Capillary Electrophoresis Separations on a Planar ... - ACS Publications

Table 2. Reproducibilities of the Migration Times of the Boundary Formed by the Leading and Terminating. Electrolytes for Defined Migration Paths on t...
0 downloads 0 Views 140KB Size
Download PDF
Anal. Chem. 2000, 72, 3596-3604

Capillary Electrophoresis Separations on a Planar Chip with the Column-Coupling Configuration of the Separation Channels Dusˇan Kaniansky,*,† Maria´n Masa´r,† Jana Bielcˇikova´,† Frantisˇek Iva´nyi,† Friedhelm Eisenbeiss,‡ Bernd Stanislawski,‡ Benedikt Grass,§ Andreas Neyer,| and Matthias Jo 1 hnck|

Department of Analytical Chemistry, Faculty of Natural Sciences, Comenius University, Mlynska´ Dolina CH-2, SK-84215 Bratislava, Slovak Republic, Merck KGaA, Frankfurter Strasse 250, D-64293 Darmstadt, Germany, ISAS, Institut fu¨r Spektrochemie und Angewandte Spektroskopie, P.O. Box 101352, D-44013 Dortmund, Germany, and Fachbereich Elektrotechnik, Universita¨t Dortmund, Otto-Hahn Strasse 6, D-44221 Dortmund, Germany

Some basic aspects of capillary electrophoresis (CE) separations on a poly(methyl methacrylate) chip provided with two separation channels in the column-coupling (CC) configuration and on-column conductivity detectors were studied. The CE methods employed in this study included isotachophoresis (ITP), capillary zone electrophoresis (CZE), and CZE with on-line ITP sample pretreatment (ITP-CZE). Hydrodynamic and electroosmotic flows of the solution in the separation compartment of the chip were suppressed, and electrophoresis was a dominant transport process in the separations performed by these methods. Very reproducible migration velocities of the separated constituents were typical under such transport conditions, and consequently, test analytes could be quantified by various ITP techniques with 1-2% RSD. The CC configuration of the separation channels provides means for an effective combination of an enhanced load capacity of the separation system with high detection sensitivities for the analytes in concentration-cascade ITP separations. In this way, for example, succinate, acetate, and benzoate could be separated also in instances when they were present in the loaded sample (1.2 µL) at 1 mmol/L concentrations while their limits of detection ranged from 8 to 12 µmol/L concentrations. A welldefined ITP concentration of the analyte(s) combined with an in-column sample cleanup (via an electrophoretically driven removal of the matrix constituents from the separation compartment) can be integrated into the separations performed on the CC chip. These sample pretreatment capabilities were investigated in ITP-CZE separations of model samples in which nitrite, phosphate, and fluoride (each at a 10 µmol/L concentration) accompanied matrix constituents (sulfate and chloride) at considerably higher concentrations. Here, both the concentration of the analytes and cleanup of the sample were included in the ITP separation in the first separation channel while the second separation channel served for the CZE separation of the ITP pretreated sample and the detection of the analytes. 3596 Analytical Chemistry, Vol. 72, No. 15, August 1, 2000

Electrophoretic methods have been proved to be very promising separation tools in lab-on-a-chip analytical systems.1-19 From recent reviews (see, e.g., refs 11-14) it is apparent that zone electrophoresis (ZE) separations, performed in free (capillary zone electrophoresis) and micellar (micellar electrokinetic chromatography) solutions, are dominantly employed in these systems. So far, the use of other basic electrophoretic methods20 (isotachophoresis and isoelectric focusing) attracted only a limited * Corresponding author: Phone: 421-7-60296 379. Fax: 421-7-65425 360. E-mail: [email protected]. † Comenius University. ‡ Merck KGaA. § ISAS. | Universita ¨t Dortmund. (1) Manz, A.; Harrison, D. J.; Verpoorte, E. M. J.; Fettinger, J. C.; Paulus, A.; Lu ¨ di, H.; Widmer, H. M. J. Chromatogr. 1992, 593, 253. (2) Harrison, D. J.; Manz, A.; Fan, Z.; Lu ¨ di, H.; Widmer, H. M. Anal. Chem. 1992, 64, 1926. (3) Harrison, D. J.; Fluri, K.; Seiler, K.; Fan, Z.; Effenhauser, C. S.; Manz, A. Science 1993, 261, 895. (4) Seiler, K.; Harrison, D. J.; Manz, A. Anal. Chem. 1993, 65, 1481. (5) Jacobson, S. C.; Hergenro¨der, R.; Koutny, L. B.; Ramsey, J. M. Anal. Chem. 1994, 66, 1114. (6) Jacobson, S. C.; Hergenro¨der, R.; Koutny, L. B.; Warmack, R. J.; Ramsey, J. M. Anal. Chem. 1994, 66, 1107. (7) Jacobson, S. C.; Hergenro¨der, R.; Koutny, L. B.; Ramsey, J. M. Anal. Chem. 1994, 66, 2369. (8) Jacobson, S. C.; Hergenro ¨der, R.; Moore, A. W., Jr.; Ramsey, J. M. Anal. Chem. 1994, 66, 4127. (9) Moore, A. W., Jr.; Jacobson, S. C.; Ramsey, J. M. Anal. Chem. 1995, 67, 4184. (10) Jacobson, S. C.; Culbertson, C. T.; Daler, J. E.; Ramsey, J. M. Anal. Chem. 1998, 70, 3476. (11) Jacobson, S. C.; Ramsey, J. M. In Handbook of Capillary Electrophoresis, 2nd ed.; Landers, J. P., Ed.; CRC Press: Boca Raton, FL, 1997; pp 827-39. (12) Effenhauser, C. S.; Bruin, G. J. M.; Paulus, A. Electrophoresis 1997, 18, 2203. (13) Campana, A. M. G.; Baeyens, W. R. G.; AboulEnein, H. Y.; Zhang, X. R. J. Microcolumn Sep. 1998, 10, 339. (14) Effenhauser, C. S. In Microsystem Technology in Chemistry and Life Science; Manz, A., Becker, H., Eds.; Springer: Berlin, 1998; pp 51-82. (15) Walker, P. A.; Morris, M. D.; Burns, M. A.; Johnson, B. N. Anal. Chem. 1998, 70, 3766. (16) Prest, J. E.; Baldock, S. J.; Bektas, N.; Fielden, P. R.; Brown, B. J. T. J. Chromatogr. A 1999, 836, 59. (17) Hofmann, O.; Che, D. P.; Cruickshank, K. A.; Mu ¨ ller, U. R. Anal. Chem. 1999, 71, 678. (18) Mao, Q. L.; Pawliszyn, J. Analyst 1999, 124, 637. (19) Rossier, J. S.; Schwarz, A.; Reymond, F.; Ferrigno, R.; Bianchi, F.; Girault, H. H. Electrophoresis 1999, 20, 727. 10.1021/ac991236s CCC: $19.00

