Rate of extraction of copper from aqueous solution

routinely for the determination of the rate of copper extraction with new commercial formulations or new batches of the same extractant. The extractio...
0 downloads 0 Views 301KB Size
Rate of Extraction of Copper from Aqueous Solutions Sir: The increasing importance of the industrial production of copper by hydrometallurgical techniques has resulted in the development of a number of organic extractants under various trade names, e.g. LIX64N (General Mills), KELEX 100 (Ashland Chemical Company), and VERSATIC 911 (Shell Chemicals Ltd.). These commercial extractants are invariably kerosene solutions of organic chelating agents. The addition of a surfactant to the kerosene solution has been found to improve the phase separability and the addition of a second chelating agent has, in several instances, increased the rate of copper extraction from the aqueous phase. A definite need has arisen therefore, for a simple method that can be used routinely for the determination of the rate of copper extraction with new commercial formulations or new batches of the same extractant. The extraction processes of industrial importance are quite rapid and it is futile to attempt to follow the kinetics of extraction by withdrawing and analyzing samples of the organic or aqueous phase at various intervals in the course of the extraction ( I ) . Two general approaches have been found useful in the study of the kinetics of rapid extractions. One method that is widely employed makes use of an apparatus (AKUFVE) for the continuous measurement of the distribution of a solute in an extraction process (2). The apparatus consists of a mixer, a centrifuge for separation of the phases, and on-line detectors for the measurement of the solute concentration in the separated phases. Large volumes of the two phases are required and the effective interfacial area that is generated within the mixer cannot be determined. Despite these shortcomings, reliable distribution data for several liquid-liquid extraction systems have been obtained with the AKUFVE apparatus. A second method that has gained wide acceptance is the single drop technique in which droplets of known volume and surface area of one of the phases are allowed to contact the second phase for a predetermined period of time (3, 4 ) . The problems that arise from the hydrodynamics of droplet systems are only moderately well characterized and the limited time of contact between the two phases restricts the amount of rate data that can be obtained. The method, however, has the advantage that the interfacial area can be treated as a variable and much useful information on the kinetics of solvent extraction has been obtained with the single drop method. We have devised a simple and rapid method for monitoring the rate of extraction of copper ions from an aqueous phase into an organic phase containing a ligand that chelates copper ions. The concentration of copper(I1) in aqueous phase can be continuously monitored without separation of the organic and aqueous phases. In addition, an experimentally reproducible interfacial area can be obtained by mixing the two phases under controlled conditions. EXPERIMENTAL Measured volumes of an aqueous phase containing copper(I1) and an organic phase such as kerosene, containing a chelating agent, were introduced into a 150-mL jacketed vessel which was thermostated. The two phases were mixed with a variable speed magnetic stirring motor and stirring bar that was l'/z inches long. The rate of stirring was determined by measurement of the frequency of the alternating current that was induced in a coil of coated copper wire wound around the jacketed glass vessel. At a stirring rate of 500 rpm or greater, the organic phase was uniformly dispersed in the aqueous layer and the rate of extraction was found t o be independent of the stirring rate. The rate of extraction of copper(I1) from the aqueous phase was monitored with an Orion Cupric Ion Selective Electrode (Model 94-29) with 1636

ANALYTICAL CHEMISTRY, VOL. 49, NO. 11, SEPTEMBER 1977

a double junction reference electrode. The electrodes immersed in a stirred two-phase system gave stable and reproducible potential differencesfor several hours at a time. Unstable readings were caused by the accumulation of the organic phase in the double junction reference electrode. This was remedied by periodically cleaning the outer compartment of the double junction and refilling it with a solution of 10% KNOB. The potential differences were measured with an Orion Model 701 digital voltmeter and rate data were obtained from potential difference readings that were recorded as a function of time on a strip chart recorder. The kinetic data were obtained as follows: -75 mL of a 1.00 X lo-' M solution of copper(I1) nitrate and 25 mL of metallurgical grade kerosene were mixed in the jacketed vessel at 23.9 h 0.1 O C and stirred at 500 rpm for several minutes until the digital voltmeter reading was constant. When the kerosene phase was evenly dispersed throughout the aqueous phase, a 2-mL sample of concentrated LIX, (obtained from General Mills) was injected into the two-phase mixture with a calibrated syringe. The potential difference between the copper ion selective electrode and the reference electrode was recorded as a function of time for about 1 2 min. In one series of experiments that is discussed below, the effect of using LIX63, LIX65N, and LIX64N on the rate of extraction of copper from an aqueous solution is clearly demonstrated.

