Watching the Conformational Changes of Maleonitriledithiolate

7230

J. Phys. Chem. A 2010, 114, 7230–7240

Watching the Conformational Changes of Maleonitriledithiolate Chromophores Inside the Inclusion Complexes with Cyclodextrins: Probed by ICD Spectra and DFT Calculations Xian Cheng, Qi Wang, Changsheng Lu,* and Qingjin Meng* Nanjing National Laboratory of Microstructures, Coordination Chemistry Institute, State Key Laboratory of Coordination Chemistry, School of Chemistry and Chemical Engineering, Nanjing UniVersity, Nanjing 210093, P.R. China ReceiVed: April 7, 2010; ReVised Manuscript ReceiVed: May 20, 2010

A series of inclusion complexes between cyclodextrins (R-, β-, γ-cyclodextrin and HP-β-cyclodextrin, HPβ-cyclodextrin ) 2-hydroxypropyl-β-cyclodextrin) and sodium maleonitriledithiolate (Na2mnt) were investigated by electronic spectra, induced circular dichroism (ICD) spectra, and quantum chemical studies. The inclusion complexes Na2mnt@R-cyclodextrin and Na2mnt@γ-cyclodextrin did not show any ICD signals, whereas Na2mnt@HP-β-cyclodextrin displayed two signs of splitting Cotton effects (CEs), with one positive CE couplet at 376 nm in the 365-410 nm region and the other negative at 277 nm in the 265-306 nm region. In addition, a dimeric host inclusion pattern of Na2mnt@HP-β-cyclodextrin in solution was determined by the method of continuous variation. Density functional theory (DFT) was used to assist assignment of the ICD signals in inclusion complexes Na2mnt@β-cyclodextrin and Na2mnt@HP-β-cyclodextrin in combination with the well-known Harata’s rule. The orientation of p f π* transition in Na2mnt chromophore was predicted by TD-DFT (time-dependent DFT) calculations to be along the CdC double bond instead of being perpendicular. Upon titrations with Zn2+ solutions, reversals of the p f π* transition-relevant ICD peak and splitting CE were experimentally observed in the cases of Na2mnt@β-cyclodextrin and Na2mnt@HP-βcyclodextrin, respectively, which strongly supported our hypotheses on their coconformations and the subsequent conformational changes of mnt chromophores occurring during the titration procedures. Therefore, on the basis of both the experimental data and TD-DFT calculations, the HP-β-cyclodextrin dimer host in inclusion complex Na2mnt@HP-β-cyclodextrin was disclosed, which in turn generated the exciton coupling between the two individually included guests and produced the splitting CEs as well as the reversals of CEs. 1. Introduction Cyclodextrins are a series of oligosaccharides consisting of six (R-cyclodextrin) or more (β-, γ-cyclodextrin, etc.) R-1,4linked D-glucopyranose residues in a cyclic array, which have slightly different cavity diameters. These molecules appear as truncated cones with primary and secondary hydroxyl groups crowning the narrower and the wider rims of their molecular cavities, respectively. Because cyclodextrins have hydrophilic exteriors and hydrophobic cavities, they are one of the most fascinating host molecules in supramolecular chemistry.1–3 They can form stable inclusion complexes with various guest molecules varying from organic compounds,4,5 amino acids and their derivatives,6,7 drugs and pharmaceutical chemicals,8 polymers,9 to metal coordination compounds.10,11 It is known that the main driving force for the formation of inclusion complexes is attributed to the hydrophobic interactions between cyclodextrin cavities and the hydrophobic guests or the hydrophobic components of guests.12 During the past several decades, much effort has been devoted to investigating the cyclodextrin inclusion complexes,13,14 most of which contained organic compounds guests. Only a little is known about the inclusion complexation between positively or negatively charged compounds and cyclodextrins.15 Therefore, the inclusion complexes between cyclodextrins and ionized guests are intriguing. Recently, induced circular dichroism (ICD) spectrometry, a sensitive and powerful tool to explore the inclusion complex* Corresponding authors. Tel: (+)(86)-25-83597266. Fax: (+)(86)-2583314502. E-mail: [email protected] (C.L.); [email protected] (Q.M.).

ation in solution,16 has been widely utilized to analyze the orientation of the aromatic molecules included in cyclodextrins or to determine the direction of a transition moment excited from a chromophore inside the cyclodextrin cavity, and the information from the solution structures of host-guest complexes greater contributes to the understanding of molecular recognition phenomena as well as the enzyme-substrate interaction or catalysis.17–19 In our previous paper, the inclusion complex between β-cyclodextrin and a bivalenced anion guest, sodium maleonitriledithiolate (Na2mnt), was reported.20 By using ICD spectrometry, we got some interesting results from the inclusion complex Na2mnt@β-cyclodextrin, which indicated the strong interaction between the host and guest. The inclusion constant of Na2mnt@β-cyclodextrin was estimated to be (2.45 ( 0.15) × 103 or (3.10 ( 0.11) × 103 M-1 in solution by different methods, and the orientation of the guest anion inside the host cavity was analyzed by structural optimization using PM3 quantum chemical method as well. Although the strong host-guest interaction has been discovered in inclusion complex Na2mnt@β-cyclodextrin, detailed elucidation of the transition-relevant ICD signals remain unknown. Therefore, in this article, we recorded ICD spectra of different types of inclusion complexes and used quantum chemical methods (TD-DFT) to investigate the inclusion complexation between Na2mnt and the selected cyclodextrins (R-, β-, and γ-cyclodextrin and HP-β-cyclodextrin). In these systems, it is interesting that almost all of the cyclodextrin hosts show

10.1021/jp103118z  2010 American Chemical Society Published on Web 06/02/2010

Watching the Conformational Changes of Na2mnt

J. Phys. Chem. A, Vol. 114, No. 26, 2010 7231

SCHEME 1: Guest Molecule Na2mnt

certain abilities to include the ionized guests, whereas their cavities are hydrophobic. With the aid of DFT calculations in the interpretation of experimental data, for the first time, we are able to elucidate the details of the p f π* transition-relevant ICD activities of maleonitriledithiolate chromophore,21 in addition to that of the π f π* transitions of the aromatic chromophores or that of the n f π* transitions of diazirines and azoalkanes systems, which have been well studied.22 It is found that the absorption maximum of Na2mnt, which was ascribed to its p f π* transition, changed subtly with different host molecules. This provided us with a unique tool for the structural assignments of Na2mnt@cyclodextrins systems in solution. Although only a few cyclodextrin inclusion complexes havebeenreportedtoshowsplittingICDsignalsexperimentally,23–26 the inclusion complex Na2mnt@HP-β-cyclodextrin was observed to present two splitting Cotton effects (CEs) couplets in solution.

Figure 1. Electronic spectra of Na2mnt ([Na2mnt] ) 1.61 × 10-4 M) in water and in the presence of increasing concentrations of R-cyclodextrin; [R-cyclodextrin] ) 0, 0.8 × 10-4, 1.61 × 10-4, 2.42 × 10-4, 3.22 × 10-4, 4.83 × 10-4, 9.66 × 10-4, 14.49 × 10-4, 19.32 × 10-4, and 28.98 × 10-4 M (a f j).

