Sweeping: concentration mechanism and applications to high-sensitivity analysis in capillary electrophoresis

Sweeping in capillary electrophoresis (CE) involves the interaction of a pseudostationary phase (PS) in the separation solution and a sample in the matrix that is free of the PS used. The PS includes not only the PSs employed in electrokinetic chromatography, but also complexation reagents such as borate. The sample matrix could have a lower, similar, or higher conductance than the separation solution. Thus, the basic condition for sweeping is a sample matrix free of the additive. The accumulation of analyte molecules during the interaction makes this interesting phenomenon very useful as an on-line preconcentration method for CE. Preconcentration occurs due to chromatographic partitioning, complexation, or any interaction between analytes and PS. Contact between analyte and PS is facilitated by the action of electrophoresis and is independent of electroosmosis. The analyte, PS, or both should have electrophoretic velocities when an electric field is applied. The extent of preconcentration is dictated by the strength of the interaction involved. From tens to several thousand-fold improvements in detector response for many neutral and charged analytes have been achieved with this technique, suggesting sweeping as a general approach to on-line preconcentration in CE. The mechanism and applications of the sweeping phenomenon under different experimental conditions are discussed in this review, with particular emphasis on a better understanding of the sweeping mechanism under reduced electric field (high conductivity) in the sample zone.     

Sweeping and new on-line sample preconcentration techniques in capillary electrophoresis

Sweeping is a powerful on-line sample preconcentration technique that improves the concentration sensitivity of capillary electrophoresis (CE). This approach is designed to focus the analyte into narrow bands within the capillary, thereby increasing the sample volume that can be injected, without any loss of CE efficiency. It utilizes the interactions between an additive [i.e., a pseudostationary phase (PS) or complexing agent] in the separation buffer and the sample in a matrix that is devoid of the additive used. The accumulation occurs due to chromatographic partitioning, complexation or any interaction between analytes and the additive through electrophoresis. The extent of the preconcentration is dependent on the strength of interaction involved. Both charged and neutral analytes can be preconcentrated. Remarkable improvements—up to several thousandfold—in detection sensitivity have been achieved. This suggests that sweeping is a superior and general approach to on-line sample preconcentration in CE. The focusing mechanism of sweeping under different experimental conditions and its combination with other on-line preconcentration techniques are discussed in this review. The recently introduced techniques of transient trapping (tr-trapping) and analyte focusing by micelle collapse (AFMC) as well as other novel approaches to on-line sample preconcentration are also described.

Strategies for the on-line preconcentration and separation of hypolipidaemic drugs using micellar electrokinetic chromatography

Three strategies were investigated for the simultaneous separation and on-line preconcentration of charged and neutral hypolipidaemic drugs in micellar electrokinetic chromatography (MEKC). A background electrolyte (BGE) consisting of 20 mM ammonium bicarbonate buffer (pH 8.50) and 50 mM sodium dodecyl sulfate (SDS) was used for the separation and on-line preconcentration of the drugs. The efficiencies of sweeping, analyte focusing by micelle collapse (AFMC), and simultaneous field-amplified sample stacking (FASS) and sweeping, were compared for the preconcentration of eight hypolipidaemic drugs in different conductivity sample matrices. When compared with a hydrodynamic injection (5 s at 50 mbar, 0.51% of capillary volume to detection window) of drug mixture prepared in the separation BGE, improvements of detection sensitivity of 60-, 83-, and 80-fold were obtained with sweeping, AFMC and simultaneous FASS and sweeping, respectively, giving limits of detection (LODs) of 50, 36, and 38 μg/L, respectively. The studied techniques showed suitability for focusing different types of analytes having different values of retention factor (k). This is the first report for the separation of different types of hypolipidaemic drugs by capillary electrophoresis (CE). The three methods were validated then applied for the analysis of target analytes in wastewater samples from Hobart city.     

