Unusual Behavior of the Aqueous Solutions of Gemini Bispyridinium

(1-4) Among the gemini surfactants, the structures most studied with regard to biological ... /CnH2n+1OOCCH2(CH3)2N + CH2CH2N + (CH3)2CH2COOCnH2n+1/2C...
1 downloads 0 Views 546KB Size
12312

J. Phys. Chem. B 2008, 112, 12312–12317

Unusual Behavior of the Aqueous Solutions of Gemini Bispyridinium Surfactants: Apparent and Partial Molar Enthalpies of the Dimethanesulfonates Emilia Fisicaro,*,† Carlotta Compari,† Mariano Biemmi,† Elenia Duce,† Monica Peroni,† Nadia Barbero,‡ Guido Viscardi,‡ and Pierluigi Quagliotto‡ Dipartimento di Scienze Farmacologiche, Biologiche e Chimiche Applicate, UniVersita` di Parma, Viale G.P. Usberti, 27/A, 43100 Parma, Italy and Dipartimento di Chimica Generale ed Organica Applicata e Centro di Eccellenza NISsSuperfici ed Interfacce Nanostrutturate, UniVersita` di Torino, Corso M. D’Azeglio, 48, 10125 Torino, Italy ReceiVed: May 14, 2008; ReVised Manuscript ReceiVed: July 8, 2008

Apparent and partial molar enthalpies at 298 K of the aqueous solutions of cationic gemini surfactants 1,1′didodecyl-2,2′-dimethylenebispyridinium dimethanesulfonate (12-Py(2)-2-(2)Py-12 MS); 1,1′-didodecyl-2,2′trimethylenebispyridinium dimethanesulfonate (12-Py(2)-3-(2)Py-12 MS); 1,1′-didodecyl-2,2′-tetramethylenebispyridinium dimethanesulfonate (12-Py(2)-4-(2)Py-12 MS); 1,1′-didodecyl-2,2′-octamethylenebispyridinium dimethanesulfonate (12-Py(2)-8-(2)Py-12 MS); 1,1′-didodecyl-2,2′-dodecamethylenebispyridinium dimethanesulfonate (12-Py(2)-12-(2)Py-12 MS) were measured as a function of concentration and are here reported for the first time. They show a very peculiar behavior as a function of the spacer length, not allowing for the determination of a -CH2- group contribution when this group is added to the spacer. The curve of the compound with a four-carbon-atom-long spacer lies between those of the compound with a spacer of 2 and 3 carbon atoms, instead of that below the latter, as expected. This surprising behavior, never found before in the literature and different from that found for the more popular m-s-m-type bisquaternary ammonium gemini surfactants, could be explained by a conformation change of the molecule, caused by stacking interactions between the two pyridinium rings, mediated by the counterion and appearing at an optimum length of the spacer. The hypothesis is also supported by the data obtained from the surface tension vs log c curves, showing that Amin, the minimum area taken at the air-water interface by the molecule, is significantly lower for 12Py(2)-4-(2)Py-12 MS than that of the other compounds of the same homologous series, and that the same compound has a greater tendency to form micelles instead of adsorbing at the air/water interface. The evaluation of the micellization enthalpies, by means of a pseudophase transition model, agrees with the exposed trends. These results confirm the great crop of information that can be derived from the study of the solution thermodynamics of aggregate systems and in particular from the curves of apparent and molar enthalpies vs concentration. Introduction Research in the field of gemini surfactantssi.e., surfactants in which at least two identical moieties are bound together by a spacer at the polar head levelshas developed very quickly, because of their advantages over monomeric surfactants, owing to their increased surface activity, lower critical micelle concentration (cmc), and useful viscoelastic properties.1-4 Among the gemini surfactants, the structures most studied with regard to biological activity and chemico-physical properties are the bisquaternary ammonium salts (bisQUATS). Cationic bisquaternary ammonium surfactants show a stronger biological activity than that of their corresponding monomers, and as a result are more active on both a molar and a weight scale, as far as germicidal activity and protein binding ability are concerned.4,5 Quite recently, the use of gemini surfactants as non-viral vectors in gene therapy has been proposed,6-9 on account of the possibility of taking advantage of their cationic * To whom correspondence should be addressed. E-mail: emilia.fisicaro@ unipr.it. † Dipartimento di Scienze Farmacologiche, Biologiche e Chimiche Applicate, Universita` di Parma. ‡ Dipartimento di Chimica Generale ed Organica Applicata e Centro di Eccellenza NIS - Superfici ed Interfacce Nanostrutturate, Universita` di Torino.

