"Copper Solvent Extraction Reagents, Past, Present and Future" Gary A.

Kordosky Cognis Corporation Abstract Copper solvent extraction reagents are now more than 35 years old. Since the introduction of LIX 63, the first oxime proposed for the extraction of copper, the active extractants on which all copper solvent extraction reagents are based have evolved to two types of extractant molecules that are amazingly similar in structure, the so called aldoximes and ketoximes. The evolution of these reagents is traced with an emphasis on those developments that resulted in improved reagent properties. While neither the ketoximes nor the aldoximes best fit all copper leach solutions the combination of ketoximes with aldoximes and aldoximes with modifiers offers the end user a wide range of copper solvent extraction reagent blends that meet most of todays needs very well. However, new leaching technologies currently under development will make greater demands on reagents and may spark further advances in reagent properties. Recent research in Cognis laboratories has resulted in several new molecules and the basic copper extraction properties of these molecules are reported. The supplier picture of today has evolved along with reagent properties. Certain events and decisions that could have significantly changed todays suppliers of copper solvent extraction reagents are presented and discussed. Introduction The largest application today for solvent extraction in the mining industry is the recovery of copper from dilute sulfuric acid leach solutions; presently about 2000 tonnes of copper is recovered annually using this technology. This application for solvent extraction, as applied to the flowsheet in Figure 1, was targeted by the fledging Minerals Group of General Mills Chemicals in about 1960. At that time solvent extraction for the recovery uranium was well known, widely practiced and economically attractive for both the end user and the supplier of reagent (1). The decision to target copper solvent extraction technology was based on the fact that the Minerals Group of General Mills Chemicals believed this technology was economically attractive for both the end user and for themselves as a supplier. There were other applications for solvent extraction competing with copper and had one these other applications looked more attractive from a combined technical and financial point of view that application would have been targeted. It is interesting, but not surprising, that the first important decision leading to the development of copper solvent extraction technology was an economic decision. Both the end user and the reagent supplier are driven by the similar economic goals and

for any technology to be successful over time the economic goals of both the user and the supplier must be met. This is an important point to realize and it is not often discussed. It is not the intent of this paper to discuss economics, but, rather to remind the reader that economics have been an important part of the development of copper solvent extraction reagents since the beginning and that it will understandably be an important consideration for future developments. Reagent Development The first key technical breakthrough in the development of the commercially successful copper solvent extraction reagents of today was the identification by Ron Swanson in the early 1960s of the oxime functionality as the functionality around which to build a copper extraction reagent that would complete the flowsheet in Figure 1. This quickly led to the development and introduction of LIX63, Figure 2 (2). While LIX 63 had many of the properties required for a copper solvent extraction reagent to be successful, it did not extract copper below pH ~3.5 and was therefore eliminated as a reagent that could be applied in the flowsheet shown in Figure 1. Fortunately, the knowledge gained in developing LIX 63 allowed the relatively quick development and introduction of LIX 64 in late 1965.(3) LIX 64, composed of LIX 65 (Figure 3) and a catalytic amount of LIX 63, became the first commercial copper SX reagent when Ranchers Exploration and Development Corporation brought their copper SX plant at the Bluebird Mine on stream in March 1968.(4) While LIX 64 had sufficiently attractive performance for Ranchers Exploration to design and build the first copper solvent extraction plant, the reagent also had significant limitations with respect to extractive strength, metal transfer kinetics, copper / iron selectivity, viscosity and phase separation. In fact, it was recommended at the time that LIX 64 not be used at concentrations greater than a copper maximum loading of about 2.5 g/l Cu. The limited extractive strength of LIX 64 meant that a minimum of three extraction stages were required to achieve 90% copper extraction from typical dump leach solutions. The limitations of LIX 64 were well known by the Minerals Group of General Mills Chemicals. In fact, even before the Ranchers copper SX plant started up General Mills chemists were busy developing the much improved reagent LIX 64N, a reagent composed of LIX 65N (Figure 3) and a catalytic amount of LIX 63.(5) LIX 64N was added to the Ranchers plant as makeup reagent in November of 1968. The improved properties of LIX 64N include greater extractive strength, faster kinetics, lower viscosity, more rapid phase separation, lower entrainment and higher copper / iron selectivity. As a result of the improved performance properties of LIX 64N a much broader range of leach solutions could be treated under a wider variety of conditions including reduced staging requirements in extraction. In addition, reagent concentrations up to a copper maximum loading of about 12.5 g/l Cu could be used with LIX 64N. LIX is a registered trademark of the Cognis Group of companies

