Host-guest interactions between cyclodextrins and surfactants with functional groups at the end of the hydrophobic tail

1 Graphical abstract In this work the influence of the incorporation of aromatic substituents at the end of the hydrophobic tail on the binding of cationic surfactants to cyclodextrins was studied. 2 HOST-GUEST INTERACTIONS BETWEEN CYCLODEXTRINS AND SURFACTANTS WITH FUNCTIONAL GROUPS AT THE END OF THE HYDROPHOBIC TAIL Victoria Isabel Martín, Francisco José Ostos, Manuel Angulo1, Antonio Márquez, Pilar López-Cornejo, Manuel López-López2, Ana Teresa Carmona3 and María Luisa Moyá* Department of Physical Chemistry, University of Seville, C/ Profesor García González 1, 41012 Seville. Spain. Tlf. 34954557175. E-mail:moy[email protected] 1NMR Service, University of Seville, Apartado 1203, E-41071 Seville, Spain 2Department of Chemical Engineering, Physical Chemistry and Materials Science, Faculty of Experimental Sciences, Campus de El Carmen, Avda. de las Fuerzas Armadas s/n, 21071 Huelva. Spain 3Department of Organic Chemistry, University of Seville, C/Profesor García González 1, 41012 Seville. Spain Key words: Surfactants, cyclodextrins, inclusion complexes, aromatic substituents, conductivity, NMR. *Author to whom all correspondence should be directed 3 Abstract The aim of this work was to investigate the influence of the incorporation of substituents at the end of the hydrophobic tail on the binding of cationic surfactants to α-, β-, and -cyclodextrins. The equilibrium binding constants of the 1:1 inclusion complexes formed follow the trend K1(α-CD)>K1(β-CD)>>K1(-CD), which can be explained by considering the influence of the CD cavity volume on the host-guest interactions. From the comparison of the K1 values obtained for dodecyltriethylammonium bromide, DTEAB, to those estimated for the surfactants with the substituents, it was found that the incorporation of a phenoxy group at the end of the hydrocarbon tail does not affect K1, and the inclusion of a naphthoxy group has some influence on the association process, slightly diminishing K1. This makes evident the importance of the contribution of hydrophobic interactions to the binding, the length of the hydrophobic chain being the key factor determining K1. However, the presence of the aromatic rings does influence the location of the host and the guest in the inclusion complexes. The observed NOE interactions between the aromatic protons and the CD protons indicate that the aromatic rings are partially inserted within the host cavity, with the cyclodextrin remaining close to the aromatic rings, which could be partially intercalated in the host cavity. To the authors´ knowledge this is the first study on the association of cyclodextrins with monomeric surfactants incorporating substituents at the end of the hydrophobic tail. 1. Introduction Cyclodextrins, CD, are cyclic oligosaccharides formed through (1-4) ether linkages of glucopyranoside units [1,2]. The most common CDs, -, -, and -CD, are composed of six, seven and eight glucose units, respectively. CDs are shaped like a truncated cone with internal cavities ranging from 5 to 8 Å. The hydroxyl functions are oriented to the exterior of the cavity, with the secondary hydroxyl groups located on the wider edge, and the primary 4 ones on the narrow edge. The C-H bonds on the ring and the nonbonding electron pairs of the glycosidic oxygen bridges point inward. As a result of this spatial arrangement of the functional groups, the cavity shows a relatively hydrophobic character while the external surfaces are hydrophilic. This is responsible for both their water solubility and their ability to form inclusion complexes with molecular guests of suitable size. This capacity to form inclusion complexes with a wide variety of molecules, together with the non-toxicity towards humans, has been the basis for the CDs large range of applications [3-9]. The understanding of the driving forces involved in the CD inclusion complex formation is fundamentally important not only in CD chemistry, but also for supramolecular chemistry as a whole. In a recent review Valente and Söderman [10] pointed out that surfactants are ideal guests for fundamental studies on the complexation with CDs since both hydrophobic and hydrophilic regions of the surfactant molecules can be systematically varied. These authors examined the effect of different surfactant architectures on the formation of inclusion complexes by considering the results obtained by several authors for single tailed, double tailed, gemini and bolaform surfactants, with special emphasis on cationic surfactants. However, to the authors´ knowledge, the influence of the incorporation of a functional group at the end of the hydrophobic surfactant tail on the surfactant:CD interactions has not been investigated. With this in mind, the surfactants triethyl(1-phenoxydodecyl)ammonium bromide (Phenoxy12) and triethyl(2-naphthoxydodecyl)ammonium bromide (Naphthoxy12) were prepared in this work and their interactions with -, -, and -cyclodextrins studied. In order to help the discussion of the results, the formation of host:guest complexes between dodecyltriethylammonium bromide (DTEAB) and CDs was also investigated. Since there is not much information about surfactants with functional groups at the end of the hydrophobic tail in the literature, a brief discussion of the physicochemical properties of Phenoxy12 and 5 Naphthoxy12 aqueous solutions was done before considering the formation of the inclusion complexes. N CH2-CH3 Br ( )11 DTEAB Phenoxy12 ON BrH Naphthoxy12 ( )11 CH2-CH3 CH2-CH3 CH2-CH3 ON ( )11 CH2-CH3 CH2-CH3 CH2-CH3 Br CH3 CH3-CH2 CH3-CH2 Scheme 1.