16546
J. Phys. Chem. B 2008, 112, 16546–16551
Aggregation Studies of n-Alkanoyl-N-methyllactitolamine Surfactants Boz˙enna Ro´z˙ycka-Roszak, Paweł Misiak,* and Beata Jurczak Department of Physics and Biophysics, Wrocław UniVersity of EnVironmental and Life Sciences, ul. Norwida 25, 50-375 Wrocław, Poland
Kazimiera A. Wilk Faculty of Chemistry, Wrocław UniVersity of Technology Wybrzez˙e Wyspian´skiego 27, 50-370 Wrocław, Poland ReceiVed: June 17, 2008; ReVised Manuscript ReceiVed: August 29, 2008
The micellization processes of n-alkanoyl-N-methyllactitolamines (CnMeLA) were studied by means of isothermal titration calorimetry (ITC) and compared with those of decanoyl-N-methyl glucamide (MEGA10). The critical micelle concentrations (cmc), the enthalpies (∆Hm) and the entropies (∆S0m) of micellization as well as the contributions of the head groups to the Gibbs free energies (∆G0m (hy)) were calculated. The molecular modeling studies including the conformational analysis and calculations of some molecular descriptors related to hydrophilic and hydrophobic interactions were done to elucidate the experimental results at the atomic level. Despite the significantly bigger size of the headgroup of CnMeLA, the value of ∆G0m (hy) for this compound is close to those for the MEGA-n and CnGluc. The introduction of gluco ring to polyoxyethylene chain of MEGA-10 decreases its hydrophobicity to a similar extent as shortening of the alkyl chain by one methylene group. Molecular modeling studies suggest that these effects came from conformational properties of the gluco ring and polyoxyethylene chain, which are different when they constitute CnMeLA headgroup and when they constitute head groups of CnGluc and MEGA-10, respectively. I. Introduction The n-alkanoyl-N-methyllactitolamines (CnMeLA) are examples of lactose-derived surfactants (Figure 1). Recalling previous findings by one of us,1 they share some performance properties (i.e., the foaming stabilizing effect and wetting behavior) with commercially available alkyl polyglucosides (Glucopon 600 EC(HH), a Henkel product). Additionally, these compounds are practically nontoxic to bacteria and yeasts, and they can be considered as readily biodegradable in the Closed Bottle Test inoculated with activated sludge and fulfill all the requirements needed for environmental acceptance.1,2 In the previous paper3 the influence of CnMeLA compounds on thermotropic phase behavior of phosphatidylcholine bilayers was compared with the influence of common sugar-based surfactants n-dodecyl-β-D-glucopyranoside (C12Gluc) and decanoyl-N-methyl glucamide (MEGA-10). The obtained results indicate that CnMeLA are very active at the level of membrane surface and less disturb the phospholipid bilayers structure than commercially used MEGA-10 and C12Gluc. CnMeLA differ from MEGA-n surfactants by substitution of one hydroxyl group of polyoxyethylene chain in headgroup by gluco ring. Thus comparing the chemical structures of CnMeLA and MEGA-10 (see Figure 1), one may think that the differences in the properties of those compounds are due to the gluco ring. However, the C12Gluc compound also has a gluco ring and its surface activity is lower and the ability to disturb the phospholipid bilayers even greater than those of CnMeLA compounds. Thus gluco ring affects the properties of CnMeLA and C12Gluc in a different way. This may be due to different hydration cospheres of the gluco ring in those compounds. * To whom correspondence should be addressed. E-mail: pawel.misiak@ up.wroc.pl. Phone: +48 71 3205293. Fax: +48 71 3205167.
