ISSN 10619348, Journal of Analytical Chemistry, 2015, Vol. 70, No. 7, pp. 837–842. © Pleiades Publishing, Ltd., 2015. Original Russian Text © A.V. Sidelnikov, D.M. Bikmeev, D.I. Dubrovskii, F.Kh. Kudasheva, V.N. Maistrenko, 2015, published in Zhurnal Analiticheskoi Khimii, 2015, Vol. 70, No. 7, pp. 721–726.

ARTICLES

Determination of Anionic Surfactants Using Methods of Impedance Spectroscopy and Chemometrics A. V. Sidelnikov, D. M. Bikmeev, D. I. Dubrovskii, F. Kh. Kudasheva, and V. N. Maistrenko Department of Chemistry, Bashkir State University, ul. Zaki Validi 32, Ufa, Republic of Bashkortostan, 450076 Russia email: [email protected] Received May 15, 2014; in final form, September 27, 2014

Abstract—A procedure is developed for the determination of anionic surfactants (ASs) in water–oil emul sions by impedance spectroscopy with the chemometric processing of experimental data. The conditions of recording impedance titration curves were optimized and the performance characteristics of the procedure were estimated. It was shown that impedance spectroscopic titration is applicable to the determination of AS in aqueous solutions and water–oil emulsions in the presence of organic and inorganic substances. Examples of determination of ASs based on sodium olefin sulfonate are given. Keywords: impedance spectroscopy, anionic surfactants, chemometrics, principal component analysis, titri metry, water–oil emulsions, sodium olefin sulfonates DOI: 10.1134/S1061934815070151

At present ASs in water media and water–oil emul sions are most often determined by spectrophotome try and potentiometry [1–3]. The potentiometric method based on surfactantselective electrodes has limitations in sensitivity and selectivity in the analysis of multicomponent aqueous–organic media and water–oil emulsions containing viscous organic com ponents, oil, etc. [3, 4]. To determine ASs in such objects, analysts usually preseparate aqueous–organic mixtures to retain the selectivity and stability of the work of sensing membranes of surfactantselective elec trodes, which substantially complicates and increases the cost of analysis. In some cases [5], test solutions were diluted with water, which significantly reduced the sensitivity of determinations. Spectrophotometric determinations of ASs are less often used for these pur poses because of the presence of colored and colloidal particles in water–oil emulsions and also the scatter ing of the light beam by the heterogeneous phases present in the samples [3]. The use of some other methods expands the possi bilities of the determination of ASs [3, 6], for example, in the case of low surfactant concentrations, low val ues of the analytical signal, and also the insufficient selectivity of the response of surfactantsensitive elec trodes, when individual surfactants cannot be deter mined in real samples. Among these methods we should mention conductometric titration, which used in the analysis of multicomponent mixtures of com plex composition [2, 3]. A promising method is also impedance spectros copy, which is based on the measurement of the dependence of an impedance of an electrochemical

cell on the frequency of alternating current [7]. The test samples of various nature and composition and processes occurring on the electrodes are character ized by different dependences of impedance compo nents (imaginary and real parts) on the frequency of alternating current, which allows the use of this method in analytical purposes for the determination of ASs. An important advantage of the method is the high sensitivity of measurements and no requirements imposed on the selectivity of electrodes, the presence of colored components, the existence of heteroge neous phases, etc. Using the chemometric principal component anal ysis (PCA), in this work we proposed a procedure for the determination of ASs by impedance spectroscopy, optimized the conditions of recording impedance titration curves, estimated the reproducibility of mea surements, and presented the results of determina tions of ASs based on sodium olefin sulfonate in water–oil emulsions. EXPERIMENTAL We used a threeelectrode electrochemical cell for impedance titration with an indicator, an auxiliary electrode, and a counter electrode, each of which was made of stainless steel with a surface area of 125 mm2, length of 50 mm, and thickness of 0.8 mm (Fig. 1). Impedance spectra were recorded on an Elins Z500P impedance meter in the alternating current fre quency range 0.5–10 kHz with the amplitude 50 mV at the zero potential of the indicator electrode.

837

838

SIDELNIKOV et al.

