π-Electron interaction of PAHs with anion-exchange silica gels modified with anionic metal-porphine and -phthalocyanine derivatives as HPLC stationary phase for preparative column in organic solvents

Talanta 63 (2004) 1035–1038

␲-Electron interaction of PAHs with anion-exchange silica gels modified with anionic metal-porphine and -phthalocyanine derivatives as HPLC stationary phase for preparative column in organic solvents Masaki Mifune a,∗ , Kenjiro Minato a , Yuji Kitamura a , Kimiya Okazaki a , Akimasa Iwado a , Hiromichi Akizawa a , Jun Haginaka b , Noriko Motohashi c , Yutaka Saito a a

b

Faculty of Pharmaceutical Sciences, Okayama University, Tsushima-Naka, Okayama 700-8530, Japan Faculty of Pharmaceutical Sciences, Mukogawa Women’s University, Koshien Kyuban-cho, Nishinomiya 663-8179, Japan c Kobe Pharmaceutical University, Motoyamakita-Machi, Higashinada-Ku, Kobe 658-8558, Japan Received 10 October 2003; received in revised form 8 December 2003; accepted 14 January 2004 Available online 21 February 2004

Abstract To develop easy-to-prepare stationary phases for HPLC, we investigated anion-exchange silica gels, Nucleosil 5SB (Nuc), modified with metal-porphines and -phthalocyanines (M-P). The modified silica gels (M-PN ) were evaluated for the availability as a stationary phase of HPLC for the separation of ␲-electron-rich polyaromatic hydrocarbons (PAHs) in polar and non-polar eluents. Separation ability of silica gels modified with Cu-phthalocyanine derivative (Cu-PCSN ) was comparable to that of the silica gels binding Cu-PCS through sulfonamide bonds; however, the latter requires troublesome procedures for the preparation. The PAHs tested interact with Cu-PCSN in non-polar organic eluents through their ␲-electrons similarly as in the case of the PYE column® , in which interaction with PAHs was reported to be only the ␲–␲-electron interaction. © 2004 Elsevier B.V. All rights reserved. Keywords: HPLC; Stationary phase; Silica gels; Metal-porphyrin; PAH; ␲–␲ Interaction

1. Introduction The high performance liquid chromatography (HPLC) is one of the measures widely used for separating ␲-electron rich compounds including carcinogenic and/or mutagenic compounds in car fumes and aerosols [1–6]. Examples of HPLC columns developed so far include the pyrenylethyl–silica gels column of which function is based on the ␲–␲-electron interaction (␲–␲ interaction) between pyrenylethyl-group and the ␲-electron rich compounds Abbreviations: M-P, metal-porphines and -phthalocyanines; Cu-PCS, Cu-phthalocyanine derivative and Cu-phthalocyanine terasulfonic acid; Cu-PCSresin , anion-exchange resin modified with Cu-PCS; TSPP, tetrakis (sulfophenyl)porphine; TCPP, tetrakis(carboxyphenyl)porphine; PP, protoporphyrin; Nuc, anion-exchange silica gels (Nucleosil 5SB); M-PN , Nuc modified with M-P, wherein the subscript “N” indicates Nuc modified with M-P ∗ Corresponding author. Tel.: +81-86-251-7952; fax: +81-86-251-7953. E-mail address: [email protected] (M. Mifune). 0039-9140/$ – see front matter © 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.talanta.2004.01.007

[7]. The silica gels binding metal-phthalocyanine [8–10] or -porphine derivatives (M-P) [11,12] also interacts with ␲-electron rich compounds including PAHs through the ␲-electrons. Although these stationary phases are useful in the separation of ␲-electron-rich compounds, the preparation thereof requires troublesome procedures. Silica gels modified with M-P were also reported to be useful only for HPLC separation of fullerenes in ␲-electron rich eluents such as toluene [13,14] and for an electrochromatography of peptides in water [15]. On the other hand, we reported that anion-exchange resins modified with Cu-phthalocyanine tetrasulfonate (Cu-PCSresin ) exhibits some ␲–␲ interaction with PAHs in aqueous methanol [16,17]. The Cu-PCSresin , however, was not necessarily suited as a stationary phase because the resins might undergo fusion and swelling by organic eluents except for extremely polar solvents. Thus, to develop HPLC-stationary phases targeting PAHs, which can be easily prepared and useful in organic eluents for the preparative HPLC column, we prepared anion-exchange silica gels modified with anionic M-P.