- Structure of the surfactants used in this work. The results obtained in this work will contribute to the understanding of the surfactant:cyclodextrin interactions. This is important in relation to the wide range of applications of both CDs and surfactants, which can be increased by taking advantage of the CD-surfactant complex formation. 2. Experimental section 2.1. Materials and chemicals Dodecyltrymethylammonium bromide, DTAB, was from Sigma-Aldrich. -, -, and -cyclodextrins of the highest purity available were also purchased from Aldrich (>99% purity, according to the manufacturer) and were kept under vacuum. DTEAB was prepared in a previous work [11] and its synthesis is briefly described in the Supplementary Material. The preparation of Phenoxy12 and Naphthoxy12 is described below. The surfactants were characterized by 1H NMR, 13C NMR and elemental analysis (CITIUS, University of Seville). D2O was supplied by Sigma. Water was MilliQ (resistivity >18 M cm). 6 2.2. Preparation of the surfactants 2.2.1. Preparation of Phenoxy12 The synthesis of Phenoxy12 was performed according to Scheme 2. Starting from commercial 1,12-dibromo-dodecane, the phenoxy group, PhO, was introduced at the end of the fatty alkyl chain by nucleophilic substitution reaction with sodium phenoxide in acetone, thus giving compound 1 (12-bromo-1-phenoxydodecane) in 29% yield. Finally, a nucleophilic displacement reaction with acetonitrile and triethylamine gave the compound 2 (triethyl(1phenoxydodecyl) ammonium bromide) in 75% yield. Its 1H NMR spectrum indicated the appearance of a triplet and a quartet signals integrating for nine and six protons, respectively, corresponding to the new three ethyl groups. Procedures for the preparation of the surfactant and intermediates are described in detail in Supplementary Material. Br (CH2)10Br +PhONa acetone Br (CH2)10 OPh N(CH)10 OPh Br triethylamine CH3CN 1 2 Scheme 2.- Synthesis of Phenoxy12 2.2.2. Preparation of Naphthoxy12 The synthesis of Naphthoxy12 was similar to that of Phenoxy12, as it is shown in Scheme 3. In this scheme the naphthoxy group is represented by NaphO. Commercial 1,12dibromo-dodecane and sodium naphthoxide were used in the nucleophilic substitution reaction to render compound 3 (12-bromo-1-naphthoxydodecane), in 52% yield. Finally, the nucleophilic displacement reaction with acetonitrile and triethylamine gave the compound 4 (triethyl(1-naphthoxydodecyl) ammonium bromide) in 97% yield. Similarly to Phenoxy12, 7 +NaphONa acetone Br triethylamine CH3CN 3 4 Br Br 10 NaphO Br 10 NNaphO 10 Scheme 3.- Synthesis of Naphthoxy12 1H NMR spectrum of Naphthoxy12 indicated the appearance of a triplet and a quartet signals integrating for nine and six protons, respectively, corresponding to the new three ethyl groups. Procedures for the preparation of the surfactants and intermediates are also described in detail in Supplementary Material. 2.3. Methods 2.3.1. Conductivity measurements Conductivity was measured with a Crison GLP31 conductimeter calibrated with KCl solutions of the appropriate concentration range. The conductimeter was connected to an external water circulator (Heto) and the whole system was placed in a room in which the temperature was kept constant within ±0.5 K. Temperature was maintained at 3030.01 K. Solutions were used within 5 h after preparation. In a typical experiment a surfactant solution was placed in the thermostated conductivity cell; then, aliquots of the CD solution, in the presence of the same surfactant concentration, were added in a stepwise manner using a programmable dispenser Crison Burette 1S (0.1 L). The specific conductivity of the solution was measured 10 min after each addition, after checking that the specific conductivity remained constant with time. Each experiment was repeated at least twice. The critical micellar concentrations of Phenoxy12 and Naphthoxy12 were estimated by means of conductivity measurements as described in ref. 12. 8 2.3.2. Surface tension measurements Surface tension was measured by a du Noüy ring method using a KSV 703 digital tensiometer (Finland) as described in ref. 12. 