character, necessary for binding and compacting DNA, and of their superior surface activity. Gene therapy is, in fact, one of the major goals pursued by postgenomic research. It is based on the principle of curing a disease caused by a known defective gene by delivering a correct copy of the gene to the diseased cells by means of a specially designed viral or synthetic vector. Although viral vectors are generally very efficient in delivering genes into a targeted cell, their use is not without the risk of adverse or immunogenic reaction or replication, depending on the virus being used. As a result, nonviral vectors have, in many cases, become a preferred means of gene delivery into eukaryotic cells, despite their still low transfection efficiency. We have obtained encouraging results in gene delivery by using very simple bisquaternary ammonium gemini surfactants, derivatives of N,N-bisdimethyl-1,2-ethanediamine of general formula /CnH2n+1OOCCH2(CH3)2N+CH2CH2N+(CH3)2CH2COOCnH2n+1/ 2Cl- (bis-CnBEC), where the subscript n stands for the number of carbon atoms of the alkyl chain bound to the carboxyl group, when formulated with DOPE [L-R-phosphatidylethanolamine dioleoyl (C18:1,[cis]-9)].9 This encouraging result urged us to design and characterize new gemini compounds starting from those having two pyridinium groups, as a polar head, bridged together by an aliphatic chain, the synthesis of which we have

10.1021/jp804271z CCC: $40.75  2008 American Chemical Society Published on Web 09/04/2008

Aqueous Solutions of Gemini Bispyridinium Surfactants

J. Phys. Chem. B, Vol. 112, No. 39, 2008 12313

recently reported,10 with the aim of achieving a better insight into the interaction of gemini surfactants with DNA and membrane, of enriching the fundamental understanding of selfaggregation thermodynamics and of providing precious information for the theoretical evaluation of their properties. Experimental Section Materials. The series of bispyridinium cationic gemini surfactants under study was prepared by us, as described in ref 10. The compounds studied are the following: 1,1′-didodecyl2,2′-dimethylenebispyridinium dimethanesulfonate (12-Py(2)2-(2)Py-12 MS); 1,1′-didodecyl-2,2′-trimethylenebispyridinium dimethanesulfonate (12-Py(2)-3-(2)Py-12 MS); 1,1′-didodecyl2,2′-tetramethylenebispyridinium dimethanesulfonate (12-Py(2)4-(2)Py-12 MS); 1,1′-didodecyl-2,2′-octamethylenebispyridinium dimethanesulfonate (12-8-12PybisMS); 1,1′-didodecyl2,2′-dodecamethylenebispyridinium dimethanesulfonate (12Py(2)-12-(2)Py-12 MS). The compounds 12-Py(2)-3-(2)Py-12 MS and 12-Py(2)-4-(2)Py-12 MS were synthesized 3× and properly characterized a further 3× in order to confirm their unexpected behavior, as far as the curves of apparent and partial molar enthalpies are concerned. These two compounds were prepared either by direct synthesis with alkylethyl methanesulfonates or by synthesis with alkylethyl trifluoromethanesulfonates, followed by proper ionic exchange step. No differences were found from either the analytical or the characterization point of view. Purity was checked by NMR, elemental analysis, and ESI-MS (positive ions). The solutions were prepared by weight using freshly boiled bidistilled water, stored under nitrogen. Solution concentrations are expressed as molality, m (mol kg-1). Calorimetric Measurements. The enthalpies of dilution were measured by means of the Thermometric TAM (flow mixing cell) microcalorimeter, equipped with 221 Nano Amplifier, at 298 K. The freshly prepared surfactant solutions, kept before injection at the experimental temperature by means of a Heto cryothermostatic bath, were diluted into the “mixing” measuring cell of the microcalorimeter in a ratio 1:1 by using CO2-free water. The solutions and the water were injected by means of a Gilson peristaltic pump, Minipuls 2, and their flows were determined by weight. Surface Tension Measurements. The surface tension, γ, was measured by using a Lauda (TE1C/3) digital tensiometer. Measurements were made using the Du Nou¨y ring (Pt/Ir alloy (80/20), circumference: 60 ( 0.2 mm, wire diameter 0.4 mm, weight: 1.6 g). Sample temperature was maintained at 25.0 ( 0.1 °C by using a circulating water thermostatic bath (ISCO GTR 2000 IIx). The data were corrected according to the Zuidema and Waters method.11 The instrument was calibrated against double-distilled (and previously deionized) water, equilibrated against atmospheric CO2, each time measurements were done. Because the dicationic gemini surfactants adsorb onto negatively charged glass surfaces, all glassware was thoroughly soaked with the solution to be measured; soaking solutions were discarded. The fresh solution was aged for several hours before surface tension measurement. Sets of measurements were taken at 15 min intervals until no significant change occurred, using a very slow ring rising velocity. These tactics ensure that the ring is completely wetted. Standard deviation of the surface tension measurements is less than 0.15 mN/m. The absence of a minimum in the γ vs log c (c is the concentration expressed in mol/L) plot in the post-cmc region showed that there was very little or no surface active impurity present in the final products.

