Note : Les descriptions sont présentées dans la langue officielle dans laquelle elles ont été soumises.
`~3~3~3
1 PC-2174
FLOTATION SEPARATION OF PENTLANDITE FROM
PYRRHOTITE USING SULFUR DIOXIDE-AIR CONDITIONING
.
TECHNICAL FIFED
The instant invention relates to beneficiati.on of complex sulfide
ores in general and, more particularly, to a process that preferentially
renders pentlandite unfloatable while leaving chalcopyrite and pyrrhotite
floatable for subsequent separation.
BACKGROUND ART
In the Sudbury district of Canada, as well as in other parts of
the world, nickel is found in a complex, finely disseminated ore with
other valuable metals including copper. The principal sulfide
mlnerali~ation consists of pentlandite (No, Fox, chalcopyrite (Quaffs)
and pyrrhotite (Fun US ) with the undesirable pyrrhotite (which itself
contains only a minor portion of nickel in solid solution being present
in amounts far greater than the pentlandite.
I
2 PC-2174
For example, the ratio of pyrrhotite to pentlandlte may be
approximately 5:1. The ores and the separation thereof have been under
study for many years and much has been discovered as a result.
The processes which have been adopted for beneflciation of these
complex ores involve the separation of the values into a nickel stream, a
copper stream, a pyrrhotite stream and reject guying or rock stream, are
essentially compromises in which overall recovery of desired metal values
and degree of concentration are balanced. Thus, in flotation of the ores
the practice of forming a bulk nickel-copper concentrate with rejection
of pyrrhotite and guying followed by further flotation to provide
separation of nickel and copper has been adopted. In effecting
nickel-copper separation, chalcopyrite is floated preferentially to
pentlandite. It is necessary to float a large proportion of pyrrhotite
to recover slow floating fine pentlandite and middling particles. Thus,
substantial quantities of pyrrhotite remain with the nickel concentrate
with the result that the nickel concentrate analyzes approximately lo% of
nickel. In contrast, the copper concentrate resulting will contain
approximately 30% of copper. The nickel concentrate still contains
copper and the copper concentrate still contains nickel, factors which
co~pllcate further processing in eke smelter. The pyrrhotite concentrate
rejected from the circuit should be as low as possible in nickel and
copper.
The fact that, in the interest of maximizing nickel (or copper
recovery, the nickel concentrate presently obtained it of relatively low
grade, means that the total content of sulfur in the nickel concentrate
fed to the smelter is higher than is wanted in terms of operating cost,
nickel throughput, losses of nickel, losses of cobalt and emissions of
sulfur dioxide. For example, raising concentrate grade from about 11% to
about 17% nickel would cut sulfur dioxide emissions approximately 25%, a
highly desirable result, and other economies can be achieved. While
provision of means for treating nickel sulfide ore to provide a nickel
concentrate of improved grade has long been recognized as an objective,
no practical means for doing so without encountering severe economic
handicaps ha heretofore been developed.
3 PC-2174
Coupled with the above, it has become increasingly important to
reduce eke quantity of sulfur dioxide (S02) emissions. Sulfur dioxide it
an undesirable byproduct of sulfide ore refining. Current gas treatment
and sulfuric acid production are exceedingly expensive. Accordingly
emphasis has been placed on removing as much pyrrhotite as possible
before the treated ore is subjected to pyrometallurgical techniques.
Obviously, the more sulfur one can remove during the initial stages of
ore processing, the less sulfur dioxide one will have to contend with
during the final refining stages.
Flotation is one of the most useful mineral dressing techniques
available to the engineer for concentrating valuable minerals contained
in ores. Many specific flotation techniques have been developed for the
treatment of specific ores and a wide selection of chemical agents has
been employed to provide a variety of beneficial results in applying such
techniques.
However, as far as can be determined, there is no commercial
flotation process that separates pentlandite from pyrrhotite. The only
other flotation process that succeeds in doing this is the subject of
Canadian patent 1,156,380 but that process has not been commercialized.
This process uses sodium carbonate and cyanide jointly which leaves
pentlandlte and chalcopyrite floatable while depressing pyrrhotite.
