J. Phys. Chem. B 2007, 111, 13645-13652
13645
Thermodynamic Studies of Drug-r-Cyclodextrin Interactions in Water at 298.15 K: Promazine Hydrochloride/Chlorpromazine Hydrochloride + r-Cyclodextrin + H2O Systems Santosh S. Terdale, Dilip H. Dagade, and Kesharsingh J. Patil* Department of Chemistry, ShiVaji UniVersity, Kolhapur - 416 004, India ReceiVed: July 12, 2007; In Final Form: September 16, 2007
Data on osmotic coefficients have been obtained for a binary aqueous solution of two drugs, namely, promazine hydrochloride (PZ) and chlorpromazine hydrochloride (CPZ) using a vapor pressure osmometer at 298.15 K. The observed critical micelle concentration (cmc) agrees excellently with the available literature data. The measurements are extended to aqueous ternary solutions containing fixed a concentration of R-cyclodextrin (R-CD) of 0.1 mol kg-1 and varied concentrations (∼0.005-0.2 mol kg-1) of drugs at 298.15 K. It has been found that the cmc values increase by the addition of R-CD. The mean molal activity coefficients of the ions and the activity coefficient of R-CD in binary as well as ternary solutions were obtained, which have been further used to calculate the excess Gibbs free energies and transfer Gibbs free energies. The lowering of the activity coefficients of ions and of R-CD is attributed to the existence of host-guest (inclusion)-type complex equilibria. It is suggested that CPZ forms 2:1 and 1:1 complexed species with R-CD, while PZ forms only 1:1 complexed species. The salting constant (ks) values are determined at 298.15 K for promazine-R-CD and chlorpromazine-R-CD complexes, respectively, by following the method based on the application of the McMillan-Mayer theory of virial coefficients to transfer free energy data. It is noted that the presence of chlorine in the drug molecule imparts better complexing capacity, the effect of which gets attenuated as a result of hydrophobic interaction. The results are discussed from the point of view of associative equilibria before the cmc and complexed equilibria for binary and ternary solutions, respectively.
1. Introduction There is great interest in the thermodynamic properties associated with biochemical processes, typically solvation, association, binding, conformational changes, unfolding, and so forth in aqueous solutions.1 The model systems studied have benefited us to advance molecular interpretation in a variety of areas. Microcalorimetry has gained wide acceptance in biochemistry to study biopolymer-ligand interactions.2 The association studies between β-cyclodextrin and various drugs have illustrated the use of Flow microcalorimetry3 by elegant means to obtain association constants and enthalpies.4 In recent times, the work related to the determination of thermodynamic equilibrium constants for host-guest-type equilibria for 18C6alkali halides has demonstrated the utility of vapor pressure osmometry.5 Thermodynamic parameters such as activity, activity coefficient, and application of the McMillan-Mayer theory of solutions to evaluate the pair and triplet free energy coefficients have enabled us to stress the importance of hydrophobic interaction and conformational changes in aqueous solutions of these supramolecular ionic entities.6,7 Parallel to such studies, the association of amphiphilic drugs in aqueous solutions is interesting; however, these are difficult because the association is micellar. There is a large amount of work reported by Attwood et. al concerning enthalpy (microcalorimetry), volumes, compressibility, and osmotic and activity coefficients in binary aqueous solutions of amphiphilic drug molecules, generally at 303 K.8-13 The phenothiazine tranquilizing drugs exhibit interesting association characteristics that are derived from their rigid, tricyclic hydrophobic groups. Attwood et. al * Corresponding author. E-mail:
[email protected].
have studied chlorpromazine hydrochloride (CPZ) over the temperature range of 273-303 K in aqueous solutions by measuring various properties14 (conductivity, vapor pressure, etc.). The results are accounted for by applying Burchfield and Woolley’s mass action model and a numerical method by PerczRodriguez.14-15 The micellization of the drug becomes increasingly exothermic with an increase in temperature, and the enthalpy change involved in the formation of the aggregates is progressively lower. According to these authors, the surface and hydrophobic contributions to the free energy are almost constant in the studied temperature range. Recently, we examined the enthalpy and entropy changes for aqueous solutions of R-cyclodextrin (R-CD) in the temperature range of 293.15-313.15 K, and the observed results are attributed to conformational changes and enthalpy-entropy compensation phenomena as well as the system behaving as an ideal system at 301 and 311 K.16 Considering all these factors, we have extended our osmotic vapor pressure studies to aqueous solutions of promazine hydrochloride (PZ) and CPZ at 298.15 K. Further, the binding of these drug molecules with R-CD in aqueous solutions of drugs at different concentrations and a fixed R-CD concentration is also studied. The activity coefficients of all three components are determined by use of the proper integration method and are used to evaluate transfer free energy data for the drug molecules. Applying a method based on the application of the McMillanMayer theory of solutions, the salting constant values are determined.5 These results are presented and discussed below. 2. Experimental Section 2.1. Materials. R-CD procured from Lancaster, London, and the drugs 2-Chloro-10-(3-dimethylaminopropyl) phenothiazine
10.1021/jp0754381 CCC: $37.00 © 2007 American Chemical Society Published on Web 11/08/2007
13646 J. Phys. Chem. B, Vol. 111, No. 48, 2007
Terdale et al. TABLE 1: Coefficients Ai in Eq 2 A2 A3 A4 A5 A6 A7 A8 A9 A10 A11 A12
Figure 1. Variation of osmotic coefficient (φ02) as a function of concentration of drug in aqueous drug solutions at 298.15 K.