The present paper is aimed to elucidate the role of gluco ring in the micellization process of CnMeLA surfactants and to shed light on the relation between biological activity and structure. Therefore the critical micelle concentrations (cmc) and enthalpy of micellization (∆Hm) for the surfactants are determined by means of isothermal titration calorimetry. Obtained values were compared with those for some n-alkyl-β-D-glucopyranoside (CnGluc) and MEGA-n compounds. As conformational properties of molecules in water solution seem to play an important role in aggregation processes, we performed molecular modeling computations aiming to give some deeper insight into molecular behavior on the atomic level. The modeling methods included sampling of the conformational space and calculating of some molecular descriptors characterizing molecular properties with respect to interactions with surrounding molecules. Obtained results were related to the experimental data. II. Materials and Methods Experimental. The n-alkanoyl-N-methyllactitolamine (CnMeLA) compounds were prepared as described by Wilk et al..1 MEGA-10 was purchased from Sigma-Aldrich Chemie company (Germany). The calorimetric measurements were performed on a TAM device (Thermal Activity Monitor, Thermometric AB, Ja¨rfa¨lla, Sweden). All experiments were performed at 25 ( 0.01 °C. The sample cell was filled with 0.9 mL of bidistilled water prior to each experiment. Concentrated surfactant solution was injected in small aliquots (5-15 µL) to the stirred sample cell using 250 µL Hamilton syringe controlled by a Lund Syringe pump 2. Enthalpy calculations were performed using ITC software.
10.1021/jp805334t CCC: $40.75 2008 American Chemical Society Published on Web 12/04/2008
n-Alkanoyl-N-methyllactitolamine Surfactants
Figure 1. Chemical structures of surfactants MEGA-n, n-alkyl-β-Dglucopyranoside (CnGluc) and n-alkanoyl-N-methyllactitolamine (CnMeLA).
Molecular Modeling. The computational modeling approach included sampling of the conformational space for single molecules of studied surfactants by means of simulated annealing method followed by final energy optimization. Subsequently for the optimal (lowest energy) conformations calculations of quantities characterizing the molecular interactions, in particular some VolSurf descriptors,4-6 were performed within the approach based on Molecular Interaction Field maps. In the molecular dynamics simulations the electrostatic interactions with water solvent was taken into account within implicit solvent model by numerical solving of Poisson-Boltzmann equation for electrostatic potential. The effective atomic charges for each compound, which are essential for subsequent modeling, were determined by means of the NDDO-type semiempirical molecular orbital calculations, performed using the MOPAC2007 package7 with the new parametrization method PM6.8 The atomic charges were fitted to the molecular surface electrostatic potential, obtained from molecular orbitals by calculation of the expected values of molecular electrostatic potential on a uniform distribution of points, i.e. within the ESP method.9 The effect of water solvent on the electronic structure of molecule was taken into account using continuum approach within the conductor-like screening model COSMO10 with dielectric constant equal to 78.4, which is that of water at 25 °C. Conformational Analysis. The conformational space of each compound was sampled in the framework of molecular dynamics (MD) method using the Sybyl 8.0 software package for molecular modeling.11 Two runs, the first of five hundred cycles and the second of thousand cycles of the Simulated Annealing (SA) dynamics, were performed for a single molecule of each compound studied. The atomic interactions were described by all-atom universal Tripos force field parameters12,13 with atomic charges obtained from semiempirical molecular orbital calculations mentioned above. The nonbonded interactions cutoff was set to 1.6 nm and implicit water solvent was taken into account by applying the distance-dependent dielectric function with dielectric constant equal to 80, as is usually used in MD simulations. Because we were mainly interested in the conformations of the head part of the molecules and the position of the head part with respect to the alkyl chain, in the first stage (the 500 cycles SA) the methylene groups of the middle part of
