Theoretical aspects of micellar liquid chromatography using C12DAPS surfactant

Fluid Phase Equilibria 147 Ž1998. 301–307

Theoretical aspects of micellar liquid chromatography using C 12 DAPS surfactant M.H. Guermouche ) , D. Habel, S. Guermouche Institut de Chimie, USTHB, BP 32, El-Alia, Bab-Ezzouar, Alger, Algeria Received 21 July 1997; accepted 11 March 1998

Abstract Theoretical aspects of micellar liquid chromatography using a zwitterionic surfactant were investigated. The micellar mobile phase consisted of n-dodecyl-N, N-Dimethylamino-3-propane-1-sulfonate Žbetter known as C 12 DAPS. aqueous solutions. Chromatography was carried out on m Bondapak C 18 column, UV detection was measured at 254 nm. Using the Armstrong equation, the partition equilibria constants of the solutes chromatographied were established between water and stationary phase; water and micelles; micelles and stationary phase. Compared to anionic or cationic surfactants, the zwitterionic surfactant gives the highest K SW values and a lowest K MW values. Therefore, for a constant surfactant concentration, capacity factors are greater on C 12 DAPS. Hydrophobic interactions with the stationary phase, electrostatic effects of the surfactant from both the micelle and the surfactant modified stationary phase explain the retention of a solute. An other way to study molecular interactions is made via the linear solvation energy relationships ŽLSER.. For the test solutes used, it seems that V1r100 Žsolute’s size. and b Žbasicity. are predominant to affect the retention. K SW LSER correlation express the binding of the solute with the micelles. The term s found is positive. C 12 DAPS has a greater dipolar environment affecting considerably the MLC partition. The term a is positive, acidic solutes binds more easily with the zwitterionic surfactant than with SDS. q 1998 Elsevier Science B.V. All rights reserved. Keywords: Micellar liquid chromatography; C 12 DAPS; Capacity factor; Thermodynamic of the retention; LSER model

1. Introduction Since the first report of Armstrong and Nome w1x, micellar liquid chromatography Ž MLC. had a solid growth. The potential applications and unique capabilities of MLC have been investigated in more than one hundred papers including several reviews w2–4x. Most of the micellar mobile phases include anionic surfactant Ž sodium dodecyl sulfate. , cationic surfactant Ž hexadecyltrimetyl or dode)

Corresponding author.

0378-3812r98r$19.00 q 1998 Elsevier Science B.V. All rights reserved. PII S 0 3 7 8 - 3 8 1 2 Ž 9 8 . 0 0 2 4 2 - 8

M.H. Guermouche et al.r Fluid Phase Equilibria 147 (1998) 301–307

302

cyltrimethyl ammonium bromide. or non-ionic surfactant Ž Brij 35. . A few papers describe the use of a zwitterionic surfactant w5,6x. C 12 DAPS Ž n-dodecyl-N, N-dimethyl-amino-3-propane-1-sulfonate. better known as sulfobetaine-12 is one of the common surfactants which maintains its zwitterionic property over all the pH range used in RPLC. Micellar mobile phases have advantages over hydroorganic mobile phases in RPLC. They offer the possibility to inject directly biologicals w5–7x, to resolve optical isomers via chiral micelles w8,9x. MLC extends the RPLC applications by the use of secondary equilibrium in the liquid phase introducing the enhancement of selectivity, by an extension of the parameters influencing the retention. However, MLC is less efficient than conventional RPLC w10,11x. In this paper, we report the theoretical aspects of MLC using C 12 DAPS as zwitterionic surfactant. The validity of the Armstrong retention is investigated. Linear solvation energy relationships ŽLSER. model is also applied to characterise solute-micelle interactions.

2. Theoretical 2.1. Thermodynamic of the retention

Micellar liquid chromatography includes three equilibria: partition equilibria of the solutes chromatographied between water and stationary phase; water and micelles; micelles and stationary phase characterised respectively by the equilibrium constants K SW , K MW and K SM . Armstrong and Nome first w1x and Deluccia et al. w4x established the thermodynamics of the retention; they verified the validity of their theory in the case of a cationic or anionic surfactant. Armstrong equation of a binding solute capacity factor indicates a linear relation between the capacity factor inverse and the micelle concentration. 1rkX s n P Ž K MW y 1 . r Ž f P K SW . P Cm q 1r Ž f P K SW . X

Ž1.