The difference between the structure of LIX 65 and LIX 65N is the R group on the phenyl ring of the extractant (Figure 3); it is a C-12 group in LIX 65 and a C-9 group in LIX 65N. This appears to be a relatively minor structural difference to give such a large difference in reagent properties, especially considering that the R group is on the opposite side of the extractant molecule from the oxime functionality. In fact, the much improved performance of LIX 64N over LIX 64 is due more to the improved manufacturing process and the higher purity of the reagent rather than the structural change in the alkyl group from C-12 to C-9. This knowledge proved to be very important at the time and again later when Shell Chemicals International decided to sell the rights to SME 529. The next significant development in copper solvent extraction reagents came from Ashland Chemical in 1968 when Kelex100 was offered as a reagent which would fill the perceived need for a copper solvent extraction reagent that could treat leach solutions having a lower pH and / or a higher copper content more effectively than LIX 64N could treat these leach solutions.(6) Kelex 120 followed in 1970.(7) Along with these reagents Ashland introduced the idea of using varying amounts of modifier, either nonylphenol or tridecanol, to facilitate the stripping of copper from strong copper extractants. Neither Kelex reagent was ever used commercially for copper recovery from dilute sulfuric acid leach solutions, however, the idea of using modifiers to alter the extraction / stripping equilibrium of a copper extractant represented a major advance in reagent development, an advance that resulted in improved reagents still in use today. In the mid 1970's Shell International Chemicals brought SME 529 (Figure 3) to the market as an alternative to LIX 64N.(8) This reagent found only limited commercial use because the poor properties of the side products from the manufacturing process offset the very good properties of the active extractant. Because SME 529 was not produced in high purity the industry did not realize at the time how effective the extractant molecule in SME 529 could be. In 1974 Birch reported the behavior of P-1 (Figure 3) from Acorga.(9) This reagent was reported to have very good properties including rapid copper transfer kinetics and excellent copper/iron selectivity. However, the reagent was a very strong copper extractant that required more than 225 g/l H2SO4 for efficient stripping, an acid concentration that is not compatible with the plating of high quality copper under currently acceptable tankhouse conditions. P-1 represented a significant advance in copper solvent extraction development since this molecule is the basis for many copper solvent extraction reagents commercially available today. In 1977 Acorga introduced the P-5000 series of reagents.(10, 11)) In introducing this series of reagents Acorga took the use of modifiers one step further than Ashland by combining the strong copper extractant P-1 with increasing amounts of nonylphenol to give a series of three reagents, P-5100, P-5200 and P-5300, where the extractive strength Acorga P-1 is produced by Avecia Kelex100 and Kelex 120 are no longer produced by Ashland Chemical Company

