CA1062474A - Extraction and separation of copper and other metals from ammoniacal solution - Google Patents

Abstract

TITLE

EXTRACTION AND SEPARATION OF COPPER AND OTHER METALS
FROM AMMONIACAL SOLUTION

INVENTORS
Gordon M. Ritcey Bernard H. Lucas ABSTRACT OF THE DISCLOSURE
Aqueous ammoniacal solutions of copper and other metals selected from zinc, cobalt and nickel are solvent extracted at an equilibrium pH of about 7 to 9 to sequentially separate copper followed by the other metals. The extraction reagent is a hydrocarbyl-substi-tuted 8-hydroxyquinoline. The loaded solvent is acid-stripped to recover the metal and recycled. The order of extraction is Cu > Zn > Co > Ni, except zinc and cobalt can be interchanged by altering process variables. The process has particular application to alkaline pressure leach solutions of ores, concentrates or tailings, or deep-sea manganese nodules.

Description

lO~Z~74 This invention deals with the solvent extrac-tion o~ ammoniacal solutions of copper and other metals such as zinc, cobalt, and nickel to sequentially separate out the copper followed by zinc or cobalt, and finally any nickel present. The solvent ex~raction reagent is a hydrocarbyl-substituted 8~hydroxyquinoline.
~ With the developmen~ of hydrometallurgical r~ processing for the treatment of ores and concentrates, there have been several leaching schemes proposed to solu-bilize the metal values. Among these have been the Cymet process, which is essentially a ferric chloride leach, ~!
`- and the Arbiter process, which is an alkaline leach using ` ~ low temperature and pressure, not unlike the ammonia pres-,',~ sure leach of Sherritt Gordon Mines Limited. In addition, there is the ammonium carbona~e leach of Caron for the ;"~ .
;l treatment o~ nickel laterites. Although acid leaching ~ii has been satisfactory for the treatment of many ores and concentrates, the advantage of an alkaline leach lies in its ability to reject iron and other metals that do not .',;. . ,. :: form ammine complexes with ammonia, and which hydrolyze at the alkaline conditions of the leach. Subsequent sepa-ration of the metal values, such as copper, nickel, 2inc `~
~ and cobalt, would thus be considerably s.implified in the l absence of iron and other impurities~ A typical leach ;l solution, resulting from alkaline pressure leaching of nickel sulfide concentrates at elevated temperatures with ammonia, has the following approximate analysis, in A~ .
! g/l: Ni, 46; Cu, 12; Co, l; (NH4)2S04, 250; and free NH3, lO0.
The hydrocarbyl-substituted 8-hydroxyquinoline ., .. :.
~ reagents are known - see U.S. Patent 3,637,711 Budde Jr.
,; . .: . .: :
et al January 25, 1972 (Ashland) and have been used ~or `:
.i .. .
",, ~ .
' ::'.'~
~! . : .

. , ,, :. ".
copper extraction at acidic pH. These reagen~s have also been used for extraction of some other metals such as zinc and cadmium at acidic pH. Ritcey and Lucas have recently investigated the use of the substituted 8-hydroxy-quinoline reagent Kelex 100 (trademark - ~shland Chemicals) , for solvent extraction of various metals at selected aci-,~ dic pH values as reported in CIM Bulletin February 1974 pages 87-92. In Canadian Patent Application 232,495, filed July 29, 1975, G. M. Ritcey et al describe the use sl - 10 of substituted quinoline reagents in sequential extraction at selected acidic pHs of metal values including mol~b-denum, copper, cobalt and zinc. Thus the use of these .
substituted qunioline reagents for solvent extraction at acidic pH has been demonstrated but no mention of solvent I extraction ~t alkaline pH using these reage~ts has been ,, noticed.
. .
,~, In accordance with this invention, aqueous ammoniacal solutions of metals comprising copper and one or more of zinc, cobalt and nickel are contacted with a ~; 20 ~ solvent extraction medium containing a hydrocarbyl-sub~
stituted 8-hydroxyquinoIine extraction reagent, at an equilibrium extraction pH of about 7 to 9.5 to preferen-tially extract~copper from the aqueous phase into the sol-$ vent phase. After separating the two phases the loaded solvent phase is stripped with acid to recover the copper values. The aqueous phase (from which copper has been removed) is again contacted with a solvent extraction !
medium at pH about 7 to 9 and zinc or cobalt, depending on the process variables such as equilibrium pH, tempera-ture and retention time; selectively extracted. After separation of the aqueous and solvent phases, the aqueous phase can again be extracted or otherwise processed to

-2-.:
. ~ .