J. Phys. Chem. B, Vol. 112, No. 51, 2008 16547 the chain were treated as the stiff entity during dynamics simulations. The terminal ethyl group and the two methylene groups from the head part of the molecules were allowed to move freely in the simulations. The second stage (1000 cycles SA) was performed using the optimal (i.e., of lowest energy) conformation of the molecule obtained from the previous stage as a starting conformation, and there were no constraints to the intermolecular atomic movements. Each SA cycle consisted of the 2 ps of molecular dynamics simulation in the temperature of 1000 K followed by 1 ps cooling with exponential lowering of temperature to 100 K. For each surfactant molecule the final conformation after each simulated annealing cycle was next optimized with respect to the total energy with the same force field and interactions parameters as used for simulations and with no constraints to the molecular flexibility. Finally, for each molecule the minimized conformations as well as the starting one were sorted according to increasing energy. Further calculations and analysis were performed for conformations with lowest energies within the range of 3kBT for T ) 300 K, that is, about 7.5 kJ/mol. The sampling of the conformational space was intended to find the molecular conformations resembling the ones appearing in premicellar solution and entering the micellar aggregate. The method described in the previous paragraph was initially tested on the n-dodecyl-N,N-dimethyl-N-benzylammonium chloride (DBeAC) surfactant for which the molecular conformations in micelle were determined from NMR studies by Ro´z˙yckaRoszak and Cierpicki.14 For the DBeAC molecule, the final conformations of lowest energy obtained from simulated annealing and subsequent minimization, especially conformation of the head part of the molecule relative to alkyl chain and the number of gauche bonds in the alkyl chain, were close to the ones described in the cited paper. Molecular Interaction Field 3D Molecular Descriptors. Some properties of the studied surfactant molecules were also characterized by means of three-dimensional (3D) molecular interaction field (MIF) descriptors calculated on the spatial grid using the GRID15 method within the VolSurf program implemented in Sybyl software. The descriptors were calculated for the OH2 and DRY probes for each molecule in each of the chosen conformations of low energy obtained as described above. Most interesting, we found the DnDRY descriptors characterizing the hydrophobic regions of the molecules and describing interaction of the molecule with the DRY probe in the energy range from -0.2 (-0.84) to -1.6 kcal/mol (-6.7kJ/ mol). III. Results and Discussion The typical calorimetric curves of the observed enthalpies of dilution ∆Hobs versus concentration c for C10MeLA, C12MeLA and also for MEGA-10, are presented in Figure 2. The data for C14MeLA are not presented, because the obtained enthalpograms were strongly nonideal, that is, the points were scattered so that interpolation by means of sigmoidal-type curve (as expected for titration data) was virtually impossible. At last, it was possible to determine cmc value reasonably well, whereas the ∆Hm value could be only roughly estimated. The pseudophase transition model of micellization was applied for interpretation of experimental data. The critical micelle concentrations cmc and the enthalpies of micellization ∆Hm were determined from the same calorimetric titration curves. The cmc and ∆Hm were determined using the “classical” method, that is, applying cubic splines to connect experimental points with the smooth line, and putting cmc equal to the
16548 J. Phys. Chem. B, Vol. 112, No. 51, 2008
Ro´z˙ycka-Roszak et al.
∆Hobs(c) ) A +
∆Hm
{1 + (2
1/θ
- 1)exp[(cmc - c)/δ]}θ
(2)
Figure 2. Typical plots of the observed enthalpies of dilution ∆Hobs vs concentration c for C10MeLA, C12MeLA, and MEGA-10. The concentration of the surfactant solution injected to the sample cell was equal to 100 mM for C10MeLA, 9 mM for C12 MeLA, and 45 mM for MEGA-10.
concentration where the numerical derivative of the line reaches maximum. The enthalpy of micellization was taken as equal to the difference of enthalpy values at cmc between the straight lines fitted to the terminal and initial points of the enthalpograms.16,17 We also applied the sigmoidal Boltzmann equation (SBE) model,18,19 which can be written in the form
∆Hobs(c) ) A +
∆Hm 1 + exp[(cmc - c) /δ]
(1)
where c is the concentration of the surfactant in the solution, ∆Hm is the enthalpy of micellization, A is the asymptote for c f -∞, and δ is scaling parameter for concentration. The above function was fitted by nonlinear least-squares method to the experimental data to obtain values of the four parameters: cmc, ∆Hm, A, and δ. The last two are of little physical relevance. In this approach, the cmc is the concentration at the inflection point of the SBE function, which has a symmetry center in it. Thus results are best if the experimental points are distributed symmetrically. However, the symmetrical sigmoidal function does not fit well to the points of our enthalpograms, because of different curvature in the lower and upper bend of the enthalpograms. Therefore, we also used the asymmetric sigmoidal function (ASF) of the form