2.3.3. NMR measurements The NMR spectra were performed in CITIUS (Research General Services for the University of Seville). NMR samples were prepared by dissolving the corresponding amount of the surfactant and/or the CD in D2O followed by a brief sonication. The solutions were kept thermostated at 303 K for at least 5 hours before carrying out the NMR experiments. NMR experiments were recorded on a Bruker Avance III 500 MHz spectrometer (500.2 MHz for 1H) equipped with a 5 mm TCI cryoprobe operating at 303 K. All 1H chemical shifts are referenced to the residual HDO signal set to 4.71 ppm [13]. Two-dimensional, 2D, rotating frame nuclear Overhauser effect experiments were performed using the Bruker standard pulse sequence (EASY-ROESY version [14]). 2048 x 256 data points were acquired with 16 transients per increment and a relaxation delay of 1.5 s. A mixing time of 250 ms was used. Data processing was performed on a 1024 x 1024 data matrix. Cosine-squared window functions were used along F1 and F2. 3. Results and discussion 3.1. Physicochemical properties of the Phenoxy12 and Naphthoxy12 aqueous solutions The critical micellar concentration, cmc, and the micellar ionization degree, , of Phenoxy12 and Naphthoxy12 in aqueous solutions were determined using conductivity measurements. Figure S1 (Supplementary Material) shows the dependence of the specific conductivity on Phenoxy12 and on Naphthoxy12 concentrations at 303 K. The Carpena method [15] was used in order to obtain the cmc and α values from the experimental results. These data are summarized in Table 1, together with that corresponding to DTEAB. The Gibbs energy of micellization, GoM, can be calculated by using eq. 1 [16]: 9 GoM= RT(2-α) ln cmc (1) where cmc is expressed in mole fraction and R and T have their usual meaning. GoM values are listed in Table 1. Comparison of the GoM values obtained for the three surfactants shows Table 1.-Critical micellar concentration, cmc, micellar ionization degree, , and Gibbs energy of micellization, GoM, for the cationic surfactants studied in this work, at 303 K. Surfactant Cmc/mM  GoM/kJ mol-1 DTEABa 14.30.4 0.350.02 -34.31.8 Phenoxy12b 3.70.2 0.400.03 -38.81.7 Naphthoxy12b 0.6410.015 0.430.03 -45.01.9 aRef. 11; bThis work. that the introduction of a phenoxy and a naphthoxy group at the end of the surfactant hydrophobic tail substantially favors micellization. The experimental observations can be explained by taking into account the transfer Gibbs energy contribution, Gotransf, to the Gibbs energy of micellization, GoM. Gotransf considers the transfer of the hydrophobic surfactant chains from the aqueous phase to the micellar interior and it is the driving force for the selfassociation process of surfactants [17]. The surfactants listed in Table 1 have a dodecyl hydrophobic chain and the corresponding Gotransf contribution would be the same for all of them. The 4-fold and a 22-fold diminution in the cmc, with respect to that of DTEAB, caused by the incorporation of a phenoxy group, C6H5O-, and of a naphthoxy group, C12H7O-, at the end of the hydrophobic tail can be rationalized by considering the additional hydrophobic contribution to Gotransf due to the transfer of the C6H5Oand C12H7Ogroups into the micelles. The large difference found between the cmc´s of Phenoxy12 and Naphthoxy12 could be accounted for by the different hydrophobicity of these two aromatic substituents. As an example, the logarithm of the octanol/water partitition coefficient, logP, is 1.46 and 2.70 for phenol and naphthol, respectively [18]. 16 orientation of the surfactant, with the aromatic moiety located close to the narrower rim of the cyclodextrin. Interactions between the CD protons and the aromatic surfactant protons are observed for Phenoxy12 and Naphthoxy12. Considering the ROESY spectra of the Phenoxy12 surfactant, one can see NOE interactions between the aromatic H1' and H2' protons (stronger for H1') of the surfactant and protons of the cyclodextrins for -, -, and -CD. For Naphthoxy12 (see Figure 4S, Supplementary Material) NOE interactions between the protons H1', H3´, H4´, and H8´ (stronger for H1´and H3´) and protons of the CDs are observed. These findings point out that in the formed inclusion complexes the cyclodextrin remains close to the aromatic rings, which could be partially intercalated in the host cavity. They also show the dynamic character of the inclusion complex formation, which associates and dissociates with a frequency that would depend, for a given guest, on the CD nature. This dynamic character could explain the really interesting fact that the interaction between the CD protons and all the aromatic protons of Phenoxy12 is only observed in the case of -CD. This cyclodextrin has the largest cavity volume of the three CDs, which would make the dissociation easier than for - and -CD, increasing the probability of interactions between the CD and the aromatic protons. The estimated equilibrium binding constants obtained in this work support this assumption (see below). 