© 2000 American Chemical Society Published on Web 06/30/2000

attention.15-19 A theoretical concept of miniaturization of the capillary electrophoresis (CE) separation system based on proportionality analysis1 is followed in the design of chip-based CE devices. This concept favors a short column (separation channel) of a very small inner diameter (cross section) as high efficiency and fast CE separations are required. For known reasons,21,22 such geometrical dimensions of the separation channel restrict the sample volumes loadable into the separation system and current miniaturized CE devices1-19 accommodate only subnanoliter sample volumes. Therefore, the separations carried out in these devices have to be monitored by very sensitive detection techniques. Laser-induced fluorescence (LIF) meets best sensitivity requirements relevant to the detection of CE analytes, and so far LIF detection has been the favorite detection technique in the chip-based CE separation systems. Recently, efforts aimed at extending the number of detection techniques applicable with these systems are apparent. Papers describing the use of holographic refractive index,23 amperometric,24 conductivity,16 chemiluminescence,25 thermal-lens absorbance,26 and Raman spectroscopy15 detectors document this. Several approaches in coupling the chips to mass spectrometry (see, e.g., refs 27-3027-30) also reflect these efforts. Sensitivities of some of these detection techniques are limited, and therefore, they provide maximum analytical benefits when employed in combinations with specific electrophoretic methods (e.g., Raman spectroscopy detection combined with isotachophoresis (ITP) on the chip15). The use of sample pretreatment procedures (see, e.g., refs 31-35) before the CE separations offers alternative solutions of these detection problems in the microfabricated CE devices. By increase of the volume of the sample taken for the analysis, concentration detectabilities of the CE analytes can be proportionally enhanced. However, this straightforward solution requires the use of the column (separation compartment) of an adequate load capacity to prevent its overloading by matrix constituents.21,22,36-39 The load capacity of the column in CE is determined, mainly, by (20) Vacı´k, J. In Electrophoresis. A Survey of Techniques and Applications, Part A: Techniques; Deyl, Z., Ed.; Elsevier: Amsterdam, 1979; pp 23-37. (21) Mikkers, F. E. P.; Everaerts, F. M.; Verheggen, T. P. E. M. J. Chromatogr. 1979, 169, 1. (22) Reijenga, J. C.; Kenndler, E. J. Chromatogr. A 1994, 659, 403. (23) Burggraf, N.; Krattiger, B.; deMello, A. J.; deRooij, N. F.; Manz, A. Analyst 1998, 123, 1443. (24) Woolley, A. T.; Lao, K. Q.; Glazer, A. N.; Mathies, R. A. Anal. Chem. 1998, 70, 684. (25) Mangru, S. D.; Harrison, D. J. Electrophoresis 1998, 19, 2301. (26) Sato, K.; Kawanishi, H.; Tokeshi, M.; Kitamori, T.; Sawada, T. Anal. Sci. 1999, 15, 525. (27) Xue, Q. F.; Dunayevskiy, Y. M.; Foret, F.; Karger, B. L. Rapid Commun. Mass Spectrom. 1997, 11, 1253. (28) Xue, Q. F.; Foret, F.; Dunayevskiy, Y. M.; Zavracky, P. M.; McGruer, N. E.; Karger, B. L. Anal. Chem. 1997, 69, 426. (29) Ramsey, R. S.; Ramsey, J. M. Anal. Chem. 1997, 69, 1174. (30) Figeys, D.; Aebersold, R. Anal. Chem. 1998, 70, 3721. (31) Jacobson, S. C.; Ramsey, J. M. Electrophoresis 1995, 16, 481. (32) Jacobson, S. C.; Ramsey, J. M. Anal. Chem. 1997, 69, 3212. (33) Khandurina, J.; Jacobson, S. C.; Waters, L. C.; Foote, R. S.; Ramsey, J. M. Anal. Chem. 1999, 71, 1815. (34) Kutter, J. P.; Ramsey, R. S.; Jacobson, S. C.; Ramsey, J. M. J. Microcolumn Sep. 1998, 10, 313. (35) Raymond, D. E.; Manz, A.; Widmer, H. M. Anal. Chem. 1994, 66, 2858. (36) Schoots, A. C.; Verheggen, Th. P. E. M.; De Vries, P. M. J. M.; Everaerts, F. M. Clin. Chem. 1990, 36, 435. (37) Kaniansky, D.; Mara´k, J.; Masa´r, M.; Iva´nyi, F.; Madajova´, V.; Sˇ imunicˇova´, E.; Zelenska´, V. J. Chromatogr. A 1997, 772, 103.