HCO 3−, CO32−, Na+, Ca2+, polymer additives, etc. A test solution (0.1–0.5 g) was weighed on an analytical bal ance, transferred to a cell for titration, and diluted with water to 10 mL; an impedance spectrum was recorded on adding an aliquot portion of a titrant. The titrant was a standard aqueous solution of a cationic surfactant, cetylpyridinium chloride (CPC) from Merk (Germany) with the concentration of the main substance not less than 99%. To reduce the degree of dilution of the studied solution in the course of titration, the concentration of the titrant was one order of magnitude higher than the concentration of AS to be determined. The block diagram of a titration installation is shown in Fig. 2. The titrant was fed with a ColeParmer syringe dosing unit, dosing rate 0.6 mL/min, total amount of the titrant 4.5 mL; the duration of titration was syn chronized with the time of recording 14 impedance spectra, each of which consisted of 50 imaginary and real components. The experimental data were processed using the Chemometrics AddIn software for Microsoft Excel [8].

Titrant supply

1 2 3

Magnetic stirrer Fig. 1. A cell for impedance spectroscopic titration: (1) counter electrode, (2) working electrode, (3) reference electrode.

The concentrations of ASs were determined in aqueous solutions and multicomponent water–oil emulsions with a salt content of 10–50 g/L. The stan dard AS was sodium dodecyl sulfate (SDS) of analyti cal grade from Panreac (Spain). Water–oil emulsions were obtained on an intense stirring of aqueous SDS solutions with highparaffin crude oil in the presence of inorganic salts. For sodium olefin sulfonate, water– oil emulsions were obtained by passing aqueous AS solutions through core samples containing highparaf fin crude oil. Depending on the experimental condi tions (pressure, concentration of AS, salt content, temperature, etc.) we obtained water–oil emulsions of different compositions, which contained oil, ASs,

Syringe pump

RESULTS AND DISCUSSION Impedance spectroscopic titration is based on the interaction of an AS with CPC with the formation of ion pairs: CnH2nSO3− + C21H38N+ → CnH2nSO3–NH38C21. Along with a decrease in the concentration of AS, the formation of ion pairs manifests itself in the curves of the dependence of imaginary (Im) and real (Re) impedance components (impedance hodograph) on the frequency of alternating current (Fig. 3). In con trast to conventional conductometric titration, the data of impedance spectroscopic titration for each point are multidimensional experimental data arrays (14 × 50).

Impedance meter

PC

Electrochemical cell

Fig. 2. Block diagram of installation for impedance spectroscopic titration. JOURNAL OF ANALYTICAL CHEMISTRY

Vol. 70

No. 7

2015

DETERMINATION OF ANIONIC SURFACTANTS USING METHODS Im, Ω 20000 18000 16000 14000 12000 10000 8000 6000 4000 2000 0 100

5

Im × 10–3, Ω 22

839

(a)

4 3 20

2 1

18 16 14 1100

2100

3100

4100

5100

6100 Re, Ω

Fig. 3. Impedance spectra of a 1% (w/w) solution of sodium dodecyl sulfate in a water–oil emulsion near the equivalence point for titrant volumes VT, mL: (1) 0.6; (2) 0.9; (3) 1.2 (titration endpoint); (4) 1.5; (5) 1.8 mL.

It is known [9] that impedance spectra for different processes occurring in an electrochemical cell (charge transfer, semiinfinite quasispherical diffusion, and also diffusion in a finite region with reflecting bound ary conditions, nonfaradaic processes under the con ditions of substance sorption–desorption on an elec trode, mixed adsorption–diffusion control, etc.) differ in shapes. Thus, the kinetics of the adsorption of sodium decyl sulfate on the surface of a mercury elec trode was studied in [10, 11] using impedance spec troscopy by the frequency dependence of admittance (imaginary component) at more negative potentials than the potential of the sorption–desorption peak. It was found that the diffusion of sodium decyl sulfate is the ratedetermining step of the adsorption process below and above the critical concentration of micelle formation. Thus, any processes occurring in the titra tion cell (on the electrode surface, in the double elec tric layer, in solution, etc.) affect not only the absolute values of impedance, as in conventional conducto metric titration, but also the shape of the impedance hodograph, which is indicative of the nature of the occurring processes. In a reaction between a cationic and an anionic surfactant proceeding under dynamic conditions, we observed changes both in the quantita tive ratio of components in solution and in its qualita tive composition, which manifested themselves in imaginary and real impedance components (Fig. 4). A change in the impedance of a system in the course of titration is described by a dependence much similar to the curve of conductometric titration [12]: it includes sections of decreasing and increasing imped ance. As an impedance spectrum recorded after the addition of an aliquot portion of a titrant was an array of experimental data rather than a single value, it was processed using a promising method of chemometrics, principal component analysis [13], which allows the JOURNAL OF ANALYTICAL CHEMISTRY

