DOI: 10.1002/cphc.200800292
Micelles as Containers for Self-Assembled Nanodevices: A Fluorescent Sensor for Lipophilicity Giuseppe Chirico,[b] Maddalena Collini,[b] Laura D’Alfonso,[b] Franck Denat,[c] Yuri A. DiazFernandez,[a] Luca Pasotti,[a] Yoann Rousselin,[c] Nicolas Sok,[c] and Piersandro Pallavicini*[a] Potentiometric titrations, fluorescence versus pH titrations, dynamic light scattering and fluorescence polarization anisotropy studies demonstrate that inside the nanodimensioned Triton X-100 micelles, 1-pyrenecarboxylic acid, PCOO , forms an apical complex with the Zn2 + cation encircled by a lipophilic cyclen ligand and hugely increasing its fluorescence. The ability of the Zn2 + -cyclen-PCOO complex plus its micellar container to act as a fluorescent sensor to evaluate the lipophilicity of molecular spe-
cies is demonstrated on the fatty acid series CH3ACHTUNGRE(CH2)xCOOH ACHTUNGRE(x=0–16). At pH 7.4 a decrease in fluorescence is observed on the addition of fatty acids that is directly related to their chain length, that is, to their tendency to enter the micellar containers, where they dislocate PCOO from the Zn2 + centre. The independent determination of fatty acid pKa values in the presence of Triton X-100 micelles confirms that our fluorescent micellar device is capable of sensing their lipophilicity.
1. Introduction The possibility offered by using water as the solvent and confining separate hydrophobic species in the small volume of the same micelle has been exploited since the 1970s. In particular, hydrophobic fluorophores and quenchers have been co-micellized and the observation of either steady-state or dynamic fluorescence quenching due to intramicellar interactions[1, 2] has been used to calculate the aggregation number of the micelle. Moreover, kinetic parameters, such as the rate constants of processes involving the fluorophore/quencher systems inside a micelle, have also been elucidated by means of fluorescence quenching.[3] The dimensions and the shape of micelles have been studied for more than 30 years, and it is well established that in the case of many traditional surfactants, a micelle is an approximately spherical object with a diameter shorter than 10 nm[4] and, accordingly, a volume smaller than 500 nm3. However, the use on purpose of micelles as nanocontainers for building self-assembled multicomponent molecular devices is still a largely unexploited area, even if they offer some unique advantages: 1) when two or more hydrophobic molecules are confined in the same micelle, they “feel each other” as if they were more concentrated, that is, inside micelles the local concentration of the contained species is huge (even if their bulk concentration is very low); 2) mobility is allowed inside micelles, where the viscosity is comparable to that of an organic solvent droplet;[5] and 3) inside micelles solvation is dramatically lower than in bulk water.[6] The most useful consequence is that dynamic interactions are promoted among molecules included in the same micelle, even if they are not linked by a covalent bond, and the micelle and its molecular content may behave as a self-assembled, multicomponent nanosized device. As in classical molecular multicomponent devices, a new overall function may be developed that is different from the mere sum of the properties brought by its molecular components. ChemPhysChem 2008, 9, 1729 – 1737
With this approach, we and other authors have used intramicellar energy and electron transfer between a fluorophore and the M2 + complex of a lipophilized ligand to obtain fluorescent micellar sensors for cations such as Cu2 + ,[7] Ni2 + ,[7c] and Hg2 + .[8] A sensor for the inositol triphosphate anion,[9] an off-on-off window-shaped fluorescent sensor for pH,[10] an AND molecular logic gate based on H + and Na + inputs[11] and an indole–quinolizine-based sensor for Cu2 + [12] have also been obtained by using micelles as containers in which interactions among micellized species are promoted. In addition, we have also investigated the role of the micellar shape and of the surfactant type in the response efficiency of self-assembled pyrene-based fluorescent sensors for the Cu2 + cation.[13] A further step forward may be made with this kind of system if the micelle is not used as a mere container but participates as a further molecular component, and its peculiar properties are exploited. In particular, it should be remembered that molecular species dissolved in a micellar solution partition between bulk water and the micelles, with the ratio of included versus bulk molecules being proportional to the lipophilicity of the molecules.[6, 14] Accordingly, a micelle is in-
[a] Dr. Y. A. Diaz-Fernandez, Dr. L. Pasotti, Prof. P. Pallavicini Dipartimento di Chimica Generale Universit3 di Pavia, viale Taramelli 12, 27100 Pavia (Italy) Fax: (+ 39) 0382528544 E-mail:
[email protected] [b] Prof. G. Chirico, Prof. M. Collini, Prof. L. D’Alfonso Dipartimento di Fisica G. Occhialini Universit3 Milano Bicocca, Piazza della Scienza 3, 20126 Milano (Italy) [c] Prof. F. Denat, Dr. Y. Rousselin, Dr. N. Sok Institut de Chimie MolAculaire ICMUB UniversitA de Bourgogne, 9 avenue Alain Savary, 21078 Dijon (France) Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/cphc.200800292.
D 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1729
P. Pallavicini et al. trinsically able to discriminate molecules on the basis of their lipophilicity. The ability of a molecular drug to dissolve in physiological solutions and to cross lipid membranes is of obvious importance in drug delivery and pharmacokinetics, and is a function of its lipophilicity.[15] Drug design takes this into account, and evaluates a molecule’s lipophilicity by measuring its partition coefficient between water and n-octanol at physiological pH (7.4) with a variety of well-established techniques (e.g. shakeflask methods or HPLC).[16] However, the recent bursting development of liposomal drug delivery may take advantage of any new method for lipophilicity evaluation, in particular if the lipophilicity of a molecule is in direct relation to its ability to penetrate into the liposome wall. On the basis of these considerations, we design a micellar sensing system capable of reporting on the lipophilicity of a molecule by means of fluorescence, and we check its properties with a simple but representative series of molecules, that is, the fatty acids n-CH3ACHTUNGRE(CH2)xCOOH with x ranging from 0 to 16. For this goal, we exploit both the promotion of coordinative interactions inside micelles and the ability of micelles to discriminate the penetration of a molecule from water to their core on the basis of its lipophilicity.
was added. All the solutions in this work are 5 K 102 m in NaNO3 as ionic strength buffer or, when necessary to buffer pH solutions, they are 5 K 102 m in 4-(2-hydroxyethyl)piperazine-1ethanesulfonic acid (HEPES). Under these concentration conditions, the average number of PCOOH molecules per micelle is 0.011 and 1.13 % of the micellar population contains one PCOOH molecule, according to a Poisson-type distribution.[17] Even at such a low concentration the strongly fluorescent PCOOH gives an intense emission spectrum (lexc = 340 nm; lem,max = 395 nm), while the inclusion of two molecules per micelle and the consequent excimer emission is ruled out. Coupled pH-metric and fluorimetric titrations disclose a sigmoidal response of fluorescence intensity (If) versus pH, as can be seen in Figure 1 A (white triangles), where the If % at the maximum emission wavelength of PCOOH (395 nm) is plotted
2. Results and Discussion 2.1. The Micellar Sensing System The Separate Components and their Behaviour in Water and in Micellar Solutions We used Triton X-100 micelles as containers and 1-pyrenecarboxylic acid (PCOOH) plus the Zn2 + complex of the N-dodecylN’,N’’,N’’’-trimethyltetraazacyclododecane ligand (C12Me3cyclen) as the molecular components of the fluorescent micellar nanosensor (see Scheme 1 for structures). PCOOH is hydrophobic
Scheme 1. Structures of compounds and Triton X-100 micelle parameters.
