Anal. Chem. 1996, 68, 758-762
Ion-Pair Extraction of Metalloporphyrins into Acetonitrile for Determination of Copper(II) Masaaki Tabata,* Midori Kumamoto, and Jun Nishimoto
Department of Chemistry, Faculty of Science and Engineering, Saga University, 1, Honjo-machi, Saga, 840 Japan
Equilibrium study of ion-pair extraction of a cationic water-soluble porphyrin [5,10,15,20-tetrakis(1-methylpyridinium-4-yl)porphyrin, H2tmpyp4+] and its metalloporphyrins (MP) into the acetonitrile layer, separated by addition of sodium chloride (4.00 mol dm-3) to a 1:1 (v/v) acetonitrile-water mixed solvent, was carried out to develop a new and useful method for the determination of a subnanogram amount of copper(II). M denotes Zn2+, Cu2+, Co3+, Fe3+, and Mn3+, and P2- is porphyrinate ion. The extraction and dissociation constants of the ion-pair complexes, defined by Kex ) [MP(ClO4)4]org[MP4+]aq-1[ClO4-]aq-4, Kdis,1 ) [MP(ClO4)3+]org[ClO4-]org[MP(ClO4)4]org-1, and Kdis,2 ) [MP(ClO4)22+]org[ClO4-]org[MP(ClO4)3+]org-1, were determined by taking into account the partition constant of sodium perchlorate (KD ) 1.82 ( 0.01). The equilibrium constants were found to be KexKdis,1 ) (7.2 ( 1.3) × 104, (6.4 ( 0.9) × 104, (1.35 ( 0.13) × 105, (4.8 ( 0.6) × 103, (1.23 ( 0.05) × 104, and (1.42 ( 0.07) × 103 at 25 °C for the free base porphyrin (H2tmpyp4+) and the metalloporphyrins of zinc(II), copper(II), cobalt(III), iron(III), and manganese(III), respectively. The Kdis,2 values were (2.9 ( 1.4) × 10-2, (3.1 ( 1.1) × 10-2, (8.0 ( 4.9) × 10-3, and (5.1 ( 2.2) × 10-2 for the free base porphyrins and the metalloporphyrins of zinc(II), copper(II), and cobalt(III), respectively. The results were developed for determination of a trace amount of copper(II) (3 × 10-8-4 × 10-6 mol dm-3) in natural water samples using H2tmpyp4+ with a molar absorptivity of 3.1 × 105 mol-1 dm3 cm-1 at a precision of 1.3% (RSD). The determination of copper(II) was not interfered by the presence of 10-4 mol dm-3 of Mn2+, Co2+, Ni2+, Hg2+, Cd2+, Ag+, Cr3+, V5+, Al3+, Mg2+, Ca2+, Br-, I-, SCN-, and S2O32- and 10-5 mol dm-3 of Fe3+, Zn2+, and Pd2+. A phase separation occurs from a mixed solvent of water and water-miscible organic solvent like acetonitrile upon addition of electrolyte to the mixed solvents, i.e., salting-out, due to decreased solubility of the organic solvent in aqueous solution.1 The separated organic solvent contains water and salts, resulting in large donor and acceptor abilities compared to those of the corresponding pure organic solvent.2 Thus, the solvent can easily extract ion-pair complexes such as tris(2,2′-bipyridine)cobalt(II) chloride3 and cadmium(II) iodide.4 Recently, the solvents have (1) Setchenov, J. Z. Phys. Chem. 1889, 4, 117. (2) Tabata, M.; Kumamoto, M.; Nishimoto. J. Anal. Sci. 1994, 10, 383. (3) Nagosa, Y. Anal. Chim. Acta 1980, 120, 279. (4) Fujinaga, T.; Nagoaa, Y. Bull. Chem. Soc. Jpn. 1980, 53, 416.
