A Cyclodextrins

Conradi, S. Vogt, C. Rohde, E. Separation of Enatiomeric Barbiturates by Capillary Electrophoresis Using a Cyclodextrin-Containing Run Buffer, /. Chem. Educ. 1997, 74, 1122-1125. 

Fig. 8. A hydrophobic inclusion complex between a chiral analyte and a cyclodextrin.

Fig. 15. Prototype examples of (a) cyclodextrins and (b) calixarenes, showing conformational stmctures and dimensions.

Several procedures are used to control the ratios of cyclodextrins produced. One is addition of a substance to the reaction mixture that can gready affect the formation of one specific cyclodextrin over another. For example, in the presence of 1-decanol and 1-nonanol, a-cyclodextrin is produced almost exclusively whereas hexane or toluene promote the production of P-cyclodextrin. Conversely both cyclodextrins are produced simultaneously in the presence of 1-heptanol (2,4). 

Immobilization. The abiUty of cyclodextrins to form inclusion complexes selectively with a wide variety of guest molecules or ions is well known (1,2) (see INCLUSION COMPOUNDS). Cyclodextrins immobilized on appropriate supports are used in high performance Hquid chromatography (hplc) to separate optical isomers. Immobilization of cyclodextrin on a soHd support offers several advantages over use as a mobile-phase modifier. For example, as a mobile-phase additive, P-cyclodextrin has a relatively low solubiUty. The cost of y- or a-cyclodextrin is high. Furthermore, when employed in thin-layer chromatography (tic) and hplc, cyclodextrin mobile phases usually produce relatively poor efficiencies. 

As a final example we consider noncovalent molecular complex formation with the macrocyclic ligand a-cyclodextrin, a natural product consisting of six a-D-glucose units linked 1-4 to form a torus whose cavity is capable of including molecules the size of an aromatic ring. Table 4-3 gives some rate constants for this reaction, where L represents the cyclodextrin and S is the substrate ... 

Table 4-3. Binding Constants and Rate Constants for Complex Formation between a-Cyclodextrin and Azo Dyes ...

Figure 5-9. Free energy reaction coordinate diagram for System 2 of Table 4-3, the formation of a cyclodextrin inclusion complex.

Recently, multidimensional GC has been employed in enantioselective analysis by placing a chiral stationary phase such as a cyclodextrin in the second column. Typically, switching valves are used to heart-cut the appropriate portion of the separation from a non-chiral column into a chiral column. Heil et al. used a dual column system consisting of a non-chiral pre-column (30 m X 0.25 mm X 0.38 p.m, PS-268) and a chiral (30 m X 0.32 mm X 0.64 p.m, heptakis(2,3-di-(9-methyl-6-(9-tert-butyldimethylsilyl)-(3-cyclodextrin) (TBDM-CD) analytical column to separate derivatized urinary organic acids that are indicative of metabolic diseases such as short bowel syndrome, phenylketonuria, tyrosinaemia, and others. They used a FID following the pre-column and an ion trap mass-selective detector following the... 

Fig. 1. Chemical structure of a-cyclodextrin. Six glucopyranose units are numbered G1 to G6. The numbers on the G1 glucopyranose refer to those of the carbon atoms...

On the other hand, the values of AH° and AS° for a-cyclodextrin-l-alkanol systems are significantly more negative than those for the corresponding P-cyclOdextrin systems. 1-Alkanols must fit closely into the cavity of a-cyclodextrin, so that the com-plexation is governed by van der Waals interaction rather than by hydrophobic interaction. 

Matsui441 has computed energies (Evdw) due to the van der Waals interaction between a-cyclodextrin and some guest molecules by the use of Hill s potential equation 451 ... 

Fig. 2. Geometries calculated (solid lines) and observed (bold dashed lines) for 1-propanol in its a-cyclodextrin adduct. G3 and G6 denote the numbers of glucopyranose units of a-cyclodextrin. H3 and H5 refer to the hydrogen atoms located inside of the cyclodextrin cavity. The hydrogen atoms for the observed geometry of 1-propanol are not shown, since their atomic coordinates have not been determined. The observed 1-propanol is twofold disordered, with site a occupied 80%, site b 20%. Interatomic distances are shown in bold italics on fine dashed lines (nm). Reproduced with permission from the Chemical Society of Japan...