2. Results and Discussion Many physical and chemical methods, such as UV, IR, NMR, cyclic voltammograms, and thermogravimetric analysis, have been established to study the host-guest inclusion complexation in cyclodextrin chemistry.29 The guest Na2mnt (Scheme 1) in our case is very soluble and contains no hydrogen atoms. Therefore, UV and ICD were used to investigate the formation of inclusion complexes Na2mnt@R-cyclodextrin, Na2mnt@HPβ-cyclodextrin, and Na2mnt@γ-cyclodextrin in solution. It is known that Harata’s rule30 and Kodaka’s rule31 are applicable in the prediction of ICD signals in cyclodextrin inclusion complexes. When inside the cyclodextrin cavities, transitions with their electric transition dipole moments parallel to the symmetry axes of the hosts give positive ICD signals, whereas the perpendicular ones bring about negative ICD signals.30 When outside the cyclodextrin cavities, a reverse conclusion dominates.31 Up to date, Harata and Kodaka’s rule is proven to be valid in the interpretation of π f π* transition-relevant ICD activities of the aromatic chromophores as well as in that of n f π* transition-relevant ICD signals of diazirines and azoalkanes systems.22 Therefore, we want to assess the validity of Harata and Kodaka’s rule in other conjugated systems, for instance by switching to the investigation of the known p f π* transitions of maleonitriledithiolate chromophores. 2.1. Inclusion Complex Na2mnt@r-cyclodextrin (1). As shown in Figure 1, one strong and broad peak was observed at ∼364 nm (curve a) for the unincluded guest (Na2mnt). When titrated with the increasing concentrations of R-cyclodextrin solutions, the guest solution exhibited changes in its electronic spectra but did not shift at all (curves b-j). It is quite different from that in the inclusion complex Na2mnt@β-cyclodextrin, where the absorption maximum shifted from 366 to 375 nm upon β-cyclodextrin titrations.20 The minor change of the UV spectra in inclusion complex 1 might result from the small cavity size of the host, which was not so suitable to include entirely Na2mnt that the microenvironment of the guest molecule was hardly influenced. So when R-cyclodextrin was added to the guest solution, the absorption was almost the same as that of the unincluded Na2mnt in aqueous solution, except for minor changes of the signal magnitude.

Figure 2. ICD (above) and UV (below) spectra of Na2mnt ([Na2mnt] ) 1.61 × 10-4 M) in water and in the presence of increasing concentrations of HP-β-cyclodextrin; HP-β-cyclodextrin ) 0, 0.8 × 10-4, 1.61 × 10-4, 2.42 × 10-4, 3.22 × 10-4, 4.83 × 10-4, 9.66 × 10-4, 14.49 × 10-4, 19.32 × 10-4, 28.98 × 10-4, 32.2 × 10-4, 64.4 × 10-4, and 128.8 × 10-4 M (a f m).

It is known that ICD is a very sensitive method to explore the inclusion complexation. Because the signal changes in electronic spectra were minor (Figure 1), the ICD spectra of inclusion complex 1 were studied as well. However, it showed horizontal lines with no signals at all. This result indicated that Na2mnt was not efficiently included by R-cyclodextrin. Thereby, the planar guest molecule could not be induced to produce any ICD signals by the chiral host. In the meantime, β-cyclodextrin has a molecular cavity large enough to include the guest, which induced the achiral guest to produce recordable ICD signals.20,32 2.2. Inclusion Complex Na2mnt@HP-β-cyclodextrin (2). The unincluded guest (Na2mnt) showed one broad peak (UV signal in curve a, Figure 2). However when titrated with the increasing concentrations of HP-β-cyclodextrin solutions, the guest solution exhibited bathochromic shift from 364 to 378 nm where the host was about 0.5- to 18-fold excessive (curve b-j). In addition, a weak absorption mounted gradually at ∼277 nm. The bathochromic effect of the intensive peak was identical to that of inclusion complex Na2mnt@β-cyclodextrin, which indicated the formation of the inclusion complex in solution. It is ascribed to the microenvironment change of Na2mnt when

7232

J. Phys. Chem. A, Vol. 114, No. 26, 2010

Figure 3. Continuous variation measurements for the inclusion complex Na2mnt@HP-β-cyclodextrin ([HP-β-cyclodextrin] + [Na2mnt] ) 3.0 × 10-4 M) in water at 20 °C (λ ) 405, 410, 425, and 430 nm).

included by HP-β-cyclodextrin compared with that of the unincluded Na2mnt in aqueous solution. As has been studied, the unincluded guest showed no ICD signals in solution because of its highly symmetric planar structure. However, the bisignate ICD signals were recorded upon inclusion complexation of Na2mnt with HP-β-cyclodextrin. The formation of inclusion complex 2 gave rise to two signs of CEs, one positive CE at 376 nm (the crossover point) in the region 365-410 nm (positive ICD absorption followed by a negative ICD absorption on going from longer to shorter wavelengths), and the other negative CE at 279 nm (the crossover point) in the region 265-306 nm (negative ICD absorption, followed by a positive ICD absorption on going from longer to shorter wavelengths). It is surprisingly in accordance with the two UV absorption bands at around 277 and 378 nm, as shown in Figure 2. In addition, the negative ICD trough at ∼365 nm fainted almost to the background when the host was about 40-fold excessive. It is well-known that an achiral guest will be induced by chiral cyclodextrins cavities to produce circular dichroism signals. Accordingly, the formation of inclusion complex 2 was confirmed by its ICD spectra. Compared with its electronic spectra (Figure 2), the ICD spectra of inclusion complex 2 are more characteristic and sensitive in identifying the host-guest inclusion complexation. In contrast with the only one positive ICD band at ∼380 nm in the case of inclusion complex Na2mnt@β-cyclodextrin,20 it is surprising that the splitting CE couplets were recorded in

Cheng et al. inclusion complex 2. The splitting CE couplets (in another word exciton-coupled ICD signals) were proposed to be generated by induced axial chirality33 via the desymmetrization of the guest upon host complexation, which originated from the strong coupling of the electric transition dipole moments between the two included Na2mnt chromophores when in close proximity (vide post, in TD-DFT Calculation Results section). On the basis of the strong UV absorption and the splitting CE couplets observed in inclusion complex 2, we intended to propose an appropriate inclusion pattern of Na2mnt@HP-βcyclodextrin in solution. Therefore, first of all, the stoichiometry of inclusion complex 2 was determined by the method of continuous variation,20 which was based on the measurements of ICD signal intensities at 405, 410, 425, and 430 nm. The results, maximum at 0.4 at all four different wavelengths recorded, strongly accounted for the 2:3 ratio of the guest to the host in solution (Figure 3) compared with the 1:1 stoichiometry identified in inclusion complex Na2mnt@β-cyclodextrin. This fact suggested a dynamically spatial complement between HP-β-cyclodextrin and Na2mnt, which enhanced a 2:3 inclusion complex between the host and guest, and the probable dimeric coconformation of inclusion complex 2 is illustrated in Scheme 2. HP-β-cyclodextrin employed here has about five hydroxyl groups randomly substituted by hydroxypropyl groups per molecule of cyclodextrin, which makes the exterior of HP-βcyclodextrin more hydrophobic with respect to the parent β-cyclodextrin.34,35 Because of the formation of hydrogen bonding network between the hydroxyl groups of the side chains and the glucose units, the 2-hydroxypropyl side groups located at the O2 positions of HP-β-cyclodextrin result in a more “spread out” but dynamically more restricted molecular configuration. It widens the cavity entrance at the secondary rim of β-cyclodextrin molecule and so enhanced the encapsulation capability of HP-β-cyclodextrin.34,35 Therefore, the more compact inclusion pattern rather than the 1:1 ratio of host-to-guest stoichiometry was set up in inclusion complex 2. Most likely, the sodium ions from the guest molecules played an important role in favoring the formation of the proposed host dimer by certain coordination interactions between Na+ cations and hydroxypropyl groups, in addition to the hydrogen bonding networks in between the two parent β-cyclodextrins. 2.3. Inclusion Complex Na2mnt@γ-cyclodextrin (3). Besides a weak peak mounting at ∼277 nm, the absorption maximum of the guest shifted from 364 to 372 nm when Na2mnt solution was titrated with the increasing concentrations of