where: k is the capacity factor of the solute; K SW is the partition constant of the solute between stationary phase and water; K MW is the partition constant of the solute between micelles and water; f is the phase volume ratio s Stationary phase volumermobile phase volume; n is the molar volume of the surfactant; Cm is the micellar concentration of the surfactants total surfactant concentration-critical micellar concentration; 1rkX against wmMx is a straight line; K SM ; partition constant of the solute between stationary phase and micelles is defined as K SWrK MW . K MW values are independent of the nature of the stationary phase in the same micellar mobile phase. K MW increases with the increasing of the electrostatic interactions. Therefore, capacity factor decreases when K MW increases. Affinity of the solute for the surfactant covered stationary phase is

M.H. Guermouche et al.r Fluid Phase Equilibria 147 (1998) 301–307

303

expressed via K SW values. Capacity factor increases when K SW increases as the affinity of the solute for the stationary phases. Therefore, K SW and K MW have antagonist effects on the retention time of a solute. 2.2. Linear solÕation energy relationships (LSER) model The LSER modelling introduced by Taft et al. w12x allows to determine a solubility-related property SP as: SP s SPo q cavity term q dipolar term q hydrogen bonding terms

Ž2.

In the case of the MLC, a multiparameter equation describes the different possible interactions as: SP s SPo q mV1r100 q sp ) q bb q a a

Ž3.

where: SP is the solute property such as ln Ž1-octanol-water partition coefficients. or ln Ž capacity factors. ; V1 is the intrinsic ŽVan der Waals. molar volume of the solute; p ) is the solvatochromic parameter which measures the solute polarityrpolarisability; b is the solute basicity; a is the solutes’ acidity; SPo, m, s, b, a are the corresponding parameters. For a number of solutes, Eq. Ž3. was applied to the 1-octanol-water partition coefficients w13x, and to the capacity factors in RP-HPLC w14–18x, micellar electrokinetic chromatography w19x, and MLC w20x using cationic and anionic surfactants but no zwitterionic surfactants.

3. Experimental 3.1. Reagents All solvents Ž spectroscopic grade. used were from Fluka ŽSwitzerland.. C 12 DAPS Ž n-dodecylN, N-dimethyl-amino-3-propane-1-sulfonate. was from Fluka Ž Switzerland.. Two experiments were carried out. To prove the validity of the Armstrong retention model, the same solutes of Ref. w21x were used. This choice allows to compare the results with those obtained with cationic or anionic surfactants. In the second experiment ŽLSER study. , test solutes were purchased from Merck Ž Germany. . 3.2. Apparatus HPLC was with a system comprising a Waters Chromatograph ALCrGPC 244 including 6000A pump, U6K universal injector and Pye Unicam PU 4020 UV detector operating at 254 nm. A guard column Ž 4 = 0.4 cm. laboratory packed with Corasil II C18 Ž Waters. along with a Ž Bondapak C18 Ž0.39 = 30 cm. analytical ŽWaters. were used. To saturate the mobile values phase, a precolumn Ž5 = 0.4 cm. homepacked with silica was placed between the pump and the injector. Several aqueous mobile phases were used. They contained C 12 DAPS with no organic modifier in the following concentrations: 3.10y3 M, 7.10y3 M, 12.10y3 M, 15.10y3 M and 20.10y3 M. For the LSER study, C 12 DAPS concentration was fixed to 15.10y3 M.

304

M.H. Guermouche et al.r Fluid Phase Equilibria 147 (1998) 301–307

4. Results and discussion 4.1. Validity of the Armstrong theory in the zwitterionic MLC According to Eq. Ž1. , 1rkX was plotted against the micelles concentration. Results are summarised in Table 1. The correlation coefficients values close to 1 appearing in this table allow to consider that the theoretical Armstrong model describes correctly the zwitterionic MLC. The studied solutes are binding solutes; their capacity factor change when the surfactant concentration is varied. Compared to anionic or cationic surfactants, the zwitterionic surfactant give a highest K SW values and a lowest K MW values. Therefore, for a constant theoretical surfactant concentration, capacity factors are greater on C 12 DAPS. As shown in Table 1, for p-nitroaniline and p-nitrophenol, no notable differences in the K MW or K SW are observed with SDS. Therefore, these solutes are not separated at any concentration of SDS. They are completely resolved on DTAB or C 12 DAPS because of the differences in there K Mw or K SW values. Hydrophobic interaction with the stationary phase, electrostatic effects of the surfactant from both the micelle and the surfactantmodified stationary phase explain the retention of a solute. For Table 1 Calculated equilibrium constants of C 12 DAPS. Comparison with SDS and DTAB data Žsee Ref. w21x. Surfactant

Slope

y-intercept

Correlation

K MW

K SW

41.6 77 80 33.2 215 188 29.2 39 111 39.2 90 89 40.0 77 305 44.2 84 192 40.0 78 75 29.2 98 91