CPZ
PZ
-56.6169 943.6803 -6614.6677 18885.6546 -18628.2414 10215.5855 -24912.3334 -57394.3334 112287.1106 65749.9591 17423.7830
3.1680 61.0980 -915.5964 2332.3669 -151.7748 -1769.6321 -1840.7656 -1304.3150 -768.5605 -402.4410 -203.4580
converted to molarity scale whenever required using density data. The density measurements of aqueous solutions of CPZ and PZ were made using an Anton Paar digital densitometer (DMA 60/602) at 298.15 K. The uncertainty in the density data was found to be (5 × 10-3 kg m-3. 2.4. Osmotic Coefficient Measurements. The osmotic coefficients of aqueous drug solutions were measured using a KNAUER K-7000 vapor pressure osmometer having temperature control of (0.001 K. The uncertainty in the osmotic coefficient data was found to be (1 × 10-3. The details about the density and osmotic pressure measurements have been reported earlier.5,17 3. Results 3.1. Binary Aqueous Drug Solutions. 3.1.1. Osmotic and ActiVity Coefficients. The osmotic coefficient (φ02) values were determined for aqueous binary drug solutions in the concentration range of ∼0.005-0.2 mol kg-1 at 298.15 K. Using the data of the osmotic coefficient, the water activity (aw) values were estimated using the equation
[ ( )]
φ02 ) - ln aw/
Figure 2. Variation of osmotic coefficient (φ02) as a function of the reciprocal of the concentration of drug in aqueous drug solutions at 298.15 K.
(CPZ) and 10-(3-dimethylaminopropyl) phenothiazine (PZ) procured from Sigma-Aldrich, were used without further purification. The structures of the drugs are given below:
R ) CH2CH2CH2N(CH3)2, and X ) H (for PZ) and Cl (for CPZ) 2.2. Karl Fischer Titration. The amount of water of hydration in R-CD (6.24 water molecules per R-CD molecule) was estimated using thermogravimetric analysis (model: thermal analyzer, TG-DTA-DSC, TA, Inc. SDT-2790) and microprocessor-controlled automatic Karl Fischer titrator (model: TKF-55, from M/s Toshniwal Company) analysis. The details about the water of hydration and the preparation of solutions of R-CD have been given elsewhere.16 2.3. Density Measurements. All the solutions were prepared using doubly quartz-distilled water on a molality basis and
x2 x1
(1)
where x1 and x2 are the mole fractions of water and the drug, respectively. The water activity was used to obtain the solvent activity coefficient (γ00). The experimental osmotic coefficient data for these systems can be expressed as a function of concentration using the polynomial equation18
φ02 ) 1 -
n 2.303 Aγz+z-xm + Aimi/2 3 i)2
∑
(2)
where φ02 is the osmotic coefficient of drug molecules in aqueous binary solutions, and Aγ is the Debye-Hu¨ckel constant equal to 0.511 for aqueous solutions at 298.15 K. The variations of the osmotic coefficient as a function of drug concentration as well as the reciprocal of drug concentration, for both systems, are given in Figures 1and 2, respectively. The data are collected in the Supporting Information. The coefficients Ai in eq 2 were obtained by the method of least-squares and are given in Table 1. The mean molal activity coefficient of a drug molecule (γ02) in binary aqueous solutions can be expressed in terms of the osmotic coefficient using the equation
ln γ02 ) (φ02 - 1) + 2
∫0
xm
(φ02 - 1) dxm xm
(3)
Using eq 2 and solving the right-hand-side integral of eq 3, one can write
Drug-R-CD Interactions in Water n
ln γ02 )
∑ i)1
J. Phys. Chem. B, Vol. 111, No. 48, 2007 13647
2+i i
Aimi/2
(4)
The mean molal activity coefficient (γ02) values were calculated using eq 4, and the data for both the systems are collected in the Supporting Information. The variation of ln γ02 with the drug concentration is shown in Figure 3. The activity coefficient data, which had been converted to mole fraction scale, were used to calculate the excess Gibbs free energy (∆GEX) of aqueous drug solutions, and the values are given in the Supporting Information for the corresponding systems. 3.1.2. Densities and Apparent Molal Volumes. The apparent molal volumes were obtained from the density data at 298.15 K. The apparent molal volumes in the premicellar region are expressed by the equation
φV ) φ0V + AVm1/2 + BVm
(5)
where φ0V is the apparent molar volume at infinite dilution, AV is the Debye-Hu¨ckel limiting law coefficient, and BV is the deviation constant. The variation of the parameter φV - AVc1/2 against c for both of the binary drug systems is shown in Figure 4. 3.1.3. Application of the McMillan-Mayer Theory. The relative magnitudes of solute-solute and solute-solvent interactions have been studied for nonelectrolytes using statistical mechanical theories.19 In dilute solutions of up to 0.1 mol kg-1, if the Debye-Hu¨ckel electrostatic contribution is subtracted from a thermodynamic parameter (e.g., ln γ02), then the remainder is linear in the molality, as it is for a nonelectrolyte.20 The mean molal activity coefficient of the solute (γ02) in the dilute concentration range can be represented as
ln γ02 ) -Rm1/2(1 + bm1/2)-1 + $m
Figure 3. Variation of the activity coefficient of a drug (ln γ02) as a function of the concentration of drug in aqueous drug solutions at 298.15 K.
