Complex Formation between Dodecylpyridinium Chloride and

A multiple regression analysis showed that the data of the corresponding standard free energy change of binding were well interpreted by the equation ...
0 downloads 0 Views 228KB Size
Langmuir 2004, 20, 8183-8191

8183

Complex Formation between Dodecylpyridinium Chloride and Multicharged Anionic Planar Substances Kiyofumi Murakami Faculty of Education, Yamaguchi University, Yoshida 1677-1, Yamaguchi 753-8513, Japan Received April 25, 2004. In Final Form: June 27, 2004 The complex formation between dodecylpyridinium chloride (DPC) and multicharged anionic planar substances, 14 azo dyes and 3 benzene- or naphthalenesulfonates, has been studied by the potentiometric titration using a surfactant selective electrode. The agreement between the observed maximum binding number and the number of anionic charges (n) on dye molecules showed n:1 complex formation. The binding isotherms were found to be composed of two types of binding; one is the noncooperative binding observed at low surfactant concentrations and the other is the cooperative binding at the higher concentrations. The microscopic binding constant for the noncooperative binding was found to take the values in the range of 50-200 mol-1 dm3 for many of the substances, but, takes more large values up to 2500 mol-1 dm3 for the substances which have a large hydrophobic part or the structure of separate hydrophobic and hydrophilic regions. A multiple regression analysis showed that the data of the corresponding standard free energy change of binding were well interpreted by the equation (in unit of kJ mol-1) ∆G° ) - 5.85 log PS - 1.68 log PD - 2.12z + 28.4, where PS and PD are the partition coefficients of the surfactants and planar substances in the 1-octanol/water system and z is the number of anionic charges on the planar molecules. At the beginning of the cooperative binding, precipitate formation was observed for almost all of the present systems. Among these, some of the dyes having the structure of separate hydrophobic and hydrophilic regions formed a needlelike crystal, which was accompanied by a hysteresis phenomenon in the binding isotherm. The stable complex formation by both the hydrophobic and electrostatic interactions between the surfactant and the planar substances was found to be important for the crystal formation. Depending on the manner of arrangement of the charged groups on the planar substances, the origin of the binding cooperativity was ascribed to the interactions between surfactants bound to one planar-substance molecule or to the association of the complexes. It was also found that the present small binding systems are useful as the model of ligand binding to protein local structures.

Introduction Ionic surfactants in aqueous solutions are known to form aggregates called micelles above a certain critical concentration (the critical micelle concentration, cmc). The interaction between micelles and dyes above the cmc has attracted much attention as the model system of the interaction between small molecules and biomembranes. Especially, the solubilization of dyes into the micelles has been studied in detail by spectrophotometric and kinetic approaches.1-5 From these studies, the detailed solubilization mechanism and the environment of solubilized small molecules have been explored. Below the cmc, on the other hand, ionic surfactants are also known to form complexes with dyes. It has sometimes been observed that the monoionic dyes such as Methyl Orange, Ethyl Orange, etc. show a characteristic blue shift in the visible absorption spectrum upon complex formation with oppositely charged surfactants.6-15 The origin of this (1) Micellization, Solubilization, and Microemulsions; Mittal, K. L., Ed.; Plenum Press: New York, 1977. (2) Takeda, K.; Tatumoto, N.; Yasunaga, T. J. Colloid Interface Sci. 1974, 47. 128-133. (3) Robinson, B. H.; White, N. C.; Mateo, C. Adv. Mol. Relax. Processes 1975, 7, 321-338. (4) Kubota, Y.; Omura, N.; Murakami, K. Bull. Chem. Soc. Jpn. 1991, 64, 814-820. (5) Dutta, R. K.; Chowdhury, R.; Bhat, S. N. J. Chem. Soc., Faraday Trans. 1995, 91, 681-686. (6) Hiskey, C. F.; Downey, T. A. J. Phys. Chem. 1954, 58, 835-840. (7) Forbes, W. F.; Milligan, B. Aust. J. Chem. 1962, 15, 841-850. (8) Quadrifoglio, F.; Crescenzi, V. J. Colloid Interface Sci. 1971, 35, 447-459. (9) Kim, B.-K.; Kagayama, A.; Saito, Y.; Machida, K. Bull. Chem. Soc. Jpn. 1975, 48, 1394-1396. (10) Vijlder, M. D. J. Chem. Soc., Faraday Trans. 1 1985, 81, 13691373.