3.3. Stoichiometry Prior to the calculation of the equilibrium binding constants of the inclusion complexes, the binding stoichiometry of the CD:Surfactant host-guest complexes has to be estimated. In order to do so Job´s method was used [26]. It is observed that when CDs are added to an aqueous ionic surfactant solution, at constant surfactant concentration, an increase in CD concentration could result in a decrease in the experimental specific conductivity. This decrease can be ascribed to the formation of CD:Surfactant, CDS, inclusion complexes, which 17 have considerably smaller ionic equivalent conductivity than those of surfactant monomers [27]. For this reason, conductivity measurements can be used in order to get information about ionic surfactants/CD interactions. Figure 3 shows some of the Job´s plots obtained for the different surfactants and cyclodextrins investigated, where the dependence of (κobs)× [CDT] on the CD molar fraction was shown, obs being the experimental specific conductivity. In all cases only 1:1 complexes, CDS, are formed under the working conditions. 0.0 0.2 0.4 0.6 0.8 1.0 0 2 4 6 8 10 XCD DTEAB:-CD Maximum at XCD=0.50 a) 103x(obs[CDT] (mol m-4) 0.0 0.2 0.4 0.6 0.8 1.0 0 2 4 6 8 10 12 XCD Phenoxy12:-CD Maximum at XCD=0.50 103x(obs[CDT] (mol m-4) b) 0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.5 1.0 1.5 2.0 Naphthoxy12:-CD XCD Maximum at XCD=0.50 c) 103x(obs[CDT] (mol m-4) Fig. 3.-Job´s plots at 303 K. a)DTEAB:-CD; b)Phenoxy12:-CD; c)Naphthoxy12:β-CD. 3.4. Formation equilibrium constants for the inclusion complexes The association process between the surfactants and the cyclodextrins has been studied by varying the CD concentration, for a constant surfactant concentration lower than the cmc. 18 The stability of the inclusion complexes can be described in terms of the equilibrium binding constants, K1. For a 1:1 complex K1 can be defined as: CD + S  CDS K1=[CDS] [CD ][S] (2) From the mass conservation law equations and taking into account that the experimental specific conductivity is the sum of the contributions coming from the surfactant free ions, the bromide counterions and the CDS inclusion complexes, the observed decrease in the molar conductance of the surfactant aqueous solutions due to the addition of CD, obs, can be expressed as [28]: ΔΛ𝑜𝑏𝑠 =Δ𝜆 2𝐾1[𝑆𝑇]{𝐾1([𝑆𝑇]+[𝐶𝐷𝑇])+ +1−((𝐾1 ([𝑆𝑇]+[𝐶𝐷𝑇])+1)2−4𝐾1 2 ([𝑆𝑇]+[𝐶𝐷𝑇]))1/2} (3) where,  is the difference in the ionic conductivities of the unassociated, S, and associated, CDS, surfactant ions, and [ST] and [CDT] are the total surfactant and cyclodextrin concentrations in the solutions. Figure 4 shows some examples of the dependence of obs on the total cyclodextrin concentration. Eq. 3 was fitted to the experimental data using a nonlinear least-square algorithm. Solid lines in Figure 4 show the result of the fittings. One can see that the agreement between the experimental and theoretical data is good. The values of the binding equilibrium constants, K1, obtained from the fittings are summarized in Table 2. Experiments with different surfactant concentrations were carried out and the results showed that [ST] does not influence the estimated K1 value. The method was also checked by determining K1 for the 1:1 inclusion complex formed between dodecyltrimethylammonium bromide, DTAB, and -cyclodextrin at 298 K. The K1 value obtained for this inclusion complex was 1.9104 M-1, in good agreement with literature data [10]. K1 values summarized in Table 2 are the average of at least four different experiments. 19 0 1 2 3 4 5 6 7 0 1 2 3 4 5 a) 103[-CDT] 104obs(-1 m2 mol-1) [Phenoxy12T]=2.0710-3 M 0 1 2 3 4 5 0 1 2 3 4 5 6b) 103[-CDT] 104obs(-1 m2 mol-1) [Phenoxy12T]=1.7410-3 M 0 2 4 6 8 10 12 14 0 1 2 3 4 103[-CDT] 104obs(-1 m2 mol-1) c) [Phenoxy12]=1.7010-3 M 0.0 0.5 1.0 1.5 2.0 2.5 3.0 0 1 2 3 4 5 6 103[-CDT] 104obs(-1 m2 mol-1) [Naphthoxy12T]=4.8510-4 M d) Fig. 4.-Dependence of obs on the total cyclodextrin concentration for the surfactants investigated at 303 K. Solid lines show the fitting of the experimental data by using eq. 3. Table 2.- Values of equilibrium binding constant, K1, estimated from the fittings of the observed molar conductance variations of the aqueous surfactant solutions upon increasing the total CD concentration, by using eq. 3. T= 303 K. Surfactant:CD K1 (M-1) DTEAB:-CD (2.40.5) 104 DTEAB:-CD (1.60.4)104 DTEAB:-CD (3.80.3)102 Phenoxy12:-CD (2.20.5)104 Phenoxy12:-CD (1.30.2)104 Phenoxy12:-CD (6.90.5)102 Naphthoxy12--CD (2.90.7)104 Naphthoxy12--CD (8.20.8)103 Naphthoxy12--CD (4.02.2)102 It is worth noting that the estimation of the equilibrium binding constants for the inclusion complexes Naphthoxy12:CDs was carried out in the presence of [surfactant]510-4 20 M, due to the low cmc of this surfactant (cmc=6.710-4 M). As a consequence, the precision of the estimated K1 values is poor, particularly for the Naphthoxy12:-CD system. The formation equilibrium constants of the inclusion complexes can also be estimated by using 1H NMR measurements. The effect of micellization in the chemical shifts of the surfactant resonances has been avoided using a fixed surfactant concentration below the cmc. The concentration of cyclodextrin was varied to obtain different molar ratios [CD]/[Surfactant]. Representative results of the 1H NMR spectra for the CD/surfactant mixtures are shown in Figure 5 for the system Phenoxy12:-CD. The 1H NMR spectra of α-, Cx=C2, C3, C4, C5, and C6 Fig. 5.