:

~--` 106Z474 recover cobalt, zinc or nickel. Where the residual aqueous phase contains substantially only nickel it may be advan-tageous to recover the nickel by sulfide precipitation, electrolysis or other means, instead of by further solvent extraction.
The primary extraction reagent used in this process is described in U.S. Patent 3,637,711 mentioned above. Preferred reagents of this type are the 7-hydro-carbyl-8-hydroxyqùinolines. One preferred group of hydro-carbyl substituents are the alkenyl groups, e.g. beta-alkenyl groups having from about 8 to 18 carbon atoms.
Alkenyl groups of the formula CH2 = CH - CH -R
where R is an alkyl group of from S to 14 carbon atoms are parti¢ularly suitable. In the tests reported below, Kelex 100, ~trademark), a beta-alkenyl-8-hydroxyquinoline reagent was used.
. . .
The aqueous ammoniacal feed solution can be ~20 a solution resulting from ammonia leaching of ores or it can be obtained from various effluents, wastes, etc. from mineral and metallurgical processing plants. Typical of the`ores that can be used are copper-molybdenum ores, copper-lead-zinc ores, copper-nickel ores, copper-nickel- ; ;
cobalt ores, etc. Of course, the entire procedure that is carried out ultimately depends on the types of metal values present in any aqueous ammoniacal solution to be treated. Deep-sea manganese nodules can also be treated.
The procedure of this invention has the impor-~; 30 tant advantages of being capable of producing high purity products with minimum pollution, minimum capital costs and minimum production costs as compared to present smelting , .

.

practice~ This is made possible particularly because of the fact that the same organic reagent can be re-used for the separate extraction of each metal value.
Usually when an ore is being processed it is first concentrated by a suitable preconcentration tech-nique and following preconcentration, the concentrate ;
is leached by atmospheric or pressure leaching. The ,., resultant leach slurry may be filtered and the filtrate, together with washings is sent to the solvent extraction circuit. The leach liquor thus obtained can be an ammo-;' ~ niacal carbonate, nitrate, chloride or sulfate solution and a typical such solution may contain for ins~ance from 5 to 50 g Cu/1. Some iron will normally be present in many of the commercial ores, but is not normally solu-bilized by the ammoniacal leaching.
A wide variety of organic diluents in which J the extraction reagent is dissolved can be used according to the invention. The minimum requirements of the diluent -that is to be used are that it be substantially water- 1 -, ~1 20 immiscible, that it will dissolve the extraction reagent and that it will provide an adequate phase separation.
1, : .
Tha diluent can be an aliphatic or aromatic hydrocarbon, halogenated hydrocarbon, etc. Examples of these diluents ¦~
include toluene, carbon tetrachloride, benzene, etc. The preferred diluents are liquid aromatic hydrocarbons boiling in the kerosene range. A variety of mixed commercial sol- ;~
vents of this type are available. The extraction reagent ~ ~:
preferably also contains a modifier which aids in solu-."~, bilizing the metal species in the organic phase and improves phase disengagement. A variety of known modifiers can be used, including TBP (tributylphosphate), isodecanol, 2-ethyl hexanol and nonylphenol. ¦~
~4~ 1`
~ 3~ .
I`"', , .'.

--- 106;~47~

It has also been found to be desirable to pre-treat the extraction reagent before using it in the process of this invention. This pre-treatment is a cleaning pro--cedure to remove undesirable impurities and can be accom-plished by contacting the solvent mixture of extractant, modifer and diluent with a metal sulfate solution such as , , copper sulfate, followed by stripping with sulfuric acid ` and water washing. - ~ -'; ,,.: :
~ The contacting of the metal-bearing aqueous -~
'. 10 ammoniacal solution with the solvent extraction reagent can be carried out by any of the well known procedures employed in liquid-liquid extraction. Although continuous ~.,,: : ; .. :
counter-current methods are preferred, batchl continuous ~;~ batch, and batch counter-current methods are also useful.
Any suitable liquid-1iquid contacting system may be em~
ployed such as a pulse column, a mixer-settler or a highly `;
~l agitated column.~ The temperature at which the mixing and yll extraction is carried out will affect the separation fac- `
tors and should be appropriately selected and controIled -~i 20 (for example see Figure 5).
The ratio of the volume of organic phase to aqueous~phase can be varied oonsiderably and the most I
efficlent ratio~in each case can be readi~y determined by one s~illed in the art. However, generally the aqueous l to organic ratio will be within the range of about 1/5 to .. ,;j . '.
5/1, depending upon the metal concentration, concentration of extraction reagent, etc.
i Copper, cobalt, zinc or nickel values can be stripped from the loaded organic phase by a variety of i 30 strong acids such as nitric, hydrochloric or sulfuric acid. The copper values are preferably stripped with -sulfuric acid having a strength of about 150 to 200 g/l H2SO4.