Sweeping of neutral analytes via complexation with borate in capillary zone electrophoresis

Summary The sweeping concept is extended to capillary zone electrophoresis (CZE) separation of neutral solutes involving complexation with borate. Analogous to the pseudostationary phase in electrokinetic chromatography (EKC), the complexing agent (borate) serves as carrier for sweeping and separation in CZE. Therefore, similar to the retention factor in EKC, the focusing effect in the present system is directly related to the association constant between the analyte and complexing agent. Theoretical and some preliminary experimental studies gerenally suggest that the electrophoretic mobility of the complex and the concentration of the complexing agent affect the resulting length of narrowed zones. Moreover, sweeping using borate is affected by pH since borate complexation is pH dependent. From around 10 to 40-fold improvement in peak heights has been observed experimentally for some neutral test analytes (monosaccharides, catechols, and nucleosides)     

Optimization of background electrolytes for capillary electrophoresis I. Mathematical and computational model

A mathematical and computational model is introduced for optimization of background electrolyte systems for capillary zone electrophoresis of anions. The model takes into account mono- or di- or trivalent ions and allows also for modeling of highly acidic or alkaline electrolytes, where a presence of hydrogen and hydroxide ions is significant. At maximum, the electrolyte can contain two co-anions and two counter-cations. The mathematical relations of the model are formulated to enable an easy algorithmization and programming in a computer language. The model assesses the composition of the background electrolyte in the analyte zone, which enables prediction of the parameters of the system that are experimentally available, like the transfer ratio, which is a measure of the sensitivity in the indirect photometric detection or the molar conductivity detection response, which expresses the sensitivity of the conductivity detection. Furthermore, the model also enables the evaluation of a tendency of the analyte to undergo electromigration dispersion and allows the optimization of the composition of the background electrolyte to reach a good sensitivity of detection while still having the dispersion properties in the acceptable range. Although the model presented is aimed towards the separation of anions, it can be straightforwardly rearranged to serve for simulation of electromigration of cationic analytes. The suitability of the model is checked by inspecting the behavior of a phosphate buffer for analysis of anions. It is shown that parameters of the phosphate buffer when used at neutral and alkaline pH values possess singularities that indicate a possible occurrence of system peaks. Moreover, if the mobility of any analyte of the sample is close to the mobilities of the system peaks, the indirect detector signals following the background electrolyte properties will be heavily amplified and distorted. When a specific detector sensitive on presence of the analyte were used, the signal would be almost lost due to the excessive dispersion of the peak.     

Flow manipulation for sweeping with a cationic surfactant in microchip capillary electrophoresis

Flow manipulation in sweeping microchip capillary electrophoresis (CE) is complicated by the free liquid communication between channels at the intersection, especially when the electroosmotic flows are mismatched in the main channel. Sweeping in traditional CE with cationic micelles is an effective way to concentrate anionic analytes. However, it is a challenge to transfer this method onto microchip CE because the dynamic coating process on capillary walls by cationic surfactants is interrupted when the sample solution free of surfactants is introduced into the microchip channels. This situation presents a difficulty in the sample loading, injection and dispensing processes. By adding surfactant at a concentration around the critical micelle concentration and by properly designing the voltage configuration, the flows in a microchip were effectively manipulated and this sweeping method was successfully moved to microchip CE using tetradecyltrimethylammonium bromide (TTAB). The sweeping effect of cationic surfactant in the sample solution was discussed theoretically and studied experimentally in traditional CE. The flows in a microchip were monitored with fluorescence imaging, and the injection and sweeping processes were studied by locating the detection point along the separation channel. A detection enhancement of up to 500-fold was achieved for 5-carboxyfluorescein.     

Determination of anthracycline antibiotics doxorubicin and daunorubicin by capillary electrophoresis with UV absorption detection.

Sweeping preconcentration and electrokinetic injection was used for the capillary electrophoretic analysis of trace amounts of biologically active anthracyclines with UV absorption detection. Phosphate buffer (100 mM), pH 2.5, with addition of 40% v/v methanol was used as background electrolyte (BGE). Sodium dodecyl sulfate (150 mM) was added to BGE in the inlet vial as the sweeping agent. The system enables effective separation of anthracyclines as well as cleanup from matrix impurities. Sweeping preconcentration of sample provides an excellent detection limit (1 x 10(-9) mol L(-1)). The method was applied for the determination of therapeutic levels of doxorubicin in real plasma samples.     

