Anal. Chem. 2002, 74, 1819-1823
In Situ Time-Resolved Fluorescence Spectroscopy in the Frequency Domain in Capillary Electrochromatography Yan He and Lei Geng*
Department of Chemistry, University of Iowa, Iowa City, Iowa 52242
In situ time-resolved fluorescence spectroscopy for capillary electrochromatography (CEC) is described in the frequency domain. Fluorescence decay of the solute molecules is collected directly in the packed stationary phase of the CEC capillary. The fluorescence lifetime profile of the solute molecules reveals the microenvironments they experience in the C18 chromatographic interface. A quartz flow cell and experimental optimization of the signal-to-noise ratio are described that enable the collection of high-quality decay data and subsequent calculation of fluorescence lifetime profiles of the solute molecules. The distribution of pyrene (PY), 1-pyrenemethanol (PY-MeOH), and 1-pyrenebutanol (PYBuOH) into the C18 stationary phase and the solute-C18 phase interactions are probed, under separation conditions for CEC. All three molecules display a Gaussian distribution of lifetimes, consistent with an ensemble of heterogeneous microenvironments in the C18 stationary phase. The least polar molecule PY diffuses deeply into and interacts extensively with the C18 phase, experiencing high hydrophobicity and significant heterogeneity of microenvironments. The retention order of PY-MeOH, PY-BuOH, and PY in CEC is determined by their interactions with the stationary phase, revealed by their fluorescence lifetime distributions. Capillary electrochromatography (CEC) combines the features of both capillary electrophoresis (CE) and liquid chromatography (LC) and has been a technique of great interest in chemical separations.1-3 In CEC separations, a high electric field is applied across a packed column to generate the electroosmotic flow (EOF) that drives the movement of the mobile phase. Compared to conventional pressure-driven flow in liquid chromatography, EOF provides a flat flow profile that leads to a great increase in separation efficiency. Since EOF is independent of the particle size and shape of the packing over a wide range, new columns and stationary phases specially designed for CEC have been developed.4-8 However, EOF is sensitive to the surface charge, * Corresponding author: (e-mail)
[email protected]; (tel) (319)-335-3167; (fax) (319)-335-1270. (1) Rathore, A. S.; Horva´th, C. J. Chromatogr., A 1997, 781, 185-195. (2) Steiner, F.; Scherer, B. J. Chromatogr., A 2000, 887, 55-83. (3) Colo´n, L. A.; Burgos, G.; Maloney, T. D.; Cintro´n, J. M.; Rodrı´guez, R. L. Electrophoresis 2000, 21, 3965-3993. (4) Fujimoto, C. Trends Anal. Chem. 1999, 18, 291-301. 10.1021/ac015679j CCC: $22.00 Published on Web 03/12/2002
© 2002 American Chemical Society
charge density, and charge distribution inside the column. The understanding of CEC mechanisms in various conditions is important for the optimization of the separation conditions. Most of the mechanistic studies of CEC to date employ intensity-based methods, such as steady-state fluorescence and UV-visible absorption. Electrochromatograms are obtained using off-column detection methods, and the column efficiency, retention factor, and selectivity can be derived from the retention time, peak shape, and peak dispersion.9-12 Although this indirect information has greatly enhanced our knowledge of CEC systems, direct observation of the interaction between the solute molecules and the stationary phase can yield detailed information on the processes that control the chemical separation. The retention mechanisms of HPLC stationary phases have been investigated in analytical chemistry for over twenty years.13-15 Due to their ability to provide dynamic and structural information on the molecular level, a wide variety of spectroscopic techniques, such as IR,16,17 NMR,18-20 Raman,21-23 and fluorescence,24-35 have (5) Zou, H. F.; Ye, M. L. Electrophoresis 2000, 21, 4073-4095. (6) Svec, F.; Peters, E. C.; Sykora, D.; Frechet, J. M. J. J. Chromatogr., A 2000, 887, 3-29. (7) Pesek, J. J.; Matyska, M. T. J. Chromatogr., A 2000, 887, 31-41. (8) Pursch M.; Sander L. C. J. Chromatogr., A 2000, 887, 313-326. (9) Smith, N.; Evans, M. B. J. Chromatogr., A 1999, 832, 41-54. (10) Moffatt, F.; Cooper, P. A.; Jessop, K. M. J. Chromatogr., A 1999, 855, 215226. (11) Gusev, I.; Huang, X.; Horva´th, C. J. Chromatogr., A 1999, 855, 273-290. (12) Wen, E.; Asiaie, R.; Horva´th, C. J. Chromatogr., A 1999, 855, 349-366. (13) Rutan, S. C.; Harris, J. M. J. Chromatogr., A 1993, 656, 197-215. (14) Tchapla, A.; He´ron, S.; Lesellier, E. J. Chromatogr., A 1993, 656, 81-112. (15) Vailaya, A.; Horva´th, C. J. Chromatogr., A 1998, 829, 1-27. (16) Rivera, D.; Poston, P. E.; Uibel, R. H.; Harris, J. M. Anal. Chem. 2000, 72, 1543-1554. (17) Sander, L. C.; Calls, J. B.; Field, L. R. Anal. Chem. 1983, 55, 1068-1075. (18) Kelusky, E. C.; Fyfe, C. A. J. Am. Chem. Soc. 1986, 108, 1746-1749. (19) Pursch, M.; Sander, L. C.; Albert, K. Anal. Chem. 1999, 71, 733A-741A. (20) Pursch, M.; Sander, L. C.; Egelhaaf, H.