was systematically decreased so that the reagent could be matched to the leach solution. Copper leach liquors having a higher copper concentration and / or a lower pH could now be effectively treated with only two extraction stages even at low temperatures. Copper solvent extraction reagents based on modified aldoximes represent a major advance in copper solvent extraction technology that is with us today and will likely be with us for years to come. With the availability of the P-5000 series of reagents end users now had to evaluate the benefits of using a modified aldoxime, such as stronger copper extraction, faster kinetics and potentially less staging, against the drawbacks of using the P-5000 series of reagents, such as higher viscosity of the organic phase, increased entrainment, greater crud generation and higher reagent consumption per tonne copper produced when compared to the use of LIX 64N. It is interesting to see how the market passed judgement on these nonylphenol modified aldoximes. The reagent Acorga P-5100, which had the lowest modifier to oxime ratio in the P-5000 series, is still in use today. In this case the industry has judged that the benefits of using this reagent outweigh the drawbacks. However, Acorga P-5200 and Acorga P-5300, which had higher modifier to oxime ratios than Acorga P-5100, were not widely used and are no longer part of the Acorga line of reagents. The industry has judged that the benefits of using these reagents do not outweigh the drawbacks with respect to entrainment, viscosity, etc. In 1979 Henkel introduced LIX 622, a reagent composed of a C-12 aldoxime with tridecanol as the modifier. LIX 622 was first tridecanol modified copper extractant to be used commercially and today tridecanol is a widely used modifier. In the early 1980s Henkel reported on patented mixtures of very strong aldoxime extractants such as LIX 860-I (Figure 3) with the moderately strong ketoxime extractants such as LIX 64N and SME 529. These mixtures result in reagents having the fast kinetics and good copper extraction characteristics of the aldoximes combined with the stability and good physical performance of the ketoximes, all without the detrimental properties of modifiers.(12) The addition of the pure aldoxime extractant LIX 860-I to existing copper SX plants using LIX 64N or SME 529 afforded these plants an easy and quick way to upgrade their plant performance and flexibility if needed or desired without the negative properties of the modifiers. Aldoxime / ketoxime mixtures have proven to be a significant advance in copper solvent extraction reagents and today these mixtures produce more copper than any other single reagent system. These patented mixtures were crucial for the survival of Henkel as a reagent supplier. Why were the aldoxime / ketoxime reagents mixtures crucial to Henkels survival as a supplier of copper solvent extraction reagents? At the time these mixtures were developed the copper industry was changing. Copper companies were willing to pay more in operating costs in order to save on capital costs and leaching techniques were producing leach solutions containing greater amounts of copper at lower pH. These changes in the industry meant that stronger copper extractants having faster kinetics were required and that LIX 64N as a reagent by itself would soon be obsolete. Henkel needed a way to keep LIX 64N as a viable reagent. Necessity being the mother of invention

Henkel tested mixtures of LIX 64N with aldoximes and to their surprise found that these mixtures performed very well. In 1984 Henkel purchased the manufacturing and applications technology and associated patent rights to SME 529 from Shell International Chemical Company. Shortly before 1984 the solvent extraction group in Shell had come to realize what Henkel had known since the development of LIX 64N, that high reagent purity was the key to good reagent performance. However, the decision to sell the SME 529 technology had been made. Several years earlier work in Henkel laboratories had shown that purified SME 529 would perform very well and having confidence in their ability to produce a cleaner product Henkel quickly bought the rights to SME 529. In late 1986 using a new manufacturing process Henkel began producing LIX 84-I (Figure 3), a significantly improved version of the old SME 529. This reagent replaced both LIX 64N and LIX 65N in the Henkel line of reagents and led to mixtures of aldoximes with LIX 84-I. In 1986 the use of high molecular weight, highly branched alcohols and esters as modifiers was reported.(13) While the authors of the paper were optimistic about certain higher molecular weight alcohols as modifiers, in practice these modifiers caused very high entrainment in operating plants and they are not used today. However, the ester modified reagents are widely used today and are particularly noted for their exceptional copper / iron selectivity and very good stability. Copper Solvent Extraction Reagents Today The improvements in copper solvent extraction reagent properties detailed above are summarized in Table 1. These improvements have come about for 4 reasons: 1. 2. 3. 4. New molecules have been developed such as the active extractants in SME 529, P-1 and LIX 860-I. Better reagent manufacturing processes have resulted in cleaner reagents; LIX 84-I compared to SME 529 is a good example. Reagent properties can be manipulated; aldoxime blends with modifiers or with ketoximes are examples. The technical and economic success of leach-solvent extractionelectrowinning plants led to new leaching techniques which in turn forced reagent suppliers to develop new reagents to effectively treat leach solutions resulting from these new leaching techniques.