phenomenon has been of great interest from the viewpoint of clarification of the circumstance and the state of dye in the complex. In one interpretation, this has been ascribed to ion-pair complex formation.8,9,12 It was suggested that both of the hydrophobic interaction between the dye ring system and the surfactant alkyl chain and the electrostatic interaction between the oppositely charged groups of the dye and surfactant, which are in a parallel alignment to each other, are important for the complex formation.12,14 In another interpretation, some aggregate forms are suggested to play a role in the spectral property.14-17 It can be easily recognized that the electrically neutral ion pair complexes have the tendency to aggregate. As a matter of fact, the complex formation between ionic dyes and oppositely charged surfactants below the cmc frequently accompanies precipitation.18 However, crystal formation from the complexes has scarcely been reported except for the systems of cationic surfactants and Methyl Orange or Methyl Red.6,8,15 Despite these many efforts, we have little of the detailed quantitative information for the complex (11) Dawber, J. G.; Fisher, D. T.; Warhurst, P. R. J. Chem. Soc., Faraday Trans. 1 1986, 82, 119-123. (12) Dutta, R. K.; Bhat, S. N. Bull. Chem. Soc. Jpn. 1993, 66, 24572460. (13) Wang, G.-J.; Engberts, J. B. F. N. Langmuir 1994, 10, 25832587. (14) Karukstis, K. K.; Savin, D. A.; Loftus, C. T.; D′Angelo, N. D. J. Colloid Interface Sci. 1998, 203, 157-163. (15) Buwalda, R. T.; Jonker, J. M.; Engberts, J. B. F. N. Langmuir 1999, 15, 1083-1089. (16) Mandal, A. K.; Pal, M. K. J. Colloid Interface Sci. 1997, 192, 83-93. (17) Fujieda, T.; Ohta, K.; Wakabayashi, N.; Higuchi, S. J. Colloid Interface Sci. 1997, 185, 332-334. (18) Mukerjee, P.; Mysels, K. J. J. Am. Chem. Soc. 1955, 77, 29372943.

10.1021/la048965+ CCC: $27.50 © 2004 American Chemical Society Published on Web 08/14/2004

8184

Langmuir, Vol. 20, No. 19, 2004

formation mechanism, since almost all of these studies below the cmc are of qualitative nature. For instance, we have no information about the magnitude of the electrostatic and hydrophobic force contributions, while these forces have been considered as acting on the ion-pair complex formation. Ionic surfactants and dyes are known to bind cooperatively or noncooperatively to proteins19-29 and linear polyelectrolytes.30-37 The origin of the cooperative binding has been thought to come from the hydrophobic and/or stacking interactions between alkyl chains of the surfactants or between aromatic parts of the dyes, which are bound to the oppositely charged protein-surface residues or polyelectrolyte groups by electrostatic interactions. In the case of the noncooperative binding of dyes to proteins, both of the hydrophobic and electrostatic interactions have also been considered to play important roles in the binding.25,26 In these interactions, the protein local structure and the manner of charge distribution on the protein surface are recognized to play essential roles. It is significant to clarify these binding mechanisms in more detail for understanding the colloidal properties of proteins, the initial stage of protein denaturation by surfactants, and the relationship between ligand bindings and protein functions. However, these macromolecular systems are too large and complex to be studied in detail. It is therefore informative to study small model systems of surfactant or dye binding to protein local structure. Ionic surfactants and dyes can be regarded as the protein local structure models, that is, surfactants can be regarded as the model of the hydrohobic grooves adjacent to a charged group and dyes as the planar hydrophobic patches having charged groups around them. In this study, the complex formation between a cationic surfactant (dodecylpyridinium chloride (DPC)) and multicharged anionic planar dyes or benzene- or naphthalenesulfonates has been investigated in terms of average number of bound surfactants, through the potentiometric titration using a surfactant selective electrode. The purpose is to consider the mechanism of the complex formation on the basis of quantitative data as well as to (19) Steinhardt, J.; Reynolds, J. A. Multiple Equilibria in Proteins; Academic Press: New York, 1969. (20) Hiramatsu, K.; Ueda, C.; Iwata, K.; Arikawa, K.; Aoki, K. Bull. Chem. Soc. Jpn. 1977, 50, 368-372. (21) Jones, M. N.; Manley, P. J. Chem. Soc., Faraday Trans. 1 1979, 75, 1736-1744. (22) Takeda, K. Bull. Chem. Soc. Jpn. 1982, 55, 1335-1339. (23) Fukushima, K.; Murata, Y.; Sugihara, G.; Tanaka, M. Bull. Chem. Soc. Jpn. 1982, 55, 1376-1378. (24) Subramanian, M.; Sheshadri, B. S.; Venkatappa, M. P. J. Biochem. 1984, 96, 245-252. (25) Murakami, K. Bull. Chem. Soc. Jpn. 1993, 66, 2808-2813. (26) Murakami, K.; Akamatsu, M.; Sano, T. Bull. Chem. Soc. Jpn. 1994, 67, 2647-2653. (27) Murakami, K. Bull. Chem. Soc. Jpn. 1998, 71, 2293-2298. (28) Murakami, K.; Tsurufuji, K. Bull. Chem. Soc. Jpn. 1999, 72, 653-659. (29) Murakami, K. Langmuir 1999, 15, 4270-4275. (30) Schwarz, G.; Klose, S.; Balthasar, W. Eur. J. Biochem. 1970, 12, 454-460. (31) Schwarz, G.; Balthasar, W. Eur. J. Biochem. 1970, 12, 461467. (32) Satake, I.; Yang, J. T. Biopolymers 1976, 15, 2263-2275. (33) Satake, I.; Gondo, T.; Kimizuka, H. Bull. Chem. Soc. Jpn. 1979, 52, 361-364. (34) Hayakawa, K.; Kwak, C. T. J. Phys. Chem. 1982, 86, 38663870. (35) Hayakawa, K.; Santerre, J. P.; Kwak, C. T. Biophys. Chem. 1983, 17, 175-181. (36) Shirahama, K.; Takashima, K.; Takisawa, N. Bull. Chem. Soc. Jpn. 1987, 60, 43-47. (37) Moriyama, Y.; Takeda, K.; Murakami, K. Langmuir 2000, 16, 7629-7633.