Figure 1. Apparent molar relative enthalpies of 1,1′-didodecyl-2,2′dimethylenebispyridinium dimethanesulfonate (12-Py(2)-2-(2)Py-12 MS, empty diamonds); 1,1′-didodecyl-2,2′-trimethylenebispyridinium dimethanesulfonate (12-Py(2)-3-(2)Py-12 MS, empty squares); 1,1′didodecyl-2,2′-tetramethylenebispyridinium dimethanesulfonate (12Py(2)-4-(2)Py-12 MS, empty triangles); 1,1′-didodecyl-2,2′-octamethylenebispyridinium dimethanesulfonate (12-Py(2)-8-(2)Py-12 MS, empty circles); 1,1′-didodecyl-2,2′-dodecamethylenebispyridinium dimethanesulfonate (12-Py(2)-12-(2)Py-12 MS, x) as a function of surfactant molality, m.

Results Calorimetric Measurements. The experimental data were expressed in terms of apparent and partial molar quantities of the solute, as is usual in solution thermodynamics, assuming infinite dilution as the reference state. Apparent and partial molar quantities were obtained from the experimental data using methods stated in detail elsewhere.12-18 For the sake of clarity, we recall that, with reference to the state of infinite dilution, the molar enthalpy of dilution, ∆Hd, is given by:

∆Hd ) LΦ,f - LΦ,i

(1)

where LΦ is the apparent relative molar enthalpy, and the indexes f and i stand for the final (after dilution) and initial (before dilution) concentrations, respectively. For ionic surfactant in the premicellar region, the apparent relative molar enthalpy can be expressed by means of a polynomial of m1/2. Stopping the serial expansion at the third term we obtain: 1

LΦ ) ALm ⁄2 + BLm + CLm3⁄2

(2)

where AL is the limiting Debye-Hu¨ckel slope for relative enthalpies accounting for the long-range electrostatic solute-solute interactions. Parameters BL and CL are obtained from the experimental points in the premicellar region by a least-squares curve fitting. In the micellar region, the apparent molar enthalpies are evaluated by means of eq 1 and, when a value of LΦ vs m not experimentally measured is needed, by graphical interpolation. The partial molar enthalpies L2 are determined by drawing the best curve for the apparent molar enthalpies vs m and then calculating the partial molar quantities as ∆(mLΦ)/∆m from points interpolated at regular intervals. Tables with all the experimental data are available as Supporting Information. Apparent and partial molar enthalpies vs m for the compounds under investigation, obtained as described above, are shown in Figures 1 and 2, respectively.

12314 J. Phys. Chem. B, Vol. 112, No. 39, 2008

Fisicaro et al.

Figure 2. Partial molar relative enthalpies of 1,1′-didodecyl-2,2′dimethylenebispyridinium dimethanesulfonate (12-Py(2)-2-(2)Py-12 MS, full diamonds); 1,1′-didodecyl-2,2′-trimethylenebispyridinium dimethanesulfonate (12-Py(2)-3-(2)Py-12 MS, full squares); 1,1′didodecyl-2,2′-tetramethylenebispyridinium dimethanesulfonate (12Py(2)-4-(2)Py-12 MS, full triangles); 1,1′-didodecyl-2,2′-octamethylenebispyridinium dimethanesulfonate (12-Py(2)-8-(2)Py-12 MS, full circles); 1,1′-didodecyl-2,2′-dodecamethylenebispyridinium dimethanesulfonate (12-Py(2)-12-(2)Py-12 MS, +) as a function of surfactant molality, m.

Surface Tension Measurements. The surface tension measurements (γ vs log c) were performed at 25 °C and gave the results reported in Table 1 and Figure 3. By using the Gibbs adsorption equation (eq 3),19,20 several parameters were determined from the γ vs log c plots: (i) the critical micelle concentration (cmc), taken as the concentrations at the point of intersection of the two linear portions of the γ vs. log c plots; (ii) the maximum surface excess concentration Γmax [mol cm-2]; (iii) the area per molecule at the interface Amin [nm2] from eq 4; (iv) the efficiency in surface tension reduction, measured by C20, i.e., the molar surfactant concentration required to reduce the surface tension of the solvent by 20 mN/m;21 (v) the effectiveness of the surface tension reduction, measured by the surface tension at the cmc, γcmc; and (vi) the cmc/C20 ratio, i.e., the measure of the tendency to form micelles relative to the tendency to adsorb at the air/water interface.