There are two distinct classes of modern copper extractants; ketoximes, where A in Figure 3 is a methyl (CH3) group, and aldoximes, where A is a hydrogen (H). A general comparison of the extraction characteristics of ketoximes and aldoximes, as well mixtures of ketoximes with aldoximes, is given in Table 2.

Table 1. Trends in Reagent Properties Since 1965 Property Extractive Strength Cu/Fe Selectivity Kinetics Stability Crud Generation Versatility 1965 Limited Poor Slow Excellent Moderate Poor 1970 Moderate Fair Moderate Excellent Good Marginal Late 1970s Strong Good Fast Good Moderate Good Today Tailored Excellent Fast Very Good Good Excellent

Table 2. Comparison of Properties for Reagents Based on Ketoximes, Modified Aldoximes, and Ketoxime/Aldoxime Mixtures Property Extractive Strength Stripping Cu/Fe Selectivity Copper Kinetics Phase Separation Stability Crud Generation Ketoxime Moderate Very Good Excellent Very Good Fast Very Good Low Aldoxime Strong Good Excellent Very Fast Fast Very Good* Variable Mixtures Customized Customized Excellent Fast Fast Very Good Low

* Dependent upon the particular modifier used. Dependent upon the leach liquor and the modifier. In many cases modifiers increase crud generation. The availability of pure aldoxime extraction reagents, such as LIX 860-I and LIX 860N-I, modified aldoxime reagents such as LIX 622N and LIX 664N, the pure ketoxime reagent LIX 84-I and various aldoxime / ketoxime nonmodified mixtures gives the operators of

copper solvent extraction plants a wide choice of reagents from which to choose, and choose they do. While some plants use a single reagent such as a modified aldoxime or a premixed aldoxime / ketoxime blend, many other plants purchase the aldoxime and ketoxime components separately and then blend a reagent best suited to their plant conditions. In fact, Chuquicamata has developed and uses a special reagent triple blend for their two copper SX plants, plants that operate very well under difficult conditions. Since ketoxime reagents are moderately strong copper extractants, they operate best when the pH of the leach solution is relatively high (above pH 1.8) and the solution is relatively warm (20 0C or higher). Ketoxime reagents strip well with only one strip stage and they can be used efficiently when the amount of acid in the tankhouse is relatively low (less than 160 g/l H2SO4). Ketoxime reagents are also highly selective over iron. Modified aldoxime reagents have excellent metallurgical properties even at low temperatures, low pHs or when the copper content of the leach liquor is very high and high copper recovery is needed. However, in many cases the use of a modified aldoxime reagent carries with it higher entrainment and/or greater crud generation resulting in increased carryover of impurities to the electrolyte and higher reagent consumption per tonne copper produced. While the nonmodified aldoxime - ketoxime mixtures also operate well at lower pH, lower temperatures and with leach liquors having a high copper content without the higher entrainment and crud generation problems of the modified aldoximes, the mixtures are slightly less effective than the modified aldoximes with respect to copper extraction. For this reason there are and will continue to be situations where the plant operator must make a choice between the better metallurgical properties of the modified aldoximes and the superior physical properties of the nonmodified aldoxime / ketoxime mixtures. Today there is not a reagent of choice for the extraction of copper from sulfuric acid leach liquors, but, rather many very good reagents and/or blends of reagents from which to choose. Copper Solvent Extraction Reagent Suppliers Currently there are two suppliers of oxime based copper solvent extraction reagents, Cognis and Avecia, interestingly two names that have not yet been discussed in this paper. Cognis, a wholly owned subsidiary of the Henkel Group, is an independent worldwide specialty chemical company that was carved from Henkel in 1999. Henkel in turn purchased General Mills Chemicals in 1977. What has remained constant through the name changes is the Minerals Group based in Tucson, Arizona which is responsible for developing and marketing the LIX Reagents and the commitment of Cognis management to solvent extraction technology. Avecia is a specialty chemical company that was formed July 1, 1999 following the management buyout of Zeneca Specialties from AstraZeneca that was financed by

