J. Chem. SOC., Faraduy Trans. 1, 1982, 78, 2033-2040 Fluorescence and Absorption Spectral Study of the Interaction between Cetylpyridinium and 2-Naphtholate Ions in Aqueous Solution BY 'SOJI A. AMIRE AND HUGH D. BURROWS* Chemistry Department, University of Ife, Ile-Ife, Nigeria Received 30th January, 198 1 Absorption spectral studies on aqueous solutions of cetylpyridinium chloride and sodium 2-naphtholate indicate the formation of cetylpyridinium naphtholate complexes. The complex is shown to be non- fluorescent, and the main cause for the decrease in 2-naphtholate fluorescence intensities with increasing cetylpyridinium concentrations below the critical micelle concentration of the surfactant involves complex formation. At higher surfactant concentrations, quenching of fluorescence of 2-naphtholate ions is also suggested to occur via a dynamic interaction between these ions and cetylpyridinium micelles.Photophysical studies of micelle-forming surfactant solutions have been shown to yield valuable information on both the structure and dynamics of such systems.l Fluorescence quenching has been shown to be particularly valuable in this respect.l? Pyridinium ions are known to be efficient quenchers of the fluorescence of aromatic hydrocarbon^,^ and it is of interest to see whether, by using a suitable long-chain alkylpyridinium compound, information can be obtained on the interaction between such a system and a fluorescent solute molecule in aqueous solution. In the present study we have looked at the interaction between cetylpyridinium cations and 2-nap ht hola te anions.EXPERIMENTAL 2-Naphthol was purified by boiling with decolourising charcoal and recrystallizing from hot water. The white crystals so obtained were washed with water and dried in an oven. Cetylpyridinium chloride (CPCI, Koch-Light pure grade) was used without further purification. 2-Naphthol was dissolved in water with sufficient 0.1 mol dm-3 sodium hydroxide to ensure complete neutralization. Stock solutions of CPCl (0.1 mol dm-3) were prepared in water. To avoid potential problems with ageing of micellar systems,2 all experiments were carried out with freshly prepared solutions. The surfactant-naphtholate systems used in this study were prepared either by mixing appropriate volumes of the stock solutions in volumetric flasks or by adding small volumes of the CPCl solution directly with a microsyringe to 2-naphtholate solutions in cuvettes and correcting for the slight volume change this introduced.Identical results were obtained with the two procedures. The pH of the solution was adjusted by adding dilute potassium hydroxide or hydrochloric acid solutions. Buffers were not used because of problems of precipitation and the potential of common buffers to complex with the cetylpyridinium system. Fluorescence spectra were recorded using Aminco-Bowman and EEL 244 fluorimeters. Absorption spectra were measured on Pye-Unicam SP6-400 and SP6-500 spectrophotometers. For absorption spectra, solutions were thermostatted ( & 0.1 "C) with a water-circulating bath. Apparent equilibrium constants were obtained from absorption spectra using a modified Hill Whilst this is essentially an empirical method, it has been shown to be valuable in systems such as the present one where multiple equilibria are pos~ible.~ The method involves 20332034 CETY LP Y R I D I N I U M-2-NAP HTHOL A TE COMPLEXES plots of log [(A -A,,)/(Am - A ) ] against log [CPCI],,,,, where A , is the absorbance in the absence of CPCl, A is that in the presence of CPC1, and A , is the absorbance of the solution at complete complex formation (saturation).Free CPCl concentrations were calculated from initial concentrations and absorption spectra. It can be simply shown that at 50% saturation the intercept on the x axis is -log K, whilst the slope of the plot, n, gives a measure of the extent of aggregation5 As defined in t h s way, the apparent equilibrium constant, K , takes the units of concentration.The K values were corrected for the parallel reaction between 2-naphtholate and hydrogen ions as de~cribed.