where θ is an additional free parameter to fit. This function allows one to take into account the different curvatures of the two bends in the smooth curve connecting the points on the enthalpograms. We observed that the cmc values obtained within the classical method are very close to those obtained from SBE and ASF models. On the other hand, we found the ∆Hm values from the sigmoidal function fitting approach overestimated. This effect can be related to the unphysical character of the sigmoidal functions used to fit to the demicellization calorimetric data. In fact, both functions are defined for any value of c, positive or negative, whereas in the physical system the concentration values are between zero and some cmax, the latter denoting the concentration of the injected micellar solution. The fitted ∆Hm being the difference between horizontal asymptotes for c f ∞ and c f -∞ (equal to A), is of restricted physical relevance. As we are interested in the transition region, that is, the one where the enthalpy changes with concentration most quickly, we can assume that critical micelle concentration is the concentration value, where the curve interpolated or fitted to the data points has the inflection point, that is, is equal to the cmc parameter of the fitted sigmoidal function. Eventually, we took the ∆Hm values obtained from the classical method. In Table 1, final values of cmc and enthalpy of micellization ∆Hm (averaged over at least three repeated measurements) for MEGA-10, C10MeLA, C12MeLA, and C14MeLA are shown in comparison with those for MEGA-10, MEGA-9, and CnGluc surfactants taken from references. The critical micelle concentrations for CnMeLA are higher than those obtained by Wilk et al.1 from surface tension measurements (shown in parentheses in Table 1). This is due to different method used. The cmc is often a method-dependent quantity, and the values obtained from tensiometry are lower than those from ITC for many compounds.19 However, for MEGA-10 both cmc and enthalpy of micellization obtained in the present work is almost the same as those obtained by means of tensiometry by Oda et al.21 and by means of ITC by Prasad et al.19 (shown in parentheses in bottom row for MEGA-10 in Table 1). The standard Gibbs free energy of micelle formation of a nonionic surfactant is related to critical micelle concentration by the formula ∆G0m ) RT ln xm, where xm is the cmc expressed in mole fraction units. The standard entropy of micelle formation was calculated from the equation ∆S0m )(∆H0m - ∆G0m)/T, where T is the temperature. Obtained values are listed in Table 1 in comparison with those available for MEGA-10 and CnGluc. On the basis of the values listed in Table 1, the contribution of methylene group of the alkyl chain to the Gibbs free energy 0 0 0 (CH2), to enthalpy ∆Hm (CH2), and entropy ∆Sm (CH2) ∆Gm of micellization for CnMeLA were estimated as equal to about -2.9 kJ/mol, -1.4 kJ/mol and 5.5 J/(mol · K), respectively. These values compare fairly well with those for MEGA-10, which are -2.88 kJ/mol, -1.40 kJ/mol, and 4.96 J/(mol · K), respectively.20 The contribution of methylene group of the alkyl chain to free energy for CnGluc estimated from literature data listed in Table 1 as -2.9 kJ/mol is also very similar to that for MEGA-n and CnMeLA surfactants.
n-Alkanoyl-N-methyllactitolamine Surfactants
J. Phys. Chem. B, Vol. 112, No. 51, 2008 16549
0 0 TABLE 1: Critical Micelle Concentration (cmc), Enthalpy of Micellization ∆Hm, Gibbs Free Energy ∆Gm , and Entropy ∆Sm Changes for Micelle Formation Process of CnMeLA, n ) 10, 12, 14, MEGA-9, MEGA-10, and CnGluc, n ) 8, 10, 12
compound
cmc [mM]
C10MeLA C12MeLA C14MeLA MEGA-9
10.4 (3.4a) 0.88 (0.45a) 0.09 (0.067a) 21.4c 6.14 (6.1d, 6.45e) 23.0f 2.2f 0.19g
MEGA-10 C8Gluc C10Gluc C12Gluc
∆Hm [kJ/mol] 6.4 3.66 ≈1.0b 5.0 (4.8d, 5.07e) 8.24f
0 ∆Gm [kJ/mol]
- 21.26 - 27.38 - 33.04 - 19.48 - 22.57 (-22.6d, -22.4e) - 19.30 - 25.11 - 31.18
0 ∆Sm [J/(mol · K)]
92.84 104.16 ≈ 114.3 93.05 (92.0d,e) 92.4
a From Wilk et al.1 b The value is only roughly estimated due to strongly nonideal enthalpogram. c From Okawauchi et al.20 d From linear interpolation of the tensiometric measurements by Oda et al.21 for 20 and 30 °C. e From Prasad et al.19 f From Heerklotz and Seelig.22 g From Shinoda et al.23