(6)
where R ) 1.173 kg1/2 mol-1/2 at 25 °C, b ) 1.0 kg1/2 mol-1/2, and ω is the nonelectrolyte solute-solute interaction parameter. According to McMillan and Mayer,21
π ) n + B/22n2 + B/222n3 + ........ kT
(7)
where n is the number density of the solute, and B*22 and B*222 are the osmotic second and third virial coefficients for a solutesolute interaction, respectively. Hill22 has shown that the coefficient A22 and so forth, in the free energy expression, may be related to the coefficient B*22 and so forth. For example,
A22υ01
)
2B/0 22
-
υ j 02
+
b011
(8)
where υ01 and υ j 02 are the molecular volume of the pure solvent and the partial molecular volume of the solute at infinite dilution, respectively, and b011 () -B/0 11) is the solute-solvent cluster integral. For a 1:1 electrolyte,20
j + B222m j2 2 ln γ/2 ) A22m
(9)
where γ/2 is the nonelectrolyte contribution to the solute activity coefficient, m j is mole ratio of solute to solvent (N2/ N1), and ω ) A22M1/2 (M1 is the molar mass of the solvent in kg mol-1).
Figure 4. Variation of (φV - 1.868c1/2) as a function of concentration of drug in aqueous drug solutions at 298.15 K. *0 TABLE 2: Solute-Solvent (NB*0 11 ) and Solute-Solute (NB22 ) Virial Coefficients of Drug Molecules in Water at 298.15 K
CPZ PZ
T/K
3 -1 NB/0 11/cm mol
3 -1 NB/0 22/cm mol
298.15 298.15
268.88 257.68
-25546.35 5916.29
Thus from eq 8 and using the relation b011 ) -υ02 + kTκT, where k is the Boltzmann constant, T is the absolute temperature, and κT is the isothermal compressibility coefficient of the pure solvent, one can write 0 NB/0 h 02 - RTκT/2 22 ) A22V1/2 + V
(10)
The value for the solute-solute virial coefficient (B/0 22) has and B/0 been calculated by obtaining the value of ω. The B/0 22 11 values at 298.15 K for CPZ and PZ are collected in Table 2. 3.2. Ternary Aqueous Mixtures of R-CD and Drug. 3.2.1. ActiVity Coefficients of Electrolytes and Nonelectrolytes. The molal activity coefficient of the nonelectrolyte γ1 (R-CD) in the ternary system is expressed as a function of concentration of the nonelectrolyte m1 and the electrolyte m2 using the equation
13648 J. Phys. Chem. B, Vol. 111, No. 48, 2007 ∞
ln γ1 )
Terdale et al.