Conductivity detection in capillary zone electrophoresis: inspection by PeakMaster

A simple rule stating that the signal in conductivity detection in capillary zone electrophoresis is proportional to the difference between the analyte mobility and mobility of the background electrolyte (BGE) co-ion is valid only for systems with fully ionized electrolytes. In zone electrophoresis systems with weak electrolytes both conductivity signal and electromigration dispersion of analyte peaks depend on the conductivity and pH effects. This allows optimization of the composition of BGEs to give a good conductivity signal of analytes while still keeping electromigration dispersion near zero, regardless of the injected amount of sample. The demands to achieve minimum electromigration dispersion and high sensitivity in conductivity detection can be accomplished at the same time. PeakMaster software is used for inspection of BGEs commonly used for separation of sugars (carbohydrates, saccharides) at highly alkaline pH. It is shown that the terms direct and indirect conductivity detection are misleading and should not be used.     

Online large volume sample staking preconcentration and separation of enantiomeric GHRH analogs by capillary electrophoresis

A capillary electrophoresis method is proposed to analyze the four most well-known growth hormone–releasing hormone (GHRH) analogs that are misused by athletes. Dimethyl-β-cyclodextrin used as a chiral selector allowed, for the first time, the separation of those basic peptide analogs, including enantiopeptides (sermorelin and CJC-1293) that differ by the chirality of only one amino acid. To increase the method sensitivity, electrokinetic preconcentration methods have been investigated. The large volume sample stacking with polarity switching (PS-LVSS) method with an injected sample volume corresponding to 80% of the capillary one was found superior to the sweeping in terms of signal enhancement factor (SEF). Acid and organic solvent addition to the sample (0.1 mM phosphoric acid with 30% methanol) led to a twofold signal improvement, when compared to water as a matrix. We increased capillary dimensions to provide a signal enhancement through the injection of a larger sample volume. Finally, using a combination of the optimized PS-LVSS preconcentration with the chiral capillary zone electrophoresis (CZE), the GHRH analogs were separated and limits of detection between 75 and 200 ng/mL were reached. This method was successfully applied to urine after a desalting step. An optimized C18 SPE was used for that purpose in order to provide low sample conductivity (<130 µS/cm) and preserve the efficiency of LVSS preconcentration. SEF of 640 was obtained with desalted urine spiked with sermorelin by comparison to the CZE (without preconcentration) method.     

Stacking due to ionic transport number mismatch during sample sweeping on microchips

Sample stacking can occur in isoconductive buffer systems as a result of ion transport mismatches that cause changes in buffer conductivity during electrophoresis. Fluorescence imaging was used to examine this effect in the sweeping of hydrophobic dyes with sodium dodecyl sulfate (SDS) on microchips. Imaging revealed the occurrence of a stacking effect in a sodium borate buffer system in which the sample buffer and SDS-containing run buffer had the same initial conductivity. Injected sample plugs were first swept by SDS micelles and the swept band was then stacked at the trailing end of the sample zone. This effect is due to changes in conductivity at both the front and back interfaces of the injected sample plug and can be modeled by moving boundary equations. Maximum signal enhancements of 86-, 160- and 560-fold were obtained for Rhodamine 560, Rhodamine B and Rhodamine 6G, respectively, by the combination of sweeping and stacking within a 1 cm section of microchannel. Based on sample sweeping/stacking and manipulation of the electric field polarity, a method of trapping and concentrating analyte from multiple injections was also demonstrated.     