Unfortunately, this process consumes a great quantity of cyanide and leaves
stable complex metal cyanides in the effluent. Other flotation methods
currently in use are at best modestly effective. For instance in what is
referred to by some as the pyrrhotite rejection circuit in Into Limlted's
Copper Cliff mill (located in Sudbury, Ontario) recovers only 60% of the
nickel as jell as 25% of the pyrrhotite.
SUMMARY OF TIE INVENTION
Accordingly, there is provided a flotation process for
expeditiously rejecting pyrrhotite while recovering most of the
pentlandite and chalcopyrlte from a mixture containing these sulfides and
associated silicate rode minerals.
~V~3~
61790-1568
The process conditions the feed, removes rock, subjects
the material to a sulfur dioxide/air stream, regulates the pi of
the flow stream to first separate the pentlandite from the stream
then floating the chalcopyrite, and leaving the pyrrhotite behind.
Thus, the invention provides a flotation process for
selectively beneficiating sulfide ore, the ore including, amongst
other things, pentlandite, chalcopyrite and pyrrhotite, the
process comprising slurring the ore, introducing a reagent to the
slurry to resorb xanthate present in the slurry, removing the
lo xanthate and liquid, adding sufficient xanthate to the ore to allow
-the sulfides to float to form a bulk sulfide concentrate while
leaving the rock as a non-float stream, diluting the bulk sulfide
concentrate to a second slurry, adjusting the pi level of the
second slurry, adding sulfur dioxide and air, and floating the
second slurry to recover chalcopyrite and pyrrhotite as a float
product while depressing the pentlandite.
BRIEF DESCRIPTION OF THE DRAWING
Figure 1 is a generalized flow sheet of the invention.
Figure 2 is a modification of the invention.
Figure 3 is a graph depicting separation efficiency.
PREFERRED MODE FOR CARRYING OUT THE INVENTION
figure 1 represents the flowchart of the process.
The feed in the following discussion is low grade
material obtained from Into Limited's Copper Cliff mill. The
stream is generated after removing the readily floatable
chalcopyrite and pentlandite and rejecting by magnetic separation
most of the monoclinic pyrrhotite. It should be appreciated,
however, that the disclosed process may be utilized with the
,,
. ,~,...
Jo J - 4 -
I
61790-1568
appropriate materials regardless of the source.
The feed in this instance analyzes about 3% chalcopyrite
(Quaffs) 5% pentlanclite (Nix foe 2S8), 40% pyrrhotite (Phase),
and 50% rock. The objective is to recover the chalcopyrite and
pentlandite as separate concentrates while rejecting all of the
pyrrhotite and rock.
Simply for ease of discussion, the ensuing treatment
applies to a one kilogram sample of feed. Larger quantities are
to be treated proportionally. In any event, the feed was slurries
to two liters of pulp.
The process begins with an operation that has been dubbed
a sulfide wash. The purpose of this step is to provide a means
for removing excess xanthate that may accompany the feed material
as a consequence of previous processing. Following addition of a
sulfide bearing reagent (sodium sulfide, sodium hydrosulfide,
ammonium sulfide, or ammonium hydrosulfide, for example) the pulp
is conditioned for a few minutes and is then -thickened during which
time the pulp is raked
Jo
- pa -
~.3~3~
5 PC-2174
continuously. Indeed, the sulfide reagent may be selected from the
sulfides or hydrosulfides of the alkali metals or alkaline earth metals.
It is also within the purview of this invention to utile a lime wash
instead of the sulfide wash.
0.5 g/kg of sodium sulfide was added, after which the pulp was
conditioned for five minutes. The purpose of the raking is to ensure
that the pulp and liquor are mixed 80 that xanthate resorbed from the
sulfides is distributed uniformly throughout the liquor. It is known
that conditions necessary for xanthate resorption are a highly negative
REDO potential and oxygen deficient conditions. Following thickening
for an extended period (17 hours) the liquor is decanted and discarded in
order to remove the resorbed xanthate.
The thickened solids are then ground at 50% solids by weight to
produce a size distribution with 35~ retained on 37 em. Following
grinding an addition of fresh xanthate is made (usually 0.07 go of
feed) and the sulfides are floated away from the rock. The flotation
time was sixteen minutes.