Murakami

discuss the binding systems as the model of the interaction between small molecules and protein local structures. Experimental Section Materials. DPC purchased from Tokyo Chemical Industry Co., Ltd., was purified from three recrystallizations from acetone solution. Disodium 1,3-benzenedisulfonates, disodium 2,6-naphthalenedisulfonate, and trisodium 1,3,6-naphthalenetrisulfonate, purchased from Tokyo Chemical Industry Co., Ltd., were purified by two recrystallizations from an aqueous sodium acetate solution and by two further recrystallizations from an aqueous ethanol solution, and were dried at 110 °C in a vacuum for 20 h. Sodium 4-(4-aminophenylazo)benzenesulfonate, sodium 6-hydroxy-5phenylazo-2-naphthalenesulfonate (Crocein Orange G), disodium 6-hydroxy-5-(4-sulfophenylazo)-2-naphthalenesulfonate (Sunset Yellow FCF), disodium 6-hydroxy-5-phenylazo-2,4-naphthalenedisulfonate (Orange G), disodium 3-hydroxy-4-(1-naphthylazo)-2,7-naphthalenedisulfonate (Bordeaux Red), trisodium 3hydroxy-4-(4-sulfonato-1-naphthylazo)-2,7-naphthalenedisulfonate (Amaranth), trisodium 6-hydroxy-5-(4-sulfonato-1-naphthylazo)-2,4-naphthalenedisulfonate (Ponceau 4R), and tetrasodium 6-hydroxy-5-(4-sulfonato-1-naphthylazo)-2,4,7-naphthalenetrisulfonate (Ponceau 6R) were purchased from Tokyo Chemical Industry Co., Ltd. Sodium 4-(2-hydroxy-1-naphthylazo)benzenesulfonate (Orange II), disodium 3-hydroxy-4-(2,4dimethylphenylazo)-2,7-naphthalenedisulfonate (Ponceau 2R), and disodium 3-hydroxy-4-(2,4,5-trimethylphenylazo)-2,7-naphthalenedisulfonate (Ponceau 3R) were purchased from Wako Pure Chemical Industries. Disodium 3-(2,4-dimethylphenylazo)-4hydroxy-2,7-naphthalenedisulfonate (Acid Red 8) and disodium 6-hydroxy-5-(1-naphthylazo)-2,4-naphthalenedisulfonate (Crystal Scarlet) were purchased from Aldrich Chemical Co. Disodium 6-hydroxy-5-(4-sulfonato-1-naphthylazo)-2-naphthalenesulfonate (Acid Red 13) was prepared by coupling sodium 4-amino1-naphthalenesulfonate with sodium 6-hydroxy-2-naphthalenesulfonate. These dyes were purified by three recrystallizations from an aqueous sodium acetate solution and by further three recrystallizations from an aqueous ethanol solution and were dried at 110 °C in a vacuum for 20 h. Chart 1 shows chemical structures of these planar substances. All the other chemicals used were of reagent grade. Methods. The binding isotherms for DPC binding to the dyes were obtained by potentiometric titration using a surfactantion-selective electrode.38 The surfactant-ion-selective membrane was composed of poly(vinyl chloride) (PVC), bis(2-ethylhexyl)phthalate, and DPC. The potentiometric measurements were made using a digital multimeter (Advantest TR6846) connected with the electrochemical cell: Ag/AgCl, KCl | salt bridge | reference solution | PVC membrane | sample solution | salt bridge | Ag/AgCl, KCl. In the cell, commercially available Ag/AgCl electrodes were used (TOA HS-205C). The temperature of the cell was maintained by circulating thermostated water. The slope of the plot of electromotive force (emf) vs log(surfactant concentration) below the cmc showed a good Nernstian slope, i.e., 58.9 mV/decade. The measurements were carried out at a dye concentration of 2.5 × 10-4 mol dm-3 and 25.0 ( 0.1 °C in the presence of 5 × 10-3 mol dm-3 NaCl. At first, 10 mL of 2.5 × 10-4 mol dm-3 dye solution was placed in the cell. Then calculated small amounts of dye and surfactant stock solutions were added using microsyringes with stirring so as to reach the desired concentrations 2 min after the stirring was stopped. The emf was read after it reached a constant value. The cmc of DPC under the present condition in the absence of dye was measured to be 1.4 × 10-2 mol dm-3 from the broken point in the Nernstian response.