Figure 1. A) Distribution diagram (c) of the species containing PCOOH/ PCOO , expressed as % concentration of each species (left vertical axis) with respect to total PCOOH, calculated at 8 K 107 m PCOOH and 1.7 K 105 m [(C12Me3cyclen)ZnACHTUNGRE(H2O)]2 + in Triton X-100 micelles. The species pertinent to each curve are indicated on the diagram. ( ! ): If % at 395 nm (right vertical axis) versus pH for PCOOH in Triton X-100 micelles. (*): If % at 395 nm (right vertical axis) versus pH for 8 K 107 m PCOOH plus 1.7 K 105 m [(C12Me3cyclen)ZnACHTUNGRE(H2O)]2 + . B) If (390 nm) for PCOO at pH 7.4 as a function of [(C12Me3cyclen)ZnACHTUNGRE(H2O)]2 + molar concentration in water (^) and in Triton X-100 micelles (*). Inset: the series of emission spectra obtained in Triton X-100 by addition of the Zn complex.
and not soluble in water whereas, due to its charge, its carboxylate PCOO is at least sparingly soluble in water (we verified its solubility up to 105 mol L1). On the other hand, both PCOOH and PCOO are well-soluble in a solution of Triton X-100, thanks to inclusion in micelles. We used a concentration of Triton X-100 in water of 0.01 m, which corresponds to a micelle concentration of 7 K 105 m,[7d, 8, 10] and 8 K 107 m PCOOH
against pH. The sigmoidal behaviour is obtained because PCOOH is strongly fluorescent, while in PCOO If is reduced to 50 %. A pKa value of 5.22 is calculated by fitting If versus pH data. The new C12Me3cyclen molecule has been chosen among many possible lipophilized tetraaza macrocyclic ligands. In its Zn2 + complex (CF3SO3 counteranion) the zinc cation is coordi-
1730
www.chemphyschem.org
D 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPhysChem 2008, 9, 1729 – 1737
Micelles as Containers for Nanodevices nated to the macrocyclic ligand with enhanced kinetic inertness, but the ring size of 12aneN4 is too small to encircle the Zn2 + cation and to impart to it a square-planar coordination. The Zn2 + cation remains out of the plane of the four nitrogen atoms,[18] an effect enhanced by tertiarization of the four amines.[19] The expected coordination number is 5, with a square-pyramidal geometry completed by an apical water molecule. The complex has thus to be formulated as [(C12Me3cyclen)ZnACHTUNGRE(H2O)]2 + . Its usefulness for our purposes is that the water molecule may be replaced by more effective ligands[18, 19] or may deprotonate on increasing the pH.[19] We determine[20] by potentiometry the pKa for the equilibrium represented in Equation (1)
½ðC12 Me3 cyclenÞZnðH2 OÞ2þ Ð ½ðC12 Me3 cyclenÞZnðOHÞþ þ Hþ
ð1Þ
obtaining 7.53ACHTUNGRE( 0.01) in water[21] and 7.66ACHTUNGRE( 0.01) in Triton X-100 micelles. To check if PCOO and [(C12Me3cyclen)Zn]2 + interact in water, we prepared a 8 K 107 m solution of PCOOH at pH 7.4 (5 K 102 m HEPES) where PCOOH is 100 % in its PCOO form (pH 7.4 was chosen as the working value throughout, because it is common practice in pharmacology to evaluate a molecule’s lipophilicity by means of its water–octanol distribution coefficient at this pH). This solution was titrated with [(C12Me3cyclen)ZnACHTUNGRE(H2O)]ACHTUNGRE(CF3SO3)2 (103 mol L1 aqueous solution). No change in the emission spectrum of PCOO was observed up to a 30-fold excess of the Zn2 + complex (Figure 1 B, white diamonds). In water the affinity of a tetracoordinated Zn2 + cation for a carboxylate fragment is low, due to the competition of water solvation, so that at low concentrations of PCOO and of the Zn2 + complex no interaction is observed, that is, the equilibrium represented in Equation (2)
½ðC12 Me3 cyclenÞZnðH2 OÞ2þ þ PCOO Ð ½ðC12 Me3 cyclenÞZnðPCOOÞþ þ H2 O
ð2Þ
is completely shifted to the left. By repeating the same titration in Triton X-100 micelles (HEPES, pH 7.4, PCOOH 8 K 107 m) even small additions of [(C12Me3cyclen)ZnACHTUNGRE(H2O)]2 + made the poorly fluorescent PCOO strongly increase its emission (Figure 1 B, inset and black circles). Upon addition of a 30-fold excess of the Zn2 + complex, If reaches a plateau with a value 600 % times higher than that of PCOO . By fitting If versus pH data log K = 5.5ACHTUNGRE( 0.1) is calculated for Equation (2) under the employed conditions, sharply indicating that inside micellar nanocontainers coordination of the carboxylate fragment to the zinc centre is dramatically promoted, boosting PCOO fluorescence. ChemPhysChem 2008, 9, 1729 – 1737
Fluorescence Polarization Anisotropy (FPA) and Dynamic Light Scattering (DLS): The Role of Micellar Inclusion in the Interactions between PCOO and [(C12Me3cyclen)ZnACHTUNGRE(H2O)]2 + It has already been shown that decreasing the librational freedom of PCOOH by decreasing the temperature of a water/glycerol solvent mixture[22] increases its emission intensity. Here, we hypothesize that a similar effect is obtained by coordination of the carboxylate fragment to the Zn2 + complex inside the micellar containers, which is promoted by the huge increase of local concentration of both included species. To verify this, we examine the shape of Triton X-100 micelles and the mobility of the included PCOO by measuring the DLS autocorrelation function (ACF) and the FPA decay. Triton X-100 forms micelles whose aggregation number (AN),[4a, 23] shape and dimensions[4a, 23, 24] have already been studied. We[7d, 13] and other authors[4a, 25] reported that an oblate ellipsoidal shape (see Scheme 1) with AN = 143 is appropriate. FPA measurements were carried out at pH 7.4 on micellar solutions containing PCOO only or [(C12Me3cyclen)ZnACHTUNGRE(PCOO)] + . The FPA decay is dominated by the pyrene wobbling relaxation q[26] (see also the Supporting Information), which is 1.0 0.2 ns for PCOO alone, whereas slowed wobbling dynamics are observed for PCOO coordinated to the Zn2 + complex with q = 2.1 0.3 ns, thus taking account of the expected decrease of librational freedom upon complexation. DLS experiments were carried out on empty Triton X-100 micelles and on micelles containing [(C12Me3cyclen)ZnACHTUNGRE(H2O)]2 + (2 K 104 m) and the experimental data were interpreted by assuming an oblate ellipsoid shape of the micelles with a major/ minor semi-axis ratio = 2.[4a, 13, 25] No appreciable difference is found between empty and complex-loaded micelles. Our estimate of the micelle minor (b) and major (a) semi-axes is 2.9 0.3 and 5.8 0.3 nm, respectively (Scheme 1), which corresponds to an average radial size (equal volume sphere pffiffiffiffiffiffiffi radius, 3 a2 b) of 4.6 0.3 nm and a volume of 410 nm3.