- Email: [email protected]
Extraction of arsenic(III) from chloride-iodide solutions by diphenyl(2-pyridyl)methane and benzene
0039-9140/x5$3 00 + 0 00 Pergamon Press Ltd
SHORT
COMMUNICATIONS
EXTRACTION OF ARSENIC(II1) FROM CHLORIDE-IODIDE SOLUTIONS BY DIPHENYL(2-PYRIDYL)METHANE AND BENZENE M. Department
of Chemical
EJAZ
and E. SIDDIQUE
Engineermg, College of Engineermg. Kmg Abdulaziz P.O. Box 9027, Jeddah, Saudi Arabia
Umversity,
and SUHAIL AHMED Pakistan
Institute
(Received
of Nuclear
1 December
Science and Technology,
1983
Ret-ised 22 April
Nilore,
Rawalpindi,
Paktstan
1985. Accepted 9 Augusr 1985)
Summary-The variation of the partition coefficient of arsenic(II1) between chloride-iodide solutions and diphenyl(2-pyridyl)methane in benzene has been studied. The effect of the concentration of hydrochloric acid and iodide in the aqueous phase has been assessed. The partition coefficients are maximal for concentrated acid solutions which are 0.024. I M m potassium iodide. Slope-analysis studies were used to elucidate the composition of the extracted species. Polymerization of the solvent species tends to coefficients of decrease the distribution arsenic with increasing concentration of diphenyl(2_pyridyl)methane, especially with trace concentrations of the element. Arsenic can be selectively separated from copper, cobalt, nickel, iron. chromium and antimony, which are usually associated with it in various ores.
I 5 have shown that arsenic can Earlier investigations be extracted as a chloro-complex by conventional solvents and Lewis-base extractants, but the distribution coefficients are too low for complete extraction in a single step. In an attempt to find a better system, we studied6 the extraction from chloride-thiocyanate medium with 0.1 M diphenyl(2-pyridyl)methane (DPPM) in benzene, but quantitative extraction still needed several equilibration steps. In the present study we examined the use of the weakly basic iodide ion for complexation. The results show that a highly selective and quantitative extraction can be achieved in a single equilibration step from 9M hydrochloric acid/0.02-1 M potassium iodide medium with 0.01 M DPPM in benzene. Under the experimental conditions for the extraction, several metal ions which are known to form anionic chloride complexes at this acidity are very poorly extracted. They can be removed almost completely from the organic phase in two or three scrub stages, which will not affect the recovery of arsenic, because of its very high extraction coefficients. EXPERIMENTAL Reagents and tracers
DPPM was obtained commercially (Aldrich) and used without further purification. Its characteristics have been reported elsewhere.7 Hydrochloric acid solutions were prepared either from BDH volumetric solution ampoules or from the Merck “pro annlysi” grade acid, and standardized if necessary. Potassium Iodide solutions were prepared from the analytical grade salt and demineralized doubly distilled water, and were kept in dark bottles.
Arsenic-76 (t, z 26.3 hr) and other tracers used were obtained by neutron irradiation of the pure metals or their oxides in the PARR-I research reactor of the Pakistan Institute of Technology or obtamed from the Radiochemical Centre, Amersham. Some were separated from the parents without a carrier. Apparatus
The radiochemical purity of the tracers was checked by gamma spectrometry with a 30-cm’ Ge(Li) detector in conjunction with a Nuclear Data model ND-4410 computerized analyser system. y-Ray count-rates were determined with a Nuclear Chicago single-channel analyser, model-872. coupled with a 7.5 x 7.5 cm thallium-doped sodium iodide well-type y-ray scintillation counter. Solid /I-emitting samples were assayed with an end-window Geiger assembly equipped with a G.E.C. tube, type EHM/2/S. Derermmarion
of disrriburion coefficients
Extractions were done at room temperature (23 + 3 ) ,in 20-ml glass vials: 1 ml of mineral acid contaimng “‘As or other element (activity _ IO4 counts. see-’ ml-‘) was put in the vial and potassium iodide solution was added to adjust the iodide concentration to the required level. The aqueous phase was then shaken for 5 min with an equal volume of organic phase (usually 0. I or 0.01 M DPPM in benzene) the phases were separated, and the activities measured. The concentration of arsenic in the original aqueous phase was < 10-5M. The organic phases were not pre-equilibrated with the aqueous phase except in solvent-effect studies.