~ In all cases studied, the Hill plots were linear over the range 15-85 % complex formation. RESULTS On addition of CPCl to 2-naphthol it was noted that the solution became pale yellow in colour. This colour formation was seen to be reversible upon dilution. The absorption spectra were run of solutions of sodium 2-naphtholate ( lop4 to lop3 mol dm-3) alone, and in the presence of varying concentrations of CPCl. Typical spectra are shown in fig. 1. The lowest energy absorption band of the 1.5 1 .o aJ 5 e 2 I) m 0.5 0 \ \ \ \ \ w avele ngt h/n rn FIG.1 .-Electronic spectra of aqueous solutions of sodium 2-naphtholate ( mol dm-3, pH 1 1.85): solid line, alone; dashed line, in the presence of 5 x mol dm-3 cetylpyridinium chloride. 2-naphtholate ion shifted by ca. 10 nm to longer wavelengths, suggesting complex formation. Addition of a similar concentration of sodium chloride to 2-naphtholate solutions caused no such red shift, indicating that complex formation was between cetylpyridinium and naphtholate ions. Isosbestic points were observed around 325 and 335 nm, suggesting that there was only one equilibrium process involved. Plots of fractional saturation, a, determined at a particular wavelength against free cetylpyridinium concentration were observed to be sigmoidal. A typical plot is shown in fig.2. For a particular 2-naphthol concentration and pH, plots of absorbance at 355 nm against absorbance at 370 or 400 nm at various CPCl concentrations were linear. Apparent equilibrium constants were determined by studying the absorbance at 355nm, the absorption maximum of the complex, as a function of CPCl concentration. A typical Hill plot obtained is shown in fig. 3. The values obtained for apparent equilibrium constants are comparable to those obtained in other studiesS. A. AMIRE A N D H. D . BURROWS 2035 1 .o 0 . 8 h 0.6 a s \ h T I s 0.4 II tt 0.; 0 2 4 6 [ CPCl] free / 1 0-4 mol dm-3 FIG. 2. FIG. 3, FIG. 2.-Plot of extent of complexing against [CPCl]f,ee for solutions of 2-naphthol (1.04 x lop4 mol dmp3, pH 8.93, 27.0 "C) in the presence of cetylpyridinium chloride studied at the absorption maximum of the complex (355 nm).FIG. 3.-Hill plot for the system 2-naphthol (1.0 x rnol dm-3, pH 9.64, L = 355 nm) in the presence of cetylpyridinium chloride at 20.6 OC. TABLE 1 .-APPARENT EQUILIBRIUM CONSTANTS AS A FUNCTION OF pH AND 2-NAPHTHOL CONCEN- TRATION FOR THE SYSTEM 2-NAPHTHOL/CETYLPYRIDINIUM CHLORIDE IN AQUEOUS SOLUTION AT 27.0 *C [2-naphthol]/ mol dmP3 K / lo4 mol dm-3 PH 9.64 8.93 9.45 11.18 9.32 1 .oo 1.04 1.96 1.90 3.02 1.09 3.08 1.99 1.59 2.68 using this m e t h ~ d , ~ and are in a range where the Hill plot can be accurately applied to spectrophotometric data. For these studies, the concentration of CPCl was always less than its critical micelle concentration (c.m.c. of cetylpyridinium chloride 9.0 x mol dm-3).6 Nevertheless, the slopes of the Hill plots were > 1 in all cases, having a value of 1.5kO.1 at pH 3 9.6 (the pK, of 2-naphthol),' but increasing to 2.6k0.3 at lower pH values.This was independent of temperature over the range studied (293-307 K), and suggests that the surfactant molecules are starting to aggregate below the c.m.c. Apparent equilibrium constants were determined from the Hill plots, and were found to increase with decreasing pH; they also appeared to show some dependence on 2-naphthol concentration (table 1). However, this apparent2036 CETY L P Y R I D I N I U M-2-N A P H T HO L A TE COMPLEXES dependence in 2-naphthol may result from pH changes. Further studies are in progress to elucidate this factor. Apparent equilibrium constants were determined at constant pH and 2-naphthol concentration as a function of temperature.A semilogarithmic plot of formation constant against inverse temperature was linear (fig. 4). The following thermodynamic parameters for complex formation were determined* from this: A H 0 = -26.0k4.4 kJ mol-1 A S 0 = - 14.5+ 11.3 J K-l mol-l. The enthalpy falls within the range of values reported for organic charge-transfer complexes between n-electron donors and acceptors.8 KIT FIG. 4.