Talanra, Vol. 30, No. 3, pp. M-159, 1983 Printed in Great Britain. All rights reserved
EXTRACTION SURFACTANTS
0039-9140/83/030155-05$03.00/O Copyright 0 1983 Pergamon Press Ltd
AND DETERMINATION OF ANIONIC WITH COPPER(ETHYLENEDIAMINE DERIVATIVE COMPLEXES
KIYOSHI SAWADA,*
SHIGEHIRO INOMATA, BUIUCHI and TOSHIO SUZUKI
GOBARA
Laboratory of Analytical Chemistry, Faculty of Science, Niigata University, Niigata 950-21, Japan (Received 14 December 1981. Revised 27 September
1982. Accepted 3 October 1982)
SmnmPry-Anionic surfactants, S- [where S- is dodecyl sulphate (DS) or dodecylbenzenesulphonate (DBS)] were extracted with a series of copper (II~thylenediamine derivative complexes, Cu(R-en): +, where R-en is ethylenediamine (en), N,N’-dimethylethylenediamine (NN’Me,en), N,N-dimethylethylenediamine (NNMe,en), N,N-diethylethylenediamine (NNE,en) or 1,2-cyclohexanediamine (Cyen). The extractton constant of the ion-pair is K,, = [Cu(R-en), r+ ‘2s -]a/[Cu(R-en): ‘1 [S -]r. The constants for extraction of the DS complexes with en, NN’Me,en, NNMe,en, NNEt,en and Cyen into chloroform were found to be log K,, = 7.93, 9.19, 8.88, 8.74 and 11.45 (+O.OS at 25”C), respectively. The extractability of the ion-pair Cu(en);+ .2S- gave a linear correlation with the acidity of the solvent. The Cu(Cyen)i+ extraction system was applied to the determination of some anionic surfactants. With use of graphitefurnace atomic-absorption spectrophotometry, a limit of detection of 5 pg/l. was obtained with a 20-ml sample of river water or sea-water.
Ion-pair extraction methods for the determination of anionic surfactants have been extensively investigated.‘z2 The most commonly used and accepted method is the Methylene Blue method developed by Longwell and Maniece, which has been improved and applied to the determination of anionic surfactants in many kinds of water.‘,* Derivatives of Methylene Blue4s5 and other cationic dyes such as Rhodamine B,6 Crystal Violet,’ and Remacryl Blue* have also been used as cationic extractants. Cationic metal complexes such as tris( 1, IO-phenantris( 1, lo-phenanthroline) copthroline) iron(II),g copper (11)‘2-‘4 per(II),rO~” and bis (ethylenediamine) have also been used for ion-pair extraction procedures. The analysis can be completed by determination of the extracted metal ion by radiometry or spectrophotometry. With atomic-absorption graphite-furnace atomic-absorption spectrophotometry it is possible to determine pg/l. levels of surfactant. In spite of its frequent use. the extraction of anionic surfactants with cationic metal complexes has not been studied in detail. In the present paper the extraction of the ion-pair of an anionic surfactant with a copper (IIbthylenediamine derivative complex has been investigated. and the effects of the solvent and particular substituted ethylenediamine have been examined. The copper(II)cyclohexanediamine complex has been applied to the determination of surfactants in natural waters.
EXPERIMENTAL Reagents The sodium dodecyl sulphate (DS) and sodium dodecylbenzenesulphonate (DBS) used were of guaranteed reagent grade (> 99.9% pure). Other reagents used were of analytical reagent grade. Extraction procedure
Mix 20 ml of aqueous phase containing copper(H) sulphate, ethylenediamine derivative and surfactant with 20 ml of organic phase in a SO-ml stoppered centrifuge tube, at 25.0 + 0.2”, and adjust the ionic strength to 0.1 with sodium sulphate. Shake the tube for 10 min (which is sufficient for complete equilibration). After separating the phases, determine the concentration of copper(I1) in the aqueous phase by atomic-absorption spectrophotometry (AAS); alternatively, determine copper in the organic phase by AAS after stripping with 0.0144 nitric acid. Recommended analytical procedure
Add 1 ml of 0.02M copper(I1) sulphate and 1 ml of freshly prepared 0.05M cyclohexanediamine solution to 20 ml of water sample in a 50-ml stoppered centrifuge tube. If necessary, neutralize the sample before adding the reagents. Add 20 ml of chloroform and shake the mixture for 10 min. Centrifuge the mixture, then draw off the aqueous phase with an aspirator. Transfer 10 ml of the chloroform phase to a 15-ml test-tube and shake with 1 ml of 0.01M nitric acid for 10 min. Let stand for 30 min, then determine the concentration of copper in the nitric acid phase by graphitefurnace AAS. If the surfactant concentration is very high, use a larger volume of nitric acid for the stripping, to optimize the copper concentration for the subsequent AAS measurements. RESULTS AND Extraction
*To whom correspondence TAL30.3-e
DISCUSSION
equilibrium
Because of the large equilibrium
should be addressed. 155
constant
for equa-
KIYOSHI SAWADAet al.