Assuming additivity of the free energy change ∆G0m, one can divide it into contributions of different fragments of a molecule as follows 0 0 0 0 ∆Gm ) ∆Gm (hy) + (n - 1)∆Gm (CH2) + ∆Gm (CH3) (3)
where ∆G0m (hy), ∆G0m (CH2), and ∆G0m (CH3) are contributions of the hydrophilic headgroup, methylene group, and terminal methyl group of alkyl chain, respectively. 0 (CH3) ) -8.78 kJ/mol, the Assuming from Tanford24 ∆Gm 0 (hy) of MEGA-10 and C10 contributions of head groups ∆Gm Gluc were calculated as 9.25 and 9.77 kJ/mol, respectively. For MEGA-10, a very similar value (9.48 kJ/mol) is reported by Okawauchi et al.20 Prasad et al.19 obtained noticeably higher 0 (hy) ) 12.43 kJ/mol, which resulted due to using value ∆Gm the incorrect number of methylene groups in the molecule (10 instead of 8). The contributions to free energy of the head groups of MEGA-10 and CnGluc are very close, which suggests that the contribution of gluco ring and polyoxyethylene chain to Gibbs free energy of micellization are very similar. The same follows from the similarity of cmc values for MEGA-9 and C8Gluc surfactants (Table 1); the surfactants have the same number of methylene groups in their alkyl chains. The headgroup of CnMeLA is significantly bigger and has more OH groups than the headgroups of MEGA-n and CnGluc. 0 (hy) of CnMeLA equals to 10.33 kJ/ Nevertheless, the ∆Gm mol and is only a little higher than each of the values obtained for MEGA-n and CnGluc headgroups. This may result from different conformational properties of the gluco ring in CnGluc and CnMeLA molecules. For comparison the calorimetric curves of the observed enthalpies of dilution (∆Hobs) were normalized to infinite dilutions (∆Hn). Kresheck25 considered the normalized curves as approximate partial molar enthalpies. The trends of normalized calorimetric curves as a function of concentration c, drawn in Figure 3, show that MEGA-10 behaves like C11MeLA. This suggests that substitution of one hydroxyl group in the polyoxyethylene chain of MEGA-10 by gluco ring decreases its hydrophobicity to the same extent as the shortening of its alkyl chain by one methylene group. Molecular Modeling. The results of molecular modeling simulations and computations were used to elucidate some details of the observed molecular behavior during micellization. In particular, the obtained optimal conformational states of the molecules can explain dependence of the micellization ability (cmc) on the length of alkyl chain of the surfactant molecule,
Figure 3. Normalized curves of enthalpies of dilution ∆Hn vs concentration c for MEGA-10, C10MeLA and C12MeLA.
calculated in the framework of linear (additive) thermodynamical approach based on measured parameters of micellization. Typical conformations of the lowest total energy of molecules MEGA-10, C8Gluc, and C10MeLA obtained from scanning of conformational space are shown in Figure 4. It can be seen from the pictures that for all three molecules the hydrophilic head is bent toward the hydrophobic chain, so that there is no room between these two parts of the molecule for any atoms to intercalate. The conformations are globular rather than elongated, resulting in mutual partial screening of the alkyl chain and hydrophilic headgroup from the interactions with surrounding molecules. In such conformations the hydrophobic interactions between alkyl fragments of surfactant molecules are weaker than they would be if the alkyl chains were more vulnerable. The conformations also suggest that the bent molecules can show the tendency to create premicellar aggregates, even more so because the compounds, being amides, can be partially ionized in water. The premicellar aggregation can result in lowering surface activity and hence lower cmc values obtained by means of tensiometric measurements than by means of ITC, especially for CnMeLA compounds. Moreover, the conformation of the CnMeLA headgroup enables creation of few intrahead group hydrogen bonds, decreasing the expected total hydrophilicity of the molecule. For MEGA-10 and CnGluc compounds, the intramolecular hydrogen bonding is much less probable. The differences in the ability of the molecules to participate in hydrophobic interactions can also be observed looking at the values of some MIF descriptors, especially the DnDRY parameters, describing interactions of molecules with the hydrophobic (DRY) probe. Table 2 shows the mean values of MIF descriptors D1DRY to D4DRY calculated by VolSurf program as described in the previous section, and polar surface area (PSA) and polar volume (PV).26 The last two are MOL-
16550 J. Phys. Chem. B, Vol. 112, No. 51, 2008
Ro´z˙ycka-Roszak et al.