SCHEME 2: Proposed Dimeric Co-conformation of Inclusion Complex 2


J. Phys. Chem. A, Vol. 114, No. 26, 2010 7233 SCHEME 3: Structure of p-Hydroxybenzoic Acid Constructed and Optimized by DFT Calculations

Figure 4. UV spectra of Na2mnt ([Na2mnt] ) 1.61 × 10-4 M) in water and in the presence of increasing concentrations of γ-cyclodextrin; [γ-cyclodextrin] ) 0, 0.8 × 10-4, 1.61 × 10-4, 2.42 × 10-4, 3.22 × 10-4, 4.83 × 10-4, 14.49 × 10-4, 19.32 × 10-4, 28.98 × 10-4 38.64 × 10-4, 43.47 × 10-4, and 57.96 × 10-4 M (a f l).

γ-cyclodextrin solutions (curves a-l, Figure 4). The observed bathochromic effect also indicated the formation of inclusion complex 3 in solution. However, inclusion complex 3 hardly showed any ICD signals in solution (Figure S1 in Supporting Information). γ-Cyclodextrin consists of eight R-1,4-linked D-glucopyranose residues (compared with seven residues in β-cyclodextrin) in a cyclic array and has a large cavity to include the guest. Because the formation of inclusion complex 3 in solution was confirmed by its electronic spectra, we proposed that the guest inside γ-cyclodextrin cavity was too flexible to be induced to produce any recordable ICD signals.30,31 3. TD-DFT Calculation Results To get a detailed understanding of the inclusion complexation events between the host and guest, quantum chemical method PM3 from the Gaussian 03 package28 has been used to study the inclusion complex Na2mnt@β-cyclodextrin in our previous study, and in this article, we utilized TD-DFT methods at the B3LYP level with a (6-31G + d, p) basis set to investigate the above-recorded absorption bands in both UV and ICD spectra (Figure 2) from the viewpoint of molecular orbits, which had never been explored before. As is known, excited states in general and optical spectra in particular can be predicted from first principles using TD-DFT methods,36 which has been proven to be a reliable method for studying excited states in broad classes of relatively large systems with good precision.37 However, before applying the TD-DFT calculations, it is still necessary to verify the accessibility of this method in our cases. TD-DFT calculations are used to estimate the energies of the relevant excited states as well as the oscillator strengths for the transitions of interests, able to assist comparison with the corresponding electronic absorption data. When assessing our TD-DFT methods with the classical inclusion complex phydroxybenzoic acid@β-cyclodextrin (the guest molecule and its computed Cartesian axes were illustrated in Scheme 3), the calculated results (Table 1) showed that the electrical dipole moment of the transition between molecular orbits (MOs) 36 and 37, which were the HOMO (highest occupied MO) and LUMO (lowest unoccupied MO) of guest molecule, was parallel to the X axis at ∼248 nm. The other two calculated transition dipole moments (between MOs 35 and 37, MOs 36 and 38) were determined to be parallel to the Y axis at ∼210 nm. Assuming that the calculated transitions at 248 and 210 nm are

equivalent to those experimentally reported at = 259 and 216 nm,38 respectively, our TD-DFT methods predicted that the ICD signals at ∼259 nm were positive and that at ∼216 nm were negative by Harata’s rule when p-hydroxybenzoic acid@βcyclodextrin was in an axial inclusion coconformation. It turned out to be in agreement with the results from the literature.38 Furthermore, our TD-DFT methods were assessed with another inclusion complex, azoalkane@β-cyclodextrin. (The guest molecule and its computed Cartesian axes were illustrated in Scheme 4.) From the calculated results (Table 2), we found that electrical dipole moment of the n f π* transition between MOs 30 and 31 (HOMO and LUMO of azoalkane) was predicted to be parallel to the X axis at ∼384 nm, in agreement with which was experimentally reported to be perpendicular to the NdN double bond at ∼370 nm. The other two calculated transition dipole moments (between MOs 30 and 33, MOs 30 and 35) were determined to be parallel to Y axis or Z axis at around 220 and 208 nm, respectively, in agreement with which was experimentally reported at ∼220 nm.39–41 Consequently, our TD-DFT methods were proven to be able to predict the same orientations of the transition dipole moments in the case if azoalkane chromophores as well. Now, we switched to the intensive ICD signals recorded in inclusion complex 2 (Figure 2). The problem of how to assign these bands remained to be addressed. Fortunately, on the basis of the TD-DFT calculations of inclusion complexes p-hydroxybenzoic acid @β-cyclodextrin and azoalkane@β-cyclodextrin assessed above, it is proven that our TD-DFT methods are good approximate computations to assign the bands of ICD signals. Therefore, TD-DFT methods were utilized further to study the UV and ICD signals of inclusion complexes Na2mnt@βcyclodextrin and Na2mnt@HP-β-cyclodextrin in the following below. From our previous study of inclusion complex Na2mnt@βcyclodextrin (the guest molecule and its computed Cartesian axes are illustrated in Scheme 5), the p f π* transition in Na2mnt was involved. Therefore, first of all, we’d like to assign this characteristic band in spectra. According to our TD-DFT calculations (Table 3), the transition 1 between MOs 36 and 37 (HOMO and LUMO of Na2mnt) at ∼428 nm indicated the most intensive absorption (oscillator strength is 0.3191). Although not necessarily relevant to the present experimental data, calculated oscillator strength can be utilized to aid with comparison of solution phase electronic absorption data.42 Therefore, it could be the most probable candidate for the assignment of the strongest absorption observed (Figure 2) because we know that the most intensive absorption in electronic spectra of Na2mnt is ascribed to its p f π* transition.20 So, the visualized molecular orbits were checked for further identification.

7234


Cheng et al.

TABLE 1: Calculated and Experimental Transition Absorptions of p-Hydroxybenzoic Acid

a Cartesian axes are illustrated in Scheme 3, and the three digitals are components of the transition dipole moment on X, Y, and Z axes, respectively. b Oscillator strength was predicted by TD-DFT calculations at the optimized geometry.