∞
TABLE 3: Coefficients Aij in Eq 15
∑ ∑ Aijm1im2j (A00 ) 0) i)0 j)0
(11)
If we expand eq 11, including all the fourth-order terms, we get5
ln γ1 ) ln γ01 + A01m2 + A11m1m2 + A02m22 + A03m23 + A21m12m2 + A12m1m22 + A31m13m2 + A22m12m22 + A13m1m23 + A04m24 (12) where γ01 is the activity coefficient of the nonelectrolyte (RCD) in binary aqueous solutions, represented as ln γ01 ) A10m1 + A20m12 + A30m13 + A40m14. It is taken from the data of binary aqueous R-CD solutions reported earlier.16 A similar expression for ln γ2 may be obtained by applying a cross-differentiation relation:
( ) ( ) ∂ ln γ1 ∂m2
)2
m1
∂ ln γ2 ∂m1
A01 A11 A02 A03 A21 A12 A31 A22 A13 A04
The result is5
ln γ2 ) ln γ02 + (1/2)A01m1 + (1/4)A11m12 + A02m1m2 + (3/2)A03m1m22 + (1/6)A21m13 + (1/2)A12m12m2 +
TABLE 4: Pair and Triplet Interaction Parameters, and Salting Coefficients of the System H2O + 0.1 M r-CD + Drug, at 298.15 K
CPZ PZ
gNE/J kg mol-2
gNNE /J kg2 mol-3
gNEE/J kg2 mol-3
ks /kg mol-1
-12412.2 -6130.1
-411.5 -193.6
42218.7 9179.4
-20.6 -9.9
∆GN(W f W + E) ) 2νmEgNE + 6νmNmEgNNE +
(1/8)A31m14 + (1/3)A22m13m2 + (3/4)A13m12m22 +
3ν2mE2gNEE + ........ (17)
2A04m1m23 (13) where γ2 is the activity coefficient of an electrolyte in a ternary aqueous system, and γ02 is that in a binary aqueous system. The coefficients Aij in the above equations may be related to an experimental quantity ∆, which is defined as23
(14)
where φ01 and φ02 are the osmotic coefficients of R-CD and the drug, respectively, in binary aqueous systems. By applying the Gibbs-Duhem relation for both binary and ternary systems, it has been shown that24
∆/m1m2 ) A01 + A11m1 + 2A02m2 + 3A03m22 + A21m12 + (3/2)A12m1m2 + A31m1 + (4/3)A22m1 m2 + 2A13m1m2 + 3
2
2
where mN and mE are the molalities of the nonelectrolyte (N) and the electrolyte (E), respectively, defined per kilogram of water, and ν is the number of ions into which the electrolyte dissociates. The terms gNE, gNNE, and so forth are the pair and triplet interaction parameters, which take into account all sources of nonideality in the ternary system. The Gibbs free energy for the transfer of electrolyte from water to aqueous nonelectrolyte solution is given by
∆GE(W f W + N) ) 2νmNgNE + 3νmN2gNNE + 6ν2mEmNgNEE + ........ (18) At low E and N concentrations, all triplet and higher-order terms can be neglected, and the pair interaction parameter gNE can be related to the familiar salting coefficient kS by
4A04m23 (15) The ∆/(m1m2) value is obtained with the help of eq 14, and the coefficients Aij can be found by the method of least-squares. The values of Aij coefficients are included in Table 3. The values of ∆/(m1m2) obtained from the least-square fit method were used for the calculation of water activity, and the reliability of the data is expressed in terms of percentage error given by5
% error in aw )
aw(calcd) - a(obsd) aw(obsd)
× 100
PZ -12.3846 -5.9074 97.3606 -36.2086 6.6816 6.7344 34.8715 5.3268 -5.0233 -1264.8717
3.2.2. Transfer Free Energies and Salting Constant. The salting constant values are determined by applying the method based on application of the McMillan-Mayer theory of solutions.5 According to this theory, the Gibbs free energy of transfer of a nonelectrolyte from water (W) to an aqueous solution containing an electrolyte (E) is given by21,25
m2
∆ ) -55.51 ln aw - m1φ01 - 2m2φ02
CPZ -26.5610 47.8138 308.4434 -1481.3790 -11.9804 -4.1792 3.4329 28.1476 -289.2126 1693.2241
(16)
The observed and calculated values of ∆/(m1m2) along with the percentage error in water activity are given in the Supporting Information. From the measured water activity, the activity coefficient of the solvent (γ0) has been calculated. The molal activity coefficient for R-CD (γ1) and the mean molal activity coefficient for the drug (γ2) in a ternary mixture are calculated using eqs 12 and 13, respectively.