1036

M. Mifune et al. / Talanta 63 (2004) 1035–1038

2. Experimental 2.1. Reagents and chemicals Tetrakis(sulfophenyl)porphine (H2 -TSPP), tetrakis(carboxyphenyl)porphine (H2 -TCPP) and protoporphyrin (H2 -PP) were purchased from commercial sources (Tokyo Kasei, Kanto Kagaku or Wako Junyaku, Japan). Metal-tetrakis(sulfophenyl)porphines (M-TSPP, M = Cu2+ , Zn2+ , Mn3+ , Co3+ ), copper(II)-tetrakis(carboxyphenyl)porphine (CuTCPP) and copper(II)-protoporphyrin (Cu-PP) were prepared from H2 -TSPP, -TCPP and -PP as described in the previous paper [9]. Copper(II)-phthalocyanine tetrasulfonic acid (Cu-PCS) was prepared from 4-sulfophthalic acid, copper(II)chloride dihydrate, urea and ammonium chloride in nitrobenzene by following the methods described in the literature [18]. Co-PCS was prepared in a similar manner as above by using cobalt(III) chloride hexahydrate. Anion-exchange silica gels, Nucleosil 5SB (porous silica gels modified with quaternary ammonium groups, 5 ␮m, Nuc), were purchased from Macherey–Nargel (Germany) and washed with acetone and dried up. ␲-Electron-rich PAHs, used as samples, were of reagent grade (Nacalai Tesque or Wako Junyaku, Japan). Sample solutions were 250 ␮g ml−1 benzene and 5-20 ␮g ml−1 PAHs solutions in methanol or in n-hexane. The PYE column® (4.6 mm × 150 mm) was purchased from Nacalai Tesque (Japan).

Fig. 1. Chromatograms of polyaromatic hydrocarbons on the M-PN columns in methanol. (1) benzene, (2) naphthalene (3) anthracene, (4) phenanthrene, and (5) pyrene; eluent: methanol, flow rate; 0.5 ml/min.

2.2. Preparation of Nucleosil 5SB modified with M-P (M-PN ) 3. Result and discussion The preparation of the Nuc modified with metal-porphyrins (M-PN ) was conducted by adding dry Nuc (4.0 g) to an aqueous solution of M-P (1.0 mmol l−1 , 100 ml), followed by incubation for 10 h at 35 ◦ C to obtain the clear mixture. All the resultant modified resins were stable to temperature, moisture and storage, and none of metal-porphyrins on Nuc was eluted with water or organic solvents tested at all. The M-PN s were packed separately into stainless-steel columns (4.0 mm × 150 mm and 4.6 mm × 150 mm). As a control, Nuc was packed into a stainless-steel column (4.0 mm × 150 mm). 2.3. Apparatus An HPLC system was assembled with a pump (Shimadzu LC-6A, Japan) a sample injector with a 20 ␮l fixed sample loop (Rheodyne model 7125, USA) and a UV/VIS detector (Shimadzu SPD-6AV) with a recorder (Shimadzu Chromatopack C-R6A). As mobile phases, methanol, 2-propanol, acetonitrile, ethylacetate, methylenechloride, chloroform, chloroform-n-hexane mixtures and n-hexane were used at a flow rate of 0.5 ml min−1 . The eluate was monitored at 254 nm.

To estimate the property of the M-PN columns in the separation of ␲-electron-rich PAHs, the interaction between a sample and M-PN was mainly evaluated on the basis of the retention time (RT) and also the difference of RT (RT) between the M-PN and the Nuc columns as described in the previous paper [17], so that even small changes can be found with accuracy. The RT values could correlate with the increment of interaction between a sample and M-PN . The abilities of M-PN as an HPLC stationary phase were first evaluated by examining the retention data of M-PN in methanol. Typical chromatograms are shown in Fig. 1. The retention times of the PAHs on the M-PN columns (4.0 mm × 150 mm) are longer than on the Nuc column (4.0 mm × 150 mm) in methanol, suggesting that the M-P on Nuc interacts with the PAHs. The RT and RT values of pyrene increase in the order of Co-, Zn-, Mn-, Fe-, H2 and Cu-TSPPN in the M-TSPPN columns. Furthermore, the Cu-PCSN column retains the PAHs tested more firmly than the Co-PCSN column. Thus, we concluded that the M-Ps containing Cu2+ ion exhibit relatively stronger interaction with the PAHs tested than other M-Ps, probably because of the planarity of Cu-P on the Nuc.