Cinven and Investcorp. Zeneca in turn was formed when ICI split its industrial and specialty chemicals divisions into two separate companies in June 1993. ICI had joined with Anglo American to develop and market the Acorga line of copper solvent extraction reagents in 1972, later ICI acquired the Anglo American interests in the Acorga line of extractants. The Mining Division responsible for developing and marketing the Acorga Reagents has remained relatively constant through these corporate changes. What Might Have Been In studying the development of copper solvent extraction reagents it became more and more interesting to note that the current suppliers could have been very different if certain events had or had not happened. For example, during the development of LIX 65 and LIX 65N the Minerals Group of General Mills Chemicals was well aware that the A group in Figure 3 could be either CH3 or H instead of C6H5. Interestingly the two potential extractants, where A is CH3 or H, are not included in the claims of the patent granted for LIX 65 and LIX 65N. The author has asked several of the individuals who were involved with the patent application for LIX 65N why the molecules where A in Figure 3 is either CH3 or H were not claimed. The following was learned. Just prior to the development of these reagents an extensive industry survey done by the Minerals Group of General Mills Chemicals found that one of the major concerns of the copper companies was the stability of the copper solvent extraction reagent under the conditions it would be used. The molecule where A is C6H5 was thought to be much more stable than the molecules where A is H or CH3 so General Mills Chemicals chose to claim only that particular class of molecules. In fact, the molecule where A = C6H5 is more stable than the molecules where A = H or CH3. However, the molecules where A = H or CH3 are sufficiently stable under the conditions of copper solvent extraction so that what General Mills Chemicals believed to be a major concern in the mid to late 1960's was a much less important issue 10 years later. Consider what might have happened if General Mills Chemicals had patented the molecules where A = CH3 or H. The rights to these molecules would have rested with General Mill Chemicals and therefore would not have been available to the companies that developed SME 529 and Acorga P-5000. In all probability the reagents based on the active extractants in SME 529 and Acorga P-1 would not have been developed at the time they were developed even though they might have been developed at a later date. In effect reagent development would have stalled, at least for some time. Furthermore, Shell Chemicals and Acorga might never have gotten into the copper solvent extraction reagent business. Think about what might have happened if Shell International Chemicals had realized the importance of producing a more highly purified reagent and done so. SME 529 would have found wide commercial acceptance, a sharp blow would have been dealt to the Minerals Division of General Mills Chemicals and Shell International Chemicals might still be a supplier today.

Speculate on what would have happened if, at the time Acorga introduced the modified C-9 aldoximes, Shell had been producing a purified SME 529. Would General Mills, as the supplier of LIX 64N, have survived as a supplier of copper solvent extraction reagents? To do so would have been a major challenge. Now consider what might have taken place if Acorga had mixed P-1 with the then available ketoxime reagents, LIX 64N and SME 529, and recognized the benefits these mixtures provide. Acorga, not Henkel, would have patented these mixtures and in doing so Acorga would have been in position to control the copper solvent extraction reagent market. While Acorga was able to patent the C-9 aldoxime they were not able to patent the C-12 analog. This molecule later became, and is still today, an important part of the Cognis line of reagents both as a stand alone, modified aldoxime reagent and in mixtures with ketoximes. While the events in what might have been is a litany of missed opportunities for one reagent supplier or another, it can be argued that the end user is much better off for these missed opportunities. Reagent properties are far superior to those of LIX 64N while at the same time the price of reagent on a copper loading basis has dropped by 30 to 40 % since LIX 64N was introduced in 1968. The Future In looking at the future leaching technologies under development Cognis has identified three goals with respect to reagent development: 1. 2. 3. Increased copper over iron selectivity, A reagent based on a molecule of intermediate extractive strength between the currently available ketoxime and aldoxime molecules, and For any new reagent the development of a manufacturing process having a cost that allows for an acceptable return on investment.