[27] Noticeably, in the PCOO + [(C12Me3cyclen)ZnACHTUNGRE(H2O)]2 + titration at pH 7.4, in which we observe fluorescence boosting, the overall concentrations were 8.0 K 107 m for PCOO and 1.67 K 105 m for [(C12Me3cyclen)ZnACHTUNGRE(H2O)]2 + when If reached 90 % of the plateau value. According to volume calculations with DLS, inside any micelle containing both one PCOO and one Zn2 + complex the local concentration of the two species is increased to 3.4 K 103 m. This value is obtained by considering one molecule as 1/6.022 K 1023 moles and dividing this number by the calculated micellar volume transformed into litres. We believe that the huge local concentration increase promotes the observed complexation of [(C12Me3cyclen)ZnACHTUNGRE(H2O)]2 + by PCOO , although a role is also played by the poor hydration inside the micellar containers. Interestingly, this strongly suggests that inclusion of lipophilized complexes and ligands in nanosized micellar containers may be a new and useful tool of general use in the well-established field of molecular recognition based on the formation of reversible coordinative bonds.[28]
D 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemphyschem.org
1731
P. Pallavicini et al. Determination of the pH Range of Existence of the [(C12Me3cyclen)ZnACHTUNGRE(PCOO)] + Complex in Micelles The complexation of PCOO to [(C12Me3cyclen)ZnACHTUNGRE(H2O)]2 + according to Equation (2) is of course influenced by pH. To determine the full pH range in which the [(C12Me3cyclen)ZnACHTUNGRE(PCOO)] + complex exists, we carried out coupled fluorimetric and pHmetric titrations of a micellar solution containing PCOOH (8.0K107 m) and [(C12Me3cyclen)ZnACHTUNGRE(H2O)]2 + (1.67 K 105 m). An increasing–plateau–decreasing trend is observed in fluorescence intensity on shifting from pH 2.5 to 11.5. Figures 2 A and
½ðC12 Me3 cyclenÞZnðPCOOÞþ þ OH Ð ½ðC12 Me3 cyclenÞZnðOHÞþ þ PCOO
ð3Þ
The emission intensity drops to that of micellized, uncoordinated PCOO (Figure 2 A and dashed spectrum in Figure 2 B at pH 11.3). The trend is sharply visualized by the If,395 versus pH profile of Figure 1 A (black circles), where 100 % If has been assigned to the emission of micellized PCOOH. Correspondence of If % at low (< 3) and high (> 11) pH with that of non-interacting, micellized PCOOH and PCOO was checked by preparing micellar solutions of PCOOH at pH 2.5 and 11.5 and adding a 30-fold excess of [(C12Me3cyclen)ZnACHTUNGRE(H2O)]2 + : in both cases no fluorescence variation was observed. Using the experimentally determined log K values for PCOOH deprotonation and for Equation (1), the log K value of Equation (2) can be determined by fitting the experimental If versus pH profile to obtain log K = 5.6ACHTUNGRE( 0.1), which is consistent with the value of 5.5 obtained by titration at pH 7.4. With these data, a distribution diagram is drawn (solid lines, Figure 1 A) that reports the percentage concentration of each species containing PCOOH or PCOO with respect to the overall PCOOH concentration. Superimposition of the experimental If versus pH values (black circles) to the diagram is perfectly consistent with the hypothesized series of equilibria and calculated log K values. What is important for our goals is that, as Figure 1 A clearly visualizes, the [(C12Me3cyclen)ZnACHTUNGRE(PCOO)] + complex exists as the prevalent species for 5.8 < pH < 7.8. 2.2. Sensing the Lipophilicity of Fatty Acids
Figure 2. A) Full series of spectra obtained with 8 K 107 m PCOOH plus 1.7 K 105 m [(C12Me3cyclen)ZnACHTUNGRE(H2O)]2 + in Triton X-100 on changing the pH. B) Spectra extracted from (A) and fully representative of the emission of PCOOH (dotted spectrum, pH 2.6), [(C12Me3cyclen)ZnACHTUNGRE(PCOO)] + (solid-line spectrum, pH 7.5) and PCOO (dashed spectrum, pH 11.3).
B display the obtained series of spectra. Starting from the featureless emission of PCOOH (pH 2.6, dotted spectrum in Figure 2 B, lmax,em = 395 nm) on increasing the pH the intensity increases, the position shifts to the blue and the spectra become sharper (Figure 2 A) due to deprotonation of PCOOH and coordination to Zn2 + according to Equation (2). For 6 < pH < 8, the [(C12Me3cyclen)ZnACHTUNGRE(PCOO)] + complex is the prevalent species and an almost superimposable series of spectra is recorded (Figure 2 A and solid-line spectrum in Figure 2 B at pH 7.5, lmax,em = 385 nm). Further increase of pH over 8 allows PCOO to be released inside the micelles due to the formation of [(C12Me3cyclen)Zn(OH)] + according to Equation (3)
1732
www.chemphyschem.org
As has been demonstrated in the previous section, in the 5.8–7.8 pH range the [(C12Me3cyclen)ZnACHTUNGRE(PCOO)] + complex exists inside micelles of Triton X-100 as the prevalent species. What we expect is that inside this pH interval the complex plus the micelle allowing its formation constitute a self-assembled nanodevice capable of signalling, through a decrease of fluorescence, the capability of a molecule to penetrate the micellar space and to compete with PCOO for coordination to the Zn2 + cation encircled by the macrocyclic ligand. If a series of molecules with the same coordinating function is examined, we also expect that the differences observed in the If variation would be proportional only to the tendency of the chosen molecule to penetrate inside the micelle, that is, to its lipophilicity. The concept is pictorially described in Scheme 2. To demonstrate this we investigate the fatty acid series CH3ACHTUNGRE(CH2)xCOOH (0 < x < 16) and work at pH 7.4 (HEPES buffer), where the [(C12Me3cyclen)ZnACHTUNGRE(PCOO)] + complex is fully formed and the acids are expected to be fully deprotonated, that is, in their carboxylate form. We titrated micellar solutions containing [(C12Me3cyclen)ZnACHTUNGRE(PCOO)] + at pH 7.4 with each acid of the chosen series, and recorded the fluorescence spectra after each addition. No fluorescence variations were observed for the acids with shorter chains (x = 0–3) up to a fivefold molar excess of added carboxylic acid with respect to the Zn2 + complex (Figure 3 A, white symbols). On stepping to acids featuring longer chains
D 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPhysChem 2008, 9, 1729 – 1737
Micelles as Containers for Nanodevices the equilibrium shown in Equation (4) CH3 ðCH2 Þx COO þ ½ðC12 Me3 cyclenÞZnðPCOOÞþ Ð ½ðC12 Me3 cyclenÞZnðCH3 ðCH2 Þx COOÞþ þ PCOO
Scheme 2. A) Low lipophilicity (RCOO remains in water); B) intermediate lipophilicity (RCOO is distributed between bulk water and micelles); C) high lipophilicity (RCOO is fully included in micelles).
