Fu Xun, Hu Zhengshui ,Wang Debao, Liu Huan, Hu Xiaopeng
(Qingdao Institute of Chem. Tech., Shandong 266042, China)

Received Feb. 14, 2000; Supported by the National Natural Science Foundation of China (No. 29971020).

Abstract Recent advances in the study of three-phase extraction systems have been reviewed. The results are emphasized on the third phase formation, extraction mechanism, and application of the newly formed phase.
Keywords  Extraction, The third phase

In liquid-liquid extraction system, the problem of three-phase (i.e., two organic phases and one aqueous phase) can be encountered. The formation of the third phase might be reported first by Healy et al^[1]. The third phase (or the middle phase, the heavy organic phase) interferes with the extraction process. Many authors have applied themselves determining when the third phase forms and knowing how to avoid its formation since 60's^[2-5]. Srinivasan et al^[6] reported that there was a Limiting Organic Concentration (LOC, when the metal loading concentration in the organic phase is higher than that value, the organic phase will split into two parts, heavy and light organic solutions) and that they measured the values for tri-alkyl phosphates diluent / Pu(IV), U(VI), Th(IV) HNO₃ system. A review of third phase formation in extraction of actinides by neutral organophosphorus extractants was made by Vasudeva Rao⁽⁷⁾. The results indicated that all experimental conditions, such as aqueous acidities, diluents, extractant structure and extracted metals, can affect the third phase formation. Das et al^[8] reported the three-phase behavior discovered in the ammonium and amine extractant systems, such as tri-octyl amine, Aliquat336. The methods of avoiding the third phase formation usually used are raising extraction temperature, adding modifiers and keeping the metal loading concentration below LOC^[9,10].
    Since the 80's, several achievements in the studies of extraction processes from the viewpoint of interfacial chemistry have been reported^[11]. Wu et al^[12] testified that reversed micelle or w/o microemulsion formed in the organic phase; Wang et al^[13] proved that micelle or o/w microemulsion formed in the aqueous phase of the extraction system; and Osseo Asare^[14] indicated that the third phase in an extraction system could be analogue to a middle phase microemulsion in a typical surfactant system. But, general speaking, studies on the three-phase extraction are lacking, the results emphasized on the distribution of metals between the three phases, the extraction mechanism, microstructure of the third phase, especially on the application of the newly formed phase are lacking. The recent advances in the studies of three-phase extraction systems will be reviewed in this paper.
1. THREE-PHASE BEHAVIOR AND THE EXTRACTION MECHANISM
1.1 Acidic extractant system
Organo-phosphoric acid could be taken as a representative of the acidic extractants. The pseudo triangular phase diagram was for the first time used by Paatero et al^[15] in studying the phase behavior of extraction system of Cynaex272 - n-hexane / NaOH aqueous solution. There were distinguished mono-phase microemulsion, two-phase, three-phase and liquid crystal regions in the phase diagram. The result indicated that the highest solubilization of water in the organic solution was found at NaA/HA=2:1. The author then purified Cyanex272 to di-(2,4,4-trimethylpentyl) phosphinic acid (HDTMPP), and investigated the effect of modifier TOPO (tri-octylphosphoric oxide) on the phase behavior of the above system^[16]. The results showed that the sodium salt of HDTMPP (NaDTMPP) could be considered as an anionic surfactant, and that both impurities and modifiers in it could affect its amphiphilic properties, then influence the phase behavior of the extraction system. One of the most obvious changes was that three-phase region became smaller in the presence of the modifier TOPO, but there was no detailed study on the newly formed middle phase.

	Fig.1 Phase behavior of HDTMPP-kerosene / H2O-NaOH-Na2SO4 system at 298K. (taken from ref.16) (a) CHA(i)=0.4mol/L, Na2SO4 0.1mol/L; (b) a=0.75, Na2SO4 0.1mol/L; (c) CHA(i)=0.4mol/L, a=0.75

Fig.2 Plot of phase volume vs. CH2SO4,b for the TBP-kerosene/ H2SO4-H2O extraction system at 298K, TBP, 10-60%. (taken from ref.17)

Fig.3 Phase diagram of the TOA·HCl-heptane-H2O system at 298K. (taken from ref. 19)