RTkS ) 2νgNE
(19)
Transfer free energies for the transfer of R-CD from water to aqueous drug solutions have been used to calculate pair and triplet interaction parameters using eq 17, and the transfer Gibbs free energy data are given in the Supporting Information, whereas the pair and triplet interaction parameters along with the corresponding salting constants are collected in Table 4. 4. Discussion 4.1. Analysis of Binary Aqueous Data Reveals the Following: The osmotic coefficient (φ02) values when plotted against the concentration of drug show break points at the critical micelle concentration (cmc). The cmc values obtained are 0.022 and 0.030 mol kg-1 for CPZ and PZ, respectively. The plots of φ02 against 1/m (Figure 2) lead to aggregation number (n) and cmc determination by linear extrapolation in the high concentration region and by using the equation26,27
Drug-R-CD Interactions in Water
J. Phys. Chem. B, Vol. 111, No. 48, 2007 13649
1 1 cmc φ02 ) + 1 n n m
(
)
(20)
Thus, from eq 20, values of n obtained are 8.3 and 6.6 for CPZ and PZ, respectively, and the cmc values obtained are 0.023 and 0.031 mol kg-1 for CPZ and PZ, respectively, which agree excellently with the literature data.15 The cmc values obtained from the variation of apparent molal volume with the concentration (Figure 4) are 0.022 and 0.030 mol kg-1for CPZ and PZ, respectively, at 298.15 K. The BV coefficient obtained is negative (-16) and positive (31) for CPZ and PZ, respectively. In general, negative BV values are due to solvent-induced solute-solute interactions or cosphere overlap effects28,29 causing water structure reinforcement. Thus, we can attribute the negative BV in the case of CPZ to hydrophobic hydration in the premicellar region, while the same is not seen distinctly in the case of PZ, where BV is positive. It has been postulated that drug molecules dimerize in the premicellar region, probably because of this positive BV parameter emerging. The φ0V values obtained are 270.0 and 258.8 cm3 mol-1 for CPZ and PZ, respectively, which are in good agreement with the reported literature data.30 The mean molal activity coefficient values indicate negative deviation for CPZ and positive deviation for PZ from the limiting law in the concentration region below the cmc values. The excess Gibbs free energy change (∆GEX) and the Gibbs free energy change due to mixing (∆Gm) are negative and decrease with increase in concentration of the drug (the data are given in the Supporting Information). As expected, the excess free energies for the solutions are less negative than ∆Gm, indicating the importance of enthalpy and entropy effects in these solutions.31 The values of solute-solute and solute-solvent virial coefficients obtained using the McMillan-Mayer theory are given in Table 2. Examining Table 2, it is observed that the values of NB*011 are similar to that of apparent molar volumes of the solute molecules. If we assume comparable hard-sphere molar volumes, it implies similar solute-solvent interaction for both of the compounds. The NB/0 22 values (i.e., a measure of the solute-solute virial interaction), when examined, are seen to be large negatives in the case of CPZ. Such negative values have also been reported for sodium tetraphenylboron.32 Thus it is certain that the incorporation of chlorine in a hydrophobic moiety causes an ion-pairing interaction uniquely special in aqueous solutions. This is reflected also in the variation of the φV - 1.868c1/2 parameter as a function of the concentration of the drugs. It is hard to account for the dimerization or selfaggregation of the molecules in the absence of excess entropy change data for such solutions. However, it is known in medicinal chemistry that CPZ is used as a reference for comparison of the activities of other drugs. It has the highest activity in the series of phenothiazine tranquilizing drugs.33 4.2. Analysis of Ternary Aqueous Data. The variations of the activity coefficients of R-CD and the drug with the square root of the concentration of the drug (expressed in moles per kilogram of water) are shown in Figures 5 and 6. Examinations of these figures reveal that the activity coefficient of R-CD decreases in the presence of drug rapidly and, after stoichiometric concentration, varies little with an increase in drug concentrations. The mean molal activity coefficients of both drug solutes show abnormal low values (CPZ ) 0.298 and PZ ) 0.529) as trace activity coefficients. The trends indicate the formation of complexed species, and, in the region of stoichiometric concentration, the activity coefficient varies rapidly with
Figure 5. Variation of the activity coefficient of 0.1 M R-CD (γ1) in aqueous solutions containing drug as a function of concentration of drug at 298.15 K.
Figure 6. Variation of the activity coefficient of a drug (γ2) in aqueous drug solutions containing 0.1 M R-CD as a function of concentration of drug at 298.15 K.
an increase in drug concentrations. The effect of R-CD is more pronounced in the case of CPZ than in PZ. These results clearly indicate complexation between R-CD and the drug, the extent being more in the case of CPZ than in PZ. The variation of excess Gibbs free energies (∆GEX) of aqueous drug solutions containing R-CD in the studied concentration range (Figure 7) indicates negative deviation from ideal behavior. The addition of low molecular weight organic additives to aqueous solutions of surfactants is known to alter the cmc. The most commonly used additives are urea, alcohols, sugars, and dioxane.34,35 This is being attributed to a change in the dielectric constant of the medium. Ions generally lower the cmc value. When our data of signal values on the osmometer were studied as a function of drug concentration (Figure 8), we found that cmc value of these drug molecules is being increased by the addition of R-CD. Thus, there is a salting-in of the micelle, and we attribute this to complexation or incorporation of drug molecules in the R-CD cavity, preventing them from associating themselves. The cmc values are altered to 0.068 and 0.073 mol kg-1 for CPZ and PZ, respectively, as a result of the presence
13650 J. Phys. Chem. B, Vol. 111, No. 48, 2007
Figure 7. Variation of the excess free energy change (∆GEX) of aqueous drug solutions containing 0.1 M R-CD as a function of concentration of drug at 298.15 K.
Terdale et al.
Figure 9. Variation of Gibbs free energies of transfer of 0.1 M R-CD (∆GNtr ) from water to drug solutions as a function of concentration of drug at 298.15 K.
