of collection.

Denver
If the Denver machine is fitted with an air supply and rotameter then the above
description for the LabTech Essa machine applies. If this is not fitted, air is controlled by
the small “tap” shaped valve at the top of the stainless steel shaft. The same control is
required via this valve for very floatable and voluminous froths.

Froth Collection
Froth is collected every 15 seconds and consists of two sweeps of the paddles one
immediately after the other from the back of the cell to the front (“A” to “B”, Figure 11).
As the paddles moved forward they are angled around the mechanism and brought back
together again before moving to the front of the cell (Figure 12). At all times the paddles
must be operated in unison as a pair so that froth is removed in a uniform manner across
the full width of the cell.

 There are four collections of concentrate in every minute,

 Collection begins on the minute and is repeated every 15 seconds,
 Collection finishes at the 45 second mark.

An example of a 4 and 7 concentrate collection test is shown in Table 3. The timing and
sequence of collections is illustrated in Table 4. Note that the test, test conditions and
pulp level are set-up followed by air rate so that froth is level with the overflow lip. As
soon as this is done the test begins and froth collection can begin immediately (ie at 5
seconds), followed by a second collection at 15 seconds. Thereafter collection is at every
15 second interval.

Eurus Mineral Consultants

Laboratory Float Procedure. Update 4 Jan 2012.
Page 14 of 21
 Back of Cell
A
Mechanism

B
Overflow Lip
Front of Cell

Figure 11

Back of Cell

B
Front of Cell

Figure 12

Eurus Mineral Consultants

Laboratory Float Procedure. Update 4 January 2012.
Page 15 of 21
 16

If 4 Concentrates are Collected

Concentrate Collected between….
1 0 min and 45 sec
2 1 min and 2m 45s
3 3 min and 9m 45s
4 10 min and 30 min

Similarly with 7 Concentrates

Concentrate Collected between….
1 0 min and 45 sec
2 1 min and 1m 45s
3 2 min and 2m 45s
4 3 min and 5m 45s
5 6 min and 11m 45s
6 12 min and 19m 45s
7 20 min and 30 min

Table 3

Min Secs
Before timing, set-up test, air
rate and froth level
0 0 Begin test
0 5 Two sweep froth collection
0 15 Two sweep froth collection
Conc 1
0 30 Two sweep froth collection
0 45 Two sweep froth collection
1 00 Two sweep froth collection
1 15 Two sweep froth collection
1 30 Two sweep froth collection
1 45 Two sweep froth collection
Conc 2
2 00 Two sweep froth collection
2 15 Two sweep froth collection
2 30 Two sweep froth collection
2 45 Two sweep froth collection
3 00 Two sweep froth collection
3 15 Two sweep froth collection
Conc 3
3 30 Two sweep froth collection
etc to end of test

Table 4

Eurus Mineral Consultants

Laboratory Float Procedure. Update 4 Jan 2012.
Page 16 of 21
Washing Down and Water Addition
After the second sweep of each collection any material adhering to the paddles should be
washed off into the concentrate collection dish. Occasionally, wash down adhered
material around the sides of the cell and mechanism.

When air rate is at the set point and froth level begins to fall below the level of the
overflow lip, pulp level should be topped with water.

Thin Froth towards the end of a Test

One of the main requirements of a flotation test is to maintain a decent froth structure and
level. If the froth becomes very thin, turn the air off, put the stopwatch on pause, add one
drop of frother, condition for 1 minute and then restart the test, stopwatch and froth
collection.

Treatment of Products
All concentrate and tailings are weighed (dish+pulp), then filtered, dried, dry weight
recorded and sent for assay. This records dry mass and contained water for each product.
If the sample contains sulphur/sulphides drying must be done in a low temperature
oven (40C) to avoid burning off any sulphur/sulphides.

Immediately after the Test

Remove the mechanism from the pulp and record the volume of pulp remaining in the
cell. Thoroughly wash down the outside and inside of the mechanism into the cell.

Particle Size Analysis

Apart from a means of concentrating economic elements and metals, the flotation process
is also a size classifier. Fines are preferentially recovered into the concentrate from the
feed stream with an upper d90 size limit of approximately 125-150µm depending on the
floatability of the ore type being treated. For example feed at 40% -75µm produces
concentrate at about 60-70% -75µm and feed at 70-80% -75µm produces concentrate at
about 90-95% -75µm. The particle size distribution across a flotation bank or circuit can be
duplicated by using a hydrocyclone model where flotation concentrate is represented by
cyclone overflow and flotation tailings by cyclone underflow.