RESULTS AND DISCUSSION The success of the method depends on the response of the cupric ion selective electrode to the aqueous phase concentration of copper(I1) in the presence of the dispersed organic phase. The electrode response was Nernstian in 75 mL of an aqueous solution of Cu2+ M-10-5 M)when the volume of the kerosene phase was increased from 2 to 25 mL and the rate of stirring maintained a t 500 rpm. The electrode can be used over a wide range of p H and temperature. The only common transition metal ions that interfere with the electrode response are Ag+, Hg2+, and Fe3+. Hence, care should be exercised in the interpretation of the rates of extraction of leach solutions that very often contain high concentrations of Fe3+. The reproducibility of the kinetic data is governed by the manner in which the kerosene phase containing the organic chelating agent (LIX) is contacted with the aqueous phase containing the copper(I1). The best results were obtained by injection of a small volume (2 mL) of a concentrated solution of LIX in kerosene into the evenly dispersed two-phase system of kerosene (25 mL) in the aqueous solution (75 mL) containing Cu2+, while stirring continuously a t 500 rpm. Meaningful kinetic data are obtained only in the first 2 min after injection of the concentrated solution of LIX. The addition of a large volume of the kerosene solution containing the LIX to the aqueous phase containing Cu2+is best avoided because the time that is required for the large volume of kerosene to be evenly dispersed, and for the electrode to attain equilibrium is at lease 30 s. Figure 1shows the relative rates at which Cu2+is extracted by LIX63, LIX65N and LIX64N in kerosene under comparable conditions. LIX64N extracts Cu2+ rapidly and equilibrium is reached in a little over 2 min. The rate of extraction of Cu2+with LIX63 is slower than with LIX64N and the rate of extraction is slowest with the LIX65N which has not reached equilibrium and is continuing to extract the Cu2+even after 3 min have elapsed. These qualitative results have been substantiated by other workers (3-5). The organic chelating agent that is present in LIX63 is an a-hydroxyoxime, and in LIX65N a benzophenone oxime. LIX64N is a mixture containing LIX65N (-45%) and LIX63 (-170).LIX64N

paratus for the extraction of copper(I1) by LIX reagents have been reported recently (3-5). The results indicate that the interface plays an important role in the extraction process. It would be of interest to use the method proposed above to verify this conclusion.

ACKNOWLEDGMENT The authors are grateful to R. N. Longwell of the Bluebird Mine, Miami, Ariz., for providing the LIX reagents used in this work.

LITERATURE CITED I. P. Allmarln, Yu. A. Zolotov, and V. A. Bodnya, Pure Appl. Chem., 25,

I

I

20

I

6o

I

I

seconds

I

I

I

140

loo

Flgure 1. Variatlon of the rate of extraction of Cu2+ from an aqueous solution Into kerosene solutions of LIX reagents. The solution consisted of 75 mL of M Cu(NO& and 25 mL of kerosene. Two mL of the

concentrated LIX reagent in kerosene was injected into the solutlon that was being continuously stirred at 500 rpm with a magnetic stirrer. Millivolt readings were recorded 5 s after injection of the LIX is clearly a superior extractant than either LIX63 or LIX65N, several explanations have been advanced for the synergistic effect of LIX64N, but convincing experimental evidence that supports any of these explanations is lacking. Kinetic data obtained by the single drop method and the AKUFVE ap-

667 (1971). J. Rydberg, Acta, Chem. Scand., 2 3 , 647 (1969). R. L. Atwood, D. N. Thatcher, and J. D. Miller, Metall. Trans., 6B, 465

,.-. -,. fIQ75)

R. J. Whewell, M. A. Hughes, and C. Hanson, J . Inorg. Nucl. Chem., 37, 2303 (1975). (5) D. S. Flett, D. N. Okuhara, and D. R. Splnk, J. Inorg. Nucl. Chem., 35, 2471 (1973).