63.5 13.8 18 93.3 41.3 47 68.4 5 23 63.0 22.4 23 97.4 8.7 99 127.8 9.4 39 53.7 19 17.2 36.6 28 29.3

Determined by Eq. Ž1. Benzene

Toluene

Phenol

Nitrobenzene

p-Nitrophenol

p-Nitroaniline

Acetophenone

Benzaldehyde

C 12 DAPS SDS Ž12. DTAB Ž12. C 12 DAPS SDS Ž12. DTAB Ž12. C 12 DAPS SDS Ž12. DTAB Ž12. C 12 DAPS SDS Ž12. DTAB Ž12. C 12 DAPS SDS Ž12. DTAB Ž12. C 12 DAPS SDS Ž12. DTAB Ž12. C 12 DAPS SDS DTAB C 12 DAPS SDS DTAB

0.45

2.40=10y2

0.997

0.73

2.50=10y2

0.922

0.70

3.44=10y2

0.998

0.48

2.55=10y2

0.997

0.65

2.50=10y2

0.966

0.86

2.26=10y2

0.966

0.47

2.95=10y2

0.977

0.46

4.27=10y2

0.988

M.H. Guermouche et al.r Fluid Phase Equilibria 147 (1998) 301–307

305

C 12 DAPS, only the hydrophobic interaction and no electrostatic attraction affect the retention of non-polar solutes such as benzene or toluene. The C12 hydrocarbon chain present in C 12 DAPS explains their high capacity factors. As described above, electrostatic effects are small when eluting non-polar solutes. Then, the nature of the surfactant will affect considerably the separation of polar and non-polar solutes. For example, nitrobenzene and toluene are well separated on zwitterionic or anionic surfactant but not on DTAB. Acidic solutes such as phenol or p-nitrophenol are repulsed electrostatically by anionic micelles and the covered stationary phase decreasing retention; the two solutes have approximately the same retention time. Opposite interactions occurs with cationic micelles, p-nitrophenol has a higher retention time than phenol. With a zwitterionic micelles, an intermediate phenomena is present; an hydrophobic interaction with the stationary phase is possible and a reduced electrostatic repulsion exists. p-nitrophenol and phenol are well separated at various surfactant concentration with a higher retention time for p-nitrophenol. 4.2. Linear solÕation energy relationships (LSER) correlation For the solutes used, V1r100, p ), b and a extracted from Refs. w13,19x are given in Table 2. These solutes are representative from three main groups: non-bonding compounds, hydrogen bond acceptors and hydrogen bond donors. LSER correlation observed for the retention is: ln kX s 1.32 q 4.80 P V1r100 q 0.24 P p ) y 3.32 P b y 0.18 P a Fig. 1 shows the plot of ln kX Žtheoretical. predicted by LSER versus ln kX Žexperimental.. A satisfying correlation is observed Ž r s 0.944.. In hydroorganic RPLC, V1r100 Žsolute’s size. and b Žbasicity. are predominant w13,15,16x. to characterise the retention. It is also the case with zwitterionic C 12 DAPS MLC Ž m s 4.8; b s y3.32. . Table 2 Solvatochromic parameters of the solutes used Solute

V1 r100

p)

b

a

Benzene Toluene Phenol Nitrobenzene p-Nitrophenol Aniline p-Nitroaniline Acetophenone Benzaldehyde Benzyl alcohol Chlorobenzene Bromobenzene Iodobenzene Anisole Benzonitrile

0.491 0.592 0.536 0.631 0.676 0.562 0.702 0.690 0.606 0.634 0.581 0.624 0.671 0.639 0.590

0.59 0.55 0.72 1.01 1.15 0.73 1.25 0.90 0.92 0.99 0.71 0.79 0.81 0.73 0.90

0.10 0.11 0.33 0.30 0.32 0.50 0.48 0.49 0.44 0.52 0.07 0.06 0.05 0.32 0.37

0 0 0.61 0 0.82 0.26 0.42 0.04 0 0.39 0 0 0 0 0

306

M.H. Guermouche et al.r Fluid Phase Equilibria 147 (1998) 301–307

Fig. 1. Logarithm of theoretical capacity factor Žpredicted by LSER correlation. against logarithm of experimental capacity factor.