The bulk sulfide concentrate produced in the foregoing step will
contain virtually all of the chalcopyrite, pentlandite and pyrrhotite
that was in the feed. Typically about 40% ox the original weight reports
to the rock tails which assay about 0.1~ Cut 0.25% No and contain
approximately 4% each of the copper and eke nickel.
The bulk sulfide concentrate diluted to 1.7Q of slurry, it heated
to 40C, the pi is downwardly adjusted to 8.1 if necessary, then sulfur
25 dioxide is added at a rate of 9 mQ/min for 50 minutes. The total S02
addition is therefore 450 my or 1.2 g. (1.2 grams/1.7 liter or 0.71
grams liter Air is added at a rate of 590 ml per minute simultaneously
during the S02 sparring to insure that the dissolved oxygen level it
maintained above about 5.4 ppm. During the addition of sulfur dioxide
and air the pulp is stirred vigorously to generate a vortex which
prevents foam from accumulating on the surface of the pulp. After 50
minutes of sulfur dioxide addition the gas flow is stopped and sufficient
lime is added to the pulp while at 40C to produce a pit of 11.5. The
pulp is conditioned for an additional 20 minutes with vigorous stirring
so as to again generate a vortex. Air was added again to maintain the
dissolved oxygen above about 5.4 ppm.
I
S PC-2174
Following this S02/air/lime conditioning the bulk sulfide
concentrate is subjected to flotation for about ten minutes.
Chalcopyrite and pyrrhotite are floated while pentlandite is depressed.
In a good test 95% of the chalcopgrlte and more than 90% of the
pyrrhotite will be in the float product lath less than 25% of the
pentlandite. The floated material is returned to the cell for a cleaning
flotation stage Lime is again added to produce a pi of 11.5 then the
pulp is subjected to flotation at ambient temperature
The non-float portion from the first and second (cleaner)
flotation stages which it identified as the pyrrhotite cleaner tails) is
combined as the nickel concentrate. The material floated in the cleaner
stage contains virtually all of the chalcopyrite and most of the
pyrrhotite.
To effect a chalcopyrite/pgrrhotite separation the floated
material is conditioned in a saturated lime solution (conditioned with
1.6 grams of lime), adding a small amount (0.1 gram) of cyanide then
conditioning with aeration until the REDO potential (A versus saturated
calmly) rises to a least -225 my then floating the chalcopyrite over a
four minute period. The non-float portion is the final pyrrhotite
20 concentrate. Typically it analyzes about 90% pyrrhotite, 1.4% No (1~5%
Nope and contains approximately 75% of the pyrrhotite along with 10% of
the pentlandite.
The copper concentrate produced from this single flotation stage
analyzed about 16% copper and contained By% of the copper in the feed
25 It also analyzed 1.2% No, 31~ pyrrhotlte with 2.7% of the nickel and 4.4
of the pyrrhotite.
More particularly, the results are shown below:
I
7 PC-2174
TABLE I
PENTLANDITE/PYRRHOTITE SEPARATION RESULTS
ASSAY (%) DISTRIBUTION (%)
Cut No S Pro Wit Cut No Pun Pro
Product
Cut Kink 1.230.331.0 5.980.3 2.7 2.4 4.4
Pro Kink 1.335.987.8 35.29.4 17.4 9.1 74.6
No Kink 12.022.227.3 16.516.S 75.4 84.8 10.9
Rig Tails 0.1 0.28 4.2 9.9
10 FEED 1.2 2.619.941.4
Ok = Rock
Pro = Pyrrhotite
Cone. = Concentrate
Pun = Pentlandlte
The following discussion relates to the particular parameters
selected.
A. pi and Conditioning Time
The effects of final pi and conditioning time during the addition
of S2 and air is shown in Figure 3. In these tests the procedure used
was that shown in Figure 2. This procedure differs from the previous on
in several aspects some of which are dependent on the sample history. No
sulfide Nash (eke thickening stage described in the previous procedure)
was done on this sample because it did not have an excess of xanthate.
Figure 3 is relevant, however, to the previous procedure
Figure 3 shows the pentlandite/pyrrhotlte separation efficiency
(SUE.) as a function of the conditioning time and final phi In this
series of tests the S02 was added in a 1.5 White mixture with air. The
actual S02 addition rate was 20 mQ/min until the pi reached the desired
value at which time the gas addition rate was reduced so that the pi was
maintained constant until the total conditioning time had elapsed. The
entire conditioning was conducted at room temperature ( 25C).