Figure 3. A) If % at 390 nm for PCOO (8 K 107 m) coordinated to [(C12Me3cyclen)ZnACHTUNGRE(H2O)]2 + (1.7 K 105 m) as a function of the molar concentration of added CH3ACHTUNGRE(CH2)xCOOH acids. White series: * x = 0, ! x = 1, & x = 2, ^ x = 3; black series: * x = 4, ! x = 5, & x = 6, ^ x = 7, ~ x = 8, hexagons x = 9; grey series: * x = 10, !x = 11, & x = 12, ^ x = 13, ~ x = 14, hexagons x = 16. The dashed arrow indicates the concentration value at which If % values have been taken to draw Figure 3 B. B) If % values measured at 4 K 105 m concentration of CH3ACHTUNGRE(CH2)xCOOH as a function of x.
(x = 4–9, black symbols in Figure 3 A), a decrease of If was observed on addition of CH3ACHTUNGRE(CH2)xCOOH with the extent of fluorescence decrease being proportional to chain length, that is, to the molecule lipophilicity. For x > 9, almost superimposable If versus concentration of acid profiles are observed (grey symbols in Figure 3 A), with If reaching a plateau at a value close to that of uncomplexed micellized PCOO . The fluorescence decrease is due to the process described in Scheme 2, that is, to ChemPhysChem 2008, 9, 1729 – 1737
ð4Þ
which proceeds to the right only if the carboxylate anion of the chosen acid is capable of penetrating inside the micellar container. Figure 3 B sharply visualizes the dependence of If on the lipophilicity of the added acid at pH 7.4, as it reports the If value found at 4.00 K 105 m of added CH3ACHTUNGRE(CH2)xCOOH as a function of x. The found sigmoidal profile indicates that ranging from acetic acid (x = 0) to pentanoic acid (x = 3) the lipophilicity is negligible and the acid carboxylate remains quantitatively in the bulk water with no effect on fluorescence (Scheme 2 A). Increasing the length of the chain (x = 4–10) results in an increase of CH3ACHTUNGRE(CH2)xCOO lipophilicity that allows penetration in the micellar container and competition with PCOO . The fatty acids belonging to this series distribute between bulk water and micelles (Scheme 2 B), with the distribution shifting towards fully micellized carboxylates on increasing the chain length, which results in a more and more pronounced If decrease. The acids featuring x > 10 have approximately the same effect on If, and give the same percentage decrease. Considering these results it should be said that for x > 10, the carboxylates of fatty acids display the same effective lipophilicity at pH 7.4, insofar as an increase of the chain length does not result in a more significant influence on the complex included in the micelle. This is a consequence of the fact that a CH3ACHTUNGRE(CH2)10 or longer chain makes the fatty acid’s carboxylate group lipophilic enough to be fully distributed inside the micellar containers with respect to bulk water (Scheme 2 C). To further stress the fundamental role of the micellar container for this sensor, we also carried out a series of experiments in dioxane/water (8:2 v/v). In this organic-enriched solvent mixture, lipophilic species are well-soluble and monodispersed without any surfactant addition while solvation is poor due to the large excess of dioxane.[7c, d, 8] First, and differently from what was observed in Triton X-100 micelles (Figure 1 A), the If versus pH profiles for PCOOH alone and for a PCOOH plus [(C12Me3cyclen)ZnACHTUNGRE(H2O)]2 + mixture are almost superimposable (see Supporting Information, Figure S1), which indicates either that the interaction between PCOO and the Zn2 + complex is only poorly promoted or that fluorescence revival on coordination of PCOO to the Zn2 + centre is not intense in this solvent mixture (measured pKa for PCOOH = 7.65 0.02). A fluorimetric titration was carried out at pH 8.5 by addition of [(C12Me3cyclen)ZnACHTUNGRE(H2O)]2 + to PCOO under the same concentration conditions as for the experiment in Triton X-100 solution: in dioxane/water a linear increase of If on addition of the Zn complex was observed (see Supporting Information, Figure S2), with a 200 % increase in If on addition of a 30-fold excess of the Zn complex. This finding suggests that even if, in dioxane/water, coordination of PCOO to 2+ [(C12Me3cyclen)ZnACHTUNGRE(H2O)] increases (although to a lesser extent) the emission intensity of the fluorophore, the tendency
D 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemphyschem.org
1733
P. Pallavicini et al. of PCOO to coordinate the Zn2 + centre is only poorly promoted outside the micellar containers (a log K value of 3.2 was calculated for Equation (2) under the titration conditions). Nevertheless, we also ran competition experiments with fatty acids in dioxane/water (8:2 v/v). We titrated solutions of PCOO ACHTUNGRE(8K107 m) plus [(C12Me3cyclen)ZnACHTUNGRE(H2O)]2 + (1.66 K 105 m) at pH 8.5, by addition of CH3ACHTUNGRE(CH2)xCOOH acids (x = 0, 6, 10, 14). In all cases, a slight decrease of PCOO fluorescence was observed on addition of a fivefold excess of acid with respect to the Zn complex, but no significant difference in the extent of the decrease was found among the examined acids. The absence of the micellar container not only depresses the tendency to form the [(C12Me3cyclen)ZnACHTUNGRE(PCOO)] + complex, even if a poorly solvating medium is used, but also deprives the system of its ability to discriminate molecules on the basis of their lipophilicity. 2.3. Determination of Observed pKa of the Fatty Acid Series in Micellar Solution To confirm the hypothesis that the degree of inclusion of fatty acids inside Triton X-100 nanocontainers is the rationale behind the If % versus x profile displayed in Figure 3 B, we undertook the experimental determination of the pKa values of the examined fatty acids in the presence of Triton X-100 micelles. Conventional potentiometric titration experiments were carried out with this goal, according to a well-established procedure,[8] and the e.m.f. versus volume of added base data were fitted[20] with non-linear least-squares regression methods to obtain pKa values. The fatty acid concentration was 5 K 104 m, Triton X-100 was 0.01 m and 5 K 102 m NaNO3 was used as the ionic strength buffer. The pKa values found have low uncertainties (lower than 0.02 log units) and are listed in Table S2 of the Supporting Information. What is important for our goals is to understand the pKa versus x trend for the CH3ACHTUNGRE(CH2)xCOOH series, so we made it visually clear in Figure 4 A (grey hexagons). It can be seen that the profile sharply parallels the If % versus x profile of Figure 3 B. It is well established in surfactant studies that inclusion in micelles of HA or BH + acids modify their observed acidity, with higher pKa values for HA and lower for BH + , as a consequence of the less efficient solvation of A and BH + species in micelles with respect to water.[6, 8, 10] In our case, for x = 0–3, acid carboxylates are not included in Triton X-100 micelles and no influence is seen on pKa values, which is consistent with what is found in the literature for the same molecules in water solution.[28] For x = 4–10, the observed acidity decreases (and pKa increases) due to a distribution of RCOO that is more and more shifted towards micellar inclusion on increasing x. For x > 10, a plateau-like trend is found, with a slight decrease of pKa with x, as the fatty acid carboxylates are lipophilic enough to be fully distributed inside micelles. As a comparison, we re-determined the pKa values for most acids of the series also in dioxane/water (8:2 v/v). The large difference observed in all cases between these values and those obtained in water or water/Triton X-100 is due to the hugely decreased solvating ability of the dioxane-rich mixture (e.g. the