Figure 10. Variation of Gibbs free energies of transfer of drug (∆GEtr) from water to 0.1 M R-CD solutions as a function of concentration of drug at 298.15 K.
Figure 8. Observed signal values against concentration of drug, at 298.15 K: (a) comparison in binary aqueous and ternary aqueous CPZ solutions; (b) comparison in binary aqueous and ternary aqueous PZ solutions.
of R-CD, from 0.022 and 0.030 mol kg-1 in the absence of R-CD, respectively.
The transfer Gibbs free energies are plotted as a function of drug concentrations in Figures 9 and 10. Examination of Figures 5 and 9 as well as Table 4 reveals that CPZ forms 2:1 and 1:1 complexed species, while PZ forms only 1:1 complexed species with R-CD. The stoichiometries of the complexes formed between R-CD and the drug have also been estimated from the data of signal value as shown in Figure 11, where the relative signal values are plotted as a function of mole ratio (i.e., drug/ R-CD). The abrupt change at a mole ratio of ∼1 indicates that the stoichiometry of the complexes is 1:1 for both of the studied systems. The small inflection at a mole ratio of ∼0.4 indicates the possibility of the existence of a weak 2:1 (two molecules of R-CD per molecule of CPZ) complex in the case of CPZ. The data of transfer free energies for the transfer of R-CD (∆GNtr ) from water to aqueous drug solutions are subjected to scrutiny by application of eq 17, while the same can also be accomplished for the transfer of drug molecules (∆GEtr) from water to aqueous R-CD solutions using eq 18. The calculated pair and triplet interaction parameters (gNE, gNNE, and gNEE) are collected in Table 4. Using the pair interaction parameter, the salting constant values are estimated for these systems using
Drug-R-CD Interactions in Water
J. Phys. Chem. B, Vol. 111, No. 48, 2007 13651 thermodynamic equilibrium constant value for 1:1 complex formation) values. These calculations yield log K values of 1.62 and 1.3 for CPZ and PZ, respectively, at 298.15 K. The value for the R-CD-CPZ complex is in fair agreement with those reported in the literature.37 Thus, our results indicate that CPZ forms a stronger complex (1:1) with R-CD molecules than does PZ. It is suggested that the phenothiazine ring participates in the association with R-CD in the interior cavity, and Nsubstituents in the drugs participate with the exterior hydrophilic surface of R-CD molecules by H-bonding. Therefore, our calculations of log K values and the negative gNNE parameters signify that, in the complexation process, the weak dispersive forces inside the R-CD cavity, H-bonding interactions, as well as hydrophobic interactions between complexed moieties play important role in governing the stability of the species formed in aqueous solutions. 5. Conclusions
Figure 11. Plot of relative signal values for aqueous ternary mixtures containing R-CD and drug as a function of mole ratio (moles of drug/ moles of R-CD) for determining the stoichiometry of complexes.
eq 19 and are also collected in Table 4. The pair interaction parameters (gNE) are negative for CPZ and PZ, which means that the pair-wise interactions are favorable for R-CD-drug complexation. The comparatively strong complex is revealed from the Gibbs free energy changes for the R-CD-CPZ system. The negative values of gNNE (Table 4) indicate the further stabilization of the complex by CD-CD hydrophobic interactions or possibly the existence of a weak 2:1 complex in the case of the R-CD-CPZ system. The positive gNEE values reveal that the ion-ion interactions are weak in both of the systems. Also, the negative salting constant indicates a salting-in effect exhibited by the R-CD for the encapsulation of the drug molecules. We have assumed in all these discussions that the anion does not play any role in the interactions; however, the incorporation of a Cl atom in the main skeleton of the drug molecule renders them more favorable for the encapsulation process. Otagiri et al.36 reported the stability constants of CPZ with R-CD in buffer solutions on a molarity scale, which are comparatively low (log K ) 2.3010). It is to be noted that these authors showed that the drug-CD complex stability is highly pH dependent, and R-CD, in the buffered solutions, does not form a complex with the phenothiazine drugs. Using ionselective electrodes, Wyn-Jones et. al36 have reported complexation constant values for 1:1 CPZ-R-CD complexed species of 120 (log K ) 2.1). In general, association constant values for R-CD complexes are reported at constant ionic strength or at different pH values. Attempts made to obtain the values of log K for CPZ and PZ using the kS values given in Table 4 yield high values as compared to the values obtained by the techniques of partition coefficient, UV spectroscopy, or other techniques. We stress here that the association constant calculations require the specialization of aggregation number. If we use the aggregation number values from our osmotic coefficient data (i.e., 8.3 and 6.6 for CPZ and PZ, respectively) and the concept that, as the cmc values of these drug molecules are very low, we are in fact transferring the micellar aggregates from binary aqueous solutions to aqueous R-CD solutions, and hence the transfer free energies have to be divided by the aggregation number (i.e., 4.15 and 3.3) of molecules to calculate the transfer free energies values per monomer of drug molecules; hence, seen in this light, the kS values are -4.19 and -3.0 for CPZ and PZ, respectively. Now following Dagade and Patil’s method,5 these kS values can be equated to -log K (where K is