Several molecules have been synthesized and tested as described below. The first set of extraction / stripping data in Table 3 was generated in the following way for experimental reagents 1 and 2. The respective extractants were diluted in a low aromatic hydrocarbon diluent at a concentration to give a copper maximum loading of about 5.6 g/l. The respective solutions were then contacted with a typical strip electrolyte having 35 g/l Cu and 160 g/l sulfuric acid for 10 minutes in order to produce a stripped organic. The respective stripped organic was then contacted with a standard aqueous feed solution having 6 g/l Cu, 3 g/l iron (III) at a pH of 2.0 for 10 minutes. These respective partially loaded organic solutions were then contacted once again with the strip electrolyte for 10 minutes to produce a typical stripped organic for two strip stages.

Finally these typical stripped organic solutions were contacted for 10 minutes with a performance feed solution having 2.5 g/l Cu and 4 g/l iron (III) at a pH of 1.6. LIX 84-I, a ketoxime reagent, LIX 860N-I, a nonmodified aldoxime reagent, and LIX 622N, a modified aldoxime reagent, are shown for comparison. The second set of extraction / stripping data in Table 3 was generated in the following manner. The extractant was diluted in a low aromatic hydrocarbon diluent to give a copper maximum loading of 3.05 g/l. This organic was then contacted 3 times with strip electrolyte, 5 minutes each contact. The stripped organic was then contacted with the performance feed solution for 5 minutes in a mixer box to give an equilibrium loaded organic. This test was run for LIX 84-I, experimental reagent 2 and experimental reagent 3. Kinetics show that all extraction and stripping reactions are at equilibrium in less than 5 minutes therefore the data reported is equilibrium data. On the one hand both experimental molecules 1 and 2 have remarkable selectivity when compared to the known reagents LIX 84-I, LIX 860N-I and LIX 622N. Unfortunately experimental reagent 1 is a very strong extractant as shown by the very high strip point while experimental reagent 2 is only a slightly stronger extractant than LIX 84-I. On the other hand experimental reagent 3 has a strip point midway between a ketoxime and a nonmodified aldoxime reagent, one of the properties Cognis was trying to incorporate into the molecule, however, the selectivity of this molecule is very poor indeed. Table 3. Copper and Iron Extraction / Stripping Data for Several Experimental Reagents Reagent 860N-I 622N 84-I XI-A XI-B XI-C 84-I XI-B M.L. g/l Cu 5.7 5.58 5.55 5.51 5.51 3.05 3 3.04 S.O. g/l Cu 2.89 1.87 0.29 3.54 0.45 1.06 0.16 0.2 Q.C. Feed L.O. g/l Cu Fe ppm 5.33 1.1 5.06 1 4.15 1.7 5.35 0.1 4.43 0.01 Performance Feed S.O. L.O. g/l Cu Fe ppm g/l Cu Fe ppm 3.21 0 5.03 6.2 2.13 0 4.25 9.6 0.33 0 2.45 6.3 3.8 0 5.24 0.4 0.56 0 2.7 1.8 0 2.83 76 0 1.92 1.3 0 2.02 0.3

Cognis has accomplished the first goal of designing and synthesizing a molecule with significantly better selectivity than the molecules that current reagents are based on. Cognis has also accomplished the second goal of designing and synthesizing a molecule having a reagent strength intermediate between the molecules current reagents are based on.