, ., , , ~,., ."~
.,~,i , j ,. , ,~ . j ., .
., i~

6~474 , This,removes the c~pper as copper sulfate solution which is ,` amenable to electrolysis. A substantial quantity of this acid remains with the solvent and this recycled acid helps -to control the equilibrium pH on re-use of the solvent phase.
, ................................ . ..
' Aspects of the invention are illustrated in the ,. . ..
attached drawings in which:
... .
,', Figure 1 is a graph depicting loading of copper ~,, vs. equilibrium pH for varying ammonium sulate concentrations.
, Figure 2 is a graph similar to Figure 1 but for loading of cobalt. ~ ' ., :, . .
~,~ ; Figure 3 is a graph of copper-zinc separation ,,~
,' (Cu/Zn ratio~ plotted against equilibrium pH for five ammo- '', ~.,; . . . .
',J~' ' nium sulfate contents. ~ ' Yigure 4 is similar to Figure 3 but,for four ~' ` ' , ammonium carbonate solutions.
. !, .
,~ Figure 5 is a graph depicting separation fac~
tors vs. retention time at room temperature and at 50C.
i Preerred aspects of the process can be sum- ' 'J marized in the following flowsheet. "
Cu, Zn, Co, Ni Ammoniacal Solution ~ ' ~' '' .
p~ 7-9.5 . , ~ . : . , Cu Loaded ~ ~ ,, Sol.~- -Stripping~ Ext~ action~ Kelex 100 !, , , pH 7-9 ~; - '. r-1 Zn Loaded ' ,.1 Sol. ~ Stripping~ -Extraction ~ Kelex 100 ,~ Solvent ,1 Raffinate pH 7.5-8.5 :i:
~ Co Loaded '1 Sol.~ -Stripping ~ Ext~ action~-- -Versatic 911* or ~ ' Solvent Kelex 100 i~
~!
~'1 Raffinate pH 7~5-8.5 :,: ~.
,=i~ Ni Loaded ~
, , Sol.<- Stripping~ Ext action ~ LIX 64N** or i~ Solvent Kelex 100 Ammol iacal Raffinate to recycle , q~ 6- , `"
i,, ' ~
!~ ,' ~06Z474 ~ ~ ~
; * Versatic 911 is a trademark for a carboxylic acid extraction reagent (Shell).
** ~IX 64N is a trademark for a substituted benzophen-oxime manufactured by General Mills Inc.
? ~ . .
After contact of the leach solution, at pH
7 to 9, with Kelex 100 to extract the copper, the raffinate is again contacted with Kelex 100 to selectively extract ~ :
the zinc at an equilibrium pH of 7 to 8. Following zinc extraction,;the cobalt could easily be extrac-ted and sepa-~ ~ 10 rated from the remaining nickel by Kelex 100 in the pH
,~ ~ range of 7~ to 10. However~ once Kelex is~loaded with co-baLtj it is diffLcult to strip unless Co2+ i5 extracted, n~which case the cobalt is then readily~stripped. The ` cobalt can also easily be extracted from~such a solution . by~use of the tertiary monocarboxyllc acid, Ver$atic 911, at a pH of 7.5 to 8.~5. Nlckel is then recovered from the raffinate by extraction wlth;~IX 64N at a pH of 7.5 to 8.5 if the nickel concentration~is greater than 1.0 g/l, or ~ `
with Kelex lOO~i lt is less than 1 g!l. The loaded sol- ~ ;
20 ~ vents can all~be stripped wlth sulfurio, nitric or hydro~
chloric acid to;~recover the metal values. The acid picked ;up~:in~the Kelex stripping clrcuits is released to the aqueous phase during the subse~uent extraation, and finally , ~ . .
reacts~wlth the ammonia to produce ammonium sul~ate, thereby maintaining pH aontrol. Extraction o~ an~ o~ the metals 1 ~ . . .
by Kele~ 100 can be accomplished from solutions containing up to 5~00 g ~(NH4)2SO4/l~or 500 g (NH~)2CO3/1~ but at this high salt concentration, extraction is best at a pH of be-