156
(A)
a= H
I
J
I -I
-3
4
log
CR__
Fig. 1. The effects of pH (A) and ethylenediamine derivative concentration (B) on the extraction of DBS with metal-en and metal-Cyen complexes. Ethylenediarnine complex: 0, Cu(I1); 0, Cd(I1); A, Ni(I1). (A) C,,, = 10-4M; C, = 5 x 10e5M; C,, = IO-*M. (B) pH = 9.0. Cyclohexanediamine complex: 0, Cu(I1). (A) C,,, = 10-5M; C, = 6 x 10-6M; C,,, = 2 x 10-4M. (B) pH = 9.0.
tion (1),i5 the presence of excess of R-en results quantitative formation of the complex Cu(R-en):+ the alkaline aqueous solution: Cu2 + + 2 R-en $ Cu (R-en): +
in in
(1)
where R-en is ethylenediamine (en), NJ’-dimethylethylenediamine (NN’Me,en), iV,N-dimethylethylenediamine (NNMe,en), iV,N-diethylethylenediamine (NNEt,en) and cyclohexanediamine (Cyen). Thus, the equilibrium for extraction of the surfactant (S-) complex can be written as with the Cu(R-en):+ Cu(R-en):+
+ 2S
+ [Cu(R-en):+
(2)
.2S-1,
where the subscript o denotes that the species is in the organic phase. The extraction constant for the ion-pair, K,, is defined as: [Cu(R-en):+
.2S-1, (3)
&’ = [Cu(R-en)~+][S-12 and the distribution D
ratio of copper(
= [Cu(R-en):+ M
D,,
as:
.2S-1, (4)
[Cu(R-en):+]
Substitution
of equation
(3) into equation
logD,=logK,,+21og[S-] The distribution
(5)
ratio of the surfactant
D = 2[Cu(R-en):+ s
en
log x;,
7.93 9.12
DS DBS
NN’Me,en 9.19
is
.2S-1,
E-1
(6)
or log Ds = log 2Kk;, + log [Cu(R-en):+] + log[S-]
(7)
Effects of pH and ethylenediamine concentration Dodecylbenzenesulphonate (DBS) was extracted with the Cu(II)-en and Cu(IItCyen complexes into chloroform at various pH values and ethylenediamine concentrations. The plot of log D, as a function of pH is shown in Fig. 1 (A). Because of dissociation of the Cu-(R-en) complex caused by the protonation of R-en in the aqueous phase, the distribution ratio shows a steep decrease in the low pH region (pH < 6). When no acid or base is added, the pH of the solution after the addition of R-en is about 9. Thus the optimum extraction pH is obtained without any adjustment.
Table 1. Extraction constants of anionic surfactants with various copper (IIbthylenediamine tive complexes R-en
(4) leads to
NNMe,en 8.88
NNEt,en 8.47
deriva-
Cyen 11.45 f 0.05 11.70 f 0.05
Determination
r
157
of anionic surfactants
(A)
-4.5
-4
-3.5 log
-4.5
CDS3
-4
-3.5
log CDBSI
Fig. 2. The effect of organic solvents on the extraction of DS (A) and DBS (B) with copper(II)ethylenediamine complex. 0 0, chloroform; 8, chloroform-1,2dichloroethane (1: 1 v/v mixture); 0 n , 1,2-dichloroethane; Q, 1,2-dichloroethane-carbon tetrachloride (1: 1 v/v mixture); $, methyl isobutyl ketone; A A, carbon tetrachloride; 0 0, benzene; 0, toluene; 0, chlorobenzene; 0, odichlorobenzene. pH = 9.0; C,, = lo-‘M; Co, = 10e4M.