its volume and by the composition of the electrolyte system in which the separation is performed.21,37-39 For known reasons,1 these parameters are changed only within narrow limits in the miniaturized CE systems, and therefore, the use of the columns (separation channels) of enhanced load capacities in the chipbased CE is rare. ITP and CZE separations in long separation channels15,40 and CZE separations by synchronized cyclic capillary electrophoresis41 may serve as examples in which this approach is in certain extents accepted. The column-coupling (CC) configuration of the separation compartment as proposed by Everaerts et al.42 is analytically very beneficial in the CE separations performed in conventional instruments.42-52 This is due, mainly, to the fact that in this instance the CE analysis can be divided into two stages in which specific advantages of electrophoretic methods (e.g., ITP and ZE) are effectively combined. For example, a well-defined concentration of the analyte and sample cleanup integrated into the separation carried out in the first stage can be followed by a sensitive detection of the analyte during the final separation of the pretreated sample in the second stage.44,51 The use of ITP in the first separation stage significantly enhances an overall analytical effect as this electrophoretic method can increase the load capacity of the separation system by a factor of 103 or more in comparison to a current single-column CZE. Consequently, some ITP combinations42-46 and CZE with on-line ITP sample pretreatment48-52 provide very favorable concentration limits of detection (CLOD) for the analytes. The use of the CC configuration of the separation compartment is not restricted to these two-stage CE separations. For example, it has been shown to be suitable to CZE separations with the sample injection by an electric splitting technique.47 Operating in a concentration-cascade mode,43 it offers improved concentration limits of detection for the ITP analytes. A transfer of the column-coupling CE separation system42-52 to a chip format is a subject of our research interest. Recently, a poly(methyl methacrylate) chip providing this CE technology was developed in our laboratories.53 This CC chip is assumed to introduce the above analytical advantages of its conventional counterparts into the chip-based CE separations. The present work was aimed at investigating some of its basic performance param(38) Mikkers, F. E. P.; Everaerts, F. M.; Peek, J. A. F. J. Chromatogr. 1979, 168, 293. (39) Bocˇek, P.; Deml, M.; Kaplanova´, B.; Jana´k, J. J. Chromatogr. 1978, 160, 1. (40) Hutt, L. D.; Glavin, D. P.; Bada, J. L.; Mathies, R. A. Anal. Chem. 1999, 71, 4000. (41) Burggraf, N.; Manz, A.; Verpoorte, E.; Effenhauser, C. S.; Widmer, H. M.; de Rooij, N. F. Sens. Actuators B 1994, 20, 103. (42) Everaerts, F. M.; Verheggen, Th. P. E. M.; Mikkers, F. E. P. J. Chromatogr. 1979, 169, 21. (43) Bocˇek, P.; Deml, M.; Jana´k, J. J. Chromatogr. 1978, 156, 323. (44) Mara´k, J.; Lasˇtinec, J.; Kaniansky, D.; Madajova´, V. J. Chromatogr. 1990, 509, 287. (45) Kaniansky, D.; Madajova´, V.; Mara´k, J.; Sˇ imunicˇova´, E.; Zelensky´, I.; Zelenska´, V. J. Chromatogr. 1987, 390, 51. (46) Dolnı´k, V.; Deml, M.; Bocˇek, P. J. Chromatogr. 1985, 320, 89. (47) Deml, M.; Foret, F.; Bocˇek, P. J. Chromatogr. 1985, 320, 159. (48) Kaniansky, D.; Mara´k, J. J. Chromatogr. 1990, 498, 191. (49) Stegehuis, D. S.; Irth, H.; Tjaden, U. R.; van der Greef, J. J. Chromatogr. 1991, 538, 393. (50) Krˇiva´nkova´, L.; Foret, F.; Bocˇek, P. J. Chromatogr. 1991, 545, 307. (51) Kaniansky, D.; Mara´k, J.; Lasˇtinec, J.; Reijenga, J. C.; Onuska, F. I. J. Microcolumn Sep. 1999, 11, 141. (52) Hirokawa, T.; Ohmori, A.; Kiso, Y. J. Chromatogr. 1993, 634, 101. (53) Grass, B.; Neyer, A.; Jo ¨hnck, M.; Siepe, D.; Eisenbeiss, F.; Weber, G.; Hergenro ¨der, R. Sens. Actuators B, submitted for publication.

Analytical Chemistry, Vol. 72, No. 15, August 1, 2000

3597

eters and at assessing its potential applicability for various ITP techniques, CZE, and CZE with on-line ITP sample pretreatment. The CE separations with suppressed hydrodynamic and electroosmotic flows in the separation compartment21,37,54-56 were favored in this work to minimize a number of sources contributing to run-to-run fluctuations of the migration velocities (times) of the separated constituents. In addition, such a transport concept was expected to reduce problems in quantitative analysis on the chip associated with electroosmotic flow.57 EXPERIMENTAL SECTION Instrumentation. A design of the poly(methyl methacrylate) CC chip used in this work is described in detail elsewhere (design no. 1, in ref 53). A schematic arrangement of its channels and their geometrical dimensions are given in Figure 1. The separations on the chip were performed in a laboratory designed and constructed CE equipment. This equipment includes the following two units. (1) An electrolyte and sample management unit (E&SMU, in Figure 1) is connected via 300 µm i.d. FEP (fluorinated ethylene propylene copolymer) capillary tubes to the inlets of the channels on the chip. Valves of this unit (V1, V2, VT, and VS, in Figure 1) serve to open these inlets on filling the channels, and they are closed during the CE runs. Pumping syringes (P1, P2, PS, and PT, in Figure 1), connected to the inlets of the corresponding valves, deliver appropriate electrolyte solutions and the sample to the channels before the separation. An outlet channel of the chip, connected to a waste container (W, in Figure 1), is permanently opened. (2) An electronic and control unit (E&CU, in Figure 1), designed and constructed by Fitek (Sˇ al’a, Slovak Republic), delivers the driving current, measures conductivity using platinum detection sensors sputtered on the channels of the chip, and interfaces the CE equipment with a PC computer. This unit includes the following modules (see also Figure 1): (i) highvoltage power supply (HV, in Figure 1), delivering the stabilized driving current in the range of 0-50 µA with a maximum voltage of 5 kV connected to the chip; (ii) high-voltage relay (HV-relay, in Figure 1), for the column-switching operation of the equipment; (iii) two conductivity detectors (CD1 and CD2, in Figure 1), decoupled from the detection sensors on the chip by transformers with PTFE insulated coils, where the detector for the first channel (CD1) is provided with a comparator circuit to identify a front boundary of the ITP zone of a selected effective mobility (needed in a control of the column-switching operation of the equipment); (iv) control unit (Control unit, in Figure 1), connecting the CE unit to a PC Pentium computer. ITP Win software (version 2.31) obtained from Kascomp (Bratislava, Slovak Republic) was used for a time-programmed control of the CE runs and for the acquisition of the detection data and their processing. (54) Everaerts, F. M.; Beckers, J. L.; Verheggen, Th. P. E. M. Isotachophoresis. Theory, Instrumentation and Applications; Elsevier: Amsterdam, 1976. (55) Bocˇek, P.; Deml, M.; Gebauer, P.; Dolnı´k, V. Analytical Isotachophoresis; VCH: Weinheim, Germany, 1988. (56) Foret, F.; Krˇiva´nkova´, L.; Bocˇek, P. Capillary Zone Electrophoresis; VCH: Weinheim, Germany, 1993. (57) Shultz-Lockyear, L. L.; Colyer, C. L.; Fan, Z. H.; Roy, K. I.; Harrison, D. J. Electrophoresis 1999, 20, 529.