Eurus Mineral Consultants

Laboratory Float Procedure. Update 4 January 2012.
Page 17 of 21
 18

A few batch tests should be done to determine this particle size relationship for the
various ore types and ore mixes. This is essential data when checking the grind of the
laboratory mill. After a test, a representative sample of final tailings can be sent for assay
and a further representative sample subjected to screening (the balance of the tailings
material can be stored in case a re-assay is necessary). The particle size distribution and %
-75µm of float feed is then determined by combining the mass and grading of tailings and
concentrate streams.

REAGENT ADDITION
Reagents should be fresh and made-up on the day. Large additions of dilute reagent such
as depressant should be added by pipette or a large barrel syringe.

In cases where it is necessary to add a reagent neat – such as a frother – the required
addition can be just one drop. Small additions of reagent need to be added using a small
barrel medical syringe with a thin hypodermic needle. One drop has a volume of
0.02cm3. In a test using 1kg in a 2.5l cell this is equivalent to 20g/t which is a sizeable
dose in a plant.

CALIBRATION, ORGANISATION AND PRACTICE

It is essential to calibrate the flotation cell and machine, practice test technique and
develop a system;
 Sufficient pipettes and/or syringe/needles should be available
 A supply of water (tap or plant) should be readily available at arms length
 A number of wash bottles which can be easily refilled in between concentrate
collection sweeps without having to move away from the cell and without
missing a concentrate collection period
 Concentrate dishes should be large enough to contain what can be a large pulp
volume, especially for the last concentrate. The containers should be labelled
before the test begins
 Practice within 15 seconds:
o Conducting two concentrate collection sweeps,
o The dexterity of holding both paddles in one hand and washing them
both into the concentrate container with a wash bottle in the other hand,

Eurus Mineral Consultants

Laboratory Float Procedure. Update 4 Jan 2012.
Page 18 of 21
 19

o Washing material on the sides of the cell and mechanism back into the cell
and
o Checking and topping up pulp level.

TEST TECHNIQUE
The set-up and execution of a rate test needs to be more organised. Remember what I
said when we did some practice tests during my visit on 24th-27th October. Conducting a
rate test is an intensive activity where the operator is continuously on the go for the
duration of the test. In a period of 15 seconds the operator needs to do some or all of,
a. Two collections of froth from back to front of the cell,
b. Wash down the paddles into the concentrate dish,
c. If needs be wash down froth adhered to the insides of the cell and around the
shaft,
d. Check pulp level,
e. Top up pulp level with water,
f. Add reagents if stage adding say collector,
g. If needs be add extra frother,
h. Change concentrate dishes and
i. Check the time.

To do all this four times a minute for a 30 minute test requires much practice in order for
the test to run smoothly. The intensity of the test requires everything to be at hand with
the items logically arranged around the cell.

For example once the test has started there is no time to fill up a wash bottle so there must
be 4 full ones at the start of the test. Label the concentrate collection dishes beforehand, as
it is easy to forget which is which if they are moved around after the test. Practice adding
one drop of reagent from a syringe. This is not easy to do within the time schedule of a-i
above. It is easy to press the plunger too hard and a stream of reagent gets added at
which point the test is ruined. Thus, if the froth is thin and it is necessary to add more
frother, one (maximum two) drop(s) will suffice. Getting this wrong and adding a stream
of frother might add (say) 120g/t in which case the froth will be voluminous, mass pull
will be unrealistically high and the test must be abandoned.

Eurus Mineral Consultants

Laboratory Float Procedure. Update 4 Jan 2012.
Page 19 of 21
 20

A supply of syringes and small needles is important. Each should be labelled and used
for a certain reagent. Each should be thorough washed at the end of the day and kept
overnight in a beaker of water.

Suggested Lay-out of the Bench for a Rate Test

Figure 13 shows a suggested arrangement of equipment around the float cell so that
everything that is needed is easily at hand. The flow of work – i.e. the movement of
concentrate collection dishes 1-4 – is assumed to be right to left. Adequate space should
be available on both sides of the float cell so that the dishes can be moved rapidly right to
left during the test.

Eurus Mineral Consultants

Laboratory Float Procedure. Update 4 Jan 2012.
Page 20 of 21
 pH meter, acid
and alkali for
automatic pH Clock/stop watch
adjustment capable of recording
Bucket of make- minutes and seconds
up water LAB CELL

Wash bottles

Beaker to add water to cell

Concentrate
Syringes of
collection dish 3
reagents

Concentrate
collection dish 2

Room for more Concentrate Concentrate

dishes collection dish 1 collection dish 4

Clipboard to
make notes of
froth condition
etc

Edge of bench

Figure 13

Eurus Mineral Consultants

Laboratory Float Procedure. Update 4 January 2012.
Page 21 of 21