1734
www.chemphyschem.org
Figure 4. A) Grey hexagons: pKa of CH3ACHTUNGRE(CH2)xCOOH, experimentally determined by potentiometric titrations in water containing Triton X-100 micelles (surfactant = 0.01 m), as a function of x (left vertical axis); white hexagons: pKa of CH3ACHTUNGRE(CH2)xCOOH, experimentally determined by potentiometric titrations in dioxane/water (8:2 v/v), as a function of x (right vertical axis). B) pKa of CH3ACHTUNGRE(CH2)xCOOH (x = 7, 9, 11, 13) as a function of surfactant concentration (acid concentration is maintained at the constant value 5 K 104 m). The species pertinent to each series of points is indicated in the plot.
pKa of acetic acid is 4.72[29] in water and 7.35 in dioxane/water), but the observed pKa trend in dioxane/water is fully representative of the effect played on the COOH acidity by mere chain lengthening. The decrease of pKa on stepping from x = 0 to x = 4 could be attributed to the increase of the electron-pushing inductive effect exerted on COO by adding CH2 units to the chain. Further increase of the chain length adds methylene groups too far from the carboxylic function to exert any further electron-pushing effect, and a plateau is observed for x > 4. Accordingly, the huge changes in pKa values observed for x > 4 in Triton X-100/water are fully attributed to inclusion of the fatty acid carboxylates in micelles. The slight pKa decrease observed even for x > 10 in Triton X-100 micelles (Figure 4 A) has to be attributed to the positioning inside the micellar nanocontainer of the COOH function of the fully micellized CH3ACHTUNGRE(CH2)xCOO molecules. It has already been shown that placing an acidic group near the hydrophobic core of a spherical micelle,[30] or in the less-solvated portion of an oblate ellipsoid micelle,[7d] is favoured by a large hydrophobic backbone of the molecule containing the acidic function. In our case, although for x > 10 the CH3ACHTUNGRE(CH2)xCOOH molecules are distributed at 100 % inside the micellar container, increasing x over 10 brings the effect of positioning the COOH group deeper in the least-solvated portion of Triton X-100 micelles. This has been verified by measuring the observed pKa values of the acids with x = 7, 9, 11 and 13 at constant acid concentration and increasing Triton X-100 concentration. The numerical results are listed in Table S2 of the Supporting Infor-
D 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPhysChem 2008, 9, 1729 – 1737
Micelles as Containers for Nanodevices mation. If a higher pKa value is measured on increasing the surfactant concentration (and consequently micelle concentration), it means that a distribution between bulk and micelle holds for the chosen acid, as increasing the micelle concentration pushes the distribution towards micellar inclusion.[6, 14] Figure 4 B sharply demonstrates that CH3ACHTUNGRE(CH2)7COO is distributed between water and micelles, while the carboxylates with x = 9, 11 and 13 are completely micelle-included at any value of the examined concentration range (or for Triton X-100 0.02 m when x = 9). Thus, for x > 10 the observed increase of the pKa on increasing x can be fully attributed to a deeper positioning of CH3ACHTUNGRE(CH2)xCOO in the micellar core.
with pharmacologically interesting molecules containing a COOH group, such as penicillins and cephalosporins.
3. Conclusions
Syntheses: The ligand 1-dodecyl-4,7,10-trimethyl-1,4,7,10-tetraazacyclododecane (C12Me3cyclen) was prepared according to the reaction described in Scheme 3. Preparative details for each step are described in detail below.
We have demonstrated that the complex between PCOO and [(C12Me3cyclen)ZnACHTUNGRE(H2O)]2 + does not form in water at low concentration of the two components, but it forms with a high observed formation constant when the two molecules are contained inside Triton X-100 micelles, indicated by a strong revival of PCOO fluorescence. From the DLS and FPA results we are able to suggest that its formation is due to the dramatically increased local concentration inside micellar containers, which is connected to their nanosized dimensions, and that the fluorescence burst is due to the decreased librational freedom of the complexed PCOO anion in the micellar containers. Determination of the distribution diagram of the PCOOH/PCOOcontaining species allowed us to establish the pH interval in which the [(C12Me3cyclen)ZnACHTUNGRE(PCOO)] + complex exists. This information allowed us work at the physiological pH of 7.4, using the complex and its micellar container as a fluorescent sensor to evaluate the lipophilicity of a series of molecules bearing the same coordinating function. To check this we chose the fatty acid series, and observed that our micellar nanodevice effectively discriminates fatty acids on the basis of their lipophilicity. The response was in the form of a fluorescence decrease whose extent was connected to their chain length, a feature that regulates the fatty acid’s ability to fully penetrate the micellar core, or to distribute between bulk water and micelles. Independent determination of another property of fatty acids, their acidity constant, which is influenced by the same effect (degree of distribution between bulk water and micelles), fully supports the interpretation of the signal output of our sensor. Herein, we have demonstrated that micellar multicomponent self-assembled devices may take advantage not only of the sum of the properties of the molecular components gathered in the same micellar container, but also of the properties of the container itself. In this case, the property used is the ability of the micelle/water interface to select the entrance of a molecule on the basis of its lipophilicity. Application of this new type of sensor to molecules of medical and pharmacological interest can be easily envisaged, as the sensing process can be applied to any species bearing a coordinating group capable of interacting with a tetracoordinated Zn2 + cation. Investigations are currently in progress to use this kind of sensor ChemPhysChem 2008, 9, 1729 – 1737
Experimental Section Materials and Reagents: Triton X-100 (tert-octylphenoxypoly(oxyethylene glycol) with an average of 9–10 oxyethylene units) was purchased from Caledon (average molecular weight MW = 647). Carboxylic acids were purchased from Sigma–Aldrich and used as received. HEPES buffer was purchased from Fluka. Dry ZnACHTUNGRE(CF3SO3)2 was purchased from Fluka and kept in a desiccator. Water was distilled twice, using deionized water (prepared with an ion-exchange apparatus) as the starting material. PCOOH was purchased from Sigma–Aldrich. 1-Benzyl-1,4,7,10-tetraazacyclododecane was purchased from CheMatech.