-J.; Raitza, M.; Wise, S. A.; Oelkrug, D.; Albert, K. J. Am. Chem. Soc. 1999, 121, 3201-3213. (21) Bryant, M. A.; Pemberton, J. E. J. Am. Chem. Soc. 1991, 113, 3629-3637. (22) Ho, M.; Pemberton, J. E. Anal. Chem. 1998, 70, 4915-4920. (23) Ho, M.; Cai, M.; Pemberton, J. E. Anal. Chem. 1997, 69, 2613-2616. (24) Lochmuller, C. H.; Colborn, A. S.; Hunnicutt, M. L.; Harris, J. M. J. Am. Chem. Soc. 1984, 106, 4077-4082. (25) Carr, J. W.; Harris, J. M. Anal. Chem. 1987, 59, 2546-2550. (26) Burns, J. W.; Bialkowski, S. E.; Marshall, D. B. Anal. Chem. 1997, 69, 38613870. (27) Kovaleski, J. M.; Wirth, M. J. J. Phys. Chem. 1996, 100, 10304-10309. (28) Wang, H.; Harris, J. M. J. Am. Chem. Soc. 1994, 116, 5754-5761. (29) Wang, H.; Harris, J. M. J. Phys. Chem. 1995, 99, 16999-17009. (30) Wirth, M. J.; Burbage, J. D. Anal. Chem. 1991, 63, 1311-1317. (31) Hansen, R. L.; Harris, J. M. Anal. Chem. 1995, 67, 492-498. (32) Wirth, M. J.; Swinton, D. J. Anal. Chem. 1998, 70, 5264-5271.
Analytical Chemistry, Vol. 74, No. 8, April 15, 2002 1819
been utilized to characterize the physical and chemical nature of LC stationary phases as well as their interactions with solutes. Because the photophysical properties of the fluorescent molecules are highly sensitive to the physicochemical environments, fluorescence spectroscopic techniques have long been used to study the structural aspects of chromatographic surfaces and RPLC retention. For example, excimer formation of surface-immobilized pyrene has been used to show that RPLC surfaces are heterogeneous in nature, their chain organizations vary substantially with different overlaying solvents,24 and these immobilized chains could undergo slow rearrangement through lateral diffusion on the surface.28 Single-molecule fluorescence spectroscopy proved the presence of both strong and weak adsorption sites on the RPLC surfaces32 and that the unblocked surface silanols are the origin of strong adsorption sites.33 The electrostatic interactions between protein molecules and the silica surface have been probed with single-molecule imaging.34 Fluorescence quenching studies25,26 showed that solute molecules can penetrate deep into the surfacebound C18 chains, locations that are inaccessible to solvent molecules. The penetration capability is dependent on the shape of the solute molecules and the mobile-phase compositions. All these studies indicate the complexity of the RPLC retention mechanism. Fluorescence spectroscopic studies have not been used to study the structure and dynamics of RPLC stationary phases under CEC conditions, where the electric field provides the driving force for the separation. On-the-fly fluorescence lifetime detection in both time36-38 and frequency39-41 domain has been actively pursued in CE separations, as an analyte-specific detector. Pyrene and its derivatives are a family of UV fluorescence probes with long fluorescence lifetimes. They are sensitive to the polarity change of surrounding microenvironments and have been widely used to study RPLC bonded phases.14,24-26,28,29,35 In this paper, we describe the experimental setup for studying solute-stationary-phase interactions in CEC using kinetic time-resolved fluorescence detection in frequency domain in the UV region. Using a UV laser with relatively low power, a high signal-to-noise ratio is achieved by optimization of optics and the detection volume is increased with a large packing cell. The fluorescence decay information is collected directly at the packed zone, on the bonded silica surface under real separation conditions. The in situ time-resolved fluorescence method presented here provides a new approach to probing molecular processes in chemical separations. EXPERIMENTAL SECTION Chemicals. Pyrene (PY), 1-pyrenebutanol (PY-BuOH), 1-pyrenemethanol (PY-MeOH), sodium tetraborate, and 83% sodium (33) Wirth, M. J.; Ludes, M. D.; Swinton, D. J. Anal. Chem. 1999, 71, 39113917. (34) Xu, N. X.; Yeung, E. S. Science 1998, 281, 1650-1653. (35) He, Y.; Geng, L. Anal. Chem. 2001, 73, 5564-5575. (36) Soper, S. A.; Legendre, B. L., Jr.; Willams, D. C. Anal. Chem. 1995, 67, 4358-4365. (37) Desilets, D. J.; Kissinger, P. T.; Lytle, F. E. Anal. Chem. 1987, 59, 18301834. (38) Lieberwirth, U.; Arden-Jacob, J.; Drexhage, K. H.; Herten, D. P.; Muller, R.; Neumann, M.; Schulz, A.; Siebert, S.; Sagner, G.; Klingel, S.; Sauer, M.; Wolfrum, J. Anal. Chem. 1998, 70, 4771-4779. (39) Li, L.-C.; McGown, L. B. Anal. Chem. 1996, 68, 2737-2743. (40) Wang, G.; Geng, L. Anal. Chem. 2000, 72, 4531-4542. (41) He, Y.; Geng, L. Anal. Chem. 2001, 73, 943-950.