Γmax ) -

∂γ 1 2.303nRT ∂log C

Amin )

(

1015 NΓmax

)

T

(3) (4)

The value of n (the number of species whose concentration at the interface changes with c) is taken as 2, although for divalent geminis (having 1 surfactant ion and 2 non-surfactant counterions), the values of both 2 and 3 have been proposed.22 For gemini surfactants, it was found that one of the two counterions is frequently firmly wedged in between the two charged headgroups, especially when the spacer is quite short.23 While the use of a different n does not affect the general trend for surface areas, the last finding enabled us to use n ) 2 with some confidence. Discussion Thermodynamic properties of the aqueous solutions of surfactants, obtained by means of direct methods, are reported in the literature in quite a systematic way for anionic, cationic, and nonionic surfactants, in particular for the hydrogenated ones.11-18,24 One of the aims of these studies is to verify the

additivity rule and to find out the contribution of each group constituting the molecule, in order to be able to predict the thermodynamic behavior of the surfactant solutions starting from the chemical structure of the surfactant itself. Following this idea, the group contributions of the methylene and perfluoromethylene groups, as well as of the polar head constituents and of the different counterions, have been reported.11-18,24 Little data have been published up to now regarding the thermodynamics of gemini surfactant solutions,13,15,25-28 and many questions about their behavior are still open to discussion. For instance, it is not completely understood whether the group contribution approach is also valid in the case of gemini surfactants and whether it is possible to use the group contribution obtained from the study of monomers to predict the properties of gemini surfactant solutions. In fact, at least in the case of enthalpies, the effect of a methylene group when added to extend the hydrophobic tail in gemini surfactants seems to be greater than that in the case of monomers,13,15 although a lot of work is needed to confirm and explain this trend. Moreover, the first and most deeply studied class of cationic gemini surfactants, the alkanediyl-R, ω-bis(dimethylalkylammonium halide), shows a well-documented sphere to rod transition for the compounds having a short spacer,29-31 whereas compounds with longer spacers only give rise to spherical micelles. Timeresolved fluorescence quenching (TRFQ) techniques for the determination of the aggregation number and Cryo-TEM for the direct visualization of the aggregates have been used to obtain the corroborating experimental evidence. At concentrations close to the cmc, micelles of the didodecyl compound with spacer consisting of three methylene groups are spherical with an aggregation number of 25: that is to say that about 50 dodecyl chains stick together.29 This number increases significantly with concentration, suggesting a change in shape toward elongated micelles, confirmed by the cryo-TEM images and by the trend of the viscosity. Small angle neutron scattering (SANS) measurements agree with the above observations.30 The evaluation of apparent and partial molar enthalpies at 298 K of the aqueous solutions of the cationic gemini surfactants propanediylR,ω-bis(octyldimethylammonium bromide) and propanediylR,ω-bis(dodecyldimethyl-ammonium bromide) as a function of concentration allowed for the determination, besides the enthalpy of micellization, of the change in enthalpy associated with this sphere to rod transition in micellar phase,13 showing once more that the apparent and partial molar enthalpies vs concentration curves are able to provide very interesting information about the behavior in solution of surfactants. These considerations prompted us to measure, by means of calorimetric techniques, the curves of the apparent and partial molar enthalpies vs m for the new class of gemini compounds under investigation, not only to acquire thermodynamic information but also to obtain evidence for structural considerations. In general for ionic surfactants, these curves, after increasing in the premicellar region, tend to level off at concentrations above the cmc, where they are almost parallel. The lowering of the curves in the micellar region, proportional to the number of carbon atoms in the alkyl chain and, in general, to the global hydrophobicity of the molecule, is attributed to the electrostatic interactions in micellar solutions.12 The trends for the compounds having the same length of the hydrophobic tail, but different spacer lengths, reported in Figure 1 for the apparent and in Figure 2 for the partial molar enthalpies, respect this general trend but also show a very peculiar behavior. In fact, as outlined in Figure 4, the effect of the addition of a -CH2- to the spacer does not give rise to a monotonic change in the values of the enthalpies in

Aqueous Solutions of Gemini Bispyridinium Surfactants

J. Phys. Chem. B, Vol. 112, No. 39, 2008 12315

TABLE 1: Surface Tension Measurementsa

compound

s

cmcb (conduc.) [mmol/L]

12-Py(2)-2-(2)Py-12 MS 12-Py(2)-3-(2)Py-12 MS 12-Py(2)-4-(2)Py-12 MS 12-Py(2)-8-(2)Py-12 MS 12-Py(2)-12-(2)Py-12 MS

2 3 4 8 12

2.07 2.09 1.93 1.32 0.75

βb

∆Hmic [kJ/mol]

cmc (surf.tens.) [mmol/L]

γcmc [dyne/cm]

Γ × 1010 [mol/cm2]

Amin [nm2]

C20 [mmol/L]

74 72 40 57 57

-3.9 -5.5 -4.4 -10.8 17.6

0.52 1.00 1.15 0.29 0.11

42.96 42.06 45.95 42.19 41.97

1.02 1.05 1.20 1.10 1.02

16.2 15.9 13.9 15.1 0.015

0.09 0.15 0.42 0.05 7.0

cmc/C20 5.8 6.7 2.7 6.0

a Cmc, from conductivity and from surface tension; β, degree of counterion association; ∆Hmic, change in enthalpy upon micellization; Γ, the maximum surface excess concentration; Amin, the area per molecule at the interface; C20, the surfactant concentration required to reduce the surface tension of the solvent by 20 mN/m; γcmc, the surface tension at the cmc; the cmc/C20 ratio, as a function of the number of carbon atoms in the spacer, s for the surfactants under investigation. b Ref. 10.