The separation efficiency is a figure of merit used to evaluate
the effectiveness of a physical separation process. It is simply
calculated by taking the difference in flotation of pentlandite and
pyrrhotite. In this work the fraction floated is based on the bulk
sulfide concentrate. The loss of pentlandite and pyrrhotite in the rock
3~3
8 PC-~174
tails is not considered. The separation efficiency measures the fraction
of the feed that was perfectly separated.
Figure 3 was obtained by using a stipples multiple regression
program to generate an equation relating the SE to the two independent
variables. Contours of SE were then generated from the regression
equations. The results indicate an optimum condition of about pi 6.0
with about 110 minutes conditioning time. The numbers outside the
contours indicate the actual experimental results.
B. Lime Wash
During an earlier trial it was learned that the feed many times
contained more xanthate than the Syria procedure could cope with at
room temperature in 60 minutes. Since the amount of xanthate associated
with the scavenger feed was also variable a method was sought by which
the excess xanthate could be resorbed and the feed to the separation
process stabilized. The method adopted was a lime wash. The lime wash
was done by adding about 5 g of lime to 1 kg of feed in a 2.27 liter
container filled with water. The slurry was agitated gently for 30
minutes then the slurry was allowed to settle for 45 minutes at which
time the liquor was decanted and discarded. The thickened slurry was
then ground, if necessary, the pi was adjusted with sulfuric acid to 9.5,
xanthate was added, then the Syria conditioning was done.
C. Temperature-Time
A series of tests was done on a sample A with this procedure. The
parameters varied were the temperature during Syria conditioning and
the duration of the conditioning period.
The optimum region was seen to be at 40C with a conditioning
time of at least 60 minutes. An average of four tests done at these
conditions was an SE of 0.77. It was determined that the Syria
conditioning causes a depression of pentlandlte flotation whereas there
is little impact on pyrrhotite. The best nickel in pyrrhotite
corresponds to the optimum region for the separation efficiency.
D. pi Adjustment with Acid
In conjunction with these tests others were done in which the pi
was adjusted to the desired value (typically a pal of 6.5) before S02 was
added. All of these tests failed to produce a significant separation
owe
9 PC-2174
which established that the separation observed following Swallower
treatment was not due simply to pi adjustment, Furthermore, it was found
that using sulfur dioxide alone to reduce the pi following the lime wash
to pi 6.5 resulted in depression of too much pyrrhotite. Needless to say
the S02 addition was too large.
E. Oxygen Level During Syria Conditioning
An attempt was made to replicate the foregoing procedure on
sample A. The test results are shown in Table II along with the
dissolved oxygen levels during Syria conditioning.
TABLE II
RESULTS OF REPLICATED TESTS SAMPLE A
TestO~(ppm) SUE.
1 6-6.5 0.78
2 4.9-6.3 0.27
3 5.1-6.1 0.67
4 6.1-6.4 0.81
6.1-6.4 0.80
These results were decidedly surprising initially. An excellent
separation was generated in test 1 (SE 0.78) however, test 2 failed to
reproduce that result. Similarly, test 3, while better than 2 was not of
the high caliber of 1. An examination of the test revealed that the
dissolved oxygen was low at the beginning of S02 addition in tests 2 and
3. A procedural change was made: the pulp was saturated with oxygen
during the heating of the pulp in test 4 and 5 so that oxygen was present
when S02 addition was started. Clearly there was only a modest change
observed in the measured values of dissolved oxygen but the results of
the separation were vastly improved. Subsequently, the procedure has
specified oxygen saturation during heating and no less than 5.4 ppm
oxygen during S02 addition.
PC-~17
F. So Addition Rate
Experiments were done on sample B with three different flow rates
of sulfur dioxide. The total addition time was fixed at 60 minutes so
the total volume of SO added was alpha changed when the flow rate was
changed. The results indicate that for this procedure with this sample
the optimum xanthate addition was 0.05 gig Furthermore it appears
that within the range tested (480 my - 600 my per 60 minutes) the
results were not affected much by the S02 addition rate. Since it is
much easier to maintain a high level of dissolved oxygen with low
addition rates of S02, a change was made in the procedure from 20 mQ/min
to 9 mQ/min. Subsequently, a stipulation was added that the final pi
could not be allowed to fall below 6. Thus, the total S02 addition time
was reduced to 50 minutes.