-Semilogarithmic plot of the apparent equilibrium constant for the 2-naphtholate-cetylpyridinium complex against inverse temperature. To gain further insight into the interaction occurring between the two ions, the fluorescence of the 2-naphtholate ion mol dm-3) was studied in the presence of varying concentrations of CPCl.Whilst the fluorescence emission and excitation spectra were found to be essentially independent of CPCl concentration, the intensity of the fluorescence was seen to decrease dramatically upon increasing the surfactant concentration (fig. 5). Addition of similar concentrations of chloride caused no such decrease in fluorescence, suggesting that the behaviour was due to interaction of cetylpyridinium and 2-naphtholate ions. This conforms with the known ability of alkylpyridinium compounds to quench the fluorescence of aromatic compo~nds.~ This decrease in intensity can arise from two causes, either static quenching by formation of a non-fluorescent complex, reaction (3), or dynamic quenching involving interaction between excited 2-naphtholate anion (BN-) and cetylpyridinium cation (CP+) (4).BN- + Av -+ lBN* monomer excitation (1) lBN* -+ BN- + kv monomer fluorescence (2) BN- + CP+ --+ (BN - CP) formation of non-fluorescent complex (3) 'BN* + CP+ -+ BN- + CP+ dynamic quenching (4) The fact that no shift in the emission spectrum was observed on complexation suggestsS. A. AMIRE AND H. D. BURROWS 2037 0 2.5 5.0 7.5 [ CPCll / 1 0-4 mol dm-3 FIG. 5.--Intensity of fluorescence of 2-naphtholate ion ( mol dmP3, ,Iexcitation = 350 nm, Aemission = 420 nm) as a function of CPCl concentration : circles, experimental data; solid line, theoretical curve from eqn (1 2). strongly that the complex was non-fluorescent. If only free naphtholate ion is fluorescing, it is possible to derive an equation relating the fluorescence intensity to its concentration and to the fraction of light absorbed at the excitation wavelength by fluorescer and complex. Using these assumptions, the intensity of light at this wavelength at any position in the cell, I, will be3 ( 5 ) I’ = I,, X 1 O-(‘BN ‘BN+&comp ‘comp) where E and c are the molar absorptivities and concentrations, respectively, of the absorbing species.Cetylpyridinium is not included in this as it does not absorb at the excitation wavelength. The amount of light absorbed by free 2-naphtholate ion in a length dl will be d I = (In 10) I,, X 1 O-(&BN ‘BN+&comp 'camp) EBN cBK dl. (6) This can be integrated to give the intensity of light absorbed over the length, x, from which emission is observed (7) (In 10) I,, EBN cBN( 1 - 1 O-(&BN CBK+EcomP ccomp) ”) (EBN CBN + Ecomp Ccomp) x I = The intensity of fluorescence, If, will be proportional to this K ’ c B N ( 1 - lO-(&BN CBNSEcompCcomp) ”) If = EBN CBN +Ecomp Ccomp where K’ is a constant for a particular fluorescer and fluorimeter geometry.If the total BN concentration, cT, remains constant, then Ccomp = CT-CBN K’cB,( 1 - 10-[CBN(EBN-Ecomp)+Ecomp ‘TI ”) and If = CBN(EBN - Ecomp) + Ecomp CT If we make the further assumption that for the denominator2038 CETYLPYRIDINIUM-2-NAPHTHOLATE COMPLEXES which is valid, except at very low CPCl concentrations, then where A and D are constant for a particular total BN concentration, excitation wavelength and fluorimeter, and B is a constant involving only the difference in molar absorptivities of BN and the complex, and x.By using values of cBN, either from absorption spectra or from apparent equilibrium constants, it is possible to construct a theoretical plot of fluorescence intensity against CPCl concentration. Good agreement was observed between theory and experiment under the conditions employed ([BN] = mol dm-3, pH 9.64, t = 25 OC, Aexcitation = 350 nm, Lemission = 420 nm) using A = 2.1 1 x lo8, B = 1000, and D = 0.018 (fig. 5). Interpret- ation of A and D is complex, as these terms involve intensity of exciting light, cT, etc. However, B involves x and the difference between the molar absorptivities at the excitation