Figure 3. Surface tension as a function of the logarithm of surfactant molarity, c of 1,1′-didodecyl-2,2′-dimethylenebispyridinium dimethanesulfonate (12-Py(2)-2-(2)Py-12 MS, full diamonds); 1,1′-didodecyl-2,2′trimethylenebispyridinium dimethanesulfonate (12-Py(2)-3-(2)Py-12 MS, full squares); 1,1′-didodecyl-2,2′-tetramethylenebispyridinium dimethanesulfonate (12-Py(2)-4-(2)Py-12 MS, full triangles); 1,1′didodecyl-2,2′-octamethylenebispyridinium dimethanesulfonate (12Py(2)-8-(2)Py-12 MS, full circles); 1,1′-didodecyl-2,2′-dodecamethylenebispyridinium dimethanesulfonate (12-Py(2)-12-(2)Py-12 MS, +).

Figure 4. Non-monotonic behavior as a function of the spacer length of the compounds with spacer constituted by 2 (diamonds), 3 (squares), and 4 (triangles) carbon atoms is emphasized.

the micellar region, that is to say that it is impossible to define a group contribution for the -CH2- group in the spacer. However, it is difficult to predict what the extent of its effect could be from the data on monomeric surfactants: if the -CH2is added to the hydrophobic chain, it causes a lowering of the enthalpic curves of about -1.5 kJ mol-1,12 but, if it is added to the polar head, the effect is exactly the opposite. For instance, we have shown that the enthalpic curves of dodecyldimethylethylammonium bromide (DEDAB) at 298 K, notwithstanding the greater total number of carbon atoms, lie above those of

dodecyltrimethylammonium bromide (DTAB), allowing us to evaluate the group contribution to the plateau value of a methylene when added to the polar head as being about +1.6 kJ mol-1 group-1 for LΦ and about +1.75 kJ mol-1 group-1 for L2. The effect is very close to that obtained by shortening the alkyl chain by one methylene group.12,32,33 The very short second chain present on the nitrogen bearing the positive charge allows for both a better charge delocalization and an increase in the size of the polar head, so that the charges are farther apart on the micelle surface of DEDAB than on that of DTAB. However, we have compared the behavior of different dodecylic surfactants having the same counterion, in order to rationalize the effect of the polar head on the enthalpic properties of their solutions.34 We have thus established a sort of thermodynamic “charge localization scale” for the polar head: the more delocalized the charge, the more similar the behavior to that of a nonionic surfactant. The lowering of the plateau value of the apparent and partial molar enthalpy curves vs m in a homologous series of surfactant is related to the increase in hydrophobicity of the alkyl chain and to the size of micelles; in contrast, for the same alkyl chain and counterion the plateau value is strictly related to the modulation of the charge density due to inductive or resonance effects. These effects do not play a great role in the value of the cmc (i.e., in the free energy of micelle formation)se.g., the cmc of DEDAB is only slightly smaller than that of DTABsalthough they strongly affect the solution enthalpies and, as a consequence, the entropies. The above considerations apply only to a very short alkyl chain bound to the polar head of a monomeric surfactant, but, if the length of the chain increases, the hydrophobic effect prevails, as has been shown for N-dodecyl-N-benzylmorpholinium chloride, in which the benzyl group, bound to the positive nitrogen, can be treated as a second chain.35 In any case, literature data regarding this subject are totally lacking. The situation becomes more complicated with the gemini surfactants under investigation: the values of cmc reported in Table 1 confirm that the addition of methylene groups to the spacer does not monotonically lower the cmc values and does not significantly increase the hydrophobicity of the overall molecule.36 The enthalpic curves of Figures 1 and 2 show that the curve of 12-Py(2)-3-(2)Py-12 MS is lower than that of 12-Py(2)-2-(2)Py-12 MS by about -2.0 kJ mol-1 for the apparent ones, a quantity greater than that found for the addition of one -CH2- in the alkyl tail of a monomeric surfactant, but the further addition of a -CH2- in the spacer (obtaining the 12-Py(2)-4-(2)Py-12 MS compound) gives rise to an inversion of the trend with the curve for the 12-Py(2)-4(2)Py-12 MS lying between 12-Py(2)-2-(2)Py-12 MS and 12Py(2)-3-(2)Py-12 MS, nearer to 12-Py(2)-2-(2)Py-12 MS. This absolutely surprising behavior has been confirmed experimentally many times, also starting from a new synthetic batch of

12316 J. Phys. Chem. B, Vol. 112, No. 39, 2008

Fisicaro et al.

