It is well-known that cyclodextrins (CyD's) form inclusion complexes with alcohols (guests) in an aqueous medium. Such complex formations have been studied by various techniques, which involve NMR 1 , calorimetric [2][3][4][5][6][7] , spectrophotometric 7-9 , and ultrasonic relaxation 10 methods. These studies, however, were exclusively concerned with α-and β-CyD's. To our knowledge, no data have been reported for γ -CyD. nAlcohols associate with α-CyD more strongly than with β-CyD. These guests are closely fitted to the relatively small α-CyD cavity, whereas they are too small in size to be in close contact with the relatively large cavity of β-CyD. 8 Then, it is expected that n-alcohols should associate with γ -CyD more weakly than with β-CyD, because the cavity size of γ -CyD is still larger than β-CyD.Despite weak association of n-alcohols with γ -CyD, these guests play an important role in complexations of γ -CyD with a certain guest of relatively large molecular size, such as pyrene.11 Accurate knowledge of alcohol-CyD association is also needed in the reversedphase LC, where CyD's have been used as a modifier to improve separation by their molecular recognizing ability. 12 Headspace GC is a useful technique to study partition of volatile solutes between gaseous and aqueous phases. This technique is of two types: one is "static headspace", in which the vapor phase is directly injected into a GC column and the other is "enriched head space", which requires trapping of volatile components onto an adsorbent prior to GC injection. The aim of the present work is to demonstrate that static headspace GC is simple and suitable for determining association constants of alcohol-CyD in aqueous medium. This report is the first to present the association constants of alcohols with γ -CyD.
Experimentaln-Alcohols (C 4 to C 9 ) and 2-pentanol (Wako Pure Chemical Co.) of the highest commercially available purity were used without further purification. α-, β-, and γ -CyD's of guaranteed grade (Nacalai Tesque Co.), dried over phosphorus pentaoxide under vacuum, were used as received. Deionized distilled water was used throughout the experiments.A known initial concentration of alcohol solution of 20 ml was prepared in a bottle of 27-ml capacity in the absence and presence of CyD of various concentrations. The bottle was sealed with a silicone rubber septum cap with PTFE liner facing toward the bottle headspace. At least three bottles were used for a single sample solution. The bottles were immersed in a 25.0±0.1˚C water bath for one day with occasional gentle swirling. After equilibration, vapor samples (200 µl) were withdrawn from the headspace of the sealed bottles with a gas-tight syringe (Shimadzu HSS-2B) and directly injected into a Shimadzu GC-14A gas chromatograph. The GC conditions were as follows: a 3.2 mm×2.1 m column packed with SE-30 (60 -80 mesh), a FID detector, and a operating temperature of 120 -160˚C.
Results and DiscussionCalibration curves were prepared in the absence of CyD using alcohol solutions of five ...