- 1H NMR spectra of Phenoxy12:-CD solutions in D2O at 303 K, with [Phenoxy12]=2.05x10-3 M. a) [β-CD]=0 M; b) [β-CD]= 4.0x10-4 M; c) [β-CD]=2.0x10-3 M; d) [β-CD]=4 x10-3 M; More β-CD concentrations were investigated but the spectra are not included in the figure for the sake of clarity. T=303 K. 8 2' 3' 1 2 3 7 4 6 5 1 a) b) c) d) Cx C1 21 β-, and -CD are shown in Figure S5 (Supplementary Material). Assuming that the condition of fast exchange on the NMR time scale applies, the measured frequency is a weighted average of the frequencies in each site, and the chemical shift can be used to measure the extent in which the equilibrium is displaced [28]. The observed chemical shift, for a 1:1 inclusion complex is [10]: δobs = XS δS +XSCDδSCD =(1−XSCD)δS+XSCDδSCD (4) where XS=[S]/[ST] and XSCD=[SCD]/[ST]. In this case: ∆𝛿𝑜𝑏𝑠 = 𝛿𝑜𝑏𝑠 −𝛿𝑆= 𝑋𝑆𝐶𝐷(𝛿𝑆𝐶𝐷 −𝛿𝑆)= 𝑋𝑆𝐶𝐷∆𝛿𝑜 (5) For a 1:1 inclusion complex, one can write [10]: K1=[SCD] [S][CD]=[SC] ([ST]−[SCD])([CDT]−[SCD]) = =XSCD (1−XSCD)([CDT]−XSCD[ST]) (6) After some algebraic manipulation and simplification [10]: Δ𝛿𝑜𝑏𝑠 =Δδ0 2𝐾1[𝑆𝑇](𝐾1([𝑆𝑇]+[𝐶𝐷𝑇])+1− −((𝐾1([𝑆𝑇]+[𝐶𝐷𝑇])+1)2−4𝐾1 2[𝑆𝑇][𝐶𝐷𝑇])1/2) (7) Eq. 7 was fitted to the experimental data using a non-linear least-square algorithm. Figure 6 shows two examples of the dependence of obs on the total cyclodextrin concentration for some nuclei. The experiments were done at least twice for each surfactant-cyclodextrin system. Since these measurements were done in order to check the reliability of the equilibrium constants values listed in Table 2, only β-cyclodextrin was used. The equilibrium constant for the Naphthoxy12:CD complexes could not be calculated from 1H NMR experiments because of the large errors due to the low surfactant concentration present in the deuterated solutions. The values of the binding equilibrium constants, K1, obtained from NMR measurements are summarized in Table 3. One can see that the K1 values listed in Tables 2 and 3 are in good agreement. 22 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 -0.04 -0.03 -0.02 -0.01 0.00 obs/ppm H(5) 103x[-CDT]/M a) DTEAB:-CD 0 1 2 3 4 5 0.000 0.010 0.020 0.030 0.040 0.050 0.060 obs/ppm H(3) 103x[-CDT] b) Phenoxy12--CD Fig. 6.- Dependence of the chemical shift obs=obs-S on total -cyclodextrin concentration for selected protons of the surfactants. Solid lines are the best fit to eq. 14. Table 3.- Values of equilibrium binding constant, K1, estimated from the fittings of the observed chemical shift variations of surfactant protons upon increasing the total CD concentration, by using eq. 14. T= 303 K. Surfactant:CD [SurfactantT](M) K1 (M-1) DTEAB:β-CD 1.9510-3 (1.60.5) 104 Phenoxy12:β-CD 2.0510-3 (1.40.4)104 N Br 6 5 4 3 21 23 The experimental results have shown that inclusion complexes are formed between the CDs and the surfactants investigated. At this point, it is interesting to consider how they are formed. The surfactants are quaternary ammonium derivatives, with identical cationic head groups and a hydrocarbon tail with twelve carbon atoms. The volume of the -N(C2H5)3+ head group is large and, besides, it is charged and to push it through the relatively non polar cavity of the CDs will be energetically expensive. As a consequence, it would be expected that the intercalation of the surfactant tail into the host cavity occurred as is shown in Figure 7. The Fig. 7.-Formation of the inclusion complexes work of Lyon et al. [29] supported this hypothesis. These authors investigated the formation of inclusion complexes between the bolaform surfactants [(CH3)3N(CH2)nN(CH3)3]Br2 (n=8, 10, 12) and ((CH3)2EtN(CH2)10NEt(CH3)2)Br2, and α-CD. They found that the replacement of one methyl by an ethyl in each of the end groups on the ((CH3)3N(CH2)10N(CH3)3)2+ surfactant results in a strong decrease in the equilibrium binding constant. Replacements of two or all of the methyls by ethyls prevent the formation of the inclusion complexes even after prolonged heating. The geometries of the aromatic substituents were optimized with a RHF wavefunction using 6-81g(d) basis set with the Gaussian 09 suit of programs [30] and their volumes were calculated. As is shown in Figure 8, the bulk of the two aromatic substituents permits the insertion of the hydrophobic tail into the host cavity of either -, - or -CD to form the inclusion complexes. Figure 7 shows that two possible inclusion complexes could be formed due to the truncated cone shape of the CD molecule. Only in the case of the α-CD:Phenoxy12 24 system, the ROESY spectrum suggests that the surfactant is preferentially oriented with the aromatic moiety located close to the narrower rim of the cyclodextrin. The experimental data would give information about the average equilibrium binding constant. Fig. 8.