Scheme 3. Synthetic route for ligand C12Me3cyclen.
1-Benzyl-4,7,10-trimethyl-1,4,7,10-tetraazacyclododecane (1), C18H32N4, MW 304.47: A solution of 37 % formaldehyde (18.9 mL, 0.68 mol) was added to a solution of 1-benzyl-1,4,7,10-tetraazacyclododecane (5 g, 19 mmol) in formic acid (21 mL). Then water (5 mL) was added and the resulting mixture was heated at reflux overnight. After removal of the solvent, the residual oil was dissolved in 13 m NaOH solution (43 mL, pH > 12). After extraction with chloroform (2 K 150 mL), the organic phase was dried over MgSO4 and the solvent was evaporated. The residual oil was taken up in pentane (150 mL) and filtered on celite. After evaporation of the solvent, 1 was obtained as a colourless oil (3.80 g, yield 66 %). 1 H NMR (300 MHz, CHCl3, 300 K): d = 2.12 (s, 6 H), 2.21 (s, 3 H), 2.47 (m, 12 H), 2.59 (t, 4 H, J = 5.1 Hz), 3.47 (s, 2 H), 7.21 ppm (m, 5 H); 13 1 C{ H} NMR (75 MHz, CHCl3, 300 K): d = (CH3) 44.2, 45.0, (CH2-a) 52.7, 55.6, 55.7, 56.2, (CH2-Ph) 60.6, (CH) 126.7, 128.0, 129.0, (-C-) 139.8 ppm; GC–MS: m/z = 304 [M + ]. 1,4,7-Trimethyl-1,4,7,10-tetraazacyclododecane (2), C11H26N4, MW 214.35: 10 % Pd/C (0.55 g, 5 mmol, 0.04 equiv) was added to a solution of 1 (3.80 g, 12.5 mmol) in MeOH/THF (1:1, 250 mL) and the
D 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemphyschem.org
1735
P. Pallavicini et al. reaction mixture was stirred in a hydrogen atmosphere at room temperature. After consumption of 280 mL of hydrogen (12.5 mmol, 1 equiv), the solution was filtered on celite. The solvent was removed and 2 was obtained as a colourless oil (2.80 g, yield 99 %). 1H NMR (300 MHz, CHCl3, 300 K): d = 2.19 (s, 3 H), 2.22 (s, 6 H), 2.34 (s, 8 H), 2.40 (t, 4 H, J = 4.1 Hz), 2.57 ppm (t, 4 H, J = 4.6 Hz); 13C{1H} NMR (75 MHz, CHCl3, 300 K): d = (CH3) 43.5, 44.7, (CH2-a) 45.8, 53.6, 55.0, 55.2 ppm; GC–MS: m/z = 215 [M + ] 1-Dodecyl-4,7,10-trimethyl-1,4,7,10-tetraazacyclododecane (C12Me3cyclen), C23H50N4, MW 382.67: Bromododecane (3.33 g, 13 mmol, 1 equiv) was slowly added to a solution of 2 (2.80 g, 13 mmol) and K2CO3 (33 mmol, 4.50 g, 2.5 equiv) in MeCN (35 mL). The resulting mixture was heated at reflux overnight. After filtration, the solvent was removed and C12Me3cyclen was obtained quantitatively as a yellow oil (4.97 g). It was purified by aluminium oxide chromatography (solvent: CHCl3). The compound was obtained as a colourless oil (3.48 g, yield 70 %). 1H NMR (300 MHz, CHCl3, 300 K): d = 1.22 (t, 3 H, J = 6.7 Hz), 1.25 (s, 20 H), 1.42 (m, 2 H), 2.21 (m, 8 H), 2.35 (m, 3 H), 2.50 (m, 10 H), 2.58 ppm (m, 4 H); 13C{1H} NMR (75 MHz, CHCl3, 300 K): d = (CH3-ACHTUNGRE(CH2)11-) 14.1, ((CH2)11) 22.6, 26.9, 27.6, 29.3, 29.6, 31.9, (CH3) 44.3 (*2), 44.6, (CH2-a) 52.8, 53.4, 55.7, 56.0, 56.7 ppm; GC–MS: m/z = 381 [M + ] Preparation of the [Zn(C12Me3cyclen)]2 + Stock Solution: C12Me3cyclen (150 mg, 0.392 mmol) was dissolved in tert-butanol (10 mL) in a 25 mL flask, to which ZnACHTUNGRE(CF3SO3)2 (142 mg, 0.392 mmol) was added. The colourless solution was refluxed for 3 h, after which the product was obtained as a waxy white solid after solvent removal with a rotary evaporator. The complex was re-dissolved in pure tert-butanol (12.0 mL) to obtain a 3.26 K 102 m solution that was used as such for preparing diluted solutions in water or tert-butanol. ESI-MS: m/z: 446, 448, 450 [M2 + /2]. Fluorimetric Titrations: Addition of [Zn(C12Me3cyclen)]2 + to PCOO : Titrating solution: A solution (307 mL, 3.26 K 102 m) of the Zn complex in tert-butanol was placed in a 10.0 mL flask, and water was added to reach the flask volume and obtain a 1.00 K 103 m solution of [Zn(C12Me3cyclen)]2 + . In a 100 mL Erlenmeyer flask, an aqueous solution (30 mL) containing HEPES (0.05 m), Triton X-100 (0.01 m), NaNO3 (0.05 m) and PCOOH (8 K 107 m) was adjusted to pH 7.4 with microadditions of 0.1 m HNO3. This solution was thermostatted at 25 8C and kept under a N2 atmosphere, and the Zn titrating solution was added in 50 mL portions (14 additions in total). After each addition a fluorescence spectrum was recorded on a 3 mL sample of the solution, which was re-added to the flask after measurement. The same conditions, except for the absence of Triton X-100, were maintained for the titration carried out in water. Coupled Fluorimetric and pH-metric Titration: An aqueous solution (30 mL) was prepared containing 0.05 m NaNO3, 0.01 m Triton X-100, 1.66 K 105 m [Zn(C12Me3cyclen)]2 + (from 500 mL of concentrated tert-butanol solution) and 8 K 107 m PCOOH. The solution was placed in a thermostatted cell at 25 8C and kept under a N2 atmosphere. Standard HNO3 (100 mL, 1.00 m) was added and then the acidic solution was titrated with 20-mL additions of standard 0.100 m NaOH, with the pH monitored by a double-electrode pH meter. After each addition of base, the pH of the solution was allowed to stabilize, then a 3-mL portion was withdrawn and placed in a quartz cell to record the emission fluorescence spectrum, after which the sample was returned to the bulk solution. Competition Titrations: Solutions (0.01–0.005 m) of the chosen CH3ACHTUNGRE(CH2)xCOOH acid were prepared in water containing 0.01 m Triton X-100, and the pH was regulated at 7.4 by addition of standard NaOH. This solution was used to titrate an aqueous solution