Binding the solute with the micelles can be expressed via K MW . In this case, LSER correlation gives: ln K MW s 0.36 q 7.71 P V1r100 q 0.60 P p ) y 5.10 P b q 0.53 P a Corresponding linear correlation is shown in Fig. 2. Compared with anionic Ž SDS. or cationic Ž C 14TAB. surfactants w20x, the term s is positive for the zwitterionic, negative for the cationic and anionic surfactants. C 12 DAPS has a greater dipolar

Fig. 2. Logarithm of theoretical K MW Žpredicted by LSER correlation. against logarithm of experimental K MW determined from Armstrong model.

M.H. Guermouche et al.r Fluid Phase Equilibria 147 (1998) 301–307

307

environment affecting considerably the MLC partition. In the other hand, the term a is positive, it is positive for C 14TAB and negative for SDS w20x. Acidic solutes binds more easily with the zwitterionic surfactant than with SDS. Further study will be carried with more solutes to explain the different interactions. 5. Conclusion Armstrong model describe correctly the retention of a series of solutes in micellar liquid chromatography with C 12 DAPS as zwitterionic surfactant. The different partitioning constants are determined. Compared to anionic or cationic surfactants, the zwitterionic surfactant gives the highest K SW values and the lowest K MW values. Hydrophobic interaction with the stationary phase, electrostatic effect of the surfactant from both the micelle and the surfactant modified stationary phase explain the retention of a solute. The C12 hydrocarbon chain present in C 12 DAPS explains the chromatographic behaviour of non-polar solutes. Acidic solutes such as phenol or p-nitrophenol are well separated because they are affected electrostatically by both the zwitterionic micelles and the covered stationary phase. LSER correlates satisfyingly the retention via ln kX. In zwitterionic C 12 DAPS MLC, V1r100 Žsolute’s size. and b Žbasicity. are predominant. LSER correlates also the binding of the solute to the micelles via K MW . The term s found is positive; C 12 DAPS has a greater dipolar environment affecting considerably the MLC partition. In the other hand, the term a found is positive. Acidic solutes binds more easily with the zwitterionic surfactant than with SDS. References w1x w2x w3x w4x w5x w6x w7x w8x w9x w10x w11x w12x w13x w14x w15x w16x w17x w18x w19x w20x w21x

D.W. Armstrong, F. Nome, Anal. Chem. 53 Ž1981. 1662. M.G. Khaledi, Biochromatogr. 3 Ž1988. 20. J.G. Dorsey, Chromatogr. 2 Ž1987. 13. F.J. Deluccia, M. Arunyanart, J. Cline-Love, Anal. Chem. 57 Ž1985. 1564. D. Habel, S. Guermouche, M.H. Guermouche, Analyst 118 Ž1993. 1511. J.P. Berry, S.G. Weber, J. Chromatogr. Sci. 25 Ž1987. 307. D. Habel, S. Guermouche, M.H. Guermouche, Biomed. Chromatogr. 11 Ž1997. 16. W. Hu, T. Takeuchi, H. Haraguchi, Chromatographia 33 Ž1992. 58. W. Hu, H. Haraguchi, Bull. Chem. Soc. Jpn. 66 Ž1993. 1967. J.G. Dorsey, M.T. De Echegaray, J.S. Landy, Anal. Chem. 55 Ž1983. 924. B.K. Lavine, S. Hendayana, J. Liq. Chrom. Rel. Technol. 19 Ž1996. 101. R.W. Taft, J.M. Abboud, M.J. Kamlet, M.H. Abraham, J. Solut. Chem. 14 Ž1985. 153. M.J. Kamlet, R.M. Doherty, M.H. Abraham, Y. Marcus, R.W. Taft, J. Phys. Chem. 92 Ž18. Ž1988. 5244. R.W. Taft, M.H. Abraham, G.R. Famini, R.M. Doherty, J.M. Abboud, M.J. Kamlet, J. Pharm. Sci. 74 Ž1985. 807. M.J. Kamlet, M.H. Abraham, P.W. Carr, R.M. Doherty, R.W. Taft, J. Chem. Soc. Perkin Trans. II 2087 Ž1988. . P.W. Carr, Microchem. J. 48 Ž1993. 4. R.S. Helburn, S.C. Rutan, J. Pampano, D. Mitchem, W.T. Patterson, Anal. Chem. 66 Ž1994. 610. P.W. Carr, R.M. Doherty, R.W. Kamlet, R.W. Taft, W. Melander, Cs. Horvath, ¨ Anal. Chem. 58 Ž1986. 2674. S. Yang, M.G. Khaledi, Anal. Chem. 67 Ž1995. 499. S. Yang, M.G. Khaledi, J. Chromatogr., A 692 Ž1995. 301. P. Yarmchuk, R. Weinberger, R.F. Hirsch, L.J. Cline Love, Anal. Chem. 54 Ž1982. 2233.