On.column pre‐concentration of alcohol dehydrogenase in capillary electrophoresis

The analysis of alcohol dehydrogenase (ADH) at low concentration using capillary electrophoresis is described. Several simple and effective ways to improve detection limits and sensitivity are investigated. These include large volume sample stacking, head column field amplified sample stacking, and sweeping. Results indicate that by using a combination of head‐column field amplified sample stacking and sweeping, fluorescently labelled alcohol dehydrogenase can be pre‐concentrated online by dissolving samples in water or other low conductivity matrices, and injecting into a high conductivity micellar buffer. The abrupt changes in conductivity cause narrowing of the analyte length and thus enhance the detection sensitivity. Combination of this approach with laser induced fluorescence detection yields a limit of detection of 5×10^–13 M. Both qualitative and quantitative aspects of this method are investigated.     

High-sensitivity analyses of metabolites in biological samples by capillary electrophoresis using dynamic pH junction-sweeping

Emerging fields of biochemical research, such as metabolomics, present challenges to current separation technologies because of the large number of metabolites present in a cell and their often low (submicromolar) concentration. Although capillary electrophoresis (CE) holds great promise as the method of choice for high-resolution separations of biological samples, it suffers from poor concentration sensitivity, especially with the use of UV detection. In CE, sweeping and dynamic pH junction represent two complementary on-line focusing techniques that have been used for sensitivity enhancement of hydrophobic and weakly acidic analytes, respectively. However, the application of either the sweeping or dynamic pH junction technique alone might, in some cases, be less effective for the analysis of certain sample mixtures. Recent work in the development of a hyphenated dynamic pH junction-sweeping technique is presented as an effective on-line method of preconcentration suitable for both hydrophilic (anionic) and hydrophobic (neutral) analytes. Sensitive analyses of flavin metabolites by CE with laser-induced fluorescence (LIF) detection is demonstrated in various biological matrixes, including cell extracts of Bacillus subtilis, pooled human plasma, as well as heat-deproteinized flavoenzymes. Enhanced analyte band narrowing and improved sensitivity is achieved for flavins using dynamic pH junction-sweeping compared to either sweeping or dynamic pH junction alone. This results in over a 1200-fold improvement in sensitivity relative to conventional injection methods, giving a limit of detection (LOD, defined as S/N = 3) of about 4.0 x 10(-12) M. Strategies for sensitive and more comprehensive analyses of other cell metabolites, including nucleotides, coenzymes, and steroids, are also discussed when using on-line focusing techniques in conjunction with multiplexed CE and UV detection.     

On-line preconcentration methods for capillary electrophoresis

The limits of detection (LOD) for capillary electrophoresis (CE) are constrained by the dimensions of the capillary. For example, the small volume of the capillary limits the total volume of sample that can be injected into the capillary. In addition, the reduced pathlength hinders common optical detection methods such as UV detection. Many different techniques have been developed to improve the LOD for CE. In general these techniques are designed to compress analyte bands within the capillary, thereby increasing the volume of sample that can be injected without loss of CE efficiency. This on-line sample preconcentration, generally referred to as stacking, is based on either the manipulation of differences in the electrophoretic mobility of analytes at the boundary of two buffers with differing resistivities or the partitioning of analytes into a stationary or pseudostationary phase. This article will discuss a number of different techniques, including field-amplified sample stacking, large-volume sample stacking, pH-mediated sample stacking, on-column isotachophoresis, chromatographic preconcentration, sample stacking for micellar electrokinetic chromatography, and sweeping.     

Use of short chain alkyl imidazolium ionic liquids for on‐line stacking and sweeping of methotrexate,flinic acid and folic acid: Their application to biological fluids