Significant challenges remain. Will Cognis be able to use the knowledge gained from the research work already done to synthesize a molecule combining the remarkable selectivity of experimental reagent 1 and 2 with the extractive strength of experimental reagent 3? Will Cognis be able to produce an improved reagent based on experimental reagent 1, 2, 3 or some yet to be synthesized molecule at an acceptable cost, especially considering the cost performance of existing reagents? The answer to these questions is not known. What is known is that the decision to develop a new copper solvent extraction reagent will be an economic decision. A new reagent will be fully developed and introduced if and only if Cognis believes such a reagent will meet Cognis investment goals. The reason for this is simple, the development of a new copper solvent extraction reagent must compete for investment funds with other projects in the company.

DUMP or HEAP LEACH

ELECTROWINNING
RAFFINATE PREGNANT LEACH SOLUTION LOADED ORGANIC PREGNANT ELECTROLYTE SPENT ELECTROLYTE

EXTRACTION
STRIPPED ORGANIC

STRIPPING

SOLVENT EXTRACTION
Figure 1. Conceptual Flowsheet for Copper Leach / Solvent Extraction / Electrowinning

OH NOH CH3(CH2)3CHCH CCH(CH2)3CH3 C 2 H5 C 2 H5

Figure 2. Structure of the Active Extractant in LIX 63

OH

NOH A

Reagent LIX65 LIX65N SME 529 & LIX84-I P1 & LIX860N-I LIX860-I

R C12H25 C9H19 C9H19 C9H19 C12H25

A C6H5 C6H5 CH3 H H

Figure 3. The General Chemical Structure for Oxime Copper Extractants

(1). J.E. House, "The Development of the LIX Reagents", 1981 Gaudin Lecture, AIME Annual Meeting, Chicago, February, 1981. (2). Swanson, R.R., and D.W. Agers, 1964, A New Reagent for the Extraction of Copper, AIME Annual Meeting, New York, February. (3). Agers, D.W., J.E. House, R.R. Swanson and J.L. Drobnick, 1965, A New Reagent for the Liquid Ion Exchange of Copper, Mining Engineering, December, pp. 76 - 80. (4). Power, K.L., 1970, Operation of the First Commercial Liquid Ion Exchange and Electrowinning Plant, AIME Annual Meeting, Denver, February. (5). Dement, E.R., and C.R. Merigold, 1970, A Progress Report on the Liquid Ion Exchange of Copper, AIME Annual Meeting, Denver, February. (6). Ashland Chemical Company, 1968, Preliminary Technical Data Sheet, Copper Solvent Extraction Reagents. (7). Hartlage, J.A., and A.D. Cronberg, 1973, Chemical and Physical Factors to be Evaluated Upon Designing a Kelex Extraction System, KIM Conference of Metallurgists, Quebec City, August. (8). van der Zeeuw, A.J., 1975, Symposium on Hydrometallurgy of the Institution of Chemical Engineers, Manchester, Institution of Chemical Engineers, Symposium Series No. 42, p. 161. (9). Birch, C.P., 1974, The Evaluation of the New Copper Extractant P-1", Proceedings of the International Solvent Extraction Conference, ISEC 74, Lyon, September. (10). Tumilty, J.A., R.F. Dalton and J.P. Massam, The Acorga P-5000 Series: A Novel Range of Solvent Extraction Reagents for Copper, The Institution of Mining and Metallurgy, London, Advances in Extractive Metallurgy, 1977. (11). Tumilty, J.A., G.W. Seward and J.P. Massam, 1979, The Acorga P-5000 Series in the Solvent Extraction of Copper: Performance Characteristics and Implications for Plant Performance Proc. of the International Solvent Extraction Conference, ISEC 77, CIM Special Volume 21. (12). Kordosky, G.A., J.M. Sierakoski, and J.E. House, 1983, The LIX 860 Series: Unmodified Copper Extraction Reagents, Proceeding of the International Solvent Extraction Conference, ISEC 83, (Denver CO, American Institute of Chemical Engineers), pp. 191-192. (13). Dalton, R.F., K.J. Severs and G. Stephens, 1986, Advances in Solvent Extraction of Copper by Optimized Use of Modifiers, Proceedings for IMMs Mining Latin America Conference, Chile, November.