SCHEME 4: Structure of Azoalkane Constructed and Optimized by DFT Calculations (All Hydrogen Atoms Are Omitted for Clarity)

Three-dimensional contours of the calculated molecular orbits were generated by the Gaussian 03 package and plotted (column 6 in Table 3). MO 36 had the highest electron density in the p orbits of sulfur atoms, whereas MO 37 featured the π* antibonding orbital surface. Therefore, the calculated p f π* transition at 428 nm by TD-DFT methods was just in accordance with the experimental UV absorption at 366 nm reported in our previous work and at 364 nm in this article. Although the calculated transition 2 between MOs 35 and 38 (HOMO-1 and LUMO+1 of Na2mnt, Table 3) had a theoretical absorption wavelength closer to the experimental data, it was determined not to be the characteristic p f π* transition by both the oscillator strength and the MO contour plots involved. Most likely, this transition (between MOs 35 and 38) overlapped with the intensive p f π* transition and was recorded as well (curve a, Figures 1, 2, and 4). The ICD signal within the same p f π* transition exhibited maximum at ∼380 nm as a positive band, whereas the UV absorption mounted at ∼375 nm in inclusion complex Na2mnt@βcyclodextrin.20 However, as learned from Table 3, the p f π* transition surprisingly was parallel to the X axis (Scheme 5), which was completely different from our previous hypothesis. Therefore, if the p f π* transition-relevant ICD signals were still subject to Harata’s rule, then the axial alignment of inclusion complex Na2mnt@β-cyclodextrin could be identified because of its positive ICD band recorded. This conclusion suggested that the guest molecule would like to adopt certain preferred coconformation in the inclusion complex Na2mnt@β-cyclodextrin in solution, where the CdC double bond of Na2mnt was parallel to the symmetry axis of the host cavity. Although in PM3 simulations this type of coconformation has a potential energy 44.8 kJ/mol higher than the lowest one in those of the

inclusion complex Na2mnt@β-cyclodextrin,20 it still could dominate in aqueous solution because the salvation effects were excluded from our TD-DFT calculations. In another word, this coconformation may be favored when the ionization of the bivalenced anion guest is taken into consideration. However, the proof from experimental data is necessary in supporting our hypothesis here (vide post). According to this result, we might be able to withdraw the conclusion in our previous study where the p f π* transition of Na2mnt was supposed to be along its Z axis. (The Cartesian axes were illustrated in Scheme 5.) The inclusion pattern of Na2mnt chromophore inside β-cyclodextrin cavity is thus defined if the ICD signal and its interpretation are subject to the well-known Harata’s rule. As a consequence, the same axial coconformation of Na2mnt inside HP-β-cyclodextrin cavity and the similar ICD signals are supposed to be disclosed in the case of inclusion complex 2 because there is almost no cavity difference between β-cyclodextrin and HP-β-cyclodextrin. As shown in Figure 2, there seemed to be four absorption bands in ICD curves (around 265, 306, 365, and 410 nm) but only two in UV curves (around 277 and 378 nm) upon inclusion complexation of HP-β-cyclodextrin with Na2mnt. Because the ICD signals of inclusion complex 2 were so complicated and not relevant to its UV signals, there must be some mechanisms responsible in our case rather than the individual electric transition dipole moments interpreted by Harata’s rule. In fact, when HP-β-cyclodextrin was much more excessive than Na2mnt, the negative ICD band at ∼365 nm almost disappeared (Figure 2, vide supra). It suggested that this ICD trough was not a real independent band but rather an auxiliary signal. From the stoichiometry determination of inclusion complex 2 above, a dimeric host has been proposed. So, the individual electric transition dipole moments of Na2mnt chromophores were capable of interacting with each other when in close proximity inside the dimeric host cavity. Therefore, the exciton-coupled mechanism was taken into consideration. Exciton-coupled circular dichroism (ECCD), in particular, has been extensively applied to various organic molecules as a microscale method to determine their absolute configurations in a nonempirical manner.43 Exciton-coupled CD is caused by induced axial chirality, which indicates that the chirality information of a chiral host (or guest) is transmitted into an asymmetric deformation of the transition moments in the achiral guest (or host). Therefore, the ICD signals with splitting CE are supposed to exhibit in inclusion complexes as a result of exciton chirality induction in otherwise symmetric guest (or host)


J. Phys. Chem. A, Vol. 114, No. 26, 2010 7235

TABLE 2: Calculated and Experimental Transition Absorption of Azoalkane

a Cartesian axes are illustrated in Scheme 4, and the three digitals are components of the transition dipole moment on X, Y, and Z axes, respectively. b Oscillator strength was predicted by TD-DFT calculations at the optimized geometry.

SCHEME 5: Structure of mnt2- Anion Constructed and Optimized by DFT Calculations

upon binding of a chiral host (or guest).44 According to the molecular exciton theory, the absorption maximum of the electronic spectrum and the crossover point of the ICD couplet should coincide, and the magnitudes of both extrema in the splitting pattern should be the same. Only a few cyclodextrin inclusion complexes have been reported to show splitting ICD signals experimentally. For example, the splitting pattern was supposed to arise from a dipole-dipole interaction between the electric transition dipole moments of the intramolecular charge-transfer band of two azo dye moieties in the inclusion complexes of modified azobenzenes with cyclodextrins.23–25 However, the azobenzene derivatives are subject to the trans-to-cis rearrangement of the NdN chromophores under the same experimental condition, which seemed to play an ambiguous role in the splitting ICD signals. As a matter of fact, one most recent paper claimed positive and negative ICD signals separately in the inclusion complexes 4-hydroxyazobenzene@PM-β-cyclodextrin and 4-aminoazobenzene@PM-β-cyclodextrin (PM-β-cyclodextrin ) permethylated β-cyclodextrin) and ascribed them to π f π* and n f π* transitions of the azo chromophore, respectively.26 In addition, the azoalkanes and diazirines guests showed no splitting ICD signals at all when included by cyclodextrins, where the trans-to-cis transformation of the NdN chromophore was disabled.39–41,45,46 On the basis of our TD-DFT calculations, there were mainly five electric transition dipole moments (numbered from 1 to 5 with oscillator strength g0.05, Table 3) disclosed from Na2mnt.

The transitions 1 (from HOMO to LUMO of the guest) and 2 (from HOMO-1 to LUMO+1) were determined to be parallel to the X axis. In the case of inclusion complex 2, both of the two transitions accounted for the 378 nm UV absorption and should produce positive ICD signals when subject to Harata’s rule. As far as the above-mentioned exciton-coupled mechanism was concerned, we consequently ascribed the splitting positive CE at ∼376 nm (the crossover point) in the region 365-410 nm (Figure 2) to the calculated transitions 1 and 2. Therefore, the TD-DFT calculations and our experimental data were in agreement with each other when incorporating both Harata’s rule and the exciton-coupled mechanism. In the same process, the other three transitions (3-5), which were perpendicular to the CdC double bond in either Y or Z directions, were proposed to contribute to the 277 nm absorption in UV spectra and were ascribed to the splitting negative CE at 279 nm (the crossover point) in the region 265-306 nm in ICD spectra shown in Figure 2. According to the experimental results, it showed that inclusion complex 2 could give rise to exciton-coupled ICD signals arising from the induced axial charity.47 As illustrated in Scheme 6, the observed ICD signals with splitting CEs from inclusion complex 2 were proposed to result from the coupling of the transition dipole moments involved. When the electric transition dipole moment (marked in blue) of the included guest interplays with the other one (marked in red) in close proximity, it brings about a pair of enantiomers a and b, and the formation of enantiomers a and b in turn generates the splitting CEs recorded experimentally in Figure 2. In addition, the asymmetric splitting CEs in Figure 2 probably result from the five random hydroxypropyl groups substitution per molecule of β-cyclodextrin, which distributed unevenly on the secondary rim of HP-βcyclodextrin. Therefore, the ratio of enantiomers a and b (Scheme 6) cannot be equal, which results in a diastereomeric excess of one of the axially chiral conformers and gives rise to an asymmetric splitting CE. The exciton-coupling model is consistent with the dimeric coconformation of inclusion complex 2 discovered from the 2:3 stoichiometry above, and the third Na2mnt guest, just in the middle of the proposed dimeric host cavities (Scheme 2), was so flexible and did no contribution to the ICD curves of inclusion complex 2. Most of the well-defined splitting CEs were reported in socalled self-inclusion complexes, where the monofunctionalized

7236


Cheng et al.