758 Analytical Chemistry, Vol. 68, No. 5, March 1, 1996
Figure 1. 5,10,15,20-Tetrakis(1-methylpyridinium-4-yl)porphyrin, H2tmpyp4+.
been used in HPLC.5-7 A further advantage is a possibility of highly charged solutes that are not extracted into normal organic solvents like chloroform. We found that a porphyrin with four positive charges [5,10,15,20tetrakis(1-methylpyridinium-4-yl)porphyrin, H2tmpyp4+ (Figure 1)] and its metalloporphyrins were extracted into the acetonitrile layer separated by salting-out using sodium chloride in the presence of perchlorate ion, even though the complexes were not extracted into chloroform and 1,2-dichloroethane. Furthermore, large molar absorptivities of porphyrins ((1-5) × 105 mol-1 dm3 cm-1) allow determination of trace amounts of metal ions.8 Only two papers, however, have been reported for the extraction of porphyrins and metalloporphyrins using surfactants due to the slight solubility of porphyrins in organic solvents.9,10 The present paper clarifies the extraction mechanism of ion-pair complexes of H2tmpyp4+ or its metalloporphyrins with perchlorate in acetonitrile to develop a new and useful method for determination of copper(II) in natural water samples. EXPERIMENTAL SECTION Reagents. Acetonitrile was purified by distillation in the presence of molecular sieves. The metalloporphyrins of copper(II), zinc(II), manganese(III), iron(III), and cobalt(III) were synthesized by the reaction of H2tmpyp4+ with an excess of metal chlorides and precipitated as the perchlorates.11 The perchlorates were washed with a small amount of perchloric acid solution (10-2 mol dm-3) and then passed through an anion-exchange resin of chloride form. The eluate containing the metalloporphyrin was evaporated and dried in a desiccator for 1 month. Hydroxyl(5) Mueller, B. J.; Lovett, R. J. Anal. Chem. 1987, 59, 1405. (6) Leggett, D. C.; Jenkins, T. F.; Miyares, P. M. Anal. Chem. 1990, 62, 1355. (7) Janjic, T. J.; Zivkovic,V.; Celap, M. B. Chromatographia 1994, 38, 447. (8) Tabata, M.; Tanaka, M. Trends Anal. Chem. 1991, 10, 128. (9) Igarashi, S.; Yotsuyanagi, T. Mikrochim. Acta 1992, 106, 37. (10) Horvath, W. J.; Hurie, C. W. Talanta 1992, 39, 487. (11) Pasternack, R. F.; Gibbs, E. J.; Villafranca, J. J. Biochemistry 1983, 22, 2406. 0003-2700/96/0368-0758$12.00/0
© 1996 American Chemical Society
Table 1. Data of Phase-Separation by Salting-Out Using Sodium Chloridea acetonitrile phase volume/cm3 NaCl/mol dm-3 H2O/mol dm-3 CH3CN/mol dm-3 ET(30)/kJ mol-1 b DII,I/kJ mol-1 c
4.17 2.49 × 10-2 4.53
aqueous phase 6.11 3.27 3.47
228 48.0
60.6
a Initially, 5 cm3 each of water and acetonitrile were taken, and 1.169 g (2.00 × 10-2 mol) of sodium chloride was added. b The value was determined from the wavelength of absorption maximum of 2,6diphenyl-4-(2,4,6-triphenylpyridinio)phenolate (DTP) in the solvent. c The value was determined from the difference in the wavelengths of two absorption maxima of bis(2,4-pentanediolato)vanadium(IV) (VO(acac)2) in the solvents.
ammonium sulfate was recrystallized from hot water. All other chemicals were used without further purification. Procedure. Sodium chloride, 1.169 g (2.00 × 10-2 mol), was added to a stoppered graduated tube containing a 5-cm3 volume of aqueous sample solution. The sodium chloride was dissolved sufficiently, a 5-cm3 volume of organic solvent added, and the mixed solvent shaken for about 1 min. The two phases were allowed to stand for a few minutes. The volumes of the organic and the aqueous phases were measured, and the concentrations of chloride ion and water in the separated organic phase were determined by argentometry and Karl Fisher titration on a Karl Fisher moisture titrator (MKL-200, Kyoto Electronics, Japan), respectively. Absorption spectra of the porphyrin and the metalloporphyrins were recorded on a Shimazu UV-vis spectrophotometer 2100. The pH of the aqueous phase after salting-out was measured on a Radiometer Ion-85 analyzer. A 1.000 × 10-2 mol dm-3 perchloric acid aqueous solution containing 3.27 mol dm-3 NaCl and 3.47 acetonitrile, which are the same concentrations as were found in the aqueous phase after salting-out, was used as a standard hydrogen ion concentration (-log [H+] ) 2.000). From pH meter readings in various hydrogen ion concentrations at 3.27 mol dm-3 of sodium chloride and 3.47 mol dm-3 of acetonitrile, the pH meter and electrode were calibrated in terms of -log [H+]. All experiments were carried out at 25 °C. RESULTS AND DISCUSSION Composition of Acetonitrile Phase. The volumes and compositions of the aqueous and acetonitrile phases after saltingout are given in Table 1. The acetonitrile phase contains a large amount of water (4.63 mol dm-3). The values of ET(30) and DII,I were determined to identify the donor and acceptor abilities of the separated solvents using 2,6-diphenyl-4-(2,4,6-triphenylpyridinio)phenolate (DTP)12 and bis(2,4-pentanediolato)vanadium(IV) (VO(acac)2),13 respectively. The ET(30) and DII,I values of the acetonitrile phase are larger than those of nitrobenzene and indicate the high donor and acceptor abilities due to a large amount of water dissolved in the solvent. This is a unique characteristic of the solvents separated by salting-out compared to other organic solvents. The high polarity of the solvent implies a possibility of extracting highly charged chemical species that (12) Reichardt, C. Solvents and Solvent Effects in Organic Chemistry; VCH: Weinheim, 1988. (13) Sone, K.; Fukuda, Y. Inorganic Thermochromism; Springer: New York, 1987.