3: tween 7.0 and~7.5. As the~salt concentration dqcreases, 30 the equilibrium pH for extraction can be increased.
Because cobaltous cobalt is readily extracted and easily stripped with 150 g H2SO4/1, and because the .,~` ~ . :

`- -- 106Z474 ~ equilibrium loading capability is not dependent on the metal .
concentration in the aqueous feed as in-the Vexsatic 911 5 ; system, then Kelex 100 could be of great advantage for the treatment of cobalt-nickel solutions arising from nickel '.?
reduction, as in the Sherritt Gordon process. ~ -; Following extraction, the solvent contains amounts of ammonia and ammonium sulfate which, if not re-moved, would be stripped with the metal, e.g. Cu, and thus contaminate the electrolyte feed to electro-winning. The ammonia and the salts are removed by water scrubbing prior to acid strlpping.
The solvent used for the illustrative tests described here consisted of a 0.5-molar solution of Kelex 100 containing 10 volume per cent (v/o~ isodecanol in an ;~
aromatic diluent, Solvesso 150 (trademark). This solvent mixture was subjected to a purification stage prior to the -~ investigations by contacting it with a copper sulfate solu-tion, followed by acid stripping with lO v/o H2SO4 and then water washing. Any impurities in the original solvent mix-ture were aqueous soluble and were discarded with the raf^
finate.
The aqueous feed solutions used in the tests were synthetic feed solutions containing one or more metali~, together with various concentrations of either (NH4~2SO4 or NH4)2CO3 Based on the~extraction results given below, an~approximate order of extractability can be defined as follows:
Cu ~ Zn > Co > Ni; SO4= ~ CO3~
although zinc and cobalt can be interchanged by control ; of equilibrium pH, temperature and retention time. (If molybdenum were present with the system, it would be ex~
tracted to a much less extent than nickel).
-~ , .
ii3,~ -8- ~
, .
, ,~ , ~Q~Z474 ~u-Ni-Co 5eparations Aqueous feed solutions were prepared, each containing about 5 g/l each of copper, nickel and co-balt, and ~arying concentrations of (NH4)2SO4 ranging from 50 to 500 g/1. These solutions were contacted with 0.5M Kelex + 10~ isodecanol in Solvesso 150 at an A/O of ~ ;
3, and at equilibrium pH values of 7O0l 8.0 and 9Ø The results, shown in part in Figures 1 and 2, indicate that ~
with an increase in (NH4)2SO4 concentration from 50 to ~ -500 g/l, and an increase in equiIibrium pH from 7 to 9, the extraction of copper decreased, that of cobalt in- ~
creased and the nickel loading remained~relatively con- `
stant, and low, ranging to 0.2 g Ni/l. The discrimi- 1 nation of Cu over Ni was high (>100) at low pH, but de-creased with increasing pH and (NH4)2SO4 concentration to 25. The discrimination of Cu over Co ranged from 15 , to 1 as the ~NH4)2SO4 and pH were increased.
From the results o~ cobalt extraction in the presence of Cu and Ni (Fig. 2), ît appeared that copper was preferentially extracted to cobalt. Because the co-~; ~ balt was co-extracted over the pH range and ~NH4)2SO4 concentrations investigated ~Fig. 2), urther investigatiolls were conducted with repeated stagewise contack of khe ~ Kelex with a feed solution containing 5 g/l each of Co3+
.
and Cu, in 300 g ~NH4)25O4/1. At a phase ratio ~A/O, of I ~ 3, a 5-minute contact time and an equilibrium pH maintained I at 8.0, the results showed that, although copper was ini~
tially extracted preferentially to cobalt, repeated con- `
tacts scrubbed and replaced the copper with Co3+~ A
slight decrease in total loading occurred with repeated ~`
contacts, because Co3~ would require three moles Kelex as opposed to Cu requiring two.