M. Mifune et al. / Talanta 63 (2004) 1035–1038

1037

Table 1 Retention times (RT, min) and RTa of PAHs on the Cu-PN b and Nuc columns in methanol Samples

Retention time (RTa ) Nuc

Cu-TSPPN

Cu-TCPPN

Cu-PPN

Cu-PCSN

Benzene Naphthalene Anthracene Phenanthrene Pyrene

3.18 3.31 3.50 3.52 3.73

3.30 3.57 4.16 4.18 5.33

3.29 3.52 4.07 4.05 4.98

3.21 3.32 3.59 3.57 3.95

3.24 (0.06) 3.68 (0.37) 5.28 (1.78) 5.53 (2.01) 18.30 (14.57)

a b

(0.12) (0.26) (0.66) (0.66) (1.60)

(0.11) (0.21) (0.57) (0.53) (1.25)

(0.03) (0.01) (0.09) (0.05) (0.22)

RT: Difference between RT on the M-PN and Nuc columns. Cu-PN ; 25 ␮mol Cu-P g−1 Nuc, eluent; methanol, flow rate; 0.5 ml min−1 .

To examine the effects of porphyrin-ring, we investigated the retention of the PAHs on the Cu-TCPPN , -PPN , -PCSN and -TSPPN columns. As we expected from the difference of the spread of ␲-electron cloud, the retention time of pyrene increased in the order of the Cu-PPN < Cu-TCPPN = Cu-TSPPN  Cu-PCSN columns (see Table 1). We then decided to investigate mainly the Cu-PCSN column in detail because its interaction is comparable with that of the column packed with the silica gels binding Cu-PCS through sulfonamide bonds [9]. To confirm that Cu-PCS on Nuc interacts with PAHs, the relationship between RT and the bound amount of Cu-PCS on Nuc (refer Fig. 2) was examined. As the bound amount increases from 2.5 to 100 ␮mol g−1 , the RT value increased linearly, although the slope varied depending on the PAH. This linearity suggests that the ␲-electron clouds of the PAHs interact with those of Cu-PCS on the Cu-PCSN column. One of the advantageous aspects of the modification of silica gels with Cu-PCS is the improvement of the abil-

Fig. 2. Effect of Cu-PCS amount on silica gels (Nuc): (䊊) benzene, (䊏) naphthalene, (䉱) anthracene, (䊐) phenanthrene, (䊉) pyrene; eluent: methanol, flow rate; 0.5 ml min−1 .

ity of recognizing the planarity of PAHs, as previously demonstrated [9]. The planarity recognition ability was then checked by using o-terphenyl and triphenylene in pairs. In the case of the Cu-PCSN column, since the RT value of o-terphenyl was nearly equal to the dead volume (t0 ), a calculation of a separation factor (α) could be insignificant. However, comparison of HPLC charts of Cu-PCSN and Cu-PCSresin [17] columns indicates that the modification with Cu-PCS clearly improve the ability of Cu-PCSN column than that of Cu-PCSresin column. As mentioned in the introduction, easy-to-prepare Cu-PCSresin was not completely suited as a stationary phase for a preparative column because organic solvents except for extremely polar solvents fuse and swell ion-exchange resins as the mother supports. Our purposes in this study is to develop easy-to-prepare stationary phases available in organic solvents. Thus, we studied the property of the Cu-PCSN column in the separation of PAHs by using an organic solvent selected from 2-propanol, acetonitrile, methylenechloride, chloroform, n-hexane and their mixtures. Notwithstanding our fear of rapid disappearance of the interaction with decrease of polarity of eluents, sufficient interactions between PAHs and Cu-PCS on Nuc were observed even in non-polar eluents. To examine HPLC parameters in non-polar eluents, we have studied new Cu-PCSN column (4.6 mm × 150 mm) with the same diameter as a commercially available packed column, the PYE column® , of which interaction with un-substituted PAHs is estimated mainly to be the ␲–␲-electron interaction. As shown in Fig. 3, the values of retention factors (k) on the PYE and Cu-PCSN columns correlate linearly with each other (r2 = 0.84). This result suggests that the ␲–␲-electron interaction is dominant on the Cu-PCSN column i.e., the ␲–␲ interaction between PAHs and the Cu-PCS on Nuc is effective for the separation of the PAHs even in a mixture of n-hexane and chloroform (3:1 by volume). The interaction in the above case seems to be the same as that in polar eluents, although further investigations would be necessary for elucidation thereof. The values of theoretical plate numbers of PAHs on the Cu-PCSN column are more than twice of those on the Cu-PCSresin column [17]. In addition, it is of interest that the Cu-PCSN column maintains the planarity recognition ability between o-terphenyl and triphenylene