Results and Discussion Binding Isotherms. Figure 1 shows the typical emf vs log[L0] plots for the binding of DPC to mono-, di-, tri-, and tetra-anionic dyes as well as the calibration curve (Nernstian response) measured in the absence of the dyes, (38) Shirahama, K.; Nishiyama, Y.; Takisawa, N. J. Phys. Chem. 1987, 91, 5928-5930.

Complex Formation between Surfactant and Planar Dyes

Langmuir, Vol. 20, No. 19, 2004 8185

Chart 1. Chemical Structures of the Planar Substances

where [L0] means the total surfactant concentration. Free surfactant concentration, [Lf], in any solution was read from the calibration curve. The amount of bound surfactant was calculated as the difference between [L0] and [Lf]

[Lb] ) [L0] - [Lf]

(1)

The average number of bound surfactants per one dye molecule was calculated by

νj ) [Lb]/[Do] where [D0] means the total dye concentration.

(2)

Figure 2 shows the binding isotherms for the binding of DPC to the above four dyes; Figure 2A, νj vs log[Lf] plot; Figure 2B, νj/[Lf] vs νj plot (Scatchard plot39). These plots show that DPC cooperatively binds to the dye molecules above certain concentrations by the amount equal to the number of anionic charges of the dyes. The agreement between the maximum binding number and the number of anionic charges (n) on the dye molecule, which was assured for all the dyes studied, shows the formation of an n:1 complex. We can see from Figure 2A the tendency that the concentration at which the cooperative binding (39) Scatchard, G. Ann. N. Y. Acad. Sci. 1949, 51, 660-672.

8186

Langmuir, Vol. 20, No. 19, 2004

Figure 1. emf vs log[L0] plots for the binding of DPC to multicharged anionic azo dyes: [, monoanionic dye (Crocein Orange G); O, dianionic dye (Acid Red 13); 2, trianionic dye (Amaranth); 0, tetra-anionic dye (Ponceau 6R).

Figure 2. Binding isotherms for the binding of DPC to multicharged anionic azo dyes: (A) νj vs log[Lf] plots; (B) νj/[Lf] vs νj plot (Scatchard plot39). The symbols used are the same as those in Figure 1.

takes place becomes large with the number of anionic charges on the dye molecules. From the comparison between the binding isotherms for the dyes having the

Murakami

Figure 3. Hysteresis phenomenon observed in the binding isotherm of DPC binding to Crystal Scarlet. The data measured in the standard procedure of the present measurements are shown by the circular symbol (O). To ensure the hysteresis phenomenon, the data were collected in increasing DPC concentration (4) and thereafter in diluting it (0). The retention time of the mixed solution in the later measurement was set to be longer than that of the ordinary measurements.

same backbone structure, although it is not shown as a figure, we found another tendency that the cooperative binding for the dyes, which have the structure of separate hydrophilic and hydrophobic regions, takes place at a lower concentration region than the region for the dyes having dispersively distributed amphiphilic groups. Furthermore, it was found that this tendency became pronounced as the hydrophobicity of the hydrophobic part became large. Precipitation, Crystallization, and Hysteresis in Binding Isotherms. At the initial stage of the cooperative binding (at low binding level), precipitation was observed for almost all of the present binding systems. Among these, a hysteresis phenomenon was observed in the binding isotherm for the mono- and dianionic dyes, which have the structure of separate hydrophobic and hydrophilic regions (Orange II, Crocein Orange G, sodium 4-(4aminophenylazo)benzenesulfonate, Bordeaux Red, Crystal Scarlet, Orange G, Acid Red 8). Figure 3 shows the hysteresis phenomenon observed for DPC-Crystal Scarlet system. In the standard procedure of the present binding measurements, νj gradually increased with an increase in surfactant concentration until the precipitation occurred (first phase: νj e 0.2), after the precipitation point νj suddenly began to increase along the line of larger binding number (second phase: νj g 1). Dilutions of the solution resulted in the decrease of νj along the inverse direction of the second phase. It can also be seen from this figure that the longer the retention time of the mixed solution becomes, the lower the νj value and the concentration at which the second phase appears become. This suggests that a phase-transition-like phenomenon is taking place in these systems. From some of these dye-surfactant solutions in the second phase, i.e., the cases of Orange II, Crocein Orange G, Bordeaux Red, Crystal Scarlet, needlelike colored crystals have been separated out. This fact suggests that the transition, observed in the binding isotherm from the first to the second phase, is the transformation from coagulates of the complex to the crystal. Figure 4 shows the crystals formed in the DPCOrange II solution ([DPC] ) [Orange II] ) 1 × 10-3 mol