. , j, .
,. . .
".

:1~62474 ;

Further investigations showed that by prepa-ring a solution containing Co2+, and maintaining the co-balt in the Co~ state during the extraction, Co2~ could be readily recovered by acid stripping. Once Co ~ is extracted, however, it must be stripped in a relatively short time. The results of an ageing study of the strip-ping characterlstics of a solvent saturated with cobalt, exposed to air and stripped using 10~ H2SO4 showed that although Co2~ is readily extracted and stripped, pro-longed exposure to air oxidizes the Co2+ to Co3+, resul-ting in a decrease in cobalt stripping.
!
.r~

Cu-Zn Separation ,: . .
I Feed solutions containing approximately 5 g/l .
,~ eaoh of copper and zinc were prepared, also containing concentrations of either (NH4)2SO4 or ~NH4)2CO3 ranging from 50 to 500 g/l. These solutions were contacted with 0.SM Kelex + 10% isodecanol in Solvesso 150 for 5 minutes at equilibrium pH values of 7.0, 8.0 and 9Ø At a pH of 7.0 and 50 g/1 salt, the metals were difficult to maintain , in solution. Figure 3 shows that an increase in both the I

(NH4)2SO4 concentration and equilibrium pH improved the ~ ratio of Cu/Zn in the loaded solvent from 5 to 80. The ¦ total metal loading as a function of pH and (NH4~2SO4 ¦~ concentration was measured. In Figure 4 sre shown com-parative results for the (NH4)2CO3 system, indicating that, ,~ although the (NH4)~CO3 system can be used, the discrimi-nation of coppsr over zinc is slightly less than in the (NH4)2SO4 system. Also, with increasing (NH~2CO3 concen-30 tration and pH, the discrimination of Cu over Zn is in-creased only~at the lower (NH4)2C03 concentrations. Above 100 g (NH~)25O4/l, the discrimination decreased with increasing pH.
. - . . ...
i --10--:' .
.
. .. . , . : , , , , :

~L~624~4 - ................................................................ ... :

EXA~lPI.E 2 (Cont . ) - ~ -Ni-Zn Separation ` .
The ammoniacal system of zinc and nickel was :~ :
then exa~ined for extraction and separation of the two metals. Feed solutions containing 5 g/l each of nickel .
.. and zinc were prepared, aIso containing quantities of either (NH4~2SO4 or (NH4)2CO3 ranging from 50 to 500 g/l i concentration. :As in the previous tests noted, these feed ,J solutions were contacte~ with 0.5M Kelex + 10~ isodecanol : ~ '.' ~ 10 TABLE 1 : :
;~ Extrac~ion and Separation of Zinc and Nickel from (NH4~2SO4 .: ~. ..... .
~ . ( 4 2 3 - :
. , ,, ,,, ~, .:
_ ~ - . .
Loaded Snlvent. Zn/Ni Ratio Concentration .. : .
, . (NH ) SO or (NH ) CO Equiv. .
,: 4 2 4(g/1) 4 2 3 pH (NH4~2SO4 (NH4~2CO3 . .
J : ~ ~ _ ~`~ 50 8 4Pp6 P3P8 9 2.5 4.1 .- ~ 100 7 9.9 ppt ~ 200 : 7 25.0 ppt ~ 8 6.7 ~.9 :~ ~ 9 9.1 6.4 i 300: 7 24.7 7.4 , . 8 17.3 2~5 '~ 9 16.5 6.6 . 500 7 14.4 5.5 I 8 8.3 4.9 : : 7.6 ., . . .. ' . ln Solvesso 150 for 5 minutes at equilibrium pH values of .
7.0, 8.0 and 9.0, and at an A/O ratio of 3. In Table 1 : .
are shown comparisons of the extractability of zinc from . 30 either (NN4)2CO3 or (NH4)2SO4 solutions in the presence -:~

:.. ~
, ~ '., ~ :

.. ~,::