Studies of the apparent molal volume of drug molecules (CPZ and PZ) in aqueous solutions as a function of concentration were performed. The two drug moieties exhibit important differences in the mode of interaction. In the premicellar region, because of solute-solute association as well as solvent-induced cosphere overlapping effects, the positive and negative values for the deviation constant in the case of PZ and CPZ, respectively, were noted. The structural details of the solute molecule thus play an important role in exhibiting micellar-type equilibria in solution phase. The aggregation number of these surfactant molecules as well as the cmc values obtained using osmotic coefficient data agree very well with those obtained from volume data as well as with data reported in the literature from conductance measurements (for CPZ). Osmotic coefficient data for the aqueous solutions of these drug molecules were used to study the concentration variation of activity coefficients of these drug molecules. The activity coefficients are higher (for PZ) and lower (for CPZ) than that predicted on the basis of the Debye-Hu¨ckel limiting law for a 1:1 electrolyte. The results about the activity coefficient of drug solutes in aqueous solutions containing R-CD indicate the formation of a complex species between R-CD and the drug. The effect of complexation is found to be more in the case of CPZ than in PZ. The cmc values get elevated as a result of the addition of R-CD. The pair (gNE) and triplet (gNNE) interaction parameters are obtained by subjecting transfer free energy data to scrutiny by applying the McMillan-Mayer solution theory of virial coefficients. We observe from the salting-in coefficient values that the formation of host-guest species-type equilibria is highly favorable and is attenuated by hydrophobic interactions in aqueous solutions. The presence of a covalent chlorine atom in the CPZ molecule imparts a high salting-in coefficient value for CPZ-R-CD complexed species in the solution phase. Glossary R-CD PZ CPZ m m1 m2 aw φ01
R-cyclodextrin promazine hydrochloride chlorpromazine hydrochloride molality of drug in binary aqueous solutions molality of R-cyclodextrin in ternary aqueous solutions containing electrolyte (drug salts) molality of electrolyte (drug salts) in ternary aqueous solutions containing R-cyclodextrin water activity osmotic coefficient of R-cyclodextrin in binary aqueous solution
13652 J. Phys. Chem. B, Vol. 111, No. 48, 2007 φ02 γ00 γ01 γ02 ∆GEX γ0 γ1 γ2 ∆GNtr ∆GEtr
osmotic coefficient of an electrolyte in binary aqueous solution activity coefficient of water in a binary mixture molal activity coefficient of R-cyclodextrin in a binary aqueous solution mean molal activity coefficient of an electrolyte in a binary aqueous solution excess Gibbs free energy change for a binary/ternary mixture activity coefficient of water in a ternary mixture molal activity coefficient of R-cyclodextrin in a ternary mixture containing an electrolyte mean molal activity coefficient of an electrolyte in a ternary mixture containing a nonelectrolyte, R-cyclodextrin Gibbs free energy change on the transfer of a nonelectrolyte (R-cyclodextrin) from water to an aqueous electrolyte (drug) solution Gibbs free energy change on the transfer of an electrolyte (drug) from water to an aqueous nonelectrolyte solution (R-cyclodextrin)
Acknowledgment. The authors are grateful to the authorities at Wissenschaftliche Gera¨tebau, Dr. Ing. Herbert Knauer GmbH, D-14163 Berlin-Zehlendorf, Germany, for the gift of the universal thermistor and related accessories required for the operation of the osmometer. The authors also wish to thank Dr. S. P. Gowindwar, Professor and Head, Department of Biochemistry, Shivaji University, Kolhapur, Maharashtra (India) for providing the drug samples. Supporting Information Available: The water activity, osmotic coefficient, and activity coefficient data obtained for binary aqueous drug solutions at 298.15 K are given in Table A1. Water activity and % error in water activity along with other parameters such as activity coefficient, free energies of transfer, excess free energy for ternary systems (H2O + 0.1 M R-CD + drug) at 298.15 K are given in Tables A2 and A3, respectively. This material is available free of charge via the Internet at http://pubs.acs.org. References and Notes (1) Franks, F. Biochemical Thermodynamics; Jones, M. N., Ed.; Elsevier: New York, 1979; Chapter 2. (2) Franks, F.; Eagland, D. Crit. ReV. Biochem. 1975, 3, 165. (3) Jolicouer, C. Methods Biochem. Anal. 1981, 27, 171. (4) Rekharsky, M. V.; Inoue, Y. Chem. ReV. 1998, 98, 1875. (5) Patil, K.; Dagade, D. J. Solution Chem. 2003, 32, 951.