3598 Analytical Chemistry, Vol. 72, No. 15, August 1, 2000

Figure 1. Block scheme of the CE equipment to separations with the closed separation compartment of the CC chips: E&CU ) electronic and control unit; HV ) high-voltage power supply [its highvoltage pole is connected to the driving electrode in the high voltage (terminating) channel of the chip, ST]; CD1, CD2 ) conductivity detectors for the first and second separation channels, respectively; HV-relay ) a high-voltage relay switching the direction of the driving current in the separation compartment (moving reeds of this relay connect to the ground pole (G) of HV either CE1 or CE2); ST ) a high-voltage (terminating) channel; S ) a 1.2 µL sample injection channel [31 × 0.2 × 0.2 mm (length, width, depth)]; SC1 ) the first separation channel [a 1.1 µL volume; 28 × 0.2 × 0.2 mm (length, width, depth)] with a platinum conductivity sensor (connected to CD1); SC2 ) the second separation channel [a 1.4 µL volume; 34.5 × 0.2 × 0.2 mm (length, width, depth)] with a platinum conductivity sensor (connected to CD2); CE1, CE2 ) counter electrodes for the first and second separation channels, respectively; L1, L2 ) separation paths for the ITP measurements of the migration velocities on the chip; E&SMU ) electrolyte and sample management unit; V1,V2,VT ) needle valves for the inlets of the separation and terminating channels; VS ) a pinch valve for the inlet of the sample injection channel; W ) waste container. P1, P2, PS, PT ) syringes for filling the first, second, sample injection, and terminating channels with the electrolyte and sample solutions, respectively.

Operation of the CE Equipment. The CE runs on the chip were prepared by filling the channels with appropriate solutions in the following order: (1) the second separation channel (SC2, in Figure 1); (2) the first separation channel (SC1); (3) the terminating channel (SCT); (4) the sample channel (S). To prevent mixing of the solutions in the separation compartment only the corresponding valve (see above) was opened on the filling of a particular channel. Excesses of the solutions were led to the waste container (W, in Figure 1) via a permanently opened outlet channel of the chip. Positions of the reeds of the high-voltage relay (HV-relay, in Figure 1) determine the direction of the driving current in the separation compartment during the CE run. Switching of these positions (column-switching) during the run was controlled (i) by the detector in the first channel (the signal of this detector is processed by a comparator circuit) and (ii) by a time program of the switching operations stored in the E&CU of the equipment. Chemicals and Electrolyte Solutions. Chemicals used for the preparation of the electrolyte solutions and the solutions of anionic model mixtures were bought from Sigma (St. Louis, MO), Serva (Heidelberg, Germany), Merck (Darmstadt, Germany), and Lachema (Brno, Czech Republic). Methylhydroxyethylcellulose

Table 1. Electrolyte Systemsa ES1 solvent leading anion concn (mmol/L) counterion additive concn (%, w/v) pH solvent terminating anion concn (mmol/L) counterion pH

water Cl10 His MHEC 0.2 6.0 water Glu 5 His 6.0

ITP ES2 ES3 water Cl20 His MHEC 0.2 6.0 water Glu 5 His 6.0

water Cl2.5 His MHEC 0.2 6.0 water Glu 5 His 6.0

ES4

ES5

ES6

water Cl5 Bala MHEC 0.2 3.6 water Cap 5 Bala 3.7

water Cl2.5 Bala MHEC 0.1 3.6 water Cap 5 Bala 3.7

water Cl8 Bala MHEC 0.2 3.6 water Asp 4 Bala 4.0

CZE solvent carrier anion concn (mmol/L) counterion additive concn (%, w/v) pH

ES7

ES8

water Asp 10 BTP MHEC 0.2 4.0

water Asp 10 Bala MHEC 0.2 3.4

a Abbreviations used: His ) histidine; Bala ) β-alanine; MHEC ) methylhydroxyethylcellulose; Glu ) Glutamate; Cap ) capronate; Asp ) aspartate; BTP ) bis-trispropane.