Figure 4. Typical examples of lowest-energy conformations of three sugar surfactants obtained from simulated annealing molecular dynamics. Left to right: MEGA-10, C8Gluc, C10MeLA. Top row: alkyl chains on the right. Bottom row: alkyl chains on front. Nitrogen atoms are shown in blue, oxygen in red, carbon in cyan, hydrogen in light gray.
TABLE 2: Mean Values of the MIF Descriptors D1DRY to D4DRY and Total Surface Area (S), Polar Surface Area (PSA), and Polar Volume (PV) for C10MeLA, C12MeLA, MEGA-10, C8Gluc, and C10Gluc compound
D1DRY
D2DRY
D3DRY
D4DRY
S/AA2
PSA/Å2
PV/Å3
C10MeLA C12MeLA MEGA-10 C8Gluc C10Gluc
171.2 215.9 182.8 146.2 189.1
66.05 85.59 71.91 54.41 73.84
33.23 41.41 34.91 21.29 31.71
16.55 20.15 17.66 8.99 13.87
703.5 770.4 597.0 553.5 619.3
215.9 217.9 143.2 155.8 153.3
278.9 291.8 161.1 257.4 264.6
PROP PSA and PV values calculated in Sybyl program. All values are means over the conformations of energies within the 3kBT range from the lowest energy one of each molecule, obtained from molecular dynamics simulations. The D1DRY to D4DRY descriptors characterize hydrophobic regions, which indicate interactions of the molecule with the hydrophobic (DRY) probe at energy levels -0.2, -0.4, -0.6, and -0.8 kcal/ mol, respectively. The polar surface area and polar volume can be related to ability of the headgroup of molecule to interact with water or polar parts of other molecules. The values of the MIF descriptors for MEGA-10 are between the respective values for C10MeLA and C12MeLA, that is, similar 0 0 , and ∆Sm shown in Table 1. to the values of cmc, ∆Hm, ∆Gm On the other hand, the DnDRY descriptors’ values are all ordered as follows: C8Gluc < C10MeLA < MEGA-10. The sequence corresponds to rising hydrophobicity of the molecules. This is in agreement with the experimental results for enthalpy changes during micellization. The DnDRY descriptors for MEGA-10 are all higher than these of C10MeLA. Because both the compounds have alkyl chain of the same length, one can conclude that properties of C10MeLA are such as its alkyl chain were shorter than it is formally, that is, as follows from the number of methylene groups. The possible reason can be the bigger headgroup, which hinders the hydrophobic interactions much more than the headgroup of MEGA-10. The PSA and PV of CnMeLA are higher than the respective values of MEGA-10 or CnGluc. On the other hand, they are significantly smaller than the sum of the PSA and of PV values of the last two compounds (even after taking into account that
it has one OH group less than the sum of head groups of MEGA10 and CnGluc molecules). The ratio of the polar to apolar surface area, PSA/ASA, where the apolar surface area ASA equals to the difference between total surface area S and polar surface area PSA (ASA ) S - PSA), can be regarded as a simple measure of the hydrophilicity/hydrophobicity ratio of the molecule. From data shown in the Table 2, one obtains PSA/ ASA ratio equal to 0.32 for MEGA-10, 0.39 for C8Gluc and 0.44 for C10MeLA. Thus, the values are ordered in the same way as the contributions of head groups to Gibbs free energy 0 (hy). ∆Gm IV. Conclusions From the comparison of the normalized calorimetric curves (Figure 3) for CnMeLA and MEGA-10, it follows that the gluco ring attached to the polyoxyethylene chain of MEGA-10 headgroup decreases hydrophobicity of the compound to a similar extent as shortening of its alkyl chain by one methylene group. The VolSurf descriptors DnDRY characterizing the hydrophobic interactions of molecules on lower energy levels are smaller for C10MeLA than for MEGA-10. This bears out the conclusion drawn from the above, that the alkyl chain of the C10MeLA molecules, being identical with that in MEGA10, behaves as if it was shorter in aggregational interactions. Molecular dynamics show that the modeled surfactant molecules are bent so that the alkyl chain and the headgroup are close to each other, resulting in partial screening the two molecular parts from interactions with surrounding molecules.