~0624~4 XAMPLE 2 (Cont.) of nickel, as a comparison of the Zn/Ni ratio in the loaded solvent from the two extraction systems. The conclusions ~ ;
that can be drawn from these results are as follows:
1. In an (NH4)2SO4 medium, the extractability (E) of zinc is greater than in an (NH4)2CO3 medlum. With an increase in either the pH or salt content, Ezn is ~e-creased~ Precipitation occurred more readily in the car-bonate system than in the sulfate system, i.e. precipitation~
occurred below 300 g (NH4)2CO3/1 at pH 7.0 compared with -precipitation occurring only at the 50 g (NH4)2SO4/1 level.
2. With the exception o~ the SO~g ~NH4~`2SO4/1 concentration, the total metal loading was greater in the (NH4)2SO4 medium than in the (NH4)2CO3 medium over the pH
range of 7 to 9 (Table 1~. Highest loadings for both sys-tems were attained in the range of 100 to 200 g/l of salt 1 , :
concentration.
3. The selectivity of zinc over nickel was ; hlgher in the sulfate system than in the carbonate s~stem over;the pH and salt concentration range investigated (Table l). The~Zn/Ni ratLo generally increased with ln-creasing sAlt ooncentration.
Ni Solution `
Preliminary results showed very low loading of the 0.5M Kelex with nickel. TestS were conducted to l determine~whether the reagent LIX 64N (General Mills Inc.) ;
¦; ~ would be more suitable. Solutions containing 5 g Ni/l and varying amounts of (NH4)2SO4 were contacted with either 0.5M Kelex or 0.28M LIX 64N at an A/O ratio of 2 and at an equilibrium pH of 8. The results indicated LIX 64N to be more suitable than Kelex 100 for the extraction of Ni from such ammoniacal solutions, and showed that the loading of ~)6Z~74 ~ ~
:
EXAMPLE 2 (Co t.) LIX 64N is not seriously afected by an increase in the (NH4)2SOg concentra~ion from 100 to 500 g/l. ;~

Effect of Retention Time and Temperature on Extraction and Separation ~. ;
Tests were performed to determine whether con-trolling the retention time and temperature would affect the kinetic~ and thus improve the metal discrimination.
A feed solution containing S g/l each of copper, nickel, cobalt and zinc, and 300 g (NH4~2SO4/1, was contacted with ~ the 0.5M Kelex solvent, at an A/O of 3/1 for varying periods I of time, at room temperature and 50C, and at an equili-brium pH of 9Ø The results in Figure 5 indicate that : : .
extraction and separation are kinetically controlled, and -the following ob-~ervations were made. ~ ! -.
r; . 1. Extractability of copper is enhanced at 50C~compared with room temperature up to a retention time of 2 mi~nutes, after which the extraction is best at room temperature. This~may be~due to increased extraction of cobalt with increased time.
, 2. Extraction o~ cobalt is enhanced at the elevated temperatures over the time period tested. ;

3. Extraction o~ zinc is enhanced at room temperature, whereas that of~nickel was extremely low ¦~
(E ~0.02) over the time period and temperature tested.

4. The separation of Cù/Co is optimum at 30 seconds at 50C, and after 2 minutes the optimum is at l`
room temperature; for Cu/Zn, elevated temperature of ex~
traction improves the separation which increases with ~ increasing retention ~time. ¦
X,.,-~.

.: i ` ' . . .. .
.. .

~062474 Extxactant Concentration and Metal Selectivity - Tests were performed at Kelex concentrations ranging from 0.05 to 1.5M. Feed solution mixtures of metal pairs of copper and cobalt, cobalt and nickel, and cobalt ~ ;
and zinc were used, each solution containing 5 g/l of each metal together with 200 g (NH4)2so4/l. The equilibrium pH
of extraction was 8.0, and the A/O ratio was varied from 3/1 to 1/10 as the solvent concentration was decreased from 1.5M to 0.05M Kelex. Contact time was 5 minutes. The selectivity of one metal over the other is shown in Table 2.
`. , , :

Solvent Concentration and Selectivity ~: . ... :
i~ , ._ , .
~. Metal Ratio in Loaded Solvent .
~ . _ Kelex Concn. Cu/Co Zn/Co Co/Ni Co/Ni (M) pH 8.0 pH 8.0 pH 8.0 pH 5.0 ~ :
, _ _ 1.5 ~ 1.1 1.3 7.8 16.3 1.0 I.3 1.3 10.2 4.8 .
0.5 ~ ~1.3 1.4 16.6 1.0 0.25 2.5 2.1 28.
. 0.10 . 3.2 2.7 .
. 0.05 7.8 3.6 31.3 .