1736
www.chemphyschem.org
(30 mL) containing HEPES (0.05 m, pH 7.4), Triton X-100 (0.01 m), [Zn(C12Me3cyclen)]2 + (1.66 K 105 m, by addition of 500 mL of the stock solution in tert-butanol) and PCOOH (8.0 K 107 m). The solutions were thermostatted at 25 8C and kept under a N2 atmosphere. The chosen acid was added in portions of 40–80 mL, up to a total acid concentration of 8 K 105 m. After each addition, a 3 mL portion of the titrated solution was transferred to a quartz cuvette to measure the fluorescence emission spectrum, after which it was returned the bulk titrated solution. Potentiometric Titrations: Water (25 mL) was made 0.05 m in NaNO3, 0.01 m in Triton X-100 and 0.001–0.0005 m with the chosen acid. The solution was placed in the cell of the automatic titrating system, thermostatted at 25 8C and kept under a N2 atmosphere. Excess acid (100 mL 1.00 m standard HNO3) was added and the acidic solution was titrated automatically with 20 mL portions of 0.200 m standard NaOH. The results were obtained as E (potential in mV at the glass electrode) versus volume of added base (in mL) and the protonation constants were calculated with the Hyperquad package[20] using E8 values for the hydrogen glass electrode determined by the Gran method.[31] In the case of titrations carried out in dioxane/water (8:2 v/v) mixtures, the same conditions and concentrations of the above case were used, except that solutions were in the solvent mixture and that no Triton X-100 was added. Also in this case, prior to each titration the E8 value for the hydrogen glass electrode was determined by the Gran method.[31] DLS and FPA: DLS measurements were performed with the 632.8 nm He–Ne laser line (Melles Griot, 25-LHP-928-230; 35 mW). The vertically polarized laser beam was focused at the centre of a thermostatted (Thermo Haake GmbH, Germany) cylindrical quartz cell (Hellma, 8 mm internal diameter) surrounded by a larger cell with an index-matching liquid. The scattered light was focused on the photocathode of a single-photon-counting photomultiplier tube (Emi 9873) mounted on a rotating goniometer. A discriminator (Hamamatsu C3866) with an appropriately chosen threshold formed the transistor–transistor logic (TTL) signals which were fed into a digital correlator board (ISS FCS Acquisition card) inserted in a personal computer equipped with the ISS Vista V3.6 software. All measurements were performed at 908 scattering angle and at a temperature of 23 0.3 8C. All the samples in water were 0.01 m in Triton X-100 and 0.05 m in NaNO3. In the samples containing only PCOOH, the fluorophore had a concentration of 4 K 105 m. In the samples containing both fluorophore and the Zn complex, PCOOH was 4 K 105 m and [Zn(C12Me3cyclen)]2 + was 1.6 K 104 m. Solutions were at pH 7.4, regulated by microadditions of standard NaOH. The samples were accurately filtered (Millipore, 0.2 mm) and scattering data were collected at 100 KHz for at least 100 s several times. The correlated data were fitted to the proper function by Origin 7.0 software. The ACF of the scattered intensity I(t), defined as [Eq. (5)]: GðtÞ hIðtÞIðt þ tÞi
ð5Þ
for particles endowed with translational diffusion constant D, is given by [Eq. (6)]: GðtÞ ¼ ½1 þ Aexpð2DK 2 t ÞhIi2
ð6Þ
where A is a constant and K = (4pn/l)sinACHTUNGRE(q/2) is the light scattering vector at the angle q. Dynamic fluorescence measurements were performed with a frequency-modulated phase fluorimeter (Digital K2, ISS, Urbana, IL).
D 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPhysChem 2008, 9, 1729 – 1737
Micelles as Containers for Nanodevices Excitation was accomplished by an argon ion laser (Spectra Physics, 2025) equipped with a multiline UV mirror (351.2 and 363.7 nm, 50 mW). Additional details can be found elsewhere.[32] Digital data acquisition and storage was provided by the ISS-DCA2D200K card inserted in a personal computer. For each data set, at least 15 logarithmically spaced frequencies were employed in the range 5–200 MHz with a cross-correlation frequency of 400 Hz. The accuracies of phase angles and modulation ratios were 0.28 and 0.004, respectively. A long-pass filter at 385 nm (Andover Co., Salem, MA) was employed to cut scattering, and lifetime measurements were performed under magic-angle conditions. A solution of dimethylPOPOP (1,4-bis(4-methyl-5-phenyloxazol-2-yl)benzene) in ethanol was used as reference sample of known lifetime (t = 1.45 ns, see Ref. [23]). The temperature of the samples was kept constant at 23 8C by a thermostatted bath (Thermo Haake GmbH, Germany). A detailed description of the collected data and of their treatment can be found in the Supporting Information.