RESULTS AND DISCUSSION
DPPM has been found to behave reagent6 and it is thought it also acts forms cations in acid solutions. The ion-association complexes is rather I055
as a solvating as a base7 and stability of its poor. Accord-
10%
SHORT
COMMUNICATIONS
(B)
(A)
-1
,L_/L
0
log
(C)
1
-2
[l-cl]
-2
-1
-,
log
[KI]
log
[K;]
Fig. 1. (A) Distribution coefficients of arsenic(W) for a. extraction from hydrochloric acid by 0 I M DPPM; h, organic phase 0. I M DPPM, aqueous phase 0.02M KI in hydrochlorrc acid; c, orgamc phase 0.01 M DPPM, aqueous phase 0.02M KI in hydrochloric acid. (B) Effect of the concentration of potassium iodide on the extraction of arsenic by 0.01 M DPPM/benzene: [HCI]: a, 7M: h, 8M; c. 9M; d IOM. (C) Effect of the concentration of potassium Iodide on the extraction of arsenic by 0.01 M DPPM/benzene: [HCI]: a, 7M; b, 8M; c, 9M; d. IOM.
ingly the extraction of tracer arsenic with DPPM (O.lM) from hydrochloric acid (Fig. lA,a) unlike that with ahphatic amines and quaternary ammonium salts,’ is almost negligible. This is evidently because of the relatively low p/c,,+ value of DPPM. The increase in extraction with increase in hydrochloric acid concentration could be due to the increased formation of extractable chloride complexes of the type AsCI; and/or the salting-out effect of the acid. The extraction of the element from 0.02M potassium Iodide in hydrochloric acid was then investigated in line with some of our previous mvestigations.“.” The results are presented in Fig. lA,b. The extraction from dilute acid solutions, like that in the absence of iodide, is very poor, but there is a very large increase concentration. Previous increasing acid with investigation” has shown that in the presence of trace concentrations of metals, this extractant has a tendency to polymerize, especially in concentrated hydrochloric acid. and hence the extraction sometimes decreases” with increasing reagent concentration. We therefore examined the effect of using 0.01 M DPPM
instead of 0.1 M and found little change m the distribution data (Fig. lA,c). Figures IB and 1C show the effect of Iodide on the extraction of tracer arsenic from different concentrations of hydrochloric acid by 0.1 M and O.OlM DPPM in benzene, respectively. The D values are higher at high iodide concentrations when O.OlM DPPM is used. The slope of the lines (Fig. IC) is close to unity for [KI] < O.O5M, indicating the involvement of one iodide ion per complex anion formed. These results led us to investigate the solvent effect. In every case the organic phase was pre-equilibrated with the corresponding aqueous phase. These studies were made with O.lM potassium iodide at four concentrations of hydrochloric acid (7, 8, 9 and IOM). The results are presented in Fig. 2. In all cases, curves were obtained with maxima at around 0.01 M DPPM. These curves also indtcate that 9M hydrochloric acid is the optimal concentration for maximum extraction of tracer arsenic. We attribute the decrease in extraction at higher DPPM concentrations to polymerization of protonated DPPM and
SHORT
1057
COMMUNICATIONS
“r
Table I. Distributton coefficients and separation factors [with respect to As(lIl)] for extraction from 0. I A4KI in 9M HCI with O.OlM DPPM in benzene Species
As(III) Na+ K+ cs+ Br-
co’+
Ni’t
CU’+ Zn’+ Cd’+ ;$I’
Fig. 2. The effect of the concentration of DPPM on the distribution coefficient of arsenic from different concentrations of hydrochloric acid: [HCI]: a, 7M; b, 8M; c, 9M; d, IOM.