Figure 2. Effect of the concentration of salting-out agents of (a) sodium chloride, (b) ammonium sulfate, (c) magnesium sulfate on the volume of organic phase.
are not extracted in conventional organic solvents like 1,2dichloroethane. Effect of Salting-Out Agents. The phase separation depends on the salting-out agents. The typical change in the volume of organic phase separated by salting-out is given in Figure 2 for salting-out agents of sodium chloride, ammonium sulfate, and magnesium sulfate. The effect of salting-out decreases in the following order: Al3+ > Mg2+ > Ca2+ > Ba2+ . Na+ > K+ > NH4+ > Li+, and SO42- > Cl-. The main driving force of the salting-out is hydration of these cations and anions. The hydration energies (kJ mol-1) of these ions are 4612 (Al3+), 2198 (Mg2+), 1592 (Ca2+), 1317 (Ba2+), 501.8 (Na+), 424.5 (K+), and 2198 (Li+).14 Except for lithium ion, the salting-out is correlated to the hydration energies of the ions. The small salting-out effect of lithium ion may be due to the large solubility of lithium chloride in acetonitrile. Interestingly, magnesium sulfate led to a salting-in effect for the 1:1 (v/v) water-acetonitrile mixed solvent. Effect of Solvent Volume Ratio on the Phase Separation. The salting-out experiments were carried out at various volume ratios of acetonitrile to aqueous solution for the salting-out agents of sodium chloride (4.0 mol dm-3), lithium chloride (2 mol dm-3), and ammonium sulfate (0.50 mol dm-3). Once phase separation occurs, the volume of acetonitrile phase increases linearly with the initial volume of the acetonitrile. The intercept of the straight line on the abscissa corresponds to the solubility of acetonitrile in the aqueous solution containing the salting-out agent. The linear increase of the volume of the phase-separated acetonitrile can be expected from the Setchenov equation: log S0/S ) ksCs, where S0, S, ks, and Cs are the solubilities of the solvent in water and aqueous salt solution, the salting-out coefficient, and the concentration of the aqueous salt solution, respectively.1 Since the concentration of an electrolyte in the aqueous phase is a given value, the solubility of acetonitrile in aqueous solution becomes constant after the phase separation. Therefore, the volume of the acetonitrile phase increased linearly with the initial volume of acetonitrile, and almost the same slope was obtained for each aqueous electrolyte solution. Partition of Sodium Perchlorate. Before analyzing the extraction of the porphyrin and the metalloporphyrins in acetonitrile using sodium perchlorate, it is necessary to determine the partition constant of sodium perchlorate between the acetonitrile and aqueous phases in the presence of sodium chloride. The concentration of perchlorate in acetonitrile was determined by the (14) Rosseinsky, D. R. Chem. Rev. 1965, 65, 467.