30 000 (Serva), purified on a mixed-bed ion exchanger (Amberlite MB-1; BDH, Poole, U.K.), was used as a suppressor of electroosmotic flow. It was added to the leading and carrier electrolyte solutions or it was applied as a coating of the inner walls of the separation channels.58 Detail compositions of the electrolyte solutions employed in our CE experiments are given in Table 1. Water purified by a Pro-PS water purification system (Labconco, Kansas City, KS) was used for the preparation of the solutions. The electrolyte solutions used in the ITP and CZE separations were filtered by disposable membrane filters of 0.8 µm pore sizes (Sigma) connected to syringes used for filling the separation compartment (P1, P2, and PT in Figure 1). RESULTS AND DISCUSSION Transport Processes and CE Separations on the CC Chip. Electroosmotic (EOF) and hydrodynamic (HDF) flows of the solution in which the CE separation is carried out may accompany electrophoretic migrations of the separated constituents. Usually employed to prolong or shorten an effective length of the separation path in the CE column,54,55,59,60 EOF and HDF also contribute to overall run-to-run fluctuations of the migration velocities of the separated constituents (∂vtot) in accordance with the law of propagation of errors:61 (58) Kaniansky, D.; Masa´r, M.; Bielcˇ´ıkova´, J. J. Chromatogr. A 1997, 792, 483. (59) Everaerts, F. M.; Vacı´k, J.; Verheggen, Th. P. E. M.; Zuska, J. J. Chromatogr. 1970, 49, 262. (60) Jandik, P.; Bonn, G. Capillary Electrophoresis of Small Molecules and Ions; VCH: Weinheim, Germany, 1993. (61) Massart, D. L.; Vandeginste, B. G. M.; Buydens, L. M. C.; De Jong, S.; Lewi, P. J.; Smeykers-Verbeke, J. Handbook of Chemometrics and Qualimetrics. Part A; Elsevier: Amsterdam, 1997.

Table 2. Reproducibilities of the Migration Times of the Boundary Formed by the Leading and Terminating Electrolytes for Defined Migration Paths on the Chipa chip no. param migration time, path “L1” (s) SD (s) migration time, path “L2” (s) SD (s)

“1” 121.35 0.92 179.29 0.35

“2” 120.29 1.16 180.18 1.86

a The measurements were carried out in the electrolyte system ES1 (Table 1) for the migration paths L1 and L2 as shown in Figure 1. Average values from a series of 10 repeated ITP runs under identical working conditions are given for both chips. The data were obtained with the same electrolyte and sample management unit.

∂vtot ) x(∂vep)2 + (∂veo)2 + (∂vhd)2

(1)

where ∂vep, ∂veo, ∂vhd are symbols characterizing random fluctuations of the electrophoretic, electroosmotic, and hydrodynamic velocities in repeated separations, respectively. Equation 1 shows that CE separations carried out without EOF and HDF offer, in general, the highest reproducibility of the migration velocities (times) of the separated constituents. Such transport conditions were preferred in this work as they were assumed to contribute to both highly reproducible CE quantitations54-56 and a reliable performance of the equipment in the separations using the column-switching (see Experimental Section). From geometrical dimensions of the channels on the CC chip (Figure 1) it is apparent that its separation compartment has a relatively small hydrodynamic resistance and, therefore, only a small pressure difference between the inlet channels can cause an undesired HDF of the solution in this compartment. To prevent such, hardly controllable, flows during the CE runs the inlet channels to the separation compartment were closed with the aid of the valves (V1, V2, VT, and VS, in Figure 1; see also Experimental Section). This solution, the separation in a (hydrodynamically) closed separation compartment, in fact, prevented HDF in the same way as employed in conventional CE instruments designed, mainly, for ITP separations.54,55 EOF in our experiments was suppressed in the way as currently preferred in the separations performed in the CE instruments with the closed separation compartment,21,37,54-56 and methylhydroxyethylcellulose, present in the electrolyte solutions used in this work (Table 1), served as an EOF suppressor. Run-to-run fluctuations of the migration velocities (times) of the separated constituents under the above transport conditions can be ascribed, mainly, to fluctuations in their electrophoretic migration velocities. Their magnitudes were estimated from repeated measurements of the migration times needed for the boundary formed by the leading and terminating electrolytes to migrate between well-defined positions in the separation compartment of the chip (L1 and L2, in Figure 1). Typical data, as obtained from the measurements performed in this respect with two chips (Table 2), reflect highly reproducible migration conditions attainable on the chip without EOF and HDF. In this context, we should note that 10-30 times higher SD values of the migration times were typical for comparative measurements with the valves in the electrolyte and sample management unit (V1, V2, VT, and VS, in Figure 1) bypassed, under otherwise identical experimental Analytical Chemistry, Vol. 72, No. 15, August 1, 2000

3599

Figure 2. Schemes of the starting arrangements of the electrolyte and sample solutions in the separation compartment for various CE techniques carried out on the CC chip: a ) single-column ITP; b ) ITP with tandem-coupled separation channels; c ) concentrationcascade ITP; d ) single-column CZE; e ) CZE with tandem-coupled separation channels; f ) CZE on-line coupled with ITP sample pretreatment; CD1, CD2 ) on-column conductivity sensors in the first and second separation channels, respectively; CE1, CE2 ) driving electrodes for the first and second separation channels, respectively; S ) sample position; LE, LE1, LE2 ) symbols for the leading electrolyte solutions; TE ) terminating electrolyte solution; BE ) background (carrier) electrolyte solution. i, i1, i2 ) symbols for the driving currents (arrows at the symbols indicate the directions of the driving currents).

conditions. Here, a residual HDF in the channels, due to a slowpressure equilibration in the pumping syringes (P1, P2, PS, and PT, in Figure 1), was found to be a source of such high run-to-run fluctuations. ITP Separation and Quantitation of the Analytes on the CC Chip. The CC configuration of the separation channels on the chip provides means for implementations of various ITP techniques as developed for conventional CE separation systems.54,55 Three of these techniques, viz., (i) single-column ITP, (ii) ITP in the tandem-coupled columns, and (iii) concentrationcascade ITP, were employed in our experiments with the chip using succinate, acetate, and benzoate as test analytes. Single-column ITP separations were carried out in the first separation channel of the chip (Figure 2a). An illustrative isota3600 Analytical Chemistry, Vol. 72, No. 15, August 1, 2000