TABLE 3: Calculated and Experimental Transition Absorption of mnt2- Anion

a

Cartesian axes are illustrated in Scheme 5, and the three digitals are components of the transition dipole moment on X, Y, and Z axes, respectively. b Oscillator strength was predicted by TD-DFT calculations at the optimized geometry. c Absorption maximum of unincluded Na2mnt. d Absorption maximum of Na2mnt in inclusion complex 2.

SCHEME 6: Proposed Exciton-Coupling between Transition Dipole Moments in Inclusion Complex 2a

a Blue and red arrows represent transition dipole moments of Na2mnt included in HP-β-cyclodextrin, and the dot line indicates that the dipole moments are behind from the front view.

cyclodextrins could interact with each other via their mutual inclusion. The main driving force for this kind of inclusion complexation was attributed to the hydrophobic interactions, for instance, between the naphthyl groups and cyclodextrin cavities.48 In those cases, it is proven that the dimerization of the cyclodextrins cavities together with the mutual inclusion of their chromophoric guests were able to bring about the exciton coupling and produce splitting CEs. However in our case, the experimental data suggested that the dimeric host cavities came into being merely via dynamic hydrogen bonding networks and in consequence induced the exciton-coupling interactions, which so far was rarely previously reported.49 It is well known that the mnt chromophore can serve as a good ligand to coordinate with transition metals, resulting in rod-like coordination compounds,50,51 and when these coordination compounds are included by β-cyclodextrin, for example in the case of [Ni(mnt)2]2-@β-cyclodextrin, they tend to adopt the rotaxane-like coconformation where the CdC double bond of mnt ligand is thus definitely perpendicular to the host cavity, no matter in solution or in crystals (data not shown). Therefore, we should be able to observe experimentally the conformational change of the mnt moieties when both Na2mnt@β-cyclodextrin

Figure 5. ICD (above) and UV (below) spectra of Na2mnt@βcyclodextrin ([Na2mnt] ) 1.61 × 10-4 M, [β-cyclodextrin] ) 6.44 × 10-4 M) in water and in the presence of increasing concentrations of ZnCl2; [ZnCl2] ) 0, 0.8 × 10-5, 1.61 × 10-5, 2.42 × 10-5, 3.22 × 10-5, 4.83 × 10-5, 6.44 × 10-5, 8.05 × 10-5, 1.61 × 10-4, and 6.44 × 10-4 M (a f j).

and Na2mnt@HP-β-cyclodextrin solution were titrated with metal ions if our hypotheses above on the coconformation of Na2mnt@β-cyclodextrin and the exciton-coupling mechanism in Na2mnt@HP-β-cyclodextrin are correct. To circumvent the complicated charge transfer transitions between metals and the coordinating mnt ligands, ZnCl2 was used in the titration experiments below because Zn2+ metal ion had a full d10 electron configuration. The inclusion complex Na2mnt@β-cyclodextrin showed one broad peak at 376 nm in its electronic absorption spectra (UV signals in curve a, Figure 5), and when it is titrated with the increasing concentrations of Zn2+ solutions, the absorption maximum shifted to 388 nm where the [Zn2+] was about 0.05 to 4-fold excessive (UV signals in curves b-j, Figure 5). In addition, the absorption at 271 nm mounted gradually. It indicated the formation of [Zn(mnt)2]2-@β-cyclodextrin in solution.


J. Phys. Chem. A, Vol. 114, No. 26, 2010 7237 SCHEME 7: Structure of [Zn(mnt)2]2- Constructed and Optimized by DFT Calculations

Figure 6. ICD (above) and UV (below) spectra of Na2mnt@HP-βcyclodextrin ([Na2mnt] ) 1.61 × 10-4 M, [HP-β-cyclodextrin] ) 6.44 × 10-4 M) in water and in the presence of increasing concentrations of ZnCl2; [ZnCl2] ) 0, 0.8 × 10-5, 1.61 × 10-5, 2.42 × 10-5, 3.22 × 10-5, 4.83 × 10-5, 6.44 × 10-5, 8.05 × 10-5, 1.61 × 10-4, and 6.44 × 10-4 M (a f j).

When in the absence of Zn2+ solutions, the inclusion complex Na2mnt@β-cyclodextrin showed a faint positive ICD band at ∼380 nm (ICD signals in curve a, Figure 5) because of the low concentration ratio (1: 4) of the guest to the host, as reported before.20 However, when the inclusion complex Na2mnt@βcyclodextrin solution was titrated with the increasing concentrations of Zn2+ solutions, a broad but strong ICD trough at ∼386 nm was generated (ICD signals in curves b-j, Figure 5) instead of the faint and positive ICD band from the untitrated solution. In the meantime, the other strong and positive ICD band was observed to mount at ∼274 nm during our titration process. Then, we switched to the case of inclusion complex 2. As shown in Figure 6, upon titration with the increasing concentrations of Zn2+ solutions, the UV spectra of inclusion complex 2 solutions exhibited almost the same changes as those recorded in that of the inclusion complex Na2mnt@β-cyclodextrin (UV curves in Figure 5). During the titration process, the formation of [Zn(mnt)2]2- in the presence of HP-β-cyclodextrin gave rise to two intensive absorption peaks situated at around 270 and 390 nm, respectively. In the meantime, the splitting CEs shifted to 386 nm (the crossover point) in the region 373-400 nm (ICD curves a-j in Figure 6), and another weak ICD band was observed to mount gradually at ∼339 nm. However, the most fascinating discovery during this experiment was the reversal of the splitting CE couplet in the region 257-288 nm. As learnt from Figure 2, one negative CE couplet at ∼279 nm (the crossover point) in the region 265-306 nm was recorded in the absence of Zn2+ solutions (curve a in Figure 6). Upon titration with the increasing concentrations of Zn2+ solutions, the positive peak of the CE couplet shifted bathochromically, whereas the negative trough fainted gradually (curves b and c in Figure 6). When the concentration of [Zn2+] was about 0.15 to 0.2 times that of the mnt ligand, the negative CE couplet completely merged into one positive ICD band at ∼272 nm (curves d and e in Figure 6), and later on, the positive CE couplet at ∼270 nm (the crossover point) in the region 257-288 nm drew its profile upon the Zn2+ solutions titrations (curves f-j in Figure 6). The reversals of ICD signals in both Figures 5 and 6 strongly indicated the conformational change of the mnt chromophoric ligand upon titration with Zn2+ solutions in the