Figure 6. Cmc (diamonds) and C20 (squares) as a function of the spacer length, s.

Figure 5. Suggested structure in solution of the molecule of 1,1′didodecyl-2,2′-tetramethylenebispyridinium dimethanesulfonate (12Py(2)-4-(2)Py-12 MS). Counterions are omitted for clarity.

the compounds. The micellar plateau region of the apparent molar enthalpies vs m of 12-Py(2)-8-(2)Py-12 MS is lowered by about -8 kJ mol-1 with respect to 12-Py(2)-4-(2)Py-12 MS, giving rise to an average group contribution of -2 kJ mol-1, comparable to that between 12-Py(2)-2-(2)Py-12 MS and 12Py(2)-3-(2)Py-12 MS, but the difference in the same region between 12-Py(2)-8-(2)Py-12 MS and 12-Py(2)-12-(2)Py-12 MS becomes lower, the group contribution for each -CH2-, thus approaching that for the same group in the alkyl chain of the monomer surfactants. Moreover, none of the trends of the curves show any peculiarity in the range of concentrations examined: that is to say that sphere to rod transitions in micellar solution do not occur. We can explain this surprising behavior arising with the lengthening of the spacer and never found before in the literature, by means of a conformation change of the molecule: when the spacer reaches the right size, the molecule doubles up, like a book, (Figure 5) giving rise to stacking interactions between the two pyridinium rings, mediated by the counterion. This arrangement is not possible when the spacer is too short (2 or 3 carbon atoms), because of the loss of sufficient conformational freedom, and when the spacer is too long, because the pyridinium rings are too far apart. So compound 12-Py(2)-4-(2)Py-12 MS behaves in a different way inside the homologous series we have examined, with regard to both enthalpic and also surface tension vs log c curves (Figure 3). We outline the strong difference between the effect of the spacer here reported and that for the m-s-m-type bisquaternary ammonium surfactants.29-31 In fact, the lengthening of the spacer in the m-s-m-type surfactants brings about a change in the structure of micelles due to the hydrophobic interactions inside the molecule, modifying the arrangement of the spacer and, as a consequence, the packing parameter of the whole molecule. The above consideration gives added value to the determination of the curves of the apparent and partial molar enthalpies vs m in the study of the behavior of surfactants in solution. In fact, they are not only able to show the presence of relevant phase transition in micellar solution (for instance sphere to rod), which appears as a small step in the micellar range of concentrations,13 but also, when inside a homologous series the group contribution additivity does not hold, they constitute evidence of a conformational change in the molecule.

The hypothesis is supported by the information obtained from eqs 3 and 4 and reported in Table 1. In fact, the minimum area taken on the surface by the molecule, Amin is significantly lower for 12-Py(2)-4-(2)Py-12 MS than the compounds of the same homologous series, suggesting a tendency of the molecules to fold. Moreover, as shown in Figure 6, the values of cmc from surface tension and C20 show a maximum when the spacer is 4 methylene long, and the further addition of 8 methylene groups to the spacer, giving 12-Py(2)-12-(2)Py-12 MS, lowers the cmc by only 1 order of magnitude. The higher value of C20, together with the lower ratio cmc/C20 and the higher value of γcmc, suggests a greater tendency of 12-Py(2)-4-(2)Py-12 MS to form micelles instead of adsorbing at the air/water interface: the more compact shape assumed by the molecule could help the assembly in micelles. Moreover, it appears from the data in Table 1 that the values of cmc from surface tension are lower than those from conductivity. Similar results have been obtained by Pinazo et al. 37 for hydrogenated gemini surfactants derived from arginine. They noticed that the cmcs obtained by fluorescence and conductivity measurements were similar to each other but somewhat higher than that determined by surface tension, a result shown very recently also by Esumi et al. 38 for trimeric surfactants. Very recently, Rosen et al. found that the occurrence of premicellar aggregates, especially for very short spacer gemini surfactants, caused the cmc, as determined by both surface tension and conductivity, to be substantially different in value.22 This finding allows for the hypothesis that, at a concentration lower than cmc, the surfactants under investigation could give small premicellar aggregates that are not surface-active, preventing the adsorption on the surface from continuing. In order to obtain the enthalpy change upon micellization, ∆Hmic, we applied a pseudophase transition model, in which the aggregation process is considered as a phase transition, taking place at equilibrium. In this model, it is assumed that, at the cmc, the partial molar properties present a discontinuity due to the formation of the pseudophase. The micellization parameters, obtained by extrapolating at the cmc the trends of the partial molar properties before and after cmc,11-18 together with the cmc values from conductometric measurements, are reported in Table 1. The micellization enthalpies reflect the same trend as that described before, without the possibility to extract a group contribution for the -CH2-, and the same inversion between 12-Py(2)-3-(2)Py-12 MS and 12-Py(2)-4-(2)Py-12 MS. Conclusions Apparent and partial molar enthalpies at 298 K of the aqueous solutions of the homologous series of cationic gemini surfactants