Stephen J. Kirchner Quintus Fernando* Department of Chemistry University of Arizona Tucson, Arizona 85721 RECEIVED for review April 26, 1977. Accepted June 9,1977. This work was supported by the Ranchers Exploration and Development Corporation, Albuquerque, N.M.

Corrosion of Stainless Steel by Organic Solvent Mixtures Sir: In the course of conducting liquid chromatographic experiments we have noted occasional problems which arise when organic liquids are allowed to remain in contact with stainless steel materials (type 316). Some corrosion effects are anticipated since it is known that formic, acetic, and propanoic acids are corrosive at room temperature (1-3). Carboxylic acids with higher molecular weight serve as corrosion inhibitors, while all organic acids are corrosive at temperatures above 300 "C (4). Other organic compounds, anhydrides, aldehydes, and those containing sulfur are also known to be corrosive to metals (5). By contrast, most of the common organic solvents are regarded as noncorrosive. We wish to report our finding that certain of these solvents, although noncorrosive individually, may display a highly corrosive attack upon stainless steel when used in the form of mixtures. Our findings can be summarized in terms of the following test. A 1.0-g piece of 316 stainless steel tubing, or of NBS Standard Reference Material No. 1155, was added to a glass vial containing 20 mL of a solvent or a solvent mixtureequally proportioned by volume, After 10 days of unagitated storage at 20-23 "C the samples showed the following changes. (I) No apparent change: with pure C C 4 (this solvent is occasionally quite troublesome in HPLC, however), tetrahydrofuran, acetone, or diethylether. (11) Slight yellow coloring of the liquid with CC14 + acetone, C C 4 + THF, C C 4 + diethylether, and CC14+ isopropyl ether + acetone. (111) Forms viscous, brown colored liquid: CCll diethylether + acetone, CC14 tetrahydrofuran acetone. The results were not dependent on storage in room light or in darkness. Mixtures I1 or I11 produced changes that could easily be detected spectrophotometrically, by 210% transmission loss after 2 or 3 h.

+

+

+

The foregoing experiments were repeated using solvents of varied purity: reagent grade, those which had been commercially redistilled (in glass), and "spectroscopy" grade. The results were clearly dependent on the chemical nature of the materials and qualitatively, at least, independent of their source or purity. Since the presence of C C 4 seemed critical to the results gathered so far, we also tested the effect of using chloroform in its place. These results were quite similar, although the reaction rates were clearly slower. The weight loss was measured in experiments similar to I1 and 111. After a period of time the steel was removed and rinsed with solvent. This removed a thin f i i from the surface of the steel. Mixed CC14 T H F caused weight loss that was linear with time up to 10% loss after 8 days. The loss from mixed CC14 + T H F acetone was also linear, but at twice the rate, up to 4 days. Then the rate accelerated to 23% loss after 24 days. The rates in I11 were boosted considerably by heating under boiling reflux. When the film was allowed to remain intact through gentle handling, the same experiments showed a weight gain of the dried sample of metal plus film. Gas chromatographic analysis of the liquid phase showed that a number of volatile products had formed. In a separate experiment, mixtures of steel and solvent were stored in the dark at room temperature for 10 days. Gas chromatography showed that the reaction products contained a series of volatile compounds, as shown in Figure 1. An attempt to use GC/MS did not provide specific structure determinations, but it became clear that most of the volatile species contained more than one chlorine atom. The volatile products in the case of the two mixtures in 111, at least, seem to be quite dangerous. Severe eye irritation and headache were experienced after a single restricted exposure to test the odor. Gas evolution from any of the mixtures in

+

+

ANALYTICAL CHEMISTRY, VOL. 49, NO. 11, SEPTEMBER 1977

1637