Methotrexate (MTX) is widely used for the treatment of many types of cancer. Folinic acid (FNA) and folic acid (FA) were usually simultaneously supplemented with MTX to reduce the side effects of a folate deficiency. This study, for the first time, included on‐line sample preconcentration by stacking and sweeping techniques under reduced or enhanced electric conductivity in the sample region using short chain alkyl imidazolium ionic liquids (ILs) as micelle forming agents for analyte focusing. Both analyte focusing by micelle collapse (AFMC) and sweeping‐MEKC had been investigated for the comparison of their effectiveness to examine simultaneously MTX, FNA and FA in plasma and urine under physiological conditions. In sweeping‐MEKC, the sample solution without micelles was hydrodynamically injected as a long plug into a fused‐silica capillary pre‐filled with phosphate buffer containing 3.0 mol/L of 1‐butyl‐3‐methylimidazolium bromide (BMIMBr). Using AFMC, the analytes were prepared in BMIMBr micellar matrix and hydrodynamically injected into the phosphate buffer without IL micelles. The conductivity ratio between BGE and sample (γ, BGE/sample) was optimized to be 3.0 in sweeping‐MEKC and 0.33 in AFMC resulting the adequate separation of analytes within 4.0 min. To reduce the possibility of BMIMBr adsorption, an appropriate rinsing protocol was used. The limits of detection were calculated as 0.1 ng/mL MTX, 0.05 ng/mL FNA and 0.05 ng/mL FA by sweeping‐MEKC and 0.5 ng/mL MTX, 0.3 ng/mL FNA and 0.3 ng/mL FA by AFMC. The accuracy was tested by recovery in plasma and urine matrices giving values ranging between 90 and 110%. Both stacking and sweeping by BMIMBr could be successfully used for the rapid, selective and sensitive determination of pharmaceuticals in complex matrices due to its fascinating properties, including high conductivity, good thermal stability and ability to form different types of interactions by electrostatic, hydrophobic, hydrogen bonding and π–π interactions. In sweeping‐MEKC, the using of BMIMBr enhanced the γ factor, k retention factor and the injected amount of sample. Consequently, this technique offers particular potential for higher sensitivity by giving 22‐ and 5‐fold sensitivity enhancement factors (SEFs) of MTX compared to CZE and AFMC, respectively.     

Recent progress of on-line sample preconcentration techniques in microchip electrophoresis

This review highlights recent developments and applications of on-line sample preconcentration techniques to enhance the detection sensitivity in microchip electrophoresis (MCE); references are mainly from 2008 and later. Among various developed techniques, we focus on the sample preconcentration based on the changes in the migration velocity of analytes in two or three discontinuous solutions system, since they can provide the sensitivity enhancement with relatively easy experimental procedures and short analysis times. The characteristic features of the on-line sample preconcentration applied to microchip electrophoresis (MCE) are presented, categorized on the basis of "field strength-" or "chemically" induced changes in the migration velocity. The preconcentration techniques utilizing field strength-induced changes in the velocity include field-amplified sample stacking, isotachophoresis and transient-isotachophoresis, whereas those based on chemically induced changes in the velocity are sweeping, transient-trapping and dynamic pH junction.     

Cyclodextrin‐assisted dispersive liquid–liquid microextraction for the preconcentration of carbamazepine and clobazam with subsequent sweeping micellar electrokinetic chromatography

A new version of dispersive liquid–liquid microextraction, namely, cyclodextrin‐assisted dispersive liquid–liquid microextraction, with subsequent sweeping micellar electrokinetic chromatography has been developed for the preconcentration and sensitive detection of carbamazepine and clobazam. α‐Cyclodextrin and chloroform were used as the dispersive agent and extraction solvent, respectively. After the extraction, carbamazepine and clobazam were analyzed using micellar electrokinetic chromatography with ultraviolet detection. The detection sensitivity was further enhanced using the sweeping technique. Under optimal extraction and stacking conditions, the calibration curves of carbamazepine and clobazam were linear over a concentration range of 2.0–200.0 ng/mL. The method detection limits at a signal‐to‐noise ratio of 3 were 0.6 and 0.5 ng/mL with sensitivity enhancement factors of 3575 and 4675 for carbamazepine and clobazam, respectively. This developed method demonstrated high sensitivity enhancement factors and was successfully applied to the determination of carbamazepine and clobazam in human urine samples. The precision and accuracy for urine samples were less than 4.2 and 6.9%, respectively.     