Terdale et al. (6) Tomar, P. A.; Kolhapurkar, R. R.; Dagade, D. H.; Patil, K. J. J. Solution Chem. 2007, 36, 193. (7) Kumbhar, R. R.; Dagade, D. H.; Terdale, S. S.; Patil, K. J. J. Solution Chem. 2007, 36, 259. (8) Attwood, D.; Dickinson, N. A.; Mosquera, V.; Villar, V. P. J. Phys. Chem. 1987, 91, 4203. (9) Attwood, D.; Mosquera, V.; Villar, V. P. J. Chem. Soc., Faraday Trans. 1 1989, 85, 3011. (10) Attwood, D.; Fletcher, P.; Boitard, E.; Dube`s, J.-P.; Tachoire, H. J. Phys. Chem. 1990, 94, 6034. (11) Attwood, D.; Mosquera, V.; Rey, C.; Vasquez, E. J. Chem. Soc., Faraday Trans. 1991, 87, 2971. (12) Attwood, D.; Boitard, E.; Dube`s, J.-P.; Tachoire, H. J. Phys. Chem. 1992, 96, 11018. (13) Ruso, J. M.; Attwood, D.; Taboada, P.; Suarez, M. J.; Sarmiento, F.; Mosquera, V. J. Chem. Eng. Data 1999, 44, 941. (14) Peˆrcz-Rodriguez, M.; Varela, L. M.; Taboada, P.; Attwood, D.; Mosquera, V. Colloid Polym. Sci. 2000, 278, 706. (15) Burchfield, T. E.; Woolley, E. M. J. Phys. Chem. 1984, 88, 2149. (16) Terdale, S. S.; Dagade, D. H.; Patil, K. J. J. Phys. Chem. B 2006, 110, 18583. (17) Patil, K.; Pawar, R.; Dagade, D. J. Phys. Chem. A 2002, 106, 9606. (18) Harned, H. S.; Owen, B. B. The Physical Chemistry of Electrolytic Solutions, 3rd ed.; Reinhold Publishing Corporation: New York, 1958. (19) Harrington, T. M.; Mole, E. L. J. Chem. Soc., Faraday Trans. 1 1982, 78, 213. (20) Harrington, T. M.; Mole, E. L. J. Chem. Soc., Faraday Trans. 1 1982, 78, 2095. (21) McMillan, W.; Mayer, J. J. Chem. Phys. 1945, 13, 276. (22) Hill, T. L. J. Am. Chem. Soc. 1957, 79, 4885. (23) Bower, V. E.; Robinson, R. A. J. Phys. Chem. 1963, 67, 1524. (24) Wen, W.-Y.; Chen, C. L. J. Phys. Chem. 1969, 73, 2895. (25) Desnoyers, J. E.; Billon, M.; Leyer, S.; Perron, G.; Morel, J. P. J. Solution Chem. 1976, 5, 681. (26) Desnoyers, J. E.; Caron, G.; DeLisi, R.; Roberts, D.; Roux, A.; Perron, G. J. Phys. Chem. 1983, 87, 1397. (27) Coudert, R.; Hajji, A. M. J. Colloid Interface Sci. 1993, 155, 173. (28) Okamoto, B. Y.; Wood, R. H.; Desnoyers, J. E.; Perron, G.; Delorme, L. J. Solution Chem. 1981, 10, 139. (29) Desnoyers, J. E.; Perron, G.; Ave´dikian, L.; Morel, J. P. J. Solution Chem. 1976, 5, 631. (30) Attwood, D.; Blundell, R.; Mosquera, V.; Garcia, M.; Rodriguez, J. Colloid Polym. Sci. 1994, 272, 108. (31) Dagade, D. H.; Kolhapurkar, R. R.; Terdale, S. S.; Patil, K. J. J. Solution Chem. 2005, 34, 415. (32) Harrington, T. M.; Taylor, C. M. J. Chem. Soc., Faraday Trans. 1 1982, 78, 3409. (33) Wilson; Gisvold Textbook of Organic Medicinal and Pharmaceutical Chemistry, 10th ed.; Delgado, J. N., Remers, W. A., Eds.; LippincottRaven Publishers: Philadelphia, PA ,1998; Chapter 14, p 450. (34) Mukerjee, P.; Mysels, K. J. Critical Micelle Concentrations of Aqueous Surfactant Sustems NSRDS-NBS 36; Superintendent of Documents: Washington, DC, 1971. (35) Kresheck, G. C. Water: A ComprehensiVe Treatise; Franks, F., Ed.; Plenum Press: New York, 1974; Vol. IV, Chapter 2. (36) Otagiri, M.; Ukema, K.; Ikeda, K. Chem. Pharm. Bull. 1975, 23, 188. (37) Takisawa, N.; Hall, D. G.; Wyn-Jones, E.; Brown, P. J. Chem. Soc., Faraday Trans. 1 1988, 84, 3059.