G. Lime Aeration Prior to Pyrrhotite Cleaning
The usual procedure included a cleaning flotation stage on the
pyrrhotite rougher concentrate. Initially the pyrrhotite rougher
concentrate was conditioned with Syria for lo minutes prior to cleaner
flotation but this was relaxed to dilution wick water previously
acidified to pi 6 with S02.
Because the lime aeration procedure for Queen separation is so
effective in depressing pentlandite while floating both chalcopyrite and
pyrrhotlte, this technique was evaluated. After 60 minutes of lime
aeration the cleaning of the pyrrhotite concentrate was greatly improved
as is shown by the results in Table III.
11 PC 2174
TABLE III
IMPACT OF LIME/AERAl`ION OF PYRRHOTITE ROUGHER CONCENTRATE
Sample Best S E Best nope
SO~/airLime fir S02/airLime/air
B 0.73 0.76 1.69 1.71
C 0.61 0.70 1.99 1.36
D 0.46 0.58 3.20 2.24
E 0.43 0.71 2.20 1.31
F 0.55 0.69 1.89 1.52
G 0.68 0.70 2.09 1.80
H 0.53 0.59 1.98 1.39
In every case the pentlandite/pyrrhotite separation efficiency
was better with the lime aeration procedure than with standard of
cleaning in pi 6 water.
An early pilot plant trial included an addition of sodium
hydrosulfide then thickening to remove the resorbed xanthate. It was
discovered very early in the program that very little xanthate was belong
decanted but there was a high xanthate concentration in the liquor
associated with the thickener under flow. It was known for a long time
that a reducing or oxygen deficient environment was required for xanthate
resorption but obviously this information was not fully utilized. To
effect xanthate removal, the thickener underlie was pumped back to the
thickener feed in a closed pipe. A bleed stream was taken from this
recycle loop for feed to the process. This solved the problem. As a
consequence of thus observation the laboratory procedure was reexamined.
It was found that in the static thickener there was a heterogeneous
distribution of xanthate with the highest concentration associated with
then thickened solids. To overcome this the solids were raked with a
slow moving rake (8 rph). Samples taken at several levels in the
thickener following 17 hours of raking showed a uniform distribution of
xanthate.
The earlier circuit included Syria conditioning of the heated
bulk sulfide concentrate, followed by a jingle stage of flotation. Prior
12 PC-2174
to the pilot plant campaign, bench scale tests had included a pyrrhotite
cleaning stage in which the dilution water had been acidified with S02 eon
a pi of 6. The run did provide a ready source of pyrrhotite rougher
concentrate. Advantage was taken of this availability to do a series of
batch bench scale cleaner flotation tests after conditioning with lime
and aerating. Pyrrhotlte flotation improved modestly with increased
aeration in lime but pentlandite flotation decreased substantially as the
aeration time was extended; as a consequence, the cleaner separation
efficiency (SUE.) increased from about 0.25 with no aeration to lust
about 0.7 after 60 minutes aeration.
H. Hot Lime Conditioning
A series of tests was undertaken to examine the effect of
elevated temperature on the lime conditioning stage. This led to the
addition of lime to the pulp at the end of the Syria conditioning which
is done at 40C. The effect of conditioning time following lime addition
is shown in Table IV.
TABLE IV
RESULTS OF TESTS WITH HOT LIME CONDITIONING
PROWAR TO PYRRHOTITE ROUGHER FLOTATION
(SAMPLE It
Lime Condition Maximum SUE.
Time (mix) Observed Test
2 0.62
25 5 0.68 2
0.60 3
~.74 4
0.70 S
I 0.69 6
3020 0.68 7
0.75 8
0.75 9
0.73 10
I
13 PC-2174
Twenty minutes conditioning lime following time addition to pi
11.5 at 40C is considered quite satisfactory.
While in accordance with provisions of the statute, there it
illustrated and described herein specific embodiments of eke invention,
those skilled in eke art will understand that changes may be made in the
form of the invention covered by the claims and the certain features of
the invention ma sometimes be used to advantage without a corresponding
use of the other features.