Complementary on-line preconcentration strategies for steroids by capillary electrophoresis

Complementary on-line preconcentration strategies are needed when analyzing different classes of solutes in real samples by capillary electrophoresis (CE) with UV detection. The performance of three different on-line preconcentration (focusing) techniques under alkaline conditions was examined in terms of their selectivity and sensitivity enhancement for a group of steroids, including classes of androgens, corticosteroids and estrogens. Electrokinetic focusing of large sample injection plugs (up to 28% of effective capillary length or 22.1 cm) directly on-capillary can be tuned for specific classes of steroids based on changes in their mobility (velocity) using a multi-section electrolyte system in CE. A dynamic pH junction was applied for the selective resolution and focusing of weakly acidic estrogens using borate, pH 11.0 and pH 8.0 in the background electrolyte and the sample, respectively. Sweeping, using an anionic bile acid surfactant and neutral gamma-cyclodextrin (gamma-CD) under alkaline conditions (pH 8), resulted in focusing and separation of the moderately hydrophobic (non-ionic) classes of steroids, such as androgen and corticosteroids. Optimal focusing and resolution of all test steroids under a single buffer condition was realized by a dynamic pH junction-sweeping format using borate, pH 11.0 and bile acid surfactant with gamma-CD in the BGE, whereas the sample is devoid of surfactant at pH 8.0. The design of selective on-line focusing strategies in CE is highlighted by the analysis of microgram amounts of ethynyl estradiol derived from a female contraceptive pill extract using the dynamic pH junction method, which resulted in over a 100-fold enhancement in concentration sensitivity.     

Stacking phenomena in electromigration: From basic principles to practical procedures

The review describes principles of procedures and techniques used in capillary zone electrophoresis for enhancement of sensitivity that are based on increasing analyte mass in its zone during the electromigration process, for which the term stacking is generally used. Attention is paid to intrinsic stacking in samples with low conductivity, transient isotachophoretic stacking applied in samples with high conductivity, and sweeping in micellar electrokinetic chromatography. Principles of these stacking schemes are explained, new procedures and instrumental arrangements are discussed, and all contributions involving stacking principles that have been published since the year 2000 are surveyed.     

On-line monitoring of pH junctions in capillary electrophoresis using Fourier transform infrared spectrometry

Fourier transform mid-infrared spectroscopic detection is proposed as an on-line detection technique for the study of on-line preconcentration processes in capillary electrophoresis (CE). The molecule-specific information contained in mid-IR spectra can be used to directly determine the chemical compositions of individual zones and their boundaries. This paper reports on pH junctions employed in myoglobin analysis. On-line mid-IR detection allowed the shape of the sample peak to be monitored as well as the chemical compositions of the surrounding zones. From this information it was possible to obtain detailed insights into the actual chemical compositions of the individual zones governing the efficiency of the preconcentration technique applied. The principle of measurement outlined here can therefore also be regarded as a promising one for investigating other on-line preconcentration techniques, like stacking, sweeping, and pH junction-sweeping among others. Fourier transform mid-infrared spectroscopic detection has been employed in pH junction experiments. This approach can be used to measure the chemical compositions of the phase boundaries formed, as well as the relative positions of the analyte in the zones. The principle of this technique is demonstrated by measuring myoglobin (acetate buffer, pH 4.5) in an ammonium BGE (pH 9.25)     

Analysis of proguanil and its metabolites by application of the sweeping technique in micellar electrokinetic chromatography

The method of applying large sample volumes in micellar electrokinetic chromatography termed sweeping is applied to determine the conservative limits of detection of some basic drugs in plasma and urine. The biguanides proguanil, 4-chlorophenylbiguanide and cycloguanil are used as models of basic drugs and the limits of detection obtained compared with those previously reported for capillary zone electrophoresis using field-amplified sample injection (FASI) and also by LC using off-line preconcentration. It is found that the sweeping method can be applied to extracts of such biological matrices. The limits of detection obtained by sweeping are improved over FASI for plasma but not for urine and the limits of detection are higher than those reported for LC, for these compounds.