formation of iodide ion-association complexes with it. Although this work was devoted mainly to tracer studies, an attempt was made to elucidate the composition of the extractable species by loading-ratio data. For this purpose we used a solution of arsenious oxide in 0.1M potassium iodide/9M hydrochloric acid medium and extracted it with O.OlM DPPM in benzene. It was found that the organic phase was saturated at an arsenic concentration of around 0.75 g/l., which confirms a 1: 1 DPPM : arsenic ratio. Comparison with our previous work6 shows that this system is much better than the corresponding thiocyanate system, where the maximum D value was only around 4. In the present case it is almost two orders of magnitude higher and a single equilibration is sufficient for quantitative extraction. To study the selectivity, we investigated the partition behaviour of several other elements under the conditions optimal for extraction of arsenic, i.e., 0.1 M DPPM/benzene, and 0.1 M potassium iodide in 9M hydrochloric acid. The results are shown in Table 1. It is seen that the extraction of arsenic is very selective. Selenium and mercury are two common toxic elements which are co-extracted and the method can be used for their preconcentration. The very high distribution ratios for arsenic also ensure its extrac-
Sr2+ Fe(II1) Au(II1) SC’+ EU’f Ce’+ Cr’+ Se(IV) Sn(IV) Hf(IV) Ru(IV) Sb(V) Cr(V1) Mo(V1) Tc(VI1)
Concentration, M
10-s 1O-4 IO J 10-O 10.” IO J IO ’ IO-5 10-J IO_’ IO_’ IO-6 10-5 IO_ 5 10-d lo-4 10-4 IO-5 ‘C.F. IO-4 10-d 1om4 10-S 1om5 *C.F. 10-5 *C.F.
Distribution coeffictent, D
441 0.0 0.0 0.0 0.01 00 0 03 0. I5 0.02 0 70 99 001 0.01 0.22 6.10 0.01 00 001 0.0 32 0.01 0.01 2.80 0.49 0.0 I2 25
Separation factor
> IOJ > IO4 > IO - 1o4 1 IO’ - lOA > IO’ - IOJ 6.3 X IO’ 447 - IO’ - IOJ - IO 72 - IO) > IOJ - IOJ > IOJ I3 - IO” - IO” I57 9x 102 > I04 35 I7
*C.F. = Carrier-free. tion from large volumes of aqueous phase. Thus under optimum conditions arsenic is more than 99% extracted when the ratio of the organic to aqueous phase volume is I : 20. REFERENCES I. G. 0. Brink, P. Kafalas, R. A. Sharp, E. L. Wetss and J. W. Irvine, J. Am. Chem. Sot., 1957. 79, 1303. 2. W. Ftscher and W. Harre. Anger. Chem.. 1954.66, 165. 3. H. C. Beard and L. A. Lyerly, Anal. Chem., 1961, 33, 1781. 4. V. A. Orlova, B. Ya. Spivakov, N. A. Sharova and T. M. Malyutina, Zh. Analit. Khim.. 1977, 32, 1591. 5. E. S. Donaldson, Tulunru, 1977. 24, 105. 6. M. Ejaz, S. M. Hasany and Z. Umar, Sepn. Ser. Technol.. 1979, 14, 431. 7. M. Ejaz, M. Iqbal, S. A. Chaudhrt and Zamiruddin. J. Radroanal. Chem., 1978. 42, 33.5. 8. M. Ejaz, Shamus-ud-Zuha, S. Ahmad, M. S. Chaudhary and M. Rashid, Mikrochim. Acta, 1980, 1, 7. 9. Y. Marcus and A. S. Kertes, ion Exchange and Solt~enr Extraction qf Metal Complexes, p. 961. Wiley, New York, 1969. IO. M. Ejaz, Anal. Chem., 1978, 90, 740. II. M. Ejaz, M. A. Quresht and Shamus-ud-Zuha, Sepn. Ser. Technol.. 1981, 16, 291. 12. M. Ejaz, S. M. Hasany and I. Hanif. J. Less-Common Metals, 1981, 77, 157.