Complex Formation between Surfactant and Planar Dyes

Langmuir, Vol. 20, No. 19, 2004 8187

neutral complex seems most feasible to associate. The vertical steps in the most left-side reactions are the selfassociations of the unbound dye, in which higher aggregates might be difficult to be formed because the electrostatic free energy of the aggregates will progressively increase with the association number. In any case, the associates of the neutral complex ((DLn)x) would be the most stable and predominant forms. Among the many possible paths including the surfactant binding steps and the association steps of the complexes, the path, which is adopted by any dye, would depend on the dye structure and the reactant concentrations. Binding Constant for the First Binding Step. Even if any path in eq 3 except the paths through D2 and D3 was adopted, the microscopic binding constant for the first bimolecular binding step can be evaluated from the data of the low surfactant concentration region in which the cooperativity does not appear as39

k1 )

Figure 4. The crystals formed in the solutions of (A) DPCOrange II and (B) DPC-Bordeaux Red systems. These crystals have been obtained by a gradual evaporation of the solvent water from the initial microcrystal solutions. The length of the measure depicted is 5 mm.

dm-3) (Figure 4A) and in the DPC-Bordeaux Red solution ([DPC] ) 1 × 10-3 mol dm-3, [Bordeaux Red] ) 5 × 10-4 mol dm-3) (Figure 4B). Thus far crystal formation from dye-surfactant solutions has scarcely been reported. Three-dimensional structure of these crystals is interesting from the structural viewpoint of clarifying the complex formation mechanism and is now under investigation. Scheme of the Complex Formation. The general scheme that takes account of the sequential bindings of surfactant molecules (L) to a multicharged dye (D) and the associations of the complexes can be expressed by eq 3.

The horizontal steps in this equation are the surfactant binding steps, which might proceed until all the charged groups are occupied by the surfactant cations. The vertical steps are the associations of the unbound dyes or of the complexes. For the sake of simplicity, the associations of differently charged complexes are not explicitly designated in this equation. From an electrostatic viewpoint, the

1 lim (νj/[Lf] ) n [Lf]f0

(4)

Here, the binding sites are assumed to be equivalent to each other. Figure 5 shows the νj/[Lf] vs log[Lf] plots for some differently charged dye systems. Table 1 summarizes the values of the microscopic binding constant evaluated as well as the values of the hydrophobicity parameter (log P) of the planar substances, which were calculated using a program of MaclogP 2.03 (BioByte Corp.). Here P means the partition coefficient in the 1-octanol/water system, which is the standard system for evaluation of the hydrophobic parameter. The calculation of log P is based on the fragment method that divides a molecule into fragments and carries out the sum of the contribution of each fragment type.40,41 k1 values for many of the dyes lie in the range of 50-200 mol-1 dm3. On the other hand, k1 for Bordeaux Red and Crystal Scarlet takes the relatively large values of 500 mol-1 dm3 and 750 mol-1 dm3, respectively. Further, those for Orange II and Crocein Orange G have still more large values, 2000 and 2500 mol-1 dm3, respectively. Figure 6 shows the log k1 vs log P plots. We can see from this figure that the binding of DPC to the dyes having the same number of charges becomes strong with an increase in hydrophobicity of dye, i.e., the hydrophobic interaction between the alkyl chain of DPC and the aromatic part of the dye as well as the electrostatic interaction between oppositely charged groups plays an important role in the binding. Further, it can be seen from the data that k1 value increases with an increase in the number of sulfonato groups on dye molecules in the region of log P < 1.5, but it can be expected that k1 decreases with the number of sulfonato groups in log P > 1.5. This means that the electrostatic interactions between oppositely charged groups are important for the systems of hydrophilic dyes, while the hydrophobic interactions are more contributable for the systems of hydrophobic dyes. The manner of arrangement of the sulfonato groups on dye molecules also affects the k1 value. For example, k1 for the dyes, which have the structure of separate hydrophobic and hydrophilic regions, has the larger values than those for the dyes having dispersively distributed hydrophilic groups; cf. the values of k1 for Acid Red 13, Bordeaux Red, and Crystal Scarlet (Table 1). This could be attributed to the difference in the tendency of forming a hydrophobic interaction between the alkyl chain of DPC (40) Leo, A. Chem. Rev. 1993, 93, 1281-1306. (41) Hansch, C.; Leo, A. Exploring QSAR: Fundamentals and Applications in Chemistry and Biology; American Chemical Society: Washington, DC, 1995; Chapters 4 and 5.

8188

Langmuir, Vol. 20, No. 19, 2004

Murakami

Figure 5. νj/[Lf] vs log[Lf] plots in a low surfactant concentration region for some differently charged dye systems: ], sodium 4-(4-aminophenylazo)benzenesulfonate; [, Crocein Orange G; O, Ponceau 2R; 4, Amaranth; 0, Ponceau 6R. Table 1. Binding Parameters for the Binding of DPC to Multicharged Azo Dyes and Benzene- and Naphthalenesulfonates substance

charges

k1/mol-1 dm3

log P

k/mol-1 dm3

u

4-(4-aminophenylazo)benzenesulfonate Orange II Crocein Orange G 1,3-benzenesulfonate 2,6-naphthalenesulfonate Sunset Yellow FCF Orange G Ponceau 2R Ponceau 3R Acid Red 13 Acid Red 8 Bordeaux Red Crystal Scarlet 1,3,6-naphthalenesulfonate Amaranth Ponceau 4R Ponceau 6R