wavelength, and, given the fact that both complex and 2-naphtholate absorptions are changing very rapidly with wavelength in this region, the best-fit value is certainly of the correct order of magnitude.The results strongly support the suggestion that the decrease in fluorescence intensity at these CPCl concentrations arises primarily from the formation of a non-fluorescent complex. At concentrations around or above the c.m.c., an additional mechanism of reduction of fluorescence intensity appears to be operating. Whilst increasing the total CPCl concentration from 5 x to 10 x lov4 mol dm-3 only decreases the free 2-naphtholate concentration by a factor of 1.7, the fluorescence intensity over the same range decreases by a factor of five. The effect shows up most clearly when the fluorescence data are replotted in the form of a Stern-Volmer plot (fig.6). At low CPCl concentrations the plot is non-linear, as expected from the formation of a non-fluorescent complex absorbing strongly at the excitation wavelength. Above ca. 5 x mol dm-3 CPCl; however, the plot* is linear, with slope 9.32 x lo4 dm3 mol-l. Quantitative interpretation of the behaviour is difficult, as the data must be corrected for light absorption by the non-fluorescent complex, and for cases where more light is absorbed by the complex than by the fluorescer errors are likely to be large.3 A rough correction for this may be made by using experimental values from absorption spectral data at the excitation wavelength. From these, the slope of the linear region is found to be ca. lo4 dm3 mol-l. The high value for this quenching constant, and the fact that the linear region starts at CPCl concentrations close to the c.m.c., suggests that this probably is due to dynamic quenching of excited 2-naphtholate ions by cetylpyridinium micelles. DISCUSSION The absorption spectral changes observed when CPCl is added to aqueous solutions of 2-naphtholate ions provide clear evidence for complex formation. That this complex has charge-transfer character is suggested by the red-shift of the lowest energy absorption, and by the known tendency of alkylpyridinium ions to form charge-transfer complexe~.~~ lo The ionisation potential of 2-naphtholate ion? is certainly favourable for the formation of such a complex. Studies of the association equilibrium suggest that more than one cetylpyridinium monomer unit is present in each complex.Whether the anion is assisting aggregation or whether there are already submicellar * Because of the wide variation in fluorescence intensities in this study, the concentration range which can be studied at a particular instrument sensitivity is limited. However, studies on different sensitivity settings indicate that the Stern-Volmer plot is linear to higher concentrations than those indicated in fig. 6. t A value of 7.83 eV is reported for 2-naphtho1," quoted in ref. (12). The ionisation potential of the anion should be even lower.S . A. AMIRE AND H. D. BURROWS 2039 80 60 * 1 +o 40 20 01 0 0 o o I I 5 10 1 0-4 [ CPCl] /mol dm-3 FIG. 6.-Stern-Volmer plot for the quenching of fluorescence of 2-naphtholate ion ( mol dm-3) by CPCl at room temperature.aggregates of cetylpyridinium ions at CPCl concentrations corresponding to the onset of complex formation is not clear. Submicellar aggregates of surfactants have been reported with carboxylic acid soaps,13 but the existence of such aggregation remains contro~ersial.~~ The increase in the slope of the Hill plots, n,. when the pH is below the pK, of 2-naphthol suggests that there is an increase in the extent of aggregation in this case. This is supported by an increase in the apparent equilibrium constant with decreasing pH. The nature of the multiple binding in these complexes is not clear. However, both the sigmoidal shape of the plot of fractional saturation against concentration (fig. 2), and the fact that n > 1 strongly suggest15 that the binding sites in the aggregates are not independent, and that binding of one ligand increases the affinity of the aggregate for binding a second ligand.The high values observed for the apparent equilibrium constants, which are comparable to those found for the binding of various ligands to haem~proteins,~ may indicate that the naphtholate ions are binding in a region where the dielectric constant is lower than that of the bulk water. Studies of fluorescence intensity as a function of surfactant concentration indicate that the complex is non-fluorescent. In 1 -(9-anthrylmethyl)pyridinium chloride, the pyridinium ring is seen to quench the anthracene fluorescence efficiently by an intramolecular me~hanism.