SHORT
COMMUNICATIONS
EXTRACTION OF ARSENIC(II1) FROM CHLORIDE-IODIDE SOLUTIONS BY DIPHENYL(2-PYRIDYL)METHANE AND BENZENE M. Department
of Chemical
EJAZ
and E. SIDDIQUE
Engineermg, College of Engineermg. Kmg Abdulaziz P.O. Box 9027, Jeddah, Saudi Arabia
Umversity,
and SUHAIL AHMED Pakistan
Institute
(Received
of Nuclear
1 December
Science and Technology,
1983
Ret-ised 22 April
Nilore,
Rawalpindi,
Paktstan
1985. Accepted 9 Augusr 1985)
Summary-The variation of the partition coefficient of arsenic(II1) between chloride-iodide solutions and diphenyl(2-pyridyl)methane in benzene has been studied. The effect of the concentration of hydrochloric acid and iodide in the aqueous phase has been assessed. The partition coefficients are maximal for concentrated acid solutions which are 0.024. I M m potassium iodide. Slope-analysis studies were used to elucidate the composition of the extracted species. Polymerization of the solvent species tends to coefficients of decrease the distribution arsenic with increasing concentration of diphenyl(2_pyridyl)methane, especially with trace concentrations of the element. Arsenic can be selectively separated from copper, cobalt, nickel, iron. chromium and antimony, which are usually associated with it in various ores.
I 5 have shown that arsenic can Earlier investigations be extracted as a chloro-complex by conventional solvents and Lewis-base extractants, but the distribution coefficients are too low for complete extraction in a single step. In an attempt to find a better system, we studied6 the extraction from chloride-thiocyanate medium with 0.1 M diphenyl(2-pyridyl)methane (DPPM) in benzene, but quantitative extraction still needed several equilibration steps. In the present study we examined the use of the weakly basic iodide ion for complexation. The results show that a highly selective and quantitative extraction can be achieved in a single equilibration step from 9M hydrochloric acid/0.02-1 M potassium iodide medium with 0.01 M DPPM in benzene. Under the experimental conditions for the extraction, several metal ions which are known to form anionic chloride complexes at this acidity are very poorly extracted. They can be removed almost completely from the organic phase in two or three scrub stages, which will not affect the recovery of arsenic, because of its very high extraction coefficients. EXPERIMENTAL Reagents and tracers
DPPM was obtained commercially (Aldrich) and used without further purification. Its characteristics have been reported elsewhere.7 Hydrochloric acid solutions were prepared either from BDH volumetric solution ampoules or from the Merck “pro annlysi” grade acid, and standardized if necessary. Potassium Iodide solutions were prepared from the analytical grade salt and demineralized doubly distilled water, and were kept in dark bottles.
Arsenic-76 (t, z 26.3 hr) and other tracers used were obtained by neutron irradiation of the pure metals or their oxides in the PARR-I research reactor of the Pakistan Institute of Technology or obtamed from the Radiochemical Centre, Amersham. Some were separated from the parents without a carrier. Apparatus
The radiochemical purity of the tracers was checked by gamma spectrometry with a 30-cm’ Ge(Li) detector in conjunction with a Nuclear Data model ND-4410 computerized analyser system. y-Ray count-rates were determined with a Nuclear Chicago single-channel analyser, model-872. coupled with a 7.5 x 7.5 cm thallium-doped sodium iodide well-type y-ray scintillation counter. Solid /I-emitting samples were assayed with an end-window Geiger assembly equipped with a G.E.C. tube, type EHM/2/S. Derermmarion
of disrriburion coefficients
Extractions were done at room temperature (23 + 3 ) ,in 20-ml glass vials: 1 ml of mineral acid contaimng “‘As or other element (activity _ IO4 counts. see-’ ml-‘) was put in the vial and potassium iodide solution was added to adjust the iodide concentration to the required level. The aqueous phase was then shaken for 5 min with an equal volume of organic phase (usually 0. I or 0.01 M DPPM in benzene) the phases were separated, and the activities measured. The concentration of arsenic in the original aqueous phase was < 10-5M. The organic phases were not pre-equilibrated with the aqueous phase except in solvent-effect studies.