1 1 1 2 2 2 2 2 2 2 2 2 2 3 3 3 4

125 2000 2500 25 50 180 100 150 150 125 200 500 750 70 167 167 125

0.712 2.34 2.34 -3.43 -2.26 -0.443 -0.443 0.555 1.00 0.73 0.448 0.73 0.73 -5.05 -2.06 -2.06 -4.84

283 12700 13500 20.1 22.2 181 482

20.2 4.9 5.5 29.5 25.7 17.5 11.7

939 1160 3500 569 4130 39.2 148 884 79.2

14.7 16.3 10.3 47.7 10.4 24.5 24.5 15.6 31.6

and the aromatic part of dye, as is schematically shown in Figure 7. In the case of the dye having the separate hydrophilic and hydrophobic regions (Figure 7A), the hydrocarbon tail of surfactant and the uncharged aromatic part of dye, which are both unstable in water, have the tendency to bind each other by dehydration of the hydrophobically hydrated water molecules (of some ordered structures) around them. In the case of the dye with a dispersed charge arrangement (Figure 7B), on the other hand, the disordered water structure around the sulfonato group42,43 weakens the hydrophobic interaction between the hydrophobic parts of the surfactant and dye. The cause of the fact that the monoanionic dye systems such as Orange II and Crocein Orange G have large k1 values can also be interpreted by this mechanism. The similar effect that the manner of charge arrangement on reactant molecules affects the hydrophobic interaction was

observed for the binding of mono- and dianionic azo dyes to bovine serum albumin (BSA).25 In this case, a large number of nonspecific binding sites for the monoanionic dyes (Orange II, Ponceau 4G), which were thought to be composed of a hydrophobic crevice and an adjacent cationic group, disappeared for the dianionic dye (Sunset Yellow FCF). The above relation between the structure of dye and the value of k1 is also in harmony with the relation between the dye structure and the concentration at which the cooperative binding takes place, the tendency of crystal formation, and the appearance of the hysteresis phenomenon. To consider quantitatively the contributions of the hydrophobic and electrostatic interactions to the standard free energy change of the first binding step (∆G°), it is convenient to use an extrathermodynamic approach.44,45 The data were analyzed by a multiple regression analysis using the hydrophobicity of dye and surfactant (log PD

(42) Tamaki, K.; Ohara, Y.; Kurachi, H.; Akiyama, M.; Odaki, H. Bull. Chem. Soc. Jpn. 1974, 47, 384-388. (43) Murakami, K.; Kimura, Y.; Saito, M. Bull. Chem. Soc. Jpn. 1997, 70, 115-121.

(44) Hansch, C.; Helmer, F. J. Polym. Sci., Part A-1 1968, 6, 32953302. (45) Helmer, F.; Kiehs, K.; Hansch, C. Biochemistry 1968, 7, 28582863.

Complex Formation between Surfactant and Planar Dyes

Langmuir, Vol. 20, No. 19, 2004 8189

Figure 6. log k1 vs log P plots for the present binding systems: [, monoanionic substances; O, dianionic substances; 2, trianionic substances; 0, tetra-anionic substance. The broken lines are the least-squares fits to the data for each class.

Figure 8. Plots of |∆G°| observed vs that calculated with eq 5: O, present data; b, data for Methyl Orange-cationic surfactant systems (ref 12); 0, data for alkyl Orange-alkyl trimethylammonium chlorides systems (ref 46).

Figure 7. Schematic representation of the effect of the manner of charge arrangement on dye molecules on the complex formation: (A) binding of the surfactant to the dye having a form of separate hydrophilic and hydrophobic regions, in which both the hydrophobic and electrostatic bonds can be formed; (B) binding of the surfactant to the dye with a dispersed charge arrangement, in which the hydrophobic bond might be weaken (see text).

parentheses show the standard deviations of the parameters. Figure 8 shows the plot of the observed |∆G°| vs the values calculated with eq 5, showing that the observed data are well interpreted by the equation. The following two points can be seen from eq 5. (1) The contribution of the surfactant hydrophobicity is larger than that of the dye hydrophobicity. This means that to lengthen the linear alkyl chain is more contributable than to spread the aromatic plane to the free energy change. (2) The contribution of the electrostatic interaction between unit opposite charges on the dye and surfactant molecules is -2.12 kJ mol-1. This value is reasonable, since it well agrees with that evaluated from the theory of ion association.25,47 This also agrees with the contribution of the electrostatic interaction between unit opposite charges for the binding of multicharged anionic dyes to BSA, which was evaluated as -2.85 kJ mol-1 by dividing the electrostatic contribution (-11.4 kJ mol-1)26 by the number of charges (+4) around the anionic-ligand binding site.48 Origin of the Cooperativity. Concerning the case of monoanionic dyes, it is clear that the cooperativity does not arise from the interaction between DPC molecules bound to more than one sulfonato groups on one dye molecule since monoanionic dyes have only one sulfonato group but must be due to the associations of the 1:1 complex K1

nd ) 34,

r ) 0.975,

F ) 194

where nd, r, and F are the number of analyzed data, the multiple correlation coefficient, and the F value, respectively. This equation is expressed so as to give the value in units of kJ mol-1. The numerical values in the (46) Gao, Y.; Motomizu, S. Bunseki Kagaku 1996, 45, 1065-1082.