Vol. 70

12 0.3 0.6 0.9 1.2 1.5 1.8 2.1 2.4 2.7 3.0 3.3 3.6 3.9 4.2 VT, mL (b) –3 Re × 10 , Ω 6.4 6.2 6.0 5.8 5.6 5.4 5.2 5.0 4.8 4.6 0.3 0.6 0.9 1.2 1.5 1.8 2.1 2.4 2.7 3.0 3.3 3.6 3.9 4.2 Vt, mL Fig. 4. Changes in (a) imaginary and (b) real impedance components (frequency 10 kHz) in the titration of a water–oil emulsion with sodium dodecyl sulfate of the concentration 1 wt %.

revelation of intrinsic regularities characteristic for a data array and their use in analytical purposes. Using principal component analysis, spectra of imaginary and real impedance components after their transfor mation were presented as points in a multidimensional space and projected onto the first principal compo nent, built along the maximum change in the experi mental data. Then the next principal component was built; it was orthogonal to the first component and directed along the next change in multidimensional data in magnitude, etc. Using the specified chemo metric operation, the experimental data were com pressed: from 50 impedance signals for each spectrum, we obtained two generalized coordinates, x and y (points in the plane PC1, PC2, see Fig. 5). As can be seen in Fig. 5, the titration curves in the plane PC1, PC2 exhibit linear sections with a clear break, depending on the volume of the added titrant. The inflection point (Vendpoint = 1.20 mL) character izes the equivalence state of the reaction in the course of titration. At a constant rate of adding the titrant, the No. 7

2015

840

SIDELNIKOV et al. (а)

PC2 (18%) 4 1.2 mL

0.9 mL

3 2 1

0.6 mL –12

–10

–8

–6

–4

–2

0

–1 2.1 mL –2

0.3 mL PC2 (10%) 4

1.5 mL 1.8 mL 2 4 PC1 (67%)

(b) 0.3 mL

3 2

–4

–2

1

1.8 mL

0 –1

2 2.1 mL

–2

4

8 6 0.6 mL

10 12 PC1 (73%)

1.5 mL

–3 –4

0.9 mL

1.2 mL –5 –6 Fig. 5. Score plots of a PCA modeling of impedance spectra for (a) imaginary and (b) real components in the titration of a water– oil emulsion with sodium dodecyl sulfate of the concentration 1 wt % (Vendpoint = 1.2 mL).

accuracy of the determination of Vendpoint depended on the rate and number of measurements of imped ance spectra. The higher the required accuracy of the result of titration, the greater number of spectra must be recorded per unit time. At the selected rates of recording impedance spectra (2 spectra/min) and rate of adding the titrant 0.6 mL/min, the error of the mea surement of Vendpoint did not exceed 0.15 mL (duration of titration 7 min, sample mass 0.1–0.5 g, range of AS concentration 0.5–1 wt %). The results of AS determi nation in water–oil emulsions of highparaffinic crude oil using SDS and Enordet sodium olefin sulfonate manufactured by Shell Chemicals are presented in the table. It can be seen that, for the selected titration con ditions, the relative standard deviation of the results of determination of both SDS and olefin sulfonate did not exceed 3%. The relative error of the result of deter mination of olefin sulfonate in water–oil emulsions on the variation of sample weight (0.1–0.4 g) for n = 3, P = 0.95 was equal to 7%. The results obtained indicated the absence of systematic errors in the determination

of ASs in the specified weight range of samples of water–oil emulsions. It should be noted that, under the studied condi tions for the determination of the titration endpoint (inflection point in the titration curve), only two prin cipal components are sufficient (Fig. 5) because, in this case, the presence of a well expressed inflection point (change of the sign of a monotonous titration curve in the score plot) is more important compared to the value of explained dispersion (fraction of disper sion VT due to the change of impedance). An increase in the number principal components virtually did not affect Vendpoint and the results of AS determination. As is known from the theory and practice of con ductometry, an important condition of recording clear curves of conductometric titration is the substitution of one ion with another with a different mobility in the course of titration reaction. In impedance spectro scopic titration, this requirement is not obligatory, because impedance spectra depend not only on the mobility of ions in solution, but also on the nature of