presence of β-cyclodextrin or HP-β-cyclodextrin. To detail the mechanism of the ICD signal reversal in our experiments, we carried out the DFT calculations as well. As shown in Scheme 7, the structure of [Zn(mnt)2]2- was optimized by our DFT calculations to be in the D2d point group, which was consistent with both the crystal data and the quantum chemical results.49,50 Therefore, the dipole moments of the same electrical transition from the two coordinating mnt moieties will interact with each other orthogonally, unless any disturbance was introduced. From the viewpoint of exciton-coupling, this type of transition-dipole-moment-couplet is not able to produce any exciton-coupling effectively and cannot give rise to any splitting CEs in turn.33 According to our TD-DFT calculations (Table 4), transitions 1 and 2 were degenerate as well as transitions 3 and 4, which took place between MOs 86 (85) and MOs 88 (87) at 389 nm (oscillator strength were 0.1080 and 0.1079, respectively). They were referred to the transitions between HOMOs and LUMOs of [Zn(mnt)2]2- and reflected the intensive absorption in UV spectra (at ∼388 nm in Figure 5 and 390 nm in Figure 6). When the MO contour plots involved were taken into account, transitions 1 and 2 almost retained the characteristic of the p f π* transition discovered in inclusion complex 2 (Table 3, vide supra) because the central zinc atom was actually in nonbonding state. Therefore, the electrical transition dipole moments of these two transitions remained intact after the Zn2+ solutions titrations. In another word, the p f π* transitions in [Zn(mnt)2]2- still kept the preference of being parallel to its CdC double bonds. As a matter of fact, our TD-DFT calculations have predicted the directions of transitions 1 and 2, just parallel to the CdC double bond of one mnt moiety or of the other (Table 4). Transitions 3 and 4 took place between MOs 85 (86) and MO 89 (which was the nonbonding orbital). They were predicted to have weak absorptions and to be in the direction parallel to the mnt CdC double bonds (Table 4). Because they were not intensive and somewhat overlapped by transitions 1 and 2, the absorptions of transitions 3 and 4 were hardly located in either Figure 5 or 6, where merely some absorption shoulder profiles were recorded at ∼350 nm. Transitions 5 and 6 were degenerate, in fact, but differed slightly in transition energy (Table 4). On the basis of our calculations, they were much intensive (oscillator strengths were 0.2252 and 0.1245, respectively) and reflected strong absorptions at around 271 or 270 nm in UV spectra (Figures 5 and 6). The directions of these two transition dipole moments were predicted to be perpendicular to the CdC double bond of the mnt moieties. Now, let us turn back to the elucidation of the ICD signal reversal in our experiments, whereas the detailed TD-DFT calculation results were taken into consideration.

7238


Cheng et al.

TABLE 4: Calculated and Experimental Transition Absorption of [Zn(mnt)2]2-

a

Cartesian axes are illustrated in Scheme 7, and the three digitals are components of the transition dipole moment on X, Y, and Z axes, respectively. b Oscillator strength was predicted by TD-DFT calculations at the optimized geometry. c Absorption maximum of unincluded [Zn(mnt)2]2-. d Absorption maximum of [Zn(mnt)2]2- in inclusion complex [Zn(mnt)2]2-@β-cyclodextrin. e Absorption maximum of [Zn(mnt)2]2- in inclusion complex [Zn(mnt)2]2-@HP-β-cyclodextrin.

Keeping in mind the fact that the mnt chromophore experimentally produces a single positive ICD band when included by β-cyclodextrin, we claim that the guest (mnt) adopts an axial alignment inside the host cavity where its CdC double bond is parallel to the symmetric axis of β-cyclodextrin. It is supported by our TD-DFT calculations when the ICD signals are subject to Harata’s rule. In the meantime, the exciton-coupling mechanism is put forward to explain the splitting CEs in the case of inclusion complex 2. However, the mnt chromophore is not able to keep the same alignment inside β-cyclodextrin cavity as the free mnt ligand does when it coordinates to Zn2+ to form a coordination anion. Because of the spatial complement between the host and guest, the [Zn(mnt)2]2- anion tends to form an inclusion complex with β-cyclodextrin at a 1:2 stoichiometry where the CdC double bonds of mnt ligands are perpendicular to the symmetric axis of the host (unpublished results). Therefore, the p f π* transition of mnt chromophore completely reversed its ICD signals upon titration with Zn2+ solutions in the case of inclusion complex Na2mnt@β-cyclodextrin (ICD curves in Figure 5). Because the two coordinating mnt ligands were orthogonal in the [Zn(mnt)2]2- anion, there was no excitoncoupling revealed. Therefore, the calculated transitions 5 and 6 (Table 4) produced a positive ICD band at ∼274 nm because

their predicted transition dipole moments were parallel to the symmetric axis of β-cyclodextrin (Figure 5). In the same way, transitions 3 and 4 (Table 4) gave rise to negative ICD bands at ∼342 nm, which were shadowed by the p f π* transition nearby (Figure 5). As illustrated in Scheme 2, we hypothesized above that the individual HP-β-cyclodextrin monomers tended to integrate with each other to form HP-β-cyclodextrin dimers via the strong interactions between the hydroxypropyl groups on the secondary rims of parent β-cyclodextrins. Upon inclusion complexation with Na2mnt, HP-β-cyclodextrin dimers induced the experimentally observed splitting CEs (Figure 2) of the guest. If the dimeric hosts are able to include the coordination anion [Zn(mnt)2]2- in the same manner as parent β-cyclodextrins do in the case of [Zn(mnt)2]2-@β-cyclodextrin, the mnt ligands will undergo conformational changes and exhibit ICD signals reversals. Although under such conditions the two mnt ligands are orthogonal in [Zn(mnt)2]2- anion as well; the dimeric HPβ-cyclodextrin can still efficiently induce certain excitoncoupling effects because of the unevenness of their hydroxypropyl groups’ substitution. That means that in the case of inclusion complex 2, most likely the splitting CEs or even the reversed splitting CEs are to be experimentally recorded upon

Watching the Conformational Changes of Na2mnt the titrations with Zn2+ solutions. In fact, one reversal (at ∼270 nm in the region 257-288 nm) from negative to positive splitting CEs was evidently observed (Figure 6). This positive splitting CE was assigned to transitions 5 and 6 in Table 4, which were predicted by our TD-DFT calculations to be perpendicular to the symmetric axis of β-cyclodextrin cavity in that of inclusion complex Na2mnt@β-cyclodextrin and parallel to the symmetric axis of β-cyclodextrin cavity in that of inclusion complex [Zn(mnt)2]2-@β-cyclodextrin. Therefore, experimentally, we observed the conformational changes of mnt chromophores during the formation of [Zn(mnt)2]2- and its subsequent inclusion complexation with β-cyclodextrin or HPβ-cyclodextrin. That is the solid proof to verify our hypotheses above on the alignments of the mnt chromophoric guests inside cyclodextrins cavities. Because of the lack of details about the unevenness of HPβ-cyclodextrin and the dimeric host structure, the weak ICD peak at ∼339 nm and the unreversed positive splitting CE at ∼386 nm in the region 373-400 nm (Figure 6) still remain unknown. According to the exciton-coupling theory,33 it probably results from the relative orientation between the coupled transition dipole moments. The subsequent work is in process in our lab now. 4. Conclusions In summary, the inclusion complexes of Na2mnt with different cyclodextrins were investigated in detail. According to the electronic spectra and ICD spectra, inclusion complexes 2 and 3 were identified, whereas R-cyclodextrin could not include Na2mnt efficiently as well. Therefore, spatial complement between the host and guest seemed a decisive factor in the formation of the cyclodextrin inclusion complexes.35 The p f π* transition-relevant ICD activities of Na2mnt chromophore in inclusion complex 2 was studied for the first time by both ICD spectra and TD-DFT calculations, besides the well-known n f π* and π f π* transitions in literatures. Our TD-DFT calculations indicated the directions of the transition dipole moments in Na2mnt chromophore and well predicted the theoretical absorptions in agreement with the experimental data. It predicted positive ICD signals at 378 nm and negative ICD signals at 277 nm, which were experimentally evidenced to coincide with the ICD couplets recorded in the case of inclusion complex 2 when incorporating both Harata’s rule and excitoncoupling mechanism. Therefore, the HP-β-cyclodextrin dimers via dynamical hydrogen bonding networks in aqueous solution were disclosed, which in turn generated the exciton coupling between the included guests and experimentally induced the splitting CEs. Upon titrations with the increasing concentrations of Zn2+ solutions, the reversals of ICD peak and splitting CE were observed in the cases of Na2mnt@β-cyclodextrin and Na2mnt@HP-β-cyclodextrin, respectively, which strongly supported our hypotheses on the axial coconformation and the conformational change occurring in the mnt inclusion complexes. In addition, our TD-DFT methods were proven to be feasible in assignment and interpretation of the ICD signals, and Harata’s rule was successfully accessed to the treatment of p f π* transition in Na2mnt chromophore. 5. Experimental Section All cyclodextrins were purchased from Seebio Biotechnology Inc. (P. R. China) (HP-β-cyclodextrin has a 70% substitution degree, about five hydroxyl groups randomly substituted by hydroxypropyl groups per molecule of cyclodextrin at its secondary rim), and the reagents employed were commercially

