No abstract

Recently, many HPLC stationary phases mainly based on π-π electron interaction (π-π interaction) have been developed for the separation of π-electron-rich compounds. Those include silica gels binding Cu 2+ -phthalocyanine 1-3 and In 3+ -porphine 4,5 and pyrene 6,7 derivatives. However, the preparation of such silica gels requires troublesome procedures 2,4,6 and the base silica gels for HPLC are relatively expensive. On the other hand, we have found that anionic metal-porphines are easily fastened on an ordinary anion-exchange resin through electrostatic and hydrophobic interactions. 8-13These facts led us to develop a stationary phase by modifying an anion-exchange resin with anionic Cu 2+ -phthalocyanine derivative (Cu-PCS r ), which would be prepared easily and cheaply. In the present paper, we would like to report briefly the utility of the resultant resin (hereinafter referred to as Cu-PCS r ) as a stationary phase for the separation of π-electron-rich compounds.

We have demonstrated in our previous studies [1][2][3] that commercially available aminopropyl-silica gels for HPLC could be converted into functional silica gels exhibiting the π-π electron interaction by a modification with metalphthalocyanine derivatives. In particular, silica gels modified with the Cu-PCS derivative (Cu-PCSD) 3 have proved to be a superior HPLC stationary phase for the separation of π-electronrich compounds compared to pyrenylethyl-silica gels 4,5 or silica gels binding porphine and metal-porphines. 6,7 However, Cu-PCSD is still unsatisfactory regarding the theoretical plate numbers (N) for some PAHs. 3 In this study, we assumed that the small N values in the previous paper were caused by an electrostatic interaction 8,9 between the solutes and the polar groups remaining on Cu-PCSD. As an attempt to elucidate the factor(s) responsible for the poor performance of the Cu-PCSD column, we tried to decrease the polarity of the silanol and/or amino groups by endcapping them with trimethylsilyl groups. This treatment resulted in an improvement of not only the N values, but also the efficiency of the separations of PAHs, thereby increasing the performance of the Cu-PCSD column. This improvement would be attributable to a decrease in the electrostatic interaction and also partly to an additional hydrophobic interaction between the trimethylsilyl-groups and the PAHs. The retention behaviors of hetrocyclic amines (HCAs), typical mutagens, 10 on the Cu-PCSD column were also changed after trimethylsilylation. Experimental Reagents Silica gels.As aminopropyl-silica gels, Develosil NH2-5 (Dev, particle size, 5 µm; specific surface area, 350 m 2 /g; average pore size, 120 Å) were purchased from Nomura Chemicals Co. Ltd. and dried at 150˚C for 2 h under a vacuum. Chemicals and reagents.Copper-phthalocyanine (Cu-PC, Fig. 1) was obtained from a commercial source (Kanto Kagaku Co. Ltd.) for the preparation of copper-phthalocyanine tetrasulfonylchloride (Cu-PCSCl, Fig. 1). Methanol used as mobile phases was of special-grade for HPLC. Special-grade dioxane was dried with molecular sieves and used for preparing modified silica gels. Trimethylchlorosilane (TMCS) and Ntrimethylsilylimidazole (TMSI) were purchased from GL Sciences Inc. Samples.π-Electron rich PAHs, benzene, naphthalene, As an attempt to elucidate the factor(s) responsible for the poor performance of a copper(II)-phthalocyanine aminopropylsilica gels (Cu-PCSD) column for HPLC, the silanol and/or amino groups remaining on Cu-PCSD were endcapped with trimethylchlorosilane (TMCS) or N-trimethylsilylimidazole (TMSI). The trimethylsilylated Cu-PCSDs (Cu-PCSD-TMCS and -TMSI) were investigated concerning their performance as an HPLC-stationary phase in the separation of π-electronrich polyaromatic hydrocarbons (PAHs), such as mutagenic anthracene and pyrene. As a result, trimethylsilylation with TMSI, which reacts only with silanol-groups, was not effective to improve the column efficiency. In contrast, trimethylsilylation by TMCS, which reacts with both t...