K3

(6)

In this case, νj can be expressed by

and log PS) and the charge number of dye (z) as independent parameters, by combining the data for the other monoanionic azo dye-cationic surfactant systems.12,46 The result is

∆G° ) -5.85 log PS - 1.68 log PD - 2.12z + 28.4 (5) (0.96) (3.7) (0.41) (0.30)

K2

D + L 798 DL 798 (DL)2 798 (DL)3 T ‚ ‚ ‚

νj )

KAPP[Lf] 1 + KAPP[Lf]

(7)

where

KAPP ) K1 + 2K12K2[D][Lf] + 3K12K2K3[D]2[Lf]2 + ‚ ‚ ‚ (8) Equation 7 has the form of simple binding with the binding constant KAPP. The apparent binding constant KAPP, which (47) Neumann, M. G.; Gehlen, M. H. J. Colloid Interface Sci. 1990, 135, 209-217. (48) Hem, X. M.; Carter, D. C. Nature 1992, 358, 209-215.

8190

Langmuir, Vol. 20, No. 19, 2004

Murakami

is equal to K1 in the limit of low DPC concentration, becomes larger than K1 for higher concentrations of DPC; this means a positive cooperativity for DPC binding. For other schemes including complex-association processes, a similar cooperative effect might also be expected. In the case of multicharged dyes, on the other hand, there are two possible causes of cooperativity. One is the sequential DPC binding, expressed by

D T DL T DL2 T ‚ ‚ ‚ T DLn

(9)

The other is the association of these complexes. In the first case (eq 9), the cooperativity results from the stabilization of the complex species due to the hydrophobic interactions between the alkyl chains of DPC molecules bound to more than one sulfonato group on one dye molecule. The intrinsic binding constant (k) for a single site and the cooperativity parameter (u) may conveniently be estimated from the data of half-saturation point by using the next two equations, which were derived for the cooperative ligand binding to linear lattice under the assumptions of nearest-neighbor interaction and of binding-site equivalency.32

u1/2 ) 4 (dθ/dln[Lf])θ)1/2

(10)

k u [Lf]θ)1/2 ) 1

(11)

where θ is the degree of binding calculated as the number of surfactant-bound sites divided by the total number of binding sites. If the sequential binding mechanism does hold, the value of k must agree with that of k1 from a self-consistency requirement. The values of u and k evaluated are also listed in Table 1. We can see from this table that the value of k relatively well agrees with that of k1 for the dianionic substances (1,3-benzenesulfonates, 2,6-naphthalenesulfonates, Sunset Yellow FCF, Bordeaux Red), the trianionic substances (1,3,6-naphthalenesulfonates, Amaranth), and the tetra-anionic dye (Ponceau 6R). This suggests that eq 9 can be applicable to the systems of these substances. The common structural feature for these substances is that the sulfonato groups are located in dispersive manners on these molecules (except the case of Bordeaux Red). We can see that u has the tendency to decrease with an increase in the distance between charged sites (cf. u ) 29.5 (1,3-benzenesulfonates), u ) 25.7 (2,6-naphthalenesulfonates), and u ) 17.5 (Sunset Yellow FCF)). This fact reasonably shows that the interaction between ligands bound to neighboring sites is stronger than that between bound ligands remote to each other. The u values (17-30) are comparable with or somewhat smaller than those for the cooperative binding of dodecyltrimethylammonium bromide (u ) 48) and tetradecyltrimethylammonium bromide (u ) 147) to sodium dextran sulfate (DxS), which were measured in our laboratory.37 The average value of u for the systems to which eq 9 is applicable was calculated as 25.6. By use of the equation

∆G{int}° ) -RT ln u

(12)

the free energy change due to the interaction between two bound ligands can be evaluated as -8.0 kJ mol-1. Converting this value to the free energy change per unit of methylene group, we obtain -670 J mol-1. This is in agreement with the corresponding free energy decrease (-525 J mol-1) per unit of methylene group for the cooperative binding of alkyl sulfates to hen egg white lysozyme.29 This agreement suggests that the interactions

Figure 9. νj vs log[Lf] plots for the binding of DPC to Ponceau 2R.