n-Alkanoyl-N-methyllactitolamine Surfactants Moreover, the two parts of the headgroup of CnMeLA, that is, gluco ring and polyoxyethylene chain also partially screen each other from interactions with water. In consequence, the interactions of gluco ring and polyoxyethylene chain with water, when they are fragments of the headgroup of CnMeLA, are weaker than when they constitute headgroups of CnGluc and MEGA0 n, respectively. These modeling results may explain why ∆Gm (hy) of CnMeLA is significantly lower than expected, taking into account that the headgroup of the compound is much bigger than the head groups of MEGA-10 and CnGluc. Acknowledgment. Calculations were performed in Wroclaw Centre for Networking and Supercomputing (WCSS). References and Notes (1) Wilk, K.; Syper, L.; Burczyk, B.; Maliszewska, I.; Jon, M.; Domagalska, B. J. Surfactants Deterg. 2001, 4, 155–161. (2) Drummond, C.; Wells, D. Colloids Surf., A 1998, 141, 131–142. (3) Ro´z˙ycka-Roszak, B.; Jurczak, B.; Wilk, K. Thermochim. Acta 2007, 453, 27–30. (4) Cruciani, G.; Pastor, M.; Guba, W. Eur. J. Pharm. Sci. 2000, 11 Suppl 2, S29-S39. (5) Cruciani, G.; Crivori, P.; Carrupt, P.-A.; Testa, B. J. Mol. Struct. 2000, 503, 17–30. (6) Crivori, P.; Cruciani, G.; Carrupt, P. A.; Testa, B. J. Med. Chem. 2000, 43, 2204–2216. (7) Stewart, J. J. P. MOPAC 2007; Stewart Computational Chemistry: Colorado Springs, CO, 2007. (8) Stewart, J. J. Mol. Model. 2007, 13, 1173–1213.
J. Phys. Chem. B, Vol. 112, No. 51, 2008 16551 (9) Besler, B. H., Jr.; Kollman, P. A. J. Comput. Chem. 1990, 11, 431– 439. (10) Klamt, A.; Schu¨u¨rmann, G. J. Chem. Soc., Perkin Trans. 2 1993, 799–805. (11) SYBYL 8.0; Tripos International: St. Louis, MO, 2007. (12) Clark, M.; Cramer, R. D., III; Van Opdenbosch, N. J. Comput. Chem. 1989, 10, 982–1012. (13) Tripos Bookshelf 8.0; Tripos International: St. Louis, MO, 2007; http://graphics.med.yale.edu:5080/TriposBookshelf/index.html. (14) Ro´z˙ycka-Roszak, B.; Cierpicki, T. J. Colloid Interface Sci. 1999, 218, 529–534. (15) GRID 22; Molecular Discovery Ltd.; http://www.moldiscovery.com. (16) Paula, S.; Su¨s, W.; Tuchtenhagen, J.; Blume, A. J. Phys. Chem. 1995, 99, 11742–11751. (17) Ro´z˙ycka-Roszak, B.; Misiak, P.; Woz´niak, E.; Mozrzymas, A.; Dega-Szafran, Z. Colloids Surf., A 2008, 318, 301–306. (18) Hait, S.; Moulik, S.; Palepu, R. Langmuir 2002, 18, 2471–2476. (19) Prasad, M.; Chakraborty, I.; Rakshit, A. K.; Moulik, S. P. J. Phys. Chem. B 2006, 110, 9815–9821. (20) Okawauchi, M.; Hagio, M.; Ikawa, Y.; Sugihara, G.; Murata, Y.; Tanaka, M. Bull. Chem. Soc. Jpn. 1987, 60, 2719–2725. (21) Oda, H.; Nagadome, S.; Lee, S.; Ohseto, F.; Sasaki, Y.; Sugihara, G. J. Surf. Sci. Technol. 1998, 14, 1–22. (22) Heerklotz, H.; Seelig, J. Biochim. Biophys. Acta 2000, 1508, 69– 85. (23) Shinoda, K.; Yamaguchi, T.; Hori, R. Bull. Chem. Soc. Jpn. 1961, 34, 237–241. (24) Tanford, C. The Hydrophobic Effect, 2nd ed.; Wiley-Interscience: New York, 1980. (25) Kresheck, G. J. Phys. Chem. B 1998, 102, 6596–6600. (26) Clark, D.; Pickett, S. Drug Disc. Today 2000, 5, 49–58.
JP805334T