J. Phys. Chem. A, Vol. 114, No. 26, 2010 7239 available and used as received without further purification. The electronic spectra were recorded on a Shimadzu UV-3100 spectrometer in 500-240 nm region with fast scan rate. The ICD measurements were performed on a JASCO J-810 circular dichroism spectrometer in the 500-240 nm region with a rate of 100 nm/min. 5.1. Preparation of the Inclusion Complexes in Solution. All inclusion complexes were prepared in solution by similar methods. The synthesis of 2-butene-dinitrile-2,3-dimercapto disodium salt (Na2mnt) was according to the method described in the literature.27 Na2mnt was dissolved to achieve 1.61 × 10-4 M aqueous solution and was then titrated with the increasing concentrations of cyclodextrins (cyclodextrins ) R-, γ-cyclodextrin, or HP-β-cyclodextrin) solutions. [R-cyclodextrin] ) 0, 0.8 × 10-4, 1.61 × 10-4, 2.42 × 10-4, 3.22 × 10-4, 4.83 × 10-4, 9.66 × 10-4, 14.49 × 10-4, 19.32 × 10-4, 28.98 × 10-4 M (a f j); [γ-cyclodextrin] ) 0, 0.8 × 10-4, 1.61 × 10-4, 2.42 × 10-4, 3.22 × 10-4, 4.83 × 10-4, 14.49 × 10-4, 19.32 × 10-4, 28.98 × 10-4 38.64 × 10-4, 43.47 × 10-4, and 57.96 × 10-4 M (a f l); [HP-β-cyclodextrin] ) 0, 0.8 × 10-4, 1.61 × 10-4, 2.42 × 10-4, 3.22 × 10-4, 4.83 × 10-4, 9.66 × 10-4, 14.49 × 10-4, 19.32 × 10-4, 28.98 × 10-4, 32.2 × 10-4, 64.4 × 10-4, and 128.8 × 10-4 M (a f m). The ratio of R-cyclodextrin to Na2mnt is set as below: 0:1, 0.5:1, 1:1, 1.5:1, 2:1, 3:1, 6:1, 9:1, 12:1, and 18:1. The ratio of γ-cyclodextrin to Na2mnt is set as below: 0:1, 0.5:1, 1:1, 1.5:1, 2:1, 3:1, 6:1, 12: 1, 18:1; 24:1, 27:1, and 36:1. The ratio of HP-β-cyclodextrin to Na2mnt is set as below: 0:1, 0.5:1, 1:1, 1.5:1, 2:1, 3:1, 6:1, 9:1, 12:1, 18:1, 20:1, 40:1, and 80:1. The titration experiments of Na2mnt with ZnCl2 solutions in the presence of cyclodextrins were performed as below: 1.61 × 10-4 M aqueous solution of Na2mnt was mixed with fourfold excessive cyclodextrins (cyclodextrins ) β-cyclodextrin or HP-β-cyclodextrin) and then titrated with the increasing concentrations of ZnCl2 solutions. [ZnCl2] ) 0, 0.8 × 10-5, 1.61 × 10-5, 2.42 × 10-5, 3.22 × 10-5, 4.83 × 10-5, 6.44 × 10-5, 8.05 × 10-5, 1.61 × 10-4, and 6.44 × 10-4 M (a f j), and the ratio of [ZnCl2] to [Na2mnt] is set as below: 0:1, 0.05: 1, 0.1:1, 0.15:1, 0.2:1, 0.3:1, 0.4:1, 0.5:1, 1:1, and 4:1. 5.2. Computational Methods. We carried out DFT calculations employing B3LYP functional with the (6-31G+d,p) basis set by using the Gaussian 03 program,28 and molecular orbits were visualized by the same package. The geometries of guest molecules (p-hydroxybenzoic acid, azoalkane, mnt2-, and [Zn(mnt)2]2-) were optimized in default-spin mode on the ground states, and the excited states of the guests were predicted using TD-DFT, as implemented in the Gaussian 03 package. Acknowledgment. We thank the National Nature Science Foundation of China (NSFC: 20771056, 20721002), the National Basic Research Program of China (2010CB923402), Jiangsu Science & Technology Department (BK2008266), and the Center of Analysis and Determining of Nanjing University for financial supports. Supporting Information Available: ICD spectra of Na2mnt ([Na2mnt] ) 1.61 × 10-4 M) in water and in the presence of increasing concentrations of γ-cyclodextrin. This material is available free of charge via the Internet at http://pubs.acs.org. References and Notes (1) Szejtli, J. Cyclodextrins and Their Inclusion Complexes; Akademiai Kiado: Budapest, Hungary, 1982. (2) Bender, M. L.; Komiyama, M. Cyclodextrin Chemistry; Springer: Berlin, 1978.