JOURNAL OF ANALYTICAL CHEMISTRY

Vol. 70

No. 7

2015

DETERMINATION OF ANIONIC SURFACTANTS USING METHODS

841

Determination of ASs in water–oil emulsions ASs Sodium dodecyl sulfate

IOS–C24–C28 mixture of sodium olefin sulfonates

IOS–C15–C28 mixture of sodium olefin sulfonates

Sample weight, g

Added AS, wt %

Found AS, wt %

0.1621

1.00

0.99

0.2942

1.00

0.98

0.3914

1.00

0.97

0.4911

1.00

1.03

0.6103

1.00

1.00

0.1402

0.50

0.49

0.3121

0.50

0.51

0.4035

0.50

0.48

0.1263

0.70

0.68

0.2128

0.70

0.72

0.3315

0.70

0.71

0.1503

0.50

0.52

0.2475

0.50

0.51

0.3951

0.50

0.49

0.1627

0.70

0.71

0.2281

0.70

0.69

0.3418

0.70

0.73

RSD, %

0.02

0.03

0.03

0.03

0.03

* Recalculated to the main component.

processes occurring at the electrode/solution inter face, i.e., adsorption, diffusion, and kinetic, which affect the shape of impedance spectra. An important advantage of the method of impedance spectroscopic titration over conductometry and potentiometry is the possibility of variation of the selectivity of response by varying the frequency range of alternating current and the potential of the electrode. The high rate of record ing cell impedance is of no less importance in contrast to potentiometry, in which an equilibrium potential is attained within several minutes. *** Thus, the results obtained point to the advantages of the method of impedance spectroscopy over the known methods of the quantitative determination of ASs in complex multicomponent homogeneous and heterogeneous systems. The method ensures the instrumental variation of the sensitivity and selectivity of AS determination and the rapidity and accuracy of determinations using different frequency ranges of alternating current and also the variation of the area and polarization potential of the working electrode. The high rate of recording impedance spectra permits the creation of an automatic impedance titrator in the future. There are prospects for the development of highly sensitive rapid methods for the determination surfactants and other trace components not only in JOURNAL OF ANALYTICAL CHEMISTRY

Vol. 70

homogeneous systems, but also at interfaces (micelles, nanoparticles, etc.) in solutions of complex composi tions. ACKNOWLEDGMENTS This work was supported by the Russian Founda tion for Basic Research, project no. 140397067r_ povolzh’e_a). REFERENCES 1. Sanchez, J. and Valle, M., CRC Crit. Rev. Anal. Chem., 2005, vol. 35, no. 1, p. 15. 2. Kulapina, E.G., Chernova, R.K., Makarova, N.M., and Pogorelova, E.S., Rev. J. Chem., 2013, vol. 3, no. 4, p. 323. 3. Bazel’, Ya.R., Antal, I.P., Lavra, V.M., and Kormosh, Zh.A., J. Anal. Chem., 2014, vol. 69, no. 3, p. 211. 4. Surfactants: A Practical Handbook, Lange, K.R., Ed., Minich: Hanser, 1999, 3rd ed. 5. Kuznetsov, V.M., Ishbulatov, R.M., Mryasova, L.M., Solominova, T.S., and Kolbin, A.M., RF Patent 2402906, Byull. Izobret., 2010, no. 31. 6. Sumina, E.G., Shtykov, S.N., and Tyurina, N.V., J. Anal. Chem., 2003, vol. 58, no. 8, p. 720. 7. Barsoukov, E. and Macdonald, J., Impedance Spectros copy: Theory, Experiment, and Applications, New York: Wiley, 2005. No. 7

2015

842

SIDELNIKOV et al.

8. Pomerantsev, A.L., Chemometrics in Excel, New York: Wiley, 2014. 9. Electroanalytical Methods: Guide to Experiments and Applications, Scholtz, F., Ed., Berlin: Springer, 2006. 10. Vollhardt, D., Modrow, U., Retter, U., Jehring, H., and Siegler, K., J. Electroanal. Chem., 1981, vol. 125, no. 1, p. 149. 11. Vollhardt, D., Retter, U., Szulzewsky, K., Jehring, H., Lohse, H., and Siegler, K., J. Electroanal. Chem., 1981, vol. 125, no. 1, p. 157.

12. Budnikov, G.K., Maistrenko, V.N., and Vyaselev, M.R., Osnovy sovremennogo elektrokhimicheskogo analiza (Foundations of Modern Electrochemical Analysis), Moscow: Mir, 2003. 13. Rodionova, O.E. and Pomerantsev, A.L., Khe mometrika v analiticheskoi khimii (Chemometrics in Analytical Chemistry), 2006. http://www.chemometrics. ru/materials/articles/chemometrics_review.pdf. Accessed April 20, 2014.

JOURNAL OF ANALYTICAL CHEMISTRY

Translated by E. Rykova

Vol. 70

No. 7

2015