Aqueous Solutions of Gemini Bispyridinium Surfactants 1,1′-didodecyl-2,2′-alkylenebispyridinium dimethanesulfonate, obtained by means of direct methods, are here reported for the first time, with the aim of enriching the fundamental understanding on self-aggregation thermodynamics of gemini surfactants. These show a very peculiar behavior as a function of the spacer length, not allowing for the determination of a -CH2- group contribution when this group is added to the spacer. The curve of the compound with spacer formed by four carbon atoms lies between those of the compound with spacer of two and three carbon atoms, instead of that below the latter, as expected. This surprising behavior, never found before in the literature, is evidence of a conformational change of the molecule caused by stacking interactions between the two pyridinium rings, mediated by the counterion and appearing at an optimum length of the spacer. The hypothesis is also supported by the data obtained from the surface tension vs log c curves, showing that Amin, the minimum area taken on the surface by the molecule, is significantly lower for 12-Py(2)-4(2)Py-12 MS than that of the other compounds of the same homologous series, and that the same compound has a greater tendency to form micelles instead of adsorbing at the air/water interface. The evaluation of the micellization enthalpies by means of a pseudophase transition model agrees with the revealed trends. These results confirm the great crop of information that can be derived from the study of solution thermodynamics of aggregate systems and in particular from the curves of apparent and molar enthalpies vs concentration. Acknowledgment. This work was supported by the contributions from the Fondazione Cassa di Risparmio di Parma and from the University of Parma (FIL 2007). The authors are grateful to Compagnia di San Paolo (Torino, Italy) and the Fondazione Cassa di Risparmio di Torino for having supplied laboratory equipment. Supporting Information Available: Tables collecting Molality (m), enthalpies of dilution (∆Hd), apparent (LΦ) and partial molar (L2) enthalpies of 1,1′-didodecyl-2,2′-dimethylenebispyridinium dimethanesulfonate (12-Py(2)-2-(2)Py-12 MS); 1,1′didodecyl-2,2′-trimethylenebispyridinium dimethanesulfonate (12-Py(2)-3-(2)Py-12 MS); 1,1′-didodecyl-2,2′-tetramethylenebispyridinium dimethanesulfonate (12-Py(2)-4-(2)Py-12 MS); 1,1′didodecyl-2,2′-octamethylenebispyridinium dimethanesulfonate (12-Py(2)-5-(2)Py-12 MS); and 1,1′-didodecyl-2,2′-dodecamethylenebispyridinium dimethanesulfonate (12-Py(2)-12-(2)Py12 MS) in water at 298 K are available free of charge via the Internet at http://pubs.acs.org. References and Notes (1) Rosen, M.; J.; Tracy, D. J. J. Surfactants Deterg. 1998, 1, 547, and references therein. (2) Menger, F. M.; Keiper, J. S. Angew. Chem., Int. Ed. 2000, 39, 1906. (3) Zana R., Novel Surfactants- Preparation, Applications, and Biodegradability. Dimeric (Gemini) Surfactants, In Surfactant Science Series; Holmberg K., Eds.; Marcel Dekker, Inc.: New York,1998; Vol. 74; p 241. (4) Fisicaro, E. Cell. Mol. Biol. Lett. 1997, 2, 45, and references therein.