    They also investigated the three-phase extraction of TBP-kerosene / H₂SO₄-TiOSO₄ (0.2mol/L) and compared it with above H₂SO₄ extraction system^[19]. In the metal system, the composition of the third phase (in terms of TBP, H₂SO₄, Ti(IV) and H₂O) was a function of C_H2SO4,b in the range of 6.3-10.2 mol/L. The extraction of Ti(IV) hardly occurred in the two-phase region, but extraction ratio increased rapidly after the third phase formed. It was attributed to the co-aggregates formed by TBP·TiOSO₄·(H₂O)₄ (written as P₃) with available TBP:H₂SO₄=1:1 complexes.
1.3Ammonium and amine extractant system
Considering tri-octyl amine as a representative, Fu et al^[20] prepared the pure salt of TOA·HCl, and proposed the triangular phase diagram of TOA·HCl-heptane-H₂O system at 298K (see Fig.3). The phase behavior was controlled by the amphiphilic properties of the salt, like a cationic surfactant.
    Most phase boundaries were straight lines. The salt-rich phases (including liquid crystal phase and heavy organic phase) presented a typical lamellar structure owing to the self-assembly of the salt molecules. The solubilization capacities of water in the polar layers and that of heptane in the apolar layers were within the values of TOA·HCl : H₂O : heptane = 1:2:3 (molar ratio).

Fig.4 Phase behavior of the extraction systems TOA-kerosene / HCl (ZnCl₂ or FeCl₃) at 298K. (taken from ref. 20)
a, acid system; b, Zn²⁺ system; c, Fe³⁺ system.

    The authors^[21] then measured the phase behavior of TOA in heptane (0.98 mol/L)/HCl (0 - 10 mol/L) system, and indicated that the third phase formed after C_HCl,i > 0.1 mol/L. The composition analysis for the middle phase (in terms of TOA, TOA·HCl, heptane and H₂O) showed that the solubilization of water and heptane were less than the above molar ratio in the range of C_HCl,i= 0 - 1.0 mol/L, i.e., the neutralized fraction of TOA by HCl a<1. It was attributed to the effect of free TOA on the polarity of the salt aggregates. The investigation also showed that the phase behavior changed when kerosene was used as the solvent or metal ion such as Zn²⁺ or Fe³⁺ was added in the HCl aqueous solution (see Fig.4). The distribution data of metals indicated that the metal was extracted mainly into the middle phase, and the metal content in the light organic phase could be negligible.
2.THE MICROSTRUCTURE OF THE MIDDLE PHASE
Acidic extractants (saponified), extractants containing nitrogen (salted), and neutral extractants can be analogues to the anionic, cationic, and non-ionic surfactants respectively, so can be the third phase in an extraction system to the middle phase microemulsion in a typical surfactant system. It is well known that the middle phase has a bi-contineuous structure, but no detailed information of the structure of the third phase formed in an extraction system has been found in the literature. We have chosen many samples of the third phase to observe their microstructures by the TEM technique after the freeze-fracture replication.
The images of the middle phase in TOA-heptane / HCl-water extraction system are shown in Fig.5 and those in TBP-kerosene / H₂SO₄-TiOSO₄ are shown in Fig.6. Their typical lamellar structures imply that the packing parameter of the amphiphilic molecules in the extraction systems is about unity, i.e., R=V/al equals about one (where V represents the molecular volume, a the section area of the polar head, and l the length of the hydrophobic group).           

     The images of the middle phases in TBP-kerosene / H₂SO₄-water extraction system are shown in Fig.7. The sample 2 was taken from the boundary B (C_H2SO4,b=10.4 mol/L, C_H2SO4,m=2.7 mol/L, see Fig.2), where the extracted complex was mainly in the form of TBP·H₂SO₄·H₂O, and it showed a typical lamellar structure owing to the complex assembly. Samples 1 and 3 were chosen from regions II and III respectively. Their TEM images indicated that both the existing complex composed of TBP:H₂SO₄ >1:1 and solubilized H₂SO₄ and water could weaken the lamellar structure. It seems that the microstructure of the third phase went through a loose state tight state loose state process with the increase of C_H2SO4,b in the three-phase region, and it can be illustrated by the transition of microemulsion of the Winsor types IIIIII with the increase of aqueous acidity.
3. APPLICATION OF THE THIRD PHASE
Two aspects of studying the applications of the newly formed phase have been reported in the literature. Firstly, the purification method for acidic organo-phosphoric extractants was established based on the principle of concentration effect of the amphiphilic compounds in the middle microemulsion. Hu et al^[22] reported that P204, P507 and Cyanex272 could be purified by striping the third phase formed in the extractant-gasoline / NaOH-Na₂SO₄ equilibrium system. Compared with the copper salt recrystallization method, this new method has many advantages such as simplicity, high recovery, high purity and less cost.
    Secondly, some nanosized ultrafine powders were prepared by direct precipitating the loaded metal in the middle phase. Yang et al^[23] prepared ZrO₂ sized 10 nm with high purity by precipitating the metal from the third phase formed in TBP-kerosene / mineral acid-Zr(IV) extraction system. The features of this method are that, the micro-environment similar to reversed micelles can confine the growth of particles; the amphiphilic species including extractant and its ion-pair with metal can prevent particle agglomeration; and the separation effect of the extraction process can ensure the high purity.
    Our laboratory investigated the preparation of ultrafine powder of TiO₂ using the middle phase formed in the TBP-kerosene / H₂SO₄-TiOSO₄ extraction system.^[19] The results indicated that the amorphous precipitate changed to anatase crystal after calcination at 600℃, and changed to rutile crystal after calcination at 1200℃. The anatase powder had a narrow size distribution of 20 nm, but the high temperature calcination would make the particles agglomerated. It is worth indicating that a rather higher metal content in the organic solution than any other methods, such as alkoxide hydrolysis, sol-gel method, and microemulsion method, promises the production of ultrafine powder on a large scale.
4. CONCLUSIONS
4.1 Acidic extractants (saponified), extractants containing nitrogen (salted), and neutral extractants can be analogues to the anionic, cationic and non-ionic surfactants respectively. The phase behavior of an extraction system is controlled by the amphiphilic properties of the extractant or extracted species. The microstructure of the third phase depends on the packing parameter of the amphiphilic molecules, R=V/al. The middle phases of the systems TBP/H₂SO₄ (metal) and TOA/HCl (metal) have lamellar structure.
4.2 The process from formation to disappearance of the third phase can be analogous to the transition process of a surfactant microemulsion system from Winsor type II, through type III to type I. The extraction mechanism in the three-phase region may be different to that in the two-phase system. In two-phase region, the transfer of metal to the organic solution by a complex reaction, but in the three-phase region, the solubilization must be considered. Metal is extracted mainly into the third phase, and the metal loading will change the phase behavior.
4.3 The artificially prepared third phase can be used in purification of extractants and production of ultrafine powders.
4.4 We suggest paying more attention to the study of three-phase extraction systems on their phase behavior, extraction mechanism, especially on the application of the newly formed third phase.