Analytical Chemistry, Vol. 68, No. 5, March 1, 1996
759
Figure 3. Distribution of sodium perchlorate into acetonitrile phase separated by salting-out of sodium chloride ([NaCl]0 ) 4.00 mol dm-3).
methylene blue method.15 Volumes of 1 cm3 of a methylene blue solution (3.91 × 10-4 mol dm-3), 0.5 cm3 of sulfuric acid (5.0 × 10-2 mol dm-3), and a sample solution containing perchlorate ion were taken into a separating funnel, and then 10 cm3 of 1,2dichloroethane was added and the mixture shaken well for 30 min. The 1,2-dichloroethane phase was transferred to a test tube with a stopper, and 0.5 g of anhydrous sodium sulfate was added to the 1,2-dichloroethane phase. The absorbance of the 1,2dichloroethane solution was measured at 655 nm and the concentration of perchlorate ion determined. Considering extraction of sodium perchlorate as an ion-pair complex and dissociation of the sodium perchlorate in the acetonitrile, we have the following equation for the partition of sodium perchlorate:
CClO4,org
DClO4 )
CClO4,aq
[NaClO4]org + [ClO4-]org ) [ClO4-]aq
) KD,ClO4(Kdis,NaClO4-1[Na+]org + 1)
(1)
where KD ) [ClO4-]org[ClO4-]aq-1 and Kdis,NaClO4 ) [Na+]org[ClO4-]org[NaClO4]org-1. Equation 1 can be rearranged to
CClO4,org ) KD,ClO4(Kdis,NaClO4-1[Na+]org + 1)CClO4,aq (2) The plot of CClO4,org vs CClO4,aq gave a straight line with zero intercept (Figure 3): Kdis,NaClO4-1[Na+]org , 1, and sodium perchlorate dissociates completely in the acetonitrile phase. The partition constant, KD, was found to be 1.82 ( 0.02 from the slope. The value is much larger than those observed in nitrobenzene (log D ) -4 at 0.1 mol dm-3 NaClO4)16 and other solvents (log D ) -1.2, -2, -2.5, and -2.24 at 0.1 mol dm-3 NaClO4) for tributyl phosphate, nitromethane, methyl isobutyl ketone, and 2-pentanone, respectively).17,18 The complete dissociation of sodium perchlorate is ascribed to the large content of water in the acetonitrile phase. The concentrations of sodium chloride were (15) Iwasaki, I.; Utsumi, S.; Kang. C. Bull. Chem. Soc. Jpn. 1963, 36, 325. (16) Sekine, T.; Dyrssen, D. Anal. Chim. Acta 1969, 45, 433. (17) Hasegawa, Y.; Ishii, T.; Sekine, T. Bull. Chem. Soc. Jpn. 1971, 44, 275. (18) Yamada, H.; Takahashi, K.; Fujii, Y.; Mizuta, M. Bull. Chem. Soc. Jpn. 1984, 57, 2847.
760 Analytical Chemistry, Vol. 68, No. 5, March 1, 1996
Figure 4. Change in apparent molar absorptivities (j) of (a) H2tmpyp4+, (b) [Zn(tmpyp)]4+, (c) [Cu(tmpyp)]4+, (d), [Co(tmpyp)Cl]4+, (e) [Mn(tmpyp)Cl]4+, and (f) [Fe(tmpyp)Cl]4+ extracted into the organic phase at various concentrations of sodium perchlorate at pH 5.4. The solid lines were calculated from the values in Table 2.
2.49 × 10-2 and 3.27 mol dm-3 in the acetonitrile and aqueous phases, respectively. Thus, the partition constant of sodium chloride was 7.61 × 10-3. Extraction of H2tmpyp4+, [MII(tmpyp)]4+, and [MIII(tmpyp)Cl] 4+ (M ) Zn, Cu, Co, Fe, and Mn) in Acetonitrile. The above cationic porphyrin and metalloporphyrins were easily extracted into the acetonitrile layer, separated by salting-out in the presence of sodium perchlorate at pH 4.8 (acetate buffer, 10-2 mol dm-3). Typical extraction data are given as a function of the concentration of perchlorate in aqueous solution in Figure 4. H2tmpyp4+, [Cu(tmpyp)]4+, and [Zn(tmpyp)]4+ were quantitatively extracted into the acetonitrile at 0.08 mol dm-3 sodium perchlorate, and [MnIII(tmpyp)Cl]4+ and [CoIII(tmpyp)Cl]4+ were extracted at perchlorate concentrations >0.3 mol dm-3. The metalloporphyrins of iron(III), cobalt(III), and manganese(III) have an octahedral structure, with water molecules in the axial positions. Thus, the metalloporphyrins of the trivalent metal ions were difficult to extract into the acetonitrile phase. On the other hand, [Cu(tmpyp)]4+ is square planar, with no bound water molecule.19 In Figure 5, the log value of absorbance ratio of the organic and aqueous phases of porphyrins is plotted against the concentration of perchlorate ion in aqueous phase. The extraction of the porphyrins decreases in the order [Cu(tmpyp)]4+ > H2tmpyp4+ ) [Zn(tmpyp)]4+ > [FeIII(tmpyp)Cl]4+ > [CoIII(tmpyp)Cl]4+ > [MnIII(tmpyp)Cl]4+. The slopes of the straight lines are 2.75 ( 0.07, 2.56 ( 0.15, 2.59 ( 0.06, 3.01 ( 0.06, 2.82 ( 0.05, and 3.01 ( 0.04 for [Cu(tmpyp)]4+, [H2tmpyp]4+, [Zn(tmpyp)]4+, [FeIII(tmpyp)Cl]4+, [CoIII(tmpyp)Cl]4+, and [MnIII(tmpyp)Cl]4+, respectively. The results indicate the binding of two or three perchlorate ions to the porphyrin in the acetonitrile phase. Thus, the partition of the free base porphyrin or the metalloporphyrins (MP4+) between the aqueous and acetonitrile phases is given by the following equation in the presence of perchlorate ions (X-): (19) Tabata, M.; Ozutsumi, K. Bull. Chem. Soc. Jpn. 1994, 67, 1608.