As the :Kelex concentration is decreased rom 1.5 to 0.05M, , . .
:~l : at a pH o-f 8.0j the selectivity of Cu/Co, Co/Ni and Zn/Co .1 .
is increased.(Table 2), although the E values for each metal, in each pair, are decreased. That is, operating l; 30 con~ltions must be selected to give a balance between maxi~
. mum metal selectivity and good extraction. Similar data .
~ for Co/Ni separation at an equilibrium pH of 5.0 are also ;'~ 14- :
~ ' .
,. . . ....
i `
., : .

106Z~74 ~ ~
. :

shown in Table 2, indicating the opposite results at a pH of 8.0; tha~ is, the discrimination of cobalt over nickel is decreased as the Kelex concentration is de-creased from 1.5 to 0.5M. The pH's in Table 2 are at equilibrium.

.~' ~ . :,, .
Scrubbing and Stripping Scrubbing tests, using either copper or zinc ammonium sulfate solutions, with 300 g (NH4)2SO4/1 at a pH of 8.0, proved unsuccessful in the removal of cobalt (Co2~ or Co3+~ from a Kelex solvent containing 11 g Co/l.
Various scrub solutions, at several concentra- `~
tions, were used in attempts to selectively remove nickel and cobalt from the copper on the solvent. The re~ults in Table 3 indicate that the scrubbing was ine~fective ; , for removal of cobalt Co2~ (as noted earlier), and that there appears to be selectivity in the stripping of nic-kel ahead of cobalt. That is, any co-extracted nickel . - .:
may be removed and separated from copper by using a dilute ~`; ;

H2SO4 scrub stage^

~, . . .
Single-Stage Stripping of Loaded Solvent ,~
Loaded 9olvent: 13.1 g/l Cu, 0.08 g/l Ni, 2.24 g/l Co2~, <0.1 g/l 2n, A/O 1/1 _ _ . .....
Strip Solution Metal Stripped ~%~

Conc. _ ' ~ ~yp~ (g/l) Cu Ni Co ! H SO 50 27.9 96 .04 ;-2 4 100 42.6 100 .04 ;~
, 30 150 59.1 100 -.Og-200 79.4 100 .O
_ . . , ,: .

~--- 106Z47~

EX~MPLE S (Cont.) Cu, Ni and Zn are readily removed by acid stripping, whereas cobalt i5 most difficult unless in the -cobaltous state. Use o~ strong acid concentrations oE HCl, HNO3 and H2SO4 (20 vol. ~) removed only about 25~ of the cobalt Co3+ in one stage, at a phase ratio of l and at a temperature of 50C. It can only be concluded that Co2+
was readily stripped and not Co3+. The use of EDTA for scrubbing or stripping was also ineffec-tive. The use of H2S has previously been reported for the removal of cobalt from LIX 64N. A loaded solvent containing 7.33 g Co~3/1 was contacted with H2S yas for 5 minutes. The resultant solvent, ater separation of the precipitate, was analyzed and showed 76% removal of the cobalt. No change ~in the solvent molarity resulted with this treatment.

Stripping-Extraction Cycle , Tests were carried out in the alkaline system ~;
'~ ~ to determine whether the presence of acid resulting from the stripping operation would be detrimental to the sub-sequent extraction and whether there was a re~uirement for .
the water wash to remove the acid. It has been reported that the acid loading occurring in the stripping circuit ~, prevents the use o~ Kelex lO0 for subsequent extraction ;~ from alkaline solutions. Feed solutions containing about .

5 g Cu~l and 300 g ~(NH~)2SO4/l, at pE values of 8 to ll, were contacted at an A/O ratio of 3 for 5 minutes with 0.5M Kelex lO0 containing lO v/o iso~ecanol in Solvesso 150. For one series of tests the solvent mixture was ~ -30 water washed prior to extraction, and in the other series the extraction was performed directly after acid stripping ~,: :,, 1~ " .:, , , ,.' :
....::