REFERENCES
[1] Healy T V, Mckay H A C. Trans. Farady Soc., 1956, 52: 633.
[2] Ochhkin A V. Fiz. Khm., 1960, 54: 1862.
[3] Kertes A S, Mckay H A C, Healy T V  et al. Solvent Extraction Chemistry of Metals. Clevelland: CRC press, 1967.
[4] Hanson C, Patal A N. J. Appl. Chem., 1969, 19: 20.
[5] Irving H, Edgington D N.  JINC, 1959, 10: 306.
[6] Srinivasan T G, Vijayasaradhi S, Dhamoduran R  et al. Solvent Extraction and Ion Exchange, 1998, 16 (4): 1001.
[7] Vasudeva R P R. Solvent Extraction Ion Exchange, 1996, 14 (6): 955.
[8] Das N  R, Lahiri S.  Sep. Tech., 1994, (5): 791.
[9] Blumberg R.  Liquid-Liquid Extraction. London: Academic Press, 1988.
[10] Xu G, Wang W, Wu J  et al. Principle of Extraction Chemistry. Shanghai: Science and Technology Press of Shanghai,  1988.
[11] Wu J, Shi N, Zhou W  et al.  Proceedings of the Second Conference of Solvent Extraction of China, Shanghai, 1991, 7.
[12] Wu J, Gao H, Chen D  et al.  Science in China B (Zhongguo Kexue), 1980, 23 (12): 1533.
[13] Wang D , Wu J , Li Y  et al.  Science in China B (Zhongguo Kexue), 1995, 38 (5): 449.
[14] Osseo-Asare K. Advances in Colloid and Interface, 1991, 37: 123.
[15] Paatero E, Ernola P, Sjoblom J et al. ISEC'88, Moscow, 1988, 124.
[16] Paatero E, Lantto T, Ernola P.  Solvent Extraction Ion Exchange, 1990, 8 (3): 371.
[17] Hu Z, Xin H, Pan Y  et al.  Applied Chemistry (Yingyong Huaxue), 1995, 12 (5): 10.
[18] Fu X, Hu X, Hu Z  et al.  Colloids and Surfaces, 1999, 152: 335.
[19] Hu Z, Hu X, Cui W  et al.  Colloids and Surfaces, 1999, 155: 383.
[20] Fu X, Liu H, Xue M  et al.  Eng. Chem. & Met. (Huagong Yejin), 1999, 20 (1): 5.
[21] Fu X, Liu H, Li Z  et al.  Solvent Extraction Ion Exchange, 1999, 17 (5): 1281.
[22] Hu Z, Pan Y, Ma W  et al.  Solvent Extraction Ion Exchange, 1995, 13 (5): 965.
[23] Yang C, Hong B, Chen J.  Powder Tech., 1996, 89: 149.

[ Back ] [ Home ] [ Up ] [ Next ] Mirror Site in USA  Europe  China  CSTNet ChinaNet