J. Phys. Chem. B, Vol. 112, No. 39, 2008 12317 (5) Viscardi, G.; Quagliotto, P.; Barolo, C.; Savarino, P.; Barni, E.; Fisicaro, E. J. Org. Chem. 2000, 65, 8197. (6) Bell, P. C.; Bergsma, M.; Dolbnya, I. P.; Bras, W.; Stuart, M. C. A.; Rowan, A. E.; Feiters, M. C.; Engberts, J. B. F. N. J. Am. Chem. Soc. 2003, 125, 1551. (7) Kirby, A. J.; Camilleri, P.; Engberts, J. B. F. N.; Feiters, M. C.; Nolte, R. J. M.; Soderman, O.; Bergsma, M.; Bell, P. C.; Fielden, M. L.; Garcia Rodriguez, C. L.; Guedat, P.; Kremer, A.; McGregor, C.; Perrin, C.; Ronsin, G.; van Eijk, M. C. P. Angew. Chem., Int. Ed. 2003, 42, 1448. (8) Chen, X.; Wang, J.; Shen, N.; Luo, Y.; Lin, L.; Liu, M.; Thomas, R. K. Langmuir 2002, 18, 6222. (9) Fisicaro, E.; Compari, C.; Duce, E.; Donofrio, G.; Roˇøycka-Roszak, B.; Wou¨niak, E. Biochim. Biophys. Acta, Gen. Subj. 2005, 1722, 224. (10) Quagliotto, P.; Viscardi, G.; Barolo, C.; Barni, E.; Bellinvia, S.; Fisicaro, E.; Compari, C. J. Org. Chem. 2003, 68, 7651. (11) Zuidema, H. H.; Waters, G. W. Ind. Eng. Chem. Anal. Ed. 1941, 13, 312. (12) Desnoyers J. E.; Perron, G.; in: Surfactant SolutionssNew Methods of InVestigation; Zana, R., Ed.; Marcel Dekker, Inc.: New York, 1987, p. 1 , and references therein. (13) Fisicaro, E.; Compari, C.; Duce, E.; Contestabili, C.; Quagliotto, P.; Viscardi, G. J. Phys. Chem. B 2005, 109, 1744. (14) Fisicaro, E.; Compari, C.; Quagliotto, P.; Viscardi, G. J. Colloid Interface Sci. 2002, 255, 410. (15) Ro´øycka-Roszak, B.; Fisicaro, E.; Ghiozzi, A. J. Colloid Interface Sci. 1996, 184, 209. (16) Fisicaro, E.; Pelizzetti, E.; Barbieri, M.; Savarino, P. J. Chem. Soc., Faraday Trans. 1991, 87, 2983. (17) Fisicaro, E.; Pelizzetti, E.; Barbieri, M.; Viscardi, G.; Savarino, P. Thermochim. Acta 1990, 168, 143. (18) De Lisi, R.; Fisicaro, E.; Milioto, S.; Pelizzetti, E.; Savarino, P. J. Solution Chem. 1990, 19, 247. (19) Alami, E.; Beinert, G.; Marie, P.; Zana, R. Langmuir 1993, 9, 1465. (20) Pe´rez, L.; Pinazo, A.; Rosen, M. J.; Infante, M. R. Langmuir 1998, 14, 2307. (21) Rosen, M. J. Surfactants and Interfacial Phenomena, second ed.; John Wiley & Sons: New York, 1989. (22) Tsubone, K.; Arakawa, Y.; Rosen, M. J. J. Colloid Interface Sci. 2003, 26, 2516. (23) Li, Z. X.; Dong, C. C.; Thomas, R. K. Langmuir 1999, 15, 4392. (24) van Os, N. M.; Haak, J. R.; Rupert, L. A. M. In Physico-Chemical Properties of Selected Anionic, Cationic and Nonionic Surfactant; Elsevier, Amsterdam, 1993. (25) Grosmaire, L.; Chorro, M.; Chorro, C.; Partyka, S.; Zana, R. Prog. Colloid Polym. Sci. 2000, 115, 31, Trends in Colloid and Interface Science XIV. (26) Grosmaire, L.; Chorro, M.; Chorro, C.; Partyka, S.; Zana, R. J. Colloid Interface Sci. 2002, 246, 175. (27) Bai, G.; Wang, J.; Yan, H.; Li, Z.; Thomas, R. K. J. Phys. Chem. B 2001, 105, 3105. (28) Bai, G.; Wang, J.; Yan, H.; Li, Z.; Thomas, R. K. Langmuir 2001, 17, 4501. (29) Danino, D.; Talmon, Y.; Zana, R. Langmuir 1995, 11, 1448. (30) Hirata, H.; Hattori, N.; Ishida, M.; Okabayashi, H.; Furusaka, M.; Zana, R. J. Phys. Chem. 1995, 99, 17778. (31) Hattori, N.; Hirata, H.; Okabayashi, H.; Furusaka, M.; O’Connor, C. J.; Zana, R. Colloid Polym. Sci. 1999, 277, 95. (32) Fisicaro, E.; Biemmi, M.; Compari, C.; Duce, E.; Peroni, M. J. Colloid Interface Sci. 2007, 305, 301. (33) De Lisi, R.; Fisicaro, E.; Milioto, S. J. Solution Chem. 1988, 17, 1015. (34) Fisicaro, E.; Pelizzetti, E.; Barbieri, M.; Viscardi, G.; Savarino, P. Thermochim. Acta 1990, 168, 143. (35) Ro´øycka-Roszak, B.; Fisicaro, E. Thermochim. Acta 1992, 205, 19. (36) Zana, R.; Benrraou, M.; Rueff, R. Langmuir 1991, 7, 1072. (37) Pinazo, A.; Wen, X.; Perez, L.; Infante, M. R.; Franses, E. I. Langmuir 1999, 15, 3134. (38) Yoshimura, T.; Yoshida, H.; Ohno, A.; Esumi, K. J. Colloid Interface Sci. 2003, 267, 167.

JP804271Z