In Fig. 1 (B), log Du is plotted as a function of log C&V where CRcn is the initial concentration of ethylenediamine derivatives added in the aqueous phase. Although the distribution ratio decreases slightly at very high CRcn, it is almost constant when the concentration of ethylenediamine derivative is sufficient to form the complex Cu(R-en):+, i.e., G&en’ 2Gu. Extraction constant Dodecyl sulphate and dodecylbenzenesulphonate were extracted into chloroform with various copper(I1) R-en complexes. Log DM is a linear function of log [S-l, the value of [S-l being calculated from [S-l = Cs - 2[Cu(R-en):+ .2S-I,, where C, is the total surfactant concentration, The plots all had a slope of 2, and thus the extraction equilibrium (2) was confirmed. The extraction constants K& obtained from the intercepts of the plots are listed in Table 1. Although increase in the molecular weight or molar
volume of R-en tends to increase K&, there is not a good correlation; in particular, Kk;,for the NNEt,en complex is smaller than that for the NN’Me,en and NNMe,en complexes. The extraction constant for DBS is larger than that for DS with both the Cu(en):+ and Cu(Cyen):+ systems. The difference between the log 4, values for the Cu-Cyen complexes of DBS and DS is smaller than that for the Cu-en complexes. The volume of Cu (Cyen): + is considerably larger than that of Cu (en): + . Thus, the extraction constant of the Cu-Cyen complex is less sensitive to change in the volume of surfactant. Eflect of the organic solvents DS and DBS were extracted with the copper(II)-e.thylenediamine complex into various solvents and their mixtures. Log DM is plotted as a function of log [DS] (A) and log [DBS] (B) in Fig. 2. The plots were all linear, with slopes (except for the
Table 2. Extraction constants of Cu(en)i+ complexes, for various solvents and their mixtures Solvent log K,
CHCl, DS DBS
7.93 9.12
CHCl,-DCEt* 7.39
DCE
DCE-Ccl,?
6.89 7.86
* 1,2_Dichloroethane. tl: 1 v/v mixture. §Methyl isobutyl ketone. $Tentative value estimated from data in the region of log [DS] from -4.4 #Mean value for benzene and its derivatives.
6.43
to - 3.8.
MIBKg
cc14
Bz#
6.06
5.54$ 6.50
6.6
KIYOSHISAWADAet al.
1.58
DS-carbon tetrachloride and DS-benzene systems) of 2.0 + 0.1. The extraction constants are listed in Table 2. The distribution ratio for n-hexane as solvent was too small for reliable data to be obtained. Thus, the efficiency of the extractants for the ion-pair Cu(R-en):+ .2Sis in the order chloroform > 1,Zdichloroethane (DCE) > methyl isobutyl ketone (MIBK) > carbon tetrachloride N benzene and its derivatives (Bz)>>n-hexane for both surfactants. The D, values for the benzene derivatives are of the same order of magnitude as that for benzene (Table 2) and show poor correlation with the polarity (e.g., the ET valueI and 2 value”) and dielectric constant of the solvents. The correlation of log KX with the solubility parameter” is also not very good, which is consistent with the poor correlation with the molar volume of the complex cation. A plot of log &, as a function of the acceptor number, which is an indicator of the acidity of the solvent,‘8 shows a good linear correlation, and the &, of n-hexane, which has an acceptor number of zero, is very small. Hence, the change in interaction of the acidic part of the ion-pair, i.e., the complex cation Cu (R-en): + , with the solvent may not contribute very significantly to the change in the extractability of the ion-pair, whereas the interaction between the solvent and the basic part of the ion-pair, i.e., the surfactant anion, may have the predominating effect on the change in extractability with change of solvent. Extraction with cadmium and nickel complexes
DBS was also extracted into chloroform with the cadmium and nickel ethylenediamine complexes. Log D, is plotted as a function of pH for the cadmium and nickel systems in Fig. 1 (A). The pH region for maximum extraction of these complexes is narrower than that for the Cu(I1) complex. The effect of the ethylenediamine concentration is shown in Fig. 1 (B). As the formation constants of the ethylenediamine complexes of Cd(I1) and Ni(I1) are smaller than that for Cu(II),” a higher ethylenediamine concentration is required to obtain the maximum distribution ratio. At high concentrations of ethylenediamine, there is a steep decrease in D, for the cadmium and nickel systems, possibly because of formation of the larger tris-ethylenediamine complexes. The log K,, values found for DBS were 9.12, 9.58 and 8.28 for the copper, cadmium and nickel ethylenediamine systems respectively. ANALYTICAL
APPLICATION
Recovery and calibration curve
The extraction
constants
for the anionic
surfac-