- Some structural information about the host molecules and the two aromatic substituents. The driving forces leading to the formation of CD:Surfactant inclusion complexes include electrostatic interactions, van der Waals interactions, hydrophobic interactions, hydrogen bonding, release of conformational strain of the CD, exclusion of cavity-bound high-energy water from the CD cavity and charge-transfer interactions [31]. Tables 2 and 3 show that the equilibrium binding constants follow the trend K1(α-CD)>K1(β-CD)>>K1(- CD). These observations can be explained by considering the volume of the cyclodextrin cavity (α-CD(V=174 Å3 [32]), β-CD (V=270 Å3 [32]) and -CD(V=472 Å3 [32]), and taking into account that the smaller the cavity is, the stronger the surfactant-CD interactions will be [10, 33-36]. Due to enthalpy-entropy compensation, release of conformational strain and exclusion of cavity-bound high-energy water do not usually play an important role in the complex formation. Van der Waals interactions and hydrophobic interactions constitute the major driving forces for cyclodextrin complexation, together with electrostatic interactions and hydrogen bonding. This is in agreement with the no substantial effects of the substituents 25 on K1. One can see that the incorporation of a phenoxy group, C6H5-O-, at the end of the dodecyl chain does not significantly affect the binding of the surfactant molecules to the three CDs investigated. On the other hand, the presence of a naphthoxy group, C10H7-O-, makes the association of the surfactant to the α-CD somewhat stronger, whereas the association to the βCD is made a little weaker. The binding of Naphthoxy12 to -CD also seems to be weaker, although the large experimental errors do not permit to reach any conclusion. This makes clear that the hydrocarbon chain length is the key structural surfactant feature determining the stability of the inclusion complexes investigated, which can be taken as evidence of the importance of the hydrophobic interactions contribution to the binding [10, 31]. A similar result was found by other authors in the study of inclusion complexes formed between anionic, cationic and non-ionic surfactant homologs and cyclodextrins [10, 37, 38]. The presence of the aromatic rings at the end of the hydrophobic tail does not substantially affect K1, however, it does influence the location of the host and the guest in the inclusion complexes. The observed NOE interactions between the aromatic protons and the CD protons indicate that the aromatic rings are partially inserted within the host cavity, with the macrocycle preferentially located at the end of the hydrocarbon tail of the surfactant, in contrast with the structure of the inclusion complexes formed with DTEAB. It was also found than in the case of the phenoxy substituent, the pattern of intermolecular NOEs observed suggests a specific orientation of the surfactant in the inclusion complex formed with α-CD, with the aromatic moiety located close to the narrower rim of the cyclodextrin. 4. Conclusions In the study of the complexation between cyclodextrins and surfactants the influence of several factors in the stability of the host-guest complexes has been investigated. The effects on the formation of the inclusion complexes of changing the size of the host cavity [10, 33-36, 39], the hydrophobic chain length of the surfactant [10, 37, 38], the nature of the 32 tension, calorimetry, and molecular dynamics of cyclodextrins with a non-ionic surfactant. J. Phys. Chem. B 111 (2008) 4383-4392. [36] S. M. Ghoreishi, M. Behpour, M. Golestaneh, Study of the inclusion complex formation between a cationic surfactant, two cyclodextrins and a drug. J. Incl. Phenom. Macrocycl. Chem. 62 (2008) 279-284. [37] M. Benko, R. Tabajdi, Z. Kiraly 976Thermodynamics of formation of b-cyclodextrin inclusion complexes with four series of surfactant homologs. J. Therm. Anal. Calorim. 112 (2013) 969-976. [38] A. Petek, M. Krajnc, A. Petek Study of host-guest interaction between -cyclodextrin and alkyltrimethylammonium bromides in water. J. Incl. Phenom. Mavrocycl. Chem. 86 (2016) 221-229. [39] S. K. Mehta, K. K. Bhasin, S. Dham, M. L. Singla, Micellar behavior of aqueous of dodecyldimethylethylammonium bromide, dodecyltrimethylammonium chloride and tetradecyltrimethylammonium chloride in the presence of -, -, HP--, and - cyclodextrins. J. Colloid Interface Sci. 321 (2008) 442-451. [40] X. Du, X. Chen, W. Lu, J. Hou, Spectrsocopic study on binding behaviors of different structural nonionic surfactants to cyclodextrins. J. Colloid Interface Sci. 274 (200) 645-651. [41] D. Ondo, Calorimetric study on the interaction of didecyldimethylammonium and decyltrimethylammonium cations with native cyclodextrins in water. J. Chem. Thermodynamics 97 (2016) 235-243. [42] E. Alami, S. A. Alami, J. Eastoe, I. Grillo, R. K. Heenan, Interactions between a nonionic gemini surfactant and cyclodextrins investigated by small-angle neutron scattering. J. Colloid Interface Sci. 255 (2002) 346-356. 