Figure 3. Isotachopherograms from the separations of succinate (1), acetate (2), and benzoate (3) by different ITP techniques. a ) a single-column separation in the first channel using CD1 (Figure 2a) to monitor the separation. The separation was carried out in the electrolyte system ES1 (Table 1) with a 10 µA driving current. The injected sample contained the anions at 300 µmol/L concentrations. b ) a separation in the tandem-coupled separation channels using CD2 (Figure 2b) to monitor the separation. The separation was carried out using the electrolyte system ES1 (Table 1) in both separation channels with a 10 µA driving current. The injected sample contained the anions at 500 µmol/L concentrations. c ) a concentration cascade ITP separation. The conductivity detector in the second channel (CD2, in Figure 2c) was used to monitor the separation. The separation in the first channel was carried out in the electrolyte system ES2 (Table 1) with a 20 µA driving current. The separation in the second channel was carried out in the electrolyte system ES3 (Table 1) with a 10 µA driving current. The injected sample contained the anions at 1000 µmol/L concentrations. Le ) leading anion (chloride), and Te ) terminating anion (glutamate).

chopherogram in Figure 3a and data in Table 3 show that rapid and reproducible ITP analyses were achieved by this technique. They also demonstrate a high production rate of the ITP separation process62 as this produced approximately 4 pmol of a pure analyte/s. Therefore, succinate, acetate, and benzoate could be resolved in the ITP separations carried out in the electrolyte system ES1 (Table 1) also in instances when their concentrations in the injected sample were as high as 300 µmol/L. Despite this, their simultaneous determination by single-column ITP was possible only within a narrow concentration range as the CLOD values for succinate (30 µmol/L), acetate (50 µmol/L), and benzoate (50 µmol/L) were too high under these electrolyte conditions. A link of such a narrow range with a limited load capacity of a 28-mm separation path in the first separation channel of the chip has a clear theoretical background.38,39 A higher load capacity was inherent to ITP in the tandemcoupled separation channels of the chip (Figures 2b and 3b). Here, (62) Kaniansky, D.; Sˇ imunicˇova´, E.; O ¨ lvecka´, E.; Ferancova´, A. Electrophoresis 1999, 20, 2786.

Table 3. Reproducibilities of the Zone Lengths for Various Concentrations of the Test Analytes Using Different ITP Techniques on the Same Chipa av length of the analyte zonea succinate concn of the analyte (µmol/L)

tZ (s)

100 150 200 250 300

2.04 3.27 4.05 5.22 6.29

100 200 300 400 500 600 200 400 600 800 1000

acetate SD (s)

tZ (s)

lactate SD (s)

tZ (s)

SD (s)

Single-Channel ITP 0.03 1.81 0.03 2.72 0.02 3.37 0.06 4.17 0.15 5.09

0.03 0.03 0.03 0.03 0.11

1.68 2.70 3.37 4.32 5.26

0.03 0.02 0.06 0.03 0.13

2.28 4.17 5.96 8.18 10.35 12.68

Tandem-Coupled ITP 0.03 2.38 0.08 3.73 0.22 5.28 0.15 6.72 0.20 8.32 0.25 10.47

0.13 0.08 0.16 0.08 0.24 0.33

1.90 3.37 4.44 6.30 8.28 10.25

0.05 0.03 0.17 0.36 0.08 0.15

2.07 4.13 5.95 8.42 9.67

Concentration-Cascade ITP 0.03 2.05 0.03 3.60 0.15 5.23 0.12 6.58 0.03 8.27

0.02 0.02 0.20 0.23 0.03

2.25 3.43 5.27 7.07 8.78

0.02 0.06 0.16 0.16 0.06

a Abbreviations used: t ) an average value of the time-based ITP zone length (each value corresponds to 3 repeated measurements); see also Z the text and Figures 2 and 3 for descriptions of the ITP techniques employed.

a 62-mm separation path made possible a complete ITP resolution of the test anions when their concentrations in the injected sample were 600 µmol/L or less. As the CLOD values for the test analytes in the second channel did not differ from those achieved in the first channel (see above), the concentration range within which they could be simultaneously determined was approximately doubled in comparison to the single-column technique. However, this gain was associated with a proportional increase of the analysis time (Figure 3b). The CE run was divided into two stages of differing functions in concentration-cascade ITP separations carried out on the chip (Figures 2c and 3c). A 20 mmol/L concentration of the leading anion in the electrolyte system employed in the first separation channel (ES 2, Table 1) approximately doubled the load capacity of this stage in comparison to the single-column ITP separations (see above). An additional contribution to the load capacity of the separation system was associated with a transfer of the leading anions to the second separation channel during the separation.43 On the other hand, a 2.5 mmol/L concentration of the leading anion in the electrolyte solution employed in the second separation channel (ES 3, Table 1) provided favorable CLOD values for the analytes [succinate (8 µmol/L), acetate (12 µmol/L), and benzoate (12 µmol/L)]. Such a combination of different functions of the separation channels made possible the ITP separations of 1200 pmol amounts of the test analytes (each at a 1 mmol/L concentration in the injection sample) and, at the same time, provided favorable conditions for their detections. Therefore, the concentration range within which the test analytes could be simultaneously determined increased about 10-fold in comparison to the singlecolumn technique. Apparently, these advantages of concentrationcascade ITP are favorable in situations when the samples containing the analytes at very differing concentrations are to be analyzed on the CC chip. The use of the concentration-cascade