In this equation, AG°CS is taken to be negligible for p- and y-cyclodextrin systems and to be constant, if there is any, for the a-cyclodextrin system. The AG W term is virtually independent of the kind of guest molecules, though it is dependent on the size of the cyclodextrin cavity. The AG dw term is divided into two terms, AG°,ec and AGs°ter, which correspond to polar (dipole-dipole or dipole-induced dipole) interactions and London dispersion forces, respectively. The former is mainly governed by the electronic factor, the latter by the steric factor, of a guest molecule. Thus, Eq. 2 is converted to Eq. 3 for the complexation of a particular cyclodextrin with a homogeneous series of guest molecules ... 

Ihb = 1, whereas Ihb = 0 when it is inert to hydrogen bonding. Since —AG,° is proportional to log 1/Kd, where Kd is the dissociation constant of a cyclodextrin complex with a guest molecule, we can derive a quantitative structure-reactivity relationship as shown, for example, in Eq. 4 ... 

Figures 4 and 5 show the plots of log 1/Kd vs. log Pe for branched or cyclic alcohol-cyclodextrin systems. Both of the plots showed considerable scatter in contrast to the plots for 1-alkanol systems (solid lines). However, a remarkable trend was found by comparing both plots. Most of the plots for an a-cyclodextrin system (Fig. 4) are located below the straight line due to Eq. 5, whereas those for a P-cyclodextrin system (Fig. 5) are located above the straight line given by Eq. 6. This shows that it is general for a bulky alcohol to associate with a-cyclodextrin less strongly and with P-cyclodextrin more strongly than a rod-like 1-alkanol if the log Pe values are the...

Fig. 4. Plots of log 1 /Kd vs. log Pe for complexes of a-cyclodextrin with branched alkanols (O) and cycloalkanols ( ). The solid line was given by the plots for an a-cyclodextrin-1-alkanol system. Numbers shown refer to the numbers in the first column of Table 2. Reproduced with permission from the Chemical Society of Japan...

These equations show that hydrophobic and steric (van der Waals) interactions are of prime importance in the inclusion processes of cyclodextrin-alcohol systems. The coefficient of Es was positive in sign for an a-cyclodextrin system and negative for a P-cyclodextrin system. These clear-cut differences in sign reflect the fact that a bulky alcohol is subject to van der Waals repulsion by the a-cyclodextrin cavity and to van der Waals attraction by the p-cyclodextrin cavity. 

Silipo and Hansch 77) have developed correlation equations for the formation of a-cyclodextrin-substituted phenyl acetate complexes (Eq. 13), a-cyclodextrin-RCOO complexes (Eq. 14), and P-cyclodextrin-substituted phenylcyanoacetic acid anion complexes (Eq. 15). 

In these equations, MR3 4, MR, and MR4 are the molar refractivities of 3- and 4-substituents, of R-, and of 4-substituents, respectively. All the equations exhibited positive coefficients of the MR terms. This suggests that the dispersion forces of substituents are actually responsible for the binding of ligands to cyclodextrin. Eq. 14 shows that the stability of a-cyclodextrin-RCOO complexes increases linearly up to MR = 4.0 and then falls off linearly. 

Upon formulating these relationships, phenols with branched alkyl substituents were not included in the data of a-cyclodextrin systems, though they were included in (3-cyclodextrin systems. In all the above equations, the n term was statistically significant at the 99.5 % level of confidence, indicating that the hydrophobic interaction plays a decisive role in the complexation of cyclodextrin with phenols. The Ibrnch term was statistically significant at the 99.5% level of confidence for (3-cyclo-dextrin complexes with m- and p-substituted phenols. The stability of the complexes increases with an increasing number of branches in substituents. This was ascribed to the attractive van der Waals interaction due to the close fitness of the branched substituents to the (3-cyclodextrin cavity. The steric effect of substituents was also observed for a-cyclodextrin complexes with p-substituted phenols (Eq. 22). In this case, the B parameter was used in place of Ibmch, since no phenol with a branched... 

The results were simple and clear-cut Only the two terms ofa° and Emin were involved for the a-cyclodextrin systems, and the two terms of k and Emin, for (S-cyclodextrin systems. This means that the stabilities of the inclusion complexes are mainly governed by the electronic and steric interactions in a-cyclodextrin systems and by the hydro-phobic and steric interactions in (i-cyclodextrin systems, regardless of the position of the substituents in the phenols. These observations agree well with those by Harata23), who showed that there is no appreciable difference in thermodynamic parameters between cyclodextrin complexes of m- and p-di substituted benzenes and that the contribution of the enthalpy term to the complexation is more significant in a-cyclodextrin systems than in P-cyclodextrin systems, where the inhibitory effect... 

---