RESULTS AND DISCUSSION
DPPM has been found to behave reagent6 and it is thought it also acts forms cations in acid solutions. The ion-association complexes is rather I055
as a solvating as a base7 and stability of its poor. Accord-
10%
SHORT
COMMUNICATIONS
(B)
(A)
-1
,L_/L
0
log
(C)
1
-2
[l-cl]
-2
-1
-,
log
[KI]
log
[K;]
Fig. 1. (A) Distribution coefficients of arsenic(W) for a. extraction from hydrochloric acid by 0 I M DPPM; h, organic phase 0. I M DPPM, aqueous phase 0.02M KI in hydrochlorrc acid; c, orgamc phase 0.01 M DPPM, aqueous phase 0.02M KI in hydrochloric acid. (B) Effect of the concentration of potassium iodide on the extraction of arsenic by 0.01 M DPPM/benzene: [HCI]: a, 7M: h, 8M; c. 9M; d IOM. (C) Effect of the concentration of potassium Iodide on the extraction of arsenic by 0.01 M DPPM/benzene: [HCI]: a, 7M; b, 8M; c, 9M; d. IOM.
ingly the extraction of tracer arsenic with DPPM (O.lM) from hydrochloric acid (Fig. lA,a) unlike that with ahphatic amines and quaternary ammonium salts,’ is almost negligible. This is evidently because of the relatively low p/c,,+ value of DPPM. The increase in extraction with increase in hydrochloric acid concentration could be due to the increased formation of extractable chloride complexes of the type AsCI; and/or the salting-out effect of the acid. The extraction of the element from 0.02M potassium Iodide in hydrochloric acid was then investigated in line with some of our previous mvestigations.“.” The results are presented in Fig. lA,b. The extraction from dilute acid solutions, like that in the absence of iodide, is very poor, but there is a very large increase concentration. Previous increasing acid with investigation” has shown that in the presence of trace concentrations of metals, this extractant has a tendency to polymerize, especially in concentrated hydrochloric acid. and hence the extraction sometimes decreases” with increasing reagent concentration. We therefore examined the effect of using 0.01 M DPPM
instead of 0.1 M and found little change m the distribution data (Fig. lA,c). Figures IB and 1C show the effect of Iodide on the extraction of tracer arsenic from different concentrations of hydrochloric acid by 0.1 M and O.OlM DPPM in benzene, respectively. The D values are higher at high iodide concentrations when O.OlM DPPM is used. The slope of the lines (Fig. IC) is close to unity for [KI] < O.O5M, indicating the involvement of one iodide ion per complex anion formed. These results led us to investigate the solvent effect. In every case the organic phase was pre-equilibrated with the corresponding aqueous phase. These studies were made with O.lM potassium iodide at four concentrations of hydrochloric acid (7, 8, 9 and IOM). The results are presented in Fig. 2. In all cases, curves were obtained with maxima at around 0.01 M DPPM. These curves also indtcate that 9M hydrochloric acid is the optimal concentration for maximum extraction of tracer arsenic. We attribute the decrease in extraction at higher DPPM concentrations to polymerization of protonated DPPM and
SHORT
1057
COMMUNICATIONS
“r
Table I. Distributton coefficients and separation factors [with respect to As(lIl)] for extraction from 0. I A4KI in 9M HCI with O.OlM DPPM in benzene Species
As(III) Na+ K+ cs+ Br-
co’+
Ni’t
CU’+ Zn’+ Cd’+ ;$I’
Fig. 2. The effect of the concentration of DPPM on the distribution coefficient of arsenic from different concentrations of hydrochloric acid: [HCI]: a, 7M; b, 8M; c, 9M; d, IOM.