33 [43] Ch. Zhou, D. Wang, M. Cao, Y. Chen. Z. Liu, Ch. Wu, H. Xu, S. Wang, Y. Wang, Self-aggregation, antibacterial activity, and mildness of cyclodextrin/cationic trimeric surfactant complexes. ACS Appl. Mater. Interfaces 8 (2016) 30811-30823. [44] M. Quiroga, M. Parajó, P. Rodríguez-Dafonte, L. garcía-Río, Kinetic study of [2]pseudorotaxane formation with an asymetrical tread. Langmuir 32 (2016) 63676375. [45] V. I. Martín, B. Sarrión, M. López-López, P. López-Cornejo, I. Robina, M. L. Moyá, Reversibility of the interactions between a novel surfactant derived from lysine and biomolecules. Colloids Surf. B 135 (2015) 346-356. [46] Gonzalez-Perez A, Dias RS, Nylander T, Lindman B. Cyclodextrin–surfactant complex: a new route in DNA decompaction. Biomacromolecules 9 (2008) 772-775. [47] Carlstedt J, Gonzalez-Perez A, Alatorre-Meda M, Dias RS, Lindman B. Release of DNA from surfactant complexes induced by 2-hydroxypropyl-beta-cyclodextrin. Int. J. Biol. Macromol. 46 (2010) 153–8. [48] Gonzalez-Perez A, Carlstedt J, Dias RS, Lindman B. Cyclodextrins in DNA decompaction. Colloid Surf. B 76 (2010) 20–7. [49] Carlstedt J, Lundberg D, Dias RS, Lindman B. Condensation and decondensation of DNA by cationic surfactant, spermine, or cationic surfactant–cyclodextrinmixtures: macroscopic phase behavior, aggregate properties, and dissolution mechanisms. Langmuir 28 (2012) 7976–89. [50] A. J. Kirby, P. Camilleri, J. F. B. N Engberts, M. C. Feiters, R. J. M. Nolte, O. Söderman, M. Bergsma, P. C. Bell, M. L. Fielden, C. L. García Rodríguez, P. Gudat, A. Kremer, C. McGregor, C. Perrin, G. Ronsin, M. C. P. van Eijk, Gemini surfactants: New synthetic vectors for gene transfection. Angew. Chem. Int. ed. 42 (2003) 14481457. 34 Figure captions Figure 1.- Concentration dependence of 1H NMR spectrum of Phenoxy12, in D2O, on surfactant concentration. a) [Phenoxy12]=2.00x10-3 M; b) [Phenoxy12]= 0.010M. T=303 K. Figure 2.- ROESY spectra of D2O solutions containing [Phenoxy12]=2.0510-3 M and [CD]=2.0010-3 M at 303 K. a) α-CD; b)β-CD; c)-CD. Figure 3.-Job´s plots at 303 K. a)DTEAB:-CD; b)Phenoxy12:-CD; c)Naphthoxy12:β-CD. Figure 4.-Dependence of obs on the total cyclodextrin concentration for the surfactants investigated at 303 K. Solid lines show the fitting of the experimental data by using eq. 3. Figure 5.- 1H NMR spectra of Phenoxy12:-CD solutions in D2O at 303 K, with [Phenoxy12]=2.05x10-3 M. a) [β-CD]=0 M; b) [β-CD]= 4.0x10-4 M; c) [β-CD]=2.0x10-3 M; d) [β-CD]=4 x10-3 M; More β-CD concentrations were investigated but the spectra are not included in the figure for the sake of clarity. T=303 K. Figure 6.- Dependence of the chemical shift obs=obs-S on total -cyclodextrin concentration for selected protons of the surfactants. Solid lines are the best fit to eq. 14. Figure 7.-Formation of the inclusion complexes Figure 8.- Some structural information about the host molecules and the two aromatic substituents. 35 Tables Table 1.-Critical micellar concentration, cmc, micellar ionization degree, , and Gibbs energy of micellization, GoM, for the cationic surfactants studied in this work, at 303 K. Surfactant Cmc/mM  GoM/kJ mol-1 DTEABa 14.30.4 0.350.02 -34.31.8 Phenoxy12b 3.70.2 0.400.03 -38.81.7 Naphthoxy12b 0.6410.015 0.430.03 -45.01.9 aRef. 11; bThis work. Table 2.- Values of equilibrium binding constant, K1, estimated from the fittings of the observed molar conductance variations of the aqueous surfactant solutions upon increasing the total CD concentration, by using eq. 3. T= 303 K. Surfactant:CD K1 (M-1) DTEAB:-CD (2.40.5) 104 DTEAB:-CD (1.60.4)104 DTEAB:-CD (3.80.3)102 Phenoxy12:-CD (2.20.5)104 Phenoxy12:-CD (1.30.2)104 Phenoxy12:-CD (6.90.5)102 Naphthoxy12--CD (2.90.7)104 Naphthoxy12--CD (8.20.8)103 Naphthoxy12--CD (4.02.2)102 Table 3.- Values of equilibrium binding constant, K1, estimated from the fittings of the observed chemical shift variations of surfactant protons upon increasing the total CD concentration, by using eq. 14. T= 303 K. Surfactant:CD [SurfactantT](M) K1 (M-1) DTEAB:β-CD 1.9510-3 (1.60.5) 104 Phenoxy12:β-CD 2.0510-3 (1.40.4)104 36 SUPPLEMENTARY MATERIAL HOST-GUEST INTERACTIONS BETWEEN CYCLODEXTRINS AND SURFACTANTS WITH FUNCTIONAL GROUPS AT THE END OF THE HYDROPHOBIC TAIL Victoria Isabel Martín,a Francisco José Ostos, Manuel Angulo,b Antonio Márquez,a Pilar López-Cornejo,a Manuel López-López,c Ana Teresa Carmona,d and María Luisa Moyáa* aDepartment de Química Física, Universidad de Sevilla, C/ Profesor García González 1, 41012 Sevilla. Spain. Tlf. 34954557175 Fax: 34954557174 E-mail: [email protected] bServicio de RMN, Universidad de Sevilla, Apartado 1203, E-41071 Sevilla, Spain cDepartment of Chemical Engineering, Physical Chemistry and Material Science, Faculty of Experimental Sciences, Campus El Carmen, Avda. De las Fuerzas Armadas s/n, 21071 Huelva,.Spain dDepartment of Organic Chemistry, University of Seville, C/Profesor García González 1, 41012 Seville. Spain *Author to whom all correspondence should be directed. 