6Z~7~
using three ~tages of 10 v/o H2SO4 at a phase ratio (O/A) of 1. The results indicated increased extraction of copper up to an equilibrium pH of 9.5 and a decrease in extraction ~ .-as the free N~3 in the raffinate is increased from 18.6 to 73.4 g/l. ~; .
With the sulfuric acid still present on the solvent from the stripping operation, there was very little difference as compared to the results obtained by first . `
water washing to remove the acid. It would therefore -:
appear that, rather than adversely affect the extraction of copper, the acid present on the solvent neutralizes the free ammonia in the feed solution, producing ammonium :
. .
sulfate, thereby maintaining a buffered system and the . .
desired pH control. The phases separated relatively 'r~
quickly, the time ranging from 25 to 35 seconds for the :.
primary break, and 55 to 75 seconds for the secondary :`
'~ break, as the equilibrium pH increased from 8.5 to 11.1 The invention can thus be defined as a met-hod of sequentially separating copper and other metals selected from zinc, cobalt and nickel, from aqueous ~:
ammoniacal solution thereof, by solvent extraction, com- `
prislng~
(a) contacting said ammoniacal solut.ion with l a solvent extraction medium containing a hydrocarbyl-sub- :~
,1~ stituted 8-hydroxyquinoline extraction reagent, at an ~.
.~ equilibrium extraction pH o about 7 to 9.5 to preferen tially extract copper from the aqueous phase into the ~' :,.
~:: solvent phase;
3~
(b) separating the aqueous phase from the ,-solvent phase and acid-stripping the solvent phase to re- ~:
cover the copper values;
. . -17- .::.
, ~ ~ , .. . .

i,., . ' :

~ ~6Z474 (c) again contacting the aqueous phase with a solvent extraction medium containing said..quinoline-type or a carboxylic acid-type extraction reagent, at an equilibrium pH of within the range 7 to 9 to further extract one of zinc, cobalt and nickel;
and ~d~ again separating the aqueous phase from the solvent phase and acid-stripping the solvent phase to recover the zinc cobalt or nickel therein.
A preerred set of extraction conditions for copper would be pH 7-9; salt concentration 50-500 g/l; and temperature 40-60C. .

' . ' ' '~ ' ' .,.~
~ . . , . . .~: .
'I '~, .

:, ., :.
,i;..- ..
~'.' . -' ''.
'' ' : ~',~',' '.' .:
~ -18~

~, . .

~ - ... ~- . .. . .:,

Claims (10)

1. A method of sequentially separating copper and other metals selected from zinc, cobalt and nickel, from aqueous ammoniacal solution thereof, by solvent ex-traction, comprising:
(a) contacting said ammoniacal solution with a solvent extraction medium containing a hydrocarbyl-sub-stituted 8-hydroxyquinoline extraction reagent, at an equilibrium extraction pH of about 7 to 9.5 to preferen-tially extract copper from the aqueous phase into the solvent phase;
(b) separating the aqueous phase from the solvent phase and acid-stripping the solvent phase to recover the copper values;
(c) again contacting the aqueous phase with a solvent extraction medium containing said quinoline-type or a carboxylic acid-type extraction reagent, at an equilibrium pH of within the range 7 to 9 to further extract one of zinc, cobalt and nickel;
and (d) again separating the aqueous phase from the solvent phase and acid-stripping the solvent phase to recover the zinc, cobalt or nickel therein.

2. The method of claim 1 wherein the hydrocarbyl substituents on the quinoline are selected from alkenyl groups having from 8 to 18 carbon atoms.

3. The method of claim 1 wherein ammonium sul-fate or ammonium carbonate is present in the aqueous solu-tion in amounts within about 50 to about 500 g/l.

4. The method of claims 1, 2 and 3 wherein the loaded solvent phase is water scrubbed to remove ammonium salts before acid stripping.

CLAIMS (Cont.)

5. The method of claims 1, 2 and 3 wherein acid remaining on the acid-stripped solvent extraction medium is recycled with the stripped medium to maintain the de-sired equilibrium extraction pH.

6. The method of claims 1, 2 and 3 wherein se-quential extraction and acid-stripping of copper, at least one of zinc and cobalt, and nickel are carried out.

7. The method of claims 1, 2 and 3 wherein co-balt is present in the ammoniacal solution and it is ex-tracted and stripped while in the cobaltous valence state.

8. The method of claims 1, 2 and 3 wherein a carboxylic acid extraction reagent is used for extraction of cobalt.

9. The method of claims 1, 2 and 3 wherein a substituted benzophenoxime extraction reagent is used for extraction of nickel.

10. The method of claims 1, 2 and 3 wherein cop-per is extracted under the conditions:
pH 7-9 Salt concentration 50-500 g/l Temperature 40-60°C