formation of iodide ion-association complexes with it. Although this work was devoted mainly to tracer studies, an attempt was made to elucidate the composition of the extractable species by loading-ratio data. For this purpose we used a solution of arsenious oxide in 0.1M potassium iodide/9M hydrochloric acid medium and extracted it with O.OlM DPPM in benzene. It was found that the organic phase was saturated at an arsenic concentration of around 0.75 g/l., which confirms a 1: 1 DPPM : arsenic ratio. Comparison with our previous work6 shows that this system is much better than the corresponding thiocyanate system, where the maximum D value was only around 4. In the present case it is almost two orders of magnitude higher and a single equilibration is sufficient for quantitative extraction. To study the selectivity, we investigated the partition behaviour of several other elements under the conditions optimal for extraction of arsenic, i.e., 0.1 M DPPM/benzene, and 0.1 M potassium iodide in 9M hydrochloric acid. The results are shown in Table 1. It is seen that the extraction of arsenic is very selective. Selenium and mercury are two common toxic elements which are co-extracted and the method can be used for their preconcentration. The very high distribution ratios for arsenic also ensure its extrac-
Sr2+ Fe(II1) Au(II1) SC’+ EU’f Ce’+ Cr’+ Se(IV) Sn(IV) Hf(IV) Ru(IV) Sb(V) Cr(V1) Mo(V1) Tc(VI1)
Concentration, M
10-s 1O-4 IO J 10-O 10.” IO J IO ’ IO-5 10-J IO_’ IO_’ IO-6 10-5 IO_ 5 10-d lo-4 10-4 IO-5 ‘C.F. IO-4 10-d 1om4 10-S 1om5 *C.F. 10-5 *C.F.
Distribution coeffictent, D
441 0.0 0.0 0.0 0.01 00 0 03 0. I5 0.02 0 70 99 001 0.01 0.22 6.10 0.01 00 001 0.0 32 0.01 0.01 2.80 0.49 0.0 I2 25
Separation factor
> IOJ > IO4 > IO - 1o4 1 IO’ - lOA > IO’ - IOJ 6.3 X IO’ 447 - IO’ - IOJ - IO 72 - IO) > IOJ - IOJ > IOJ I3 - IO” - IO” I57 9x 102 > I04 35 I7
*C.F. = Carrier-free. tion from large volumes of aqueous phase. Thus under optimum conditions arsenic is more than 99% extracted when the ratio of the organic to aqueous phase volume is I : 20. REFERENCES I. G. 0. Brink, P. Kafalas, R. A. Sharp, E. L. Wetss and J. W. Irvine, J. Am. Chem. Sot., 1957. 79, 1303. 2. W. Ftscher and W. Harre. Anger. Chem.. 1954.66, 165. 3. H. C. Beard and L. A. Lyerly, Anal. Chem., 1961, 33, 1781. 4. V. A. Orlova, B. Ya. Spivakov, N. A. Sharova and T. M. Malyutina, Zh. Analit. Khim.. 1977, 32, 1591. 5. E. S. Donaldson, Tulunru, 1977. 24, 105. 6. M. Ejaz, S. M. Hasany and Z. Umar, Sepn. Ser. Technol.. 1979, 14, 431. 7. M. Ejaz, M. Iqbal, S. A. Chaudhrt and Zamiruddin. J. Radroanal. Chem., 1978. 42, 33.5. 8. M. Ejaz, Shamus-ud-Zuha, S. Ahmad, M. S. Chaudhary and M. Rashid, Mikrochim. Acta, 1980, 1, 7. 9. Y. Marcus and A. S. Kertes, ion Exchange and Solt~enr Extraction qf Metal Complexes, p. 961. Wiley, New York, 1969. IO. M. Ejaz, Anal. Chem., 1978, 90, 740. II. M. Ejaz, M. A. Quresht and Shamus-ud-Zuha, Sepn. Ser. Technol.. 1981, 16, 291. 12. M. Ejaz, S. M. Hasany and I. Hanif. J. Less-Common Metals, 1981, 77, 157.