7240


(3) D’Souza, V. T.; Lipkowitz, K. B. Chem. ReV. 1998, 98, 1741– 1742. (4) Udachin, K. A.; Ripmeester, J. A. J. Am. Chem. Soc. 1998, 120, 1080–1081. (5) Liu, Y.; Zhao, Y. L.; Zhang, H. Y.; Yang, E. C.; Guan, X. D. J. Org. Chem. 2004, 69, 3383–3390. (6) Clark, J. L.; Stezowski, J. J. J. Am. Chem. Soc. 2001, 123, 9880– 9888. (7) Clark, J. L.; Booth, B. R.; Stezowski, J. J. J. Am. Chem. Soc. 2001, 123, 9889–9895. (8) Monti, S.; Sortino, S.; Guidi, G. D.; Marconi, G. New J. Chem. 1998, 22, 599–604. (9) Liu, Y.; Zhao, Y. L.; Zhang, H. Y.; Li, X. Y.; Liang, P.; Zhang, X. Z.; Xu, J. J. Macromolecules 2004, 37, 6362–6369. (10) Liu, Y.; Zhao, Y. L.; Zhang, H. Y.; Song, H. B. Angew. Chem., Int. Ed. 2003, 42, 3260–3263. (11) Fu, H.; Zhou, Y. H.; Chen, W. L.; Deqing, Z. G.; Tong, M. L.; Ji, L. N.; Mao, Z. W. J. Am. Chem. Soc. 2006, 128, 4924–4925. (12) (a) Ogino, H. J. Am. Chem. Soc. 1981, 103, 1303–1304. (b) Wylie, R. S.; Macartney, D. H. J. Am. Chem. Soc. 1992, 114, 3136–3138. (13) Zhang, X.; Nau, W. M. Angew. Chem., Int. Ed. 2000, 39, 544– 547. (14) Shimizu, H.; Kaito, A.; Hatano, M. Bull. Chem. Soc. Jpn. 1999, 52, 2678–2684. (15) Lu, C. S.; Ren, X. M.; Hu, C. J.; Zhu, H. Z.; Meng, Q. J. Chem. Pharm. Bull. 2001, 49, 818–821. (16) Harata, K.; Uedaira, H. Bull. Chem. Soc. Jpn. 1975, 48, 375–378. (17) Matsui, Y.; Yokoi, T.; Mochida, K. Chem. Lett. 1976, 5, 1037– 1040. (18) Breslow, R.; Doherty, J. B.; Guillot, G.; Lipsey, C. J. Am. Chem. Soc. 1978, 100, 3227–3229. (19) D’Souza, V. T.; Bender, M. L. Acc. Chem. Res. 1987, 20, 146– 152. (20) Lu, Z. D.; Lu, C. S.; Meng, Q. J. J. Inclusion Phenom. Macrocyclic Chem. 2008, 61, 101–106. (21) Lu, C. S.; Zhang, W. W.; Ren, X. M.; Hu, C. J.; Zhu, H. Z.; Meng, Q. J. J. Chem. Soc., Dalton Trans. 2001, 3052–3055. (22) (a) Mieusset, J. L.; Krois, D.; Pacar, M.; Brecker, L.; Giester, G.; Brinker, U. H. Org. Lett. 2004, 6, 1967–1970. (b) Bakirci, H.; Zhang, X. Y.; Nau, W. M. J. Org. Chem. 2005, 70, 39–46. (c) Peter, W.; Vladimir, B. A.; Udo, H. B. J. Org. Chem. 2006, 71, 3274–3277. (23) Harada, N.; Nakanishi, K. Acc. Chem., Res. 1972, 5, 257–263. (24) Suzuki, M.; Kajtar, M. Carbohydr. Res. 1991, 214, 25–33. (25) Suzuki, M.; Kajtar, M. Carbohydr. Res. 1992, 223, 71–80. (26) Shi, J.; Guo, D. S.; Ding, F.; Liu, Y. Eur. J. Org. Chem. 2009, 923–931. (27) Ba¨hr, G.; Schleitzer, G. Chem. Ber. 1957, 90, 438–443. (28) Frisch, M. J. G.; Trucks, W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven, T.; Kudin, K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz,

Cheng et al. P.; Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.; Gonzalez, C.; Pople, J. A. Gaussian 03, revision B.03; Gaussian, Inc.: Pittsburgh, PA, 2003. (29) (a) Saenger, W. Angew. Chem., Int. Ed. Engl. 1980, 19, 344–362. (b) Wenz, G. Angew. Chem., Int. Ed. Engl. 1994, 33, 803–822. (c) D’Souza, V. T.; Lipkowitz, K. B. Chem. ReV. 1998, 98, 1741–2076. (30) Harata, K.; Uedaira, H. Bull. Chem. Soc. Jpn. 1975, 48, 375–378. (31) Kodaka, M. J. Am. Chem. Soc. 1993, 115, 3702–3705. (32) (a) Schneider, H. J.; Hacket, F.; Ru¨diger, V.; Ikeda, H. Chem. ReV. 1998, 98, 1755–1785. (b) Connors, K. A. Chem. ReV. 1997, 97, 1325– 1357. (33) Allenmark, S. Chirality 2003, 15, 409–422. (34) Yong, C. W.; Clive, W.; William, S. Pharm. Res. 2008, 25, 1092– 1099. (35) Castronuovo, G.; Niccoli, M.; Varriale, L. Tetrahedron 2007, 63, 7047–7052. (36) Stein, T.; Kronik, L.; Baer, R. J. Am. Chem. Soc. 2009, 131, 2818– 2820. (37) (a) Marques, M.; Rubio, A.; Ullrich, C. A.; Burke, K.; Nogueira, F.; Gross, A. Time-Dependent Density Functional Theory; Springer: Berlin, 2006. (b) Burke, K.; Werschnik, J.; Gross, E. K. U. J. Chem. Phys. 2005, 123, 062206. (c) Chelikowsky, J. R.; Kronik, L.; Vasiliev, I. J. Phys.: Condens. Matter 2003, 15, R1517–R1547. (38) Shimizu, H.; Kaito, A.; Hatano, M. Bull. Chem. Soc. Jpn. 1979, 52, 2678–2684. (39) Mayer, B.; Zhang, X.; Nau, W. M.; Marconi, G. J. Am. Chem. Soc. 2001, 123, 5240–5248. (40) Krois, D.; Brinker, U. H. J. Am. Chem. Soc. 1998, 120, 11627– 11632. (41) Zhang, X.; Nau, W. M. Angew. Chem., Int. Ed. 2000, 39, 544– 547. (42) Waters, T.; Wang, X. B.; Yang, X.; Zhang, L. R.; O’hair, A. J.; Wang, L. S.; Wedd, A. G. J. Am. Chem. Soc. 2004, 126, 5119–5129. (43) (a) Harada, N.; Nakanishi, K. In Circular Dichroic Spectroscopy: Exciton Coupling in Organic Stereochemistry; University Science Books: Mill Valley, CA, 1983. (b) Nakanishi, K. ; Berova, N. In Circular Dichroism: Principles and Applications; Nakanishi, K., Berova, N., Woody, R. W., Eds.; VCH Publishers: New York, 1994; pp 361-398. (44) Kikuchi, Y.; Kobayashi, K.; Aoyama, Y. J. Am. Chem. Soc. 1992, 114, 1351–1358. (45) Krois, D.; Brinker, U. H. J. Am. Chem. Soc. 1998, 120, 11627– 11632. (46) Bobek, M. M.; Krois, D.; Brinker, U. H. Org. Lett. 2000, 2, 1999– 2002. (47) Hosoi, S.; Kamiya, M.; Ohta, T. Org. Lett. 2001, 3, 3659–3662. (48) (a) Gao, X. M.; Zhang, Y. L.; Tong, L. H.; Ye, Y. H.; Ma, X. Y.; Liu, W. S.; Inoue, Y. J. Inclusion Phenom. Macrocyclic Chem. 2001, 39, 77–80. (b) Gao, X. M.; Tong, L. H.; Zhang, Y. L.; Hao, A. Y.; Inoue, Y. Tetrahedron Lett. 1999, 40, 969–972. (c) Park, J. W.; Song, H. E.; Lee, S. Y. J. Phys. Chem. B 2002, 106, 5177–5183. (49) Pietro, B.; Giancarlo, M.; Sandra, M.; Bernd, M.; Gottfried, K.; Gottfried, G. Chem.sEur. J. 2000, 6, 1578–1591. (50) Gareth, R.; Lewis, Dance, I. J. Chem. Soc., Dalton Trans. 2000, 3176–3185. (51) Waters, T.; Wang, X. B.; Woo, H. K.; Wang, L. S. Inorg. Chem. 2006, 45, 5841–5851.

JP103118Z

Watching the Conformational Changes of Maleonitriledithiolate

Recommend Documents