[MPX4]org
Kdis,1
Kdis,2
[MPX3]org+ + X–org
[MPX2]org2+ + 2X–org (3)
Kex
[MP]aq4+ + 4X–aq
where X- denotes perchlorate ion and Kex ) [MPX4]org[MP4+]aq-1[X-]aq-4, Kdis,1 ) [MPX3+]org[X-]org[MPX4]org-1, and Kdis,2 ) [MPX22+]org[X-]org[MPX3+]org-1. The apparent molar absorptivity (j) of the porphyrins in the organic phase is correlated to the above Kex, Kdis,1, and Kdis,2 as follows:
j )
Absorg ) CMP 1[MPX3+]org + 2[MPX22+]org [MP4+]aq(VaqVorg-1) + [MPX3+]org + [MPX22+]org
(4)
) 1KexKdis,1KD[X-]3aq + 2KexKdis,1Kdis,2[X-]2aq KD2(VaqVorg-1) + KexKdis,1KD[X-]3aq + KexKdis,1Kdis,2[X-]2aq
Figure 5. Plot of logarithmic value of absorption ratio of organic phase to aqueous phase versus the concentration of perchlorate ion in the aqueous phase for the distribution of (b) [Cu(tmpyp)4+], (O) H2tmpyp4+, (9) [Zn(tmpyp)]4+, (0) [Fe(tmpyp)Cl]4+, (2) [Co(tmpyp)Cl]4+, and (4) [Mn(tmpyp)Cl]4+. Table 2. Distribution Constants of Free Base and Metalloporphyrins of H2tmpyp4+
(5)
The values of Kex, Kdis,1, and Kdis,2 were determined using a leastsquares minimization program, assuming 1 ) 2. Since the extracted chemical species [MPX4] completely dissociates to [MPX3+], the Kdis,1 value could not be determined. The extraction and dissociation constants determined are summarized in Table 2. The solid lines in Figure 4 were calculated using the determined values. Effect of pH on Extraction of the Porphyrin and Metalloporphyrins. Copper(II), manganese(III), and iron(III) porphyrins were quantitatively extracted in acetonitrile at a wide pH range (1-12). But the free base porphyrin (H2tmpyp4+) and [Zn(tmpyp)]4+ were not extracted into acetonitrile at pH < 2 due to protonation of H2tmpyp4+ 20 and dissociation of [Zn(tmpyp)]4+ to zinc(II) ion,21 respectively. Extraction of iron(III) and cobalt(III) porphyrins decreases gradually at higher pH due to formation of hydroxo- and dihydroxometalloporphyrins like [CoIII(tmpyp)OH]4+ and [CoIII(tmpyp)(OH)2]3+.22 From the higher extraction constants of [Cu(tmpyp)]4+ and the lower extraction constants of H2tmpyp4+ and zinc(II) at pH 1, we infer the possibility of separation analysis of [Cu(tmpyp)]4+ from H2tmpyp4+ and [Zn(tmpyp)]4+ and confirmed the following determination method for copper(II) based on the present phase separation phenomenon by salting-out. Determination of Copper(II) by Extraction in Acetonitrile. (a) Calibration Curve. The above results indicate a quantitative extraction of [Cu(tmpyp)]4+ into the acetonitrile layer, and that H2tmpyp4+ was hardly extracted at all into the acetonitrile at pH