tants are much larger with the copper(II)cyclohexanediamine complex than with copper(II)-ethylenediamine complex (Table 1). As can be seen from equation (7), the distribution ratio of the surfactant, D,, decreases with decrease in the equilibrium concentration of surfactant in the aqueous phase, [S-l, when the concentration of Cu(R-en);+ is kept constant. It follows that a fairly large value of KX is required for the determination of low concentrations of surfactant. For example, in the copper(II)-ethylenediamine method, the distribution ratio of a lo-pg/l. DS sample is 0.059 (i.e., the degree of extraction of DS is about 5.6%) for equal volumes of the aqueous [lo-‘M Cu(R-en):+] and organic (chloroform) phases. On the other hand, in the copper(II)-cyclohexanediamine method, D, is about 13 (i.e., - 93% extraction of the surfactant) under the same conditions. Thus, the latter method is advantageous for the determination of low surfactant concentrations. Over the DS concentration range of 10~10,000 pg/l. the recommended procedure gave an apparent 100 f 3% extraction at high concentrations and over 90% at low, in agreement (within 5%) with that calculated by using the K,, values shown in Table 1. Limit of detection
It might be possible to lower the limit of detection by using a large volume of sample, but tests with surfactant-free water give blank values of several tenths of a pg of copper per litre, corresponding to a DS concentration of several pg/l. Thus, the limit of detection is taken as 5pg/l. (as DS). Interferences
The effect of inorganic anions, which may be extracted as Cu(Cyen)i+ ion-association complexes, on the detection limit, has been examined by use of extraction tests in the absence of surfactant. The concentrations of anions listed in Table 3 produce contributions below the detection limit for DS (< 5 pg/l.). A large ion such as I- or ClO, is extracted at relatively low concentration levels, but the concentrations of these ions in natural waters are much less than those listed in Table 3. The concentrations listed in Table 4 cause an error of less than +5% in the determination of DS at the 100~pg/l. level. Sulphide ion up to the 5 mg/l. level does not interfere if iron(I1) is added to give [Fe* +] = 50 mg/l. When the concentration of sulphide ion is less than 0.5 mg/l., it is not necessary to add the iron(I1). The cations and anions cause negative and positive interferences, respectively. The allowable concentrations of foreign ions listed in Tables 3 and
Table 3. Concentrations of anions giving a signal below the limit of detection for Anion Concn., M
SOi1
Cl0.5
DS (5 pg/l.) Br3 x 10-j
I10-j
NO; 10-s
CIO; 10-d
Determination
159
of anionic surfactants
Table 4. Concentration of foreign ions causing < kS% error for extraction of DS at the 100~pg/l. level Concentration 0.75M 1000 mg/l. 500 ma/l. 100 mg/l. 50 mg/l. 5 ma/l.
Ion SO:-, Cl-, Na+ ClO,, NO;, Br-, K+, Ca*+, M$+, Ni*+ 1Zn2+ 7 Mn2+, ICo’+, Al’+ Fe2+, Fe’+ CN-, S2-*
NH:
*With 50mg/l. Fez+ added; 0.5 mg/l. in the absence of added Fe2 + .
4 are higher than those present in sea-water. As can be seen from Tables 1 and 2, the extraction constants for DBS larger than those for DS. Consequently, the interference of foreign ions is less significant in determination of DBS than of DS. The concentrations of anionic surfactant found in some natural waters by the recommended procedure were as follows: sea-water (off-shore, Japan Sea, Niigata Prefecture), not detected; river water (Shin river, Niigata, Japan), 310 pg/l., expressed as DS; lake water (Toyano lake, Niigata, Japan), 590 pg/l., expressed as DS.
REFERENCES
1. R. A. Llenado and R. A. Jamieson, Anal. Chem., 1981, 53, 174R. 2. M. J. Fishman, D. E. Erdmann and T. R. Steinheimer, ibid., 1981, 53, 182R. 3. J. Longwell and W. D. Maniece, Analyst, 1955,80, 167. 4. L. K. Wang and P. J. Panzardi, Anal. Chem., 1975,47, 1472.
5. K. Toei and H. Fujii, Anal. Chim. Actu, 1977, 90, 319. 6. Z. Marczenko, Microchim. Acta, 1977 II, 651. 7. S. Motomizu, S. Fujiwara and K. Toei, Anal. Chim. Acta, 1981, 128, 185; S. Motomizu, S. Fujiwara, A. Fujiwara and K. Toei, Anal. Chem., 1982, 54. 392. 8. S. Janeva and R. Borissova-Pangarova, Tulanta, 1978, 25, 279. 9. C. G. Taylor and J. Waters, Analyst, 1972, 97, 533. 10. A. Le Bihan and J. Courtot-Coupez, Anal. Lett., 1977, 10, 759. 11. Idem, Bull. Sot. Chim. France, 1970, 406. 12. P. T. Crisp, J. M. Eekert and N. A. Gibson, Anal. Chim. Acta, 1975, 78, 391. 13. P. T. Crisp, J. M. Eekert, N. A. Gibson, G. F. Kirkbright and T. S. West, ibid., 1976, 87, 97. 14. M. J. Gagnon, Water Res., 1979, 13, 53. 15. L. G. Sill&n and A. E. Martell, Stabilitv Constants of Metal-fan Complexes. The Chemical Society, London, 1964. 16. K. Dimroth, C. Reichardt, T. Siepmann and F. Bohlmann, Ann., 1963, 661, 1; 1963, 669, 95. 17. E. Kosower, J. Am. Chem. Sot., 1958, 80, 3253, 3261, 3267. 18. J. H. Hildebrand and R. L. Scott, Regular Solutions. Prentice-Hall, Englewood Cliffs, New York, 1962. 19. U. Mayer, V. Gutmann and W. Gerger, Monatsh. Chem., 1975, 106, 1235.