between the surfactant molecules bound to these dyes or to the protein-surface residues are essentially the same hydrophobic interaction. For the other dyes that commonly have the structure of separate hydrophobic and hydrophilic regions, on the other hand, k values are from 5 to 18 times larger than k1 values (Table 1), suggesting that the sequential binding model (eq 9) should be discarded for these dye systems. Rather, associations of the complexes must play important roles in the total free energy decrease likely in the case of the monoanionic dyes. The structural characteristics suggests that the alkyl chain of the surfactant strongly interacts with the aromatic plane of these dyes (Figure 7A) and the stable neutral complex might have a planelike structure, which would be likely to stack successively to form crystals. This observation well agrees with the experimental result that these dyes have the strong tendency to form needlelike crystals on complexation with the surfactant. The stacking of the neutral complex (crystallization) might be the origin of the observed cooperativity for these systems. Among the dyes listed in Table 1, dianionic dye Ponceau 2R showed the different binding behavior that DPC cooperatively binds to Ponceau 2R in two phases (Figure 9), while the binding for all the other dianionic dyes take place in one phase. The first binding phase takes place until νj reaches unity in the concentration region of [Lf] from 1 × 10-4 to 2 × 10-4 mol-1 dm3, which is comparable to the cooperative binding region for the other dianionic dyes, and the subsequent second phase takes place from νj ) 1 to νj ) 2 in the higher concentration region from 1.5 × 10-3 to 4 × 10-3 mol-1 dm3, which corresponds to the cooperative binding region for the naphthalene and benzenesulfonates. The fact that the cooperativity appeared at νj < 1 suggests that some association of the charged complex is involved in the first phase to form a multicharged associate, subsequently cooperative sequential DPC bindings to the associate might take place in the second phase. The structure of Ponceau 2R is similar to those of Ponceau 3R and Acid Red 8 (Chart 1) but differs from them in the value of hydrophobicity (cf. Ponceau 2R and Ponceau 3R in Table 1) or in the arrangement of the substituent groups (cf. Ponceau 2R and Acid Red 8 in Chart 1). Although these structural differences must be the origin of the difference in the binding behavior, it seems difficult at present stage to interpret it in terms of the difference in interactions, which arises from the structural differences. However, the binding behavior of Ponceau 2R may be regarded as the intermediate behavior between the cooperative binding due to the association of the complexes and that due to the sequential DPC binding to one dye molecule. Aspects as the Model of Ligand Binding to Protein Local Structure. The results presented in this paper

Complex Formation between Surfactant and Planar Dyes

Langmuir, Vol. 20, No. 19, 2004 8191

showed the following facts. (1) The manner of charge arrangement on dye molecules, i.e., the form of separate hydrophobic and hydrophilic parts or the form of dispersively distributed hydrophilic groups, highly affects the k1 value for the first bimolecular binding step. This was interpreted by the ability of forming hydrophobic interaction between hydrophobic parts (Figure 7). In fact, this charge-arrangement effect has been observed in the system of mono- and dianionic dyes binding to BSA.25 (2) The contribution of the electrostatic interaction between oppositely charged groups to the total free energy change has quantitatively been evaluated (eq 5). The result was found to be consistent with the theory of ion association25,47 and with the data obtained for multicharged anionic dyesBSA systems.26 (3) The value of the cooperativity parameter (u), which gives the free energy change due to the interaction between bound surfactants, was found to be consistent with the data obtained for the cooperative surfactants binding to protein (lysozyme)29 and linear polymer (DxS).37 It might also be possible to presume the structure of binding sites on protein molecules as follows. The coefficients of the first and second terms in eq 5 represent the contributions of hydrophobicity of reactants to the free energy change, in terms of the values for the 1 decade increase in the partition coefficient. If we regard the surfactant and the dye as the binding site and the ligand, respectively, eq 5 means that the free energy of binding decreases by the amount of -1.68 kJ mol-1 for a 1-decade increase in ligand hydrophobicity. Contrarily, if we assume reverse roles, the corresponding free energy decrease amounts to -5.85 kJ mol-1. The contribution of ligand hydrophobicity to the free energy change of binding for the dye-BSA binding systems26 can be calculated from the data to be -4.91 kJ mol-1. Comparing these values,

we can presume that the binding site on BSA has the structure of a wide hydrophobic cavity like the planar dyes, rather than the narrow hydrophobic crevice like the shape of the surfactant. These observations show that the present small binding systems are the useful model systems of ligand binding to protein local structures. That is, we can presume the bound state of the ligand and the nature of interactions in real protein systems, from the data of these model systems. Conclusion DPC forms n:1 complexes with multicharged anionic planar substances such as azo dyes and benzene- and naphthalenesulfonates in cooperative manners. The dyes having the structure of separate hydrophobic and hydrophilic regions have the tendency to form needlelike crystals, exhibiting a hysteresis phenomenon in the binding isotherm. The formation of a stable complex by both the hydrophobic and electrostatic interactions between the surfactant and dyes is important for the strong binding and crystal formation. Depending on the manner of distribution of the hydrophobic and hydrophilic parts on dye molecules, the cooperativity originates from two processes; i.e., one is the sequential bindings of DPC to one dye molecule and the other is the association of the complexes. The present small binding systems are useful in considering the mechanism of ligand binding to protein local structures. Acknowledgment. I thank Ms. Yuko Yoshida for experimental assistance. LA048965+