ITP technique in a conventional way (a complete transfer of the leading ions from the first to the second separation channel) is linked with a relatively long analysis time.43 In our experiments this drawback was eliminated by a defined transfer of the leading ions from the first to the second separation channel via a proper switching of the direction of the driving current (Figure 2c). Although ITP separations of multicomponent mixtures were not studied in detail in the present work, an isotachopherogram obtained from the separation of a 14-component test mixture of anions at pH 3.5 (Figure 4) may serve as an illustration of potentialities of the CC chip in such separations. The concentration-cascade ITP technique was found effective also in this instance. Recently, phenomena accompanying injections of the samples on the CE chips by a double “T” injection technique were studied.57 From the quoted work one can deduce that EOF in the injection channel is a potential source of analytical errors in the quantitations of the CE analytes. The data obtained with this injection technique under our working conditions (Tables 3 and 4) indicate that a suppression of EOF has a positive impact on the reproducibility of the sample injection in a general sense. In this context, we should also note that the reproducibility of the ITP quantitation as achieved in this work did not differ significantly from that attainable for similar zonelengths in conventional ITP separation systems.54,55 CZE and ITP-CZE Separations on the CC Chip. The volume of the sample injection channel on the present chip (1.2 µL) is comparable to 1.1 and 1.4 µL volumes of its separation channels (Figure 1). From the point of view of high-efficiency CZE separations this is an apparent volume disproportion.21,22,56,60 A negative impact of such a disproportion on the separation efficiencies in conventional CE separation systems is prevented, Analytical Chemistry, Vol. 72, No. 15, August 1, 2000

3601

Figure 4. An isotachopherogram from the separation of a 14component model mixture of anions using concentration-cascade ITP. CD2 (Figure 2c) was used to monitor the separation. The separation in the first channel was carried out in the electrolyte system ES4 (Table 1) with a 10 µA driving current. The separation in the second channel was carried out in the electrolyte system ES5 with a 5 µA driving current. The injected sample contained the anions at 200 µmol/L concentrations. Zone assignments: Le ) leading anion (chloride); 1 ) chlorate; 2 ) methanesulfonate; 3 ) dichloroacetate; 4 ) phosphate; 5 ) citrate; 6 ) isocitrate; 7 ) glucuronate; 8 ) β-bromopropionate; 9 ) succinate; 10 ) glutarate; 11 ) acetate; 12 ) suberate; 13 ) propionate; 14 ) valerate; Te ) terminating anion (capronate). Table 4. Parameters of the Regression Equations for the Calibration Lines of the Analytes and for Different ITP Techniques Performed on the Same Chipa a (s)

b (s/(µmol‚L-1))

r

c (µmol‚L-1)

n

100-300 100-300 100-300

15 15 15

succinate acetate benzoate

-0.024 0.217 -0.067

Single-Channel ITP 0.021 0.9977 0.016 0.9979 0.018 0.9977

succinate acetate benzoate

0.212 0.246 0.830

Tandem-Coupled ITP 0.020 0.9980 0.016 0.9957 0.015 0.9978

100-500 100-500 100-500

15 15 15

succinate acetate benzoate

0.202 0.522 0.350

Concentration-Cascade ITP 0.0097 0.9961 0.0077 0.9982 0.0084 0.9968

200-1000 200-1000 200-1000

15 15 15

a Abbreviations used: a ) intercept; b ) slope; r ) correlation coefficient; c ) concentration span of the analyte in the injected sample; n ) number of data points.

for example, by an electric field stacking of the injected sample.63 This stacking technique was favored in our CZE experiments carried out in the first separation channel (Figure 2d) and in the tandem-coupled separation channels (Figure 2e). A typical electropherogram as obtained in the CZE separations of model samples in the first separation channel is given in Figure 5a. Here, sulfate, fluoride, and phosphate (these anions dissolved at 10 µmol/L concentrations in a 10% solution of the carrier electrolyte were taken for the separation) were resolved in less than 40 s. On the other hand, the shapes of their peaks clearly show that the stacking effect was not efficient enough to focus the injected (63) Chien, R.-L.; Burgi, D. S. Anal. Chem. 1992, 64, 489A.

3602 Analytical Chemistry, Vol. 72, No. 15, August 1, 2000

Figure 5. Electropherograms from CZE separations of sulfate (1), fluoride (2), and phosphate (3) on the CC chip. a ) the separation in the first channel using the electrolyte system ES7 (Table 1) and a 40 µA driving current. The injected sample contained the anions at 10 µmol/L concentrations (dissolved in a 10% (v/v) solution of the carrier electrolyte). b ) the separation in the tandem-coupled separation channels using CD2 (Figure 2e) to monitor the separation. The same working conditions and the same sample were as in (a).

sample into a volume negligible relative to that of the separation channel.21,22,56,60 Similar conclusions can be drawn from an electropherogram (Figure 5b) as obtained, under identical working conditions and with the same sample, from the separation performed in the tandem-coupled separation channels (Figure 2e). Our attempts to enhance the sample stacking via a 5-fold increased concentration of the carrier electrolyte failed as a high conductivity of the carrier electrolyte negatively influenced performance of the conductivity detection (measurements of very small conductivity changes on a high conductivity background). However, this is a general disadvantage of the conductivity detection in CZE21,56 and a reduction of the volume of the injection channel on the chip may solve this problem. On the other hand, the CC configuration of the separation channels of the chip offers an alternative solution, viz., the sample injection with an electric sample splitting. This solution may be advantageous, as it requires no modification of the chip when the electronic and control unit of the CE equipment is appropriately adapted.47 An experimental evaluation of both alternatives will be a subject of our future research. The CC configuration of the separation channels on the chip makes it suitable for CZE separations with on-line ITP sample pretreatment (ITP-CZE) in the way as described for conventional column-coupling CE systems.48-51,64 Therefore, this combination of electrophoretic methods on the chip (Figure 2f) is assumed to provide low CLOD values for the CE analytes also in instances when these are present in samples containing matrix constituents (64) Kaniansky, D.; Zelensky´, I.; Hybenova´, A.; Onuska, F. I. Anal. Chem. 1994, 66, 4258.

Table 5. Reproducibility of the Transfer of a Fraction of the Separated Sample Constituents from the First to the Second Channel of the CC Chipa analyte zone length succinate

acetate

tZ (s)

SD (s)

tZ (s)

4.09

0.03

3.22

First Channel 0.03 3.42

2.31

Second Channel 0.03

2.49

0.05

SD (s)

benzoate tZ (s)

total

SD (s)

ttot (s)

SD (s)

0.03

10.73

0.08

4.79

0.06

a Abbreviations used: t ) an average length of the zone (for 10 Z repeated runs). See the text and Figure 6 for further details.