1038

M. Mifune et al. / Talanta 63 (2004) 1035–1038

Acknowledgements The authors thank Ms. Minako Yamane and Ms. Kurumi Tanaka for their technical supports. This work was supported by a Grant-in-Aid for science research from the Japan Society for the Promotion Science (#14572031).

References

Fig. 3. Correlation diagram of retention factors (k) between the Cu-PCSN and PYE columns. Eluent; a mixture of n-hexane and chloroform (7:3 by volume), flow rate; 0.5 ml min−1 .

(α = 14.1). These results suggest that the Cu-PCSN column have enough ability in non-polar eluent for the preparative column. In conclusion, by the modification with metal-porphyrins, anion-exchange silica gels can be converted easily into functional silica gels which can be used for the separation of PAHs as preparative columns adaptable to polar and non-polar organic eluents. We hope that the present modified silica gels, M-PN , is useful as a stationary phase for the separation of ␲-electron rich compounds such as pharmaceuticals, mutagens and pollutants.

[1] N. Kishikawa, M. Wada, N. Kuroda, S. Akiyama, K. Nakashima, J. Chromatogr. B 789 (2003) 257–264. [2] W. Brack, T. Kind, H. Hollert, S. Schrader, M. Möder, J. Chromatogr. A 986 (2003) 55–66. [3] C. Mao, S.A. Tucker, J. Chromatogr. A 966 (2002) 53–61. [4] A.-J. Madoulet, A. Abarnou, A.-M. Le Guellec, V. Loizeau, F. Leboulenger, J. Chromatogr. A 886 (2000) 153–173. [5] S. Moret, L. Conte, J. Chromatogr. A 882 (2000) 245–253. [6] G. Boysen, S.S. Hecht, Mut. Res. 543 (2003) 17–30. [7] N. Tanaka, Y. Tokuda, K. Iwaguchi, M. Araki, J. Chromatogr. 239 (1982) 761–772. [8] Y. Saito, M. Mifune, J. Odo, Y. Otsuki, M. Mitsuhashi, Y. Mori, A.E.H. Gassim, J. Haginaka, Anal. Sci. 7 (1991) 805–806. [9] M. Mifune, Y. Shimomura, Y. Saito, Y. Mori, M. Onoda, A. Iwado, N. Motohashi, J. Haginaka, Bull. Chem. Soc. Jpn. 71 (1998) 1825– 1829. [10] M. Mifune, Y. Mori, M. Onoda, A. Iwado, N. Motohashi, J. Haginaka, Y. Saito, Anal. Sci. 14 (1998) 1127–1131. [11] C.E. Kibbey, M.E. Meyerhoff, J. Chromatogr. 641 (1993) 49–55. [12] S. Chen, U. Ruedel, M.E. Meyerhoff, J. Chromatogr. A 859 (1999) 121–132. [13] J. Xiao, M.E. Meyerhoff, J. Chromatogr. A 715 (1995) 19–29. [14] T. Takeuchi, J. Chu, T. Miwa, Analusis 24 (1996) 271–274. [15] J. Charvátová, V. Kašiˇcka, T. Barth, Z. Deyl, I. Mikš´ık, V. Král, J. Chromatogr. A 1009 (2003) 73–80. [16] M. Mifune, T. Kanai, M. Urabe, A. Iwado, J. Haginaka, N. Motohashi, Y. Saito, Anal. Sci. 15 (1999) 581–584. [17] M. Mifune, A. Iwado, K. Okazaki, H. Akizawa, J. Haginaka, N. Motohashi, Y. Saito, Anal. Sci. 16 (2000) 177–179. [18] J.H. Weber, D.H. Busch, Inorg. Chem. 4 (1965) 469–471.