1820
Analytical Chemistry, Vol. 74, No. 8, April 15, 2002
silicate solution were purchased from Aldrich (Milwaukee, WI) and used as received. Spherisorb ODS2 5-µm stationary phase was obtained from Phenomenex (Torrance, CA). It is an endcapped monomeric packing with average surface area of 200 m2/ g, pore size of 80 Å, and C18 phase coverage of 2.72 µmol/m2. LiChrospher Si60 7-µm silica beads were a gift from EM Science (Gibbstown, NJ). Polyimide-coated capillary with 50-µm i.d. and 375-µm o.d. was purchased from Polymicro Technologies (Phoenix, AZ). Acetonitrile (HPLC grade) was obtained from EM Science. All solutions were prepared with water purified with a MilliQ system (MilliQ-Plus, Millipore, Bedford, MA). Buffer solution consisting of 75% CH3CN and 25% 4 mM Na2B4O7 at pH 8.5 was used throughout the experiments. CEC/Time-Resolved Fluorescence Detector Interface. The capillary electrophoresis system was built in-house and interfaced with the sample chamber of the SLM48000 MHF spectrofluorometer (Jobin Yvon, Edison, NJ). The capillary is mounted on an aluminum bracket holder that is attached to a miniaturized threeaxis translation stage fixed inside the sample chamber. Two short pieces of Teflon tubing with 380-µm i.d. glued to the holder serve to guide the capillary through the optical detection zone. A 360µm-o.d. capillary can be inserted directly into the holder. To install a column with the packed quartz cell, which has a larger diameter than the capillary column, the upper portion of the holder can be removed first. One end of the column is inserted into the lower portion of the holder. The upper portion is slid into the other end of the column and fixed back to the holder. The capillary is held at an angle of ∼20° with respect to the incident laser beam to minimize scattering off the capillary walls.42 Because the capillary can be tightly mounted on the holder, this configuration is rigid enough against mechanical vibration noise. The injection end of the capillary and the high-voltage end from a Spellman CZE1000 power supply (Hauppauge, NY) are enclosed in a Teflon block. Fluorescence Lifetime Measurements. Typically, a packed column was conditioned with buffer for 1 h. A solution of pyrene or its derivatives with appropriate concentration was diluted from the stock solution with buffer and then introduced into the packed column with pressure or high electric field. All solutions were degassed with an ultrasonicator for 5 min before injection. The base frequency for MHF was 4.1 or 5.0 MHz, and the correlation frequency was 7 Hz. The excitation is provided by a 325-nm HeCd laser. Fluorescence emission is selected through two 345-nm long-pass filters to remove the scattering. The scattering reference signal from the silica beads was selected with a combination of 325-nm band-pass filter and neutral density filters. The scattered light was used as the lifetime reference for experimental convenience and, more importantly, to reduce the effect of photon migration.43 The excitation photon wave needs to diffuse through the highly scattering medium of silica beads to reach the fluorescent probe molecules. The re-emitted fluorescence photon wave will again have to migrate through the packed stationary phase to reach the detection optics. This process of photon migration introduces an additional phase delay in the frequency domain measurements that complicates data analysis. When the scattered light from the beads is used as the fluorescence lifetime reference, frequency domain measurements could be reliably (42) Lee, T. T.; Yeung, E. S. J. Chromatogr., A 1992, 595, 319-325. (43) Hutchinson, C. L.; Troy, T. L.; Sevick-Muraca, E. M. Appl. Opt. 1996, 35, 2325-2332.
made.43 The color effect of the PMT, which could be reduced by using reference fluorophores, will remain; but for highly scattering media such as the column packed with silica beads in our experiments, it is imperative and more important to remove the effects of photon migration in fluorescence lifetime measurements. If not mentioned otherwise, each reference file was an average of 100 measurements or 14.3 s of data and each sample measurement was collected for 300 s. Since the kinetic data are stored directly in the time domain, several in-house programs written in C++ were used to extract the intensity profile, the time domain decay profile, and the frequency domain decay profile. The phase and demodulation data were then subject to analysis using a nonlinear least-squares method (NLLS), as described previously.35 The lifetime profile was best fit with a Gaussian distribution for all conditions. A minor component (