Acknowledgements Financial support by Universit3 Italo-Francese and Egide (Galileo/ GalilAe grant), MIUR (PRIN-COFIN 2005 grant) and INSTM/MIUR (FIRB grant RBNE019H9 K 002) is gratefully acknowledged. Keywords: fluorescent probes · lipophilicity · micelles · molecular devices · self-assembly [1] M. Wolszczak, J . Miller, J. Photochem. Photobiol. A 2002, 147, 45–54. [2] R. G. Alargova, I. I. Kochijashky, M. L. Sierra, R. Zana, Langmuir 1998, 14, 5412–5418. [3] a) P. P. Infelta, M. Gratzel, J. K. Thomas, J. Phys. Chem. 1974, 78, 190–195. [4] a) R. J. Robson, E. A. Dennis, J. Phys. Chem. 1977, 81, 1075–1078; b) K. Streletzky, G. D. J. Phillies, Langmuir 1995, 11, 42–47; c) P. H. Nelson, G. C. Rutledge, T. A. Hatton, J. Chem. Phys. 1997, 107, 10777–10781; d) E. Feitosa, W. Brown, Langmuir 1998, 14, 4460–4465. [5] Solution Behaviour of Surfactants (Eds.: K. L. Mittal, E. J. Fendler), Plenum Press, New York, 1982. [6] a) F. Grieser, C. J. Drummond, J. Phys. Chem. 1988, 92, 5580–5593; b) C. J. Drummond, F. Grieser, T. W. Healy, J. Chem. Soc. Faraday Trans. 1989, 85, 551–560; c) N. O. Mchedlov-Petrossyan, A. V. Plichko, A. S. Shumaker, Chem. Phys. Rep. 1996, 15, 1661–1678; d) G. P. Gorbenko, N. O. Mchedlov-Petrossyan, T. A. Chernaya, J. Chem. Soc. Faraday Trans. 1998, 94, 2117–2125. [7] a) P. Grandini, F. Mancin, P. Tecilla, P. Scrimin, U. Tonellato, Angew. Chem. 1999, 111, 3247–3250; Angew. Chem. Int. Ed. 1999, 38, 3061–3064; b) M. Berton, F. Mancin, G. Stocchero, P. Tecilla, U. Tonellato, Langmuir 2001, 17, 7521–7528; c) Y. Diaz-Fernandez, A. Perez-Gramatges, V. Amendola, F. Foti, C. Mangano, P. Pallavicini, S. Patroni, Chem. Commun. 2004, 1650–1652; d) P. Pallavicini, L. Pasotti, S. Patroni, Dalton Trans. 2007, 5670–5677. [8] P. Pallavicini, Y. A. Diaz-Fernandez, F. Foti, C. Mangano, S. Patroni, Chem. Eur. J. 2007, 13, 178–187. [9] K. Niikura, E. V. Anslyn, J. Org. Chem. 2003, 68, 10156–10157. [10] Y. A. Diaz-Fernandez, F. Foti, C. Mangano, P. Pallavicini, S. Patroni, A. Perez-Gramatges, S. Rodriguez-Calvo, Chem. Eur. J. 2006, 12, 921–930.
ChemPhysChem 2008, 9, 1729 – 1737
[11] S. Uchiyama, G. D. McClean, K. Iwai, A. P. de Silva, J. Am. Chem. Soc. 2005, 127, 8920–8921. [12] A. Mallick, M. C. Mandal, B. Haldar, A. Chakrabarty, P. Das, N. Chattopadhyay, J. Am. Chem. Soc. 2006, 128, 3126–3127. [13] a) Y. DSaz-FernTndez, S. RodrSguez-Calvo, A. PUrez-Gramatges, C. Mangano, P. Pallavicini, Chem. Phys. Lett. 2004, 398, 245–249; b) Y. DSaz-FernTndez, S. RodrSguez-Calvo, A. PUrez-Gramatges, P. Pallavicini, S. Patroni, C. Mangano, J. Colloid Interface Sci. 2007, 313, 638–644. [14] N. O. Mchedlov-Petrossyan, N. A. Vodolazkaya, A. O. Doroshenko, J. Fluoresc. 2003, 13, 235–248. [15] H. Van de Waterbeemd, D. A. Smith, B. C. Jones, J. Comput. Aided Mol. Des. 2001, 15, 273. [16] a) R. A. Scherrer, S. M. Howard, J. Med. Chem. 1977, 20, 53–58; b) F. Lombardo, M. Y. Shalaeva, K. A. Tupper, F. Gao, J. Med. Chem. 2001, 44, 2490–2497. [17] M. Tachiya, Chem. Phys. Lett. 1975, 33, 289–292. [18] a) T. Koike, E. Kimura, J. Am. Chem. Soc. 1991, 113, 8935–8941; b) E. Kimura, T. Koike, J. Am. Chem. Soc. 1996, 118, 10963–10970. [19] a) G. A. Kalligeros, E. L. Blinn, Inorg. Chem. 1972, 11, 1145–1148; b) E. K. Barefield, F. Wagner, Inorg. Chem. 1973, 12, 2435–2439. [20] All the data obtained from potentiometric and fluorimetric titrations were fitted with the least-squares non-linear regression package Hyperquad; see P. Gans, A. Sabatini, A. Vacca, Talanta 1996, 43, 1739–1753. [21] The Zn2 + complex of C12Me3cyclen is an amphiphilic molecule that forms micelles itself if sufficiently concentrated. Using the standard method based on the I1/I3 intensity ratio of plain pyrene fluorescence (see, for example: A. Dominguez et al., J. Chem. Ed. 1997, 74, 1227), we measured a critical micelle concentration (cmc) of 5.4 K 103 m at pH 6 (+ 2 charged complex) and of 1.8 K 103 m at pH 9 (coordinated OH , + 1 charged complex). Noticeably, in all the experiments described herein, we worked well under the found cmc values. [22] N. V. Nucci, B. Zelent, J. M. Vanderkooi, J. Fluoresc. 2008, 18, 41–91. [23] G. D. J. Phillies, J. Stott, S. Z. Ren, J. Phys. Chem. 1993, 97, 11563–11568. [24] a) L. Ge, X. Zhang, R. Guo, Polymer 2007, 48, 2681–2691; b) K. Y. Mya, A. M. Jamieson, A. Sirivat, Polymer 1999, 40, 5741–5749. [25] a) T. Saitoh, H. Hoshino, T. Youtsuyanagi, J. Chem. Soc. Faraday Trans. 1994, 90, 479–486; b) D. T. Cramb, S. C. Beck, J. Photochem. Photobiol. A: Chemistry 2000, 134, 87–95. [26] J. R. Lakowicz, Principles of Fluorescence Spectroscopy , Springer, Singapore, 1996, pp. 383–400. [27] Literature reports[4a] a = 5.4 and b = 2.7 nm. The slightly larger value that we found is connected to the well-known tendency of Triton X100 micelles to increase their dimensions in the presence of concentrated electrolytes (see, for example: I. D. Charlton, A. P. Doherty, J. Phys. Chem. B 2000, 104, 8327). Our DLS measurements were carried out in the same environment as that used for pH-variable fluorimetric and potentiometric titrations, that is, with 0.05 m NaNO3 as ionic strength buffer. [28] M. Kruppa, B. Kçnig, Chem. Rev. 2006, 106, 3520–3560. [29] IUPAC Stability Constants Database (SC-Database), Academic Software, UK. [30] R. A. Bissell, A. J. Bryan, A. P. de Silva, C. P. McCoy, J. Chem. Soc. Chem. Commun. 1994, 405–407. [31] G. Gran, Analyst 1952, 77, 661–671. [32] L. D’Alfonso, M. Collini, G. Baldini, Eur. J. Biochem. 2003, 270, 2497– 2504.
Received: May 19, 2008 Published online on July 23, 2008
D 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemphyschem.org
1737