Figure 6. Isotachopherograms from a transfer of the sample fraction from the first to the second channel using the column switching on the CC chip. a ) an isotachopherogram from the separation performed in the first channel [CD1 (Figure 1) was used to monitor the separation] in the electrolyte system ES1 (Table 1) with a 10 µA driving current. b ) an isotachopherogram for the sample fraction [marked by a dash-line box on the isotachopherogram (a)] transferred into the second channel [CD2 (Figure 1) was used to monitor the separation]. The working conditions were the same as in (a). The injected sample contained succinate (1), acetate (2), and benzoate (3) (each at a 200 µmol/L concentration).

at considerably higher concentrations. Here, a transfer of the analyte containing sample fraction from the ITP stack into the CZE stage and a removal of the matrix constituents from the separation compartment after the separation in the ITP stage are integrated into the CE separation.51 These key operations are performed by a proper switching of the direction of the driving current in the separation compartment during the CE run. This function of the CC chip was evaluated from ITP runs carried out under identical working conditions in both separation channels. Here, the column-switching program of the run (see Experimental Section) switched the directions of the driving current in such a way that only the fraction of our interest (marked by a dash-line box on an isotachopherogram in Figure 6a) was transferred from the first to the second channel. From the data characterizing this transfer (Table 5) we can see that not only total sizes of the transferred sample fractions but also relative contents of the test analytes in the fractions reproduced very satisfactory. Undoubtedly, these data indicate that reliable column-switching operations on the chip in the ITP-CZE separations can be expected. Model mixtures, containing chloride and sulfate as macroconstituents and nitrite, fluoride, and phosphate as microconstituents, were used in our feasibility experiments with the ITP-CZE combination on the CC chip. Here, the separation scheme (Figure 2f) followed the one as proposed for the ITP-CZE separations of these anions in conventional column-coupling CE equipment.64 The sample was separated by ITP in the first channel, and main parts of the sample macroconstituents (migrating in a front part

Figure 7. Electropherograms from the CZE stage in the ITP-CZE separations of anions on the CC chip. The ITP separation in the first channel, monitored by CD1 (Figure 2f), was carried out in the electrolyte system ES6 (Table 1) with a 5 µA driving current. The CZE separation of the transferred sample fraction in the second channel, monitored by CD2 (Figure 2f), was carried out in the electrolyte system ES8 (Table 1) with a 7 µA driving current. The injected sample contained chloride (600 µmol/L), sulfate (800 µmol/ L), nitrite (1) (10 µmol/L), fluoride (2) (10 µmol/L), and phosphate (3) (10 µmol/L). a, b ) repeated runs with the same sample under identical working conditions.

of the ITP stack) were led out of the separation channel to the counter electrode of the first channel (CE1 in Figure 2f). Only a rear part of the ITP zone of macroconstituents and focused microconstituents (migrating behind the macroconstituents under our separating conditions) were electrophoretically transferred into the second channel for the CZE separation and detection. Electropherograms as obtained in this channel for the same sample in repeated ITP-CZE runs (Figure 7) reflect a good repeatability of our experiments with the ITP-CZE combination on the CC chip. In addition, they illustrate excellent detectabilities for nitrite, fluoride, and phosphate (each of these microconstituents was present in the injected sample at a 10 µmol/L concentration) in a sample containing the macroconstituents at significantly Analytical Chemistry, Vol. 72, No. 15, August 1, 2000

3603

higher concentrations [sulfate (800 µmol/L) and chloride (600 µmol/L)]. CONCLUSIONS Some specific advantages of the CC chip, especially in the ITP and ITP-CZE separations, are apparent from the results presented in this work. In ITP, for example, its CC configuration of the separation channels makes possible the separations by an analytically very beneficial concentration-cascade technique. In addition, this configuration provides means for a direct integration of the ITP concentration of the analytes and in-column sample cleanup into the CE separations carried out on the chip and, in fact, it offers ways for significant simplifications of sample pretreatment procedures. CZE separations with on-line ITP sample pretreatment as carried out in our feasibility experiments with model samples of inorganic anions outlined some of these potentialities. In this context, we should also note rapid ITP resolutions of nanomol amounts of the test analytes as currently achieved in our experiments (see Figure 3). Such resolution rates38 attainable by ITP on the CC chip may be very favorable, especially in instances when rapid pretreatments of the samples of highly ionic matrixes are required (e.g., various samples of biological and environmental origins). Employed with suppressed EOF and the closed separation compartment (to prevent HDF), the CC chip provides working conditions under which only the electrophoretic transport of the separated constituents is effective. Very reproducible ITP quantitations as achieved in this work (see, Table 3) can be attributed, at least partially, to the use of this approach. The use of the CC

3604

Analytical Chemistry, Vol. 72, No. 15, August 1, 2000

chip is not restricted to the CE separations with the closed separation compartment, and it can be employed in the separations with the opened separation compartment56,60,65 as well. This, however, requires a corresponding adaptation of the electrolyte and sample management unit (see Experimental Section) to control or eliminate HDF in the separation compartment by other means. The conductivity detectors monitored the CE separations carried out on the CC chip. Although very convenient for the detection of the ITP zones, the conductivity detection has certain limitations in the CZE separations (measurements of small conductivity changes due to the zones on a relatively high conductivity background of the carrier electrolyte). The CC chip itself is compatible also with other detection techniques currently employed in conventional CE equipment,54-56,60 and undoubtedly, the use of these on-column and/or postcolumn techniques can significantly enhance its overall analytical utility. ACKNOWLEDGMENT This work was supported by Merck (Darmstadt, Germany) and, in part, by a grant from the Slovak Grant Agency for Science under Project No. 1/7247/20.

Received for review October 29, 1999. Accepted May 9, 2000. AC991236S (65) Jorgenson, J. W.; Lukacs, K. D. Anal. Chem. 1981, 53, 1298.