37 Preparation of dodecyltriethylammonium bromide, DTEAB The DTEAB was prepared in a previous work (ref.11) following the method of Guo et al. (J. Polym. Sci. A 2009, 47, 434-449). Briefly, stoichiometric amounts of 1bromododecane and trietylamine were heated, under reflux, in acetone for 20 h at 75ºC. The crude product was recrystallized 5 times from acetone and washed with ether. The product was obtained as a white solid (26.7%). The purity of DTEAB was checked by NMR measurements and mass spectrometry. Experimental procedures for triethyl(1-phenoxydodecyl)ammonium bromide, Phenoxy12 and triethyl(2-naphthoxydodecyl)ammonium bromide, Naphthoxy12. General techniques. The characterization of the compound was performed by its spectral data. 1H and 13C-NMR spectra were obtained for solutions in D2O on a Bruker Avance III 500 MHz spectrometer (500.2 MHz for 1H) equipped with a 5 mm TCI cryoprobe operating at 303 K. All 1H NMR chemical shifts are referenced to the residual HDO signal set to 4.71 ppm; J values are given in Hz and δ in ppm. The NMR spectra for all compounds were performed in CITIUS (Research General Service for the University of Seville). The completion of the reactions were monitored by TLC (silica gel HF254 (Merck) hexane and DCM:MeOH=3:1) with detection by UV light and charring with Pancaldi. Elemental analysis of the surfactant was also carried out. 12-Bromo-1-phenoxydodecane (1) A solution of 1,12-dibromo-dodecane (1g, 3.05 mmoles) and sodium phenolate (0.71g, 6.12 mmol) in dry acetone (70 mL), was stirred under Ar and heated at 62 °C for 30 min and then concentrated to dryness at reduced pressure. The residue was dissolved in dichloromethane and extracted successively with water (4x15 mL). The organic phase was dried with Na2SO4 and concentrated to dryness under reduced pressure. The residue was purified using column chromatography with silica gel and hexane. Product 1 was obtained as a white amorphous solid (0.3 g, 29%). Triethyl(1-phenoxydodecyl)ammonium bromide (2), Phenoxy12 A solution of 12-bromo-1-phenoxydodecane (0.246g, 0.79 mmol) and triethylamine (2.5mL, 17.9 mmol) in acetonitrile (19 mL), was stirred under Ar and heated at 90 °C for 1 day. The progress of the reaction was controlled by TLC (DCM:MeOH=3:1). The reaction mixture was then concentrated to dryness at reduced pressure. Subsequently, 10 ml of hexane was added and the mixture was stirred for 15 min and then filtered under vacuum (this 38 procedure was repeated three times). Product 2 was obtained as a white amorphous solid (0.24 g, 75%). 1H RMN (500 MHz, D2O): (ppm)=7.45 (t, 2H, Ph), 7.15-7.07 (m, 3H, Ph), 4.16 (t, 2H, PhO-CH2-(CH2)11), 3.36-3.26 (m, 6H, N+(CH2)3(CH3)3), 3.21-3.13 (m, 2H, CH2N+(CH2)3(CH3)3), 1.88-1.79 (m, 2H, Ph-O-CH2-CH2), 1.76-1.66(m, 2H, CH2-CH2N+(CH2)3(CH3)3), 1.56-1.47 (m, 2H, Ph-O-(CH2)2-CH2), 1.47-1.34 (m, 14H, (CH2)7), 1.31 (t, 9H, 3J = 7 Hz, N+(CH2)3(CH3)3). 13C RMN (75.4 MHz, CDCl3): δ (ppm) = 157.1, 129.24, 127.65, 123.63, 118.7, 106.7 (Ph), 67.9 (Ph-O-CH2-(CH2)11), 56.5 (CH2-N+(CH2)3(CH3)3), 52.7 (N+(CH2)3(CH3)3), 29.6, 26.4, (Ph-O-CH2-(CH2)9), 21.3(CH2-CH2-N+(CH2)3(CH3)3, N+(CH2)3(CH3)3) 12-Bromo-1-naphthoxydodecane (3) A solution of 1,12-dibromo-dodecane (2g, 6.10 mmoles) and sodium naphtholate (2.03g, 12.19 mmol) in dry acetone (140 mL), was stirred under Ar and heated at 65 °C for 30 min and then concentrated to dryness at reduced pressure. The residue was purified using column chromatography with silica gel and cyclohexane. Product 1 was obtained as a white amorphous solid (2.11 g, 52.26%). Triethyl(2-naphthoxydodecyl)ammonium bromide (4), Naphthoxy12 A solution of 12-bromo-1-naphthoxydodecane (0.640g, 1.64 mmol) and triethylamine (2.73mL, 19.62 mmol) in acetonitrile (50 mL), was stirred under Ar and heated at 90 °C for 1 day. The progress of the reaction was controlled by TLC (DCM:MeOH=3:1). The reaction mixture was then concentrated to dryness at reduced pressure. Subsequently, 10 ml of cold cyclohexane was added and the mixture was stirred for 30 min and then filtered under vacuum (this procedure was repeated three times). Product 2 was obtained as a white amorphous solid (0.624 g, 97.43%). 1H NMR (500 MHz, D2O, 303 K): (ppm)=7.38 (m, 2H, Naph), 7.05 (m, 1H, Naph), 7.02 (m, 2H, Naph), 4.09 (t, 2H, Naph-O-CH2-, 6.5 Hz), 3.23 (q, 6H, N+(CH2CH3)3, 7.3 Hz), 3.10 (m, 2H, -CH2-N+(CH2CH3)3), 1.76 (m, 2H, Naph-O-CH2-CH2-), 1.63 (m, 2H, -CH2-CH2-N+-), 1.44 (m, 2H, Naph-O-(CH2)2-CH2-), 1.38-1.26 (m, 14H, -(CH2)7-(CH2)2-N+-), 1.23 (bt, 9H, - N+(CH2CH3)3). 13C NMR (75.4 MHz, CDCl3): δ= 134.72 (C4’, C5’and C8’), 129.40-127.72 (C4’, C5’and C8’), 126.79-126.41 (C6'), 123.58 (C7’), 119.12 (C1’ and C3’), 106.68 (C1’and C3’) 68.12 (C8), 57.74 (C3), 53.74 (C2), 29.59-29.25 (C7, C6 and C5), 26.60-26.19 (C6 and C5), 22.23(C4), 8.27 (C1). 39 0.0 0.5 1.0 1.5 2.0 0 20 40 60 80 100 120 cmc = 6.41x10-4 M = 0.43 103x[Naphthoxy12]/M /S cm-1 T=303 K b) Figure S1.- Dependence of the specific conductivity, /S cm-1, on surfactant concentration. a)Phenoxy12; b)Naphthoxy12. T=303 K. The solid lines correspond to the Carpena fittings. 103x[Phenoxy12]/M 02468 0 100 200 300 400 500 /S cm-1 cmc=3.7x10-3 M =0.40 T=303 K a) 40 -8 -7 -6 -5 -4 44 48 52 56 60 exc=1.64x10-6 mol m-2 Amin=99x10-20 m2 ln([Phenoxy12]/M) mN m-1 Figure S2.- Dependence of the surface tension, , on ln([Phenoxy12). T=303 K. 41 Figure S3.- Concentration dependence of 1H NMR spectrum of Naphthoxy12, in D2O, on surfactant concentration. a) [Naphthoxy12]=5.00x10-4 M; b) [Naphthoxy12]=1.00x10-3 M. T=303 K. b) a)