~ For such efficient quenching, however, the pyridinium and aromatic species must be close to each other.lH n.m.r. studies of aqueous solutions of various phenols in the presence of surfactant systems strongly suggest that they are solubilised in the palisade layer of the micelle, with the hydroxy group pointing towards the bulk aqueous so1ution.l69 l7 Specific charge-transfer interactions in the present case favour the 2-naphtholate and cetylpyridinium ions being adjacent, even in the absence of full micellar aggregation.2040 C E T Y L P Y R I D I N I U M-2-N A P H T HO L A TE C 0 M P L E X E S At higher CPCl concentrations the Stern-Volmer plot of the quenching of 2-naphtholate fluorescence is linear. This appears to be due to dynamic quenching of 2-naphtholate excited state by cetylpyridinium micelles. Using the corrected Stern-Volmer quenching constant, and the lifetime of the 2-naphtholate fluorescence (8.1 ns),18 an apparent second-order rate constant of ca.10l2 dm3 mol-l s-l was obtained for this quenching. A ‘true’ rate constant ( i e . in terms of concentration of micelles) can be estimated by dividing this apparent rate constant by the micellar aggregation number ( n = 95 for cetylpyridinium chloride).lg This gives a value ca. 1O1O dm3 (mole of micelle)-l s-l. This is comparable to the diffusion-controlled rate in aqueous solution, and can be contrasted with the rate constant for reaction between hydrated electrons and cetylpyridinium micelles [ca. 5 x 10l2 dm3 (mole of micelle)-l S - ~ ] . ~ O The extremely high rate constant in this latter case is suggested to arise from the high electrical potential of the micelle double layer.In the present case, the surface charge may be partially neutralised by naphtholate ions already bound to the micelle. K. Kalyanasundaram, Chem. SOC. Rev., 1978, 7, 453. * See, for example, H. D. Burrows, S. J. Formosinho, M. F. J. R. Paiva and E. J. Rasburn, J. Chem. SOC., Faraday Trans. 2, 1980, 76, 685, and references therein. R. A. Hanna, D. R. Rosseinsky and T. P. White, J. Chem. SOC., Faraday Trans. 2, 1974, 70, 1522. A. V. Hill, J. Physiol. (London), 1912, 40, 4. A. C. Anusiem, J. G. Beetlestone and D. H. Irvine, J. Chem. SOC. A , 1968, 960. E. J. Fendler and J. H. Fendler, Advances in Physical Organic Chemistry, ed. V. Gold (Academic Press, London, 1970), vol. 8, p. 276. A. Albert and E. P. Serjeant, The Determination of Ionization Constants (Chapman and Hall, London, 2nd edn, 1971), p. 87. R. S. Mulliken and W. B. Person, Molecular Complexes (Wiley, New York, 1969), p. 89. E. M. Kosower, Progress in Physical Organic Chemistry, ed. A. Streitwieser and R. W. Taft (Intersci- ence, New York, 1965), vol. 3, p. 81. lo R. Foster, Organic Charge-transfer Complexes (Academic, London, 1969). l 1 K. Beakers and A. Szent-Gyorgyi, Red. Trav. Chim. Pays-Bas, 1962, 81, 255. l2 J. B. Birks, Photophysics of Aromatic Molecules (Wiley-Interscience, London, 1970), p. 459. l3 D. Eagland and F. Franks, Trans. Faraday SOC., 1965, 61, 2468. l4 L. R. Fisher and D. G. Oakenfull, Chem. SOC. Rev., 1977, 6, 25, and references therein. l5 E. Antonini, Science, 1967, 158, 1417. l6 J. J. Jacobs, R. A. Anderson and T. R. Watson, J. Pharm. Pharmacol., 1971, 23, 148. li F. Tokiwa and K. Aigami, Kolloid Z. 2. Polym., 1971, 246, 688. A. Weller, 2. Phys. Chem. (Frankfurt am Main), 1958, 17, 224. l9 E. W. Anacker, J , Phys. Chem., 1958, 62, 41. 2o M. Gratzel, J. K. Thomas and L. K. Patterson, Chem. Phys. Lett., 1974, 29, 393. (PAPER 1/144)